Citation
Identification of new citrus norisoprenoids in orange juice using time intensity GC-0 and GC-MS

Material Information

Title:
Identification of new citrus norisoprenoids in orange juice using time intensity GC-0 and GC-MS
Creator:
Mahattanatawee, Kanjana
Publication Date:
Language:
English
Physical Description:
xii, 85 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Carotenoids ( jstor )
Flavors ( jstor )
Insects ( jstor )
Juices ( jstor )
Larvae ( jstor )
Mortality ( jstor )
Norisoprenoids ( jstor )
Orange juice ( jstor )
Pests ( jstor )
Dissertations, Academic -- Food Science and Human Nutrition -- UF
Food Science and Human Nutrition thesis, Ph. D
City of Lakeland ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2004.
Bibliography:
Includes bibliographical references.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Kanjana Mahattanatawee.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
022439961 ( ALEPH )
880637417 ( OCLC )
880438684 ( OCLC )

Downloads

This item has the following downloads:


Full Text











IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE
USING TIME INTENSITY GC-O AND GC-MS
















By

KANJANA MAHATTANATAWEE














A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004
































Copyright 2004

by

Kanjana Mahattanatawee
































This dissertation is dedicated to my late father.















ACKNOWLEDGMENTS

I would especially like to thank to Dr. Russell L. Rouseff, my supervisor, for his friendship and guidance throughout my graduate program. He provided me with intellectual, thoughtful discussion; encouragement; and time. Without his help and support I would not have come from Thailand and continued at the University of Florida. His high ethical standards and philosophical views will never be forgotten. I would like to thank Dr. R.M. Goodrich for all of her support, guidance and friendship. She was very kind and helpful whenever I needed her help. I also thank the other members of my supervisory committee, Dr. D.H. Powell and Dr. M.R. Marshall, Jr, for their kindness and valuable, thoughtful discussion toward my research. Especially warm thanks go to Dr. Kevin L. Goodner for his advice and support on instrumental. My special thanks go to John Smoot and Dr. Filomina Valim who assisted me any time I needed them. My appreciation goes to my dear friend Dr. Fahiem El-Borai Kora who assisted me whenever I needed his help. I extend my appreciation to all my friends at Citrus Research and Education Center (CREC) in Lake Alfred for their friendship and support. I would like also to thank my professors back in Thailand expecially Dr. Twee Hormchong, who taught me what good scientists and teachers are. My great thanks go to my beloved family. My lovely wonderful mother, father, brothers and sisters always encouraged me to follow my dream, and without their love and support, I could never be who I am today. Finally, my appreciation goes to the Siam University for giving me the financial support necessary to obtain my Ph.D from the University of Florida.


iv
















TABLE OF CONTENTS

pae

ACKNOW LEDGM ENTS ............................................................................................... iv

LIST OF TABLES ........................................................................................ viii

LIST OF FIGURES ........ ................................................ ix

ABSTRACT...................................... ......................................................................... xii

CHAPTER

1 INTRODUCTION ............................................. ..................................................... 1

2 LITERATURE REVIEW ....................................... ........... ...................3

Orange Juice Aroma ................................................................. .............................. 3
Carotenoids ......................................................................... .................................... 3
Norisoprenoids ................................................ ........................................................ 4
Norisoprenoids Formation from Carotenoids............................. ..............
Apple ................................................................ ....................................
Tomato .................................................. ............................................... 8
Saffron ................................................. .....................................................9
Grape and W ine ........................................................... 10
Gas Chromatography-Olfactometry ..................................... ............................. 11
Solid Phase M icroextraction..... ............................................................................ 13
Orange Juice Norisoprenoids................................................................................. 14

3 HPLC DETERMINATION OF CAROTENOID NORISOPRENOID
PRECURSORS IN ORANGE JUICE ..................................................................... 15

Introduction.................................. .........................................................................15
Objectives ........................................................................................................... 17
M aterials and M ethods ........................................................................................ 17
Carotenoid Extraction............................................ ........................................ 17
Carotenoid Saponification ................................................................ 17
HPLC Procedure.. ..........................................................................................18
Results and Discussion .........................................................................................19
Carotenoids of Interest ................................................................................ 19
Hydrolysis Conditions ......................................................................................20


v









HPLC Separation......................................... ........................................... 21
Carotenoid Identification.......................................... .....................................23
C onclusions......................................................... ..................................................25

4 IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING
TIME INTENSITY GC-O AND GC-MS........................................... 26

Introduction......................................................... ..................................................26
O bjectives ................................................................................................................... 28
M aterials and M ethods ............................................... ..........................................28
Orange Juice Samples and Processing ....................................... .....28
C hem icals ...................................................... ................................................29
Orange Juice Headspace Extraction ......................................... ......29
Gas Chromatography: GC-FID and GC-Olfactometer...................... ....30
Gas Chromatography-Mass Spectrometry ..................................... .....31
Aroma Peak Identification..............................................................................32
Results and D iscussion .................................................................. .......................32
Extraction and Concentration of Juice Norisoprenoids ....................................32
GC-Olfactometry ...............................................................35
Mass Spectrometry Norisoprenoid Identifications ......................... 36
C onclusion ............................................................ ................................................4 1

5 QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT
OF NORISOPRENOIDS IN ORANGE JUICE ..................................... .....43

Introduction........................... ................................................................................43
O bjectives .................................................................................................. 45
M aterials and M ethods .................................................................. .......................45
Quantification of Norisoprenoids in Orange Juice ..............................................45
Determination of the Relative Impact of Norisoprenoids in Orange Juice .........47
Results and D iscussion ..................................................................... ....................47
Quantification of Norisoprenoids in Orange Juice ........................................47
Norisoprenoid Quantification using Standard Additions .................................50
Determination of Relative Aroma Impact of Norisoprenoids ..........................54
Norisoprenoid Contribution to Total Floral Aroma ..................................... 58
C onclusion .......................................................... ..................................................6 1

6 THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL
SO LU T IO N .............................................................................. ............................62

Introduction......................................................... ..................................................62
O bjective .............................................................................. ................................. 63
M aterials and M ethods ..................................................................... ....................63
C rystallization................................................... .............................................63
Model Solutions...................................................................................................64
A nalytical M ethods .........................................................................................64
Results and discussion ......................................................................................... 65


vi









GC-O Analysis of p-Carotene Decomposition at 350C ......................................68
M S Identification .................. .........................................................................70
Conclusion .......................................................... ..................................................72

7 CONCLUSIONS .................................................... ..............................................73

LIST OF REFERENCES ................................................. ............................................75

BIOGRAPHICAL SKETCH .............................................................84












































vii















LIST OF TABLES

Table page

3-1 HPLC retention times, spectral characteristics of orange juice carotenoids ............24

4-1 Identification, retention characteristics and aroma descriptions of aroma active
com pounds in fresh orange juice........................................................................... 36

5-1 Reproducibility of SPME exposure time 45 min at 400C .....................................49

5-2 Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique.......................................... ......................................53

5-3 Concentration of P-damascenone in fresh, pasteurized and reconstituted
concentrate ............................................................................ .............................53

5-4 Aroma active compounds in orange juice grouped by citrusy/minty....................55

5-5 Aroma active compounds in orange juice grouped by metallic/mushroom/
geranium .......................................................... ...................................................56

5-6 Aroma active compounds in orange juice grouped by roasted/cooked/meaty/
spice........... ................................................................................................ 56

5-7 Aroma active compounds in orange juice grouped by fatty/soapy/green .............56

5-8 Aroma active compounds in orange juice grouped by sulfury/solventy/medicine ..57 5-9 Aroma active compounds in orange juice grouped by floral ................................57

5-10 Aroma active compounds in orange juice grouped by sweet/fruity ......................57

5-11 Aroma active compounds in orange juice grouped by green/grassy.....................58

5-12 Norisoprenoids in orange juice and peel oil.........................................60

6-1 Aroma active compounds from 1-carotene thermal degradation in model solution
pH 3.8, storage at 350C for 2 weeks .................................................69






viii















LIST OF FIGURES

Figure page

2-1 Examples of carotene and xanthophyll carotenoid structures ...................................4

2-2 Major fragment classes of carotenoid biodegradation ...............................................5

2-3 General steps for the conversion of carotenoids into flavor compounds, showing
the formation of P-ionone and P-damascenone from 0-carotene and neoxanthin
respectively...... ... .................................................................................................. 6

2-4 Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters) ...........................................8

2-5 Steven's law, comparing two difference compounds: A= compound A, B=
com pound B ....................................................... ................................................12

3-1 Possible degradation pathways for the formation of 0-cyclocitral and -ionone
from 0-carotene.............................................................................................. 15

3-2 Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of 1-damascenone .......................................... ........19

3-3 Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH, MTBE
ternary solvent gradient ............................................................22

3-4 Absorbance spectra for 1-carotene (a), leutoxanthin (b), and neoxanthin (c), peak
24, 12 and 4 respectively................................................ ............ ..............23

4-1 GC-FID (top) and average time-intensity of four GC-O runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and 23
correspond to norisoprenoids, all numbers refers to compounds in Table 4-1 ......34 4-2 Comparison between total ion chromatogram and selected ion chromatograms
(SIC) A: 0-cyclocitral, B: 0-ionone, C: o-ionone ........................................ 37

4-3 Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190 ......................................... .......38




ix









4-4 Upper spectra from orange juice MS at RT = 17.68 bottom spectra of 13cyclocitral from database NIST 2002...................................................................40

4-5 Upper spectra from orange juice MS at RT = 21.94, bottom spectra from
standard 1-ionone using identical ion trap MS at identical retention time. .............40

4-6 Upper spectra from orange juice MS at RT = 20.87, bottom spectra from
standard a-ionone using identical ion trap MS at identical retention time. .............41

5-1 Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 40oC, = 1-cyclocitral, m = 1-damascenone, A= a-ionone,
= 1-ionone ...................................................... ................................................48

5-2 Standard addition data for -cyclocitral peak area vs. added concentration in fresh
orange juice. Regression line calculated from peak area at selected mass 137.......50

5-3 Standard addition data for c-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
a-ionone concentration .............................................................51

5-4 Standard addition data for 1-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
3-ionone concentration ..............................................................52

5-5 Standard addition -damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190.........52

5-6 Standard addition data of -damascenone peak area vs. added concentration in
reconstituted from concentrate orange juice ...........................53

5-7 Aroma group profiles of fresh (+), pasteurized (U), and reconstituted from
concentrate (0) orange juice ..................................... ................. 58

5-8 Upper bar norisoprenoids contribute mainly to the total floral category, fresh =
78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent nonnorisoprenoids including linalool and unknown (LRI = 1255)............................60

6-1 The standard 1-carotene (99% purity) as received (no purification) ....................65

6-2 Headspace volatiles from 1-carotene in model solution pH 3.5 at 0 day.................66

6-3 Headspace volatiles from -carotene in model solution pH 3.5 after storage 1 day
at 350C: 1 = -ionone, a = sweet/raspberry ............................................66

6-4 Degradation of 1-carotene in model solution at difference carbon bonds ............67




x








6-5 Headspace volatiles from p-carotene in model solution pH 3.5, after storage 2
w eeks at 350C ..................................................... ...............................................69

6-6 Selected ion chromatogram (SIC) of model solution headspace volatiles after
storage 2 weeks at 350C .........................................................70

6-7 Upper spectra from model solution MS at RT 19.61, bottom spectra from
standard -cyclocitral using identical ion trap MS at identical retention time. ......70

6-8 Upper spectra from model solution MS at RT 20.46, bottom spectra from
standard P-homocyclocitral using identical ion trap MS at identical retention
time ....... .................................... 71

6-9 Upper spectra from model solution MS at RT 25.13, bottom spectra from standard
1-ionone using identical ion trap MS at identical retention time ..........................71



































xi















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE USING TIME INTENSITY GC-O AND GC-MS By

Kanjana Mahattanatawee

May 2004

Chair: Russell L. Rouseff
Major Department: Food Science and Human Nutrition

Numerous analytical studies have quantified the major volatiles in orange juice in an effort to duplicate this aroma. However, when combined, the resulting aroma does not duplicate that of orange juice, suggesting that important aroma components were missing. Citrus carotenoids have been studied primarily for their role as pigments and have generally been ignored as a source of aroma compounds. Carotenoids can be degraded into smaller (C9-C13), more volatile products called norisoprenoids. Norisoprenoids have been identified as aroma impact in foods containing carotenoids (i.e. tea, grapes, tomato, and saffron). Therefore, carotenoid-decomposition may be responsible for a portion of orange juice aroma and also aroma changes associated with thermal processing or elevated temperature storage.

Three norisoprenoids, 0-cyclocitral, 3-damascenone, ct-ionone, were fully

identified in fresh and pasteurized orange juice for the first time. Beta-ionone was also detected but had been previously identified. Identification was based upon SPME


xii









headspace, GC-O, and GC-MS data. Only two norisoprenoids (p-damascenone and

-ionone) were detected in reconstituted juice that had been thermally concentrated. Peel oil from the same fruit contained only -damascenone and 3-cyclocitral.

Concentrations of -cyclocitral, -damascenone, a-ionone, and -ionone were

determined using standard addition SPME and GC-MS and found to be 145, 0.09, 47 and 83 tg/L respectively. The concentration of 3-damascenone increased from 0.09 to 0.85 Atg/L after thermal concentration and reconstitution. Orange juice norisoprenoids contribute approximately 8-10% of total aroma intensity as determined from combined aromagram peak heights and 60-80% of the total floral-category.

Known norisoprenoids precursors (-carotene, a-carotene, a-cryptoxanthin,

0-cryptoxanthin, and neoxanthin) were identified in Valencia orange juice using C30 reverse phase HPLC with photodiode array detection.

Thermal decomposition products of -carotene in citric acid solutions buffered at pH 3.8 were examined during 350C storage using GC-O and GC-MS. Beta-cyclocitral, P-homocyclocitral, -damascone and 0-ionone were detected after 2 weeks thus demonstrating that -carotene can produce norisoprenoids. Since half of the a-carotene, a-cryptoxanthin, -cryptoxanthin structures share the identical structure as -carotene, these carotenoids must be considered potential norisoprenoid sources as well.













xiii














CHAPTER 1
INTRODUCTION

The delicate aroma of fresh orange juice is the result of a complex mixture of volatiles blended in specific proportions. Numerous analytical studies (1-5) have identified and quantified the major volatiles in orange juice in an effort to duplicate this aroma. However, when combined, the identified volatiles could not duplicate orange juice aroma, suggesting that important aroma components were missing. Early orange juice gaschromatography olfactometry (GC-O) studies (4, 6, 7) have shown that many of the aroma-active compounds in orange juice exist as low-level volatiles that are difficult to detect using typical flame ionization detector (FID) or mass spectrometer (MS) detectors. Furthermore, these studies demonstrated that the major volatiles in orange juice have little to no aroma activity. Recent orange juice GC-O studies (5) quantified the 25 most intense aroma-active compounds in fresh juice, using isotope dilution analysis. Model solutions of the aroma components in orange juice based on GC-O studies have come closer to duplicating the aroma of fresh orange juice than model systems based on the composition of the volatiles found in highest concentration.

Carotenoids are too large (C40) to be volatile under normal conditions. Because they contain a highly conjugated double bond structure, they can be degraded by enzyme, chemical, and/or thermal reactions to form a wide range of structures, depending on which double bond is broken. Some of their smaller (C9-CI3), volatile, decomposition products are called norisoprenoids. Norisoprenoids have been shown to have significant aroma impact in fruits, vegetables and spices such as grapes, apples, lychee, starfruit,


1






2


mango, tomato, saffron, cured tobacco, and black tea (8-16). Only a single norisoprenoid, 0-ionone, has been identified to date in fresh orange juice (4, 5)

More than 50 carotenoids have been separated and identified from the juice of three varieties of Citrus sinensis (Shamouti, Valencia, and Washington Navel) using column chromatography combined with thin layer chromatography (TLC) (17). Some of these carotenoids (such as 0-carotene, a-carotene, neoxanthin, 0-crytoxanthin, lutein, violaxanthin, and canthaxanthine) have the structural potential to form potent norisoprenoid fragments (18-22). Furthermore, 3-carotene in tomato products has been shown to produce 0-ionone and 0-cyclocitral (23). Beta-ionone and a-ionone have been generated from 0-carotene and a-carotene respectively in carrots (24). Neoxanthin in grapes has been shown to be a source of 3-damascenone (25). Prior orange juice carotenoid studies were primarily directed toward the contribution of carotenoids to juice color and for vitamin A content. They have been generally ignored as precursors of aroma compounds. Since orange juice has so many carotenoids that could serve as precursors for a wide range of norisopernoids, the objectives of this research were to:

1. Confirm the presence of possible carotenoid norisoprenoid precursors in orange juice
using HPLC and photodiode array detection. (Chapter 3)

2. Determine if additional norisoprenoid are present in orange juice. Characterize and
identify these new norisoprenoids. (Chapter 4)

3. Determine the relative aroma impact of carotenoid degradation products
(norisoprenoids) to the total aroma impact of orange juice in fresh, pasteurized, and
reconstituted from concentrate juice. (Chapter 5)

4. Develop quantitative procedures to isolate and quantify orange juice norisoprenoids
using static headspace SPME with GC-MS. (Chapter 5)

5. Determine i.f 1-carotene can form norisoprenoid degradation products at 350C storage
in model solutions. (Chapter 6).














CHAPTER 2
LITERATURE REVIEW

Orange Juice Aroma

The aroma of fresh orange juice is composed of a complex mixture of aldehydes, esters, ketones, alcohols and terpenes blended in specific proportions. Numerous studies (1-5) aimed at identifying the flavor volatiles in orange juice have led to the identification of about 200 volatiles, but no single aroma character impact for orange flavor has ever been reported. GC-olfactometry and orange juice volatile quantification have been used to gain a more accurate understanding of their contribution to orange flavor (2, 5-7). Early orange juice GC-O studies have demonstrated that the orange juice volatiles present in highest concentration have little to no aroma activity and many aroma active compounds exist as low-level volatiles that are difficult to detect using typical instrumental detectors.

Carotenoids

Carotenoids are primarily responsible for the colors of many plants, birds, and insects; but also serve as plant photoprotection agents during photosynthesis, and as essential human nutrients. However, the least-appreciated role of carotenoids is their function as aroma precursors.

Carotenoids are tetraterpenes (C40 ) resulting from the joining together of eight molecules of isoprene (C5) through "tail-to-tail" condensation. Most carotenoids have a C40 carbon skeleton. The ends may or may not be cyclized into six membered rings. If the ends are not cyclized, the molecule is termed acyclic. There are two main groups of carotenoids:


3






4


the hydrocarbon group, which contain only carbon and hydrogen; and the xanthophyll group, which contain carbon, hydrogen, and oxygen. Oxygen in xanthophylls is usually found as either hydroxyl-(monols, diols and polyols), epoxy- (5,6 and 5,8-epoxides), methoxy, aldehyde, oxo, carboxy and/or esters. Hydroxyl substitution primarily occurs at the C3 position in the ionone ring; and a carbonyl substitution usually occurs at the C4 position in the 3-ionone ring. In most of the cyclic carotenoids, the 5,6- and 5',6'-double bonds are the most susceptible to epoxidation. The unconjugated double bond in the aionone E ring does not undergo epoxidation. Allenic carotenoids have a C=C=C grouping at one end of the central chain, and acetylenic carotenoids have a -C-C- bond in position 7,8 and/or 7',8' (26, 27). Figure 2-1 shows acyclic carotene (lycopene), bicyclic carotene (0-carotene), the monol 1-cryptoxanthin, and the diol zeaxanthin.




Lycopene


-carotene



H-O fcryptoxanthin OH



FHO Zeaxanthin


Figure 2-1. Examples of carotene and xanthophyll carotenoid structures.

Norisoprenoids

Norisoprenoids are volatile C9-C13 fragments from the degradation of the C40

carotenoids. The formation of norisoprenoids from carotenoids is thought to proceed via






5


enzymatic and nonenzymatic pathways. Nonenzymatic cleavage reactions include photooxygenation (18), auto-oxidation (28, 29), and the thermal degradation processes (30, 31). The in vivo cleavage of the carotenoid chain is generally considered to be catalyzed by dioxygenase (lipoxidase and peroxidase) systems and require molecular oxygen and other cofactors for activity. The polyene chain of carotenes is readily oxidized, giving rise to cyclic and acyclic compounds (often having an oxygen-containing functional group on a trimethylcyclohexane ring, or an oxygen-containing functional group on the allyllic side chain). Although all the in-chain double bonds seem to be vulnerable to enzymatic attack, in actuality the formation of major fragment classes with 10, 13, 15 or 20 carbon atoms are most common (see Fig. 2-2). In fruit tissues, the bio-oxidative cleavage of the 9,10 (9',10') double bond seems to be the most preferred (15, 32, 33).




12' 10' 8' 4' ,

3 4 51 8 10 12 14 7' C20 (Retinoids)
Cl0 (e.g. Plant Hormones)
Cl3 (e.g. 0-ionone)
Clo (e.g. Safranal and 0-cyclocitral

Figure 2-2. Major fragment classes of carotenoid biodegradation. Norisoprenoids Formation from Carotenoids

Norisoprenoids can be generated from carotenoids via either direct cleavage or

cleavage followed by subsequent reactions. In the latter process, three steps are required to generate an aroma compound from the parent carotenoid: 1) the initial dioxygenase cleavage, 2) subsequent enzymatic transformations of the initial cleavage product giving






6


rise to polar intermediates (aroma precursors), and 3) acid-catalyzed conversions of the

nonvolatile precursors into the aroma active form (32). One example illustrating these

reaction is the formation of B-damascenone from neoxanthin (Fig. 2-3). The primary

oxidative cleavage product of neoxanthin, grasshopper ketone, must be enzymatically

reduced before finally being acid-catalyzed converted into the odoriferous ketone. In the

direct process, the target compound is immediately obtained after the initial cleavage

(i.e., formation of a- and 0-ionone directly from a- and 0- carotene) (34).

Carotenoid
0-carotene neoxanthin
Step 1
o oxidative cleavage Primary cleavage 0H

Step 2 HO
-ionone Enzymatic
transformation
OH
Non-volatile metabolite H (aroma precursor)
HO
Step 32H20
Acid catalyzed -2H20 conversions
Aroma compound

1-damascenone

Figure 2-3. General steps for the conversion of carotenoids into flavor compounds,
showing the formation of 0-ionone and 1-damascenone from 1-carotene and
neoxanthin respectively (Winterhalter, P., Rouseff, R. Carotenoids-Derived
Aroma Compounds: An Introduction. In Carotenoid-derived Aroma
Compounds; P. Winterhalter and R. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002, Fig. 4, page 12).

Recent studies have shown that some of the volatile C13-compounds are not free,

uncomplexed plant constituents; but rather are derived from less or nonvolatile precursors

such as polyols, glycosides, and glucose esters. Carotenoid degradation is initiated by






7


oxidative cleavage of the intact carotenoid. After further enzymatic transformation steps, the primary cleavage products are converted into reactive Clo to C13 fragments of the initial carotenoid. These volatile fragments can be stabilized and made nonvolatile by glycosylation (which involves glycosyltransferases of those norisoprenoid compounds possessing a hydroxyl group) (35, 36).

Glycosilation stabilizes and solubilizes norisoprenoids in plant systems.

Degradation of the glycoconjugates librates the potent volatile and can produce profound aroma changes. This process can be acid-catalyzed (e.g., during fruit processing) (9) or enzymatic (e.g., during fermentation) (37). Another important class of precursors is the polyols, which upon (allylic) elimination of water is transformed into volatile forms. An example is the reactive allyl-1,6-diol that, under gentle reaction conditions (natural pH, room temperature), is converted into isomeric theaspiranes (38). A third class of carotenoid-derived aroma precursor is glucose esters (e.g., C1o-compounds derived from the central part of the carotenoid chain, which is left after the cleavage of the endgroups) that gives rise to isomeric marmelo lactones, key aroma constituents of quince fruit (see Fig. 2-4) (39).

Norisoprenoids have been shown to have significant aroma impact in fruits (apple, mango, grape, starfruit, lychee, passion fruit, nectarine, etc.) (9-12, 19, 36, 40) vegetables (tomato (41)), spices (saffron (14) and paprika (42)), leaf products (tobacco (43) and tea

(44)) as well as roses (45), wine (46) and oak wood (47). Apple

Beta-Damascenone is a potent aroma compound found in a variety of natural

products, with a threshold of 0.002 g/L in water (48). Eight separate B-damascenone






8


precursors have been detected in apples (Malus domestica Borkh. cv. Empire) (49). The most abundant precursor, present at 4.6 ng/g, was the 9 (or 3) -a-L-arabinofuranosyl(1,6)-o3-D-glucopyranoside of the acetylenic diol. The second most abundant precursor, present at 3.1 ng/g, is a more polar glycoside of the acetylenic diol. (49). BetaDamascenone contributed 32% of the total aroma potency of heated apple juice, but only

1.6% of the total aroma of fresh apple juice as determined by GC-O. Thus, most of the 1-damascenone in heated apple juice was generated from nonvolatile precursors during thermal processing (9).

OH



o -H20

Allyl-1,6-diol Theaspiranes OH

Ho O H+
HO HO -H20
HO
P-D-Gentiobioside of 3-Hydroxy-3-ionol OH H+
H20 o

-Glu -H20 Marmelo Lactone

Figure 2-4. Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters). Tomato

One of the most marked differences between the fresh tomato and the paste is the almost complete loss of the major contributor to fresh tomato aroma, (Z)-3-hexenal. The most notable increase is with the potent odorant, 3-damascenone, which shows a 10 fold





9


increase in concentration in the paste (50). Beta-ionone in tomatoes seems to be formed mainly by an oxidative mechanism. It was not detected among the glycoside hydrolysis products. The compound -damascenone was shown to be produced in fruits from hydrolysis of glycosides via an intermediate acetylenic compound megastigm-5-en-7yne-3,9-diol. This appears to be the final step in tomato volatile norisoprenoid formation

(51). Three experimental lines of tomato: a high-3-carotene line; a high-lycopene line; and a low-carotenoid line were examined for their norisoprenoid content. In fresh tomato, the high 0-carotene line produced the highest concentrations of B-ionone (17 gg/L, versus 1 lpg/L in the low-carotenoid line) and P-cyclociral produced 30 gg/L in the high carotene line versus 0 pg/L in the low-carotene line). Both norisoprenoids are known biological or chemical degradation products of -carotene. The high lycopene line, however, did not show any significantly higher concentration of the expected lycopene degradation products, 6-methyl-5-hepten-2-one, 6-methyl-5-hepten-2-ol, or geranylacetone. It did show a significantly higher value for geranial (21 gg/L) compared to that of the common line (12 plg/L). Geranial could be considered a lycopenedegradation product (41).

Saffron

Safranal (monoterpene aldehyde, C10H140) is the characteristic impact compound of saffron (dried stigmas of Crocus sativus), formed in saffron during drying and storage by hydrolysis of picrocrocin. Picrocrocin was the colorless glycoside of the aglycone, 4-hydroxy-2,6,6-trimethyl- -carboxaldehyde- -cyclohexene (HTCC), which was the main substance responsible for the bitter taste of saffron. Safranal was not present in the fresh stigma. Its concentration in saffron depended strongly on both the drying and





10


storage conditions. Additional flavor compounds in saffron were formed upon cooking of the spice (52). Aroma isolates of saffron have been prepared by simultaneous distillation extraction (SDE) at pH 2.6 as well as liquid-liquid extraction using pentane: diethyl ether (1:1) as solvent. Aroma activity and relative aroma strength was determined using aroma extract dilution analysis (AEDA). Compounds with high FD-factors were safranal and 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien- -one as well as linalool and isophorone. The 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-1-one was only detected in the SDE isolate and not in the liquid-liquid extract. This result shown the presence of certain forms of precursors, which upon heat treatment are converted into the aroma compound 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien- -one (53). Grape and Wine

Norisoprenoids are important aroma constituents of grape and wine. They are

thought to arise from carotenoid breakdown; and occur in grapes as glycosidically bound precursors. The major carotenoids in grape are 3-carotene and lutein, representing nearly 85% of the total carotenoids. These are accompanied by minor carotenoids such as neoxanthin, violaxanthin, lutein-5,6-epoxide, zeaxanthin, neochrome, flavoxanthin, and luteoxanthin (54). Grape carotenoids decrease progressively during maturation, with a concomitant increase of the volatile compounds. This degradation would occur during berry metabolism either enzymatically or by chemical pathway in acid medium (54, 55). This would account for the presence of volatile compounds, such as 1-ionone and 1-damascenone, identified in grape (56) and possibly originating in carotenoids (36). Many norisoprenoids occur in grapes as glycosidic precursors. Enzymatic and acid





11


cleavage during crushing, fermentation, and bottle-aging result in cleavage of the bound sugar moiety releasing the free norisoprenoid aglycone (57, 58).

Gas Chromatography-Olfactometry

Gas chromatography-olfactometry (GC-O) is a technique that allows the effluent from the GC column to be evaluated for aroma activity using the human nose. The effluent from the GC column is usually split between an FID detector and sniff port. The human being detects which of the volatiles eluting from a GC column are aroma active, as well as to describe aroma quality, and to estimate aroma intensity. The FID detector is used as a general mass detector. Some of the GC-O techniques available are Charm@ Analysis (59), Aroma Extraction Dilution Analysis (AEDA) (60), and OSME (61) which is a time intensity method. Charm Analysis and ADEA are based on the determination of odor detection thresholds of the compounds through a series of dilutions. Both define aroma strength in terms of its dilution strength. OSME determines intensities based upon magnitude estimation using a variable potentiometer to estimate intensity. Da Silva et al.

(61) suggested that dilution techniques might not give accurate values of aroma intensity, since the odorants may have different dose-response functions above their thresholds. Steven's law (62) establishes that the odor intensity (1) of a compound increases as a power function (n, which varies from compound to compound) of the concentration within a certain range of concentration (C) directly above the detection threshold (T). The law is commonly expressed as:

I= k (C-T),

where k represents the proportionality constant. Response will increase once the threshold concentration is exceeded. Even though not defined by the above equation, a limit will be reached where the sensory response will no longer increase with increasing






12


concentration. This point is defined as saturation. When a sample is diluted below the odor-detection threshold, there will be no sensory response. Steven's law suggests that two different compounds (A and B) at the same concentration, with similar detection thresholds but with different exponents (n values), will produce different dose-odor intensity profiles (Fig. 2-5). Individual odors will contribute differently to the overall food aroma intensity, depending on their concentration and n value (61).



saturation

1so-intensity,"

a---- -- --- ------ -- -- -- --;1a

iA/n tens
0 threshold ,


I I
I I
I I I I I
10 100 1000 10000 100000 Dose

Figure 2-5. Steven's law, comparing two difference compounds: A= compound A, B=
compound B.

The OSME is a time intensity procedure that determines the intensity of the

perceived odor without dilution. In this method, the trained assessors sniff the effluents from GC mixed with humidified air, and directly record the odor intensity and duration of each odor active component while describing its odor quality. Intensities of individual components are plotted versus elution time; and the resultant graph is known as an aromagram.





13


Solid Phase Microextraction

Solid phase microextraction (SPME) is a relative new technique whereby analytes of interest partition from the sample matrix into a polymeric coated silica fiber, developed by Pawliszyn and co-workers (63). It is a simple, rapid, solventless technique to sample static headspace volatiles. A 1 or 2 cm length of fused silica fiber, coated with a polymer, is bonded to a stainless steel plunger and installed in a holder that looks like a modified microliter syringe. The plunger moves the fused silica fiber into and out of a hollow needle. To use the unit, the analyst draws the fiber into the needle, passes the needle through the septum that seals the sample vial, and depresses the plunger, exposing the fiber to the sample or the headspace above the sample. Organic analytes adsorb to the coating on the fiber. After adsorption equilibrium is attained, the fiber is drawn into the needle, and the needle is withdrawn from the sample vial. Finally, the needle is introduced into the gas chromatographic (GC) injector, where the adsorbed analytes are thermally desorbed and delivered to the GC column.

The application of headspace SPME to flavor volatile compounds has been

employed in the study of flavor volatiles in orange juice using a PDMS coated fiber (64), a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65), PDMS and polyacrylate fiber (66). The partition coefficients of the polymeric coatings for the analyses differed markedly. For example, terpenes such as a-pinene, P-myrcene, y-terpinenes, and limonene are all nonpolar, and were extracted to a higher degree into the nonpolar PDMS coating (66). Corresponding PDMS extracted the least amount of the more highly polar volatiles, PDMS/DVB and Carbowax/DVB had partition coefficients higher than that of PDMS for the most polar molecules (67). The Carboxen-PDMS fiber coating was more





14


selective for terpenes than early eluting alcohols and aldehydes (6). Polyacrylate was more effective in extracting highly polar compounds such as methanol and ethanol (66).

Orange Juice Norisoprenoids

Only a single norisoprenoid (p-ionone) has been reported and completely identified in fresh orange juice (4, 5). Recently, -damescenone has been reported in heated orange juice, but not completely identified (65). With so many carotenoid precursors present in orange juice, it seems highly likely that additional norisoprenoids would also be present. The primary objective of this study was to determine if these additional norisoprenoids were present in orange.














CHAPTER 3
HPLC DETERMINATION OF CAROTENOID NORISOPRENOID PRECURSORS IN ORANGE JUICE

Introduction

The color of orange juice is due to a complex mixture of plant pigments called carotenoids. Over 50 carotenoids have been identified in orange juice including 1carotene, a-carotene, 1-cryptoxanthin and neoxanthin (17, 68). In addition to acting as plant pigments and free radical scavengers (produced during photosynthesis), these large highly conjugated molecules can break down forming smaller, highly potent aroma volatiles called norisoprenoid (15, 39, 69, 70). The structure of 1-carotene is shown below. If this molecule hydrolyzes between carbon atoms 9 and 10, a C13 norisoprenoid, 0-ionone is formed. If the molecule hydrolyzes between carbon atoms 7 and 8, then a Clo norisoprenoid, 0-cyclocitral is formed.


0-carotene

8 10



10-cyclocitral 1 [-ionone


Figure 3-1. Possible degradation pathways for the formation of -cyclocitral and 3ionone from 0-carotene.

Beta-cyclocitral, 0-damascenone, a-ionone, and P-ionone are some of the volatiles reported in tomato (23), wine (71), tobacco (22) and tea (16) as aroma active compounds.



15






16


In each case carotenoids have been shown to be their precursors. Citrus carotenoids have been examined using a variety of chromatographic techniques such as column and thin layer chromatography, TLC, and more recently high performance liquid chromatography, HPLC (68, 72, 73). Once separated, individual carotenoids have been identified primarily by their unique "three fingered" visible absorbance patterns. In orange juice, most oxygen containing carotenoids are esterified with C12-C18 fatty acids (74) thus increasing their size and structural complexity. The most common practice is to deesterify (hydrolyze) these esters so that each carotenoid will elute as a single peak rather than several smaller peaks with various fatty acids attached. However, even with hydrolysis, the large numbers and subtle structural differences of orange juice carotenoids provide a severe separation challenge. To complicate matters further, carotenoids are sensitive to heat, light and oxygen, thus artifacts are readily formed during sample preparation and/or analysis steps. HPLC equipped with a photodiode array detector is the preferred analytical technique of choice to separate and quantify carotenoids without artifact formation. Both normal phase and reverse phase chromatography have been employed to separate these plant pigments, but the reverse phase approach offer the most advantages. The most common reverse phase column is C-18 and most of the early carotenoid separations employed this column. However, in recent years a C-30 reverse phase column has been developed especially for carotenoid separations. Several investigators (68, 75, 76) have employed this column with ternary solvent gradient and photodiode array detector to isolate and identify the complex mixture of carotenoids in orange juice.





17


Objectives

The objective of this study was to confirm the presence of specific carotenoids in Valencia orange juice which could serve as norisoprenoid precursors. The specific carotenoids of interest include: a-cryptoxanthin, 0-cryptoxanthin, a-carotene, -carotene and neoxanthin because they possess the structural features needed to serve as precursors to the newly identified norisoprenoids. (See Objective #1) Materials and Methods

Carotenoid Extraction

The carotenoid extraction method according to Lee et al. (75) was carried out with slight modification. A 25 mL aliquot of Valencia juice was extracted with 50 mL of a mixed solvent (hexane:acetone:ethanol, 50:25:25) using a Omni mixer homogenizer (model no. 700, Lourdes, Vernitron Medical Products, Inc. Carlstadt, NJ). It was extracted for 5 min at medium speed in ice bath, and centrifuged (CR412, Jouan, Inc., Winchester, VA) for 10 min at 4000 rpm and 10'C. The top layer of hexane containing pigments was collected and concentrated to dryness in rotary evaporator. Carotenoid Saponification

Saponification was carried out according to Noga and Lenz (77) with slight

modification. The dried pigment was redissolved with 2 mL of methyl tert.-butyl ether (MTBE), and placed in a 40 mL vial. Two mL of 10% methanolic potassium hydroxide (KOH) was added to the sample and the headspace was blanketed with nitrogen before closing. The sample was wrapped with aluminum foil to protect it from light, and placed at room temperature for 1 hour. The sample was then transferred to separatory funnel to which 5 mL of water was added and 2 mL of 0.1% butylated hydroxyl toluene (BHT) in MTBE, and the aqueous layer removed. Additional water rinses were carried out until





18


free of alkali. The MTBE layer was then filtered through a small glass column filled with deactivated glass wool (Restek Corporation, PA) and anhydrous sodium sulfate (Fisher Scientific, NJ) to remove residue water from MTBE layer. Each sample was concentrated by evaporation with nitrogen, and the volume adjusted with 0.1% BHT in MTBE to 1 mL and placed in sealed amber vials under refrigeration (4oC) until analyzed. HPLC Procedure

Carotenoid pigments were analyzed according to Rouseff et al.(68) by reverse phase HPLC using ternary gradient of water, methanol, and MTBE with photo diode array detection (PDA] by reverse phase HPLC using ternary gradient of water, methanol (MeOH), and MTBE with photodiode array detection (PDA). The 4.6 mm i.d. x 250 mm YMC CarotenoidTM 5 pm column (YMC, Inc., Waters Corporation, MA) was used. The chromatographic system consisted of autosampler, LC pump, and PDA detector (Surveyor, ThermoFinnigan, CA). The PDA was set to scan from 280 to 550 nm. Three separate data channel were set to record the absorbances at 350, 430, and 486 nm with spectral bandwidths of 1 nm. Data were collected, stored, and integrated, using the Atlas software (Atlas 2003, Thermo Electron Corporation, Cheshire, UK). All reagents used were HPLC grade (Fisher Scientific, NJ). One standard, -carotene, was purchased from Acros (Acros, NJ). The initial ternary gradient composition consisted of 90% MeOH, 5% water, and 5% MTBE. The solvent composition changed in a linear fashion to 95% MeOH and 5% MTBE at 12 min. After the next 8 min (at 20 min) the solvent composition was 86% MeOH and 14% MTBE. At this composition the solvent composition was gradually changed to 75% MeOH and 25% MTBE at 30 min. The final composition was 50% MeOH and 50% MTBE at 50 min. Intial conditions were





19


reestablished within 2 min and reequilibrated for 15 min before the next injection. Flow rate was 1 mlJmin and injection volume was 10 gL.

Results and Discussion Carotenoids of Interest

Although over 50 carotenoids have been identified in orange juice, only a few

possess the structural requirements to produce potent norisoprenoids. The structures of the carotenoids which have been shown to produce norisoprenoids of interests in other food systems (15, 23, 31, 39, 69, 70, 78) are shown in Fig. 3-2. Hydrolysis points are indicated with arrows and resulting norisoprenoid indicated as text.


10 8onone


0-cyclocitral ~ e ioonoe --cydodtral
n-carotene

7 9 10 10 8 a-carotene
-cyocitral -ionone a-carotene a-ionone
hy o --io-ionone



10 0-cryptoxanthin hydroxy--ionone
0-cyclocir 0-lonone OH a-cryptoxanthin
D-cydocitral 0-ionone hydroxy-a-ionone OH


neoxanthln
HO OH


Figure 3-2. Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of P-damascenone.

It is worth noting that the structures of the left half of the first four carotenoids are identical. Each of these four carotenoids can produce either 3-cyclocitral or 8-ionone.





20


The right half of these four carotenoids differ considerably. The right half of 0-carotene can produce both 0-cyclocitral and 3-ionone because it is identical to the left half of the molecule. The right half of 0-carotene can produce a-ionone and neither (a- or 1cryptoxanthin produce norisoprenoids which were observed in orange juice. The final carotenoid of interest, neoxanthin, has been shown to produce 1-damescenone in a three step process (34).

Hydrolysis Conditions

As previously discussed, citrus carotenoids must be hydrolyzed to simplify the separation due to the complexity from the multiple natural esters formed from C12-C18 saturated fatty acids (74). Hydrolysis conditions must be optimized in order to free the esterified carotenoids into a single form but not so long as to promote alkaline hydrolysis of the carotenoids. Concentration of alkali, reaction time and temperature are the variables of interest. In recent years, most carotenoid studies have employed 0.1 M KOH and room temperature so only reaction time was optimized for this study. Chromatograms with no saponification showed 80% of the total peak area eluting as an unresolved band of peaks during the last quarter of the chromatogram. As saponification time increased, the number of peaks at the end of the chromatogram diminished and the peaks were more evenly distributed during the chromatographic run. Saponification times in excess of one hour did not reduce the number of late eluting peaks and total carotenoid peak area was lower at 4 hours and overnight saponification compared to the one hour saponification. Therefore the one hour saponification was used for the remainder of the study.





21


HPLC Separation

C-30 carotenoid columns with ternary gradient solvent systems and photodiode array detectors have been employed to separate and identify carotenoids in citrus (68, 75, 76). In this study saponified carotenoids were separated using a C-30 carotenoid column with ternary solvent gradient system of water, methanol, and MTBE with photodiode array detection. The resulting separation is shown in Figure 3-3. More than twenty-four carotenoids were separated as distinct peaks and sixteen of these peaks were identified based on their spectral characteristics (Table 3-1), relative elution order compared to literature values and authentic standards. As seen from the chromatogram in Fig. 3-3, peaks 11 and 20 corresponding to cis-violaxanthin and -cryptoxanthin (15.76 and 12.34 percentage of total peak area, respectively). They have been reported as the major carotenoids in earlier studies. Beta-cryptoxanthin is well accepted as the major contributor to the orange color of the juice (79) because of its relatively high concentration an overall absorbance in the red/orange range of the spectrum. The last four peaks (21-24) are due to a variety of carotenes which are not completely resolved. Both a- and -carotene are of particular interest in this study because of their ability to produce a range of norisoprenoids which have been observed in other food products. In addition, peaks 18 and 20 were well resolved and corresponded a- and -cryptoxanthin from the match of retention times and spectral characteristics compared to authentic standards. The final peak of interest was neoxanthin and this compound corresponds to peak 4 which is neither particularly well resolved nor large.

The large number and similarity of orange juice carotenoids make separation difficult. Thus even under the best chromatographic systems, some peaks will not be






22


well resolved (i.e., peaks 8-10) and make accurate identification difficult. Lutein and zeaxanthin (peaks 14 and 15, respectively) are usually difficult to resolve as they differ only in the position of a single double bond in one of the terminal rings. These pigments can be separated on a ZnCO3-MgCO3 column and the separation requires several hours

(80) but they are completely resolved in this chromatographic system. However lutein is barely resolved from mutatoxanthin (peak 13) even though mutatoxanthin contains an extra 5,8 epoxide group. Phytofluene and a-cryptoxanthin (peak 19 and 18, respectively) were not well resolved chromatographically, but could easily be separately quantified using different monitoring wavelengths as their respective absorbance maxima differ by approximately 100 nm.

1000 1




S15
cI 6
0500 1 16


12
241 18




0 10 20 30 40 50 Retention Time (min)

Figure 3-3. Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH,
MTBE ternary solvent gradient. Spectral characteristics for each numbered
peak are summarized in Table 3-1. See HPLC experimental section for
additional details.





23


Carotenoid Identification

Shown in Figure 3-4 is an overlay of peaks 4, 12 and 24. The height of their

spectra corresponds to their relative peak heights since the spectra were taken from the apex of each peak. These were chosen to show the range and diversity of these spectra which are not conveyed when just tabulated peak maxima are tabulated. The shape of the absorbance band as well as the location of the absorbance maxima are all highly characteristic of individual carotenoids. This information taken with retention time can


100


(b)


"d 50


(C)

0

280 320 360 400 440 480 520
Wavelength (nm)

Figure 3-4. Absorbance spectra for -carotene (a), leutoxanthin (b), and neoxanthin (c),
peak 24, 12 and 4 respectively.


be used to identify specific carotenoids, especially if their spectral and chromatographic characteristics have been reported elsewhere. The spectra and relative retention times of a-, 1-carotene, a-, B-cryptoxanthin and neoxanthin matched their published values and were used as confirmation of the presence of these peaks in orange juice. It should be pointed out that these five carotenoids have been previously reported in orange juice (17).






24


As shown in Table 3-1 the absorbance maxima observed exactly matched those published

in the literature or differed at most by 2 nm as in the case of the central peak for 3cryptoxanthin. Since the wavelength accuracy of most photodiode array detectors is only

+ 1 nm, the agreement is excellent. Since the carotenoids of interest have the same

elution and spectral characteristics as a-, 0-carotene, a-, P-cryptoxanthin and neoxanthin,

it is reconfirmed that they are present in orange juice and could potentially serve as

norisoprenoid precursors.

Table 3-1. HPLC retention times, spectral characteristics of orange juice carotenoids Peak RT a Observed (nm) Literature (nm) Carotenoid (min) Peak Peak Peak Peak Peak Peak Ref.b
no. (min) 2 3 1 2 3
1 Valenciaxanthin 5.52 351 369 390 351 369 390 E
2 6.00 371 391 414 3 6.92 420 435 465
4 Neoxanthin 7.35 416 438 468 415 439 467 A
5 11.60 410 431 454 6 12.95 416 438 467
7 Neochrome 14.68 400 422 448 399.5 421.5 447.5 B
8 15.55 408 429 415 9 15.90 383 402 425 10 16.83 411 430 462
11 Cis-violaxanthin 17.60 415 437 464 414 437 464 C 12 Leutoxanthin 19.07 399 418 443 399.5 419.5 441.5 B 13 Mutatoxanthin 20.08 405 429 451 404 427 452 D 14 Lutein 20.92 420 445 471 424.5 445.5 471.5 B 15 Zeaxanthin 23.80 425 450 476 425 450 478 A 16 Isolutein 24.70 418 441 468 418 439.5 467.5 B
17 26.40 429 445 469
18 a-cryptoxanthin 28.15 420 445 472 420 444 472 D 19 Phytofluene 28.83 331 348 367 331 348 367 A 20 0-cryptoxanthin 31.57 425 451 477 425 449 476 A
0-carotene,5,8:
21 5',8'-diepoxide 34.30 380 400 424 380 400 425 A 22 a-carotene 36.33 420 446 472 420 445 472 D 23 f-carotene 39.28 379 401 425 378 400 425 A 24 1 -carotene 39.77 425 451 477 425 450 478 A

aRT = retention time, bA= Britton (81); B= Rouseff et al. (68); C= DeRitter and Prucell
(82); D= Farin et al. (83); E= Curl and Bailey (84).





25


Conclusions

Carotenoids in Valencia orange juice were extracted using mixed solvent (hexane:acetone:ethanol, 50:25:25) and subsequently saponified. The saponified carotenoids were separated using a C-30 carotenoid column with a ternary gradient solvent system. Twenty-four carotenoids were separated as distinct peaks and sixteen of these peaks were identified based on their spectral characteristics (Table 3-1), relative elution order compared to literature values and authentic standards. Although they have been reported previously, the presence of ca-cryptoxanthin, 3-cryptoxanthin, a-carotene, 1-carotene and neoxanthin in orange juice was confirmed by comparing retention and spectral properties with standards or literature values. These specific carotenoids were of interest because they possess the direct structural segments needed to serve as precursors potent aroma norisoprenoids.













CHAPTER 4
IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING TIME INTENSITY GC-O AND GC-MS

Introduction

Early GC-O studies (4, 6, 7) have shown that many aroma active compounds in orange juice exist as low-level volatiles that are difficult to detect using typical FID or MS detectors. Furthermore, these studies also demonstrated that the orange juice volatiles present in highest concentration have little to no aroma activity. Recently, the 25 most intense aroma active compounds in fresh juice, as determined by dilution analysis (5), were quantified using isotope dilution analysis (5). Beta-ionone is the only orange juice norisoprenoid, which has been fully identified (4, 5). Even though it has a moderately intense aroma, it was not one of the 25 odorants recently quantified using isotope-dilution analysis (5). Norisoprenoids are volatile C9-C13 fragments with extremely low aroma thresholds which can be formed from the degradation of C40 carotenoids. This degradation can be the result of in vivo enzymatic reaction, or post harvest thermal degradation in foods containing carotenoids. They are also observed from the release of glycosidically bound norisoprenoids which were originally from carotenoid decompositions as in the case of wine (58). Norisoprenoids have been shown to have significant aroma impact in fruits, vegetables and spices such as grapes (8), apples (9), lychee (10) starfruit (11), mango (12), tomato (13), saffron (14) cured tobacco

(15) and black tea (16). During the ripening of red raspberries, ac-ionone and 3-ionone increased to produce the characteristic raspberry aroma (85). In heated apple juice,



26






27


1-damascenone contributed 32% of the total aroma potency of the juice, and only 1.6% of the total aroma potency of fresh (unheated) apple juice (9). Safranal is a potent aroma in saffron formed during drying and storage by hydrolysis from picrocrocin, a monoterpene glycoside (86). Beta-cyclocitral, 1-ionone and 1-damascenone were detected in fresh tomato. Only 1-ionone and 1-damascenone are the important to tomato aroma because their concentrations (4 and 1 gpg/L respectively) are higher than their odor threshold (0.007 and 0.002 gLg/L respectively). Beta-damascenone shows a 10-fold increase in concentration in heated tomato juice which was concentrated to tomato paste

(50). Buttery et al. (41) examined both low carotene and high 0-carotene tomato lines for norisoprenoids. They found that the high 1-carotene line contained the highest concentrations of 1-ionone and 1-cyclocitral. Both norisoprenoids are known biological or chemical degradation products of 0-carotene.

Carotenoids are widely distributed in the plant kingdom and orange juice is

particularly rich and a complex source of these compounds (87). Lutein, zeaxanthin, P-crytoxanthin, a-carotene and p-carotene have been determined in unsaponified orange juice carotenoids extracted by ethyl acetate (73). Thirtynine carotenoid pigments were separated and identified in saponified orange juice carotenoids using HPLC (68). Among these, 1-carotene, a -carotene, neoxanthin P-crytoxanthin, lutein, violaxanthin, and canthaxanthine have the structural potential to form potent norisoprenoid fragments (18-22). These carotenoids have been confirmed to be present in the Valencia juice used in this study based upon HPLC retention time and spectral characteristics data. (See Chapter 3)





28


Objectives

Since orange juice has so many carotenoids that could serve as precursors for a wide range of norisoprenoids, the objective of this study was to determine if more than one aroma active norisoprenoid was present in fresh or heat-treated orange juice. If additional norisoprenoids are found, they should be characterized and identified. (See Objective #2)

Materials and Methods

Orange Juice Samples and Processing

Late-season Valencia oranges (from Haines City Citrus Growers Association,

Haines City Florida) were juiced using an FMC juice extractor at the Citrus Research and Education Center (CREC), Lake Alfred, Florida. The oranges were juiced using a commercial FMC juice extractor model 291 with standard juice settings. An FMC model 035 juice finisher (FMC Corp., Lakeland, FL) was used with a 0.02 inch screen. The finished juice had a Brix value of 11.70, an acid content of 0.67% citric acid, a Brix/acid ratio of 17.5 and an oil level of 0.0196%. The freshly squeezed juice was divided into three groups. In Group 1, fresh orange juice was immediately chilled and NaCl (36 g/100 mL of juice) was added to inhibit enzymatic reactions. In Group 2, fresh orange juice was pasteurized using UHT/HTST lab Microthermics tubular pasteurizer Model 25 (Microthermic Corp., Raleigh, NC) at 1950F (90.5oC), held for 12 seconds and filled at 410F (50C). The oil level of the pasteurized juice was 0.0168%. In Group 3, fresh orange juice was concentrated to 650 Brix using a thermally accelerated short-time evaporator (TASTE) built by Cook Machinery, Dunedin, Florida. The concentrate was then reconstituted to 11.730 Brix by diluting with deionized water, but without restoring volatiles.






29


Chemicals

Standard aroma compounds were obtained from the following sources: methional, ethyl 2-methylpropanoate, ethyl 2-methylbutyrate, 1-octanol, 2-acetyl-2-thiazoline, (E,Z)-2,6-nonadienal, (E)-2-nonenal, (E,E)-2,4-decadienal, 1-octen-3-one, ethyl hexanoate, (E,E)-2,4-nonadienal, L-carvone, E-2 octenal, terpinolene, a-terpinyl acetate, a-terpineol, Z-4-decenal, neral, geranial, 4,5-epoxy-E-2-decenal, P-ionone and P-cyclocitral were purchased from Aldrich (Milwaukee, WI). Octanal, limonene, linalool, nonanal, hexanal, decanal, dodecanal, 1,8 cineole, citronella, terpinen-4-ol, 1-sinensal, P-myrcene, nootkatone ethyl butyrate, acetaldehyde, geraniol, and nerol were obtained as gifts from SunPure (Lakeland, FL). Alpha-ionone was obtained as a gift from Danisco (Lakeland, FL). The (Z)-2-nonenal was found in purchase of (E)-2nonenal at the 5-10% level. The (E,Z)-2,4-nonadienal and (E,Z)-2,4-decadienal were found in the purchase of (E,E)-2,4-nonadienal and (E,E)-2,4-decadienal respectively. Their identities were confirmed by mass spectra, retention indices and odor qualities. Beta-damascenone and p-1-Menthen-8-thiol were obtained from Givaudan (Lakeland, FL). The 4-mercapto-4-methyl-2-pentanone and 4-mercapto-4-methyl-2-pentanol were

synthesized in our laboratory. The 3-mercaptohexan-1-ol was bought from Interchim (Montlucon, France).

Orange Juice Headspace Extraction

A 10 mL aliquot of orange juice was added to a 40 mL glass vial containing a

micro stirring bar and sealed with a screw-top cap that contained a Teflon-coated septa. The bottle and contents were placed in a combination water bath and stirring plate set at 400C, and gently stirred. After equilibrating for 45 min a SPME fiber (50/30 mm






30


DVB/Carboxen/PDMS on a 2 cm StableFlex fiber, Supelco, Bellefonte, PA) was inserted into the headspace of the sample bottle and exposed for 45 min. The fiber was then removed from the headspace and inserted into the heated GC injector port at 2200C where the volatiles were thermally desorbed for 5 min. Gas Chromatography: GC-FID and GC-Olfactometer

Separation was accomplished with a HP-5890 GC (Palo Alto, CA ) using either a DB-wax column (30 m x 0.32 mm. i.d. x 0.5 mm, J&W Scientific; Folsom, CA) or Zebron ZB-5 column (30 m x 0.32 mm. i.d. x 0.5 mm, Phenomenex, Torrance, CA). Column oven temperature (for DB-wax) was programmed from 40 to 2400C (or 40 to 2650C for ZB-5) at 70 C/min with a 5 min hold. Helium was used as carrier gas at flow rate of 1.55 mlimin. Injector and detector temperature were 2200C and 2900C, respectively. A narrow-diameter injection port liner (0.75 mm.) was used to improve peak shape and chromatographic efficiency for SPME thermal desorption. The entire separation was conducted in the splitless mode. A GC splitter (Gerstel, Baltimore, MD) split the column effluent between the FID and olfactometer (equipped with a highvolume sniffing port, DATU, Geneva, NY) in a 1:2 ratio, respectively as described by Bazemore et al. (6). A time-intensity approach was used to evaluate odor quality and intensity at the sniffing port during the GC run. Assessors rated aroma intensity continuously throughout the chromatographic separation process using a linear potentiometer that supplied a continuous signal to the chromatographic software. Retention times and verbal descriptors were recorded to permit aroma descriptors to be coupled with computerized aroma time-intensity plots. Two olfactometry panelists were trained in GC-sniffing with standard solution of 11 compounds typically found in orange juice (ethyl butanoate, cis-3-hexenol, tran-2-hexenal, ca-pinene, myrcene, linalool,






31


13-citronellol, carvone, terpin-4-ol, geranial, and neral). The panelists sniffed the effluent of aroma standard from GC-O with optimum positioning and breathing technique. The intensity of each standard was recorded on a sliding scale (varying from none to strong intensity) and panelists were provided verbal descriptors of aroma quality. For additional experience, the extract of aroma volatiles from commercial orange juice was provided to panelists under identical conditions. Panelists were accepted on they demonstrated an ability to replicate aroma peak times for at least 80 % of the components in the test mixture.

Two trained panalists evaluated the volatiles of orange juice (extracted by SPME) in duplicate, thus producing four individual time-intensity aromagrams. Average intensity from the four runs was calculated for each odorant. If no peak was detected in a run, its value was treated as missing, not zero. An indication of aroma activity with similar aroma descriptors, at the same retention was required from at least half the panel results before a peak could be considered aroma-active. Averaged time-intensity aromagrams were constructed by plotting average intensity versus retention time. Chromatograms and aromagrams were recorded and integrated using Chromperfect version 5.0, Justice Laboratory Software (Palo Alto, CA). Identification of the aromaactive components was based on the combination of sensory descriptors, standardized retention indices, and identification confirmed by comparison with standards and GC-MS spectra.

Gas Chromatography-Mass Spectrometry

Orange juice headspace volatiles were extracted by SPME and introduced to the GC-MS. Volatiles were separated and analyzed using a Finnigan GCQ ion trap mass






32


spectrometer (Finnigan, Palo Alto, CA) equipped with a DB5, 60M x 0.25 mm I.D., capillary column (J&W Scientific, Folsom, CA). The injector temperature and transfer line temperature were 200 and 250 oC, respectively. Helium was used as the carrier gas at 1 ml/min. The oven temperature program consisted of a single thermal gradient from 40 to 275 oC at 7oC/min. The MS was set to scan from mass 40 to 300 at 2.0 scans/s in the positive ion electron impact mode. The ionization energy was set at 70 eV. Aroma Peak Identification

Initial identification was based on the combination of matches with standardized alkane retention index values (Kovat's Index) using two dissimilar column materials (e.g., DB-wax and ZB-5) and aroma characteristics. If the aroma component was sufficiently concentrated, fragmentation patterns were compared with library spectra (NIST 2002 and Wiley (6th Edition) databases using the spectral fit criterion. Only those compounds with spectral fit values equal to or greater than 800 were considered as possible identification candidates. Whenever standards could be obtained, they were used as a confirmation of identification, by comparing the resulting fragmentation pattern, retention index value and aroma descriptor (88).

Results and Discussion

Extraction and Concentration of Juice Norisoprenoids

Solid phase microextraction (SPME) was used to extract and concentrate orange juice volatiles because it is a rapid, solventless headspace sampling technique (6). When solvent extraction was used, early eluting peaks were obscured by the large quantities of solvent. Early eluting aroma peaks such as acetaldehyde have been shown to be important in orange juice flavor (89) but could not be examined using GC-O in solventextracted samples. Although solvent extraction would not have presented a problem in






33


determining norisoprenoids, as they elute fairly late, one of the secondary objectives in the overall study was to determine the relative contribution of norisoprenoids to the total aroma of orange juice. Solvent extracted juice samples would have been unsatisfactory for this purpose for the above stated reason.

The application of headspace SPME to flavor volatile compounds has been

employed in the study of flavor volatiles in tomato and strawberry fruits using PDMS, PDMS/DVB, and Carbowax/DVB coated fiber (67), in orange juice using a PDMS coated fiber (64), a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65), PDMS and polyacrylate fiber (66). The partition coefficients of the polymeric coatings for the analyses differed markedly. For example, terpenes such as a-pinene, P-myrcene, y-terpinenes, and limonene are all nonpolar, and were extracted to a higher degree into the nonpolar PDMS coating (66). Corresponding PDMS extracted the least amount of the more highly polar volatiles, PDMS/DVB and Carbowax/DVB had partition coefficients higher than that of PDMS for the most polar molecules (67). The CarboxenPDMS fiber coating was more selective for terpenes than early eluting alcohols and aldehydes (6). Polyacrylate was more effective in extracting highly polar compounds such as methanol and ethanol (66). Due to the wide range of volatile compounds from orange juice and for the increased fiber capacity, the headspace volatiles in this study were extracted and concentrated using the 50/30 mm DVB/Carboxen/PDMS coating on a

2 cm StableFlex fiber.

In examining adsorption curves for 1-cyclocitral, 3-damascenone and a- and

0- ionone on the chosen fiber (see Fig. 5-1) it was concluded that 45 min. represents a rough compromise for all four analytes between minimal exposure time and maximum





34


peak area. For example 0-damascenone and 1-ionone reaching more than 80% of their final equilibrium value within 45 min. It is a rare SPME analysis that employs true equilibrium exposure time. If exposure time can be carefully controlled, then exposure times of as little as 5 min. can be employed. These very short exposure times are usually limited to analytes in relatively high concentration and even then the reproducibility is not that good. In this study, very short exposure times were not an option as the analytes of interest were present in very low concentration.










SIA





-t 13 14
S 1 620
5 9 1617
2 4 7 12 1819
W 3 8 210 23 24

12 14 16 18 20 Time (min)

Figure 4-1. GC-FID (top) and average time-intensity of four GC-O runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and
23 correspond to norisoprenoids, all numbers refers to compounds in Table
4-1





35


GC-Olfactometry

In this study, a total of 59 aroma active components were detected in SPME

headspace samples from fresh orange juice (orange juice group 1) Since the primary goal of this study was to determine if additional aroma active norisoprenoids were present in orange juice, GC-O was employed primarily in the region where B-ionone and other norisoprenoid standards eluted. Using standards of 3-cyclocitral, a-ionone, p-ionone and 1-damascenone, the retention time region was established between 12 and 20 min, and the resulting aromagram and concurrent chromatogram is shown in Fig. 4-1.

As noted in Fig. 4-1, four aroma peaks corresponding to peaks 5, 19, 21 and 23

were observed at the identical retention times as 3-cyclocitral, 13-damescenone, a-ionone and 0-ionone respectively. It is also apparent from the relative intensities shown in Table 3-1, that these potential norisoprenoid peaks were among the more intense aromas. Beta-ionone was the most intense and -cyclocitral was the weakest aroma compound of all the four potential norisoprenoids observed. When the samples were rerun on a DBwax column the four aroma peaks also were found at retention index values that corresponded with the four potential norisoprenoids. Furthermore, the aroma quality of each juice norisoprenoid corresponded exactly with the aroma description of standards. Since these compounds have the same retention behavior on two very dissimilar chromatographic columns and also have the same aroma quality as standards, they are probably 1-cyclocitral, 0-damescenone, a-ionone and 0-ionone respectively. This represents the first time that 1-cyclocitral, and a-ionone have been reported in orange juice. Beta-damescenone had recently been reported in heated orange juice but its identity was not confirmed by supporting instrumental methods (65).






36


Table 4-1. Identification, retention characteristics and aroma descriptions of aroma
active compounds in fresh orange juice
Linear retention Relativec
No. Compound Aroma descriptor ZB-5 DB-wax intensity
1 Terpinen-4-olb Metallic, musty 1175 1619 5 2 Z-4-decenala Green, metallic, soapy 1188 1542 7 3 Decanalb Green, soapy 1198 1508 7 4 (E,E)24-nonadienala Fatty, reen 1209 1702 7

6 Nerola Lemongrass 1222 1798 5 7 Neralb Lemongrass 1236 1692 7 8 L-carvoneb Minty 1242 1747 8 9 Unknown Metallic/ woody 1247 6 10 Geraniola Citrus, geranium 1265 1853 9 11 Unknown Soapy, almond 1274 7 12 1-p-menthene-8-thiola Grapefruit 1281 1619 7 13 (E,Z)-2,4-decadienala Metallic, geranium 1293 1759 4 14 Geraniala Green, minty 1310 1742 4 15 (E,E)-2,4-decadienala Fatty, green 1314 1819 4 16 a-teroinvl-acetatea Sweet 1349 1663 6 17 4,5-epoxy-E-2-decenala Metallic, fatty 1375 2010 6 18 Unknown Sweet nutty 1380 7

20 Dodecanala Soapy 1403 1722 5

22 Unknowna Fermented, rancid butter 1459 5

24 Unknown Nutty 1510 8
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard
b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard, and MS
c Averages of normalized intensities (10) evaluated by two trained panelists in four replications

Mass Spectrometry Norisoprenoid Identifications

Headspace volatiles from fresh orange juice were analyzed using capillary GC with

an ion trap mass spectrometer. To achieve greater selectivity for the norisoprenoids of

interest, selected ion chromatograms were reconstructed in the retention region where

norisoprenoid standards were found to elute. The selectivity achieved is demonstrated in

Fig. 4-2. Specific m/z values were evaluated to provide the best peak height for each

norisoprenoid of interest as well as minimizing interference from non-norisoprenoid





37


components as well as noise. The following ions were monitored for the specific norisoprenoids: 1-cyclocitral, m/z = 137 and 152; 1-damascenone, m/z = 175 and 190; a-ionone, m/z = 177 and 192; 1-ionone, m/z = 177 and 192.

Although only a single ion has been shown for each norisoprenoid, two or more selective ions were employed to detect the presence of specific norisoprenoids. For example, the selected ion chromatogram using m/z 137 was more intense than that from m/z = 152 but not as specific for 1-cyclocitral. The selected ions of m/z = 177 and 192 were extracted for the determination of c-ionone and 1 -ionone. Selected ion chromatograms at m/z 177 provided excellent signal strength and selectivity for 1 ionone. The a-ionone was obviously present at much lower concentrations than Bionone. The SIC chromatogram at m/z = 192 (Fig. 4-2) provides more selectivity for aionone but better signal and noise ratio was obtained at m/z 177.

1&
10 17.6

~20.06 21.-4 1TIC
20L 18-3L IMI' ia m A 20.4ft$ Ya LiM JJI.
0 1 .68
C A
60 m/z =137
S 20 .12 17.97 iR4 1&8. 19.31 19.702f01 20.56 21.2921.49 22.08
0.94

60 m/z = 177
20 16.29 16.Z517.0017.3717.6 18.09 18.69 19.2 19A7 19.70 20.08 20.47 20.87 A., 21

mC /z= 192
2073 17 os8344 o19619331 57 2014 21 21
16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0
Time (min)

Figure 4-2. Comparison between total ion chromatogram and selected ion
chromatograms (SIC) A: 1-cyclocitral, B: 3-ionone, C: a-ionone.






38





TIC
. 10.06

1 2II


20
260
2420






A2 om/z = 90
200 1 80
S 140

20 260






chromatograms using SIM at m/z 190. Middle chroatogram 0-damascenone
200
180 10.06
160
140 20
8.50 9.b0 9.50 10:00 105b 11700 1i:5 12:00 Time (min)

Figure 4-3. Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190. Middle chromatogram 13-damascenone
detected from orange juice, and lower overlay chromatogram of spiked (A)
and non-spiked (B) of orange juice with standard P-damascenone.

The 0-damascenone, selected ion chromatograms using m/z = 175 and 190 (two masses highly characteristic for 0-damascenone) did not provide a clear signal at the expected retention time of 1-damascenone using the ion trap MS. Beta-damascenone had been detected by GC-O at the expected retention time with the characteristic aroma but not detected by either FID or SIC ion-trap MS, suggesting that B-damascenone, if present, was there at very low levels. Beta-damascenone has an extremely low odor threshold, which is below the detection limits of most instrumental detectors (0.002 gg/L). However, by employing quadrupole mass spectrometer in the single ion





39


monitoring mode at least a 10x greater sensitivity (lower detection level) can be achieved because all the ions of a single mass are continuously measured rather than measured for an instant before monitoring other masses. Using selected ion monitoring m/z values 175 and 190, -damascenone was detected at the expected retention time (see Fig. 4-3 for the case at m/z 190). The combined GC-O and two SIM peaks at the exact retention time of 0-damascenone, confirm its presence in orange juice.

Although selected ion chromatograms strongly suggest the presence of the other norisoprenoids of interest, they do not offer absolute proof. They only indicate that a juice volatile elutes at the identical retention time as the norisoprenoid of interest, and this volatile contains the same mass fragment. The combination of this information with the GC-O information provides three independent pieces of information strongly suggesting the presence of specific norisoprenoids. However, to absolutely confirm the presence of O-cyclocitral, ac-ionone, and 0-ionone, their spectra from the juice MS data at the retention times of each respective norisoprenoid was obtained and compared with reference spectra in standard libraries or compared with that obtained from authentic standards. The resulting match for the case of 0-cyclocitral is shown in Fig. 4-4. It is readily apparent that although the relative ion abundances are not the same (usually a function of instrument to instrument variation) an excellent spectral match has occurred and that the presence of 1-cyclocitral in orange juice is confirmed. Comparing the relative abundances of ions m/z 137 vs 152 in the upper spectrum of Fig. 4-4, it can be readily appreciated why examining the selective ion chromatogram at 137 provided a better signal to noise ratio than the chromatogram at 152, the molecular ion for 13cyclocitral.








40



41
100
900

so
70- Exact Mass =152
Molecular Formula =C10H160 60- 137
50 79
S401
30 67 3 109 123

10 94 95
10 42 65( 107 11 24 152

100 137
90


70- 3? 67 81 123 152
1109

SO 55
40
30 7 91


20 I I I I51
10 IL I I i12 11196 1191241 53

40 60 so0 100 120 140 160 IBO


m/z




Figure 4-4. Upper spectra from orange juice MS at RT = 17.68 bottom spectra of 3cyclocitral from database NIST 2002.


1 77
100 177
90 0
so


60
so Exact Mass =192
4o Molecular Formula =C13H200
30

20 5lOS 133
91 189 S977 204 10 65 67 14 161 I100




1 70





613 179 13
40




1- 6 77 9 1 0 1 7 1 162

40 60 0 1 00 120 140 160 I O 200



m/z



Figure 4-5. Upper spectra from orange juice MS at RT = 21.94, bottom spectra from

standard 0-ionone using identical ion trap MS at identical retention time.








41




However, a-ionone has not been previously reported and is shown in Figure 4-6.



The spectral match in this case is good considering the very low levels of a-ionone


present, but not perfect. Even with careful background subtraction (which was done for


all the previous spectra as well), there will be a fair amount of extraneous peaks simply


due to random noise. However, the major fragment ions of m/z 192 (M), 177, 163, 136,


121, 109, 93, 91, and 77 are all present, more than enough to confirm the presence of aionone in orange juice.





100 1
100 93 0
90
91 177 70
60 Exact Mass=192
s50 Molecular Formula =C13H200
40 192 109
277 0-136
20 43 67 22 147
10 A S5 11 1 _6 186 207
0 1 o , I ,, ,, i I, ,l l L .. .. 1 1J, 1 1
100o
90


70- 91
60


40- 177
7
30
20- 109
10- 51 65 1 10 1 119 136 1 9192

40 60 80 100 120 140 160 180 200


m/z


Figure 4-6. Upper spectra from orange juice MS at RT = 20.87, bottom spectra from

standard a-ionone using identical ion trap MS at identical retention time.




Conclusion


Four norisoprenoids in fresh orange juice (0-cyclocitral, 0-damascenone, a-ionone,



and 03-ionone) have been conclusively identified through the combined information from





42


GC-O retention index matches with standards on two dissimilar chromatographic column materials, aroma descriptor matches and GC-MS matches of both retention time and fragmentation spectra. Of these four norisoprenoids, 0-ionone had been reported in two previous orange juice GC-O studies (4, 5). There is one previous mention of 3damascenone in heated juice, but no MS or independent instrumental confirmation data was presented (65). 3-cyclocitral and a-ionone were detected in orange juice for the first time in this study and confirming MS data for 3-damascenone was presented for the first time.













CHAPTER 5
QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT OF NORISOPRENOIDS IN ORANGE JUICE Introduction

Chromatographic data is often used to determine the relative concentrations of

components in a volatile mixture. Within the linear range of the detector, integrated peak area is proportional to the amount of that component in the sample. The four techniques commonly employed to quantify chromatographic components are normalization; internal standards; external standards; and standard addition methods (90). Only a few of these have been employed to determine the amounts of specific volatiles in orange juice volatiles to better understand their contribution to orange flavor (2, 5). Buettner and Schieberle (5) employed stable isotope dilution assay to quantify 25 volatiles from a solvent extract of hand-squeezed Valencia orange juice. The juice was spiked with a known amount of the labeled internal standard and the juice was extracted with diethyl ether and subsequently analyzed by GC-MS. Standard curves of the labeled and unlabeled reference odorants were used to establish a relationship between peak area and concentration. Odor activity values (OAV, concentrations of the odorants divided by their odor threshold) were determined to estimate their respective odor contributions. The highest OAVs were calculated for (s)-ethyl 2-methylbutanoate, ethyl butanoate,

(Z)-3-hexenal, ethyl 2-methylpropanoate, acetaldehyde, and (R)-limonene.

Moshonas and Shaw (2) quantified the volatiles from orange juice using dynamic headspace GC with a pressurized purge and trap apparatus. Concentrations for each



43






44


volatile were calculated using the standard addition procedure. Regression equations were developed from peak area data from four different concentrations of each compound added to a juice base. Odor activity values were calculated for each component measured (although they were not identified as OAV values). Compounds which exceeded their threshold by the greatest amounts (highest OAV values) and thus most likely to contribute to fresh orange flavor included: limonene, myrcene, a-pinene, decanal, octanal, ethyl butanoate, and linalool. The differences between these two studies which both claim to determine the components most responsible for fresh orange juice flavor are worth noting.

SPME is a rapid, solventless static headspace procedure. It can be used for the

quantitative analysis of flavor and fragrance compounds. The standard addition method has been used primarily because the concentration in the headspace (volatility) will be influenced by the sample matrix (66, 91, 92). Boa et al. (92) reported that reliability problems of headspace SPME quantification is associated with the matrix and could be reduced by employing the standard addition method or employing isotopically labeled internal standards. Headspace SPME with standard additions were used in the present study because SPME can extract and concentrate orange juice headspace volatiles which transfer them directly into the injector of a GC in a simple, straightforward manner. Just as important for this study, the nonvolatile carotenoids will not be extracted. If the nonvolatile carotenoids were present they might degrade when exposed to the heat (2000C) of GC injection port and possibly produce artifact norisoprenoids. The major problem with the standard addition approach is that several injections at each standard addition level are required in order to obtain a single result. Thus, depending on the





45


number of levels and the number of injections per level, this procedure can be time consuming. However, once a pseudocalibration curve produced, the calculated slope can be used for other samples of similar matrix. Thus, it is not essential that the standard addition be employed for each and every sample, but the slope of the pseudocalibration curve should be checked from time to time. All solutions analyzed must fall within the linear range of the detector response (90).

Objectives

The primary goal of this study was to quantify the norisoprenoids in orange juice using static headspace SPME with standard additions and GC-MS. (Objective 4) A secondary goal was to determine the relative aroma contribution of all four norisoprenoids to the total orange juice aroma. (Objective 3) Materials and Methods

Quantification of Norisoprenoids in Orange Juice

The adsorption (amount vs. exposure time) curves for the SPME fiber (50/30 mm DVB/Carboxen/PDMS) employed in this study was determined by varying exposure time from 5 to 150 minutes. Since native concentrations were so low, orange juice samples were fortified with 8.6, 4, 5.4, and 5.27 ppm 0-cyclocitral, P-damascenone, wa-ionone, and 1-ionone respectively so that adsorption characteristics could be more accurately determined. Ten milliliters of the fortified juice were transferred to 40 ml vial with screw cap coated with Teflon. After 45 min at 40oC, the headspace volatiles were extracted using SPME (as described in Chapter 4)

Headspace SPME and GC-MS were used to quantify norisoprenoids in orange juice using the standard addition method. Each standard (B-cyclocitral, 8-damascenone,





46


c-ionone, and 1-ionone) was added separately to the orange juice sample to obtain the final concentration of each norisoprenoid from 0 to 2 ppm. Beta-damascenone was the only exception; its added concentrations ranged from 0 to 0.02 ppm. Sampling was accomplished by adding a 10 mL aliquot of the juice to a 40 ml glass vial containing a micro-stirring bar sealed and a Teflon coated septa. Samples were equilibrated at 400C for 45 minutes and gently stirred before a SPME fiber was inserted into the headspace of the sample bottle and exposed for another 45 min. The fiber was then removed from the headspace and inserted into the GC-MS. GC conditions were the same as in Chapter 4. Each sample was prepared and injected at lease twice. Quantitative measurements were made using integrated peak areas from selected ion chromatograms. The ions chosen to reconstruct these single ion chromatograms were at m/z 137, 177, 177, and 190 and were fairly unique for -cyclocitral, p-ionone, ca-ionone, and P-damascenone respectively. The latter m/z values corresponded to the respective molecular ion of 1-damascenone.

In order to quantify the low levels of -damascenone, a quadrupole MS (Agilent 5973 Network Mass Selective Detector, Agilent Technologies, CA) was employed using selected ion monitoring (SIM) mode at m/z 190. It was equipped with HP an Innowax 30 m x 0.25 ipm x 0.25 lm capillary column (Agilent/J&W HP Innowax, Scientific Instrument Services, Inc., NJ) and autosampler (Gerstel Multi Purpose Sampler MPS2, Gerstel Inc., MD). The oven temperature program consisted of two ramps from 90 to 1600C at 60C/min and from 1600C to 2500C at 1200C/min (in order to shorter the GC running time after the 1-damascenone was eluted). Each sample was analyzed from the response at m/z 190. A graph of SIM 190 peak area versus concentration was prepared






47


and the amount of 13-damascenone in the sample determined from the regression line equation.

Determination of the Relative Impact of Norisoprenoids in Orange Juice

The aroma active compounds from 3 types of orange juice (fresh, pasteurized, and reconstituted from concentrate) were separated and identified using GC-O (chapter 4). Intensities of aroma active compounds of each run were normalized so the highest intensity was given a score of 10. The normalized intensities of all the runs were then averaged, providing a similar aroma activity was detected at least half the time at that retention time. If the compound was not detected in one run its value was treated as missing, not zero. Aroma-active compounds from the entire GC-O trial were categorized into eight groups based on similar aroma description. These eight groups were 1) citrusy/minty; 2) metallic/mushroom/geranium; 3) roasted/cooked/meaty/spice; 4) fatty/soapy/green; 5) sulfury/solventy/medicine; 6) floral; 7) sweet/fruity; and 8) green/grassy. The sums of the total olfactometry intensities for each aroma group was determined and presented in spider web (radar graph) for each of the four juice types. The contribution of norisoprenoids to orange juice was calculated from the total intensity of norisoprenoids to the total intensity of all aroma active volatiles in the juice.

Results and Discussion

Quantification of Norisoprenoids in Orange Juice

The amount of volatile compounds found on the SPME coating depends on exposure time, temperature, sample volume, headspace volume, and sample concentration. In this study only exposure time was varied in order to determine the time needed for equilibrium concentrations for each analyte to be established. All other factors remained constant. Equilibrium time between SPME fiber and headspace of






48


fortified juices was indicated when little to no increase in peak area was observed with additional exposure time. The equilibrium time for -cyclocitral, 1-damascenone, a-ionone, and 1-ionone were 75, 90, 115, and 120 min. respectively (Fig. 5-1). The

100000


80000

60000


40000


20000


0 20 40 60 80 100 120 140 160 Exposure time (min)

Figure 5-1. Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 400C, = 3-cyclocitral, a = 3-damascenone, A= a-ionone,
= 3-ionone.

results show that the time needed to reach equilibrium depends on the polarity and the relative molecular mass of each norisoprenoid. Since 75-120 minutes to reach the equilibrium would be too long to wait for practical purposes and may alter the volatile profiles from thermally induced reactions, a shorter exposure time was chosen for routine analyses. It can be seen from the adsorption curves for each compound that 45 min. represents a rough compromise for all four analytes between minimal exposure time and maximum peak area. For example P-damascenone and 1-ionone reaching more than 80% of their final equilibrium value within 45 min. It is a rare SPME analysis that employs true equilibrium exposure time. If exposure time can be carefully controlled, then exposure times of as little as 5 min. can be employed. These very short exposure





49


times are usually limited to analytes in relatively high concentration and even then the reproducibility is not that good. In this study, very short exposure times were not an option as the analytes of interest were present in very low concentrations.

The reproducibility (analytical precision) of a fortified juice using SPME-GC-FID was determined in five replicates at 40'C with 45 min exposure. The relative standard deviations (RSD) obtained were 1.7, 1.7, 0.4, 1.4 % for P-cyclocitral, P-damascenone, aionone, and P-ionone respectively (Table 5-1). It should be kept in mind that the orange juice had been fortified with 8.6, 4, 5.4, and 5.27 ppm 3-cyclocitral, P-damascenone, aionone, and -ionone respectively. The low RSD indicated that the SPME and GC analytical conditions employed in this study could quantify norisoprenoids in orange Table 5-1. Reproducibility of SPME exposure time 45 min at 400C
Replicate P-cyclocitral P-damascenone a-ionone P-ionone
1 84101 27181 46618 23072 2 84149 26797 46715 22861 3 84129 26534 46813 22589 4 81092 26382 46570 22337 5 82037 25961 46319 22335 Average 83101 26571 46607 22639 STDVa 1443 456 186 325 RSDb 1.7 1.7 0.4 1.4 a Standard deviation, b relative standard deviation juice in a highly reproducible manner. However, it should be pointed out the concentrations used to fortify the sample were considerably higher than would ever be found in an orange juice sample. Typical juice concentrations are 50 to 1000 times lower so that typical RSD's for unfortified juice samples range from 20- 50% which might seem high, but still very acceptable for analyses at the sub 9lg/L level the complex matrix of orange juice. The volatility of flavor compounds can be changed according to the






50


sample matrices. Boa et al.(92) reported that the combination of SPME with the standard addition method reduce the problem of matrix effects and improved the precision of the procedure.

Norisoprenoid Quantification using Standard Additions

The norisoprenoids -cyclocitral, 1-damascenone, a-ionone, and 1-ionone were

quantified in orange juice using the standard addition method (Fig. 5-2, 5-3, 5-4, 5-5, and 5-6). The integrate peak areas at specific m/z 137, 177, 177 and 190 for 3-cyclocitral,

800000
y = 373321.26x + 54248.27 R2= 0.86
600000
2 500000
400000
300000 200000 100000
0
0 0.5 1 1.5 2 0-cyclocitral concentration (ppm)

Figure 5-2. Standard addition data for 0-cyclocitral peak area vs. added concentration in
fresh orange juice. Regression line calculated from peak area at selected mass
137.

1-ionone, a-ionone, and -damascenone respectively were plotted versus the concentration of the spiked standards. The amount of each norisoprenoid (Table 5-2 and Table 5-3) was calculated from the regression equation where the calculated value was determined at y = 0.

As seen from the plots of Fig. 5-2 through Fig. 5-6, the correlation coefficients for the standard addition data was at least 0.99 in all cases except for 1-cyclocitral (Fig. 5-2)






51


where it was 0.86. One way analysis of variance (ANOVA) show that there are a highly significant differences (P_0.01) among orange juice spiked with different concentration of standard a-ionone (Fig. 5-3) and 0-ionone (Fig. 5-4). However there were no significant differences within sample. In contrast for -cyclocitral (Fig. 5-2), there were no significant differences between the two samples spiked with standard 1-cyclocitral at concentration 0.54 and 1.1 ppm. This suggests that an error occurred during analysis with at least one data point.

A quadrupole MS provides at least 10x greater sensitivity in the SIM mode than an ion trap MS under the same conditions and was thus used to quantify 0-damascenone in fresh and pasteurized juice when the ion trap failed to detect this compound. The calculated slope from the fresh juice data was also employed to determine the concentration of 0-damascenone in pasteurized juice as it was thought the matrix effects would be the same for both samples.

350000
y = 257750x + 11996
300000 R2= 0.99

250000

S200000

150000 100000 50000


0 0.2 0.4 0.6 0.8 1 1.2
a-ionone concentration (ppm)
Figure 5-3. Standard addition data for ca-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. a-ionone concentration.






52



1200000
y = 1009203.69x + 83575.55 1000000 = .

800000

a 600000400000 200000


0 0.2 0.4 0.6 0.8 1 1.2 0-ionone concentration (ppm)

Figure 5-4. Standard addition data for P-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. 0-ionone concentration




350000y = 14498114.63x + 1229.61 300000- R2= 0.99

250000-t 200000150000100000500000 I
0 0.005 0.01 0.015 0.02 0.025 0-damascenone concentration (ppm)

Figure 5-5. Standard addition -damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190.






53


500000450000- y = 19924780.33x + 16956.31
400000 R2= 0.96
350000300000S250000200000150000100000 50000
0
0 0.005 0.01 0.015 0.02 0.025 1-damascenone concentration (ppm)

Figure 5-6. Standard addition data of 1-damascenone peak area vs. added concentration in reconstituted from concentrate orange juice. GC- quadrupole mass
spectrometer in SIM mode at m/z 190.

The concentration of 1-damascenone in reconstituted from concentrate was calculated from separate standard addition data (Fig. 5-6) as it was thought that the matrix would be substantially different.

Table 5-2. Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique
Norisoprenoids Concentration (ptg/L) Threshold (pg/L in water)a OAV 3-cyclocitral 145 5 25 0-damascenone 0.09 0.002 45 c-ionone 47 0.4 118 0-ionone 83 0.007 11857
a Buttery and Teranishi (50)


Table 5-3. Concentration of -damascenone in fresh, pasteurized and reconstituted
concentrate
orange juice concentration (g/L) OAV
Fresh 0.09 45 Pasteurized 0.18 90 Reconstituted 0.85 425





54


The calculated concentrations of -cyclocitral, 0-damascenone, a-ionone and

p-ionone in fresh orange juice were 145, 0.09, 47, and 83 Rlg/L respectively (Table 5-2). The aroma active compounds in orange juice have been studied by GC-O methods (4-7). Only 1-ionone was reported (4, 5) but has not been reported the concentration of this volatile. The concentration of 1-damascenone in 3 types of orange juice: fresh, pasteurized, reconstituted from concentrate were 0.09, 0.18, and 0.85 gg/L respectively (Table 5-3). This data suggest that there is precursors present in juice and generate 3damascenone during thermal processing. These precursors are probably carotenoids like neoxanthin, but could also be glycosided forms of 1-damascenone. These precursors can generate aroma volatiles in foods that have undergone thermal processing as reported for tomato paste (23) and heated apple juice (9). It has been previously reported that citrus juice pulp and cloud (insoluble solids) can retain considerable volatiles (93, 94). Therefore 0-damascenone may have been trapped in the pulp during thermal concentration and might not be completely removed during evaporation. Its partial loss may also been partially compensated by newly P-damascenone generated from thermally unstable carotenoids during thermal concentration. Determination of Relative Aroma Impact of Norisoprenoids

The odor activity value (OAV) is a rough way of determining relative aroma contribution of various substances. It is determined by dividing the analytical concentration by the aroma threshold. The OAV of 0-cyclocitral, 0-damascenone, a-ionone and P-ionone were 25, 45, 118, and 11857 respectively (Table 5-2). The OAV value shows that P-ionone is predicted to have the greatest contribution compared to the other norisoprenoids. Hinterholzer and Schieberle, (4) analyzed the volatiles from orange






55


juice by solvent extraction and determined the aroma contribution by aroma extract dilution analysis (AEDA). The value from AEDA was recorded as flavor dilution (FD) factor (the highest dilution factor of the particular aroma active compounds which can be perceived by human nose). The most odor active compound by this method was ethyl butanoate (FD 1024) but the FD factor of 1-ionone was onlyl6. These same authors (5) quantified twenty-five odor active compounds by stable isotope dilution assay and estimated their respective odor contributions by OAV values. Unfortunately they did not quantify 0-ionone. The OAV of ethyl butanoate (the compound with the highest dilution value from AEDA (4)) was 1192 (concentration 1192 pg/L, odor threshold 1 glg/L). If one compares this OAV value with the OAV of 1-ionone in present study (11857), p-ionone could be the most aroma active compound in orange juice. The apparent conflict in the two sets of data suggests that 0-ionone may not have been well extracted in the AEDA study.



Table 5-4. Aroma active compou nds in orange juice group by citrusy/mint
Compounds Description LRI Relative ZB-5 DB-wax intensity
Unknown Orange peel 963 1,8 cineole Minty, camphor 1026 1232 5c, 6d Nonanal Orange peel, soapy 1090 1398 6c, 6d 3-mercapto hexan-1-ol Grapefruit 1121 7c, 7d Citronellal Minty,camphor 1160 1489 5c, 7d, 5e Nerol Lemomgrass 1222 1798 5c, 5d Neralb Lemomgrass 1236 1692 7c, 7d L-carvoneb Minty 1242 1747 8c, 8d Geraniol Citrusy,geranium 1265 1853 9c, 7d, 4e 1-p-menthene-8-thiol Grapefruit 1281 1619 7c, 7d, 5e Geranial Green,minty 1310 1742 4c,4d, 4e Nootkatoneb Sweet,sour, grapefruit 1824 7c, 5d






56


Table 5-5. Aroma active compounds in orange juice grouped by
metallic/mushroom/geranium
LRI Relative
Compounds Description ZB-5 DB- intensity wax
1-octene-3-one Metallic, mushroom 974 1308 6c, 7d, 5e 3-myrcene Geranium,plastic 979 1163 7c, 7d 8e Octanalb Metallic, orange peel 998 1299 8c, 8d 8e E-2-octenal Metallic, fatty, green 1052 1449 4d Terpinoleneb Metallic, citrusy 1070 1296 6c,5d Unknown Green, metallic 1100 6d Unknown Metallic, pungent 1128 5d Z-2-nonenal Green, metallic 1141 1515 4c, 6d, 5e Terpinen-4-olb Metallic, musty 1175 1619 5, 6d Unknown Metallic, woody 1247 6C, 6d (E,Z)-2,4-decadienal Metallic, geranium 1293 1759 4C, 7d 4,5-epoxy-E-2-decenal Metallic, fatty 1375 2010 6c,6d, 4e Unknown Aquarium, metallic 1589 5c, 6d 13-sinensal Aquarium 1698 2244 8c, 8d


Table 5-6. Aroma active compounds in orange juice grouped by
roasted/cooked/meat y/spice
LRI
Compoundsa Description DB- Relative ZB-5 intensity
wax
Methional Cooked potato 904 1464 8C, 8, 7 2-acetyl-2-thiazoline Cooked jusmine rice 1104 1766 6C, 6d, 7e Unknown Spice 1317 6d Unknown Sweet, nutty 1380 7, 8d Unknown Fermented, rancid 1459 5, 7d Unknown Green, overipe orange 1461 4d, 4e Unknown Nutty 1510 8c, 9d Unknown spice 1718 7C, 7d


Table 5-7. Aroma active compounds in orange juice grou by fatty/soapy green
Compounds Description LRI Relative DB- intensity
wax
Hexanalb Green, fatty 794 1083 7, 6d, 7e 1-octanol Green, soapy 1065 1565 8, 7d, 5e E-2-nonenal Soapy 1153 1542 8c, 10d,6e
(Z)-4-decenal Green, metallic, soapy 1188 1542 7C, 7d Decanalb Green, soapy 1198 1508 7c, 8d, 5e (E,E)-2,4-nonadienal Fatty, green 1209 1702 7c, 7d Unknown Soapy, almond 1274 7C, 7d (E,E)-2,4-decadienal Fatty, green 1314 1819 4C, 6d Dodecanal Soapy 1403 1722 5c, 6d






57


Table 5-8. Aroma active compounds in orange juice grouped by
sulfury/solventy/medicine
LRI
Relative
Compoundsa Description ZB-5 DB- intensity wax
Acetaldehyde Fresh alcohol 445 732 6c, 6d Carbon disulfide Sulfur, fermented cabbage 678 6c, 7d, 6e Dimethyl sulfide Solventy, plastic 691 6c, 6d, 4e Dimethyl disulfide Plastic 772 1074 4c, 4d, 7e Unknown Fermented, sulfur 818 5c, 6d 2-methyl-3-furanthiol Meaty, vitamin, medicine 865 1305 7c, 7d, 6e 4-mercapto-4-methyl-2-pentanone Sulfury, grapefruit 944 1389 7C, 5d Dimethyl trisulfide Sulfur, sweaty 968 1392 3c, 5d,8e 4-mercapto-4-methyl-2-pentanol Sweaty,grapefruit,guava 1039 7c, 7d, 7e Unknown Solventy 1167 5d Dimethyl tetrasulfide Sulfury, musty 1225 6e




Table 5-9. Aroma active compounds in orange juice group by floral LRI Relative
Compoundsa Description Z-5 DB- intensity wax
Linaloolb Floral 1094 1551 8c, ,5e 0-cyclocitralb Mild floral, hay-like 1214 1632 6c, 4d Unknown Tobacco,sweet, floral 1255 6e 3-damascenoneb Tabacco, apple, floral 1383 1829 7C, 8d, 8e a-iononeb Floral 1426 1863 8c, 8d P-iononeb Floral, raspberry 1484 1951 8c, 9d 7e




Table 5-10. Aroma active compounds in orange juice grouped by sweet/fruity LRI Relative
Compoundsa Description Z-5 DB- intensity wax
Ethyl-2-methylpropanoate Sweet, fruity 758 966 6c, 6 Ethyl butyrateb Sweet, fruity 795 1034 4C, 6d 8 Ethyl-2-methylbutyrate Sweet, fruity 846 1051 5c, 6d Ethyl hexanoateb Sweet 994 1242 6c, 7d (-terpinyl acetate Sweet 1349 1663 6c,6d






58


Table 5-11. Aroma active compounds in orange juice grouped by green/grassy
LRI Relative Compoundsa Description ZB-5 DB-wax intensity 3-(Z)-hexen-1-ol / (E)-2-hexenalb Green, grass 854 1226/1391 6c, 7d (E,Z)-2,6-nonadienal Green 1148 1593 7, 8 d 7e
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard, and MS
C fresh orange juice d pasteurized orange juice a reconstituted from concentrate





1. Citrusy/minty 100



60

40


I 3. Roasted/cooked/meaty/spice
7. Sweet fruity




4. Fatty/soapy/green

6. Floral


5. Sulfury/solventy/medicine Figure 5-7. Aroma group profiles of fresh (*), pasteurized (0), and reconstituted from
concentrate (0) orange juice.

Norisoprenoid Contribution to Total Floral Aroma

All four of the identified norisoprenoids in orange juice have a general floral

aroma quality (Table 5-9). When all the relative intensities of these four norisoprenoids were combined (in each type of orange juice) and compared with the total aroma





59


intensity, their contribution to total juice aroma were 7.8, 7.6, and 8.7 % in fresh, pasteurized, and reconstituted from concentrate respectively. Fig. 5-7 illustrates aroma group profiles from the total intensity in similar aroma quality groups from GC-O aromagrams, comparing among fresh, pasteurized, and reconstituted from concentrate. Comparing fresh and pasteurized orange juice, the major aroma category differences were in the metallic/mushroom/geranium and in roasted/cooked/meaty/spice categories which were higher in pasteurized juice than fresh juice. This difference come from three addition aroma active compounds (E-2-octenal, and two unknown at LRI 1100 and LRI 1128) in metallic/mushroom/geranium category and two additional unknown ( LRI 1317 and LRI 1461) in roasted/cooked/meaty /spice category in pasteurized juice (Table 5-5 and Table 5-6). In fresh juice citrusy/minty category was higher than pasteurized juice. The differences were the higher intensity of compounds in citrusy/minty category in fresh juice than in pasteurized juice. The contribution of floral quality was the same in fresh and pasteurized orange juice. From the aroma group profile of reconstituted from concentrate there were many volatile compounds lost due to the thermal evaporation process.

Fig. 5-8 shows the contribution of aroma active compounds in just the floral category. This group includes linalool, P-cyclocitral, P-damascenone, c-ionone, P-ionone and one unknown (LRI 1255) generated after thermal processing. The norisoprenoids in orange juice contribute the majority of floral aroma in floral category, specifically 78, 78, and 59% in fresh, pasteurized and reconstituted from concentrate respectively.






60



40 35

S30
25

20



10

5 0
Fresh Pasteurized Pumpout Orange juice

Figure 5-8. Upper bar norisoprenoids contribute mainly to the total floral category, fresh
= 78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent nonnorisoprenoids including linalool and unknown (LRI = 1255) generated
during thermal processing.



Table 5-12. Norisoprenoids in orange juice and peel oil Reconstituted Hand
Norisoprenoids Fresh Pasteurized Peel oil concentrate squeezed
f-cyclocitral X X X X
-damascenone X X X X X ct-ionone X X X P-ionone X X X X
X = indicates presence of norisoprenoids in the various samples.

Four norisoprenoids, P-cyclocitral, 1-damascenone, ca-ionone, and 0-ionone were detected in both fresh and pasteurized juice (Table 5-12). Only two norisoprenoids 3damascenone and 0-ionone were detected in reconstituted from concentrate, indicating that these two compounds could be generated from precursors during thermal evaporation and/or they were retained by the pulp during the evaporation process.






61


Conclusion

Concentrations of four orange juice norisoprenoids were determined using SPME with the standard addition method. The concentrations of P-cyclocitral, 1-damascenone, a-ionone, and p-ionone in fresh orange juice were 145, 0.09, 47, and 83 ptg/L respectively. The OAV (determined by dividing the analytical concentration by the aroma threshold) of -cyclocitral, P-damascenone, a-ionone, and P-ionone were 25, 45, 118, and 11857 respectively. The OAV values suggest that 3-ionone provides the greatest aroma contribution compared to the other norisoprenoids. The concentration of 1-damascenone increaded with thermal processing, indicating that there are precursors in juice which generate 0-damascenone during elevated temperatures. Combined, the four norisoprenoids contribute 8-10% of the total aroma impact. The norisoprenoids have a general floral character and contribute the majority (60-80%) of the floral character to orange juice.













CHAPTER 6
THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL SOLUTION Introduction

Carotenoids are unstable in both the presence of heat and/or light. The thermal

degradation of carotenoids produces a range of volatile products and norisoprenoids are the most potent aroma compounds of all the volatiles produced. The formation of specific norisoprenoids from the thermal degradation of carotenoids during heat treatment of food products have been reported. Beta-ionone, ct-ionone, and 1-damascenone have been reported in tomato paste (23) and black tea (95). These norisoprenoids also were detected from the thermal degradation of carotenoids in model systems such as 3carotene in water at 970C, 3 hrs (31), 1% solution of 1-carotene heated at 1880F for 72 hrs. (96), and thermal degradation of crystallize 1-carotene at 2400F in a vaccum (97). However the norisoprenoids formed in those model systems were formed at high temperature. At best these studies could be considered accelerated storage studies. In order to have a model more representative of the conditions that a "real world" juice might be exposed to, a 350C storage study was carried out. The model solution was buffered to pH 3.8 (orange juice pH) using citric acid and tripotassium citrate. Sugars and amino acids were not added to reduce the possibility that they could be possible norisoprenoid sources. The concept that carotenoids could act as a source of norisoprenoids is relatively new (15, 39, 69, 70). These studies have indicated that specific carotenoids need be present in order to produce specific norisoprenoids.




62





63


Therefore juice carotenoids may be the source of some aromas due to thermal degradation during processing and subsequence storage.

Twenty-four carotenoids from orange juice were isolated in present study (i.e., neoxanthin, 1-carotene. o-carotene, and -cryptoxanthin, see chapter 3). One of the carotenoids isolated in orange juice, 1-carotene, was studied in model aqueous solution to determine which aroma active compounds could be produced via thermal degradation and thus unequivocally demonstrate that a carotenoid found in orange juice can act as a precursor of norisoprenoids. Beta-carotene was chosen because it was commercially available, relatively inexpensive and should be the most unstable as it cannot be esterified to improve thermal stability.

Objective

Determine if aroma active norisoprenoids are generated from 1-carotene via

thermal degradation using model solutions adjusted to orange juice pH (3.8) and stored at 350C. (Objective 5)

Materials and Methods

Crystallization

Beta-carotene (99% purity, purchase from Acros) was recrystallized before using to remove aroma active impurities. The method of recrystallization followed that of Schiedt and Liaaen-Jensen (98) with minor modification. Beta-carotene was dissolved in the smallest possible volume of petroleum ether (Fisher Scientific, NJ), filtered through glass wool in a funnel, and ethanol (Fisher Scientific, NJ) was added drop-wise until turbidity was observed. The mixture was left at room temperature for about an hour and the temperature was then lowered gradually to 60C (refrigerator) and finally to -200C (deepfreeze) over night or until the crystals formed. The crystals were collected on a fine






64


sintered-glass, washed on the filter with cold ethanol and dried with the flow of nitrogen gas. The headspace volatiles of recrystallized 1-carotene were checked by SPME before using. It was ready to use when no aroma active volatiles were detected. Model Solutions

Acetone (Fisher Scientific, NJ) was chosen to dissolve the recrystallized 1-carotene because it was a polar solvent and would facilitate the transfer of 1-carotene into the model aqueous solution. One milligram of recrystallized P-carotene was dissolved in acetone and diluted to citrate buffer pH 3.8 (citric acid 1.2 g., tripotassium citrate 0.6 g. adjust pH to 3.8 by 1 N. NaOH, (Fisher Scientific, NJ)). Ten milliliters of the solution were added into 40 ml vial with Teflon coated screw cap and wrapped with aluminum foil kept in 350C for up to 1 month.

Analytical Methods

The headspace volatiles of the model solution were extracted by Solid Phase

Microextraction (SPME,50/30m DVB/Carboxen/PDMS, Supelco). The solution was equilibrated at 400C with gentle agitation (by stirring bar) for 45 min and then inserted the SPME fiber to the headspace of the model solution in order to extract and concentrate the headspace volatile by the fiber for another 45 min. The fiber was injected to GC (A HP-5890 GC (Palo Alto, CA) with either a DB-Wax or ZB-5 column whose effluent was split between an olfactometer or flame ionization detector (FID). Column oven temperature was programmed from 40 to 2400C at 7 oC/min with a 5 min hold. The aroma active compounds detected by GC-O were identified from their aroma quality and retention index by comparison with standards and confirmed by GC-MS (as describer in chapter 4).






65


Results and discussion

Before beginning storage study, the high purity 13-carotene (99% purity) was

evaluated for aroma active impurities using GC-O of the material in the same manner as the storage study. This demonstrated the potency of very minor impurities (less than 1%) and the need to recrystallize the standard 1-carotene to remove aroma active impurity before beginning the storage study (Fig. 6-1). No effort was made to identify these impurities only to remove them. Freshly recrystallized 0-carotene was used in all model solution storage studies. Before storage, a day 0 (control) was examined using GC-O to make certain no aroma active volatiles were detected (Fig. 6-2).














0


CrO

C C


10 15 20 25 Time (min)

Figure 6-1. The standard 1-carotene (99% purity) as received (no purification)





66




















10 15 20 25 Time (min)

Figure 6-2. Headspace volatiles from 13-carotene in model solution pH 3.5 at 0 day





















10 15 20 25 Time (min)

Figure 6-3. Headspace volatiles from -carotene in model solution pH 3.5 after storage 1
day at 350C: 1 = -ionone, a = sweet/raspberry






67


After one day of storage at 350C, only a single aroma active compound, 0-ionone, was detected (Fig. 6-3). This suggests that 0-ionone is either one of the most common thermal decomposition products and/or it has one of the lowest aroma thresholds. Betaionone does have one of the lowest aroma thresholds (see Table 5-2), but is probably also a common decomposition fragment and as shown in Fig. 6-3 might represent a certain weakness in the C9-C10 double bond.






i'


% % ~-ionone 0-damascone

CHO CHO

0-cyclocitral P-homocyclocitral


Figure 6-4. Degradation of -carotene in model solution at difference carbon bonds

After two weeks storage, five distinct aroma active peaks were observed. Four of these appeared to correspond to distinct FID peaks. It is interesting to note that P-ionone is still the highest peak and that all of the predicted decomposition products shown in Fig. 6-4 were observed (e.g., peak 1 was due to 0-cyclocitral, peak 2 was due to

-homocyclocitral). It appears that oxidative degradation of 1-carotene at double bond C9-C10, is the most preferable and 0-ionone was reported as the major product from 1-carotene degradation (32, 99). Beta-ionone was reported as an off-flavor of dehydrated carrot stored in oxygen. When dehydrated carrot was stored in the presence of oxygen its





68


color, due to P-carotene, was destroyed and simultaneously an off-flavor (violet-like) developed (24).

The polyene-carotenes are apparently oxidized at the first conjugated diene bonds. Oxidation is more prevalent adjacent to the methyl group and it is possible that inductive effects of the methyl group make the double bond adjacent to it more susceptible to oxidation (100). When 1-carotene in benzene or tetrachloromethane is allowed to react with molecular oxygen in the absence of light at 300C, 1-ionone can be formed within the first few hours. As the oxidation progressed a number of shorter chain products are formed, including 0-cyclocitral (101). The aroma active volatile compounds formed indicates that oxidative scission of 1-carotene can occur at carbon bond C7-C8, C8-C9 and C9-Clo of 1-carotene to generate Clo: 0 -cyclocitral, C11: 1-homocyclocitral C13: 1-damascone and C13: 1-ionone respectively (see figure 6-4). The same degradation position and volatile compounds (except 1-homocyclocitral) have been reported by photo-oxygenation (102) autoxidation with molecular oxygen at 300C in the dark conditions (101) and oxidation in water at 970C (31). At the scission C8-C9 of 1-carotene, 1-homocyclocitral (C11) was formed in present model condition but at the same scission position, dihydroactinodiolide (C11) was formed and has been reported in the different model conditions (31, 101, 102). GC-O Analysis of 1-Carotene Decomposition at 350C

The GC-O data for the two week storage sample is in Table 6-1. It should be noted that the same retention and aroma descriptors observed earlier for standard and juices were also observed for this storage sample. It is interesting to see just how similar the values in Table 6-1 are with their corresponding components in Table 5-9. In making this






69



5

o 4





o 1



3 b c
b d


10 15 20 25 Time (min)

Figure 6-5. Headspace volatiles from -carotene in model solution pH 3.5, after storage
2 weeks at 350C : 1 = P-cyclocitral, 2 = P-homocyclocitral, 3 = -damascone, 4 = unknown, 5 = 1-ionone, a = sweet/floral/hay-like, b = sweet/floral/haylike, c = sweet/apple, d = sweet/raspberry, e = sweet.

Table 6-1. Aroma active compounds from 0-carotene thermal degradation in model
solution pH 3.8, storage at 350C for 2 weeks
LRI
Compounds Aroma description ZB-5 DB-wa MS ZB-5 DB-wax
P-cyclocitral Sweet, floral, hay-like 1228 1632 x 1-homocyclocitral Sweet, floral, hay-like 1262 1780 x 1-damascone Sweet, floral 1425 1835
-ionone Sweet, raspberry 1495 1960 x

comparison, it will be noted that neither 1-homocyclocitral or O-damascone was found in orange juice. Their retention characteristics and aroma descriptors exactly matched that of authentic standards, providing enough evidence for at least a tentative identification. Positive identification of these compounds was achieved from the MS data. The MS fragmentation patterns for the identified norisoprenoids are shown in the following figures which should offer conclusive proof as to their identity.








70




MS Identification




t 12TIC 14 222 ooJ ..
i2 1 i 1 A 21A 1, 1 1 876_1 9 47 Q- 97 G A2

0 :1 61


ood A m/z = 137


S 2 1 9 .67 4 .8 8 2 .88 4 7 2 300 2



"- B m/z = 151


20 57 19 12.87 299 8 .8 Z44
U 2 .13


15) m/z = 177




18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0


Time (min)



Figure 6-6. Selected ion chromatogram (SIC) of model solution headspace volatiles after

storage 2 weeks at 350C : A = B-cyclocitral, B = P-homocyclocitral, C = Bionone.


1 00 7
90

1 09 I 2l
0 1 137
79 91 I Exact Mass =152 s 0 M olecular Formula =C10H 160
40


20 11o


90-j
BO- 109 123
70 at

50 77 95
60 7v

40
20 119


S ,50 57 . .. 1
40 60 BO 1 O0 1 20 1 40 1 60 1 0

m/z



Figure 6-7. Upper spectra from model solution MS at RT 19.61, bottom spectra from

standard B-cyclocitral using identical ion trap MS at identical retention time.









71






1 51
1 00- 107

90
so I O16 -133

70 91
,s1

50 Exact Mass =166
40 Molecular Formula =C11H180 40 95


206
0 1 09 4 52





80 107

70 91
105
60- 79
so- 1

40 67 7

102



0 I
I -- 51 1 QB 1 67
I 3 rd8 I 121 1 5 I3 11 4 I 8 1 $ 40 60 80 1 O0 120 140 1 60 1 BO


rn/z




Figure 6-8. Upper spectra from model solution MS at RT 20.46, bottom spectra from

standard 0-homocyclocitral using identical ion trap MS at identical retention

time







100 90



70


so_ Exact Mass =192
Molecular Formula =C13H200
40 30


91 162 ,o ,o5 ,17 135 147

100 90



70 60

so40 3020
91 162
10 51 65 7171 93 lO 10 7 15 I159 1.

O M :, --.""' I I I I, 141 J,"3
40 60 so 100 120 140 1 60 1 o 200



m/z


Figure 6-9. Upper spectra from model solution MS at RT 25.13, bottom spectra from

standard 0-ionone using identical ion trap MS at identical retention time.






72


Conclusion

The norisoprenoids detected in this study were also detected in other foods that have been thermally processed e.g.,tea (95), tomato paste (50). Therefore the results of this study indicate that -carotene could be a precursor of -cyclocitral and 0-ionone in orange juice. Furthermore it could be a precursor of norisoprenoids during thermal processing and subsequence storage at relative high temperature and it is reasonable to assume that any appreciable change in carotenoids content of orange juice will have an effect on flavor.













CHAPTER 7
CONCLUSIONS

Four norisoprenoids, 0-cyclocitral, 1-damascenone, a-ionone and B-ionone were identified in orange juice using headspace SPME, GC-O, GC-FID and GC-MS. Three of them, P-cyclocitral, 0-damascenone, a-ionone were identified and confirmed by GC-MS for the first time. Their concentrations in fresh orange juice were determined using SPME with standard addition technique. Odor activity values (OAV) were calculated using published threshold values. Calculated OAV values suggest that 0-ionone provided the greatest contribution to total floral aroma in orange juice compared to the other three norisoprenoids. The concentration of 1-damascenone increased almost 10 fold after thermal processing, indicating there are thermally unstable precursors which generate 0-damascenone at elevated temperatures. All four of the norisoprenoids in orange juice contribute 8-10% floral aroma to the total aroma quality of the orange juice and are the major contributors (60-80%) in the floral category.

Several carotenoids were identified using HPLC with photodiode array detection. Twenty-four carotenoids were separated as distinct peaks and sixteen of these peaks were identified based on their spectral characteristics, relative elution order compared to literature values and authentic standards. The identified carotenoids include: a-carotene,

-carotene, a-cryptoxanthin, 0-cryptoxanthin and neoxanthin, which are known as norisoprenoid precursors. These specific carotenoids were of interest because they





73






74


possess the direct structural segments needed to serve as precursors to the newly identified norisoprenoids.

To demonstrate that carotenoids could serve as norisoprenoids precursors,

3-carotene was studied in a model system at 350C storage. GC-O and GC-MS data confirmed the presence of -cyclocitral and 0-ionone in these solutions in as little as two weeks. This was direct proof that 1-carotene can degrade to form specific norisoprenoids under conditions an orange juice might encounter.















LIST OF REFERENCES

1 Moshonas, M. G.; Shaw, P. E. Quantitative analysis of orange juice flavor
volatiles by direct-injection gas chromatography. J. Agric. Food Chem. 1987, 35,
161-165.

2 Moshonas, M. G.; Shaw, P. E. Quantitative determination of 46 volatile
constituents in fresh, unpasteurized orange juices using dynamic headspace gas
chromatography. J. Agric. Food Chem. 1994, 42, 1525-1528.

3 Tonder, D.; Petersen, M. A.; Poll, L.; Olsen, C. E. Discrimination between freshly
made and stored reconstituted orange juice using GC odor profiling and aroma
values. Food Chem. 1998, 61, 223-229.

4 Hinterholzer, A.; Schieberle, P. Identification of the most odor-active volatiles in
fresh, hand-extracted juice of Valencia late oranges by odor dilution techniques.
Flavour Frag. J. 1998, 13, 49-55.

5 Buettner, A.; Schieberle, P. Evaluation of aroma differences between handsqueezed juices from Valencia late and Navel oranges by quantitation of key
odorants and flavor reconstitution experiments. J. Agric. Food Chem. 2001, 49,
2387-2394.

6 Bazemore, R.; Goodner, K.; Rouseff, R. Volatiles from unpasteurized and
excessively heated orange juice analyzed with solid phase microextraction and
GC-olfactometry. J. Food Sci. 1999, 64, 800-803.

7 Marin, A. B.; Acree, T. E.; Hotchkiss, J. H.; Nagy, S. Gas chromatographyolfactometry of orange juice to assess the effects of plastic polymers on aroma
character. J. Agric. Food Chem. 1992, 40, 650-654.

8 Kotseridis, Y.; Baumes, R. Identification of impact odorants in bordeaux red
grape juice, in the commercial yeast used for its fermentation, and in the produced
wine. J. Agric. Food Chem. 2000, 48, 400-406.

9 Zhou, P. G.; Cox, J. A.; Roberts, D. D.; Acree, T. E. Beta-Damascenone
precursors in apples. Prog. Flavour Precursor Stud. Proc. Int. Conf. 1993, 261273.

10 Ong, P. K. C.; Acree, T. E. Gas chromatography/olfactory analysis of Lychee
(Litchi chinensis Sonn.). J. Agric. Food Chem. 1998, 46, 2282-2286.



75






76


11 MacLeod, G.; Ames, J. M. Volatile components of starfruit. Phytochemistry 1990,
29, 165-172.

12 Olle, D.; Baron, A.; Lozano, Y. F.; Sznaper, C.; Baumes, R.; Bayonove, C.;
Brillouet, J. M. Microfiltration and reverse osmosis affect recovery of mango
puree flavor compounds. J. Food Sci. 1997, 62, 1116-1119.

13 Ishida, B. K.; Mahoney, N. E.; Ling, L. C. Increased lycopene and flavor volatile
production in tomato calyces and fruit cultured in vitro and the effect of 2-(4Chlorophenylthio)triethylamine. J. Agric. Food Chem. 1998, 46, 4577-4582.

14 Tarantilis, P. A.; Polissiou, M. G. Isolation and identification of the aroma
components from saffron (Crocus sativus). J. Agric. Food Chem. 1997, 45, 459462.

15 Enzell, C. Biodegradation of carotenoids an important route to aroma
compounds. Pure Appl. Chem. 1985, 57, 693-700.

16 Sanderson, G. W.; Co, H.; Gonzalez, J. G. Biochemistry of tea fermentation: role
of carotenes in black tea aroma formation. J. Food Sci. 1971, 36, 231-236.

17 Gross, J.; Gabai, M.; Lifshitz, A. Comparative study of the carotenoid pigments in
juice of Shamouti, Valencia, and Washington oranges, three varieties of Citrus
sinensis. Phytochemistry 1972, 11, 303-308.

18 Isoe, S.; Hyeon, S. B.; Sakan, T. Photo-oxygenation of carotenoids. I. Formation
of dihydroactinidiolide and 0-ionone from 0-carotene. Tetrahedron Lett. 1969,
279-281.

19 Demole, E.; Enggist, P.; Winter, M.; Furrer, A.; Schulte-Elte, K. H.; Egger, B.;
Ohloff, G. Megastigma-5,8-dien-4-one, an aroma constituent of the yellow
passion fruit and Virginia tobacco. Helv. Chim. Acta 1979, 62, 67-75.

20 Strauss, C. R.; Wilson, B.; Anderson, R.; Williams, P. J. Development of
precursors of C13 nor-isoprenoid flavorants in Riesling grapes. Am. J. Enol. Vitic.
1987, 38, 23-27.

21 Williams, P. J.; Sefton, M. A.; Francis, I. L. Glycosidic Precursors of Varietal
Grape and Wine Flavor. In Flavor Precursors: Thermal and Enzymatic
Conversions; R. Teranishi; G. R. Takeoka and M. Guntert, Eds.; American
Chemical Society: Washington, DC, 1992; pp 74-86.

22 Enzell, C. R. Influence of Curing on the Formation of Tobacco Flavor. In Flavour
'81; P. Schreier, Ed.; de Gruyter: Berlin, New York, 1981; pp 449-478.

23 Buttery, R. G.; Teranishi, R.; Flath, R. A.; Ling, L. C. Identification of additional
tomato paste volatiles. J. Agric. Food Chem. 1990, 38, 792-795.






77


24 Ayers, J. E.; Fishwick, M. J.; Land, D. G.; Swain, T. Off-flavor of dehydrated
carrot stored in oxygen. Nature 1964, 203, 81-82.

25 Vernin, G. Terpenoids and norisoprenoids in grape berry aromas and in wines. 1.
Origins and formation. Rivista Italiana EPPOS 1997, 8, 19-31.

26 Gross, J. Pigments in Fruits; Academic Press Inc.: London, 1987; 303 pp.

27 Weedon, B. C. L.; Moss, G. P. Structure and nomenclature. In Carotenoids; G.
Britton; S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel Boston
Berlin, 1995; pp 27-70.

28 Mordi, R. C. Mechanism of 0-carotene degradation. Biochem. J. 1993, 292, 310312.

29 Lutz-Roder, A.; Jezussek, M.; Winterhalter, P. Nickel peroxide induced oxidation
of canthaxanthin. J. Agric. Food Chem. 1999, 47, 1887-1891.

30 Ouyang, J. M.; Daun, H.; Chang, S. S.; Ho, C. T. Formation of carbonyl
compounds from 1-carotene during palm oil deodorization. J. Food Sci. 1980, 45,
1214-1217.

31 Kanasawud, P.; Crouzet, J. C. Mechanism of formation of volatile compounds by
thermal degradation of carotenoids in aqueous medium. 1. -Carotene
degradation. J. Agric. Food Chem. 1990, 38, 237-243.

32 Winterhalter, P. Carotenoid-Derived Aroma Compounds: Biogenetic and
Biotechnological Aspects. In Biotechnology for Improved Foods and Flavors; G.
R. Takeoka; R. Teranishi; P. J. Williams and A. Kobayashi, Eds.; American
Chemical Society: Washington, DC, 1996; pp 295-308.

33 Weeks, W. W. Carotenoids: A Source of Flavor and Aroma. In Biogeneration of
Aromas; H. P. Thomas and C. Rodney, Eds.; American Chemical Society:
Washington, D.C., 1986; pp 157-166.

34 Winterhalter, P.; Rouseff, R. Carotenoid-Derived Aroma Compounds: an
Introduction. In Carotenoid-Derived Aroma Compounds; P. Winterhalter and R.
Rouseff, Eds.; American Chemical Society: Washington, DC, 2002; pp 1-17.

35 Knapp, H.; Straubinger, M.; Stingl, C.; Winterhalter, P. Analysis of
Norisoprenoid Aroma Precursors. In Carotenoid-Derived Aroma Compounds; P.
Winterhalter and R. L. Rouseff, Eds.; American Chemical Society: Washington,
DC, 2002; pp 20-35.

36 Baumes, R.; Wirth, J.; Bureau, S.; Gunata, Y.; Razungles, A. Biogeneration of
C13-norisoprenoid compounds: experiments supportive for an apo-carotenoid
pathway in grapevines. Anal. Chim. Acta 2002, 458, 3-14.






78


37 Gunata, Z.; Dugelay, I.; Sapis, J. C.; Baumes, R.; Bayonove, C. Role of enzymes
in the use of the flavor potential from grape glycosides in winemaking. Prog.
Flavour Precursor Stud. Proc. Int. Conf 1993, 219-234.

38 Winterhalter, P. Oxygenated C13-Norisoprenoids: Important Flavor Precursors. In
Flavor Precursors: Thermal and Enzymatic Conversions; R. Teranishi; G. R.
Takeoka and M. Guntert, Eds.; American Chemical society: Washington, DC,
1992; pp 98-115.

39 Lutz, A.; Winterhalter, P. Isolation of additional carotenoid metabolites from
quince fruit (Cydonia oblonga Mill.). J. Agric. Food Chem. 1992, 40, 1116-1120.

40 Aubert, C.; Ambid, C.; Baumes, R.; Guenata, Z. Investigation of Bound Aroma
Constituents of Yellow-Fleshed Nectarines (Prunus persica L. Cv. Springbright).
Changes in Bound Aroma Profile during Maturation. J. Agric. Food Chem. 2003,
51, 6280-6286.

41 Buttery, R. G.; Teranishi, R.; Ling, L. C.; Flath, R. A.; Stem, D. J. Quantitative
studies on origins of fresh tomato aroma volatiles. J. Agric. Food Chem. 1988, 36,
1247-1250.

42 Wilkins, C. K. Paprika: relationships between aroma profile data and both GC and
HPLC data. Lebensm- Wiss.- Technol. 1992, 25, 212-218.

43 Enzell, C. R.; Wahlberg, I. Tobacco isoprenoids precursors of important aroma
constituents. Pure Appl. Chem. 1990, 62, 1353-1356.

44 Kawakami, M.; Ganguly, S. N.; Banerjee, J.; Kobayashi, A. Aroma composition
of oolong tea and black tea by brewed extraction method and characterizing
compounds of darjeeling tea aroma. J. Agric. Food Chem. 1995, 43, 200-207.

45 Suzuki, M.; Matsumoto, S.; Fleischmann, H.-P.; Shimada, H.; Yamano, Y.; Ito,
M.; Watanabe, N. Identification of 0-damascenone Progenitors and Their
Biogenesis in Rose Flowers (Rosa damascena Mill.). In Carotenoid-Derived
Aroma Compounds; P. Winterhalter and R. L. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002; pp 89-101.

46 Kotseridis, Y.; Baumes, R. L.; Skouroumounis, G. K. Quantitative determination
of free and hydrolytically liberated .beta.-damascenone in red grapes and wines
using a stable isotope dilution assay. J. Chromatogr., A 1999, 849, 245-254.

47 Sefton, M. A.; Francis, I. L.; Williams, P. J. Volatile norisoprenoid compounds as
constituents of oak woods used in wine and spirit maturation. J. Agric. Food
Chem. 1990, 38, 2045-2049.

48 Ohloff, G. Importance of minor components in flavors and fragrances. Perfum.
Flavor. 1978, 3, 11-12, 14-16, 18-22.






79


49 Roberts, D. D.; Mordehai, A. P.; Acree, T. E. Detection and Partial
Characterization of Eight 1-Damascenone Precursors in Apples (Malus domestica
Borkh. Cv. Empire). J. Agric. Food Chem. 1994, 42, 345-349.

50 Buttery, R. G.; Teranishi, R.; Ling, L. C.; Turnbaugh, J. G. Quantitative and
sensory studies on tomato paste volatiles. J. Agric. Food Chem. 1990, 38, 336340.

51 Buttery, R. G.; Ling, L. C. Volatile Components of Tomato Fruit and Plant Parts:
Relationship and Biogenesis. In Bioactive Volatile Compounds from Plants; R.
Teranishi; R. G. Buttery and H. Sugisawa, Eds.; American Chemical Society:
Washington, DC, 1993; pp 23-34.

52 Lozano, P.; Delgado, D.; Gomez, D.; Rubio, M.; Iborra, J. L. A non-destructive
method to determine the safranal content of saffron (Crocus sativus L.) by
supercritical carbon dioxide extraction combined with high-performance liquid chromatography and gas chromatography. J. Biochem. Biophys. Methods 2000,
43, 367-378.

53 Knapp, H.; Straubinger, M.; Witte, A.; Winterhalter, P. Aroma Formation in
Saffron. In Frontiers of Flavour Science; P. Winterhalter, Ed.: Freising, Germany,
2000; pp 440-444.

54 Razungles, A.; Bayonove, C. L.; Cordonnier, R. E.; Sapis, J. C. Grape
carotenoids: changes during the maturation period and localization in mature
berries. Am. J. Enol. Vitic. 1988, 39, 44-48.

55 Razungles, A.; Gunata, Z.; Pinatel, S.; Baumes, R.; Bayonove, C. Content of
terpenes, norisoprenoids and their precursors in several varieties of grapes. Sci.
Aliments 1993, 13, 59-72.

56 Schreier, P.; Drawert, F.; Junker, A. Identification of volatile constituents from
grapes. J. Agric. Food Chem. 1976, 24, 331-336.

57 Williams, P. J.; Sefton, M. A.; Wilson, B. Nonvolatile Conjugates of Secondary
Metabolites as Precursors of Varietal Grape Flavor Components. In Flavor
Chemistry: Trends and Development; R. Teranishi; R. G. Buttery and F. Shahidi,
Eds.; American Chemical Society: Washington, DC, 1989; pp 35-48.

58 Schneider, R.; Razungles, A.; Augier, C.; Baumes, R. Monoterpenic and
norisoprenoidic glycoconjugates of Vitis vinifera L. cv. Melon B. as precursors of
odorants in Muscadet wines. J. Chromatogr., A 2001, 936, 145-157.

59 Acree, T. E.; Barnard, J.; Cunningham, D. G. A procedure for the sensory
analysis of gas chromatographic effluents. Food Chem. 1984, 14, 273-286.





80


60 Schieberle, P.; Grosch, W. Identificierung von Aromastoffen ausder Kruste von
Roggenbrot Verleich mit den Aromastoffen der Krume. Z. Lebensm.-Unters.
Forsch. 1984, 479-483.

61 da Silva, M. A. A. P.; Lundahl, D. S.; M.R., M. The Capabilities and
Psychophysics of Osme: a New GC-Olfactometry Technique. In Trends in Flavor
Research; H. Maarse and V. D. Heij, Eds.; Elsevier: New York, 1994; pp 191210.

62 Stevens, S. S. Concerning the psychophysical power law. Quart. J. Exp. Psychol.
1964, 16, 383-385.

63 Arthur, C. L.; Pawliszyn, J. Solid phase microextraction with thermal desorption
using fused silica optical fibers. Analytical Chemistry 1990, 62, 2145-2148.

64 Jia, M.; Zhang, Q. H.; Min, D. B. Optimization of Solid-Phase Microextraction
Analysis for Headspace Flavor Compounds of Orange Juice. J. Agric. Food
Chem. 1998, 46, 2744-2747.

65 Bezman, Y.; Rouseff, R. L.; Naim, M. 2-Methyl-3-furanthiol and methional are
possible off-flavors in stored orange juice: aroma-similarity, NIF/SNIF GC-O,
and GC analyses. J. Agric. Food Chem. 2001, 49, 5425-5432.

66 Steffen, A.; Pawliszyn, J. Analysis of Flavor Volatiles Using Headspace SolidPhase Microextraction. J. Agric. Food Chem. 1996, 44, 2187-2193.

67 Song, J.; Fan, L.; Beaudry, R. M. Application of Solid Phase Microextraction and
Gas Chromatography/Time-of-Flight Mass Spectrometry for Rapid Analysis of
Flavor Volatiles in Tomato and Strawberry Fruits. J. Agric. Food Chem. 1998, 46,
3721-3726.

68 Rouseff, R.; Raley, L.; Hofsommer, H.-J. Application of diode array detection
with a C-30 reversed phase column for the separation and identification of
saponified orange juice carotenoids. J. Agric. Food Chem. 1996, 44, 2176-2181.

69 Winterhalter, P.; Lutz, A.; Schreier, P. Isolation of a glucosidic precursor of
isomeric marmelo lactones from quince fruit. Tetrahedron Lett. 1991, 32, 36693670.

70 Lutz, A.; Winterhalter, P. Bio-oxidative cleavage of carotenoids: important route
to physiological active plant constituents. Tetrahedron Lett. 1992, 33, 5169-5172.

71 Kotseridis, Y.; Baumes, R. L.; Bertrand, A.; Skouroumounis, G. K. Quantitative
determination of b-ionone in red wines and grapes of bordeaux using a stable
isotope dilution assay. Journal of Chromatography, A 1999, 848, 317-325.






81


72 Lee, H. S.; Castle, W. S. Seasonal Changes of Carotenoid Pigments and Color in
Hamlin, Early gold, and Budd Blood Orange Juices. J. Agric. Food Chem. 2001,
49, 877-882.

73 Pupin, A. M.; Dennis, M. J.; Toledo, M. C. F. HPLC analysis of carotenoids in
orange juice. Food Chem. 1999, 64, 269-275.

74 Philip, T.; Chen, T. S.; Nelson, D. B. Liquid chromatographic profiles of major
carotenoid esters in commercially processed California Navel and Valencia
orange juice concentrates. J. Chromatogr. 1988, 442, 249-265.

75 Lee, H. S.; Castle, W. S.; Coates, G. A. High-performance liquid chromatography
for the characterization of carotenoids in the new sweet orange (Earlygold) grown
in Florida, USA. J. Chromatogr., A 2001, 913, 371-377.

76 Mouly, P. P.; Gaydou, E. M.; Lapierre, L.; Corsetti, J. Differentiation of several
geographical origins in single-strength Valencia orange juices using quantitative
comparison of carotenoid profiles. J. Agric. Food Chem. 1999, 47, 4038-4045.

77 Noga, G.; Lenz, F. Separation of citrus carotenoids by reversed-phase highperformance liquid chromatography. Chromatographia 1983, 17, 139-142.

78 Marais, J. 1,1,6-Trimethyl-1,2-dihydronaphthalene (TDN): a possible degradation
product of lutein and b-carotene. South African J. Enol. Viticulture 1992, 13, 5255.

79 Chen, C. S.; Shaw, P. E.; Parish, M. E. Orange and Tangerine Juice. In Fruit
Juice Processing Technology; S. Nagy; C. S. Chen and P. E. Shaw, Eds.;
AGScience, Inc.: Auburndale, 1993; pp 110-165.

80 Stewart, I.; Wheaton, T. A. Continuous flow separation of carotenoids by liquid
chromatography. J. Chromatogr. 1971, 55, 325-336.

81 Britton, G. UV/Visible Spectroscopy. In Carotenoids, Vol. ]A: Isolation and
Analysis; G. Britton; S. Liaaen-Jenson and H. Pfander, Eds.; Birkhauser Verlag:
Basel, Boston, Berlin, 1995; pp 13-62.

82 De Ritter, E.; Purcell, A. E. Carotenoid analytical Methods. In Carotenoids as
Colorants and Vitamin A Precursors; J. C. Bauemfeind, Ed.; Academic Press:
London, 1981; pp 815-923.

83 Farin, D.; Ikan, R.; Gross, J. The carotenoid pigments in the juice and flavedo of a
mandarin hybrid (Citrus reticulata) cv. Michal during ripening. Phytochemistry
1983, 22,403-408.

84 Curl, A. L.; Bailey, G. F. The carotenoids of Navel oranges. J. Food Sci. 1961,
26, 442-447.





82


85 Robertson, G. W.; Griffiths, D. W.; Woodford, J. A. T.; Birch, A. N. E. Changes
in the chemical composition of volatiles released by the flowers and fruits of the
red raspberry (Rubus idaeus) cultivar glen Prosen. Phytochemistry 1995, 38,
1175-1179.

86 Castellar, M. R.; Montijano, H.; Manjon, A.; Iborra, J. L. Preparative highperformance liquid chromatographic purification of saffron secondary
metabolites. J. Chromatogr. 1993, 648, 187-190.

87 Gross, J.; Gabai, M.; Lifshitz, A. Carotenoids in juice of Shamouti orange. J.
Food Sci. 1971, 36, 466-473.

88 Valim, M. F.; Rouseff, R. L.; Lin, J. Gas Chromatographic-Olfactometric
Characterization of Aroma Compounds in Two Types of Cashew Apple Nectar. J.
Agric. Food Chem. 2003, 51, 1010-1015.

89 Ahmed, E. M.; Dennison, R. A.; Shaw, P. E. Effect of selected oil and essence
volatile components on flavor quality of pumpout orange juice. J. Agric. Food
Chem. 1978, 26, 368-372.

90 Poole, C. F. The essence of chromatography; Elsevier: Amsterdam, Boston,
London, New York, Oxford, Paris, san Diego, San Francisco, Singapore, Sidney,
Tokyo, 2003.

91 Zhang, Z.; Pawliszyn, J. Headspace solid-phase microextraction. Anal. Chem.
1993, 65, 1843-1852.

92 Bao, M.; Mascini, M.; Griffini, O.; Burrini, D.; Santianni, D.; Barbieri, K.
Headspace solid-phase microextraction for the determination of trace levels of
taste and odor compounds in water samples. Analyst (Cambridge, United
Kingdom) 1999, 124, 459-466.

93 Radford, T.; Kawashima, K.; Friedel, P. K.; Pope, L. E.; Gianturco, M. A.
Distribution of volatile compounds between the pulp and serum of some fruit
juices. J. Agric. Food Chem. 1974, 22, 1066-1070.

94 Lin, J.; Rouseff, R. L.; Barros, S.; Naim, M. Aroma Composition Changes in
Early Season Grapefruit Juice Produced from Thermal Concentration. J. Agric.
Food Chem. 2002, 50, 813-819.

95 Sanderson, G. W.; Grahamm, H. N. Formation of black tea aroma. J. Agric. Food
Chem. 1973, 21, 576-585.

96 Day, W. C.; Erdman, J. G. Ionene; a thermal degradation product of 1-carotene.
Science 1963, 141, 808.

97 Mader, I. 3-Carotene: thermal degradation. Science 1964, 144, 533-534.






83


98 Schiedt, K.; Liaaen-Jensen, S. Isolation and Analysis. In Carotenoids; G. Britton;
S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel, Boston, Berlin,
1995; pp 81-108.

99 Ravichandran, R. Carotenoid composition, distribution and degradation to flavor
volatiles during black tea manufacture and the effect of carotenoid
supplementation on tea quality and aroma. Food Chem. 2002, 78, 23-28.

100 Stevens, M. A. Relationship Between Polyene-Carotene Content and Volatile
Compound Composition of tomatoes. J. Amer. Soc. Hort. Sci. 1970, 95, 461-464.

101 Mordi, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Ingold, K. U.; Lindsay,
D. A. Exploratory study of 1-carotene autoxidation. Tetrahedron Lett. 1991, 32,
4203-4206.

102 de Heij, H. T.; van Dort, H. M. Autoxidation of carotenes. Prog. Flavour
Precursor Stud. Proc. Int. Conf. 1993, 409-413.















BIOGRAPHICAL SKETCH

Kanjana Mahattanatawee was born in Bangkok, Thailand on 17 May, 1965. She

received a B.Sc. in biology with a major in Microbiology in 1988 from Sri-Nakharinwirot University, Thailand. She continued to pursue her Master of Science degree in the area of industrial microbiology at the Department of Microbiology, Chulalongkom University, Bangkok Thailand from 1988-1991. From 1991-1992 she worked as a researcher, in the Department of Microbiology, Chulalongkorn University, Thailand. In 1992-1993 Kanjana was awarded a UNESCO scholarship to earn her Diploma in Microbiology and Biotechnology from Osaka University, Japan. Kanjana was appointed to a position as Lecturer, Department of Food Technology, Faculty of Science, Siam University from 1993-1999. From 1995-1997 she was an adjunct lecturer, Faculty of Environment and Natural Resource, Mahidol University. Kanjana conducted research and taught two microbiology courses (Industrial Microbiology and Fermentation Technology) for undergraduate students at the Faculty of Science, Siam University, Bangkok, Thailand.

She was awarded a scholarship from Siam University to pursue her Ph.D. In

Spring 1999, she enrolled in the graduate program at the Department of Food Science and Human Nutrition at the University of Florida under Dr. R.L. Rouseff's supervision. She considers herself very fortunate to be enrolled in one of the greatest graduate programs in flavor chemistry, with excellent scientists who are a pleasure to work with. She completed her research for her Ph.D. degree at the Citrus Research and Education Center (CREC) in Lake Alfred, Florida.


84






85


After completing her Ph.D. program, Kanjana plans to work as a postdoctoral

researcher to gain more experience in this subject area. Later, she will return to Thailand to fulfill an appointed position as an associate professor at Siam University. Kanjana will teach and conduct research. She hopes to deliver the excitement and enthusiasm to her students in Thailand that she has experienced from her professors in the U.S.A.









I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Russell L. Rouseff, Chair
Professor of Food Science and Human Nutrition

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of P ilosophy.


Maurice R. Marshall, Jr.
Professor of Food Science and Human Nutrition

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


Re ee M.Xoodrich
Assistant Professor of Food Science and Human Nutrition

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy.


David H. Powell
Faculty Scientist of Chemistry

This dissertation was submitted to the Graduate Faculty of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy.

May 2004



Dean, College of Agriculture and fe Sc



Dean, Graduate School




Full Text
39
monitoring mode at least a lOx greater sensitivity (lower detection level) can be achieved
because all the ions of a single mass are continuously measured rather than measured for
an instant before monitoring other masses. Using selected ion monitoring m/z values 175
and 190, P-damascenone was detected at the expected retention time (see Fig. 4-3 for the
case at m/z 190). The combined GC-0 and two SIM peaks at the exact retention time of
P-damascenone, confirm its presence in orange juice.
Although selected ion chromatograms strongly suggest the presence of the other
norisoprenoids of interest, they do not offer absolute proof. They only indicate that a
juice volatile elutes at the identical retention time as the norisoprenoid of interest, and
this volatile contains the same mass fragment. The combination of this information with
the GC-0 information provides three independent pieces of information strongly
suggesting the presence of specific norisoprenoids. However, to absolutely confirm the
presence of P-cyclocitral, a-ionone, and P-ionone, their spectra from the juice MS data at
the retention times of each respective norisoprenoid was obtained and compared with
reference spectra in standard libraries or compared with that obtained from authentic
standards. The resulting match for the case of P-cyclocitral is shown in Fig. 4-4. It is
readily apparent that although the relative ion abundances are not the same (usually a
function of instrument to instrument variation) an excellent spectral match has occurred
and that the presence of P-cyclocitral in orange juice is confirmed. Comparing the
relative abundances of ions m/z 137 vs 152 in the upper spectrum of Fig. 4-4, it can be
readily appreciated why examining the selective ion chromatogram at 137 provided a
better signal to noise ratio than the chromatogram at 152, the molecular ion for P-
cyclocitral.


90
Figure 7-3. Feeding scars on water hyacinth (Eichhornia crassipes) leaf after exposure
to Neochetina eichhorniae adults. Black marks represent feeding scars
marked with a fine tip marker to aid in counting (other side counted but not
shown).


CHAPTER 2
HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL INSECT
PESTS
Non-Protein Amino Acids
One avenue of pest management explored in the field of biorational pesticides is
the use nonprotein amino acids. Secondary plant materials such as these serve many
functions in insect-plant relationships from attractants and repellents to crude insecticides
(Dahlman 1980). Only a few nonprotein amino acids have been examined as a potential
means to control insect pests. L-canavanine and its by-product of detoxification, L-
canaline, have been studied extensively, with a variety of effects ranging from
developmental deformities to aberrant adult behavior (Dahlman and Rosenthal 1975;
1976; Rosenthal et al. 1995). L-canavanine is found mainly in leguminous plants,
including several economic species (Bell 1978; Felton and Dahlman 1984). It is believed
that plants produce this allelochemical for protection against feeding by phytophagous
insects and herbivores (Rosenthal 1977). The mode of action for canavanine can be
traced to several metabolic processes, including disruption of DNA/RNA and protein
synthesis, arginine metabolism, uptake, anomalous canavanyl protein formation, and the
reduction of active transport of K+ in the midgut epithelium (Kammer et al. 1978;
Racioppi and Dahlman 1980; Rosenthal 1977; Rosenthal et al. 1977; Rosenthal and
Dahlman 1991). In contrast, canaline possesses neurotoxic characteristics with an
unknown mode of action (Kammer et al. 1978). The species of choice for studies
involving nonprotein amino acids has been the tobacco homworm (THW), Manduca
sexta (L.) (Lepidoptera: Sphingidae).
7


headspace, GC-O, and GC-MS data. Only two norisoprenoids (p-damascenone and
P-ionone) were detected in reconstituted juice that had been thermally concentrated. Peel
oil from the same fruit contained only P-damascenone and P-cyclocitral.
Concentrations of P-cyclocitral, P-damascenone, a-ionone, and P-ionone were
determined using standard addition SPME and GC-MS and found to be 145, 0.09, 47 and
83 /rg/L respectively. The concentration of P-damascenone increased from 0.09 to 0.85
fig/L after thermal concentration and reconstitution. Orange juice norisoprenoids
contribute approximately 8-10% of total aroma intensity as determined from combined
aromagram peak heights and 60-80% of the total floral-category.
Known norisoprenoids precursors (P-carotene, a-carotene, a-cryptoxanthin,
P-cryptoxanthin, and neoxanthin) were identified in Valencia orange juice using C30
reverse phase HPLC with photodiode array detection.
Thermal decomposition products of P-carotene in citric acid solutions buffered at
pH 3.8 were examined during 35C storage using GC-0 and GC-MS. Beta-cyclocitral,
p-homocyclocitral, P-damascone and P-ionone were detected after 2 weeks thus
demonstrating that P-carotene can produce norisoprenoids. Since half of the a-carotene,
a-cryptoxanthin, P-cryptoxanthin structures share the identical structure as P-carotene,
these carotenoids must be considered potential norisoprenoid sources as well.
xiii


92
(Error bars @ 95%; F(00S)4> u= 0.98, F=3.33; /* =0.038)
1400
Control 0.1% 0.5% 1.0% Proline
Figure 7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated
with L-methionine and Proline. No statistical differences were observed
between treatments (Tukeys MST, P=0.038).


44
O 1
3 4 5
Days of Exposure
Figure 4-1. Mortality of Colorado potato beetle larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nTOtai=560).
Proline (1.0%) and Bit were included for comparison as positive and
negative controls. Data were adjusted using Abbotts formula for
control mortality.


11
cleavage during crushing, fermentation, and bottle-aging result in cleavage of the bound
sugar moiety releasing the free norisoprenoid aglycone (57, 58).
Gas Chromatography-Olfactometry
Gas chromatography-olfactometry (GC-O) is a technique that allows the effluent
from the GC column to be evaluated for aroma activity using the human nose. The
effluent from the GC column is usually split between an FID detector and sniff port. The
human being detects which of the volatiles eluting from a GC column are aroma active,
as well as to describe aroma quality, and to estimate aroma intensity. The FID detector is
used as a general mass detector. Some of the GC-0 techniques available are Charm
Analysis (59), Aroma Extraction Dilution Analysis (AEDA) (60), and OSME (61) which
is a time intensity method. Charm Analysis and ADEA are based on the determination
of odor detection thresholds of the compounds through a series of dilutions. Both define
aroma strength in terms of its dilution strength. OSME determines intensities based upon
magnitude estimation using a variable potentiometer to estimate intensity. Da Silva et al.
(61) suggested that dilution techniques might not give accurate values of aroma intensity,
since the odorants may have different dose-response functions above their thresholds.
Stevens law (62) establishes that the odor intensity (I) of a compound increases as a
power function (n, which varies from compound to compound) of the concentration
within a certain range of concentration (C) directly above the detection threshold (7).
The law is commonly expressed as:
I = k (C-T)n,
where k represents the proportionality constant. Response will increase once the
threshold concentration is exceeded. Even though not defined by the above equation, a
limit will be reached where the sensory response will no longer increase with increasing


91
Figure 7-4. Mortality of Neochetina eichhorniae on treated water hyacinth
leaves. Data corrected for control mortality using Abbotts
formula.


71
Each row contained the 4 treatment plots of 10 plants (control (0% L-methionine), 0.1%
L-methionine, 0.5% L-methionine and 1.0% L-methionine in deionized water solutions)
in a Latin square design. Plants within treatment plots were spaced 3 feet apart while
treatment plots were 9 feet apart. Figure 6-1 shows the diagrammatic representation of
the field plot.
Plant Yield
Before beginning the experiment, all developing eggplants were removed from
the plants in an effort to standardize the treatments and ensure all eggplant development
occurred after the exposure of methionine. Treatments were administered using a KQ 3L
CO2 (Weed Systems, Inc.; Hawthorne, FL) backpack sprayer charged to 30 lbs PSI and a
3-nozzle boom to ensure complete coverage of the plant (Figure 6-2). Each treatment
consisted of a 3L application over the 4 representative groups. The adjuvant Silwett
L-77 (0.5% concentration) was included to improve the residual effect of the
methionine under the field conditions. Plants were sprayed a total of nine times at
approximately two-week intervals. Fruits were harvested at various times during the
study and were weighed in the field using a Tokyo Electronics hand-held digital scale.
Pest Introduction
Neonate CPB larvae were reared on excised eggplant leaves for two days at 27C,
60% relative humidity and 16L/8D photoperiod in FRJUs to ensure healthy individuals
for the test. Larvae were transferred to the field plants using a camel hairbrush and the
branch marked with flagging tape. Introduction was made after the last spray treatment
in November. Ten larvae were placed on each plant for a total sample size of 1,600
individuals. Plants were inspected for the next 5 days and larvae encountered noted.


37
were treated with an extract from potato foliage suggesting induced host preference,
attraction, and dependence on this compound in the extent of sustained feeding and
development. A combination of sensory structures may be involved for the detection of
specific amino acids and host plant compounds, which may explain the selection of
methionine depleted host plants to avoid problems with the CAATCH1 system present in
the midgut of the THW.
The difference in the LC50 for the artificial and natural diets was striking
considering the concentrations were the same. One possible explanation is the
L-methionine on the natural diet was more readily available than that found in the
artificial diet. With the artificial diet, the L-methionine is presumably spread throughout
the diet and would therefore take longer for the THW to ingest enough to adversely affect
the CAATCH1 system. In contrast, the L-methionine was found on the surface of the
leaf in higher concentrations than that of the artificial diet and was also freely available
once ingested. Thus, larvae were exposed to a higher concentration of L-methionine with
less work to digest, resulting in lower survivorship in the same period of time.
The 1.0%L-methionine concentration had the same mortality, feeding and
developmental rates for THW, as did the Btk treatments (Figure 3-9). The 0.3%
L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments were virtually the
same for mortality (Figure 3-9), developmental rate (Figure 3-10) and total leaf material
consumed (Figure 3-11) and statistically the same as the 1.0% L-methionine
concentration and the Btk treatment. The similar mortality rate observed for the higher
concentrations of L-methionine and Btk is encouraging considering the resistance to Bt
seen in many insect species because of reduced receptor activity and binding (Bills et al.


CHAPTER 5
QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT OF
NORISOPRENOIDS IN ORANGE JUICE
Introduction
Chromatographic data is often used to determine the relative concentrations of
components in a volatile mixture. Within the linear range of the detector, integrated peak
area is proportional to the amount of that component in the sample. The four techniques
commonly employed to quantify chromatographic components are normalization;
internal standards; external standards; and standard addition methods (90). Only a few of
these have been employed to determine the amounts of specific volatiles in orange juice
volatiles to better understand their contribution to orange flavor (2, 5). Buettner and
Schieberle (5) employed stable isotope dilution assay to quantify 25 volatiles from a
solvent extract of hand-squeezed Valencia orange juice. The juice was spiked with a
known amount of the labeled internal standard and the juice was extracted with diethyl
ether and subsequently analyzed by GC-MS. Standard curves of the labeled and
unlabeled reference odorants were used to establish a relationship between peak area and
concentration. Odor activity values (OAV, concentrations of the odorants divided by
their odor threshold) were determined to estimate their respective odor contributions.
The highest OAVs were calculated for (s)-ethyl 2-methylbutanoate, ethyl butanoate,
(Z)-3-hexenal, ethyl 2-methylpropanoate, acetaldehyde, and (R)-limonene.
Moshonas and Shaw (2) quantified the volatiles from orange juice using dynamic
headspace GC with a pressurized purge and trap apparatus. Concentrations for each
43


37
components as well as noise. The following ions were monitored for the specific
norisoprenoids: P-cyclocitral, m/z = 137 and 152; P-damascenone, m/z = 175 and 190;
a-ionone, m/z = 177 and 192; P-ionone, m/z = 177 and 192.
Although only a single ion has been shown for each norisoprenoid, two or more
selective ions were employed to detect the presence of specific norisoprenoids. For
example, the selected ion chromatogram using m/z 137 was more intense than that from
m/z = 152 but not as specific for P-cyclocitral. The selected ions of m/z = 177 and 192
were extracted for the determination of a-ionone and P -ionone. Selected ion
chromatograms at m/z 177 provided excellent signal strength and selectivity for P -
ionone. The a-ionone was obviously present at much lower concentrations than P-
ionone. The SIC chromatogram at m/z = 192 (Fig. 4-2) provides more selectivity for a-
ionone but better signal and noise ratio was obtained at m/z 177.
*'\ TIC
m/z = 137
17.97 18.46 1&3E 19.3j19.70mm 20.56 21.2921.49 22.08
M 18.69 1911947 19.70 20.08 20.47
20.87
m/z = 177
21.85
m/z = 192
16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0
Time (min)
Figure 4-2. Comparison between total ion chromatogram and selected ion
chromatograms (SIC) A: P-cyclocitral, B: P-ionone, C: a-ionone.


70
surfactant that has wetting and spreading properties (Helena Chemicals 2002) and was
found to be compatible with solutions of L-methionine.
The objectives for this portion of the study were to examine the effects of a
methionine and Silwet L-77 mixture on a crop plant (eggplant) in terms of yield (both
fruit weight and total yield) and to evaluate this mixture as an insecticide under natural
conditions.
Materials and Methods
Preliminary Investigation of Silwet L-77 and L-methionine
Adult CPBs were obtained from the University of Florida Horticultural Unit,
Gainesville and held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware
cloth (to facilitate cleaning) and held at 27C, 60% relative humidity and 16L/8D
photoperiod in FRIUs. Twenty-four adults were exposed used in each of the 5
treatments, with 4 replicates per treatment (nTOtai=120). Adults were used because of the
lack of sufficient numbers of larvae to test. Excised leaves were dipped in solutions of
deionized H2O containing different concentrations of methionine and Silwett L-77
(0.5% concentration), 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine and
controls of deionized H2O and deionized H2O +Silwet L-77. The additional control
was to determine the possible insecticidal properties of Silwet L-77 alone and to make
sure the addition of this adjuvant did not affect mortality or deter feeding.
Plot Design
Eggplants {Solarium melongena L.,Classic variety) were grown and maintained
at the University of Florida Horticultural Unit, Gainesville, from 18 June to 04 November
2001. Eight, one hundred ft. rows of plants were used for this study, with two rows on
each side consisting of buffer rows and four rows in the middle used for the experiments.


72
Conclusion
The norisoprenoids detected in this study were also detected in other foods that
have been thermally processed e.g.,tea (95), tomato paste (50). Therefore the results of
this study indicate that (3-carotene could be a precursor of (3-cyclocitral and P-ionone in
orange juice. Furthermore it could be a precursor of norisoprenoids during thermal
processing and subsequence storage at relative high temperature and it is reasonable to
assume that any appreciable change in carotenoids content of orange juice will have an
effect on flavor.


109
Mallinckrodt Baker, Inc. 2001. L-methionine Material Safety Data Sheet MSDS
Number M2108. JT Baker Inc. Internet URL:httpV/www jtbakcr.com
/msds/englishhtml/M2108.htm. Accessed April 2004.
Marrone, P.G. and S.C. Macintosh. 1993. Resistance to Bacillus Ihuringiensis and
resistance management, pp. 221-236. IN, P.F. Entwistle, J.S. Cory, M.J. Bailey
and S. Higgs (eds.), An Environmental Biopesticide: Theory and Practice. John
Wiley and Sons New York. 330pp.
McPherson, R.M. and D.C. Jones. 2002. Tobacco Insects: Summary of losses from
insect damage and costs of control in Georgia-2001. University of Georgia
Integrated Pest Management. Internet URL: http://entomology.ent.uga.edu/IPM/
si 01/tobacco.htm. Accessed April 2004.
Melangeli, C., G.A. Rosenthal and D.L. Dahlman. 1997. The biochemical basis for L-
canavanine tolerance by the tobacco budworm Heliothis virescens (Noctuidae).
Proc. Natl. Acad. Sci. USA. 94: 2255-2260.
Mitchell, B.K. 1974. Behavioral and electrophysiological investigations on the
responses of larvae of the Colorado potato beetle (Leptinotarsa decent!ineata) to
amino acids. Ent. Exp. & Appl. 17:255-264.
Mitchell, B.K. and L.M. Schoonhoven. 1974. Taste receptors in Colorado potato beetle
larvae. J. Insect Physiol. 20:1787-1793.
Mittler, T.E. 1967a. Effect of amino acid and sugar concentrations on the food uptake of
the aphid Myzus persicae. Ent Exp. & Appl. 10: 39-51.
Mittler, T.E. 1967b. Gustation of dietary amino acids by the aphid Myzus persicae. Ent.
Exp. & Appl. 10: 87-96.
Munyaneza, J. and J.J. Obrycki. 1998. Development of three populations of
Coleomegilla maculata (Coleptera: Coccinellidae) feeding on eggs of Colorado
potato beetle (Coleptera: Chrysomelidae). Environ. Entomol. 27: 117-122.
Nakajima,N. S. Hiradate and Y. Fujii. 2001. Plant growth inhibitory activity of L-
canavanine and its mode of action. J. Chem. Ecol. 27(1): 19-31.
Nation, J. 2001. Insect Physiology and Biochemistry. CRC Press, Boca Raton. 496pp.
Neishman, O.N. and K. Vulinec. 2001. Florida Crop/Pest Management Profiles:
Eggplant CIR 1264. Pesticide Information Office, Food Science and Human
Nutrition Department, Florida Cooperative Extension Service, Institute of Food
and Agricultural Science, University of Florida. Internet URL:
http://www.edis.ifas.ufl.edu/ BODY_PI045Jitm. Accessed April 2004.


63
Therefore juice carotenoids may be the source of some aromas due to thermal
degradation during processing and subsequence storage.
Twenty-four carotenoids from orange juice were isolated in present study (i.e.,
neoxanthin, P-carotene. a-carotene, and P-cryptoxanthin, see chapter 3). One of the
carotenoids isolated in orange juice, p-carotene, was studied in model aqueous solution to
determine which aroma active compounds could be produced via thermal degradation
and thus unequivocally demonstrate that a carotenoid found in orange juice can act as a
precursor of norisoprenoids. Beta-carotene was chosen because it was commercially
available, relatively inexpensive and should be the most unstable as it cannot be esterified
to improve thermal stability.
Objective
Determine if aroma active norisoprenoids are generated from P-carotene via
thermal degradation using model solutions adjusted to orange juice pH (3.8) and stored at
35C. (Objective 5)
Materials and Methods
Crystallization
Beta-carotene (99% purity, purchase from Acros) was recrystallized before using to
remove aroma active impurities. The method of recrystallization followed that of Schiedt
and Liaaen-Jensen {98) with minor modification. Beta-carotene was dissolved in the
smallest possible volume of petroleum ether (Fisher Scientific, NJ), filtered through glass
wool in a funnel, and ethanol (Fisher Scientific, NJ) was added drop-wise until turbidity
was observed. The mixture was left at room temperature for about an hour and the
temperature was then lowered gradually to 6C (refrigerator) and finally to -20C (deep
freeze) over night or until the crystals formed. The crystals were collected on a fine


45
number of levels and the number of injections per level, this procedure can be time
consuming. However, once a pseudocalibration curve produced, the calculated slope can
be used for other samples of similar matrix. Thus, it is not essential that the standard
addition be employed for each and every sample, but the slope of the pseudocalibration
curve should be checked from time to time. All solutions analyzed must fall within the
linear range of the detector response (90).
Objectives
The primary goal of this study was to quantify the norisoprenoids in orange juice
using static headspace SPME with standard additions and GC-MS. (Objective 4) A
secondary goal was to determine the relative aroma contribution of all four
norisoprenoids to the total orange juice aroma. (Objective 3)
Materials and Methods
Quantification of Norisoprenoids in Orange Juice
The adsorption (amount vs. exposure time) curves for the SPME fiber (50/30 mm
DVB/Carboxen/PDMS) employed in this study was determined by varying exposure time
from 5 to 150 minutes. Since native concentrations were so low, orange juice samples
were fortified with 8.6, 4, 5.4, and 5.27 ppm P-cyclocitral, [3-damascenone, a-ionone, and
P-ionone respectively so that adsorption characteristics could be more accurately
determined. Ten milliliters of the fortified juice were transferred to 40 ml vial with screw
cap coated with Teflon. After 45 min at 40C, the headspace volatiles were extracted
using SPME (as described in Chapter 4)
Headspace SPME and GC-MS were used to quantify norisoprenoids in orange juice
using the standard addition method. Each standard (P-cyclocitral, p-damascenone,


62
Figure 5-6. Mortality of yellow fever mosquito larvae exposed to various
concentrations of L-leucine (nTotai=240). Data were adjusted
using Abbotts formula for control mortality. Note the
overlap in trend lines for all treatments.


15
production (Sang and King 1961). Lack of methionine in the diet of the female may also
explain the transfer of methionine in the ejaculate of the male during fertilization
(Bownes and Partridge 1987). Methionine plays another role in insect biochemistry,
especially in juvenile hormone biosynthesis, inhibitory allatostatins, and storage proteins
known as hexamerins. Audsley et al. (1999) found that in vitro rates of juvenile hormone
synthesis in females of the tomato moth (Mamestra olercea (L.) (Lepidoptera:
Noctuidae)) were dependent on the concentration of methionine present in the incubation
medium. Tobe and Clarke (1985) found a direct relationship between methionine
concentration and juvenile hormone biosynthesis in the cockroach, Diploptera punctata
(Eschscholtz) (Blattodea: Blaberidae), further supporting the idea that methionine plays
an important role in insect biochemistry.
Storage proteins, or hexamerins, act as a storehouse for amino acids that can be
sequestered for later use in the developmental cycle (Pan and Telfer 1996). Many
Lepidoptera have been identified with hexamerins containing high concentrations of
methionine and are metabolized during the last larval stage, and presumably used for egg
production (Wheeler et al. 2000).
Methionine as a potential pesticide has not been overlooked entirely. Tzeng
(1988) tested a methionine and riboflavin mixture and found it successful in controlling
various pests, including the larvae of Culex spp. (Dptera: Culicidae). The mode of
action for this mixture was attributed to a photodynamic reaction and the production of
oxygen rich radicals (Tzeng et al. 1990). Their research led to the use of this methionine
compound as a control agent for sooty mold of strawberry (Tzeng and Devay 1989;
Tzeng et al. 1990) but not as an insecticide.


113
Tzeng, D.D., M.H. Lee, K.R. Chung and J.E. Devay. 1990. Products in light-mediated
reactions of free methionine-riboflavin mixtures that are biocidal to
microorganisms. Can. J. Microbiol. 36(7): 500-506.
Wadsworth, D.J. 1995. Animal health products, pp. 257-284. IN C.R.A. Godfrey (ed.)
Agrochemicals From Natural Products, Marcel Dekker, New York. 424pp.
Walker, T.J., J.J. Gaffney, A.W. Kidder and A.B. Ziffer. 1993. Florida Reach-Ins:
Environmental chambers for entomological research. American Entomologist
39(3): 177-182.
Weinzierl, R., T. Henn and P.G. Koehler. 1998. Microbial insecticides. ENY-275,
Cooperative Extension Service, Institute of Food and Agricultural Services,
University of Florida. 13pp.
Wheeler, D.E., I. Tuchinskaya, N.A. Buck and B.E. Tabahnik. 2000. Hexameric storage
proteins during metamorphosis and egg production in the diamondback moth,
Plutella xylostella (Lepidoptera). J. Insect. Physiol. 46: 951-958.
Womack, M. 1993. The yellow fever mosquito, Aedes aegypti. Wing Beats 5(4): 4.
Wright, R. 1995. Know Your Friends: Wasp Parasites of Greenbugs. Midwest
Biological Control News Online, 11:9.
Young, V.R. and A.E. El-Khoury. 1996. Human amino acid requirements: A re-
evaluation. Food and Nutrition Bulletin 17(3): 191-203.
Zar, J.H. 1999. Biostatistical Analysis, 4th ed. Prentice Hall. New Jersey. 663pp.
Zeh, M., A.P. Casazza, O. Kreft, U. Roessner, K. Bieberich, L. Willmitzer, R. Hoefgen
and H. Hesse. 2001. Antisense inhibition of threonine synthase leads to high
methionine content in transgenic potato plants. Plant Physiol. 127: 792-802.


GC-0 Analysis of (3-Carotene Decomposition at 35C 68
MS Identification 70
Conclusion 72
7 CONCLUSIONS 73
LIST OF REFERENCES 75
BIOGRAPHICAL SKETCH 84
vii


ACKNOWLEDGMENTS
I thank Jim Cuda and Bruce Stevens for giving me the financial and intellectual
freedom that made this work possible. I want to thank Jim for housing me in his lab and
providing the facilities to perform this work, and Bruce for allowing me to take his initial
work and elaborate on it as well as including me as a co-inventor of the research
presented. Most of all, I would like to express my sincere appreciation to Judy Gillmore.
Without her support and help this research would not have been completed. Judy was
integral in every aspect of this endeavor and put up with more than her fair share of my
research. I extend heartfelt thanks to George Gerencser, James Maruniak, Simon Yu,
and Susan Webb for serving as members of my supervisory committee. I would like to
also thank Jim Lloyd, Jerry Butler, and Carl Barfield for all the experiences and
knowledge shared. Finally, I want to express my deepest, eternal gratitude to my fellow
graduate students Jim Dunford and Heather Smith, for providing support and guidance
that only colleagues, intellectual equals, and close friends can give. I can only hope to
repay them for their help by providing the same amount of support for their endeavors as
they did mine.
m


16
Discovery of novel means for controlling various insect pests is one tenant of
IPM. The amino acid methionine, an environmentally safe organic compound, appears to
be a candidate for further study. Before it can be considered for use in controlling insects
pests, several issues must be addressed, including the determination of concentrations
needed to provide effective control, compatibility with current application systems, safety
to nontarget organisms (i.e., beneficial or biological-control agents), and to phytotoxicity.
Research Objectives
Our overall goal was to evaluate the effects of L-methionine, and its amino acid
analogues, on the CAATCH1 system putatively in the midgut/hindgut as a means to
control different insect pests. The working hypothesis is that the L-methionine only
affects the CAATCH1 system and no other system, especially those involving Na+
channels or pumps (i.e., nervous tissue). The L-isomer of methionine was chosen
because of the inability of most insect species to utilize the D-isomer. Ideal targets for
this research are those pests that cause severe damage to agricultural systems and to
human health. Specific objectives were to
Examine the effects of L-methionine as an insecticide on the larvae of M. sexta
(Tobacco homworm), L. decemlineata (Colorado potato beetle) and A. aegypti
(Yellow-fever mosquito) under various conditions
Determine any adverse effects of L-methionine on plant health to ensure its safe
use in a cropping system
Examine the effects of L-methionine on various nontarget insect species to ensure
the environmental safety of L-methionine and thus its compatibility with natural
enemies in the context of IPM.


12
No Amino Acid
Proline
Methionine
Figure 2. The CAATCH1 system identified from the midgut of the tobacco
homworm (modified from Quick and Stevens 2001). In the
presence of no amino acids, ion flow is similar for both K+ and
Na+ (A). With the addition of an amino acid, flows are changed.
When proline is added (B), the transport occurs but the binding of
the amino acid increases the ion flow, notably Na+. However,
when methionine is added (C) transport occurs and the binding of
the amino acid greatly decreases the flow of Na+ while K+ is
increased


55
Figure 5-1. Bioassay setup for yellow fever mosquito larvae exposed to various
concentrations of amino acids and Bti. Jars contained 500mL of
solution and were covered with screen to prevent the escape of
emerging adults.


60
Days of Exposure
Figure 5-4. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Tris-buffered L-methionine (nxotai=240). Data
were adjusted using Abbotts formula for control mortality.
Note the longer exposure because of the bioassay involving
neonates instead of 3rd instars. Note the overlap in some of the
trend lines on Day 1 with the 0.3% L-methionine and 0.5% L-
methionine treatments.


60
40 -
35 -
30 -
£
o
Fresh Pasteurized Pumpout
Orange juice
Figure 5-8. Upper bar norisoprenoids contribute mainly to the total floral category, fresh
= 78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent non-
norisoprenoids including linalool and unknown (LRI = 1255) generated
during thermal processing.
Table 5-12. Norisoprenoids in orange juice and peel oil
Norisoprenoids
Fresh
Pasteurized
Reconstituted
concentrate
Hand
squeezed
Peel oil
P-cyclocitral
X
X
X
X
P-damascenone
X
X
X
X
X
a-ionone
X
X
X
P-ionone
X
X
X
X
X = indicates presence of norisoprenoids in the various samp
es.
Four norisoprenoids, P-cyclocitral, P-damascenone, a-ionone, and P-ionone were
detected in both fresh and pasteurized juice (Table 5-12). Only two norisoprenoids P-
damascenone and P-ionone were detected in reconstituted from concentrate, indicating
that these two compounds could be generated from precursors during thermal evaporation
and/or they were retained by the pulp during the evaporation process.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in sqqpe and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jamefe P. Cuda, Chair
Assistant Professor of Entomology and
Nematology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Bruce R. Stevens, Cochair
Professor Physiology and Functional
Genomics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Professor of Physiology and Functional
Genomics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jane^ E. Maruniak
Associate Professor of Entomology and
Nematology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Simon S.J. Yu
Professor of Entomology and
Nematology


24
As shown in Table 3-1 the absorbance maxima observed exactly matched those published
in the literature or differed at most by 2 nm as in the case of the central peak for (3-
cryptoxanthin. Since the wavelength accuracy of most photodiode array detectors is only
+ 1 nm, the agreement is excellent. Since the carotenoids of interest have the same
elution and spectral characteristics as a-, (3-carotene, a-, (3-cryptoxanthin and neoxanthin,
it is reconfirmed that they are present in orange juice and could potentially serve as
norisoprenoid precursors.
Table 3-1. HPLC retention times, spectral characteristics of orange juice carotenoids
Peak
no.
Carotenoid
RTa
(min)
Observed (nm)
Literature (nm)
Ref.b
Peak
1
Peak
2
Peak
3
Peak
1
Peak
2
Peak
3
1
Valenciaxanthin
5.52
351
369
390
351
369
390
E
2
6.00
371
391
414
3
6.92
420
435
465
4
Neoxanthin
7.35
416
438
468
415
439
467
A
5
11.60
410
431
454
6
12.95
416
438
467
7
Neochrome
14.68
400
422
448
399.5
421.5
447.5
B
8
15.55
408
429
415
9
15.90
383
402
425
10
16.83
411
430
462
11
Cis-violaxanthin
17.60
415
437
464
414
437
464
C
12
Leutoxanthin
19.07
399
418
443
399.5
419.5
441.5
B
13
Mutatoxanthin
20.08
405
429
451
404
427
452
D
14
Lutein
20.92
420
445
471
424.5
445.5
471.5
B
15
Zeaxanthin
23.80
425
450
476
425
450
478
A
16
Isolutein
24.70
418
441
468
418
439.5
467.5
B
17
26.40
429
445
469
18
oc-cryptoxanthin
28.15
420
445
472
420
444
472
D
19
Phytofluene
28.83
331
348
367
331
348
367
A
20
3-cryptoxanthin
31.57
425
451
477
425
449
476
A
3-carotene,5,8:
21
5',8'-diepoxide
34.30
380
400
424
380
400
425
A
22
a-carotene
36.33
420
446
472
420
445
472
D
23
^-carotene
39.28
379
401
425
378
400
425
A
24
3-carotene
39.77
425
451
477
425
450
478
A
aRT = retention time, bA= Britton (81); B= Rouseff et al. (68); C= DeRitter and Prucell
(82); D= Farin et al. (83); E= Curl and Bailey (84).


21
Figure 3-2. Setup for whole plant studies involving tobacco homworm. Top and
portions of the sides were replaced with fine mesh to allow for
airflow and to reduce condensation.


CHAPTER 4
IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING TIME
INTENSITY GC-0 AND GC-MS
Introduction
Early GC-0 studies (4, 6, 7) have shown that many aroma active compounds in
orange juice exist as low-level volatiles that are difficult to detect using typical FID or
MS detectors. Furthermore, these studies also demonstrated that the orange juice
volatiles present in highest concentration have little to no aroma activity. Recently, the
25 most intense aroma active compounds in fresh juice, as determined by dilution
analysis (5), were quantified using isotope dilution analysis (5). Beta-ionone is the only
orange juice norisoprenoid, which has been fully identified (4, 5). Even though it has a
moderately intense aroma, it was not one of the 25 odorants recently quantified using
isotope-dilution analysis (5). Norisoprenoids are volatile C9-C13 fragments with
extremely low aroma thresholds which can be formed from the degradation of C40
carotenoids. This degradation can be the result of in vivo enzymatic reaction, or post
harvest thermal degradation in foods containing carotenoids. They are also observed
from the release of glycosidically bound norisoprenoids which were originally from
carotenoid decompositions as in the case of wine (55). Norisoprenoids have been shown
to have significant aroma impact in fruits, vegetables and spices such as grapes (5),
apples (9), lychee (10) starfruit (11), mango (72), tomato (13), saffron (14) cured tobacco
(75) and black tea (16). During the ripening of red raspberries, a-ionone and (3-ionone
increased to produce the characteristic raspberry aroma (55). In heated apple juice,
26


73
Figure 6-2. Weed Systems, Inc. KQ 3L CO2 backpack back sprayer used for application
of L-methionine and Silwett L-77 solutions. Boom consisted of three
nozzles (middle top and end of each arm). In total, 3L were applied per
treatment every two weeks from 09 July to 31 August 2001.


56
Data Analysis
Sample sizes of all experiments were selected according to the guidelines of
Robertson and Preisler (1991) for optimal sample size and data analysis. Probit analysis
and determination of mean Lethal Concentration (LC50) were performed using PROBIT
Version 1.5 (Ecological Monitoring Research Division, USEPA) after Abbotts
correction for control mortality (Abbott 1925). Probit analysis was performed on
different concentrations (0.1%, 0.3%, 0.5%, 0.7% and 1.0%) of L-methionine, Tris-
buffered L-methionine, D-methionine, Beta-alanine, proline and L-leucine for 24,48, 72
and 168 hours (the end of the trials). Survival data were normalized to the previous value
when control mortality was greater than the treatment mortality, to produce a smoother
trend line. Statistical analyses were performed on the data using Minitab Version 12.
Analysis (Minitab, Inc; State College, PA) of the data included One-way ANOVA and
separation of means using Tukeys Multiple Comparison test (Zar 1999).
Results
Bioassav
Mortality of YFM larvae in both the unbuffered L-and D-methionine trials was
similar with low or no mortality, at the 0.1% concentrations (Figures 5-2 and 5-3). The
0.3% concentration had lower mortality with D-methionine (45%) than L-methionine
(75%) and greater than 80% mortality for the 0.5% concentration for both isomers.
Higher concentrations of both D-and L-methionine forms produced 100% mortality of
the larvae within 2 days after treatment.
Greater than 40% mortality was observed for the buffered 0.1% L-methionine
concentration with complete mortality for the remaining treatments within 5 days of


9
proteins (produced by the assimilation of L-canavanine) into newly synthesized proteins;
the proteases involved do not efficiently degrade enough to prevent some damage from
occurring in the insect (Rosenthal and Dahlman 1986; 1988).
Surprisingly, L-canavanine also was shown to increase the effectiveness of
Bacillus thuringiensis in vivo by altering membrane properties, mainly gut permeability,
and active transport in the midgut of the THW (Felton and Dahlman 1984). However,
despite the possible synergistic relationship between the relatively safe Bt product and
this amino acid, no further research has been conducted to evaluate the combination for
future commercial use.
Other species of insects have also been tested for susceptibility to canavanine with
a variety of results. Larvae of Drosophilia melanogaster Meigen (Dptera:
Drosophilidae) showed no deleterious response to lower concentrations of canavanine,
but showed mortality increased at concentrations over 1,000 ppm (Harrison and Holiday
1967). Lower concentrations also were ineffective against adult Pseudosarcophaga
affinis (Fallen) (Dptera: Calliphoridae), with no effect on oocyte development (Hegdekar
1970). Dahlman et al. (1979) examined four species of muscoid flies and found greater
than 70% mortality at the higher concentration (800 ppm) and decreased pupal weights as
concentrations of canavanine increased.
Despite the toxicity of canavanine to some insects, others have evolved
detoxifying mechanisms to deal with high concentrations of this compound. Rosenthal et
al. (1978) attributed the detoxification of canavanine in the bruchid Caryedes brasiliensis
Thunberg (Coleptera: Bruchidae) to the beetles ability to convert canavanine to
canaline, another toxic amino acid. Canaline is metabolized through reductive
deamination to homoserine and ammonia, with the overall result being the detoxification


72
73
74
75
76
77
78
79
80
81
82
83
84
81
Lee, H. S.; Castle, W. S. Seasonal Changes of Carotenoid Pigments and Color in
Hamlin, Early gold, and Budd Blood Orange Juices. J. Agrie. Food Chem. 2001,
49, 877-882.
Pupin, A. M.; Dennis, M. J.; Toledo, M. C. F. HPLC analysis of carotenoids in
orange juice. Food Chem. 1999, 64, 269-275.
Philip, T.; Chen, T. S.; Nelson, D. B. Liquid chromatographic profiles of major
carotenoid esters in commercially processed California Navel and Valencia
orange juice concentrates. J. Chromatogr. 1988, 442, 249-265.
Lee, H. S.; Castle, W. S.; Coates, G. A. High-performance liquid chromatography
for the characterization of carotenoids in the new sweet orange (Earlygold) grown
in Florida, USA. J. Chromatogr., A 2001, 913, 371-377.
Mouly, P. P.; Gaydou, E. M.; Lapierre, L.; Corsetti, J. Differentiation of several
geographical origins in single-strength Valencia orange juices using quantitative
comparison of carotenoid profiles. J. Agrie. Food Chem. 1999, 47, 4038-4045.
Noga, G.; Lenz, F. Separation of citrus carotenoids by reversed-phase high-
performance liquid chromatography. Chromatographia 1983,17, 139-142.
Marais, J. l,L6-Trimethyl-l,2-dihydronaphthalene (TDN): a possible degradation
product of lutein and b-carotene. South African J. Enol. Viticulture 1992,13, 52-
55.
Chen, C. S.; Shaw, P. E.; Parish, M. E. Orange and Tangerine Juice. In Fruit
Juice Processing Technology, S. Nagy; C. S. Chen and P. E. Shaw, Eds.;
AGScience, Inc.: Aubumdale, 1993; pp 110-165.
Stewart, I.; Wheaton, T. A. Continuous flow separation of carotenoids by liquid
chromatography. J. Chromatogr. 1971, 55, 325-336.
Britton, G. UV/Visible Spectroscopy. In Carotenoids, Vol. 1A: Isolation and
Analysis; G. Britton; S. Liaaen-Jenson and H. Pfander, Eds.; Birkhauser Verlag:
Basel, Boston, Berlin, 1995; pp 13-62.
De Ritter, E.; Purcell, A. E. Carotenoid analytical Methods. In Carotenoids as
Colorants and Vitamin A Precursors; J. C. Bauemfeind, Ed.; Academic Press:
London, 1981; pp 815-923.
Farin, D.; Ikan, R.; Gross, J. The carotenoid pigments in the juice and flavedo of a
mandarin hybrid (Citrus reticulata) cv. Michal during ripening. Phytochemistry
1983, 22, 403-408.
Curl, A. L.; Bailey, G. F. The carotenoids of Navel oranges. J. Food Sci. 1961,
26, 442-447.


33
determining norisoprenoids, as they elute fairly late, one of the secondary objectives in
the overall study was to determine the relative contribution of norisoprenoids to the total
aroma of orange juice. Solvent extracted juice samples would have been unsatisfactory
for this purpose for the above stated reason.
The application of headspace SPME to flavor volatile compounds has been
employed in the study of flavor volatiles in tomato and strawberry fruits using PDMS,
PDMS/DVB, and Carbowax/DVB coated fiber (67), in orange juice using a PDMS
coated fiber (64), a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65),
PDMS and polyacrylate fiber (66). The partition coefficients of the polymeric coatings
for the analyses differed markedly. For example, terpenes such as a-pinene, (3-myrcene,
y-terpinenes, and limonene are all nonpolar, and were extracted to a higher degree into
the nonpolar PDMS coating (66). Corresponding PDMS extracted the least amount of
the more highly polar volatiles, PDMS/DVB and Carbowax/DVB had partition
coefficients higher than that of PDMS for the most polar molecules (67). The Carboxen-
PDMS fiber coating was more selective for terpenes than early eluting alcohols and
aldehydes (6). Polyacrylate was more effective in extracting highly polar compounds
such as methanol and ethanol (66). Due to the wide range of volatile compounds from
orange juice and for the increased fiber capacity, the headspace volatiles in this study
were extracted and concentrated using the 50/30 mm DVB/Carboxen/PDMS coating on a
2 cm StableFlex fiber.
In examining adsorption curves for (3-cyclocitral, (3-damascenone and a- and
(3- ionone on the chosen fiber (see Fig. 5-1) it was concluded that 45 min. represents a
rough compromise for all four analytes between minimal exposure time and maximum


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE
USING TIME INTENSITY GC-0 AND GC-MS
By
Kanjana Mahattanatawee
May 2004
Chair: Russell L. Rouseff
Major Department: Food Science and Human Nutrition
Numerous analytical studies have quantified the major volatiles in orange juice in
an effort to duplicate this aroma. However, when combined, the resulting aroma does not
duplicate that of orange juice, suggesting that important aroma components were missing.
Citrus carotenoids have been studied primarily for their role as pigments and have
generally been ignored as a source of aroma compounds. Carotenoids can be degraded
into smaller (C9-C13), more volatile products called norisoprenoids. Norisoprenoids have
been identified as aroma impact in foods containing carotenoids (i.e. tea, grapes, tomato,
and saffron). Therefore, carotenoid-decomposition may be responsible for a portion of
orange juice aroma and also aroma changes associated with thermal processing or
elevated temperature storage.
Three norisoprenoids, (3-cyclocitral, (3-damascenone, a-ionone, were fully
identified in fresh and pasteurized orange juice for the first time. Beta-ionone was also
detected but had been previously identified. Identification was based upon SPME
xii


Mean Head Capsule Width (mm)
64
(Error Bars @ 95%; F 2.5
2
1.5
1
0.5
0
Figure 5-8. Mean head capsule widths of yellow fever mosquito larvae
exposed to various Tris buffered (7.0 pH) concentrations of L-
methionine (niotai-320). Proline (1.0%) and Bti were included
for comparison as positive and negative controls. Error bars
denote 2 SE. Bars within treatments having the same letter are
not statistically different (Tukeys MST, PO.OOl).


87
Results
Coleomesilla maculata
There was virtually no difference between the control and treatment groups for
either the artificial or natural diet tests after correction for control mortality. Mortality
was slightly higher for the control groups than the 1.0% L-methionine treatment (Figures
7-1 and 7-2). Further analysis was not necessary because of the identical numbers.
Neochetina eichhorniae
Total mortality for the treatments was less than 20% for all treatments, with the
individual treatments having similar results (Figure 7-4). Feeding damage ranged
between 2,000 and 4,000 scars per treatment and an average of 10.7 to 16.9 scars per
survivor during the course of the experiment (Figure 7-5). No statistical differences were
observed between the treatment and control groups
Lvsiphlebus testaceipes
In total, 188 and 232 aphid mummies with exit holes were found on treatment and
control plants, respectively. Means for each treatment were not statistically different for
each collection period or overall based on One-way ANOVA (Figure 7-6) with the only
exception being the second and last collection period.
Discussion
In general, L-methionine did not have the same toxic effect on the non-target
organisms tested when compared to the pest species exposed to the compound in
previous chapters. The pink spotted ladybird beetle adults actually showed the least
amount of susceptibility to L-methionine. Survival of the adult beetles was higher in the
1.0% L-methionine treatments than the control for both the artificial and natural diet


76
(Error Bars @ 95%; F(1)3fl56 =2.6626, F =0.37137; P= 0.77377)
Control 0.1% 0.5% 1.0%
(n=195) (n= 191) (n=175) (n=174)
Control 0.1% 0.5% 1.0%
(n=T95) (n=191) (n=175) (n=174)
Figure 6-4. Effects of L-methionine and Silwett L-77 on eggplant yield
(A) and mean weight in grams of fruit (B) from 09 June to
31 August 2001. Error bars denote 2 SE. There was no
statistical difference for either eggplant yield or mean
eggplant weight (Tukeys MST, Z^O.05).


7
oxidative cleavage of the intact carotenoid. After further enzymatic transformation steps,
the primary cleavage products are converted into reactive Cio to C13 fragments of the
initial carotenoid. These volatile fragments can be stabilized and made nonvolatile by
glycosylation (which involves glycosyltransferases of those norisoprenoid compounds
possessing a hydroxyl group) (35, 36).
Glycosilation stabilizes and solubilizes norisoprenoids in plant systems.
Degradation of the glycoconjugates librates the potent volatile and can produce profound
aroma changes. This process can be acid-catalyzed (e.g., during fruit processing) (9) or
enzymatic (e.g., during fermentation) (37). Another important class of precursors is the
polyols, which upon (allylic) elimination of water is transformed into volatile forms. An
example is the reactive allyl-l,6-diol that, under gentle reaction conditions (natural pH,
room temperature), is converted into isomeric theaspiranes (38). A third class of
carotenoid-derived aroma precursor is glucose esters (e.g., Cio-compounds derived from
the central part of the carotenoid chain, which is left after the cleavage of the endgroups)
that gives rise to isomeric marmelo lactones, key aroma constituents of quince fruit (see
Fig. 2-4) (39).
Norisoprenoids have been shown to have significant aroma impact in fruits (apple,
mango, grape, starfruit, lychee, passion fruit, nectarine, etc.) (9-12, 19, 36, 40) vegetables
(tomato (41)), spices (saffron (14) and paprika (42)), leaf products (tobacco (43) and tea
(44)) as well as roses (45), wine (46) and oak wood (47).
Apple
Beta-Damascenone is a potent aroma compound found in a variety of natural
products, with a threshold of 0.002 pg/L in water (48). Eight separate (3-damascenone


LIST OF REFERENCES
1 Moshonas, M. G.; Shaw, P. E. Quantitative analysis of orange juice flavor
volatiles by direct-injection gas chromatography. J. Agrie. Food Chem. 1987, 35,
161-165. *
2 Moshonas, M. G.; Shaw, P. E. Quantitative determination of 46 volatile
constituents in fresh, unpasteurized orange juices using dynamic headspace gas
chromatography. J. Agrie. Food Chem. 1994, 42, 1525-1528.
3 Tonder, D.; Petersen, M. A.; Poll, L.; Olsen, C. E. Discrimination between freshly
made and stored reconstituted orange juice using GC odor profiling and aroma
values. Food Chem. 1998, 61, 223-229.
4 Hinterholzer, A.; Schieberle, P. Identification of the most odor-active volatiles in
fresh, hand-extracted juice of Valencia late oranges by odor dilution techniques.
Flavour Frag. J. 1998, 13, 49-55.
5 Buettner, A.; Schieberle, P. Evaluation of aroma differences between hand-
squeezed juices from Valencia late and Navel oranges by quantitation of key
odorants and flavor reconstitution experiments. J. Agrie. Food Chem. 2001, 49,
2387-2394.
6 Bazemore, R.; Goodner, K.; Rouseff, R. Volatiles from unpasteurized and
excessively heated orange juice analyzed with solid phase microextraction and
GC-olfactometry. J. Food Sci. 1999, 64, 800-803.
7 Marin, A. B.; Aeree, T. E.; Hotchkiss, J. H.; Nagy, S. Gas chromatography-
olfactometry of orange juice to assess the effects of plastic polymers on aroma
character. J. Agrie. Food Chem. 1992, 40, 650-654.
8 Kotseridis, Y.; Baumes, R. Identification of impact odorants in bordeaux red
grape juice, in the commercial yeast used for its fermentation, and in the produced
wine. J. Agrie. Food Chem. 2000, 48, 400-406.
9 Zhou, P. G.; Cox, J. A.; Roberts, D. D.; Aeree, T. E. Beta-Damascenone
precursors in apples. Prog. Flavour Precursor Stud. Proc. Int. Conf. 1993, 261-
273.
10 Ong, P. K. C.; Aeree, T. E. Gas chromatography/olfactory analysis of Lychee
(Litchi chinensis Sonn.). J. Agrie. Food Chem. 1998, 46, 2282-2286.
75


85
Neochetina eichhorniae
Adults of the MWHW were used in this study since the larvae and pupae are
buried deep in plant tissue and therefore not likely to come into contact with methionine
that could be present in a body of water. Specimens were supplied by Hydromentia, Inc.
(Ocala, FL), from areas around South Florida. Weevils were maintained following the
procedures outlined by Haag and Boucias (1991), with small petri dishes fitted with
moistened filter paper and freshly cut water hyacinth leaves. Water hyacinth plants were
collected from Lake Alice on the campus of the University of Florida and maintained in
the University of Florida, Department of Entomology and Nematology greenhouse.
Treatments consisted of cut leaves dipped in deionized H2O (control) or solutions
containing 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine or 1.0% proline.
Prior to weevil exposures, each leaf was inspected for feeding scars or damage
and noted to ensure the counts were based on current feeding. Each treatment consisted
of 4 replicates with n=5 per replicate (n=20 per treatment and total n=100). Weevils and
hyacinth leaves were held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a
hardware cloth (to facilitate cleaning) and maintained at 27 C, 60% relative humidity
and 16L/8D photoperiod in FRJUs. Fresh leaves were provided every 4 days; exposed
leaves were preserved in sealed plastic bags and placed in a refrigerator until scars could
be counted. Feeding damage was determined (with the use of an Olympus Tokyo Model
213598 stereo microscope) by the total number of scars present with each counted scar
marked with a fine tipped permanent marker (Figure 7-3).
Statistical analyses of the weevil data were performed using Minitab Version 12
(Minitab, Inc.; State College, PA). Feeding scars on control and treatment leafs were


78
37 Gunata, Z.; Dugelay, I.; Sapis, J. C.; Baumes, R.; Bayonove, C. Role of enzymes
in the use of the flavor potential from grape glycosides in winemaking. Prog.
Flavour Precursor Stud. Proc. lnt. Conf. 1993, 219-234.
38 Winterhalter, P. Oxygenated C13-Norisoprenoids: Important Flavor Precursors. In
Flavor Precursors: Thermal and Enzymatic Conversions', R. Teranishi; G. R.
Takeoka and M. Guntert, Eds.; American Chemical society: Washington, DC,
1992; pp 98-115.
39 Lutz, A.; Winterhalter, P. Isolation of additional carotenoid metabolites from
quince fruit (Cydonia oblonga Mill.). J. Agrie. Food Chem. 1992, 40, 1116-1120.
40 Aubert, C.; Ambid, C.; Baumes, R.; Guenata, Z. Investigation of Bound Aroma
Constituents of Yellow-Fleshed Nectarines (Prunus prsica L. Cv. Springbright).
Changes in Bound Aroma Profile during Maturation. J. Agrie. Food Chem. 2003,
51, 6280-6286.
41 Buttery, R. G.; Teranishi, R.; Ling, L. C.; Flath, R. A.; Stem, D. J. Quantitative
studies on origins of fresh tomato aroma volatiles. J. Agrie. Food Chem. 1988, 36,
1247-1250.
42 Wilkins, C. K. Paprika: relationships between aroma profile data and both GC and
HPLC data. Lebensm-Wiss.-Technol. 1992, 25, 212-218.
43 Enzell, C. R.; Wahlberg, I. Tobacco isoprenoids precursors of important aroma
constituents. Pure Appl. Chem. 1990, 62, 1353-1356.
44 Kawakami, M.; Ganguly, S. N.; Banerjee, J.; Kobayashi, A. Aroma composition
of oolong tea and black tea by brewed extraction method and characterizing
compounds of darjeeling tea aroma. J. Agrie. Food Chem. 1995, 43, 200-207.
45 Suzuki, M.; Matsumoto, S.; Fleischmann, H.-P.; Shimada, H.; Yamano, Y.; Ito,
M.; Watanabe, N. Identification of (3-damascenone Progenitors and Their
Biogenesis in Rose Flowers (Rosa damascena Mill.). In Carotenoid-Derived
Aroma Compounds', P. Winterhalter and R. L. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002; pp 89-101.
46 Kotseridis, Y.; Baumes, R. L.; Skouroumounis, G. K. Quantitative determination
of free and hydrolytically liberated .beta.-damascenone in red grapes and wines
using a stable isotope dilution assay. J. Chromatogr., A 1999, 849, 245-254.
47 Sefton, M. A.; Francis, I. L.; Williams, P. J. Volatile norisoprenoid compounds as
constituents of oak woods used in wine and spirit maturation. J. Agrie. Food
Chem. 1990, 38, 2045-2049.
48 Ohloff, G. Importance of minor components in flavors and fragrances. Perfum.
Flavor. 1978, 3, 11-12, 14-16, 18-22.


61
Figure 5-5. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Proline (nTOtai=240). Data were adjusted using
Abbotts formula for control mortality. Note the overlap of
trend lines for all treatments except the 0.7% L-methionine and
1.0% L-methionine treatments.


95
body of the host, and not through direct contact with the foliage where the compound was
applied. There is a possibility for the parasitoid having higher methionine requirements;
based on filarial worm infected Aedes aegypti (L.) (Dptera; Culicidae) females and the
associated drop in methionine levels in the haemolymph (Jaffe and Chrin 1979). This
makes alternatives such as L-methionine safe for use around beneficial insects like the
greenbug parasitoid.
Overall, the results indicate that the PSLB (C. maclala), the MWHW (JV.
eichhorniae) and the GBP, (L. tes tace ipes) were not adversely affected by exposure to
L-methionine in excess concentrations in a variety of artificial and natural diets.
Survivorship and feeding rates were not statistically different between control and
treatment groups for each species. From these data, it can be concluded that
L-methionine is safe for use with beneficial insects and could be considered biorational
in that it showed no adverse effects on non-target species. It also should be stressed that
additional testing on other beneficial insects would be, on a case by case basis, necessary
to examine the safety and biorational qualities of L-methionine.


47
Feeding rates of CPB also were found to be statistically different among treatments
(Figure 4-4). Three distinct groups were observed with the first group containing the
Control and 0.1% L-methionine treatments while the second group of the 0.1% L-
methionine and 0.3% L-methionine, treatments were found to be statistically the same.
The 0.5% L-methionine, 0.7% L-methionine, 1.0% L-methionine and Btt treatments were
statistically different from the other groups. Overlap occurred with the proline treatment
across all groups indicating no statistical difference with the rest of the treatments.
Preference Tests
The amount of Control and 1.0% L-methionine leaf tissue consumed during the
preference tests was found not to be statistically different (Figure 4-5). In addition, the
mean head capsule width (i,e development) showed no relationship with either treatment
based upon the low correlation coefficients.
Discussion
The 1.0% L-methionine concentration produced the same larval mortality, feeding
and developmental rates for CPB, as did the Btt treatments (Figures 4-1,4-3, and 4-4).
The 0.3% L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments took 4
days longer for complete control (Figure 4-1), but were statistically different for the
developmental rates for the same treatments (Figure 4-3). As was the case with the THW
survivorship, the 0.1% L-methionine concentration was not different from that of the
Control. This may indicate a threshold of methionine that can be tolerated by the THW,
and CPB to some extent, evidenced by the low mortality observed for this treatment.
The Preference tests did not indicate any preference of leaf disks with or without
L-methionine. The high mortality (90%) of the CPB larvae could be explained by a


22
well resolved (i.e., peaks 8-10) and make accurate identification difficult. Lutein and
zeaxanthin (peaks 14 and 15, respectively) are usually difficult to resolve as they differ
only in the position of a single double bond in one of the terminal rings. These pigments
can be separated on a ZnC03-MgC03 column and the separation requires several hours
(80) but they are completely resolved in this chromatographic system. However lutein is
barely resolved from mutatoxanthin (peak 13) even though mutatoxanthin contains an
extra 5,8 epoxide group. Phytofluene and a-cryptoxanthin (peak 19 and 18, respectively)
were not well resolved chromatographically, but could easily be separately quantified
using different monitoring wavelengths as their respective absorbance maxima differ by
approximately 100 nm.
Retention Time (min)
Figure 3-3. Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH,
MTBE ternary solvent gradient. Spectral characteristics for each numbered
peak are summarized in Table 3-1. See HPLC experimental section for
additional details.


53
requirements of methionine in the amounts of 0.0007mg/ml for the YFM. This amino
acid also is considered essential for other species of mosquito in untraceable (in those
studies) amounts (Chen, 1958; Singh and Brown, 1957). Given the high alkalinity found
in the midgut of the YFM as well as other mosquito species, this physiological condition
indicates the possibility of the presence of the CAATCH1 system in larval mosquitoes
(Dadd, 1975).
The purpose of this portion of the study was to examine the survival and
development of YFM larvae exposed to water treated with excess L-methionine (adults
were not tested given the feeding nature). In addition to L-methionine, other amino acids
were tested in an effort to see if their response (i.e., survivorship) was similar CAATCH1
responses to methionine found by Feldman et al. (2000).
Materials and Methods
Bioassav
The bioassay experiments consisted of six treatments (control, 0.1%, 0.3%, 0.5%,
0.7% and 1.0%) each with four replicates. Both L-methionine and D-methionine were
tested along with proline, Beta-alanine and L-leucine to examine the other amino acids
that were found to be reactive to the CAATCH-1 system (Feldman et al., 2000).
Bt-isrealiensis (Aquabac @ a rate of 2.3 mL/m2; Biocontrol Network, Brentwood, TN)
and proline also were included in some trials of L-methionine to allow for comparison of
both positive and negative effects. Amino acids were weighed using a Denver
Instruments Co. XD2-2KD digital scale and added to glass quart jars containing 500ml of
deionized FLO. Concentrations were based on the proportion of lg/100ml for a 1%
solution and for corresponding concentrations. Solutions were allowed to sit at room


CHAPTER 7
CONCLUSIONS
Four norisoprenoids, P-cyclocitral, p-damascenone, a-ionone and P-ionone were
identified in orange juice using headspace SPME, GC-O, GC-FID and GC-MS. Three of
them, P-cyclocitral, P-damascenone, a-ionone were identified and confirmed by GC-MS
for the first time. Their concentrations in fresh orange juice were determined using
SPME with standard addition technique. Odor activity values (OAV) were calculated
using published threshold values. Calculated OAV values suggest that P-ionone provided
the greatest contribution to total floral aroma in orange juice compared to the other three
norisoprenoids. The concentration of P-damascenone increased almost 10 fold after
thermal processing, indicating there are thermally unstable precursors which generate
P-damascenone at elevated temperatures. All four of the norisoprenoids in orange juice
contribute 8-10% floral aroma to the total aroma quality of the orange juice and are the
major contributors (60-80%) in the floral category.
Several carotenoids were identified using HPLC with photodiode array detection.
Twenty-four carotenoids were separated as distinct peaks and sixteen of these peaks were
identified based on their spectral characteristics, relative elution order compared to
literature values and authentic standards. The identified carotenoids include: a-carotene,
P-carotene, a-cryptoxanthin, P-cryptoxanthin and neoxanthin, which are known as
norisoprenoid precursors. These specific carotenoids were of interest because they
73


61
Conclusion
Concentrations of four orange juice norisoprenoids were determined using SPME
with the standard addition method. The concentrations of P-cyclocitral, P-damascenone,
a-ionone, and P-ionone in fresh orange juice were 145, 0.09, 47, and 83 p.g/L
respectively. The OAV (determined by dividing the analytical concentration by the
aroma threshold) of p-cyclocitral, p-damascenone, a-ionone, and p-ionone were 25, 45,
118, and 11857 respectively. The OAV values suggest that P-ionone provides the
greatest aroma contribution compared to the other norisoprenoids. The concentration of
p-damascenone increaded with thermal processing, indicating that there are precursors in
juice which generate P-damascenone during elevated temperatures. Combined, the four
norisoprenoids contribute 8-10% of the total aroma impact. The norisoprenoids have a
general floral character and contribute the majority (60-80%) of the floral character to
orange juice.


81
Overall, it appears that L-methionine can be used in a natural setting to control
CPB larvae without affecting crop production. The adjuvant Silwett L-77 worked well
with L-methionine in controlling CPB larvae but not the adults. The lack of effectiveness
on the adults may be attributed to their ability to stop feeding and living off of reserves
acquired during the larval stage until suitable food sources can be found. It is unknown if
L-methionine, alone or in combination with Silwett L-77 adversely affects fecundity of
the adults.


31
P-citronellol, carvone, terpin-4-ol, geranial, and neral). The panelists sniffed the effluent
of aroma standard from GC-0 with optimum positioning and breathing technique. The
intensity of each standard was recorded on a sliding scale (varying from none to strong
intensity) and panelists were provided verbal descriptors of aroma quality. For additional
experience, the extract of aroma volatiles from commercial orange juice was provided to
panelists under identical conditions. Panelists were accepted on they demonstrated an
ability to replicate aroma peak times for at least 80 % of the components in the test
mixture.
Two trained panalists evaluated the volatiles of orange juice (extracted by SPME)
in duplicate, thus producing four individual time-intensity aromagrams. Average
intensity from the four runs was calculated for each odorant. If no peak was detected in a
run, its value was treated as missing, not zero. An indication of aroma activity with
similar aroma descriptors, at the same retention was required from at least half the panel
results before a peak could be considered aroma-active. Averaged time-intensity
aromagrams were constructed by plotting average intensity versus retention time.
Chromatograms and aromagrams were recorded and integrated using Chromperfect
version 5.0, Justice Laboratory Software (Palo Alto, CA). Identification of the aroma-
active components was based on the combination of sensory descriptors, standardized
retention indices, and identification confirmed by comparison with standards and GC-MS
spectra.
Gas Chromatography-Mass Spectrometry
Orange juice headspace volatiles were extracted by SPME and introduced to the
GC-MS. Volatiles were separated and analyzed using a Finnigan GCQ ion trap mass


57
Table 5-8. Aroma active compounds in orange juice grouped by
sulfury/solventy/medicine
Compounds2
Description
LI
ill
Relative
intensity
ZB-5
DB-
wax
Acetaldehyde
Fresh alcohol
445
732
6C, 6d
Carbon disulfide
Sulfur, fermented cabbage
678
6C, 7d, 6e
Dimethyl sulfide
Solventy, plastic
691
6C, 6d, 4e
Dimethyl disulfide
Plastic
772
1074
4C, 4d, T
Unknown
Fermented, sulfur
818
5C, 6d
2-methyl-3-furanthiol
Meaty, vitamin, medicine
865
1305
7C, 7d, 6e
4-mercapto-4-methyl-2-pentanone
Sulfury, grapefruit
944
1389
7C, 5d
Dimethyl trisulfide
Sulfur, sweaty
968
1392
3C, 5d,8e
4-mercapto-4-methyl-2-pentanol
Sweaty .grapefruit,guava
1039
7C, 7d, T
Unknown
Solventy
1167
5d
Dimethyl tetrasulfide
Sulfury, musty
1225
6e
Table 5-9. Aroma active compounds in orange juice grouped by floral
Compounds2
Description
LI
RI
Relative
intensity
ZB-5
DB-
wax
Linaloolb
Floral
1094
1551
8C, 8d, 5e
(3-cyclocitralb
Mild floral, hay-like
1214
1632
6C, 4d
Unknown
Tobacco,sweet, floral
1255
6e
(3-damascenoneb
Tabacco, apple, floral
1383
1829
7C, 8d, 8e
a-iononeb
Floral
1426
1863
OO
o
00
o.
P-iononeb
Floral, raspberry
1484
1951
8C, 9d, T
Table 5-10. Aroma active compounds in orange juice grouped by sweet/fruity
Compounds2
Description
L
RI
Relative
intensity
ZB-5
DB-
wax
Ethyl-2-methylpropanoate
Sweet, fruity
758
966
6C, 6d
Ethyl butyrateb
Sweet, fruity
795
1034
4C, 6d, 8e
Ethyl-2-methylbutyrate
Sweet, fruity
846
1051
5C, 6d
Ethyl hexanoateb
Sweet
994
1242
6C, 7d
a-terpinyl acetate
Sweet
1349
1663
6c,6d


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL
CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL
IMPORTANCE
By
Lewis Scotty Long
May, 2004
Chair: James P. Cuda
Cochair: Bruce R. Stevens
Major Department: Department of Entomology and Nematology
Integrated pest management (IPM) strategies were developed in an effort to
control pests with fewer pesticides. However, because of the misuse of pesticides and the
failure to adopt IPM practices pesticide use is higher than ever. An alternative to
conventional broad-spectrum pesticides is the use of biorational compounds; those that
pose minimal risk to the environment and are specific to the target pests.
The recent discovery of the CAATCH1 system in the midgut of the tobacco
homworm (THW), Manduca sexta, has revealed a novel means to control certain insect
pests. This membrane protein works in alkaline conditions as both an amino acid
transporter and also independently as a cation channel. However, the amino acid
L-methionine blocks amino acid transport thus altering the ion flow.
xi


68
mechanisms and many are the result of a combination of different pesticide classes. The
CAATCH1 system is one that could be exploited in cases where the only alternative is
applying different or higher rates of pesticides to break resistance. Further research is
needed to determine compatibility of Bti and L-methionine for cases in which resistance
is observed in natural populations. Given the safety of L-methionine and the similar time
required for 100% mortality (when compared to Bti), this compound could represent a
viable alternative to traditional biorational compounds used in the management of the
YFM or other susceptible pest mosquito species.


34
(Error Bars @ 95%; F(o.os)7,i52=2 .37, F=\8.2; P<0.001)
Figure 3-11. Total leaf area consumed by tobacco homworm larvae exposed to
excised eggplant leaves treated with various concentrations of L-
methionine (nrotai=320). Proline (1.0%) and Btk were included for
comparison as positive and negative controls. Error bars denote 2 SE.
Bars within treatments having the same letter are not statistically
different (Tukeys MSTP, P<0.001).


36
Discussion
The initial studies involving the high concentrations of L-methionine (i.e., 3.0-10.0%,
which are outside the range normally encountered in nature) showed that a concentration
of 1.0% L-methionine was sufficient enough to provide good control of THW larvae
reared on both artificial and natural diets. The 0.1%L-methionine concentration
remained similar to that of the control for developmental and feeding trials (Figure 3-9),
indicating a level of methionine that can be tolerated to some extent, as seen in the low
mortality of this treatment. This is in stark contrast to the mortality seen in the excised
leaf trials in which the same concentration had over 60% mortality (Figure 3-6). One
possible explanation could be the amount of L-methionine present on the leaf disk being
low enough and ingested at a slower rate than that of the whole leaf, which was left in the
chamber with the larvae until the leaf was either completely consumed or too wilted for
the larvae to ingest.
The preference tests did show some preference towards control leaf disks over the
1.0%L-methionine treated disks as seen in the correlation analysis of the diet consumed
and the mean head capsule width of the larvae. Despite the lack of a statistical difference
between the amount consumed, the larvae could have fed on the treated disks and then
switched to the control disks based on a physiological cue. It is unclear if THW larvae
possess specialized sensory structures to detect amino acids like those found in other
Lepidoptera (Beck and Henee 1958; Dethier and Kuch 1971; Schoonhoven 1972), but the
possible switch from the methionine rich treatment to the control leaf disks does indicate
some sort of mechanism for detection. Del Campo and Renwick (2000) found THW
larvae were induced to feeding on plants outside of their normal diet when the plants


10
storage conditions. Additional flavor compounds in saffron were formed upon cooking
of the spice (52). Aroma isolates of saffron have been prepared by simultaneous
distillation extraction (SDE) at pH 2.6 as well as liquid-liquid extraction using pentane:
diethyl ether (1:1) as solvent. Aroma activity and relative aroma strength was determined
using aroma extract dilution analysis (AEDA). Compounds with high FD-factors were
safranal and 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one as well as linalool and
isophorone. The 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one was only detected
in the SDE isolate and not in the liquid-liquid extract. This result shown the presence of
certain forms of precursors, which upon heat treatment are converted into the aroma
compound 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one (55).
Grape and Wine
Norisoprenoids are important aroma constituents of grape and wine. They are
thought to arise from carotenoid breakdown; and occur in grapes as glycosidically bound
precursors. The major carotenoids in grape are P-carotene and lutein, representing nearly
85% of the total carotenoids. These are accompanied by minor carotenoids such as
neoxanthin, violaxanthin, lutein-5,6-epoxide, zeaxanthin, neochrome, flavoxanthin, and
luteoxanthin (54). Grape carotenoids decrease progressively during maturation, with a
concomitant increase of the volatile compounds. This degradation would occur during
berry metabolism either enzymatically or by chemical pathway in acid medium (54, 55).
This would account for the presence of volatile compounds, such as p-ionone and
p-damascenone, identified in grape (56) and possibly originating in carotenoids (36).
Many norisoprenoids occur in grapes as glycosidic precursors. Enzymatic and acid


57
Figure 5-2. Mortality of yellow fever mosquito larvae exposed to various
concentrations of L-methionine (nTOtai=240). Data were
adjusted using Abbotts formula for control mortality.


no
Nester, E.W., L.S. Thomashow, M. Metz and M. Gordon. 2002. One Hundred Years of
Bacillus thruingiensis: A Critica Scientific Assessment Report from the
American Academy of Microbiology. 16p.
Onifade, A.A., O.O. Oduguwa, A.O. Fanimo, A.O. Abu, T.O. Olutunde, A. Arije and G.
M. Babatunde. 2001. Effects of supplemental methionine and lysine on the
nutritional value of housefly larvae meal (Musca domestica) fed to rats. Biores.
Technol. 78:191-194.
Palumbo, R.E. and D.L. Dahlman. 1978. Reduction of Manduca sexta fecundity and
fertility by L-canavanine. J. Econ. Entomol, 71:674-676.
Pan, M.L. and W.H. Telfer. 1996. Methionine-rich hexamerin and arylphorin as
precursor reservoirs for reproduction and metamorphosis in female Luna moths.
Arch. Insect Biochem. Physiol. 32:149-162.
Patterson, K.D. 1992. Yellow fever epidemics and mortality in the United States, 1693-
1905. Soc. Sci. Med. 34(8): 855-865.
Perfect, T.J. 1992. IPM in 2000, pp.47-53. IN A.A.S.A. Kadir and H.S. Barlow (eds.),
Pest Management and the Environment in 2000. CAB International, Oxford, UK.
401pp.
Quick, M. and B.R. Stevens. 2001. Amino acid transporter CAATCH1 is also an amino
acid-gated cation channel. J. Bio. Chan. 276(36): 33143-33418.
Racioppi, J.V. and D.L. Dahlman. 1980. Effects of L-canavanine on Manduca sexta
(Sphingidae: Lepidoptera) larval hemolymph solutes. Comp. Biochem. Physiol.
67:35-39.
Ragsdale, D. and E.B. Radcliffe. 1999. Colorado potato beetle management. University
of Minnesota Cooperative Extension Service. Internet URL: http://ipmworld.umn.
edu/aphidalert/CPB~DWR.html. Accessed April 2004.
Robertson, J.L. and H.K. Priesler. 1991. Pesticide bioassays with arthropods. CRC
Press, Inc. Boca Raton, 127pp.
Rock, G.C. 1971. Utilization of D-isomers of the dietary, indispensable amino acids by
Argyrotaenia velutinana larvae. J. Insect Physiol. 17:2157-2168.
Rock, G.C. and E. Hodgson. 1971. Dietary amino requirements for Heliothis zea
determined by dietary deletion and radiometric techniques. J. Insect. Physiol. 17:
1087-1097.
Rock, G.C., BG. Lign, and E. Hogson. 1973. Utilization of methionine analoges by
Argyrotaenia velutinana larvae. Arm. Entol. Soc. Amer. 66(1): 177-179.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Russell L. Rouseff, Chair
Professor of Food Science and Human Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Maurice R. Marshall, Jr.
Professor of Food Science and Human Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Reriee M./ioodrich
Assistant Professor of Food Science and Human
Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
David H. Powell
Faculty Scientist of Chemistry
This dissertation was submitted to the Graduate Faculty of the College of Education and
to the Graduate School and was accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.
May 2004
Dean, Graduate School


TABLE OF CONTENTS
Elge
ACKNOWLEDGMENTS iii
LIST OF FIGURES viii
ABSTRACT xi
CHAPTERS
1 THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE
CONVENTIONAL PESTICIDES THE ONLY ANSWER? 1
Introduction 1
Importance of IPM in Florida and Surrounding States 2
Problems Associated with Pesticide Misuse 4
Biorational Compounds- An Alternative to Chemical Pesticides 5
2 HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL
INSECT PESTS 7
Non-Protein Amino Acids 7
Essential Amino Acids 10
The Cation-Anion Modulated Amino Acid Transporter with Channel
Properties (CAATCH1) System 9
Methionine 13
Research Objectives 16
3 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT
OF THE TOBACCO HORNWORM, Manduca sexta, UNDER
LABORATORY CONDITIONS 17
Introduction 17
Materials and Methods 18
Diets and Survivorship 18
Feeding and Development 20
Preference Tests 22
Data Analysis 24
Results 24
v


40
<
o
X!
3
3
X
<
>
4>
3
0)
04
m/z
Figure 4-4. Upper spectra from orange juice MS at RT = 17.68 bottom spectra of |3-
cyclocitral from database NIST 2002.
O
C
a
T3
C
3
X
<
>

4
13
04
m/z
Figure 4-5. Upper spectra from orange juice MS at RT = 21.94, bottom spectra from
standard P-ionone using identical ion trap MS at identical retention time.


58
Days of Exposure
Figure 5-3. Mortality of yellow fever mosquito larvae exposed to various
concentrations of D-methionine (nTOtai=240). Data were
adjusted using Abbotts formula for control mortality.


Relative Abundance
71
m/z
Figure 6-8. Upper spectra from model solution MS at RT 20.46, bottom spectra from
standard P-homocyclocitral using identical ion trap MS at identical retention
time
Figure 6-9. Upper spectra from model solution MS at RT 25.13, bottom spectra from
standard (3-ionone using identical ion trap MS at identical retention time.


14
in its use as a feed supplement for livestock under the trade name of Alimet (Novus,
Inc., St. Louis, MO).
In addition to vertebrates, methionine also is considered an essential amino acid
for insects (Nation 2001). Based on research on nutritional requirements for insects, the
amount of methionine needed in a diet for survival ranged from as little as 0.0007 mg/mL
(for Aedes aegypti (L.) (Diptera: Culicidae) to as high as 100 mg/mL (for Heliothis zea
(Broddie) (Lepidoptera: Noctuidae)) (Dadd and Krieger 1968; Eymann and Friend 1985;
Friend et al. 1957; Kaldy and Harper 1979; Kasting et al. 1962; Koyama 1985; Koyama
and Mitsuhashi 1975; Rock and Hodgson 1971; Singh and Brown 1957). Methionine
occurs naturally as the L-isomer while the D-isomer (an optical enantiomer) is toxic to
many insects and is not found in nature (Anand and Anand 1990). A few exceptions are
known, (mainly Diptera and Lepidoptera) that actually are capable of metabolizing the
normally unusable D-isomer (Dimond et al. 1958; Geer 1966; Rock 1971; Rock et al.
1973; Rock et al. 1975). The requirement for small amounts of this amino acid (as
compared to other amino acids) may be a result of the ability for some insects to
synthesize methionine from cysteine (a common sulfur containing amino acid) thus
reducing the need to take in exogenous sources of methionine. Jaffe and Chrin (1979)
found that A. aegypti adults were able to synthesize methionine from homocysteine with
the aid of a methionine synthetase. They found this enzyme similar to those common in
other metazoans, and found that the levels of methionine synthetase increased with the
presence of filarial parasites. They hypothesized that this increase in methionine
synthetase was a result of the parasite depleting the host of methionine.
Fertility and fecundity also have been associated with methionine in some insects
(mainly D. melanogaster,) with the possibility if it being a limiting factor during egg


105
Dimond, J.B., A.O. Lea and D.M. Delong. 1958. Nutritional requirements for
reproduction in insects. Proc. 10 Int Congr. Entomol. 2:135-137.
Droux, M, B.Gakire, L. Denis, S. Ravanel, L. Tabe, A.G. Lappartient, D. Job. 2000.
Methionine biosynthesis in plants: biochemical and regulatory aspects. Pp. 73-92.
/VBrunold, C., Rennenberg, H., De Kok, L.J., Stulen, L, Davidian, J.C. (eds.):
Sulfur Nutrition and Sulfur Assimilation in Higher Plants. Molecular,
Biochemical and Physiological Aspects. Paul Haupt Publishers. 447pp.
Durham, S. 2000. Hairy vetch thwarts Colorado potato beetle. Agricultural Research
Service, United States Department of Agriculture. Internet URL:
http://www.ars.usda.gov/is/pr/2000/000413.htm. Accessed April 2004.
Dwyer, J. 1999. Research Links 2000 Tobacco Homworm. Carleton College,
Department of Biology. Internet URL: http://www.acad.carieton.edu/curncular
/BIOL/resources/rlink. Accessed April 2004.
Ehler L.E. and D.G. Bottrell. 2000. The illusion of integrated pest management. Issues
in Science and Technology Online. National Academies and the University of
Texas (Dallas). Internet URL: http://www.nap.edU/issues/16.3/dder.htm.
Accessed April 2004.
Elliott, N.C., J.A. Webster, and S.D. Kindier. 1999. Developmental response of
Lysiphlebus testaceipes to temperature. Southwest Entomol. 24: 1-4.
Eymann, M. and W.G. Friend. 1985. Development of onion maggots (Dptera:
Anthomyiidae) on bacteria-free onion agar supplemented with vitamins and amino
acids. Ann. Entomol. Soc. Am. 78:182-185.
Feldman, D.H., W.R. Harvey and B.R. Stevens. 2000. A novel electrogenic amino acid
transporter is activated by K+ or Na+, is alkaline pH-dependent, and is CT-
independent. J. Biol. Chem. 275:24518-24526
Felton, G.W. and D.L. Dahlman. 1984. Allelochemical induces stress: Effects of L-
canavanine on the pathogenicity of Bacillus tkuringiensis in Manduca sexta. J.
Invert Path. 44: 187-191.
Ferro, D.N. 1985. Pest status and control strategies of the Colorado potato beetle. IN
Ferro, D.N. and R.H. Voss (eds.) Proceedings of the Symposium on the Colorado
potato beetle, XVII Inemational Congress of Entomology.
Fisher Scientific Internationa]. 2004. Online Catalog. Fisher Science International
Internet URL: https://wwwl.fishersci.com/index.jsp. Accessed April 2004.
Florida FIRST. 1999. Putting Florida FIRST: Focusing LFAS resources on solutions for
tomorrow. University of Florida, Institute of Food and Agricultural Sciences.
16pp.


21
HPLC Separation
C-30 carotenoid columns with ternary gradient solvent systems and photodiode array
detectors have been employed to separate and identify carotenoids in citrus (68, 75, 76).
In this study saponified carotenoids were separated using a C-30 carotenoid column with
ternary solvent gradient system of water, methanol, and MTBE with photodiode array
detection. The resulting separation is shown in Figure 3-3. More than twenty-four
carotenoids were separated as distinct peaks and sixteen of these peaks were identified
based on their spectral characteristics (Table 3-1), relative elution order compared to
literature values and authentic standards. As seen from the chromatogram in Fig. 3-3,
peaks 11 and 20 corresponding to cis-violaxanthin and (3-cryptoxanthin (15.76 and 12.34
percentage of total peak area, respectively). They have been reported as the major
carotenoids in earlier studies. Beta-cryptoxanthin is well accepted as the major
contributor to the orange color of the juice (79) because of its relatively high
concentration an overall absorbance in the red/orange range of the spectrum. The last
four peaks (21-24) are due to a variety of carotenes which are not completely resolved.
Both a- and (3-carotene are of particular interest in this study because of their ability to
produce a range of norisoprenoids which have been observed in other food products. In
addition, peaks 18 and 20 were well resolved and corresponded a- and (3-cryptoxanthin
from the match of retention times and spectral characteristics compared to authentic
standards. The final peak of interest was neoxanthin and this compound corresponds to
peak 4 which is neither particularly well resolved nor large.
The large number and similarity of orange juice carotenoids make separation
difficult. Thus even under the best chromatographic systems, some peaks will not be


41
However, a-ionone has not been previously reported and is shown in Figure 4-6.
The spectral match in this case is good considering the very low levels of a-ionone
present, but not perfect. Even with careful background subtraction (which was done for
all the previous spectra as well), there will be a fair amount of extraneous peaks simply
due to random noise. However, the major fragment ions of m/z 192 (M+), 177, 163, 136,
121, 109, 93, 91, and 77 are all present, more than enough to confirm the presence of a-
ionone in orange juice.
m/z
Figure 4-6. Upper spectra from orange juice MS at RT = 20.87, bottom spectra from
standard a-ionone using identical ion trap MS at identical retention time.
Conclusion
Four norisoprenoids in fresh orange juice (p-cyclocitral, (3-damascenone, a-ionone,
and (3-ionone) have been conclusively identified through the combined information from


23
Figure 3-3. Chambers used for tobacco homworm and Colorado potato beetle
preference tests. Two treatments (control and 1.0% L-
methionine) were used to determine if any larvae exhibited any
preference or avoidance to L-methionine. Treatments were
alternated in the chamber and neonates were released in the center
of the dish and allowed to search for food. The filter paper in the
bottom of the dish was moistened to prevent desiccation of the
leaf disks and the test specimens.


63
Days of Exposure
Figure 5-7. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Beta-alanine (nrotai=240). Data were adjusted
using Abbotts formula for control mortality.


25
Conclusions
Carotenoids in Valencia orange juice were extracted using mixed solvent
(hexane:acetone:ethanol, 50:25:25) and subsequently saponified. The saponified
carotenoids were separated using a C-30 carotenoid column with a ternary gradient
solvent system. Twenty-four carotenoids were separated as distinct peaks and sixteen of
these peaks were identified based on their spectral characteristics (Table 3-1), relative
elution order compared to literature values and authentic standards. Although they have
been reported previously, the presence of a-cryptoxanthin, (3-cryptoxanthin, a-carotene,
(3-carotene and neoxanthin in orange juice was confirmed by comparing retention and
spectral properties with standards or literature values. These specific carotenoids were of
interest because they possess the direct structural segments needed to serve as precursors
potent aroma norisoprenoids.


CHAPTER 4
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER LABORATORY
CONDITIONS
Introduction
Leptinotarsa decemlineta (Say) (Coleptera: Chrysomelidae), the Colorado
potato beetle (CPB), is considered an economic pest throughout North America. The
larvae and adults of the CPB feed on a wide variety of solanaceous crop plants and are
responsible for $150 million in losses and control related costs (Durham 2000). To
further complicate matters, the CPB is resistant to numerous pesticides, including various
pyrethroids and carbamates (Bills et al. 2004). Historically, CPB management relied
heavily on chemical control methods that led to the development of resistance to different
pesticides in several areas of the eastern United States (Forgash 1985; Gauthier et al.
1981). Control of CPB without the use of chemicals is further complicated given the
species ability to develop resistance and the limitations on the use of resistant varieties of
potato (Ragsdale and Radcliffe 1999). The use of plant varieties that are resistant to CPB
and other pests also run the risk of developing tolerance to chemical pesticides in other
pest species (Sorenson et al. 1989). Despite the success of Bacillus thuringiensis-
tenebrionis (Btt) and the biocontrol agents Podisus maculiventris Say (Hemiptera:
Pentatomidae) and Edovum puttleri Grissel (Hymenoptera: Eulophidae), more biorational
alternatives are necessary for controlling CPB to prevent yet another devastating threat to
the potato industry because of this insects ability develop resistance and overcome
control methods (Boucher 1999; Ferro 1985; Tipping et al. 1999). This makes the CPB
39


27
treatment after 10 days of exposure. The 0.1% L-methionine concentration had lowest
larval mortality with approximately 30% observed for the trial.
The excised leaf trials exhibited higher mortality rates associated with the
treatments than did the artificial diet trials. Again, complete mortality was observed with
the 3.0% L-methionine thru 10.0% L-methionine concentrations after 1 day of exposure
(Figure 3-6). Greater than 90% in the 0.5% L-methionine and 1.0% L-methionine
treatments, followed by 80% mortality in the 0.3% L-methionine treatment, and greater
than 60% mortality occurred in the 0.1% L-methionine treatment after 8 days.
Whole plant trials produced results similar to the excised leaf trials with greater
than 90% larval mortality observed with the 1.0% L-methionine treated plants after 14
days (Figure 3-7). Mortalities exceeding 20% and 60% were observed for the 0.1%
L-methionine and 0.5% L-methionine treatments, respectively, during the same time
interval.
PROBIT analysis of a sample size of n-rota 1,540 for 7 treatments (0.1%
L-methionine, 0.3% L-methionine, 0.5% L-methionine, 1.0% L-methionine, 3.0%
L-methionine, 5.0% L-methionine and 10.0% L-methionine) revealed an overall LC50 of
0.66% (32.3 mg/g leaf material) concentration for the artificial diet and 0.074% (4.39
mg/g leaf material) concentration for the natural diet after 9 days of exposure (Figure
3-8). The LC50 for the THW exposed to artificial diet was approximately half the value
of that for the natural diet for the 24 to 72 hour exposure period. The LC50 for the
artificial diet of 1.08% (52.3 mg/g leaf material) for 24 h dropped to 1.0% (48.5 mg/g leaf
material) after 48 h and to 0.57% (28.0 mg/g leaf material) after 72 h. As for the natural
diet, the LC50 of 0.53% (26.1 mg/g leaf material) was found to be lower than the artificial


5
such as Dengue fever, yellow fever, and West Nile virus to name just a few, put countless
millions more at risk. It would be dangerous to think that these diseases only occur in
underdeveloped countries and not the United States. Integrated Pest Management
practices also should be adopted for controlling the medical and veterinarian important
insect vectors of these and other diseases.
Biorational Compounds: An Alternative to Traditional Chemical Insecticides
One way to reduce this reliance on traditional chemical pesticides and delay
resistance is by increasing the variety and use biorational compounds. Biorational
compounds are effective against selected pest species but are innocuous to nontarget or
beneficial organisms; and have limited affect (if any) on biological control agents
(Stansly et al. 1996). Biorational compounds include detergents, oils, pheromones,
botanical products, microbes, and systemic and insect growth regulators (Perfect 1992;
Wienzierl et al. 1998). Their safety lies in the low toxicity of the compound to nontarget
organisms and the compounds short residual activity in the field. For example, Bacillus
thuringiensis isrealensis (Bti) currently is one of the most widely used microbial
pesticides for controlling aquatic dipteran pests (i.e., mosquitoes and black flies) because
of its selectivity to this group and minimal nontarget effects (Glare and OCallaghan
1998). However, resistance to Bt products has occurred in many species of lepidoptera
from overuse of Bacillus thuringiensis kurstaki, and in some mosquito species to Bti, thus
showing the need for alternatives to these compounds that are still effective (Brogdon and
McAllister 1998; Marrone and Macintosh 1993). In addition to resistance, other
problems are associated with the use of microbial control agents. Cook et al. (1996)
discussed potential hazards, not properly identified in the planning stages, of
displacement of native microorganisms, allergic responses in susceptible humans and


4
all risk of insect damage by using more applications and stronger pesticides (Schuster et
al. 1996).
Problems Associated with Pesticide Misuse
The use of pesticides is not completely ruled out under IPM strategies, but rather
IPM encourages responsible use to minimize environmental harm and to protect human
safety and health (Deedat, 1994). However, the misuse (both intentional, in terms of
more is better; and unintentional, as in agricultural runoff) also has resulted in
resistance in some of the target pests. For example, surveys in North Carolina have
shown that the Colorado potato beetle has become resistance to fenvalerate, carbofuran,
and azinphosmethyl as a result of control failures in the field (Heim et al. 1990).
Resistance to insecticides has also been observed in more than 450 arthropod pests
(Romoser and Stoffolano 1998). Bills et al. (2004) found a 38% increase in the number
of registered compounds used as pesticides from 1989-2000, and also a 16% increase in
pesticide resistance of arthropod species worldwide.
Losses are not limited to agricultural systems alone. Across Africa for example,
populations of insecticide-resistant mosquitoes are the result of a variety of mechanisms,
including exposure to pesticide residues in agricultural runoff, mutation of target sites,
and migration of resistant populations into areas where there were no previous problem
(FIC-NIH 2003). Parts of southwest Asia have seen a resurgence of malaria in some
areas where it was considered eradicated (due to a combination of resistance and the
economics associated with control of mosquito vectors) (Deedat 1994). The importance
of this example becomes even more relevant when one considers that one million
individuals die every year as a result of malaria, with upwards of 500 million cases per
year (Centers for Disease Control 2003). The existence of other mosquito-bome diseases


28
Figure 3-6. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nTotai= 1,540) on excised eggplant leaves. Data were
adjusted using Abbotts formula for control mortality. Note the overlap in
trend lines for the 3.0% L-methionine-10.0% L-methionine concentrations
after Day 1.


13
Solid Phase Microextraction
Solid phase microextraction (SPME) is a relative new technique whereby analytes
of interest partition from the sample matrix into a polymeric coated silica fiber,
developed by Pawliszyn and co-workers (63). It is a simple, rapid, solventless technique
to sample static headspace volatiles. A 1 or 2 cm length of fused silica fiber, coated with
a polymer, is bonded to a stainless steel plunger and installed in a holder that looks like a
modified microliter syringe. The plunger moves the fused silica fiber into and out of a
hollow needle. To use the unit, the analyst draws the fiber into the needle, passes the
needle through the septum that seals the sample vial, and depresses the plunger, exposing
the fiber to the sample or the headspace above the sample. Organic analytes adsorb to the
coating on the fiber. After adsorption equilibrium is attained, the fiber is drawn into the
needle, and the needle is withdrawn from the sample vial. Finally, the needle is
introduced into the gas chromatographic (GC) injector, where the adsorbed analytes are
thermally desorbed and delivered to the GC column.
The application of headspace SPME to flavor volatile compounds has been
employed in the study of flavor volatiles in orange juice using a PDMS coated fiber (64),
a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65), PDMS and polyacrylate
fiber (66). The partition coefficients of the polymeric coatings for the analyses differed
markedly. For example, terpenes such as a-pinene, P-myrcene, y-terpinenes, and
limonene are all nonpolar, and were extracted to a higher degree into the nonpolar PDMS
coating (66). Corresponding PDMS extracted the least amount of the more highly polar
volatiles, PDMS/DVB and Carbowax/DVB had partition coefficients higher than that of
PDMS for the most polar molecules (67). The Carboxen-PDMS fiber coating was more


84
The purpose of this portion of the study was to examine the effects of L-
methionine on selected nontarget species that are both important in terms of being
beneficial in controlling other pest species and also represent different feeding guilds that
would come into contact with this compound in different ways (e.g., on prey items, on
plant surfaces, hosts of parasitoids).
Materials and Methods
Coleomeeilla maculata
Adults were obtained from ENTOMOS, LLC (Gainesville, Florida), and were
held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth stage
inserted (to facilitate cleaning) at 27C, 60% relative humidity and 16L/8D photoperiod
in FRIUs. Natural diet consisted of excised cotton leafs infested with aphids (Aphis
gossypii Glover (Hemiptera: Aphididae)). Leaves were then dipped into either a 1.0%
L-methionine solution or 0% L-methionine (control) mixed with deionized H2O. Five
adults were used in each replicate for a total n=30 for each treatment. Leaves were
replaced every other day from 27 October 2002 to 07 November 2002. Artificial diet was
obtained from ENTOMOS and prepared according to their guidelines with the exception
of the inclusion of methionine for the 1.0% L-methionine treatment (wt/wt). Diets were
replaced every other day from 27 October 2002 to 07 November 2002. Ten adults were
used for each replicate for a total n=60 for each treatment. Data was normalized to 0%
mortality when the treatments were corrected for control mortality (i.e., when the control
mortality was greater than that of the treatment).


LCS0 (% L-methionine Concentration)
45
Figure 4-2. Concentrations (%) of L-methionine concentrations required for
the mortality of 50% of the test population of Colorado potato
beetle after 8 days exposure (nrotai=220). Number range
following value is the 95% confidence limits. Determination of
LC50 was performed using PROBIT Version 1.5 (Ecological
Monitoring Research Division, USEPA), including Abbotts


LIST OF FIGURES
Figure page
2-1 Examples of carotene and xanthophyll carotenoid structures 4
2-2 Major fragment classes of carotenoid biodegradation 5
2-3 General steps for the conversion of carotenoids into flavor compounds, showing
the formation of P-ionone and P-damascenone from P-carotene and neoxanthin
respectively 6
2-4 Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters) 8
2-5 Stevens law, comparing two difference compounds: A= compound A, B=
compound B 12
3-1 Possible degradation pathways for the formation of P-cyclocitral and P-ionone
from P-carotene 15
3-2 Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of P-damascenone 19
3-3 Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH, MTBE
ternary solvent gradient 22
3-4 Absorbance spectra for P-carotene (a), leutoxanthin (b), and neoxanthin (c), peak
24, 12 and 4 respectively 23
4-1 GC-FID (top) and average time-intensity of four GC-0 runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and 23
correspond to norisoprenoids, all numbers refers to compounds in Table 4-1 34
4-2 Comparison between total ion chromatogram and selected ion chromatograms
(SIC) A: P-cyclocitral, B: P-ionone, C: a-ionone 37
4-3 Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190 38
IX


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ERXM6CEXM_XPAN37 INGEST_TIME 2015-04-01T19:38:04Z PACKAGE AA00029854_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


29
Figure 3-7. Survivorship of tobacco homworm larvae exposed to various
concentrations of L-methionine (niotai=256) on whole plants. L-
methionine was applied using a hand-held sprayer in the amount of
10 mL/treatment. Data were adjusted using Abbotts formula for
control mortality.


CHAPTER 3
HPLC DETERMINATION OF CAROTENOID NORISOPRENOID PRECURSORS IN
ORANGE JUICE
Introduction
The color of orange juice is due to a complex mixture of plant pigments called
carotenoids. Over 50 carotenoids have been identified in orange juice including P-
carotene, a-carotene, P-cryptoxanthin and neoxanthin (17, 68). In addition to acting as
plant pigments and free radical scavengers (produced during photosynthesis), these large
highly conjugated molecules can break down forming smaller, highly potent aroma
volatiles called norisoprenoid (15, 39, 69, 70). The structure of P-carotene is shown
below. If this molecule hydrolyzes between carbon atoms 9 and 10, a C13 norisoprenoid,
P-ionone is formed. If the molecule hydrolyzes between carbon atoms 7 and 8, then a C10
norisoprenoid, P-cyclocitral is formed.
Figure 3-1. Possible degradation pathways for the formation of P-cyclocitral and P-
ionone from P-carotene.
Beta-cyclocitral, P-damascenone, a-ionone, and P-ionone are some of the volatiles
reported in tomato (23), wine (71), tobacco (22) and tea (16) as aroma active compounds.
15


Bioassays involving the tobacco homworm, Colorado potato beetle (CPB),
Leptinotarsa decemlineata, and the yellow fever mosquito (YFM), Aedes aegypti, were
conducted to determine the insecticidal properties of this compound. L-methionine in
artificial and natural diets resulted in the mortality of 50 to 100% in concentrations of
0.3% and higher for THW and CPB. Feeding rates and larval development also were
affected with reduced levels (>0.1%) of L-methionine. Bioassay trials involving YFM
larvae were similar, concentrations greater than 0.1% L-methionine produced mortality
rates of 70 to 100%. All three species responded better to higher concentrations of L-
methionine than to Bacillus thuringiensis, the most commonly used and commercially
available biorational pesticide.
Field trials and non-target tests also were performed to determine L-methionine
effectiveness under natural settings and safety to other organisms. Eggplant yield was
not reduced by the application of L-methionine, which effectively controlled CPB larvae
on the plants. Furthermore, several beneficial insects that were tested (a predator, a
herbivore, and a parasitoid) were not affected by the addition of L-methionine to their
diets.
Based on these results, L-methionine was found to be effective in controlling
selective agriculturally and medically important insect pest species, yet posed little threat
to the crop plants applied to or to non-target organisms. The use of L-methionine as a
pesticide, its relationship with insects and its possible use in delaying insecticide
resistance were also examined.
xu


CHAPTER 7
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF
SELECTED NONTARGET SPECIES
Introduction
A biorational pesticide is defined as one that is effective against pest species but
innocuous to non-target organisms and not disruptive to biological control agents and
beneficial species (Stansly et al. 1996). To test L-methionine as a potential pesticide and
determine if it could be considered biorational, it was necessary to examine the effects of
this compound on selected nontarget species that could possibly come into contact with
it, either directly while on the plant or indirectly via incidental contact or as a host that
has come into direct contact with this compound. The species chosen reflect a variety of
non-target organisms, mainly those that were shown to be important in controlling some
pest species. The pink spotted ladybird beetle, Coleomegilla maculata (DeGeer), the
mottled water hyacinth weevil, Neochetina eichhorniae Warner, and the greenbug
parasitoid, Lysiphlebus testaceipes (Cresson) all are beneficial insects that have been
effective against pests in the state of Florida and also are common and readily available.
Each species also represents a different feeding guild (predator, herbivore and parasitoid,
respectively) to ensure a thorough examination of the possible effects of methionine as it
might be encountered in under natural conditions.
The pink spotted ladybird beetle (PSLB) is an abundant polyphagus species that is
known to feed on many lepidopteran and coleopteran pests, including the Colorado
potato beetle, in which it was responsible for over 50% of the predation on eggs and early
82


CHAPTER 5
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY
CONDITIONS
Introduction
Integrated Pest Management practices are not restricted to agricultural pests.
Medically important insect pests are responsible for epidemics that have changed the
course of human existence, from bubonic plague spread by the Oriental rat flea
(Xenopsylla cheops Rothschild (Siphonaptera: Pulicidae)), to malaria carried by
anopheline mosquitoes. One medically important species that has had a significant
impact on human existence is the yellow fever mosquito (YFM), Aedes aegypti (L.)
(Dptera: Culicide). This cosmopolitan species is found worldwide and is the primary
vector for human dengue and yellow fever despite concerted efforts at eradication in the
United States (Womack, 1993). In the United States alone, upwards of 150,000 lives
were lost to yellow fever in the period starting in the late 18th century and into the early
20th century (Patterson, 1992). However, because of the development of a vaccine,
yellow fever has been replaced by Dengue which is now second only to malaria as a
worldwide threat (Gubler, 1998). Because Dengue fever is also vectored by the YFM, it
poses a risk by affecting tens of millions of people worldwide (Gubler and Clark, 1995).
The inclusion of the YFM in this study was an effort borne of curiosity because of
the lack of knowledge of the CAATCH1 system in other insects and the availability of
specimens for study. Mosquito larvae are particulate feeders and have dietary
52


78
were observed thereafter. By Day 4 the 1.0% and 0.5% treatment were the only
treatments that were statistically different from the control. There was substantial
unexplained attrition of CPB larvae in the field for all treatments, which leveled off by
Day 3. Data from day 5 was discounted because of the onset of a severe cold front that
made it difficult to separate the effects of the weather from the treatments affects.
Discussion
The results of the field studies show that, using conventional application
techniques, a mixture of methionine and Silwett L-77 did not appear to affect eggplant
yield. Furthermore, the same combination produced substantial control of CPB larvae
under natural field conditions after four days. Dahlman (1980) found that L-canavanine,
a non-protein amino acid, could be used in the same manner for control of THW on
tobacco, but the widespread use of this compound was limited by the cost ($107.85 for lg
L-canavanine versus $3.35 for lg of L-methionine (Fisher Scientific International 2004)),
adverse effect on plant development (Nakajima et al. 2001), and toxicity to vertebrates
(Rosenthal 1977). Although complete coverage of the plant was not feasible,
approximately 2.5 grams to 7.5 grams of L-methionine was applied to the plants in each
of the treatment plots. Each plant, based on the amount applied, received approximately
7.5xl06 pg for the 1.0% L-methionine treatment, 3.8xl05 pg for the 0.5% L-methionine
treatment and 2.5x104 pg for the 0.1% L-methionine treatment. This compares to only 4pg of
L-canavanine, which resulted in decreased size, fecundity, and mortality of THW under
field conditions (Dahlman 1980). It should be noted that the toxicity of L-canavanine is
well documented and has a different mode of action than L-methionine and cannot be


50
sample matrices. Boa et al.(92) reported that the combination of SPME with the standard
addition method reduce the problem of matrix effects and improved the precision of the
procedure.
Norisoprenoid Quantification using Standard Additions
The norisoprenoids P-cyclocitral, (3-damascenone, a-ionone, and p-ionone were
quantified in orange juice using the standard addition method (Fig. 5-2, 5-3, 5-4, 5-5, and
5-6). The integrate peak areas at specific m/z 137, 177, 177 and 190 for P-cyclocitral,
Figure 5-2. Standard addition data for P-cyclocitral peak area vs. added concentration in
fresh orange juice. Regression line calculated from peak area at selected mass
137.
P-ionone, a-ionone, and p-damascenone respectively were plotted versus the
concentration of the spiked standards. The amount of each norisoprenoid (Table 5-2 and
Table 5-3) was calculated from the regression equation where the calculated value was
determined at y = 0.
As seen from the plots of Fig. 5-2 through Fig. 5-6, the correlation coefficients for
the standard addition data was at least 0.99 in all cases except for P-cyclocitral (Fig. 5-2)


Ill
Rock, G.C., A. Khan, and E Hodgson. 1975. The nutritional value of seven D-amino
acids and a-keto acids for Argyrotaenia velutinana, Heliothis zea and Phormia
regina. J. Insect Physiol. 21:693-703.
Romoser, W.S. and J. G Stoffolano, Jr. 1998. The Science of Entomology, Sedition.
McGraw-Hill. Newyork, 605pp.
Rosen, D., F.D. Bennett, and J.L Capinera. 1996. Preface, pp. V-vi. IN D. Rosen, F.D.
Bennett and J.L. Capinaera (eds.), Pest Management in the Subtropics: Biological
Control- a Florida Perspective. Intercept Limited, Andover, UK. 737pp.
Rosenthal, G. A. 1977. The biological effects and mode of action of L-canavanine, a
structural analogue of L-arginine. Q. Rev. Biol. 52(2): 155-178.
Rosenthal, G.A. and D.L. Dahlman. 1975. Non-protein amino acid-insect interactions.
II. Effects of canaline-urea cycle amino acids growth and development of the
tobacco homworm, Manduca sexta (L.) (Sphingidae). Comp. Biochem. Physiol.
52:105-108.
Rosenthal, G.A. and D.L. Dahlman. 1988. Degradation of aberrant proteins by larval
tobacco homworm, Manduca sexta (L) (Sphingidae). Arch. Insect Biochem
Physiol. 8: 165-172.
Rosenthal, G. A. and D.L. Dahlman. 1991. Incorporation of L-canavanine into proteins
and the expression of its antimetabolic effects. J. Ag. and Food Chon. 39:987-
990.
Rosenthal, G.A., D.L. Dahlman, P.A. Crooks. S.N. Phuket, and L.S. Trifonov. 1995.
Insecticidal properties of some derivatives of L-canavanie. J. Agrie. Food Chem.
43:2728-2734.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1976. A novel means for dealing with
L-canavanine, a toxic metabolite. Science 192: 256-258.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1977. Degradation and detoxification
of canavanine by a specialized seed predator. Science 196:658-660.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1978. L-canaline detoxification: A
seed predators biochemical defense. Science 202: 528-529.
Rosenthal, G.A., P. Nkomo and D.L. Dahlman. 1998. Effect of long-chained esters on
the insecticidal properties of L-canavanine. J. Agrie. Food Chem. 46(1): 296-299.
Royer, T.A., K.L. Giles, S.D. Kindier and N.C. Elliott. 2001. Developmental response
of three geographic isolates of Lysiphlebus testaceipes (Hymenoptera: Aphididae)
to temperature. Environ. Entomol. 30(4): 637-641.


CHAPTER 1
THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE CONVENTIONAL
PESTICIDES THE ONLY ANSWER?
Introduction
Integrated Pest Management (IPM), the sustainable approach to the management
of pest species using a combination of biological, chemical and cultural methods to
reduce economic, environmental, and public health risk, was a result of economic losses
associated with years of overuse of chemical control leading to resistance problems. The
use of IPM strategies have certainly decreased pesticide usage and encouraged the use of
methods that ensure a safer environment but many feel that it is not enough. After three
decades of research efforts in the United States, IPM as it was envisioned in the 1970s
was practiced on less than 8% of U.S. crop acreage based on Consumers Union
estimateswell short of the national commitment to implement IPM on 75% of the total
U.S. acreage by the end of the 1990s (Ehler and Bottrell 2000). This means that farm
practices have changed little since the national IPM initiative was established in 1994 to
implement biologically based alternatives to pesticides for controlling arthropod pests. It
should be noted that the low percentage of IPM practices on commercial U.S. farmland
may possibly be related to the lack of sufficient reporting means and actually may be
higher than believed when the local growers and homeowners are included. However,
the United States is considered the worlds largest user of chemical pesticides, accounting
for nearly 50% of total worldwide production and shows no sign of slowing (Deedat
1994). Pesticides remain the primary tool of pest consultants and farmers, because of the
lack of economic incentives to adopt alternative strategies that require more effort to
1


67
After one day of storage at 35C, only a single aroma active compound, P-ionone,
was detected (Fig. 6-3). This suggests that p-ionone is either one of the most common
thermal decomposition products and/or it has one of the lowest aroma thresholds. Beta-
ionone does have one of the lowest aroma thresholds (see Table 5-2), but is probably also
a common decomposition fragment and as shown in Fig. 6-3 might represent a certain
weakness in the C9-C10 double bond.
Figure 6-4. Degradation of P-carotene in model solution at difference carbon bonds
After two weeks storage, five distinct aroma active peaks were observed. Four of
these appeared to correspond to distinct FED peaks. It is interesting to note that P-ionone
is still the highest peak and that all of the predicted decomposition products shown in Fig.
6-4 were observed (e.g., peak 1 was due to P-cyclocitral, peak 2 was due to
P-homocyclocitral). It appears that oxidative degradation of p-carotene at double bond
C9-C10, is the most preferable and P-ionone was reported as the major product from
P-carotene degradation (32, 99). Beta-ionone was reported as an off-flavor of dehydrated
carrot stored in oxygen. When dehydrated carrot was stored in the presence of oxygen its


101
and resistant alike because of the difference in mode of action. Once the population was
reduced, and the corresponding resistant genotype, Btk could be used once more at a
lower concentration, closer to that of the susceptible population. This system could also
be used for the reduction of Bt toxin resistance in the CPB and YFM if the compounds
are compatible.
In conclusion, it appears that L-methionine can be used as an insecticide to
control insect pests of economic and medical importance. The target site (CAATCH1) is
known and found in the midgut/hindgut (presumed) in at least three pest species (tobacco
homworm, Colorado potato beetle and the yellow fever mosquito) and possibly more.
The compound (L-methionine) is a safe compound that is already used for livestock feed
supplements, has very low mammalian toxicity, and is compatible with insecticide
application systems. Non-target organisms were not affected with the application of L-
methionine, further supporting its use as a biorational insecticide. With increasing
resistance to current insecticides in the study organisms, alternatives such as L-
methionine are needed now more than ever to further support of Integrated Pest
Management strategies.


13
Several amino acids were found to initiate the blocking action of ion flow through the
CAATCH1 protein, including threonine, leucine, and methionine, with the latter
producing the greater response, based on CAATCH1 research (Feldman et al. 2000;
Stevens et al. 2002; Quick and Stevens 2001).
Methionine
The amino acid methionine is considered essential in the diets of many organisms.
Methionine is considered an indispensable amino acid in humans. Because the body does
not synthesize it, uptake of methionine must occur in the diet. The recommended daily
allowance of methionine for a healthy lifestyle ranges from 13 to 27 mg/kg/day for
infants to full-grown adults (Young and El-Khoury 1996). This amino acid is linked to a
decrease in histamine levels, increased brain function, and is found in a variety of
sources; with the highest concentration in various seeds, greens, beef, eggs, chicken, and
fish (Dietary Supplement Information Bureau 2000). Recently, research has centered on
the genetic modification of crop plants to overproduce methionine to increase its
nutritional quality (Zeh et al. 2001). Wadsworth (1995) discussed using methionine as a
feed supplement, as an aid in the therapy of ketosis in livestock, and as a treatment for
urinary infections in domestic pets. Onifade et al. (2001) examined the use of housefly
larvae as protein foodstuffs, and found an increase in body weight gain and erythrocyte
counts in rats whose diets were supplemented with fly larvae and methionine. Likewise,
Koo et al. (1980) suggested dry face fly pupae could be used as a dietary supplement and
foodstuff extender for poultry because of the high concentration of methionine. The
environmental safety of methionine is well known, as it poses no risk to vertebrates due
to a rather high oral LD50 of 36g/kg~' observed in rats (Mallinckrodt Baker 2001) and also


This dissertation is dedicated to my late father.


implement, produce unpredictable results, and require new knowledge (Barfield and
Swisher 1994; Ehler and Bottrell 2000).
Importance of IPM in Florida and Surrounding States
Considerable effort has been devoted to developing IPM programs in Florida
because of its unique pest problems and crop production systems, sensitivity to chemical
pollutants, and increased urbanization (Capinera et al. 1994; Rosen et al. 1996). The
necessity for developing IPM protocols for Floridas major plant and animal pests was
underscored in a new statewide initiative. In November 1999, the Institute of Food and
Agricultural Sciences (IFAS) at the University of Florida launched Putting Florida FIRST
Focusing IFAS Resources on Solutions for Tomorrow (Florida FIRST 1999). The
Florida FIRST initiative was created (with input from stakeholders) to define the role of
IFAS in shaping Floridas future in the 21st century. Increasing concerns (expressed
repeatedly by Floridas scientific community and the general public) about environmental
contamination, food safety issues, and human and animal health problems resulting from
the indiscriminate use (and often misuse) of pesticides are making existing methods for
pest management obsolete. Successful implementation of true IPM, as it was
envisioned by those who envisioned the original concept, will have the added benefit of
helping Florida ... enhance natural resources, provide consumers with a wide variety of
safe and affordable foods,... provide enhanced environments for homes, work places
and vacations, maintain a sustainable food and fiber system, and improve the quality of
life... (Florida FIRST 1999).
This effort to promote IPM programs in the state of Florida also benefits the
surrounding states. For example, solanaceous crops produced in the southeastern U.S.
(such as tomato, tobacco, eggplant, peppers and potato) are subjected to the same


49
50
51
52
53
54
55
56
57
58
59
79
Roberts, D. D.; Mordehai, A. P.; Aeree, T. E. Detection and Partial
Characterization of Eight (3-Damascenone Precursors in Apples (Malus domestica
Borkh. Cv. Empire). J. Agrie. Food Chem. 1994, 42, 345-349.
Buttery, R. G.; Teranishi, R.; Ling, L. C.; Tumbaugh, J. G. Quantitative and
sensory studies on tomato paste volatiles. J. Agrie. Food Chem. 1990, 38, 336-
340.
Buttery, R. G.; Ling, L. C. Volatile Components of Tomato Fruit and Plant Parts:
Relationship and Biogenesis. In Bioactive Volatile Compounds from Plants', R.
Teranishi; R. G. Buttery and H. Sugisawa, Eds.; American Chemical Society:
Washington, DC, 1993; pp 23-34.
Lozano, P.; Delgado, D.; Gomez, D.; Rubio, M.; Iborra, J. L. A non-destructive
method to determine the safranal content of saffron (Crocus sativus L.) by
supercritical carbon dioxide extraction combined with high-performance liquid
chromatography and gas chromatography. J. Biochem. Biophys. Methods 2000,
43, 367-378.
Knapp, H.; Straubinger, M.; Witte, A.; Winterhalter, P. Aroma Formation in
Saffron. In Frontiers of Flavour Science', P. Winterhalter, Ed.: Freising, Germany,
2000; pp 440-444.
Razungles, A.; Bayonove, C. L.; Cordonnier, R. E.; Sapis, J. C. Grape
carotenoids: changes during the maturation period and localization in mature
berries. Am. J. Enol. Vitic. 1988, 39, 44-48.
Razungles, A.; Gunata, Z.; Pinatel, S.; Baumes, R.; Bayonove, C. Content of
terpenes, norisoprenoids and their precursors in several varieties of grapes. Sci.
Aliments 1993,13, 59-72.
Schreier, P.; Drawert, F.; Junker, A. Identification of volatile constituents from
grapes. J. Agrie. Food Chem. 1976, 24, 331-336.
Williams, P. J.; Sefton, M. A.; Wilson, B. Nonvolatile Conjugates of Secondary
Metabolites as Precursors of Varietal Grape Flavor Components. In Flavor
Chemistry: Trends and Development', R. Teranishi; R. G. Buttery and F. Shahidi,
Eds.; American Chemical Society: Washington, DC, 1989; pp 35-48.
Schneider, R.; Razungles, A.; Augier, C.; Baumes, R. Monoterpenic and
norisoprenoidic glycoconjugates of Vitis vinifera L. cv. Melon B. as precursors of
odorants in Muscadet wines. J. Chromatogr., A 2001, 936, 145-157.
Aeree, T. E.; Barnard, J.; Cunningham, D. G. A procedure for the sensory
analysis of gas chromatographic effluents. Food Chem. 1984, 14, 273-286.


55
juice by solvent extraction and determined the aroma contribution by aroma extract
dilution analysis (AEDA). The value from AEDA was recorded as flavor dilution (FD)
factor (the highest dilution factor of the particular aroma active compounds which can be
perceived by human nose). The most odor active compound by this method was ethyl
butanoate (FD 1024) but the FD factor of P-ionone was onlyl. These same authors (5)
quantified twenty-five odor active compounds by stable isotope dilution assay and
estimated their respective odor contributions by OAV values. Unfortunately they did not
quantify (3-ionone. The OAV of ethyl butanoate (the compound with the highest dilution
value from AEDA (4)) was 1192 (concentration 1192 pg/L, odor threshold 1 pg/L). If
one compares this OAV value with the OAV of P-ionone in present study (11857),
P-ionone could be the most aroma active compound in orange juice. The apparent
conflict in the two sets of data suggests that P-ionone may not have been well extracted in
the AEDA study.
Table 5-4. Aroma active compounds in orange juice grouped by citrusy/minty
Compounds3
Description
L
RI
Relative
intensity
ZB-5
DB-wax
Unknown
Orange peel
963
8C, 8d
1,8 cineole
Minty, camphor
1026
1232
5C, 6d
Nonanal
Orange peel, soapy
1090
1398
6C, 6d
3-mercapto hexan-l-ol
Grapefruit
1121
7C, 7d
Citronellal
Minty,camphor
1160
1489
5C, 7d, 5e
Nerol
Lemomgrass
1222
1798
5C, 5d
Neralb
Lemomgrass
1236
1692
7C, 7d
L-carvoneb
Minty
1242
1747
8C, 8d
Geraniol
Citrusy,geranium
1265
1853
9C, 7d, 4e
1 -p-menthene-8-thiol
Grapefruit
1281
1619
7C, 7d, 5e
Geranial
Green,minty
1310
1742
4c,4d, 4e
Nootkatoneb
Sweet,sour, grapefruit
1824
7C, 5d


Diets and Survivorship 24
Feeding and Development 31
Choice Tests 31
Discussion 36
4 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF
THE COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER
LABORATORY CONDITIONS 39
Introduction 39
Materials and Methods 40
Survivorship 40
Feeding and Development 41
Preference Tests 41
Data Analysis 42
Results 43
Survivorship 43
Feeding and Development 43
Preference Tests 47
Discussion 47
5 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY
CONDITIONS 52
Introduction 52
Materials and Methods 53
Bioassay 53
Growth and Development 54
Data Analysis 56
Results 56
Bioassay 56
Growth and Development 59
Discussion 66
6 EVALUATION OF L-METHIONINE UNDER NATURAL FIELD CONDITIONS69
Introduction 69
Materials and Methods 70
Preliminary Investigation of Silwet L-77 and L-methionine 70
Plot Design 70
Fruit Yield 71
Pest Introduction 71
Data Analysis 74
vi


112
Schoonhoven, L.M. 1972. Plant recognition by lepidopterous larvae. Insect/Plant
Relationships, Symposium of the Roy. Entomol. Soc. London 6:83-93.
Schardt,J.D. 1987. 1987 Florida Aquatic Flora Survey Report. Florida. Department of
Natural Resources, Bureau of Aquatic Plant Management. Tallahassee, FL. 49 pp.
Schuster, D.J, J.E. Funderburk and P.A. Slansly. 1996. IPM in tomatoes, pp. 387-408.
IN D. Rosen, F.D. Bennett and J.L. Capinaera (eds.), Pest Management in the
Subtropics: Integrated Pest Management- A Florida Perspective. Intercept
Limited, Andover, UK. 737pp.
Singh, K.R.P. and A.W.A. Brown. 1957. Nutritional requirements of Aedes aegypti. J.
Insect Physiol. 1(1): 199-220.
Sorenson, C.E., R.L. Fery and G.G. Kennedy. 1989. Relationship between Colorado
potato beetle (Coleptera: Chrysomelidae) amd tobacco homworm (Lepidoptera::
Sphingidae) resistance in Lycopersicon hirsutum f. glagratum. J. Econ. Entomol.
82(4): 1743-1748.
Stansly, P.A., T.X. Liu, D.J. Schuster and D.E. Dean. 1996. Role of biorational
insecticides in management of Bemisia, pp. 605-615. IN D. Gerling and R. T.
Mayer, Jr. (eds.) Bemisia: 1995 Taxonomy, Biology, Damage Control and
Management. Andover, Hants, UKD. 702pp.
Stevens, B.R., D.H. Feldman, Z. Liu and W.R. Harvey. 2002. Conserved tyrosine-147
plays a critical role in the ligand-gated current of the epithelial cation/anion acid
transporter/channel CAATCH1. J. Exp. Biol. 205: 2545-2553.
Sugarman, D. and W. Jakinovitch, Jr. 1986. Behavioral gustatory responses of adult
cockroaches, Periplaneta americana to D and L amino acids. J. Insect Physiol.
32(1): 35-41.
Tobe, S.S. and N. Clarke. 1985. The effect of L-methionine concentration on juvenile
hormone biosynthesis by corpora allata of the cockroach Diploptera punctata.
Insect Biochem. 15(2): 175-179.
Tipping, P.W., C.A. Holko, A.A. Abdul-Baki, and J.R. Aldrich. 1999. Evaluating
Edovum puttleri Grissell and Podiscus maculiventris (Say) for augmentative
biological control of Colorado potato beetle in tomatoes. Biol. Control 16: 35-42.
Tzeng, D.D. 1988. Photodynamic action of methionine-riboflavin mixture and its
application in the control of plant diseases and other plant pests. Plant Protection
Bulletin 30(2): 87-100.
Tzeng, D.D. and J.E. Devay. 1989. Biocidal activity of mixtures of riboflavin against
plant pathogenic fungi and bacteria and possible modes of action. Mycologia
81(3): 404-412.


3
defoliation and fruit damage from various lepidopteran and coleopteran pests that also
threaten Florida. The tomato pinworm [Keiferia lycopersicella (Walshingham)
(Lepidoptera: Gelechidae)], armyworms [Spodoptera spp. (Lepidoptera: Noctuidae)], the
Colorado potato beetle [Leptinotarsa decemlineata (Say) (Coleptera: Chrysomelidae)],
and homworms [Manduca spp. (Lepidoptera: Sphingidae)] are some examples of pests
that threaten both conventional producers and homeowners alike. For example, the
estimated loss from and the cost of control of the tobacco homworm, the number-one
pest in tobacco crops in Georgia, reached $1.5 (and $2.3 million), respectively, for the
years 1996-1997 (Jones and McPherson 1997). From 1992-1998, tomato, eggplant, and
pepper producing areas in the Southeast had a total of 1,247,000 pounds of endosulfan
applied over 270,000 acres (Aerts and Neshiem 1999; Neshiem and Vulinec 2001). The
cost of insecticides applied in Florida tomato production alone for 1993-1994 amounted
to approximately $ 1,052/hectare for a total of $2.1 million; and rose to $2550/acre,
totaling $103M for the 1996-1997 season (Aerts and Neshiem 1999; Schuster et al.
1996). The use of pesticides in Florida tomato production is high because tomatoes
account for 30% of the total vegetable-crop value and 13% of the total vegetable acreage
for the state, with 99% of production aimed toward the fresh market (Schuster et al.
1996). For Florida potato producers, the cost of applying pesticides from 1995-1996
was $11.5M, and 96% of total Florida eggplant-crop acreage was treated with chemical
insecticides (mainly methomyl and endosulfan) (Neshiem and Vulinec 2001). In addition
to the monetary cost of pesticide use, commonly used insecticides such as endosulfan and
fenvalerate show a high degree of toxicity to parasitoids of the tomato pinworm, thus
negating the benefits of predation by natural enemies (Schuster et al. 1996). These
figures may be the result of the more is better attitude of producers who want to avoid


6-5. Mortality of Colorado potato beetle larvae on eggplants treated with L-methionine
and Silwett L-77 87
7-1. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated
artificial diet 88
7-2. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated
cotton plant leaves infested with aphids 89
7-3. Feeding scars on water hyacinth {Eichhornia crassipes) leaf after exposure to
Neochetina eichhorniae adults 90
7-4. Mortality of Neochetina eichhorniae on treated water hyacinth leaves 91
7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated with
L-methionine and Proline 92
7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with
L-methionine 93
x


IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE
USING TIME INTENSITY GC-0 AND GC-MS
By
KANJANA MAHATTANATAWEE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004


Mean Head Capsule Width (mm)
33
(Errer Bars @ 95%; ^(o.os)7,i 52=2.37, F=18.2; P <0.001)
6
5
4
3
2
1
0
Control 0.1% 0.3% 0.5% 0.7% 1.0% Proline Btk
Figure 3-10. Mean head capsule widths of tobacco homworm larvae exposed
to excised eggplant leaves treated with various concentrations of
L-methionine (nrota]=320). Proline (1.0%) and Btk were included
for comparison as positive and negative controls. Error bars
denote 2 SE. Bars within treatments having the same letter are
not statistically different (Tukeys MST, P0.001).


Results 74
Effects of L-methionine and Silwett L-77 on Colorado Potato Beetle Adults
Under Laboratory Conditions 74
Effects of L-methionine and Silwett L-77 on yield 74
Survival of CPB larvae 74
Discussion 78
7 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
NON-TARGET SPECIES 82
Introduction 82
Materials and Methods 84
Coleomegilla maculata 84
Neochetina eichhorniae 85
Lysiphlebus testaceipes 86
Data Analysis 86
Results 87
Coleomegilla maculata 87
Neochetina eichhorniae 87
Lysiphlebus testaceipes 87
Discussion 87
8 SUMMARY AND DISCUSSION 96
LIST OF REFERENCES 102
BIOGRAPHICAL SKETCH 114
Vll


65
Results and discussion
Before beginning storage study, the high purity P-carotene (99% purity) was
evaluated for aroma active impurities using GC-0 of the material in the same manner as
the storage study. This demonstrated the potency of very minor impurities (less than 1%)
and the need to recrystallize the standard P-carotene to remove aroma active impurity
before beginning the storage study (Fig. 6-1). No effort was made to identify these
impurities only to remove them. Freshly recrystallized P-carotene was used in all model
solution storage studies. Before storage, a day 0 (control) was examined using GC-0 to
make certain no aroma active volatiles were detected (Fig. 6-2).
10 15 20 25
Time (min)
Figure 6-1. The standard P-carotene (99% purity) as received (no purification)


77
Control
0.10%
0.50%
1.00%
20
Days after treatment
Figure 6-5. Mortality of Colorado potato beetle larvae on eggplants treated with
L-methionine and Silwett L-77. Mortality of larvae corrected using
Abbotts formula (Abbott, 1925). Analysis performed on arcsin
transformed data. Error bars denote 2 SE. Data points having by the
same letter are not statistically different (Tukeys MST, P=0.05)


60
61
62
63
64
65
66
67
68
69
70
71
80
Schieberle, P.; Grosch, W. Identificierung von Aromastoffen ausder Kruste von
Roggenbrot Verleich mit den Aromastoffen der Krume. Z. Lebensm.-Unters.
Forsch. 1984, 479-483.
da Silva, M. A. A. P.; Lundahl, D. S.; M.R., M. The Capabilities and
Psychophysics of Osme: a New GC-Olfactometry Technique. In Trends in Flavor
Research-, H. Maarse and V. D. Heij, Eds.; Elsevier: New York, 1994; pp 191-
210.
Stevens, S. S. Concerning the psychophysical power law. Quart. J. Exp. Psychol.
1964,16, 383-385.
Arthur, C. L.; Pawliszyn, J. Solid phase microextraction with thermal desorption
using fused silica optical fibers. Analytical Chemistry 1990, 62, 2145-2148.
Jia, M.; Zhang, Q. H.; Min, D. B. Optimization of Solid-Phase Microextraction
Analysis for Headspace Flavor Compounds of Orange Juice. J. Agrie. Food
Chem. 1998, 46, 2744-2747.
Bezman, Y.; Rouseff, R. L.; Naim, M. 2-Methyl-3-furanthiol and methional are
possible off-flavors in stored orange juice: aroma-similarity, NEF/SNIF GC-O,
and GC analyses. J. Agrie. Food Chem. 2001, 49, 5425-5432.
Steffen, A.; Pawliszyn, J. Analysis of Flavor Volatiles Using Headspace Solid-
Phase Microextraction. J. Agrie. Food Chem. 1996, 44, 2187-2193.
Song, J.; Fan, L.; Beaudry, R. M. Application of Solid Phase Microextraction and
Gas Chromatography/Time-of-Flight Mass Spectrometry for Rapid Analysis of
Flavor Volatiles in Tomato and Strawberry Fruits. J. Agrie. Food Chem. 1998, 46,
3721-3726.
Rouseff, R.; Raley, L.; Hofsommer, H.-J. Application of diode array detection
with a C-30 reversed phase column for the separation and identification of
saponified orange juice carotenoids. J. Agrie. Food Chem. 1996, 44, 2176-2181.
Winterhalter, P.; Lutz, A.; Schreier, P. Isolation of a glucosidic precursor of
isomeric marmelo lactones from quince fruit. Tetrahedron Lett. 1991, 32, 3669-
3670.
Lutz, A.; Winterhalter, P. Bio-oxidative cleavage of carotenoids: important route
to physiological active plant constituents. Tetrahedron Lett. 1992, 33, 5169-5172.
Kotseridis, Y.; Baumes, R. L.; Bertrand, A.; Skouroumounis, G. K. Quantitative
determination of b-ionone in red wines and grapes of bordeaux using a stable
isotope dilution assay. Journal of Chromatography, A 1999, 848, 317-325.


LIST OF REFERENCES
Abbott, W.S. 1925. A method for computing the effectiveness of an insecticide. J.
Econ. Entomol. 18:265-267.
Aerts, M.J. and O.N. Neishiem. 1999. Florida Crop/Pest Management Profiles:
Tomatoes. CIR 1238. Pesticide Information Office. Food Science and Human
Nutrition Department, Florida Cooperative Extension Service, Institute of Food
and Agricultural Science, University of Florida. Internet URL:
http://www.edis.ifas.ufl.edu/ BODY_PI039.htm. Accessed April 2004.
Anand, R. and M. Anand. 1990. Nutritive effect of the D isomers of the essential amino
acids in casein diet on Dacus cucurbitae (Coquillett) maggots. Indian J. Entomol.
52(4): 525-528.
Andow, D.A., and S.J. Risch. 1985. Predation in diversified agroecosystems: Relations
between a coccinellid predator Coleomegilla maculata and its food. J. Appl.
Ecol. 22: 57-372.
Audsley, N., R.J. Weaver and J.P. Edwards. 1999. Juvenile hormone synthesis by
corpora allata of tomato moth, Lacanobia olercea (Lepidoptera: Noctuidae), and
the effects of allatostatins and allatotropin in vitro. Eur. J. Entomol. 96: 287-293.
Barfield, C.S. and M.E. Swisher. 1994. Integrated pest management: ready for export?
Historical context and internationalization of IPM. Food Reviews Intemat. 10(2):
215-267.
Baumhover, A.H., W.W. Cntelo, J.M. Hobgod, Jr., C.M. Knott, and J.J. Lam, Jr. 1977.
An improved method for mass rearing the tobacco horn worm. Agricultural
Research Service, Unites States Department of Agriculture ARS-S-167,13 pp.
Beck, S.D. and W. Hnec. 1958. Effects of amino acids on feeding behavior of the
European com borer, Pyraustra nubilialis (Hibn.). J. Insect Physiol. 2:85-96.
Bell, E.A. 1978. Toxins in seeds, pp. 143-161. IN J. Harboume (ed.), Biochemical
Aspects of Plant and Animal Coevolution. Academic Press, New York. 435pp.
Berge, M.A., G.A. Rosenthal and D.L. Dahlman. 1986. Tobacco budworm, Heliothis
virescens (Noctuidae) resistance to L-canavanine, a protective allelochemical.
Pest. Biochem and Physiol. 25: 319-326.
102


54
The calculated concentrations of P-cyclocitral, P-damascenone, a-ionone and
p-ionone in fresh orange juice were 145, 0.09, 47, and 83 pg/L respectively (Table 5-2).
The aroma active compounds in orange juice have been studied by GC-0 methods (4-7).
Only P-ionone was reported (4, 5) but has not been reported the concentration of this
volatile. The concentration of P-damascenone in 3 types of orange juice: fresh,
pasteurized, reconstituted from concentrate were 0.09, 0.18, and 0.85 pg/L respectively
(Table 5-3). This data suggest that there is precursors present in juice and generate P-
damascenone during thermal processing. These precursors are probably carotenoids like
neoxanthin, but could also be glycosided forms of P-damascenone. These precursors can
generate aroma volatiles in foods that have undergone thermal processing as reported for
tomato paste (23) and heated apple juice (9). It has been previously reported that citrus
juice pulp and cloud (insoluble solids) can retain considerable volatiles (93, 94).
Therefore P-damascenone may have been trapped in the pulp during thermal
concentration and might not be completely removed during evaporation. Its partial loss
may also been partially compensated by newly P-damascenone generated from thermally
unstable carotenoids during thermal concentration.
Determination of Relative Aroma Impact of Norisoprenoids
The odor activity value (OAV) is a rough way of determining relative aroma
contribution of various substances. It is determined by dividing the analytical
concentration by the aroma threshold. The OAV of P-cyclocitral, P-damascenone,
a-ionone and P-ionone were 25, 45, 118, and 11857 respectively (Table 5-2). The OAV
value shows that P-ionone is predicted to have the greatest contribution compared to the
other norisoprenoids. Hinterholzer and Schieberle, (4) analyzed the volatiles from orange


27
p-damascenone contributed 32% of the total aroma potency of the juice, and only 1.6%
of the total aroma potency of fresh (unheated) apple juice (9). Safranal is a potent aroma
in saffron formed during drying and storage by hydrolysis from picrocrocin, a
monoterpene glycoside (86). Beta-cyclocitral, P-ionone and P-damascenone were
detected in fresh tomato. Only P-ionone and P-damascenone are the important to tomato
aroma because their concentrations (4 and 1 pg/L respectively) are higher than their odor
threshold (0.007 and 0.002 pg/L respectively). Beta-damascenone shows a ~ 10-fold
increase in concentration in heated tomato juice which was concentrated to tomato paste
(50). Buttery et al. (41) examined both low carotene and high P-carotene tomato lines for
norisoprenoids. They found that the high P-carotene line contained the highest
concentrations of P-ionone and P-cyclocitral. Both norisoprenoids are known biological
or chemical degradation products of P-carotene.
Carotenoids are widely distributed in the plant kingdom and orange juice is
particularly rich and a complex source of these compounds (87). Lutein, zeaxanthin,
P-crytoxanthin, a-carotene and P-carotene have been determined in unsaponified orange
juice carotenoids extracted by ethyl acetate (73). Thirtynine carotenoid pigments were
separated and identified in saponified orange juice carotenoids using HPLC (68). Among
these, P-carotene, a -carotene, neoxanthin P-crytoxanthin, lutein, violaxanthin, and
canthaxanthine have the structural potential to form potent norisoprenoid fragments
(18-22). These carotenoids have been confirmed to be present in the Valencia juice used
in this study based upon HPLC retention time and spectral characteristics data. (See
Chapter 3)


83
instars (Andow and Risch 1985; Giroux et al. 1995; Griffin and Yeargan 2002; Groden et
al. 1990; Hazzard et al. 1991; Hilbeck and Kennedy 1996; Munyaneza and Obrycki
1998). This species is widespread throughout North America, and has been shown to
provide effective biological control in several crop species, including com, crucifers,
tomato and potato (Hoffman and Frodsham 1993). However, the PSLB was found to be
susceptible to carbaryl and menthamidophos, the same pesticides used for the control of
many aphid species (Hoffman and Frodsham 1993).
Since its introduction into the United States in 1884, water hyacinth (Eichhornia
crassipes (Mart.) Solms-Laubach) has infested waterways of the southeast that has cost
upwards of $2 million to control in Florida alone (Schardt 1987). The mottled water
hyacinth weevil (MWHW), native to Argentina, was first released in Florida in 1972 and
subsequently to other states and countries in an effort to control water hyacinth (Center
1994). The genus is restricted to feeding on members of Pontederiaceae, with the
MWHW feeding mainly on the introduced water hyacinth; it can be found virtually
everywhere the host plant is present (Haag and Habeck 1991; Center et al. 1998).
The greenbug parasitoid (GBP) is an important natural enemy of many cereal
aphids. This species is known for the production of mummies, the bodies of
parasitized aphids that act as a protective case for the developing wasp pupa, and is
considered by many to be tolerant to cold temperatures (Elliott et al. 1999; Knutson et al.
1993; Wright 1995). However, this greenbug parasitoid is an insect and is just as
susceptible to pesticides despite the protective case of the immature form (Knutson et al.
1993).


98
to the safety of this compound and residual found on the plant does not pose the same
risk to the human population.
It is difficult to understand how a compound such as methionine can be
considered essential and deadly within the same organism. To understand this
dichotomy, an examination of the role of this compound and how it relates to
metabolism, development and reproduction is necessary.
Although the diet of the THW is lacking high concentrations of methionine, the
use of hexamerins may account for the levels needed for the biosynthesis of JH. The
larvae take in methionine, metabolizing what is needed and storing the rest for later on
during metamorphosis. In contrast, the larvae of the diamondback moth (Plutella
xylostella (Lepidoptera: Plutellidae)), feeding mainly on methionine-rich crucifers, lack
hexamerins with high methionine concentrations (Wheeler et al. 2000). The levels of
methionine encountered in a normal diet are below what the CAATCH1 proteins are
capable of processing and may also be affected by the presence of symbiotic bacteria that
is responsible for methionine oxidation in some insects (Gasnier-Fauchet and Nardon
1986a; 1986b). It is when the concentration exceeds the handling capacity of the midgut
that problems occur. The time it takes to digest material containing natural amounts of
methionine could be long enough for the CAATCH1 system to recover from exposure.
The difference between the artificial and natural diet LC50 for the THW (Figure 3-8)
appears to support the idea that bound methionine (i.e., incorporated into the diet and not
applied topically) takes longer to cause problems for the organism (if any) versus the
relatively quick kill associated with the free methionine present on the leaf surface. The


30
DVB/Carboxen/PDMS on a 2 cm StableFlex fiber, Supelco, Bellefonte, PA) was inserted
into the headspace of the sample bottle and exposed for 45 min. The fiber was then
removed from the headspace and inserted into the heated GC injector port at 220C
where the volatiles were thermally desorbed for 5 min.
Gas Chromatography: GC-FID and GC-Olfactometer
Separation was accomplished with a HP-5890 GC (Palo Alto, CA ) using either a
DB-wax column (30 m x 0.32 mm. i.d. x 0.5 mm, J&W Scientific; Folsom, CA) or
Zebron ZB-5 column (30 m x 0.32 mm. i.d. x 0.5 mm, Phenomenex, Torrance, CA).
Column oven temperature (for DB-wax) was programmed from 40 to 240C (or 40 to
265C for ZB-5) at 7 C/min with a 5 min hold. Helium was used as carrier gas at flow
rate of 1.55 mL/min. Injector and detector temperature were 220C and 290C,
respectively. A narrow-diameter injection port liner (0.75 mm.) was used to improve
peak shape and chromatographic efficiency for SPME thermal desorption. The entire
separation was conducted in the splitless mode. A GC splitter (Gerstel, Baltimore, MD)
split the column effluent between the FID and olfactometer (equipped with a high-
volume sniffing port, DATU, Geneva, NY) in a 1:2 ratio, respectively as described by
Bazemore et al. (6). A time-intensity approach was used to evaluate odor quality and
intensity at the sniffing port during the GC run. Assessors rated aroma intensity
continuously throughout the chromatographic separation process using a linear
potentiometer that supplied a continuous signal to the chromatographic software.
Retention times and verbal descriptors were recorded to permit aroma descriptors to be
coupled with computerized aroma time-intensity plots. Two olfactometry panelists were
trained in GC-sniffing with standard solution of 11 compounds typically found in orange
juice (ethyl butanoate, cis-3-hexenol, tran-2-hexenal, a-pinene, myrcene, linalool,


g of L-methionine/g of Leaf Material
L-methionine Concentration (%)
Y = 8.65E-04 + 4.76E-03X
R-Sq = 98.6 %
95% Confidence Intenal
+/- 2SE (SE=.001997)
Figure 3-4. Amount of L-methionine present on leaf surface after treatment.
Excised leaves were weighed, dipped into various concentrations
of L-methionine, allowed to dry, and then re-weighed. Difference
assumed to be the amount of L-methionine remaining on leaf
surface (T=22.43, and P<0.001).


64
sintered-glass, washed on the filter with cold ethanol and dried with the flow of nitrogen
gas. The headspace volatiles of recrystallized P-carotene were checked by SPME before
using. It was ready to use when no aroma active volatiles were detected.
Model Solutions
Acetone (Fisher Scientific, NJ) was chosen to dissolve the recrystallized P-carotene
because it was a polar solvent and would facilitate the transfer of P-carotene into the
model aqueous solution. One milligram of recrystallized P-carotene was dissolved in
acetone and diluted to citrate buffer pH 3.8 (citric acid 1.2 g., tripotassium citrate 0.6 g.
adjust pH to 3.8 by 1 N. NaOH, (Fisher Scientific, NJ)). Ten milliliters of the solution
were added into 40 ml vial with Teflon coated screw cap and wrapped with aluminum
foil kept in 35C for up to I month.
Analytical Methods
The headspace volatiles of the model solution were extracted by Solid Phase
Microextraction (SPME,50/30pm DVB/Carboxen/PDMS, Supelco). The solution was
equilibrated at 40C with gentle agitation (by stirring bar) for 45 min and then inserted
the SPME fiber to the headspace of the model solution in order to extract and concentrate
the headspace volatile by the fiber for another 45 min. The fiber was injected to GC (A
HP-5890 GC (Palo Alto, CA) with either a DB-Wax or ZB-5 column whose effluent was
split between an olfactometer or flame ionization detector (FID). Column oven
temperature was programmed from 40 to 240C at 7 C/min with a 5 min hold. The
aroma active compounds detected by GC-0 were identified from their aroma quality and
retention index by comparison with standards and confirmed by GC-MS (as describer in
chapter 4).


43
Results
Survivorship
Mortality of CPB larvae on treated excised eggplant leaves ranged from
approximately 20% for the 0.1% L-methionine treatment after 4 days, 80% mortality for
the 0.3% L-methionine treatment after 8 days of exposure and 100% for the remaining
concentrations with the highest dose of 1.0% L-methionine exhibiting complete control
of CPB in 3 days post treatment (Figure 4-1). Some mortality (50%) was observed for
the proline (1.0%) treatment while the Bit larval treatment mortality was similar to the
1.0% L-methionine treatment, resulting in 100% mortality after 5 days.
PROBIT analysis of a sample size of ntotai=L320 for 6 treatments (Control), 0.1%
L-methionine, 0.3% L-methionine, 0.5% L-methionine, 0.7% L-methionine and 1.0%
L-methionine) revealed an overall LC50 of 0.218% concentration for the CPB after 8 days
of exposure (Figure 4-2). The LC50 of 2.9% for 24 hours dropped to 1.1% after 48 hours
and to 0.22% after 72 hours.
Feeding and Development
Mean head capsule widths between treatments were found to be statistically
different (Figure 4-3). Four distinct groups were observed, with the Control, 0.1%
L-methionine and proline treatments forming the first group. The second group of
proline and 0.5% L-methionine were statistically the same and likewise the third group of
the 0.3% L-methionine, 0.5% L-methionine, and 0.7% L-methionine treatments. The
final group of Bit and 1.0% L-methionine treatments was statistically different from all
other treatments.


I dedicate this work to Karen, my wife and best friend. I thank her for putting up with
living as a graduate student for the last 5 years in fulfillment of my childhood dream of
being a Doctor. She has been my pillar of support, and I would not have made it this
far without her love and understanding.


80
enough to significantly lower the fecundity of the females and possibly interfere with
other behaviors such as mating.
During the course of this portion of the study, some anecdotal data were collected
based on personal observations. Predators (mainly arachnids) were observed on the
plants until the end of the experiment. Other insects also were observed feeding on plants
after treatments including piercing-sucking insects (i.e., aphids, coreids and cicadellids)
with foliage feeders such as caterpillars rarely encountered except found only on control
plants. Attempts to control predators via manual removal were unsuccessful, and
predation may have contributed to the observed decrease in CPB. Because predators
were present on all treatments, loss from predation was corrected with the use of Abbotts
formula. The presence of natural enemies indicates the selectivity of the L-methionine in
the field. The amount of methionine ingested by the predators was probably very small
because they fed on other insects not plant material.
Another set of observations on the safety of L-methionine was the exposure of
potted eggplants to high (1.0% methionine in distilled H2O solution). In total, five plants
were sprayed daily with the methionine solution and compared to five plants sprayed
with water alone for 14 days. The only difference in the plants was the browning of the
leaf tips and edges of the methionine sprayed plants. This also was seen in the excised
leaf experiments with THW and CPB. A possible reason for this occurrence was the
excess sulfur in the methionine might have burned the leaves. As mentioned earlier, the
concentration was very high and also applied daily. Applications of the same
concentration did not affect the plants in the field plots, indicating that treatments
conducted at 2-week intervals would be safe for the plant.


LC50 (% Concentration)
65
1.4
1.06
(0.92-1.4)
0.8
0.6
0.4
0.2
0
1.20
(0.95 2
24h
48h
72h
Overall (240h)
L-methionine
L-methionine (Buffered)
D-methionine
Beta-alanine
0.41(0.21-0.74)
0.44 (0.39 0.48)
0.50 (0.43 0.59)
0.33 (0.29 -0.38)
0.35 (0.28- 0.42)
0.32 (0.27 0.36)
0.34 (0.27-0.41)
(0.16-0.22)
Figure 5-9. Concentrations (%) resulting in 50% mortality (LC50) of yellow fever
mosquito larvae exposed to various amino acids after 10 days
(nT<#ai=240 for each amino acid). Number range following value is the
95% confidence limits. Proline and L-leucine were also tested but did
not exhibit sufficient mortality to allow for Probit Analysis.


LC50 (% L-methionine Concentration)
30
24h 48h 72h Overall (216h)
Figure 3-8. Concentrations (%) of L-methionine required for the mortality of 50%
of test population of tobacco homworm after 9 days exposure
(nTota)~L540; n=180 for 3.0% L-methionine 10.0% L-methionine,
n=200 for remainder). Number range following value is the 95%
confidence limits. Determination of LC50 was performed using
PROBIT Version 1.5 (Ecological Monitoring Research Division,


32
spectrometer (Finnigan, Palo Alto, CA) equipped with a DB5, 60M x 0.25 mm I.D.,
capillary column (J&W Scientific, Folsom, CA). The injector temperature and transfer
line temperature were 200 and 250 C, respectively. Helium was used as the carrier gas
at 1 ml/min. The oven temperature program consisted of a single thermal gradient from
40 to 275 C at 7C/min. The MS was set to scan from mass 40 to 300 at 2.0 scans/s in
the positive ion electron impact mode. The ionization energy was set at 70 eV.
Aroma Peak Identification
Initial identification was based on the combination of matches with standardized
alkane retention index values (Kovats Index) using two dissimilar column materials
(e.g., DB-wax and ZB-5) and aroma characteristics. If the aroma component was
sufficiently concentrated, fragmentation patterns were compared with library spectra
(NIST 2002 and Wiley (6th Edition) databases using the spectral fit criterion. Only those
compounds with spectral fit values equal to or greater than 800 were considered as
possible identification candidates. Whenever standards could be obtained, they were
used as a confirmation of identification, by comparing the resulting fragmentation
pattern, retention index value and aroma descriptor (88).
Results and Discussion
Extraction and Concentration of Juice Norisoprenoids
Solid phase microextraction (SPME) was used to extract and concentrate orange
juice volatiles because it is a rapid, solventless headspace sampling technique (<5). When
solvent extraction was used, early eluting peaks were obscured by the large quantities of
solvent. Early eluting aroma peaks such as acetaldehyde have been shown to be
important in orange juice flavor (89) but could not be examined using GC-0 in solvent-
extracted samples. Although solvent extraction would not have presented a problem in


11
The Cation-Anion Modulated Amino Acid Transporter With Channel Properties
(CAATCHl) System
Cation-Anion modulated Aminoacid Transporter with Channel properties
(CAATCHl) is a recently cloned insect-membrane protein isolated from larval
midgut/hindgut nutritive absorptive epithelium. This membrane protein exhibits a unique
polypeptide and nucleotide sequence related to, but different from, mammalian Na+-, Cf-
coupled neurotransmitter transporters (Feldman et al. 2000). Using a unique PCR-based
strategy, the gene encoding CAATCHl was cloned from the digestive midgut of THW
larvae. The unique biochemical, physiological, and molecular properties of CAATCHl
indicate that it is a multifunction protein that mediates thermodynamically uncoupled
amino acid uptake, functions as an amino acid-modulated gated alkali cation channel, and
is likely a key protein in electrolyte and organic-solute homeostasis of pest insects (Quick
and Stevens 2001). In the presence of no amino acids, the cations K+ and Na+ are
transported through the membrane via the channel (Figure 2A). When exposed to
proline, the amino acid is transported through the membrane with an increase in cation
flow, especially Na+ (Figure 2B). However, when exposed to methionine, the amino acid
transport is stopped and cation flow is altered, mainly the increased flow of K+ and the
decreased flow of Na+ (Figure 2C). The CAATCHl system works in alkaline conditions,
at a pH optimum ~ 9.5. This alkaline condition is found in the midgut of several species
(Nation 2001) and has been attributed to a variety of causes, from the detoxification of
plant allelochemicals to amino acid uptake (Giordana et al., 2002; Leonardi et al. 2001).


BIOGRAPHICAL SKETCH
Kanjana Mahattanatawee was bom in Bangkok, Thailand on 17 May, 1965. She
received a B.Sc. in biology with a major in Microbiology in 1988 from Sri-Nakharinwirot
University, Thailand. She continued to pursue her Master of Science degree in the area
of industrial microbiology at the Department of Microbiology, Chulalongkom University,
Bangkok Thailand from 1988-1991. From 1991-1992 she worked as a researcher, in the
Department of Microbiology, Chulalongkom University, Thailand. In 1992-1993
Kanjana was awarded a UNESCO scholarship to earn her Diploma in Microbiology and
Biotechnology from Osaka University, Japan. Kanjana was appointed to a position as
Lecturer, Department of Food Technology, Faculty of Science, Siam University from
1993-1999. From 1995-1997 she was an adjunct lecturer, Faculty of Environment and
Natural Resource, Mahidol University. Kanjana conducted research and taught two
microbiology courses (Industrial Microbiology and Fermentation Technology) for
undergraduate students at the Faculty of Science, Siam University, Bangkok, Thailand.
She was awarded a scholarship from Siam University to pursue her Ph.D. In
Spring 1999, she enrolled in the graduate program at the Department of Food Science and
Human Nutrition at the University of Florida under Dr. R.L. Rouseff s supervision. She
considers herself very fortunate to be enrolled in one of the greatest graduate programs in
flavor chemistry, with excellent scientists who are a pleasure to work with. She
completed her research for her Ph.D. degree at the Citrus Research and Education Center
(CREC) in Lake Alfred, Florida.
84


10
of the two antimetabolites. This process actually increases the nitrogen intake from the
foodstuff (from the increase of ammonia) (Rosenthal et al. 1976; Rosenthal et al. 1977).
Another insect, the tobacco budworm (Heliothis virescens (Fab.) (Lepidoptera:
Noctuidae)) was able to metabolize far more canavanine then the bruchid beetle larva
ever takes in during its development, suggesting that generalists may have more than a
single detoxification mechanism for compounds they may encounter (Berge et al. 1986).
Metabolism of L-canavanine by the tobacco budworm was attributed to the gut enzyme
canavanine hydrolase, and may have been the result of feeding on canavanine-containing
plants of the Fabaceae (Melangeli et al. 1997).
Essential Amino Acids
In despite of the extensive toxicological research conducted on nonprotein amino
acids, another group of amino acids, the essential ones, has been overlooked. One reason
this avenue for research has not been pursued is that we do not want to give pests
convenient access to an integral part of their diet. The fear of creating a super insect
(that has been provided with compounds that actually aid in its development) is a rational
one. Mittler (1967a; 1967b) found an increase in gustation in Myzus persicae (Sulzer)
(Hemiptera: Aphididae), with amino acid levels as low as 0.2% concentration in a
sucrose solution. Likewise, Sugarman and Jakinovich (1986) found increased gustatory
response to both D-and L-methionine by Periplaneta americana (L) (Blattodea:
Blattidae) adults. Another reason that essential amino acids have not been examined for
use as a pesticide is the knowledge regarding the limited mode of action these compounds
could be involved with (i.e., an active site or systemic response). Recent studies on the
membrane proteins of insects show the possibility of a biophysiological system that can
be exploited for insect control with certain essential amino acids.


35
GC-Olfactometry
In this study, a total of 59 aroma active components were detected in SPME
headspace samples from fresh orange juice (orange juice group 1) Since the primary goal
of this study was to determine if additional aroma active norisoprenoids were present in
orange juice, GC-0 was employed primarily in the region where p-ionone and other
norisoprenoid standards eluted. Using standards of P-cyclocitral, a-ionone, P-ionone
and P-damascenone, the retention time region was established between 12 and 20 min,
and the resulting aromagram and concurrent chromatogram is shown in Fig. 4-1.
As noted in Fig. 4-1, four aroma peaks corresponding to peaks 5, 19, 21 and 23
were observed at the identical retention times as P-cyclocitral, P-damescenone, a-ionone
and P-ionone respectively. It is also apparent from the relative intensities shown in Table
3-1, that these potential norisoprenoid peaks were among the more intense aromas.
Beta-ionone was the most intense and P-cyclocitral was the weakest aroma compound of
all the four potential norisoprenoids observed. When the samples were rerun on a DB-
wax column the four aroma peaks also were found at retention index values that
corresponded with the four potential norisoprenoids. Furthermore, the aroma quality of
each juice norisoprenoid corresponded exactly with the aroma description of standards.
Since these compounds have the same retention behavior on two very dissimilar
chromatographic columns and also have the same aroma quality as standards, they are
probably P-cyclocitral, P-damescenone, a-ionone and P-ionone respectively. This
represents the first time that p-cyclocitral, and a-ionone have been reported in orange
juice. Beta-damescenone had recently been reported in heated orange juice but its
identity was not confirmed by supporting instrumental methods (65).


89
100
80
g
13 60
S 40
20
0
Figure 7-2. Mortality of Coleomegilla maculata adults after exposure to L-
methionine treated cotton plant leaves infested with aphids. Data
corrected for control mortality using Abbotts formula.
Control- ND
1.0% L-methionine ND
Survivorship of 1.0%L-methionine Grp> Control Grp
4 5 6 7 8 9
Days After Exposure
10 11 12


BIOGRAPHICAL SKETCH
Lewis Scotty Long was bom in Calhoun, Georgia on August 20,1971. He
graduated from Madisonville High School (Madisonville, Tennessee) in May 1989. On a
biology scholarship, Lewis attended Middle Tennessee State University (MTSU), where
he earned his BS in May 1994. On graduation, he took a job as an aquatic biologist for
Aquatic Resources Center (Franklin, Tennessee). Lewis worked there specializing in
taxonomy of mayflies, stoneflies, caddisflies, and freshwater molluscs (snails and
mussels). While still employed at Aquatic Resources Center, he started his graduate
studies in 1996 at MTSU and continued the work he had started during his undergraduate
years. In May of 1999, Lewis graduated with his MS. After receiving his MS, Lewis
moved to Florida and entered the PhD program at the University of Florida, Department
of Entomology and Nematology. He worked with Dr. Bill Peters (Florida A&M
University) on the worldwide taxonomic revision of an understudied group of mayflies.
However, Dr. Peters unexpectedly passed away in 2000, and Lewis took this unfortunate
event as a chance to broaden his expertise in entomology. In 2000, he took a part-time
job with Drs. James Cuda and Bruce Stevens on research that was in the patent process.
This was the research that Lewis undertook for his dissertation. Lewis also served as a
teaching assistant for the department for classes such as Bugs and People, Life Sciences
for Education Majors, Principles of Entomology, and Medical and Veterinary
Entomology. He served as primary instructor for Insect Classification and Immature
114


26
Figure 3-5. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nxotai-480) in artificial diet. Data were adjusted using
Abbotts formula to account for control mortality. Note the overlap in trend
lines for the 3.0% L-methionine-10.0% L-methionine concentrations after
Day 1 and the 0.3% L-methionine and 0.5% L-methionine treatments from
Day 1 to Day 10.


28
Objectives
Since orange juice has so many carotenoids that could serve as precursors for a
wide range of norisoprenoids, the objective of this study was to determine if more than
one aroma active norisoprenoid was present in fresh or heat-treated orange juice. If
additional norisoprenoids are found, they should be characterized and identified. (See
Objective #2)
Materials and Methods
Orange Juice Samples and Processing
Late-season Valencia oranges (from Haines City Citrus Growers Association,
Haines City Florida) were juiced using an FMC juice extractor at the Citrus Research and
Education Center (CREC), Lake Alfred, Florida. The oranges were juiced using a
commercial FMC juice extractor model 291 with standard juice settings. An FMC model
035 juice finisher (FMC Corp., Lakeland, FL) was used with a 0.02 inch screen. The
finished juice had a Brix value of 11.7, an acid content of 0.67% citric acid, a Brix/acid
ratio of 17.5 and an oil level of 0.0196%. The freshly squeezed juice was divided into
three groups. In Group 1, fresh orange juice was immediately chilled and NaCl (36 g/100
mL of juice) was added to inhibit enzymatic reactions. In Group 2, fresh orange juice
was pasteurized using UHT/HTST lab Microthermics tubular pasteurizer Model 25
(Microthermic Corp., Raleigh, NC) at 195F (90.5C), held for 12 seconds and filled at
41F (5C). The oil level of the pasteurized juice was 0.0168%. In Group 3, fresh
orange juice was concentrated to 65 Brix using a thermally accelerated short-time
evaporator (TASTE) built by Cook Machinery, Dunedin, Florida. The concentrate was
then reconstituted to 11.73 Brix by diluting with deionized water, but without restoring
volatiles.


23
Carotenoid Identification
Shown in Figure 3-4 is an overlay of peaks 4, 12 and 24. The height of their
spectra corresponds to their relative peak heights since the spectra were taken from the
apex of each peak. These were chosen to show the range and diversity of these spectra
which are not conveyed when just tabulated peak maxima are tabulated. The shape of the
absorbance band as well as the location of the absorbance maxima are all highly
characteristic of individual carotenoids. This information taken with retention time can
Figure 3-4. Absorbance spectra for (3-carotene (a), Ieutoxanthin (b), and neoxanthin (c),
peak 24, 12 and 4 respectively.
be used to identify specific carotenoids, especially if their spectral and chromatographic
characteristics have been reported elsewhere. The spectra and relative retention times of
a-, (3-carotene, a-, (3-cryptoxanthin and neoxanthin matched their published values and
were used as confirmation of the presence of these peaks in orange juice. It should be
pointed out that these five carotenoids have been previously reported in orange juice (77)


19
Figure 3-1. Rearing chamber for tobacco homworm and Colorado potato
beetle larvae used in the artificial and excised leaf diet tests.
Hardware cloth stage supporting the leaf allowed for easy
clean up and minimized disease problems by preventing
larvae from coming in contact with fecal material (paper liner
not shown).


40
an excellent candidate for the evaluation of L-methionine as a possible means of
controlling this devastating pest.
Because little information is available on the insecticidal properties of
L-methionine, several baseline experiments were necessary to determine what
concentrations of this amino acid to test. Therefore, it was necessary to test L-methionine
and CPB interaction in a variety of ways including survivorship of both larvae and adults,
development of larvae when exposed to different concentrations of the amino acid, and
preference tests. The purpose of this portion of this study was to conduct bioassays to
determine if exposure to L-methionine was detrimental to the survival and development
of the CPB and to determine if L-methionine could be used to control this species.
Materials and Methods
Eggs of CPB were obtained under UDSA permit from the insectary of the New
Jersey Department of Agriculture and held in 26.4L x 19.2W x 9.5H (cm) clear plastic
boxes with a hardware cloth (to facilitate cleaning) and held at 27 C, 60% relative
humidity and 16L/8D photoperiod in FRIUs (Figure 3-1). Excised eggplant leafs were
placed in the chambers with the neonates and they were allowed to feed for 2 days after
eclosin before being transferred to experiments. A camel hair brush was used for
transferring the neonates to minimize the risk of damaging the larvae.
Survivorship
Larvae and adults of the CPB were tested in preliminary experiments with the
highest concentration (1.0% L-methionine (wt/wt)) observed in tests done on the THW in
the previous chapter. The diet for the larvae and adults consisted of excised eggplant
leaves (,Solarium melongena L.,Classic variety (Family: Solanaceae)) from plants
grown and maintained at the University of Florida, Department of Entomology and


ACKNOWLEDGMENTS
I would especially like to thank to Dr. Russell L. Rouseff, my supervisor, for his
friendship and guidance throughout my graduate program. He provided me with
intellectual, thoughtful discussion; encouragement; and time. Without his help and
support I would not have come from Thailand and continued at the University of Florida.
His high ethical standards and philosophical views will never be forgotten. I would like
to thank Dr. R.M. Goodrich for all of her support, guidance and friendship. She was very
kind and helpful whenever I needed her help. I also thank the other members of my
supervisory committee, Dr. D.H. Powell and Dr. M.R. Marshall, Jr, for their kindness and
valuable, thoughtful discussion toward my research. Especially warm thanks go to Dr.
Kevin L. Goodner for his advice and support on instrumental. My special thanks go to
John Smoot and Dr. Filomina Valim who assisted me any time I needed them. My
appreciation goes to my dear friend Dr. Fahiem El-Borai Kora who assisted me whenever
I needed his help. I extend my appreciation to all my friends at Citrus Research and
Education Center (CREC) in Lake Alfred for their friendship and support. I would like
also to thank my professors back in Thailand expecially Dr. Twee Hormchong, who
taught me what good scientists and teachers are. My great thanks go to my beloved
family. My lovely wonderful mother, father, brothers and sisters always encouraged me
to follow my dream, and without their love and support, I could never be who I am today.
Finally, my appreciation goes to the Siam University for giving me the financial support
necessary to obtain my Ph.D from the University of Florida.
IV


51
where it was 0.86. One way analysis of variance (ANOVA) show that there are a highly
significant differences (P<0.01) among orange juice spiked with different concentration
of standard a-ionone (Fig. 5-3) and P-ionone (Fig. 5-4). However there were no
significant differences within sample. In contrast for P-cyclocitral (Fig. 5-2), there were
no significant differences between the two samples spiked with standard P-cyclocitral at
concentration 0.54 and 1.1 ppm. This suggests that an error occurred during analysis
with at least one data point.
A quadrupole MS provides at least lOx greater sensitivity in the SIM mode than an
ion trap MS under the same conditions and was thus used to quantify P-damascenone in
fresh and pasteurized juice when the ion trap failed to detect this compound. The
calculated slope from the fresh juice data was also employed to determine the
concentration of P-damascenone in pasteurized juice as it was thought the matrix effects
would be the same for both samples.
Figure 5-3. Standard addition data for a-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. a-ionone concentration.


20
Department of Entomology and Nematology green and shade houses. Excised leaves
were dipped in solutions of deionized H2O containing different concentrations of
methionine; depending on the experiment and exposed to larvae in the same rearing
chambers as the artificial diet trials under the same conditions. Survivorship data were
pooled from several different trials for data analysis.
In total, 64 potted eggplants were used for the whole-plant portion of the study.
Plants were held in FRIUs under the same conditions as the artificial and excised leaf
trials, in 38H x 15D (cm) plexigls cylinders (Figure 3-2). Four THW neonates were
placed on each plant for a total of 64 larvae (16 replicates) per treatment (nx0tai=256
larvae). The treatment of L-methionine was applied to the test plants (using a hand-held
sprayer calibrated to deliver approximately 10 mL of solution to each plant) before the
addition of larvae.
Feeding and Development
To test L-methionine on the developmental rates of THW, larvae were exposed to
excised eggplant leaves dipped in solutions containing the same concentrations of L-
methionine used in the artificial diet trials. Additional treatments of proline (1.0%) and
Bt-kurstaki (Dipel 86% WP at 3.5 grams/liter; Bonide, Oriskany, NY) were included as
positive and negative controls, respectively. Leaves were scanned photometrically using
the Cl 203 Area Meter with conveyor attachment (CID, Inc.; Camas, WA) to measure
leaf consumption before and after exposure to larvae. The difference in leaf areas
resulting from the missing leaf tissue was assumed to be the amount eaten by the
developing larvae. Larval head capsule widths were measured at the time of death or the


48
(Error Bars @ 95%; F(0.05)7,312=1.14;F=40.1; P<0.001)
400
Control 0.1% 0.3% 0.5% 0.7% 1.0% Proline Btt
Figure 4-4. Total leaf area consumed by Colorado potato beetle larvae exposed
to excised eggplant leaves treated with various concentrations of L-
methionine (nTOtai=:320). Proline (1.0%) and Btt were included for
comparison as positive and negative controls. Error bars denote 2
SE. Bars within treatments having the same letter are not statistically
different (Tukeys MST, P0.001).


46
a-ionone, and (3-ionone) was added separately to the orange juice sample to obtain the
final concentration of each norisoprenoid from 0 to 2 ppm. Beta-damascenone was the
only exception; its added concentrations ranged from 0 to 0.02 ppm. Sampling was
accomplished by adding a 10 mL aliquot of the juice to a 40 ml glass vial containing a
micro-stirring bar sealed and a Teflon coated septa. Samples were equilibrated at 40C
for 45 minutes and gently stirred before a SPME fiber was inserted into the headspace of
the sample bottle and exposed for another 45 min. The fiber was then removed from the
headspace and inserted into the GC-MS. GC conditions were the same as in Chapter 4.
Each sample was prepared and injected at lease twice. Quantitative measurements were
made using integrated peak areas from selected ion chromatograms. The ions chosen to
reconstruct these single ion chromatograms were at m/z 137, 177, 177, and 190 and were
fairly unique for P-cyclocitral, P-ionone, a-ionone, and P-damascenone respectively.
The latter m/z values corresponded to the respective molecular ion of P-damascenone.
In order to quantify the low levels of P-damascenone, a quadrupole MS (Agilent
5973 Network Mass Selective Detector, Agilent Technologies, CA) was employed using
selected ion monitoring (SIM) mode at m/z 190. It was equipped with HP an Innowax 30
m x 0.25 pm x 0.25 pm capillary column (Agilent/J&W HP Innowax, Scientific
Instrument Services, Inc., NJ) and autosampler (Gerstel Multi Purpose Sampler MPS2,
Gerstel Inc., MD). The oven temperature program consisted of two ramps from 90 to
160C at 6C/min and from 160C to 250C at 120C/min (in order to shorter the GC
running time after the P-damascenone was eluted). Each sample was analyzed from the
response at m/z 190. A graph of SIM 190 peak area versus concentration was prepared


16
In each case carotenoids have been shown to be their precursors. Citrus carotenoids have
been examined using a variety of chromatographic techniques such as column and thin
layer chromatography, TLC, and more recently high performance liquid chromatography,
HPLC (68, 72, 78). Once separated, individual carotenoids have been identified
primarily by their unique three fingered visible absorbance patterns. In orange juice,
most oxygen containing carotenoids are esterified with C12-C18 fatty acids (74) thus
increasing their size and structural complexity. The most common practice is to de-
esterify (hydrolyze) these esters so that each carotenoid will elute as a single peak rather
than several smaller peaks with various fatty acids attached. However, even with
hydrolysis, the large numbers and subtle structural differences of orange juice carotenoids
provide a severe separation challenge. To complicate matters further, carotenoids are
sensitive to heat, light and oxygen, thus artifacts are readily formed during sample
preparation and/or analysis steps. HPLC equipped with a photodiode array detector is the
preferred analytical technique of choice to separate and quantify carotenoids without
artifact formation. Both normal phase and reverse phase chromatography have been
employed to separate these plant pigments, but the reverse phase approach offer the most
advantages. The most common reverse phase column is C-18 and most of the early
carotenoid separations employed this column. However, in recent years a C-30 reverse
phase column has been developed especially for carotenoid separations. Several
investigators (68, 75, 76) have employed this column with ternary solvent gradient and
photodiode array detector to isolate and identify the complex mixture of carotenoids in
orange juice.


EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL
CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL
IMPORTANCE
By
LEWIS SCOTTY LONG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004

Copyright 2004
by
Lewis Scotty Long

ACKNOWLEDGMENTS
I thank Jim Cuda and Bruce Stevens for giving me the financial and intellectual
freedom that made this work possible. I want to thank Jim for housing me in his lab and
providing the facilities to perform this work, and Bruce for allowing me to take his initial
work and elaborate on it as well as including me as a co-inventor of the research
presented. Most of all, I would like to express my sincere appreciation to Judy Gillmore.
Without her support and help this research would not have been completed. Judy was
integral in every aspect of this endeavor and put up with more than her fair share of my
research. I extend heartfelt thanks to George Gerencser, James Maruniak, Simon Yu,
and Susan Webb for serving as members of my supervisory committee. I would like to
also thank Jim Lloyd, Jerry Butler, and Carl Barfield for all the experiences and
knowledge shared. Finally, I want to express my deepest, eternal gratitude to my fellow
graduate students Jim Dunford and Heather Smith, for providing support and guidance
that only colleagues, intellectual equals, and close friends can give. I can only hope to
repay them for their help by providing the same amount of support for their endeavors as
they did mine.
m

I dedicate this work to Karen, my wife and best friend. I thank her for putting up with
living as a graduate student for the last 5 years in fulfillment of my childhood dream of
being a Doctor. She has been my pillar of support, and I would not have made it this
far without her love and understanding.

TABLE OF CONTENTS
Elge
ACKNOWLEDGMENTS iii
LIST OF FIGURES viii
ABSTRACT xi
CHAPTERS
1 THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE
CONVENTIONAL PESTICIDES THE ONLY ANSWER? 1
Introduction 1
Importance of IPM in Florida and Surrounding States 2
Problems Associated with Pesticide Misuse 4
Biorational Compounds- An Alternative to Chemical Pesticides 5
2 HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL
INSECT PESTS 7
Non-Protein Amino Acids 7
Essential Amino Acids 10
The Cation-Anion Modulated Amino Acid Transporter with Channel
Properties (CAATCH1) System 9
Methionine 13
Research Objectives 16
3 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT
OF THE TOBACCO HORNWORM, Manduca sexta, UNDER
LABORATORY CONDITIONS 17
Introduction 17
Materials and Methods 18
Diets and Survivorship 18
Feeding and Development 20
Preference Tests 22
Data Analysis 24
Results 24
v

Diets and Survivorship 24
Feeding and Development 31
Choice Tests 31
Discussion 36
4 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF
THE COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER
LABORATORY CONDITIONS 39
Introduction 39
Materials and Methods 40
Survivorship 40
Feeding and Development 41
Preference Tests 41
Data Analysis 42
Results 43
Survivorship 43
Feeding and Development 43
Preference Tests 47
Discussion 47
5 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY
CONDITIONS 52
Introduction 52
Materials and Methods 53
Bioassay 53
Growth and Development 54
Data Analysis 56
Results 56
Bioassay 56
Growth and Development 59
Discussion 66
6 EVALUATION OF L-METHIONINE UNDER NATURAL FIELD CONDITIONS69
Introduction 69
Materials and Methods 70
Preliminary Investigation of Silwet L-77 and L-methionine 70
Plot Design 70
Fruit Yield 71
Pest Introduction 71
Data Analysis 74
vi

Results 74
Effects of L-methionine and Silwett L-77 on Colorado Potato Beetle Adults
Under Laboratory Conditions 74
Effects of L-methionine and Silwett L-77 on yield 74
Survival of CPB larvae 74
Discussion 78
7 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
NON-TARGET SPECIES 82
Introduction 82
Materials and Methods 84
Coleomegilla maculata 84
Neochetina eichhorniae 85
Lysiphlebus testaceipes 86
Data Analysis 86
Results 87
Coleomegilla maculata 87
Neochetina eichhorniae 87
Lysiphlebus testaceipes 87
Discussion 87
8 SUMMARY AND DISCUSSION 96
LIST OF REFERENCES 102
BIOGRAPHICAL SKETCH 114
Vll

LIST OF FIGURES
Figure Pagi
2-1. The CAATCH1 system identified from the midgut of the tobacco homworm 15
3-1. Rearing chamber for tobacco homworm and Colorado potato beetle larvae used in
the artificial and excised leaf diet tests 19
3-2. Setup for whole plant studies involving tobacco homworm 21
3-3. Chambers used for tobacco homworm and Colorado potato beetle preference tests 23
3-4. Amount of L-methionine present on leaf surface after treatment 25
3-5. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (n-rotar^O) in artificial diet 26
3-6. Survivorship of THW larvae exposed to various concentrations of L-methionine
(nTota 1,540) on excised eggplant leaves 28
3-7. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nrotai=256) on whole plants 29
3-8. Concentrations (%) of L-methionine required for the mortality of 50% of test
population of tobacco homworm after 9 days exposure (nT0tai=1,540; n=180
for 3.0% L-methionine 10.0% L-methionine, n=200 for remainder) 30
3-9. Mortality of tobacco homworm larvae exposed to various concentrations of L-
methionine (ntota 160) on excised eggplant leaves for feeding and
development trials 32
3-10. Mean head capsule widths of tobacco homworm larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nTOtai=320) 33
3-11. Total leaf area consumed by tobacco homworm larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nrotai=320) 34
3-12. Mean leaf consumption by tobacco homworm in the preference tests 35
4-1. Mortality of Colorado potato beetle larvae exposed to excised eggplant leaves treated
with various concentrations of L-methionine (nTOtai=560) 44
4-2. Concentrations (%) of L-methionine concentrations required for the mortality of
50% of the test population of Colorado potato beetle after 8 days exposure
(niotai=220) 45
vui

4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to excised
eggplant leaves treated with various concentrations of L-methionine
(niotai=320) 46
4-4. Total leaf area consumed by Colorado potato beetle larvae exposed to excised
eggplant leaves treated with various concentrations of L-methionine
(nrotar=320) 48
4-5. Mean leaf consumption by Colorado potato beetle in the preference tests 49
5-1. Bioassay setup for yellow fever mosquito larvae exposed to various concentrations
of amino acids and Bti 55
5-2. Mortality of yellow fever mosquito larvae exposed to various concentrations of
L-methionine (nTOtai=240) 57
5-3. Mortality of yellow fever mosquito larvae exposed to various concentrations of
D-methionine (nrotai=240) 58
5-4. Mortality of yellow fever mosquito larvae exposed to various concentrations of Tris-
buffered L-methionine (nrotai~240) 60
5-5. Mortality of YFM larvae exposed to various concentrations of Proline (nrotai=240) 61
5-6. Mortality of yellow fever mosquito larvae exposed to various concentrations of
L-leucine (nTotai=240) 62
5-7. Mortality of YFM larvae exposed to various concentrations of Beta-alanine
(nTotai=240) 63
5-8. Mean head capsule widths of yellow fever mosquito larvae exposed to various Tris
buffered (7.0 pH) concentrations of L-methionine (niotai=320) 64
5-9. Concentrations (%) resulting in 50% mortality (LC50) of yellow fever mosquito
larvae exposed to various amino acids after 10 days (nrotar=240 for each amino
acid) 65
6-1. Overview of the design layout used to study the effects of L-methionine and Silwett
L-77 solutions on yield of eggplant 72
6-2. Weed Systems, Inc. KQ 3L CO2 backpack back sprayer used for application of
L-methionine and Silwett L-77 solutions 73
6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant leaves treated
with L-methionine and the adjuvant Silwett L-77 (nTotai=120) 75
6-4. Effects of L-methionine and Silwett L-77 on eggplant yield (A) and mean weight
in grams of fruit (B) from 09 June to 31 August 2001 76
IX

6-5. Mortality of Colorado potato beetle larvae on eggplants treated with L-methionine
and Silwett L-77 87
7-1. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated
artificial diet 88
7-2. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated
cotton plant leaves infested with aphids 89
7-3. Feeding scars on water hyacinth {Eichhornia crassipes) leaf after exposure to
Neochetina eichhorniae adults 90
7-4. Mortality of Neochetina eichhorniae on treated water hyacinth leaves 91
7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated with
L-methionine and Proline 92
7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with
L-methionine 93
x

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL
CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL
IMPORTANCE
By
Lewis Scotty Long
May, 2004
Chair: James P. Cuda
Cochair: Bruce R. Stevens
Major Department: Department of Entomology and Nematology
Integrated pest management (IPM) strategies were developed in an effort to
control pests with fewer pesticides. However, because of the misuse of pesticides and the
failure to adopt IPM practices pesticide use is higher than ever. An alternative to
conventional broad-spectrum pesticides is the use of biorational compounds; those that
pose minimal risk to the environment and are specific to the target pests.
The recent discovery of the CAATCH1 system in the midgut of the tobacco
homworm (THW), Manduca sexta, has revealed a novel means to control certain insect
pests. This membrane protein works in alkaline conditions as both an amino acid
transporter and also independently as a cation channel. However, the amino acid
L-methionine blocks amino acid transport thus altering the ion flow.
xi

Bioassays involving the tobacco homworm, Colorado potato beetle (CPB),
Leptinotarsa decemlineata, and the yellow fever mosquito (YFM), Aedes aegypti, were
conducted to determine the insecticidal properties of this compound. L-methionine in
artificial and natural diets resulted in the mortality of 50 to 100% in concentrations of
0.3% and higher for THW and CPB. Feeding rates and larval development also were
affected with reduced levels (>0.1%) of L-methionine. Bioassay trials involving YFM
larvae were similar, concentrations greater than 0.1% L-methionine produced mortality
rates of 70 to 100%. All three species responded better to higher concentrations of L-
methionine than to Bacillus thuringiensis, the most commonly used and commercially
available biorational pesticide.
Field trials and non-target tests also were performed to determine L-methionine
effectiveness under natural settings and safety to other organisms. Eggplant yield was
not reduced by the application of L-methionine, which effectively controlled CPB larvae
on the plants. Furthermore, several beneficial insects that were tested (a predator, a
herbivore, and a parasitoid) were not affected by the addition of L-methionine to their
diets.
Based on these results, L-methionine was found to be effective in controlling
selective agriculturally and medically important insect pest species, yet posed little threat
to the crop plants applied to or to non-target organisms. The use of L-methionine as a
pesticide, its relationship with insects and its possible use in delaying insecticide
resistance were also examined.
xu

CHAPTER 1
THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE CONVENTIONAL
PESTICIDES THE ONLY ANSWER?
Introduction
Integrated Pest Management (IPM), the sustainable approach to the management
of pest species using a combination of biological, chemical and cultural methods to
reduce economic, environmental, and public health risk, was a result of economic losses
associated with years of overuse of chemical control leading to resistance problems. The
use of IPM strategies have certainly decreased pesticide usage and encouraged the use of
methods that ensure a safer environment but many feel that it is not enough. After three
decades of research efforts in the United States, IPM as it was envisioned in the 1970s
was practiced on less than 8% of U.S. crop acreage based on Consumers Union
estimateswell short of the national commitment to implement IPM on 75% of the total
U.S. acreage by the end of the 1990s (Ehler and Bottrell 2000). This means that farm
practices have changed little since the national IPM initiative was established in 1994 to
implement biologically based alternatives to pesticides for controlling arthropod pests. It
should be noted that the low percentage of IPM practices on commercial U.S. farmland
may possibly be related to the lack of sufficient reporting means and actually may be
higher than believed when the local growers and homeowners are included. However,
the United States is considered the worlds largest user of chemical pesticides, accounting
for nearly 50% of total worldwide production and shows no sign of slowing (Deedat
1994). Pesticides remain the primary tool of pest consultants and farmers, because of the
lack of economic incentives to adopt alternative strategies that require more effort to
1

implement, produce unpredictable results, and require new knowledge (Barfield and
Swisher 1994; Ehler and Bottrell 2000).
Importance of IPM in Florida and Surrounding States
Considerable effort has been devoted to developing IPM programs in Florida
because of its unique pest problems and crop production systems, sensitivity to chemical
pollutants, and increased urbanization (Capinera et al. 1994; Rosen et al. 1996). The
necessity for developing IPM protocols for Floridas major plant and animal pests was
underscored in a new statewide initiative. In November 1999, the Institute of Food and
Agricultural Sciences (IFAS) at the University of Florida launched Putting Florida FIRST
Focusing IFAS Resources on Solutions for Tomorrow (Florida FIRST 1999). The
Florida FIRST initiative was created (with input from stakeholders) to define the role of
IFAS in shaping Floridas future in the 21st century. Increasing concerns (expressed
repeatedly by Floridas scientific community and the general public) about environmental
contamination, food safety issues, and human and animal health problems resulting from
the indiscriminate use (and often misuse) of pesticides are making existing methods for
pest management obsolete. Successful implementation of true IPM, as it was
envisioned by those who envisioned the original concept, will have the added benefit of
helping Florida ... enhance natural resources, provide consumers with a wide variety of
safe and affordable foods,... provide enhanced environments for homes, work places
and vacations, maintain a sustainable food and fiber system, and improve the quality of
life... (Florida FIRST 1999).
This effort to promote IPM programs in the state of Florida also benefits the
surrounding states. For example, solanaceous crops produced in the southeastern U.S.
(such as tomato, tobacco, eggplant, peppers and potato) are subjected to the same

3
defoliation and fruit damage from various lepidopteran and coleopteran pests that also
threaten Florida. The tomato pinworm [Keiferia lycopersicella (Walshingham)
(Lepidoptera: Gelechidae)], armyworms [Spodoptera spp. (Lepidoptera: Noctuidae)], the
Colorado potato beetle [Leptinotarsa decemlineata (Say) (Coleptera: Chrysomelidae)],
and homworms [Manduca spp. (Lepidoptera: Sphingidae)] are some examples of pests
that threaten both conventional producers and homeowners alike. For example, the
estimated loss from and the cost of control of the tobacco homworm, the number-one
pest in tobacco crops in Georgia, reached $1.5 (and $2.3 million), respectively, for the
years 1996-1997 (Jones and McPherson 1997). From 1992-1998, tomato, eggplant, and
pepper producing areas in the Southeast had a total of 1,247,000 pounds of endosulfan
applied over 270,000 acres (Aerts and Neshiem 1999; Neshiem and Vulinec 2001). The
cost of insecticides applied in Florida tomato production alone for 1993-1994 amounted
to approximately $ 1,052/hectare for a total of $2.1 million; and rose to $2550/acre,
totaling $103M for the 1996-1997 season (Aerts and Neshiem 1999; Schuster et al.
1996). The use of pesticides in Florida tomato production is high because tomatoes
account for 30% of the total vegetable-crop value and 13% of the total vegetable acreage
for the state, with 99% of production aimed toward the fresh market (Schuster et al.
1996). For Florida potato producers, the cost of applying pesticides from 1995-1996
was $11.5M, and 96% of total Florida eggplant-crop acreage was treated with chemical
insecticides (mainly methomyl and endosulfan) (Neshiem and Vulinec 2001). In addition
to the monetary cost of pesticide use, commonly used insecticides such as endosulfan and
fenvalerate show a high degree of toxicity to parasitoids of the tomato pinworm, thus
negating the benefits of predation by natural enemies (Schuster et al. 1996). These
figures may be the result of the more is better attitude of producers who want to avoid

4
all risk of insect damage by using more applications and stronger pesticides (Schuster et
al. 1996).
Problems Associated with Pesticide Misuse
The use of pesticides is not completely ruled out under IPM strategies, but rather
IPM encourages responsible use to minimize environmental harm and to protect human
safety and health (Deedat, 1994). However, the misuse (both intentional, in terms of
more is better; and unintentional, as in agricultural runoff) also has resulted in
resistance in some of the target pests. For example, surveys in North Carolina have
shown that the Colorado potato beetle has become resistance to fenvalerate, carbofuran,
and azinphosmethyl as a result of control failures in the field (Heim et al. 1990).
Resistance to insecticides has also been observed in more than 450 arthropod pests
(Romoser and Stoffolano 1998). Bills et al. (2004) found a 38% increase in the number
of registered compounds used as pesticides from 1989-2000, and also a 16% increase in
pesticide resistance of arthropod species worldwide.
Losses are not limited to agricultural systems alone. Across Africa for example,
populations of insecticide-resistant mosquitoes are the result of a variety of mechanisms,
including exposure to pesticide residues in agricultural runoff, mutation of target sites,
and migration of resistant populations into areas where there were no previous problem
(FIC-NIH 2003). Parts of southwest Asia have seen a resurgence of malaria in some
areas where it was considered eradicated (due to a combination of resistance and the
economics associated with control of mosquito vectors) (Deedat 1994). The importance
of this example becomes even more relevant when one considers that one million
individuals die every year as a result of malaria, with upwards of 500 million cases per
year (Centers for Disease Control 2003). The existence of other mosquito-bome diseases

5
such as Dengue fever, yellow fever, and West Nile virus to name just a few, put countless
millions more at risk. It would be dangerous to think that these diseases only occur in
underdeveloped countries and not the United States. Integrated Pest Management
practices also should be adopted for controlling the medical and veterinarian important
insect vectors of these and other diseases.
Biorational Compounds: An Alternative to Traditional Chemical Insecticides
One way to reduce this reliance on traditional chemical pesticides and delay
resistance is by increasing the variety and use biorational compounds. Biorational
compounds are effective against selected pest species but are innocuous to nontarget or
beneficial organisms; and have limited affect (if any) on biological control agents
(Stansly et al. 1996). Biorational compounds include detergents, oils, pheromones,
botanical products, microbes, and systemic and insect growth regulators (Perfect 1992;
Wienzierl et al. 1998). Their safety lies in the low toxicity of the compound to nontarget
organisms and the compounds short residual activity in the field. For example, Bacillus
thuringiensis isrealensis (Bti) currently is one of the most widely used microbial
pesticides for controlling aquatic dipteran pests (i.e., mosquitoes and black flies) because
of its selectivity to this group and minimal nontarget effects (Glare and OCallaghan
1998). However, resistance to Bt products has occurred in many species of lepidoptera
from overuse of Bacillus thuringiensis kurstaki, and in some mosquito species to Bti, thus
showing the need for alternatives to these compounds that are still effective (Brogdon and
McAllister 1998; Marrone and Macintosh 1993). In addition to resistance, other
problems are associated with the use of microbial control agents. Cook et al. (1996)
discussed potential hazards, not properly identified in the planning stages, of
displacement of native microorganisms, allergic responses in susceptible humans and

6
animals, and eventual toxicity to nontarget organisms. Because of these problems,
alternatives are needed to prevent another crisis like the one from which IPM originally
arose.

CHAPTER 2
HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL INSECT
PESTS
Non-Protein Amino Acids
One avenue of pest management explored in the field of biorational pesticides is
the use nonprotein amino acids. Secondary plant materials such as these serve many
functions in insect-plant relationships from attractants and repellents to crude insecticides
(Dahlman 1980). Only a few nonprotein amino acids have been examined as a potential
means to control insect pests. L-canavanine and its by-product of detoxification, L-
canaline, have been studied extensively, with a variety of effects ranging from
developmental deformities to aberrant adult behavior (Dahlman and Rosenthal 1975;
1976; Rosenthal et al. 1995). L-canavanine is found mainly in leguminous plants,
including several economic species (Bell 1978; Felton and Dahlman 1984). It is believed
that plants produce this allelochemical for protection against feeding by phytophagous
insects and herbivores (Rosenthal 1977). The mode of action for canavanine can be
traced to several metabolic processes, including disruption of DNA/RNA and protein
synthesis, arginine metabolism, uptake, anomalous canavanyl protein formation, and the
reduction of active transport of K+ in the midgut epithelium (Kammer et al. 1978;
Racioppi and Dahlman 1980; Rosenthal 1977; Rosenthal et al. 1977; Rosenthal and
Dahlman 1991). In contrast, canaline possesses neurotoxic characteristics with an
unknown mode of action (Kammer et al. 1978). The species of choice for studies
involving nonprotein amino acids has been the tobacco homworm (THW), Manduca
sexta (L.) (Lepidoptera: Sphingidae).
7

8
L-canavanine exhibits a range of insecticidal effects in artificial diets when
exposed to the THW. Dahlman (1977) demonstrated a reduction in consumption of
artificial diet containing less than 1% canavanine (w/v) which resulted in a lower body
mass and increased developmental time to the adult stage. Fecundity and fertility also
was affected by L-canavanine. Rosenthal and Dahlman (1975) showed that
concentrations as low as 0.5 mM L-canavanine in the diets of the THW resulted in the
reduction of ovarial mass of adults, while Palumbo and Dahlman (1978) showed that
concentrations of L-canavanine in agar-based diets resulted in the reduction of
chorionated oocyte production in concentrations between 1.0 and 2.0 mM.
Under natural conditions, L-canavanine was found to retard development, and
increased the susceptibility of exposed larvae to biotic and abiotic mortality factors
(Dahlman 1980). However, field applications of L-canavanine were shown to be
impractical because of the expense involved in synthesizing L-canavanine from its
source, the jack bean (Canavalia ensiformis (L.) DC. (Family: Fabaceae)).
Other sources of L-canavanine (i.e., analogues and homologues) were sought in
an attempt to find a more practical source of the amino acid. Structural homologues of
canavanine were examined and found to contribute to pupal deformities (and to a lesser
degree, to mortality) (Rosenthal et al. 1998). Long-chain esters of L-canavanine were
found to be more toxic than the parent compound when injected or added to an artificial
diet exposed to last instar of THW specimens (Rosenthal et al. 1998). Adding amino
acids other than arginine (the parent compound to L-canavanine) to diets containing L-
canavanine increased deformities and mortality of THW larvae and was attributed to the
structure and position of the functional groups on the added compounds (Dahlman and
Rosenthal 1982). Although the THW has an effective means of degrading aberrant

9
proteins (produced by the assimilation of L-canavanine) into newly synthesized proteins;
the proteases involved do not efficiently degrade enough to prevent some damage from
occurring in the insect (Rosenthal and Dahlman 1986; 1988).
Surprisingly, L-canavanine also was shown to increase the effectiveness of
Bacillus thuringiensis in vivo by altering membrane properties, mainly gut permeability,
and active transport in the midgut of the THW (Felton and Dahlman 1984). However,
despite the possible synergistic relationship between the relatively safe Bt product and
this amino acid, no further research has been conducted to evaluate the combination for
future commercial use.
Other species of insects have also been tested for susceptibility to canavanine with
a variety of results. Larvae of Drosophilia melanogaster Meigen (Dptera:
Drosophilidae) showed no deleterious response to lower concentrations of canavanine,
but showed mortality increased at concentrations over 1,000 ppm (Harrison and Holiday
1967). Lower concentrations also were ineffective against adult Pseudosarcophaga
affinis (Fallen) (Dptera: Calliphoridae), with no effect on oocyte development (Hegdekar
1970). Dahlman et al. (1979) examined four species of muscoid flies and found greater
than 70% mortality at the higher concentration (800 ppm) and decreased pupal weights as
concentrations of canavanine increased.
Despite the toxicity of canavanine to some insects, others have evolved
detoxifying mechanisms to deal with high concentrations of this compound. Rosenthal et
al. (1978) attributed the detoxification of canavanine in the bruchid Caryedes brasiliensis
Thunberg (Coleptera: Bruchidae) to the beetles ability to convert canavanine to
canaline, another toxic amino acid. Canaline is metabolized through reductive
deamination to homoserine and ammonia, with the overall result being the detoxification

10
of the two antimetabolites. This process actually increases the nitrogen intake from the
foodstuff (from the increase of ammonia) (Rosenthal et al. 1976; Rosenthal et al. 1977).
Another insect, the tobacco budworm (Heliothis virescens (Fab.) (Lepidoptera:
Noctuidae)) was able to metabolize far more canavanine then the bruchid beetle larva
ever takes in during its development, suggesting that generalists may have more than a
single detoxification mechanism for compounds they may encounter (Berge et al. 1986).
Metabolism of L-canavanine by the tobacco budworm was attributed to the gut enzyme
canavanine hydrolase, and may have been the result of feeding on canavanine-containing
plants of the Fabaceae (Melangeli et al. 1997).
Essential Amino Acids
In despite of the extensive toxicological research conducted on nonprotein amino
acids, another group of amino acids, the essential ones, has been overlooked. One reason
this avenue for research has not been pursued is that we do not want to give pests
convenient access to an integral part of their diet. The fear of creating a super insect
(that has been provided with compounds that actually aid in its development) is a rational
one. Mittler (1967a; 1967b) found an increase in gustation in Myzus persicae (Sulzer)
(Hemiptera: Aphididae), with amino acid levels as low as 0.2% concentration in a
sucrose solution. Likewise, Sugarman and Jakinovich (1986) found increased gustatory
response to both D-and L-methionine by Periplaneta americana (L) (Blattodea:
Blattidae) adults. Another reason that essential amino acids have not been examined for
use as a pesticide is the knowledge regarding the limited mode of action these compounds
could be involved with (i.e., an active site or systemic response). Recent studies on the
membrane proteins of insects show the possibility of a biophysiological system that can
be exploited for insect control with certain essential amino acids.

11
The Cation-Anion Modulated Amino Acid Transporter With Channel Properties
(CAATCHl) System
Cation-Anion modulated Aminoacid Transporter with Channel properties
(CAATCHl) is a recently cloned insect-membrane protein isolated from larval
midgut/hindgut nutritive absorptive epithelium. This membrane protein exhibits a unique
polypeptide and nucleotide sequence related to, but different from, mammalian Na+-, Cf-
coupled neurotransmitter transporters (Feldman et al. 2000). Using a unique PCR-based
strategy, the gene encoding CAATCHl was cloned from the digestive midgut of THW
larvae. The unique biochemical, physiological, and molecular properties of CAATCHl
indicate that it is a multifunction protein that mediates thermodynamically uncoupled
amino acid uptake, functions as an amino acid-modulated gated alkali cation channel, and
is likely a key protein in electrolyte and organic-solute homeostasis of pest insects (Quick
and Stevens 2001). In the presence of no amino acids, the cations K+ and Na+ are
transported through the membrane via the channel (Figure 2A). When exposed to
proline, the amino acid is transported through the membrane with an increase in cation
flow, especially Na+ (Figure 2B). However, when exposed to methionine, the amino acid
transport is stopped and cation flow is altered, mainly the increased flow of K+ and the
decreased flow of Na+ (Figure 2C). The CAATCHl system works in alkaline conditions,
at a pH optimum ~ 9.5. This alkaline condition is found in the midgut of several species
(Nation 2001) and has been attributed to a variety of causes, from the detoxification of
plant allelochemicals to amino acid uptake (Giordana et al., 2002; Leonardi et al. 2001).

12
No Amino Acid
Proline
Methionine
Figure 2. The CAATCH1 system identified from the midgut of the tobacco
homworm (modified from Quick and Stevens 2001). In the
presence of no amino acids, ion flow is similar for both K+ and
Na+ (A). With the addition of an amino acid, flows are changed.
When proline is added (B), the transport occurs but the binding of
the amino acid increases the ion flow, notably Na+. However,
when methionine is added (C) transport occurs and the binding of
the amino acid greatly decreases the flow of Na+ while K+ is
increased

13
Several amino acids were found to initiate the blocking action of ion flow through the
CAATCH1 protein, including threonine, leucine, and methionine, with the latter
producing the greater response, based on CAATCH1 research (Feldman et al. 2000;
Stevens et al. 2002; Quick and Stevens 2001).
Methionine
The amino acid methionine is considered essential in the diets of many organisms.
Methionine is considered an indispensable amino acid in humans. Because the body does
not synthesize it, uptake of methionine must occur in the diet. The recommended daily
allowance of methionine for a healthy lifestyle ranges from 13 to 27 mg/kg/day for
infants to full-grown adults (Young and El-Khoury 1996). This amino acid is linked to a
decrease in histamine levels, increased brain function, and is found in a variety of
sources; with the highest concentration in various seeds, greens, beef, eggs, chicken, and
fish (Dietary Supplement Information Bureau 2000). Recently, research has centered on
the genetic modification of crop plants to overproduce methionine to increase its
nutritional quality (Zeh et al. 2001). Wadsworth (1995) discussed using methionine as a
feed supplement, as an aid in the therapy of ketosis in livestock, and as a treatment for
urinary infections in domestic pets. Onifade et al. (2001) examined the use of housefly
larvae as protein foodstuffs, and found an increase in body weight gain and erythrocyte
counts in rats whose diets were supplemented with fly larvae and methionine. Likewise,
Koo et al. (1980) suggested dry face fly pupae could be used as a dietary supplement and
foodstuff extender for poultry because of the high concentration of methionine. The
environmental safety of methionine is well known, as it poses no risk to vertebrates due
to a rather high oral LD50 of 36g/kg~' observed in rats (Mallinckrodt Baker 2001) and also

14
in its use as a feed supplement for livestock under the trade name of Alimet (Novus,
Inc., St. Louis, MO).
In addition to vertebrates, methionine also is considered an essential amino acid
for insects (Nation 2001). Based on research on nutritional requirements for insects, the
amount of methionine needed in a diet for survival ranged from as little as 0.0007 mg/mL
(for Aedes aegypti (L.) (Diptera: Culicidae) to as high as 100 mg/mL (for Heliothis zea
(Broddie) (Lepidoptera: Noctuidae)) (Dadd and Krieger 1968; Eymann and Friend 1985;
Friend et al. 1957; Kaldy and Harper 1979; Kasting et al. 1962; Koyama 1985; Koyama
and Mitsuhashi 1975; Rock and Hodgson 1971; Singh and Brown 1957). Methionine
occurs naturally as the L-isomer while the D-isomer (an optical enantiomer) is toxic to
many insects and is not found in nature (Anand and Anand 1990). A few exceptions are
known, (mainly Diptera and Lepidoptera) that actually are capable of metabolizing the
normally unusable D-isomer (Dimond et al. 1958; Geer 1966; Rock 1971; Rock et al.
1973; Rock et al. 1975). The requirement for small amounts of this amino acid (as
compared to other amino acids) may be a result of the ability for some insects to
synthesize methionine from cysteine (a common sulfur containing amino acid) thus
reducing the need to take in exogenous sources of methionine. Jaffe and Chrin (1979)
found that A. aegypti adults were able to synthesize methionine from homocysteine with
the aid of a methionine synthetase. They found this enzyme similar to those common in
other metazoans, and found that the levels of methionine synthetase increased with the
presence of filarial parasites. They hypothesized that this increase in methionine
synthetase was a result of the parasite depleting the host of methionine.
Fertility and fecundity also have been associated with methionine in some insects
(mainly D. melanogaster,) with the possibility if it being a limiting factor during egg

15
production (Sang and King 1961). Lack of methionine in the diet of the female may also
explain the transfer of methionine in the ejaculate of the male during fertilization
(Bownes and Partridge 1987). Methionine plays another role in insect biochemistry,
especially in juvenile hormone biosynthesis, inhibitory allatostatins, and storage proteins
known as hexamerins. Audsley et al. (1999) found that in vitro rates of juvenile hormone
synthesis in females of the tomato moth (Mamestra olercea (L.) (Lepidoptera:
Noctuidae)) were dependent on the concentration of methionine present in the incubation
medium. Tobe and Clarke (1985) found a direct relationship between methionine
concentration and juvenile hormone biosynthesis in the cockroach, Diploptera punctata
(Eschscholtz) (Blattodea: Blaberidae), further supporting the idea that methionine plays
an important role in insect biochemistry.
Storage proteins, or hexamerins, act as a storehouse for amino acids that can be
sequestered for later use in the developmental cycle (Pan and Telfer 1996). Many
Lepidoptera have been identified with hexamerins containing high concentrations of
methionine and are metabolized during the last larval stage, and presumably used for egg
production (Wheeler et al. 2000).
Methionine as a potential pesticide has not been overlooked entirely. Tzeng
(1988) tested a methionine and riboflavin mixture and found it successful in controlling
various pests, including the larvae of Culex spp. (Dptera: Culicidae). The mode of
action for this mixture was attributed to a photodynamic reaction and the production of
oxygen rich radicals (Tzeng et al. 1990). Their research led to the use of this methionine
compound as a control agent for sooty mold of strawberry (Tzeng and Devay 1989;
Tzeng et al. 1990) but not as an insecticide.

16
Discovery of novel means for controlling various insect pests is one tenant of
IPM. The amino acid methionine, an environmentally safe organic compound, appears to
be a candidate for further study. Before it can be considered for use in controlling insects
pests, several issues must be addressed, including the determination of concentrations
needed to provide effective control, compatibility with current application systems, safety
to nontarget organisms (i.e., beneficial or biological-control agents), and to phytotoxicity.
Research Objectives
Our overall goal was to evaluate the effects of L-methionine, and its amino acid
analogues, on the CAATCH1 system putatively in the midgut/hindgut as a means to
control different insect pests. The working hypothesis is that the L-methionine only
affects the CAATCH1 system and no other system, especially those involving Na+
channels or pumps (i.e., nervous tissue). The L-isomer of methionine was chosen
because of the inability of most insect species to utilize the D-isomer. Ideal targets for
this research are those pests that cause severe damage to agricultural systems and to
human health. Specific objectives were to
Examine the effects of L-methionine as an insecticide on the larvae of M. sexta
(Tobacco homworm), L. decemlineata (Colorado potato beetle) and A. aegypti
(Yellow-fever mosquito) under various conditions
Determine any adverse effects of L-methionine on plant health to ensure its safe
use in a cropping system
Examine the effects of L-methionine on various nontarget insect species to ensure
the environmental safety of L-methionine and thus its compatibility with natural
enemies in the context of IPM.

CHAPTER 3
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
TOBACCO HORNWORM, Manduca sexta, UNDER LABORATORY CONDITIONS
Introduction
Manduca sexta (L.) (Lepidoptera: Sphingidae), the tobacco homworm (THW), is
a widespread species considered an economic pest throughout North and South America.
The caterpillar is known for its voracious appetite. In Georgia, the THW was responsible
for between approximately $1.2 to $1.5 million in losses and costs for control annually in
tobacco from 1997 to 2001 (Jones and McPherson 1997; McPherson and Jones 2002). In
addition to its well-earned reputation as an agricultural pest of solanaceous crops, the
THW has shown to be resistant to common pesticides (such as endrin and endosulfan),
with the possibility of cross-resistance (Bills et al. 2004).
The THW also is very important to scientific research outside the arena of
economic entomology, with studies ranging from molecular-based research to ecological
and physiological research, mainly because of its availability and ease in culturing
(Dwyer 1999). One research area of interest to scientists involves the chemistry and
physiology of the midgut. Insect control (or the development of new insecticides) was
probably not the main purpose of the research that resulted in identifying the CAATCH1
protein, yet it became the basis of our research project.
Because little information is available on the insecticidal properties of
methionine, several baseline experiments were necessary to determine that concentrations
of this amino acid to test It also was necessary to test I .-methionine and THW
17

18
interaction in a variety of ways, including artificial diet, natural diet (excised leaves,
whole plant, and choice tests. The purpose of this portion of this study was to determine
whether L-methionine was detrimental to the survival and development of the THW and
to determine if L-methionine could be used to control this species.
Materials and Methods
Eggs of THW were obtained from the insectary of North Carolina State
University, and were held in 26.4L x 19.2W x 9.5H (cm) clear plastic rearing chambers
with a hardware cloth (to facilitate cleaning) (Figure 3-1). Florida Reach-In Units
(FRIUs) were used to control the environment for the rearing containers (Walker et al.
1993) Containers were held at 27 C, 60% relative humidity, and a 16L:8D photoperiod
in FRIUs with either artificial or natural diet (excised eggplant leaves or whole plants)
depending on the pending experiment. Neonates were allowed to feed for 2 days after
eclosin before being transferred to treatment groups. A camel hair brush was used for
transferring larvae, to minimize the risk of damage.
Diets and Survivorship
The artificial diet was prepared using the procedures outlined in Baumhover et al.
(1977) with the inclusion of L-methionine for the treatment concentrations of 0.1%,
0.3%, 0.5%, 1.0%, 3.0%, 5.0% and 10.0% (wt/wt). The artificial diet was changed on a
regular basis to prevent desiccation and fungal growth. Larvae were exposed to the
artificial diet in the clear plastic rearing chambers with a hardware cloth, and kept in the
FRIUs programmed with the aforementioned environmental constants.
Natural diets consisted of excised eggplant leaves (Solarium melongena
L.,Classic variety) of potted plants grown and maintained at the University of Florida,

19
Figure 3-1. Rearing chamber for tobacco homworm and Colorado potato
beetle larvae used in the artificial and excised leaf diet tests.
Hardware cloth stage supporting the leaf allowed for easy
clean up and minimized disease problems by preventing
larvae from coming in contact with fecal material (paper liner
not shown).

20
Department of Entomology and Nematology green and shade houses. Excised leaves
were dipped in solutions of deionized H2O containing different concentrations of
methionine; depending on the experiment and exposed to larvae in the same rearing
chambers as the artificial diet trials under the same conditions. Survivorship data were
pooled from several different trials for data analysis.
In total, 64 potted eggplants were used for the whole-plant portion of the study.
Plants were held in FRIUs under the same conditions as the artificial and excised leaf
trials, in 38H x 15D (cm) plexigls cylinders (Figure 3-2). Four THW neonates were
placed on each plant for a total of 64 larvae (16 replicates) per treatment (nx0tai=256
larvae). The treatment of L-methionine was applied to the test plants (using a hand-held
sprayer calibrated to deliver approximately 10 mL of solution to each plant) before the
addition of larvae.
Feeding and Development
To test L-methionine on the developmental rates of THW, larvae were exposed to
excised eggplant leaves dipped in solutions containing the same concentrations of L-
methionine used in the artificial diet trials. Additional treatments of proline (1.0%) and
Bt-kurstaki (Dipel 86% WP at 3.5 grams/liter; Bonide, Oriskany, NY) were included as
positive and negative controls, respectively. Leaves were scanned photometrically using
the Cl 203 Area Meter with conveyor attachment (CID, Inc.; Camas, WA) to measure
leaf consumption before and after exposure to larvae. The difference in leaf areas
resulting from the missing leaf tissue was assumed to be the amount eaten by the
developing larvae. Larval head capsule widths were measured at the time of death or the

21
Figure 3-2. Setup for whole plant studies involving tobacco homworm. Top and
portions of the sides were replaced with fine mesh to allow for
airflow and to reduce condensation.

22
end of the trial (using an Olympus Tokyo Model 213598 stereomicroscope with a optical
micrometer) to monitor larval development
Trials to determine the total amount of L-methionine applied to excised leaves
also were included to quantify how much of the amino acid was physically present on
leaves at the different concentration levels. Leaves were weighed before dipping into the
control (0%) and L-methionine solutions (0.1%-10%), allowed to air dry for 30 min and
weighed again. The difference was assumed to be the actual amount of L-methionine
residue on the leaf. This value then was used to determine the total amount of
L-methionine on the leaf surface of the excised leaves and the amount of L-methionine
consumed per gram of leaf material, based on calculations of the physical amount of the
compound for each % concentration.
Preference Tests
It was unknown if the additional methionine acted to attract or repel larvae.
Neonate larvae were used in the choice tests to determine if there was a preference
between the control (deionized H2O) and the Treatments (1.0% L-methionine). Leaves
were obtained from potted plants maintained in the outdoor shade house. The tests
consisted of 4 leaf disks (30 mm diameter) dipped into the control solution and placed
into the chamber alternately with four leaf disks (30 mm diameter) dipped into the
treatment solution and replicated with a total of 10 chambers. Each chamber consisted of
a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The
filter disk was moistened routinely with deionized H2O to prevent the leaf disks from
desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental
constants described previously. The leaf disks also were scanned photometrically and

23
Figure 3-3. Chambers used for tobacco homworm and Colorado potato beetle
preference tests. Two treatments (control and 1.0% L-
methionine) were used to determine if any larvae exhibited any
preference or avoidance to L-methionine. Treatments were
alternated in the chamber and neonates were released in the center
of the dish and allowed to search for food. The filter paper in the
bottom of the dish was moistened to prevent desiccation of the
leaf disks and the test specimens.

24
larval head capsule measurements made using the same procedures described in the
Feeding and Development section.
Data Analysis
Sample sizes of all experiments were chosen according to the guidelines
recommended by Robertson and Preisler (1991) for optimal sample size and data
analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were
performed using PROBIT Version 1.5 (Ecological Monitoring Research Division,
USEPA) after Abbotts correction for control mortality (Abbott 1925). Survival data
were normalized to the previous value when control mortality was greater than the
treatment mortality, to produce a smoother trend line. Statistical analysis was performed
on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the
data included One-way ANOVA and separation of significant means using Tukeys
Multiple Comparison and Pearson Correlation was performed on the choice trial data to
examine possible relationships between development and consumption of treated leaf
material (Zar 1999). Regression analysis using lest squares were performed on the leaf
weights before and after the L-methionine treatment for the equation used to convert %
concentration to mg/g plant material (Figure 3-4).
Results
Diets and Survivorship
The artificial diet resulted in 100% mortality of THW larvae for the 3.0%
L-methionine to 10.0% L-methionine treatment after only one day of exposure (Figure
3-5). Approximately 80% mortality was observed in the 1.0% L-methionine treatment
after 4 days, and 50% mortality for both the 0.3% L-methionine and 0.5% L-methionine

g of L-methionine/g of Leaf Material
L-methionine Concentration (%)
Y = 8.65E-04 + 4.76E-03X
R-Sq = 98.6 %
95% Confidence Intenal
+/- 2SE (SE=.001997)
Figure 3-4. Amount of L-methionine present on leaf surface after treatment.
Excised leaves were weighed, dipped into various concentrations
of L-methionine, allowed to dry, and then re-weighed. Difference
assumed to be the amount of L-methionine remaining on leaf
surface (T=22.43, and P<0.001).

26
Figure 3-5. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nxotai-480) in artificial diet. Data were adjusted using
Abbotts formula to account for control mortality. Note the overlap in trend
lines for the 3.0% L-methionine-10.0% L-methionine concentrations after
Day 1 and the 0.3% L-methionine and 0.5% L-methionine treatments from
Day 1 to Day 10.

27
treatment after 10 days of exposure. The 0.1% L-methionine concentration had lowest
larval mortality with approximately 30% observed for the trial.
The excised leaf trials exhibited higher mortality rates associated with the
treatments than did the artificial diet trials. Again, complete mortality was observed with
the 3.0% L-methionine thru 10.0% L-methionine concentrations after 1 day of exposure
(Figure 3-6). Greater than 90% in the 0.5% L-methionine and 1.0% L-methionine
treatments, followed by 80% mortality in the 0.3% L-methionine treatment, and greater
than 60% mortality occurred in the 0.1% L-methionine treatment after 8 days.
Whole plant trials produced results similar to the excised leaf trials with greater
than 90% larval mortality observed with the 1.0% L-methionine treated plants after 14
days (Figure 3-7). Mortalities exceeding 20% and 60% were observed for the 0.1%
L-methionine and 0.5% L-methionine treatments, respectively, during the same time
interval.
PROBIT analysis of a sample size of n-rota 1,540 for 7 treatments (0.1%
L-methionine, 0.3% L-methionine, 0.5% L-methionine, 1.0% L-methionine, 3.0%
L-methionine, 5.0% L-methionine and 10.0% L-methionine) revealed an overall LC50 of
0.66% (32.3 mg/g leaf material) concentration for the artificial diet and 0.074% (4.39
mg/g leaf material) concentration for the natural diet after 9 days of exposure (Figure
3-8). The LC50 for the THW exposed to artificial diet was approximately half the value
of that for the natural diet for the 24 to 72 hour exposure period. The LC50 for the
artificial diet of 1.08% (52.3 mg/g leaf material) for 24 h dropped to 1.0% (48.5 mg/g leaf
material) after 48 h and to 0.57% (28.0 mg/g leaf material) after 72 h. As for the natural
diet, the LC50 of 0.53% (26.1 mg/g leaf material) was found to be lower than the artificial

28
Figure 3-6. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nTotai= 1,540) on excised eggplant leaves. Data were
adjusted using Abbotts formula for control mortality. Note the overlap in
trend lines for the 3.0% L-methionine-10.0% L-methionine concentrations
after Day 1.

29
Figure 3-7. Survivorship of tobacco homworm larvae exposed to various
concentrations of L-methionine (niotai=256) on whole plants. L-
methionine was applied using a hand-held sprayer in the amount of
10 mL/treatment. Data were adjusted using Abbotts formula for
control mortality.

LC50 (% L-methionine Concentration)
30
24h 48h 72h Overall (216h)
Figure 3-8. Concentrations (%) of L-methionine required for the mortality of 50%
of test population of tobacco homworm after 9 days exposure
(nTota)~L540; n=180 for 3.0% L-methionine 10.0% L-methionine,
n=200 for remainder). Number range following value is the 95%
confidence limits. Determination of LC50 was performed using
PROBIT Version 1.5 (Ecological Monitoring Research Division,

31
diet at 24 h and dropped to 0.4% (19.9 mg/g leaf material) at 48 h and 0.25% (12.8 mg/g
leaf material) after 72 h exposure. Overall, the LC50 at the end of the experiment for the
natural diet was well below the value for the artificial diet, with close to a 90% reduction.
Feeding and Development
Mortality of THW for the developmental tests ranged from approximately 30%
for the 0.1% L-methionine treatment and over 40% for the proline treatment (Figure 3-9).
Complete mortality for the 0.3% L-methionine occurred after 7 days while the 0.5%
L-methionine treatment took only 5 days. The Btk treatment mortality was similar to the
0.7% L-methionine and 1.0-%L-methionine treatment, resulting in 100% mortality after 1
day of exposure to the amino acid. Both the mean head capsule width and amount of leaf
material consumed showed significant differences between treatments, with the control,
0.1% L-methionine and proline treatments being different that the remaining treatments
(Figures 3-10 and 3-11).
Preference Tests
The amount of control and 1.0% L-methionine leaf tissue consumed during the
preference tests were found not to be statistically different (Figure 3-12). In addition to
the amount of leaf material consumed between treatments not being different, the mean
head capsule width (i.e., development) showed a correlation with the amount of control
diet consumed (Pearson Correlation Coefficient 0.885, P<0.001) while no correlation to
the Treatment diet consumed (Pearson Correlation Coefficient 0.630, P=0.051) (Figure
3-11).

32
Figure 3-9. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (njota 160) on excised eggplant leaves for feeding and
development trials. Proline (1.0%) and Btk were included for comparison
as positive and negative controls. Data were adjusted using Abbotts
formula for control mortality. Note the overlap in the 0.7% L-methionine,
1.0% L-methionine and Btk treatments at Day 1.

Mean Head Capsule Width (mm)
33
(Errer Bars @ 95%; ^(o.os)7,i 52=2.37, F=18.2; P <0.001)
6
5
4
3
2
1
0
Control 0.1% 0.3% 0.5% 0.7% 1.0% Proline Btk
Figure 3-10. Mean head capsule widths of tobacco homworm larvae exposed
to excised eggplant leaves treated with various concentrations of
L-methionine (nrota]=320). Proline (1.0%) and Btk were included
for comparison as positive and negative controls. Error bars
denote 2 SE. Bars within treatments having the same letter are
not statistically different (Tukeys MST, P0.001).

34
(Error Bars @ 95%; F(o.os)7,i52=2 .37, F=\8.2; P<0.001)
Figure 3-11. Total leaf area consumed by tobacco homworm larvae exposed to
excised eggplant leaves treated with various concentrations of L-
methionine (nrotai=320). Proline (1.0%) and Btk were included for
comparison as positive and negative controls. Error bars denote 2 SE.
Bars within treatments having the same letter are not statistically
different (Tukeys MSTP, P<0.001).

35
(Bror Bars @ 95%; /r(0.05)i, 18=5.98, F-1.64; P =0.217)
Control 1.0% Treatment
Figure 3-12. Mean leaf consumption by tobacco homworm in the preference
tests. Error bars denote 95% SE, and treatments were found not to
be statistically different. However, there was correlation between
the control diet consumed and mean head capsule width (Pearson
Correlation Coefficient 0.885, P=0.001) while no correlation was
found between the Treatment diet consumed and mean head
capsule width (Pearson Correlation Coefficient 0.630, P=0.05).

36
Discussion
The initial studies involving the high concentrations of L-methionine (i.e., 3.0-10.0%,
which are outside the range normally encountered in nature) showed that a concentration
of 1.0% L-methionine was sufficient enough to provide good control of THW larvae
reared on both artificial and natural diets. The 0.1%L-methionine concentration
remained similar to that of the control for developmental and feeding trials (Figure 3-9),
indicating a level of methionine that can be tolerated to some extent, as seen in the low
mortality of this treatment. This is in stark contrast to the mortality seen in the excised
leaf trials in which the same concentration had over 60% mortality (Figure 3-6). One
possible explanation could be the amount of L-methionine present on the leaf disk being
low enough and ingested at a slower rate than that of the whole leaf, which was left in the
chamber with the larvae until the leaf was either completely consumed or too wilted for
the larvae to ingest.
The preference tests did show some preference towards control leaf disks over the
1.0%L-methionine treated disks as seen in the correlation analysis of the diet consumed
and the mean head capsule width of the larvae. Despite the lack of a statistical difference
between the amount consumed, the larvae could have fed on the treated disks and then
switched to the control disks based on a physiological cue. It is unclear if THW larvae
possess specialized sensory structures to detect amino acids like those found in other
Lepidoptera (Beck and Henee 1958; Dethier and Kuch 1971; Schoonhoven 1972), but the
possible switch from the methionine rich treatment to the control leaf disks does indicate
some sort of mechanism for detection. Del Campo and Renwick (2000) found THW
larvae were induced to feeding on plants outside of their normal diet when the plants

37
were treated with an extract from potato foliage suggesting induced host preference,
attraction, and dependence on this compound in the extent of sustained feeding and
development. A combination of sensory structures may be involved for the detection of
specific amino acids and host plant compounds, which may explain the selection of
methionine depleted host plants to avoid problems with the CAATCH1 system present in
the midgut of the THW.
The difference in the LC50 for the artificial and natural diets was striking
considering the concentrations were the same. One possible explanation is the
L-methionine on the natural diet was more readily available than that found in the
artificial diet. With the artificial diet, the L-methionine is presumably spread throughout
the diet and would therefore take longer for the THW to ingest enough to adversely affect
the CAATCH1 system. In contrast, the L-methionine was found on the surface of the
leaf in higher concentrations than that of the artificial diet and was also freely available
once ingested. Thus, larvae were exposed to a higher concentration of L-methionine with
less work to digest, resulting in lower survivorship in the same period of time.
The 1.0%L-methionine concentration had the same mortality, feeding and
developmental rates for THW, as did the Btk treatments (Figure 3-9). The 0.3%
L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments were virtually the
same for mortality (Figure 3-9), developmental rate (Figure 3-10) and total leaf material
consumed (Figure 3-11) and statistically the same as the 1.0% L-methionine
concentration and the Btk treatment. The similar mortality rate observed for the higher
concentrations of L-methionine and Btk is encouraging considering the resistance to Bt
seen in many insect species because of reduced receptor activity and binding (Bills et al.

38
2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and
many are the result of a combination of different pesticide classes. The CAATCH1
system is one that could be used in cases where the only alternative is by adding more
pesticides or at higher rates to break resistance. Further research is needed to determine
compatibility of the different Bt insecticides and L-methionine with each other for cases
in which Bt resistance is observed in natural populations. Given the safety of
L-methionine and the shorter time required for 100% mortality (when compared to Btk
results of this study), this compound could represent a viable alternative for pesticides
currently used in the management of the THW.

CHAPTER 4
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER LABORATORY
CONDITIONS
Introduction
Leptinotarsa decemlineta (Say) (Coleptera: Chrysomelidae), the Colorado
potato beetle (CPB), is considered an economic pest throughout North America. The
larvae and adults of the CPB feed on a wide variety of solanaceous crop plants and are
responsible for $150 million in losses and control related costs (Durham 2000). To
further complicate matters, the CPB is resistant to numerous pesticides, including various
pyrethroids and carbamates (Bills et al. 2004). Historically, CPB management relied
heavily on chemical control methods that led to the development of resistance to different
pesticides in several areas of the eastern United States (Forgash 1985; Gauthier et al.
1981). Control of CPB without the use of chemicals is further complicated given the
species ability to develop resistance and the limitations on the use of resistant varieties of
potato (Ragsdale and Radcliffe 1999). The use of plant varieties that are resistant to CPB
and other pests also run the risk of developing tolerance to chemical pesticides in other
pest species (Sorenson et al. 1989). Despite the success of Bacillus thuringiensis-
tenebrionis (Btt) and the biocontrol agents Podisus maculiventris Say (Hemiptera:
Pentatomidae) and Edovum puttleri Grissel (Hymenoptera: Eulophidae), more biorational
alternatives are necessary for controlling CPB to prevent yet another devastating threat to
the potato industry because of this insects ability develop resistance and overcome
control methods (Boucher 1999; Ferro 1985; Tipping et al. 1999). This makes the CPB
39

40
an excellent candidate for the evaluation of L-methionine as a possible means of
controlling this devastating pest.
Because little information is available on the insecticidal properties of
L-methionine, several baseline experiments were necessary to determine what
concentrations of this amino acid to test. Therefore, it was necessary to test L-methionine
and CPB interaction in a variety of ways including survivorship of both larvae and adults,
development of larvae when exposed to different concentrations of the amino acid, and
preference tests. The purpose of this portion of this study was to conduct bioassays to
determine if exposure to L-methionine was detrimental to the survival and development
of the CPB and to determine if L-methionine could be used to control this species.
Materials and Methods
Eggs of CPB were obtained under UDSA permit from the insectary of the New
Jersey Department of Agriculture and held in 26.4L x 19.2W x 9.5H (cm) clear plastic
boxes with a hardware cloth (to facilitate cleaning) and held at 27 C, 60% relative
humidity and 16L/8D photoperiod in FRIUs (Figure 3-1). Excised eggplant leafs were
placed in the chambers with the neonates and they were allowed to feed for 2 days after
eclosin before being transferred to experiments. A camel hair brush was used for
transferring the neonates to minimize the risk of damaging the larvae.
Survivorship
Larvae and adults of the CPB were tested in preliminary experiments with the
highest concentration (1.0% L-methionine (wt/wt)) observed in tests done on the THW in
the previous chapter. The diet for the larvae and adults consisted of excised eggplant
leaves (,Solarium melongena L.,Classic variety (Family: Solanaceae)) from plants
grown and maintained at the University of Florida, Department of Entomology and

41
Nematology green and shade houses. Excised leaves were dipped in solutions of
deionized H2O containing different concentrations of methionine and held in the clear
plastic boxes and held at the aforementioned environmental conditions (Figure 3-1).
Additional treatments of proline (1.0%) and Bt-tenebrionis (Novodor FC @12.4 mL/L;
Valent Biosciences, Libertyville, IL) were included as positive and negative controls,
respectively. Survivorship data were pooled from several different trials for data
analysis.
Feeding and Development
To test L-methionine on the developmental rates of CPB, larvae were exposed to
excised eggplant leaves dipped in different concentrations of L-methionine under the
same conditions as the survivorship trials. Additional treatments of proline (1.0%) and
Btt were included as positive and negative controls, respectively. Leaves were scanned
photometrically using the Cl 203 Area Meter with conveyor attachment (CID, Inc.,
Camas, WA) before exposure to the larvae and measuring after leaf consumption. The
difference in leaf areas resulting from the missing leaf tissue was assumed to be the
amount eaten by the developing larvae. Larval head capsule widths were measured at the
time of death or the end of the trial (using an Olympus Tokyo Model 213598
stereomicroscope with an ocular micrometer) as an evaluation of larval development.
Preference Tests
It was unknown if the additional methionine acted to attract or repel larvae.
Neonate larvae were used in the choice tests to determine if there was a preference
between the Control (deionized H20) and the treatments (1.0% L-methionine). Leaves
were obtained from potted plants maintained in the outdoor shade house. The tests

42
consisted of 4 leaf disks (30 mm diameter) dipped into the Control solution and placed
into the chamber alternately with four leaf disks (30 mm diameter) dipped into the
treatment solution and replicated with a total of 10 chambers. Each chamber consisted of
a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The
filter disk was moistened routinely with deionized H2O to prevent the leaf disks from
desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental
constants described previously. The leaf disks also were scanned photometrically and
larval head capsule measurements made using the same procedures described in the
Feeding and Development section.
Data Analysis
Sample sizes of all experiments were chosen according to the guidelines
recommended by Robertson and Preisler (1991) for optimal sample size and data
analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were
performed using PROBIT Version 1.5 (Ecological Monitoring Research Division,
USEPA) after Abbotts correction for control mortality (Abbott 1925). Survival data
were normalized to the previous value when control mortality was greater than the
treatment mortality, to produce a smoother trend line. Statistical analysis was performed
on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the
data included One-way ANOVA and separation of significant means using Tukeys
Multiple Comparison and Pearson Correlation was performed on the choice trial data to
examine possible relationships between development and consumption of treated leaf
material (Zar 1999).

43
Results
Survivorship
Mortality of CPB larvae on treated excised eggplant leaves ranged from
approximately 20% for the 0.1% L-methionine treatment after 4 days, 80% mortality for
the 0.3% L-methionine treatment after 8 days of exposure and 100% for the remaining
concentrations with the highest dose of 1.0% L-methionine exhibiting complete control
of CPB in 3 days post treatment (Figure 4-1). Some mortality (50%) was observed for
the proline (1.0%) treatment while the Bit larval treatment mortality was similar to the
1.0% L-methionine treatment, resulting in 100% mortality after 5 days.
PROBIT analysis of a sample size of ntotai=L320 for 6 treatments (Control), 0.1%
L-methionine, 0.3% L-methionine, 0.5% L-methionine, 0.7% L-methionine and 1.0%
L-methionine) revealed an overall LC50 of 0.218% concentration for the CPB after 8 days
of exposure (Figure 4-2). The LC50 of 2.9% for 24 hours dropped to 1.1% after 48 hours
and to 0.22% after 72 hours.
Feeding and Development
Mean head capsule widths between treatments were found to be statistically
different (Figure 4-3). Four distinct groups were observed, with the Control, 0.1%
L-methionine and proline treatments forming the first group. The second group of
proline and 0.5% L-methionine were statistically the same and likewise the third group of
the 0.3% L-methionine, 0.5% L-methionine, and 0.7% L-methionine treatments. The
final group of Bit and 1.0% L-methionine treatments was statistically different from all
other treatments.

44
O 1
3 4 5
Days of Exposure
Figure 4-1. Mortality of Colorado potato beetle larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nTOtai=560).
Proline (1.0%) and Bit were included for comparison as positive and
negative controls. Data were adjusted using Abbotts formula for
control mortality.

LCS0 (% L-methionine Concentration)
45
Figure 4-2. Concentrations (%) of L-methionine concentrations required for
the mortality of 50% of the test population of Colorado potato
beetle after 8 days exposure (nrotai=220). Number range
following value is the 95% confidence limits. Determination of
LC50 was performed using PROBIT Version 1.5 (Ecological
Monitoring Research Division, USEPA), including Abbotts

Head Capsule Width (mm
46
(Error Bars @ 95%; F(o.o5)7,3i2=1.14;F=576.71; P<0.001)
Figure 4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to
excised eggplant leaves treated with various concentrations of L-
methionine (nTOtai=320). Proline (1.0%) and Bt were included for
comparison as positive and negative controls. Error bars denote 2 SE.
Bars within treatments having the same letter are not statistically
different (Tukeys MST, PO.OOl).

47
Feeding rates of CPB also were found to be statistically different among treatments
(Figure 4-4). Three distinct groups were observed with the first group containing the
Control and 0.1% L-methionine treatments while the second group of the 0.1% L-
methionine and 0.3% L-methionine, treatments were found to be statistically the same.
The 0.5% L-methionine, 0.7% L-methionine, 1.0% L-methionine and Btt treatments were
statistically different from the other groups. Overlap occurred with the proline treatment
across all groups indicating no statistical difference with the rest of the treatments.
Preference Tests
The amount of Control and 1.0% L-methionine leaf tissue consumed during the
preference tests was found not to be statistically different (Figure 4-5). In addition, the
mean head capsule width (i,e development) showed no relationship with either treatment
based upon the low correlation coefficients.
Discussion
The 1.0% L-methionine concentration produced the same larval mortality, feeding
and developmental rates for CPB, as did the Btt treatments (Figures 4-1,4-3, and 4-4).
The 0.3% L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments took 4
days longer for complete control (Figure 4-1), but were statistically different for the
developmental rates for the same treatments (Figure 4-3). As was the case with the THW
survivorship, the 0.1% L-methionine concentration was not different from that of the
Control. This may indicate a threshold of methionine that can be tolerated by the THW,
and CPB to some extent, evidenced by the low mortality observed for this treatment.
The Preference tests did not indicate any preference of leaf disks with or without
L-methionine. The high mortality (90%) of the CPB larvae could be explained by a

48
(Error Bars @ 95%; F(0.05)7,312=1.14;F=40.1; P<0.001)
400
Control 0.1% 0.3% 0.5% 0.7% 1.0% Proline Btt
Figure 4-4. Total leaf area consumed by Colorado potato beetle larvae exposed
to excised eggplant leaves treated with various concentrations of L-
methionine (nTOtai=:320). Proline (1.0%) and Btt were included for
comparison as positive and negative controls. Error bars denote 2
SE. Bars within treatments having the same letter are not statistically
different (Tukeys MST, P0.001).

Mean Amount of Leaf Material Consumed (ci)i
49
(Error Bars @ 95%; F(0.05)i,i8=5.98, F=1.64; P =0.217)
Control 1.0% L-methionine
Figure 4-5. Mean leaf consumption by Colorado potato beetle in the preference
tests. Error bars denote 95% SE, and treatments were found not to
be statistically different. No correlation between either Control or
Treatment Diet consumed and mean head capsule width was found
(Pearson Correlation Coefficient 0.466, P=0.175 and 0.665,
P=0.036, respectively).

50
combination of the early consumption of the treated disks and mortality occurring after
48 hours, when a lower concentration is required for mortality. The larvae could have
fed on the treated disks and then switched to the Control based on a physiological cue.
Mitchell (1974) and Mitchell and Schoonhoven (1974) examined the taste receptors of
CPB and found physiological and behavioral responses to some amino acids, mainly
gamma aminobutyric acid (GABA) and alanine. They discussed the possibility that host
selection in solanaceous plants may have been the result of these chemosensory structures
and the concentration of amino acids in the leaves. It should be noted that both studies
excluded methionine and no electrophysiological data were collected on the response of
CPB to this amino acid. This is not surprising considering the fact that the diet of the
CPB is low in methionine and therefore would not be a candidate for the inclusion in
feeding stimulatory studies (Cibula et al. 1967). It is unknown if these sensory structures
can detect methionine and possibly act as a means to avoid plant material high in this
amino acid. This appears to be contradicted by the data in Figure 4-5, in which there was
no difference between the treatments. The larvae feeding on the Control treatment,
consuming the majority and then moving to the 1.0% L-methionine treatment, could
explain the lack of difference.
There are some differences between some of the Feeding and Development
treatments should be noted. The mean head capsule of the larvae in the 0.5%
L-methionine treatment was higher than the 0.3% L-methionine treatment while the
amount of leaf material consumed for the same treatment were the same indicating
another factor involved with the greater head capsule width. The differences could be the
result of the larger size of females and possibly could have included more females.

51
The higher concentrations of L-methionine that produced mortality similar to the
Btt is encouraging considering the occurrence of resistance to this compound seen in
many pest insect species because of reduced receptor activity and binding (Bills et al.
2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and
many are the result of exposure to a combination of different pesticide classes. The
Methionine-CAATCHl system could be exploited in cases where the only alternative is
applying different pesticides or using higher rates to break resistance. Further research is
needed to determine compatibility with Bt and L-methionine for cases in which resistance
is observed in natural populations. Given the safety of L-methionine and the shorter time
required for 100% mortality (when compared to Btt), this compound could represent a
new biorational tool for the management of the CPB.

CHAPTER 5
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY
CONDITIONS
Introduction
Integrated Pest Management practices are not restricted to agricultural pests.
Medically important insect pests are responsible for epidemics that have changed the
course of human existence, from bubonic plague spread by the Oriental rat flea
(Xenopsylla cheops Rothschild (Siphonaptera: Pulicidae)), to malaria carried by
anopheline mosquitoes. One medically important species that has had a significant
impact on human existence is the yellow fever mosquito (YFM), Aedes aegypti (L.)
(Dptera: Culicide). This cosmopolitan species is found worldwide and is the primary
vector for human dengue and yellow fever despite concerted efforts at eradication in the
United States (Womack, 1993). In the United States alone, upwards of 150,000 lives
were lost to yellow fever in the period starting in the late 18th century and into the early
20th century (Patterson, 1992). However, because of the development of a vaccine,
yellow fever has been replaced by Dengue which is now second only to malaria as a
worldwide threat (Gubler, 1998). Because Dengue fever is also vectored by the YFM, it
poses a risk by affecting tens of millions of people worldwide (Gubler and Clark, 1995).
The inclusion of the YFM in this study was an effort borne of curiosity because of
the lack of knowledge of the CAATCH1 system in other insects and the availability of
specimens for study. Mosquito larvae are particulate feeders and have dietary
52

53
requirements of methionine in the amounts of 0.0007mg/ml for the YFM. This amino
acid also is considered essential for other species of mosquito in untraceable (in those
studies) amounts (Chen, 1958; Singh and Brown, 1957). Given the high alkalinity found
in the midgut of the YFM as well as other mosquito species, this physiological condition
indicates the possibility of the presence of the CAATCH1 system in larval mosquitoes
(Dadd, 1975).
The purpose of this portion of the study was to examine the survival and
development of YFM larvae exposed to water treated with excess L-methionine (adults
were not tested given the feeding nature). In addition to L-methionine, other amino acids
were tested in an effort to see if their response (i.e., survivorship) was similar CAATCH1
responses to methionine found by Feldman et al. (2000).
Materials and Methods
Bioassav
The bioassay experiments consisted of six treatments (control, 0.1%, 0.3%, 0.5%,
0.7% and 1.0%) each with four replicates. Both L-methionine and D-methionine were
tested along with proline, Beta-alanine and L-leucine to examine the other amino acids
that were found to be reactive to the CAATCH-1 system (Feldman et al., 2000).
Bt-isrealiensis (Aquabac @ a rate of 2.3 mL/m2; Biocontrol Network, Brentwood, TN)
and proline also were included in some trials of L-methionine to allow for comparison of
both positive and negative effects. Amino acids were weighed using a Denver
Instruments Co. XD2-2KD digital scale and added to glass quart jars containing 500ml of
deionized FLO. Concentrations were based on the proportion of lg/100ml for a 1%
solution and for corresponding concentrations. Solutions were allowed to sit at room

54
temperature (23C) to permit the amino acid to fully dissolve before the addition of the
larvae. An additional trial of L-methionine buffered with Tris to a pH of 7.0 using a
Fisher Scientific Accumet pH 900 was conducted to determine if mortality was attributed
to a change in pH or exposure to the L-methionine.
Larvae of YFM (third instar) were obtained from the mosquito colony maintained
at the Department of Entomology and Nematology, University of Florida. Larvae were
transferred to the treatment jars using a camel hair, with 10 larvae per replicate for a total
of 40 Iarvae/treatment and niotai=240 for each amino acid bioassay experiment (Figure
5-1). Approximately 0.5g of finely ground fish food was added to serve as a larval food
source and nylon window screen was used to cover the tops of the jar to prevent the
escape of any emerged adults. Jars were held at 23C on a dedicated laboratory bench
top for approximately one week. The numbers of larvae surviving were recorded each
day.
Growth and Development
This experiment used the same Materials and Methods as the bioassay portion
with the exception of neonate larvae instead of 3rd instars. Eggs were placed in a tray of
water and held at 23 C for 2 days after eclosin. Neonates were removed using a camel
hair paintbrush and placed into each jar, with 10 larvae per replicate for a total of 40
larvae/treatment (nTOtai~240). Larval exuviae or dead larvae were removed and used to
examine growth rates by measuring the head capsules. Larvae head capsule widths were
measured (using an Olympus Tokyo Model 213598 stereomicroscope with an ocular
micrometer) as an evaluation of larval development.

55
Figure 5-1. Bioassay setup for yellow fever mosquito larvae exposed to various
concentrations of amino acids and Bti. Jars contained 500mL of
solution and were covered with screen to prevent the escape of
emerging adults.

56
Data Analysis
Sample sizes of all experiments were selected according to the guidelines of
Robertson and Preisler (1991) for optimal sample size and data analysis. Probit analysis
and determination of mean Lethal Concentration (LC50) were performed using PROBIT
Version 1.5 (Ecological Monitoring Research Division, USEPA) after Abbotts
correction for control mortality (Abbott 1925). Probit analysis was performed on
different concentrations (0.1%, 0.3%, 0.5%, 0.7% and 1.0%) of L-methionine, Tris-
buffered L-methionine, D-methionine, Beta-alanine, proline and L-leucine for 24,48, 72
and 168 hours (the end of the trials). Survival data were normalized to the previous value
when control mortality was greater than the treatment mortality, to produce a smoother
trend line. Statistical analyses were performed on the data using Minitab Version 12.
Analysis (Minitab, Inc; State College, PA) of the data included One-way ANOVA and
separation of means using Tukeys Multiple Comparison test (Zar 1999).
Results
Bioassav
Mortality of YFM larvae in both the unbuffered L-and D-methionine trials was
similar with low or no mortality, at the 0.1% concentrations (Figures 5-2 and 5-3). The
0.3% concentration had lower mortality with D-methionine (45%) than L-methionine
(75%) and greater than 80% mortality for the 0.5% concentration for both isomers.
Higher concentrations of both D-and L-methionine forms produced 100% mortality of
the larvae within 2 days after treatment.
Greater than 40% mortality was observed for the buffered 0.1% L-methionine
concentration with complete mortality for the remaining treatments within 5 days of

57
Figure 5-2. Mortality of yellow fever mosquito larvae exposed to various
concentrations of L-methionine (nTOtai=240). Data were
adjusted using Abbotts formula for control mortality.

58
Days of Exposure
Figure 5-3. Mortality of yellow fever mosquito larvae exposed to various
concentrations of D-methionine (nTOtai=240). Data were
adjusted using Abbotts formula for control mortality.

59
exposure (Figures 5-4). The 1.0%L.methionme treatment caused 100% mortality after 2 days
while the Bti treatment took 3 days to reach the same level of control. The proline
treatment caused less than 10% mortality.
In contrast to methionine, survival of YFM larvae exposed to proline and
L-leucine was higher, with only approximately 20% mortality for the higher 0.7% proline
and 1.0% proline concentrations (Figure 5-5) and less than 3% mortality with the highest
L-leucine concentration (Figure 5-6). Beta-alanine mortality was similar to the
L-methionine treatments with between 75% and 83% mortality for the 0.5% Beta-alanine
thru 1.0% Beta-alanine concentrations, respectively, greater than 40% mortality with the
0.3% Beta-alanine, and less than 5% mortality for the 0.1% Beta-alanine concentrations
(Figure 5-7).
Growth and Development
Developmental rates of YFM larvae resulted in three distinct groups, with the
control and proline treatments, producing virtually identical results; both were
statistically different from the 0.1%L-methionine treatment and the remaining
L-methionine treatments (Figure 5-8). The Bti treatment was statistically the same as the
0.3% L-methionine to 1.0% L-methionine treatments, with very little growth taking
place.
Probit analysis for unbuffered L-methionine (nrotai=40 for 5 treatments; 0.1%,
0.3%, 0.5%, 0.7% and 1.0%) revealed an overall LC50 of 0.19% concentration for the
YFM after 7 days of exposure (Figure 5-9). The LC50 of 1.2% for 24 hours dropped to
0.41% after 48 hours and to 0.24% after 72 hours. When the L-methionine treatments
(same concentrations) were buffered to a pH Of 7.0, the values dropped to 0.64% for 24

60
Days of Exposure
Figure 5-4. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Tris-buffered L-methionine (nxotai=240). Data
were adjusted using Abbotts formula for control mortality.
Note the longer exposure because of the bioassay involving
neonates instead of 3rd instars. Note the overlap in some of the
trend lines on Day 1 with the 0.3% L-methionine and 0.5% L-
methionine treatments.

61
Figure 5-5. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Proline (nTOtai=240). Data were adjusted using
Abbotts formula for control mortality. Note the overlap of
trend lines for all treatments except the 0.7% L-methionine and
1.0% L-methionine treatments.

62
Figure 5-6. Mortality of yellow fever mosquito larvae exposed to various
concentrations of L-leucine (nTotai=240). Data were adjusted
using Abbotts formula for control mortality. Note the
overlap in trend lines for all treatments.

63
Days of Exposure
Figure 5-7. Mortality of yellow fever mosquito larvae exposed to various
concentrations of Beta-alanine (nrotai=240). Data were adjusted
using Abbotts formula for control mortality.

Mean Head Capsule Width (mm)
64
(Error Bars @ 95%; F 2.5
2
1.5
1
0.5
0
Figure 5-8. Mean head capsule widths of yellow fever mosquito larvae
exposed to various Tris buffered (7.0 pH) concentrations of L-
methionine (niotai-320). Proline (1.0%) and Bti were included
for comparison as positive and negative controls. Error bars
denote 2 SE. Bars within treatments having the same letter are
not statistically different (Tukeys MST, PO.OOl).

LC50 (% Concentration)
65
1.4
1.06
(0.92-1.4)
0.8
0.6
0.4
0.2
0
1.20
(0.95 2
24h
48h
72h
Overall (240h)
L-methionine
L-methionine (Buffered)
D-methionine
Beta-alanine
0.41(0.21-0.74)
0.44 (0.39 0.48)
0.50 (0.43 0.59)
0.33 (0.29 -0.38)
0.35 (0.28- 0.42)
0.32 (0.27 0.36)
0.34 (0.27-0.41)
(0.16-0.22)
Figure 5-9. Concentrations (%) resulting in 50% mortality (LC50) of yellow fever
mosquito larvae exposed to various amino acids after 10 days
(nT<#ai=240 for each amino acid). Number range following value is the
95% confidence limits. Proline and L-leucine were also tested but did
not exhibit sufficient mortality to allow for Probit Analysis.

66
hours, and to 0.11% for 48-168 hours and remained constant since the trial lasted longer
because of the use of neonates instead of 3rd instars. The D-methionine treatments were
similar with 0.44% for 24 and 48 hours, 0.33% for 72 hours and 0.32% after 168 hours.
While not as striking as the others, Beta-alanine had a LC50 concentration of 1.1% after
24 hours, dropped to 0.5% after 48 hours and leveled off around at 0.35% after 72 and
168 hours. Probit analysis of the Proline and L-leucine treatments was not performed, as
the mortality associated with those treatments was too low (Figures 5-5 and 5-6).
Discussion
Although not commonly encountered, the D- form of methionine had virtually the
same effect as the L- form on larval mosquito mortality. The D-and L-methionine trials
showed that the D- form had lower mortality associated with it than the more reactive
L-counterpart. Insects do not commonly use the D- form of amino acids, although
D-methionine is metabolized by some orders to a limited extent (Ito and Inokuchi, 1981).
The YFM could be an example of this phenomenon.
Because of the nature of the CAATCH1 system in the alkaline midgut, buffering
may have acted to increase the effectiveness of the system. Buffering the solutions did
result in an increase in mortality, with even the lowest concentration of 0.1%
L-methionine exhibiting a two-fold increase with the buffered form (Figure 5-4).
Complete mortality was reached sooner with the buffered forms even for concentrations
that did not reach 100% in the unbuffered form. In a field setting, the addition of
L-methionine would be buffered naturally by the chemical properties of the bodies of
water to which it was applied and similar results would be expected.

67
Jaffe and Chrin (1979) found the adults of YFM females infected with Brugia, a
filiaral parasite, were depleted of free form methionine because of the infection and were
able to make up the difference by converting homocysteine to methionine with a special
synthetase. The ability of YFM adults to synthesize methionine from homocysteine may
be present in the larvae as well. This could be the result of the lack of methionine in the
diet and possible evidence of the CAATCH1 system being present in at least the adult
stage. The susceptibility of the larvae to L-methionine also could be the result of
overexposure to a compound that is normally not encountered in high concentrations
(>0.1%). However, the alkalinity of the particulate feeding larvae and the high mortality
to L-methionine suggests that the CAATCH1 system is present and could be exploited in
other species with similar midgut characteristics (Dadd, 1975).
The survival of YFM larvae exposed to both Beta-alanine and L-leucine was
unusual in that they each had the opposite effect on the YFM larvae. L-leucine was
expected to have similar blocking properties as L-methionine based on CAATCH1
research (Feldman et al., 2000). Instead, almost no mortality was observed indicating the
possibility of another system involved with the transport of this amino acid. Conversely,
beta-alanine was not found to be reactive with the CAATCH1 system based on the work
of Feldman et al. (2000). The unusually high larval mortality associated with this amino
acid may be the result of a yet to be discovered midgut property.
The similar mortalities observed for the higher concentrations of L-methionine
and Bti is encouraging considering the resistance to this compound that has been
documented in many insect species because of reduced receptor activity and binding
(Bills et al., 2004; Nester et al., 2002). Resistance in insects involves a variety of

68
mechanisms and many are the result of a combination of different pesticide classes. The
CAATCH1 system is one that could be exploited in cases where the only alternative is
applying different or higher rates of pesticides to break resistance. Further research is
needed to determine compatibility of Bti and L-methionine for cases in which resistance
is observed in natural populations. Given the safety of L-methionine and the similar time
required for 100% mortality (when compared to Bti), this compound could represent a
viable alternative to traditional biorational compounds used in the management of the
YFM or other susceptible pest mosquito species.

CHAPTER 6
FIELD EVALUATION OF L-METHIONINE AS AN INSECTICIDE
Introduction
The role of methionine in animal systems is well known and only recently
understood in plants. Methionine is required for protein synthesis; it is a precursor to
several important biochemical compounds including ethylene and polyamines, sulfate
uptake and assimilation, and also acts as an activator of threonine-synthase (Giovanelli et
al. 1980; Droux et at. 2000; Bourgis et at. 2000; Zeh et at. 2001). Recently, research has
focused on the transgenic modification of crop plants to overproduce methionine in order
to increase their nutritional quality without affecting other biochemical processes (Zeh et
al 2001). However, little work has been conducted on the effects of exogenous
methionine and it became important to understand the role of externally applied
methionine on plant health.
Furthermore, the application of L-methionine to plants exposed to natural
conditions presents additional problems in terms of how long the residue remains on the
plant. Observations of other experiments using L-methionine revealed the tendency of
this compound to crystallize after the aqueous portion evaporated forming a brittle, crusty
coating that is easily removed. This coating does not appear to interfere with respiration
and transpiration at the concentrations studied (1% and lower). To prevent the loss of
L-methionine from the plants in a natural setting, the adjuvant Silwett L-77 (Helena
Chemical; Collierville, TN) was included in this portion of the study in an effort to
increase residual activity on the plant. Silwet L-77 is a nonionic organosilicate
69

70
surfactant that has wetting and spreading properties (Helena Chemicals 2002) and was
found to be compatible with solutions of L-methionine.
The objectives for this portion of the study were to examine the effects of a
methionine and Silwet L-77 mixture on a crop plant (eggplant) in terms of yield (both
fruit weight and total yield) and to evaluate this mixture as an insecticide under natural
conditions.
Materials and Methods
Preliminary Investigation of Silwet L-77 and L-methionine
Adult CPBs were obtained from the University of Florida Horticultural Unit,
Gainesville and held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware
cloth (to facilitate cleaning) and held at 27C, 60% relative humidity and 16L/8D
photoperiod in FRIUs. Twenty-four adults were exposed used in each of the 5
treatments, with 4 replicates per treatment (nTOtai=120). Adults were used because of the
lack of sufficient numbers of larvae to test. Excised leaves were dipped in solutions of
deionized H2O containing different concentrations of methionine and Silwett L-77
(0.5% concentration), 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine and
controls of deionized H2O and deionized H2O +Silwet L-77. The additional control
was to determine the possible insecticidal properties of Silwet L-77 alone and to make
sure the addition of this adjuvant did not affect mortality or deter feeding.
Plot Design
Eggplants {Solarium melongena L.,Classic variety) were grown and maintained
at the University of Florida Horticultural Unit, Gainesville, from 18 June to 04 November
2001. Eight, one hundred ft. rows of plants were used for this study, with two rows on
each side consisting of buffer rows and four rows in the middle used for the experiments.

71
Each row contained the 4 treatment plots of 10 plants (control (0% L-methionine), 0.1%
L-methionine, 0.5% L-methionine and 1.0% L-methionine in deionized water solutions)
in a Latin square design. Plants within treatment plots were spaced 3 feet apart while
treatment plots were 9 feet apart. Figure 6-1 shows the diagrammatic representation of
the field plot.
Plant Yield
Before beginning the experiment, all developing eggplants were removed from
the plants in an effort to standardize the treatments and ensure all eggplant development
occurred after the exposure of methionine. Treatments were administered using a KQ 3L
CO2 (Weed Systems, Inc.; Hawthorne, FL) backpack sprayer charged to 30 lbs PSI and a
3-nozzle boom to ensure complete coverage of the plant (Figure 6-2). Each treatment
consisted of a 3L application over the 4 representative groups. The adjuvant Silwett
L-77 (0.5% concentration) was included to improve the residual effect of the
methionine under the field conditions. Plants were sprayed a total of nine times at
approximately two-week intervals. Fruits were harvested at various times during the
study and were weighed in the field using a Tokyo Electronics hand-held digital scale.
Pest Introduction
Neonate CPB larvae were reared on excised eggplant leaves for two days at 27C,
60% relative humidity and 16L/8D photoperiod in FRJUs to ensure healthy individuals
for the test. Larvae were transferred to the field plants using a camel hairbrush and the
branch marked with flagging tape. Introduction was made after the last spray treatment
in November. Ten larvae were placed on each plant for a total sample size of 1,600
individuals. Plants were inspected for the next 5 days and larvae encountered noted.

Barrier Barrier
Rows Rows
Figure 6-1. Overview of the design layout used to study the effects of
L-methionine and Silwett L-77 solutions on yield of
eggplant. Rows were four feet apart with individual
plants three feet apart and treatments nine feet apart.
Each letter represents a group of ten eggplants.

73
Figure 6-2. Weed Systems, Inc. KQ 3L CO2 backpack back sprayer used for application
of L-methionine and Silwett L-77 solutions. Boom consisted of three
nozzles (middle top and end of each arm). In total, 3L were applied per
treatment every two weeks from 09 July to 31 August 2001.

74
Data Analysis
Data from the fruit and the CPB experiments were analyzed with ANOVA using
Minitab Version 12. Survivorship of CPB was corrected using Abbotts formula (Abbott
1925) to account for control mortality, mean separation was performed using Tukeys
multiple comparison procedure (Zar 1999). Data for both the eggplant weight mean per
treatment and also mean number of eggplants per treatment were analyzed using paired
t-test.
Results
Effects of L-methionine and Silwett L-77 on CPB Adults Under Laboratory Conditions
Little mortality was observed with the adult CPB at the 1.0% L-methionine
concentration (Figure 6-3). The 0.5% L-methionine concentration had the highest
mortality of all the treatments at approximately 20% with the other treatments showing
no adverse effects after correction for control mortality.
Effects of L-methionine and Silwett L-77 on yield
In total, 735 eggplants were collected during the course of this study from 09 June
to 31 August 2001. Mean weight and yield of eggplants between the treatments were not
statistically different from each other (Figures 6-4). Control plants produced 195 fruits
with a mean weight of 276.9 grams, followed by the 0.1% treatment with 191 fruits at
281.2 grams. The 0.5% and 1.0% treatments yielded 175 and 174 fruits with mean
weights of 295.7 grams and 283.6 grams, respectively.
Survival of CPB larvae
No statistical difference in survivorship of CPB larvae was observed between the
three treatments for the first day after exposure (Figure 6-5) but treatment differences

75
Figure 6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant
leaves treated with L-methionine and the adjuvant Silwett L-77
(nTotai=120). Data corrected for control mortality using Abbotts formula.
Note the overlap in trend lines for the Control treatments and 0.1%L-methk>nine
treatment.

76
(Error Bars @ 95%; F(1)3fl56 =2.6626, F =0.37137; P= 0.77377)
Control 0.1% 0.5% 1.0%
(n=195) (n= 191) (n=175) (n=174)
Control 0.1% 0.5% 1.0%
(n=T95) (n=191) (n=175) (n=174)
Figure 6-4. Effects of L-methionine and Silwett L-77 on eggplant yield
(A) and mean weight in grams of fruit (B) from 09 June to
31 August 2001. Error bars denote 2 SE. There was no
statistical difference for either eggplant yield or mean
eggplant weight (Tukeys MST, Z^O.05).

77
Control
0.10%
0.50%
1.00%
20
Days after treatment
Figure 6-5. Mortality of Colorado potato beetle larvae on eggplants treated with
L-methionine and Silwett L-77. Mortality of larvae corrected using
Abbotts formula (Abbott, 1925). Analysis performed on arcsin
transformed data. Error bars denote 2 SE. Data points having by the
same letter are not statistically different (Tukeys MST, P=0.05)

78
were observed thereafter. By Day 4 the 1.0% and 0.5% treatment were the only
treatments that were statistically different from the control. There was substantial
unexplained attrition of CPB larvae in the field for all treatments, which leveled off by
Day 3. Data from day 5 was discounted because of the onset of a severe cold front that
made it difficult to separate the effects of the weather from the treatments affects.
Discussion
The results of the field studies show that, using conventional application
techniques, a mixture of methionine and Silwett L-77 did not appear to affect eggplant
yield. Furthermore, the same combination produced substantial control of CPB larvae
under natural field conditions after four days. Dahlman (1980) found that L-canavanine,
a non-protein amino acid, could be used in the same manner for control of THW on
tobacco, but the widespread use of this compound was limited by the cost ($107.85 for lg
L-canavanine versus $3.35 for lg of L-methionine (Fisher Scientific International 2004)),
adverse effect on plant development (Nakajima et al. 2001), and toxicity to vertebrates
(Rosenthal 1977). Although complete coverage of the plant was not feasible,
approximately 2.5 grams to 7.5 grams of L-methionine was applied to the plants in each
of the treatment plots. Each plant, based on the amount applied, received approximately
7.5xl06 pg for the 1.0% L-methionine treatment, 3.8xl05 pg for the 0.5% L-methionine
treatment and 2.5x104 pg for the 0.1% L-methionine treatment. This compares to only 4pg of
L-canavanine, which resulted in decreased size, fecundity, and mortality of THW under
field conditions (Dahlman 1980). It should be noted that the toxicity of L-canavanine is
well documented and has a different mode of action than L-methionine and cannot be

79
compared directly. However, the cost for the amount of L-canavanine required for large-
scale application far exceeds that of the largest amount of L-methionine needed.
Despite the lack of a statistical difference between treatments for both mean
weight and mean yield of eggplant, there were some interesting disparities within the
data. First, there was an observable difference in mean weight of the eggplants between
the treatments and the control. All eggplant weights were greater for the treatments than
the control, with the 0.5% L-methionine concentration treatment producing the highest
mean eggplant weight. It would appear that excess methionine decreases the number of
fruit produced, but those fewer eggplants weighed more. Further research is needed to
better understand the differences observed during this study.
The addition of Silwet L-77 did not appear to adversely affect survival of CPB
as seen in the preliminary tests on the adults and on the larvae during the field release
(Figures 6-3 and 6-5). The low adult mortality observed could be attributed to the ability
of this species to stop feeding and fly to a more suitable food source. Because the adults
were unable to move to an untreated leaf, they were observed sitting motionless on the
underside of the leaves. This was not observed in either of the controls as they were seen
actively feeding the majority of the time.
One aspect of this research that was not examined is that of fertility and fecundity
of adults exposed to excess amounts of L-methionine. Despite the fact that methionine is
used for egg production in many insect species, excess concentrations may act as a
deterrent to feeding causing the adults to stop feeding and to seek other food sources.
The lag time from the cessation in feeding to finding another food source may be long

80
enough to significantly lower the fecundity of the females and possibly interfere with
other behaviors such as mating.
During the course of this portion of the study, some anecdotal data were collected
based on personal observations. Predators (mainly arachnids) were observed on the
plants until the end of the experiment. Other insects also were observed feeding on plants
after treatments including piercing-sucking insects (i.e., aphids, coreids and cicadellids)
with foliage feeders such as caterpillars rarely encountered except found only on control
plants. Attempts to control predators via manual removal were unsuccessful, and
predation may have contributed to the observed decrease in CPB. Because predators
were present on all treatments, loss from predation was corrected with the use of Abbotts
formula. The presence of natural enemies indicates the selectivity of the L-methionine in
the field. The amount of methionine ingested by the predators was probably very small
because they fed on other insects not plant material.
Another set of observations on the safety of L-methionine was the exposure of
potted eggplants to high (1.0% methionine in distilled H2O solution). In total, five plants
were sprayed daily with the methionine solution and compared to five plants sprayed
with water alone for 14 days. The only difference in the plants was the browning of the
leaf tips and edges of the methionine sprayed plants. This also was seen in the excised
leaf experiments with THW and CPB. A possible reason for this occurrence was the
excess sulfur in the methionine might have burned the leaves. As mentioned earlier, the
concentration was very high and also applied daily. Applications of the same
concentration did not affect the plants in the field plots, indicating that treatments
conducted at 2-week intervals would be safe for the plant.

81
Overall, it appears that L-methionine can be used in a natural setting to control
CPB larvae without affecting crop production. The adjuvant Silwett L-77 worked well
with L-methionine in controlling CPB larvae but not the adults. The lack of effectiveness
on the adults may be attributed to their ability to stop feeding and living off of reserves
acquired during the larval stage until suitable food sources can be found. It is unknown if
L-methionine, alone or in combination with Silwett L-77 adversely affects fecundity of
the adults.

CHAPTER 7
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF
SELECTED NONTARGET SPECIES
Introduction
A biorational pesticide is defined as one that is effective against pest species but
innocuous to non-target organisms and not disruptive to biological control agents and
beneficial species (Stansly et al. 1996). To test L-methionine as a potential pesticide and
determine if it could be considered biorational, it was necessary to examine the effects of
this compound on selected nontarget species that could possibly come into contact with
it, either directly while on the plant or indirectly via incidental contact or as a host that
has come into direct contact with this compound. The species chosen reflect a variety of
non-target organisms, mainly those that were shown to be important in controlling some
pest species. The pink spotted ladybird beetle, Coleomegilla maculata (DeGeer), the
mottled water hyacinth weevil, Neochetina eichhorniae Warner, and the greenbug
parasitoid, Lysiphlebus testaceipes (Cresson) all are beneficial insects that have been
effective against pests in the state of Florida and also are common and readily available.
Each species also represents a different feeding guild (predator, herbivore and parasitoid,
respectively) to ensure a thorough examination of the possible effects of methionine as it
might be encountered in under natural conditions.
The pink spotted ladybird beetle (PSLB) is an abundant polyphagus species that is
known to feed on many lepidopteran and coleopteran pests, including the Colorado
potato beetle, in which it was responsible for over 50% of the predation on eggs and early
82

83
instars (Andow and Risch 1985; Giroux et al. 1995; Griffin and Yeargan 2002; Groden et
al. 1990; Hazzard et al. 1991; Hilbeck and Kennedy 1996; Munyaneza and Obrycki
1998). This species is widespread throughout North America, and has been shown to
provide effective biological control in several crop species, including com, crucifers,
tomato and potato (Hoffman and Frodsham 1993). However, the PSLB was found to be
susceptible to carbaryl and menthamidophos, the same pesticides used for the control of
many aphid species (Hoffman and Frodsham 1993).
Since its introduction into the United States in 1884, water hyacinth (Eichhornia
crassipes (Mart.) Solms-Laubach) has infested waterways of the southeast that has cost
upwards of $2 million to control in Florida alone (Schardt 1987). The mottled water
hyacinth weevil (MWHW), native to Argentina, was first released in Florida in 1972 and
subsequently to other states and countries in an effort to control water hyacinth (Center
1994). The genus is restricted to feeding on members of Pontederiaceae, with the
MWHW feeding mainly on the introduced water hyacinth; it can be found virtually
everywhere the host plant is present (Haag and Habeck 1991; Center et al. 1998).
The greenbug parasitoid (GBP) is an important natural enemy of many cereal
aphids. This species is known for the production of mummies, the bodies of
parasitized aphids that act as a protective case for the developing wasp pupa, and is
considered by many to be tolerant to cold temperatures (Elliott et al. 1999; Knutson et al.
1993; Wright 1995). However, this greenbug parasitoid is an insect and is just as
susceptible to pesticides despite the protective case of the immature form (Knutson et al.
1993).

84
The purpose of this portion of the study was to examine the effects of L-
methionine on selected nontarget species that are both important in terms of being
beneficial in controlling other pest species and also represent different feeding guilds that
would come into contact with this compound in different ways (e.g., on prey items, on
plant surfaces, hosts of parasitoids).
Materials and Methods
Coleomeeilla maculata
Adults were obtained from ENTOMOS, LLC (Gainesville, Florida), and were
held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth stage
inserted (to facilitate cleaning) at 27C, 60% relative humidity and 16L/8D photoperiod
in FRIUs. Natural diet consisted of excised cotton leafs infested with aphids (Aphis
gossypii Glover (Hemiptera: Aphididae)). Leaves were then dipped into either a 1.0%
L-methionine solution or 0% L-methionine (control) mixed with deionized H2O. Five
adults were used in each replicate for a total n=30 for each treatment. Leaves were
replaced every other day from 27 October 2002 to 07 November 2002. Artificial diet was
obtained from ENTOMOS and prepared according to their guidelines with the exception
of the inclusion of methionine for the 1.0% L-methionine treatment (wt/wt). Diets were
replaced every other day from 27 October 2002 to 07 November 2002. Ten adults were
used for each replicate for a total n=60 for each treatment. Data was normalized to 0%
mortality when the treatments were corrected for control mortality (i.e., when the control
mortality was greater than that of the treatment).

85
Neochetina eichhorniae
Adults of the MWHW were used in this study since the larvae and pupae are
buried deep in plant tissue and therefore not likely to come into contact with methionine
that could be present in a body of water. Specimens were supplied by Hydromentia, Inc.
(Ocala, FL), from areas around South Florida. Weevils were maintained following the
procedures outlined by Haag and Boucias (1991), with small petri dishes fitted with
moistened filter paper and freshly cut water hyacinth leaves. Water hyacinth plants were
collected from Lake Alice on the campus of the University of Florida and maintained in
the University of Florida, Department of Entomology and Nematology greenhouse.
Treatments consisted of cut leaves dipped in deionized H2O (control) or solutions
containing 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine or 1.0% proline.
Prior to weevil exposures, each leaf was inspected for feeding scars or damage
and noted to ensure the counts were based on current feeding. Each treatment consisted
of 4 replicates with n=5 per replicate (n=20 per treatment and total n=100). Weevils and
hyacinth leaves were held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a
hardware cloth (to facilitate cleaning) and maintained at 27 C, 60% relative humidity
and 16L/8D photoperiod in FRJUs. Fresh leaves were provided every 4 days; exposed
leaves were preserved in sealed plastic bags and placed in a refrigerator until scars could
be counted. Feeding damage was determined (with the use of an Olympus Tokyo Model
213598 stereo microscope) by the total number of scars present with each counted scar
marked with a fine tipped permanent marker (Figure 7-3).
Statistical analyses of the weevil data were performed using Minitab Version 12
(Minitab, Inc.; State College, PA). Feeding scars on control and treatment leafs were

86
compared with a One-way ANOVA and mean separation was performed using Tukeys
Multiple Comparison test (Zar, 1999).
Lvsiphlebus testaceipes
To test the effects of methionine on the GBP, cotton plants (Gossypium sp.;
Family: Malvacae) were grown and maintained at the University of Florida, Department
of Entomology and Nematology green and shade houses from 07 October 2002 to 25
November 2002. Aphids (A. gossypii Glover) were supplied from other experiments
using this organism and kept on plants within a sealed greenhouse to prevent unwanted
parasitism. Plants were maintained in the sealed greenhouse, infested with aphids and
then placed in the open shadehouse area to encourage parasitation. In total, 20 plants
were used for 2 treatments, 1.0% L-methionine and 0% L-methionine (Control) mixed
with deionized H20. Plants were sprayed weekly (12 October 2002 through 17
November 2002) with approximately 10 ml of solution using a hand-held spray bottle.
Counts of parasitized aphids began approximately two weeks after placing plants outside
to ensure adequate time for parasitism (Royer et al. 2001). Counts were made using a
hand lens and counter; mummies with exit holes were enumerated and removed. A
few parasitized aphids were removed and held in glass vials to ensure correct
identification of the parasitoid.
Data Analysis
Data from the parasitoid experiments were analyzed using Minitab Version 12
(Minitab, Inc.; State College, PA). Control and experimental plants were compared
against one another with a One-way ANOVA and separation of significant means was
performed with Tukeys Multiple Comparison test (Zar, 1999).

87
Results
Coleomesilla maculata
There was virtually no difference between the control and treatment groups for
either the artificial or natural diet tests after correction for control mortality. Mortality
was slightly higher for the control groups than the 1.0% L-methionine treatment (Figures
7-1 and 7-2). Further analysis was not necessary because of the identical numbers.
Neochetina eichhorniae
Total mortality for the treatments was less than 20% for all treatments, with the
individual treatments having similar results (Figure 7-4). Feeding damage ranged
between 2,000 and 4,000 scars per treatment and an average of 10.7 to 16.9 scars per
survivor during the course of the experiment (Figure 7-5). No statistical differences were
observed between the treatment and control groups
Lvsiphlebus testaceipes
In total, 188 and 232 aphid mummies with exit holes were found on treatment and
control plants, respectively. Means for each treatment were not statistically different for
each collection period or overall based on One-way ANOVA (Figure 7-6) with the only
exception being the second and last collection period.
Discussion
In general, L-methionine did not have the same toxic effect on the non-target
organisms tested when compared to the pest species exposed to the compound in
previous chapters. The pink spotted ladybird beetle adults actually showed the least
amount of susceptibility to L-methionine. Survival of the adult beetles was higher in the
1.0% L-methionine treatments than the control for both the artificial and natural diet

88
Figure 7-1. Mortality of Coleomegilla maculata adults after exposure to L-methionine
treated artificial diet. Data corrected for control mortality using Abbotts
formula.

89
100
80
g
13 60
S 40
20
0
Figure 7-2. Mortality of Coleomegilla maculata adults after exposure to L-
methionine treated cotton plant leaves infested with aphids. Data
corrected for control mortality using Abbotts formula.
Control- ND
1.0% L-methionine ND
Survivorship of 1.0%L-methionine Grp> Control Grp
4 5 6 7 8 9
Days After Exposure
10 11 12

90
Figure 7-3. Feeding scars on water hyacinth (Eichhornia crassipes) leaf after exposure
to Neochetina eichhorniae adults. Black marks represent feeding scars
marked with a fine tip marker to aid in counting (other side counted but not
shown).

91
Figure 7-4. Mortality of Neochetina eichhorniae on treated water hyacinth
leaves. Data corrected for control mortality using Abbotts
formula.

92
(Error bars @ 95%; F(00S)4> u= 0.98, F=3.33; /* =0.038)
1400
Control 0.1% 0.5% 1.0% Proline
Figure 7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated
with L-methionine and Proline. No statistical differences were observed
between treatments (Tukeys MST, P=0.038).

Mean Number of Parasitized
Mummies/Plant
93
(Error bars @ 95%; 1=4.41: F=3.25; P =0.005)
16
10/22/02 10/29/02 11/5/02 11/12/02 11/19/02 11/26/02
Figure 7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with
L-methionine. Ten plants were used for each treatment and held in the
shade house at the University of Florida, Department of Entomology and
Nematology from 22 October to 25 November 2002. No statistical
differences were observed except for the second and final collection date.

94
trials. One possible explanation for this observation could be that the excess
L-methionine increased the dietary quality of the artificial and natural diets for the PSLB
in the treatments. However, because only adults were available, further tests are needed
to determine if the larvae, also predaceous on the same pests as the adults, are sensitive to
this compound. It should be noted that the midgut properties (i.e., alkalinity) for this
species are not well known and may not even have the CAATCH1 proteins present in the
midgut.
The mottled water hyacinth weevil also appears not to be adversely affected by
exposure to excess amounts of L-methionine despite its herbivorous habit like the THW
and CPB. Another weevil within the same family {Anthonomus granis Boheman
(Coleptera: Curculionidae)) is known to have an acidic midgut and the same could apply
to the MWHW based on these results (Nation 2001). Therefore, this species and possibly
other weevils may not be affected by compounds like L-methionine because of the lack
of an alkaline midgut needed for the CAATCH1 protein to operate (Feldman et al. 2000;
Quick and Stevens 2001). Again, further research is necessary to determine if
CAATCH1 proteins are present in this weevil species.
The greenbug parasitoid also was unaffected by exposure to the excess
L-methionine found on treated leaves infested with aphids. Dadd and Krieger (1968)
found higher methionine requirements for the greenbug Myzus persicae Sulzer
(Hemiptera: Aphididae) when cysteine is scarce because of its ability to transform excess
methionine to much needed sulfur and could possibly explain the parasitoids tolerance to
high methionine concentrations. Because of the life cycle of the GBP, and many other
parasitoids, direct contact with compounds such as L-methionine would occur inside the

95
body of the host, and not through direct contact with the foliage where the compound was
applied. There is a possibility for the parasitoid having higher methionine requirements;
based on filarial worm infected Aedes aegypti (L.) (Dptera; Culicidae) females and the
associated drop in methionine levels in the haemolymph (Jaffe and Chrin 1979). This
makes alternatives such as L-methionine safe for use around beneficial insects like the
greenbug parasitoid.
Overall, the results indicate that the PSLB (C. maclala), the MWHW (JV.
eichhorniae) and the GBP, (L. tes tace ipes) were not adversely affected by exposure to
L-methionine in excess concentrations in a variety of artificial and natural diets.
Survivorship and feeding rates were not statistically different between control and
treatment groups for each species. From these data, it can be concluded that
L-methionine is safe for use with beneficial insects and could be considered biorational
in that it showed no adverse effects on non-target species. It also should be stressed that
additional testing on other beneficial insects would be, on a case by case basis, necessary
to examine the safety and biorational qualities of L-methionine.

CHAPTER 8
SUMMARY AND DISCUSSION
The creation and implementation of Integrated Pest Management (IPM) strategies
to combat pest species were developed as a response to the economic losses associated
with the overuse of chemical control. However. IPM strategies are not widely used
because of the lack of alternatives and the ease of use of pesticides. This has resulted in
the resistance to pesticides in many insect species, including economic and medical pests.
In an effort to provide alternatives to traditional chemical control, biorational methods
have been investigated and one such avenue is the use of non-protein amino acids.
Chapter 2 covered the history of the use of non-protein amino acids as a pesticide,
and discussed the CAATCH1 system and the safety of L-methionine. Only a handful of
these amino acids have been investigated as a means of controlling insect pests but still
lack the practicality and cost effectiveness as current chemical control methods. Recent
discovery of a new midgut membrane protein, CAATCH1, has revealed a new possibility
in insect control. The CAATCH1 system works in alkaline conditions and responds to
different amino acids, mainly the reduction in ion flow after exposure to methionine, an
essential amino acid required for normal development and metabolism of many species
including humans. The use of a compound such as methionine would be an excellent
addition to the IPM arsenal because of its relative safety to vertebrates and warrants
further study as a pesticide.
Chapters 3,4, and 5 were dedicated to examining the effects of L-methionine, a
common analog of methionine, on three different economic and medically important
96

pests. The tobacco horn worm (THW), Colorado potato beetle (CPB) and the yellow
fever mosquito (YFM) were tested and found to be susceptible to concentrations greater
than 0.1%. Diets, both natural and artificial, containing this compound resulted in the
complete mortality of THW and also in the natural diet for CPB. Development and
feeding rates were also affected by the addition of L-methionine to diets for THW and
CPB. Survivorship and developmental rates of YFM were also affected by the addition
of this amino acid to the larval habitat.
In Chapter 6 it was found that the field application of L-methionine under natural
conditions was able to control CPB. It was also determined that L-methionine was
compatible with Silwett L-77, a commonly used adjuvant, and showed no detrimental
effects on crop yield of eggplant.
Finally, the application of a compound such as L-methionine has to be able to
control the pests that it is used against and not have an effect on beneficial organisms that
may come into contact with this compound. Chapter 7 detailed the results of tests that
involved various beneficial insects from different feeding guilds (herbivore, predator and
parasitoid) showed that L-methionine does not appear to pose a threat to nontarget
organisms.
One aspect of the use of a compound like L-methionine that is very important is
the relative safety. The health hazards related to the contamination of the environment
with pesticides are well documented and in the recent years have resulted of the review
and removal of several insecticides from commercial and private use. The use of
L-methionine as an insecticide would alleviate the dangers associated with other
pesticides. The approved use as a nutritional supplement for livestock feed is a testament

98
to the safety of this compound and residual found on the plant does not pose the same
risk to the human population.
It is difficult to understand how a compound such as methionine can be
considered essential and deadly within the same organism. To understand this
dichotomy, an examination of the role of this compound and how it relates to
metabolism, development and reproduction is necessary.
Although the diet of the THW is lacking high concentrations of methionine, the
use of hexamerins may account for the levels needed for the biosynthesis of JH. The
larvae take in methionine, metabolizing what is needed and storing the rest for later on
during metamorphosis. In contrast, the larvae of the diamondback moth (Plutella
xylostella (Lepidoptera: Plutellidae)), feeding mainly on methionine-rich crucifers, lack
hexamerins with high methionine concentrations (Wheeler et al. 2000). The levels of
methionine encountered in a normal diet are below what the CAATCH1 proteins are
capable of processing and may also be affected by the presence of symbiotic bacteria that
is responsible for methionine oxidation in some insects (Gasnier-Fauchet and Nardon
1986a; 1986b). It is when the concentration exceeds the handling capacity of the midgut
that problems occur. The time it takes to digest material containing natural amounts of
methionine could be long enough for the CAATCH1 system to recover from exposure.
The difference between the artificial and natural diet LC50 for the THW (Figure 3-8)
appears to support the idea that bound methionine (i.e., incorporated into the diet and not
applied topically) takes longer to cause problems for the organism (if any) versus the
relatively quick kill associated with the free methionine present on the leaf surface. The

99
target ingests the methionine first as it feeds ensuring the overload CAATCH1 system
and eventually death.
As for the stored methionine, it is released from the storage proteins as needed to
synthesize juvenile hormone and allow for transformation in addition to other functions.
The remaining methionine is then used for protein synthesis in the tissues around the
ovaries to boost yolk production, as seen in the transfer of methionine from male to
female Drosophila species (Bownes and Partridge 1987). In the THW, the presence of
hexamerins with high methionine content may be an alternative to the male contribution
possibly found in its ejaculate. Methionine-rich hexamerins are common in Lepidoptera
and have been shown to provide the larvae a source of amino acids during the synthesis
of these proteins during the last stage of larval development (Wheeler et al. 2000). In
addition to the need for methionine for metabolism and reproduction, the release of
methionine may also in part account for the decrease in ion transport of the posterior
region of the midgut during larval molts and the wandering stage present before pupation.
Currently, little is known regarding the mechanisms involved with the decrease of ion
transport during these developmental stages (Lee et al. 1998). Clearly there appears to be
more to the role methionine plays in the development of some insects other than the
vague designation of essential amino acid.
Insects have evolved to deal with limiting resources, such as methionine, and have
successfully found effective strategies like hexamerin storage or alternate pathways to
deal with such problems. No attempt to link together all the aspects of the role of
methionine in a whole organism or system context. It appears that methionine actually
may play a role far more important than that of just an essential amino acid. From the

100
synthesis of homocysteine to produce methionine to the presence of methionine rich
hexamerins and allophorins and protein synthesis, the role of methionine in plant-insect
interactions may be larger than originally theorized.
The production of methionine overproducing plants could also be used in future
IPM strategies. Preliminary results indicate that genetically modified plants do produce
enough methionine to affect the survivorship of caterpillars feeding on the plant
(unpublished data). This could be used in crops in which improved nutritional quality is
important as well as the insecticidal properties of the additional methionine. However,
there appears to be a sublethal level (0.1%) of L-methionine in which THW and CPB can
tolerate and survive with little mortality (Figures 3-9 and 4-1). Any system that makes
use of a crop that can overproduce compounds like L-methionine would have to be able
to express levels greater than this level to avoid any resistance/tolerance.
This research has also provided more possibilities for the use of compounds such
as L-methionine in the YFM portion of this study. The amino acid Beta-alanine provided
similar levels of control, as did the methionine trials (Figure 5-7). Although unexpected
(as discussed in Chapter 5), it shows that there are several other systems that can possibly
be exploited in controlling some insects.
Further research is necessary to determine if the combination of a compound like
methionine and a pesticide already in use would result in the increase in toxicity or the
decrease in the concentration of pesticide used. If compatibility between methionine and
Bacillus thuringiensis does exists, then it is possible that resistance could be broken in a
given population. For example, if a population of THW started to show resistance to
Bacillus thuringiensis kurstaki then methionine could be used to remove both susceptible

101
and resistant alike because of the difference in mode of action. Once the population was
reduced, and the corresponding resistant genotype, Btk could be used once more at a
lower concentration, closer to that of the susceptible population. This system could also
be used for the reduction of Bt toxin resistance in the CPB and YFM if the compounds
are compatible.
In conclusion, it appears that L-methionine can be used as an insecticide to
control insect pests of economic and medical importance. The target site (CAATCH1) is
known and found in the midgut/hindgut (presumed) in at least three pest species (tobacco
homworm, Colorado potato beetle and the yellow fever mosquito) and possibly more.
The compound (L-methionine) is a safe compound that is already used for livestock feed
supplements, has very low mammalian toxicity, and is compatible with insecticide
application systems. Non-target organisms were not affected with the application of L-
methionine, further supporting its use as a biorational insecticide. With increasing
resistance to current insecticides in the study organisms, alternatives such as L-
methionine are needed now more than ever to further support of Integrated Pest
Management strategies.

LIST OF REFERENCES
Abbott, W.S. 1925. A method for computing the effectiveness of an insecticide. J.
Econ. Entomol. 18:265-267.
Aerts, M.J. and O.N. Neishiem. 1999. Florida Crop/Pest Management Profiles:
Tomatoes. CIR 1238. Pesticide Information Office. Food Science and Human
Nutrition Department, Florida Cooperative Extension Service, Institute of Food
and Agricultural Science, University of Florida. Internet URL:
http://www.edis.ifas.ufl.edu/ BODY_PI039.htm. Accessed April 2004.
Anand, R. and M. Anand. 1990. Nutritive effect of the D isomers of the essential amino
acids in casein diet on Dacus cucurbitae (Coquillett) maggots. Indian J. Entomol.
52(4): 525-528.
Andow, D.A., and S.J. Risch. 1985. Predation in diversified agroecosystems: Relations
between a coccinellid predator Coleomegilla maculata and its food. J. Appl.
Ecol. 22: 57-372.
Audsley, N., R.J. Weaver and J.P. Edwards. 1999. Juvenile hormone synthesis by
corpora allata of tomato moth, Lacanobia olercea (Lepidoptera: Noctuidae), and
the effects of allatostatins and allatotropin in vitro. Eur. J. Entomol. 96: 287-293.
Barfield, C.S. and M.E. Swisher. 1994. Integrated pest management: ready for export?
Historical context and internationalization of IPM. Food Reviews Intemat. 10(2):
215-267.
Baumhover, A.H., W.W. Cntelo, J.M. Hobgod, Jr., C.M. Knott, and J.J. Lam, Jr. 1977.
An improved method for mass rearing the tobacco horn worm. Agricultural
Research Service, Unites States Department of Agriculture ARS-S-167,13 pp.
Beck, S.D. and W. Hnec. 1958. Effects of amino acids on feeding behavior of the
European com borer, Pyraustra nubilialis (Hibn.). J. Insect Physiol. 2:85-96.
Bell, E.A. 1978. Toxins in seeds, pp. 143-161. IN J. Harboume (ed.), Biochemical
Aspects of Plant and Animal Coevolution. Academic Press, New York. 435pp.
Berge, M.A., G.A. Rosenthal and D.L. Dahlman. 1986. Tobacco budworm, Heliothis
virescens (Noctuidae) resistance to L-canavanine, a protective allelochemical.
Pest. Biochem and Physiol. 25: 319-326.
102

103
Bills, P.S., D. Mota-Sanchez and M. Whalon. 2004. The Database of Arthropods
Resistant to Pesticides. Michigan State University Center for Integrated Plant
Systems. Internet URL: http://www.cips.msu.edu/resistance/nndb/. Accessed
April 2004.
Boucher, T. J. 1999. Using IPM on CPB saves money, insecticides. Yankee Grower
1(2): 7-9.
Bourgis, F., S. Roje, M.L. Nuccio, D.B. Fisher, M.C. Tarczynski, C. Li, C. Herschbach,
H. Rennenberg, M.J. Pimenta, T. Shen, D.A. Gage and A.D. Hanson. 2000. S-
methymethionine has a major role in pholem, sulfer transport and is synthesized
by a novel methyltransferase. Pp. 283-284. IN C. Brunold (ed.), Sulfiir Nutrition
and Sulfur Assimilation in Higher Plants. Paul Haupt, Bern, Switzerland. 427pp.
Bownes, M. and L. Partridge. 1987. Transfer of molecules from ejaculate to females in
Drosophila melanogaster and Drospohila pseudoobscura. J. Insect Physiol.
33(12): 941-947.
Brogdon, W.G and J.C. McAllister. 1998. Insecticide resistance and vector management.
Emerging Infect. Diseases 4(4): 605-613.
Capinera, J.L, F.D. Bennett and D. Rosen. 1994. Introduction: Why biological control
and IPM are important to Florida, pp.3-8. In D. Rosen, F.D. Bennett and J.L.
Capinera (eds.), Pest Management in the Subtropics: Biological Control- a Florida
Perspective. Intercept Limited, Andover, UK. 737pp.
Center, T.D. 1994. Biological control of weeds, Chapter 23. pp.481-521. IN: D. Rosen,
F.D. Bennett, J.L. Capinera, (eds.), Pest Management in the Subtropics:
Biological Control-The Florida Experience. Intercept, Ltd., Andover, Hampshire,
UK. 737pp.
Center, T.D., F.A. Dray and V.V. Vandriver, Jr. 1998. Biocontrol with insects: The
water hyacinth weevils. Florida Cooperative Extension Service, Institute of Food
and Agricultural Sciences, University of Florida. Internet URL:
http://edis.ifas.ufl.edu/scripts/htmlgen.exe?body&DOCUMUMENT_AG01.
Accessed April 2004.
Centers for Disease Control (CDC). 2003. Malaria: General Information. Centers for
Disease Control. Internet URL: http://www.cdc.gov/travel/malinfo.htm. Accessed
April 2004.
Chen, P.S. 1958. Studies on the protein metabolism of Culexpipens L.-I. Metabolic
changes of free amino acids during larval and pupal development. J. Ins. Physiol.
2:38-51.
Cibula, A.B., R.H. Davidson, F.W. Fisk and J.B. LaPidus. 1967. Relationship of free
amino acids of some Solanaceous plants to growth and development of

104
Leptinotarsa decemlineata (Coleptera: Chrysomelidae). Annals Ent. Soc Am.
60(3): 626-631.
Cook, R.J., W.L. Bruckart, J.R. Coulson, M.S. Goettel, R.A. Humber, R.D. Lumsden,
J.V. Maddox, M.L. McManus, L. Moore, S.F. Meyer, P.C. Quimby, Jr., J.P.
Stack, and J.L. Vaughn. 1996. Safety of microorganisms intended for pest and
plant disease control: A framework for scientific evaluation. Biological Control
7: 333-351.
Dadd, R.H. 1975. Alkalinity within die midgut of mosquito larvae with alkaline-active
digestive enzymes. J. Insect Physiol. 21:1847-1853.
Dadd, R.H. and D.L. Krieger. 1968. Dietary amino acid requirements of the aphid,
Myzus persicae. J. Insect Physiol. 14: 741-774.
Dahlman, D.L. 1980. Field tests ofL-canavanine for control of tobacco homworm. J.
Econ. Entomol, 73:279-281.
Dahlman, D.L., F. Herald and F.W. Knapp. 1979. L-canavanine effects on growth and
development of four species of Muscidae. J. Econ. Entomol. 72: 678-679.
Dahlman, D.L. and G.A. Rosenthal. 1975. Non-protein amino acid-insect alterations (1)
Growth effects and symptomology of L-canavanine consumption by tobacco
homworm, Manduca sexta (L). Comp. Biochem Physiol. 51: 33-36.
Dahlman, D.L. and G A. Rosenthal. 1976. Further studies on the effect of L-canavanine
on the tobacco homworm, Manduca sexta. Insect Physiol. 22:265-271.
Dahlman, D.L. and G.A. Rosenthal. 1982. Potentiation of L-canavanine-induced
developmental anomalies in the tobacco homworm, Manduca sexta, by some
amino acids. J. Insect Physiol. 28(10): 829-833.
Deedat,Y.D. 1994. Problems associated with the use of pesticides: An overview. Insect
Sci. Applic. 15(3): 247-251.
Del Campo, M., and J.A.A. Renwick. 2000. Induction of host specificity in larvae of
Manduca sexta: Chemical dependence controlling host recognition and
developmental rate. Chemecol. 10: 115-121.
Dethier, V.G. and J.H. Kuch. 1971. Electrophysiological studies of gustation in
lepidopterous larvae 1. Comparative sensitivity to sugars, amino acids and
glycosides. Z. Vergl. Phys. 72: 343-363.
Dietary Supplement Information Bureau. 2000. Methionine Dietary Supplement
Education Alliance. Internet URL: http://www.supplementinfo.org/index.htm
Accessed April 2004.

105
Dimond, J.B., A.O. Lea and D.M. Delong. 1958. Nutritional requirements for
reproduction in insects. Proc. 10 Int Congr. Entomol. 2:135-137.
Droux, M, B.Gakire, L. Denis, S. Ravanel, L. Tabe, A.G. Lappartient, D. Job. 2000.
Methionine biosynthesis in plants: biochemical and regulatory aspects. Pp. 73-92.
/VBrunold, C., Rennenberg, H., De Kok, L.J., Stulen, L, Davidian, J.C. (eds.):
Sulfur Nutrition and Sulfur Assimilation in Higher Plants. Molecular,
Biochemical and Physiological Aspects. Paul Haupt Publishers. 447pp.
Durham, S. 2000. Hairy vetch thwarts Colorado potato beetle. Agricultural Research
Service, United States Department of Agriculture. Internet URL:
http://www.ars.usda.gov/is/pr/2000/000413.htm. Accessed April 2004.
Dwyer, J. 1999. Research Links 2000 Tobacco Homworm. Carleton College,
Department of Biology. Internet URL: http://www.acad.carieton.edu/curncular
/BIOL/resources/rlink. Accessed April 2004.
Ehler L.E. and D.G. Bottrell. 2000. The illusion of integrated pest management. Issues
in Science and Technology Online. National Academies and the University of
Texas (Dallas). Internet URL: http://www.nap.edU/issues/16.3/dder.htm.
Accessed April 2004.
Elliott, N.C., J.A. Webster, and S.D. Kindier. 1999. Developmental response of
Lysiphlebus testaceipes to temperature. Southwest Entomol. 24: 1-4.
Eymann, M. and W.G. Friend. 1985. Development of onion maggots (Dptera:
Anthomyiidae) on bacteria-free onion agar supplemented with vitamins and amino
acids. Ann. Entomol. Soc. Am. 78:182-185.
Feldman, D.H., W.R. Harvey and B.R. Stevens. 2000. A novel electrogenic amino acid
transporter is activated by K+ or Na+, is alkaline pH-dependent, and is CT-
independent. J. Biol. Chem. 275:24518-24526
Felton, G.W. and D.L. Dahlman. 1984. Allelochemical induces stress: Effects of L-
canavanine on the pathogenicity of Bacillus tkuringiensis in Manduca sexta. J.
Invert Path. 44: 187-191.
Ferro, D.N. 1985. Pest status and control strategies of the Colorado potato beetle. IN
Ferro, D.N. and R.H. Voss (eds.) Proceedings of the Symposium on the Colorado
potato beetle, XVII Inemational Congress of Entomology.
Fisher Scientific Internationa]. 2004. Online Catalog. Fisher Science International
Internet URL: https://wwwl.fishersci.com/index.jsp. Accessed April 2004.
Florida FIRST. 1999. Putting Florida FIRST: Focusing LFAS resources on solutions for
tomorrow. University of Florida, Institute of Food and Agricultural Sciences.
16pp.

106
Fogarty International Center and the U.S. National Institutes of Health (FIC-NIH). 2003.
Multilateral Initiative on Malaria. U.S. National Institutes of Health. Internet
URL: http://mim.nih.gov/english/index.html. Accessed April 2004.
Forgash, AJ. 1985. Insecticide resistance in the Colorado potato beetle, pp. 33-52. IN
D.N. Ferro and R.H. Voss (eds.) Proceedings of the Symposium on Colorado
Potato Beetle. XVII International Congress of Entomology, Massachusetts
Agricultural Experiment Station Bulletin 704. Amherst Massachusetts.
Friend, W.G., R.H. Backs and L.M. Cass. 1957. Studies on amino acid requirements of
larvae of the onion maggot, Hylema antiqua (MG.), under aseptic conditions.
Can. J. Zool. 35: 535-543.
Gasnier-Fauchet, F. and P. Nardon. 1986a. Comparison of sarcosine and methionine
sulfoxide levels in symbiotic and aposymbiotic larvae of two sibling species,
Sitophilus oryzae and Sitophilus zeamais (Coleptera: Curculionidae). Insect
Biochemistry 17(1): 17-20.
Gasnier-Fauchet, F. and P. Nardon. 1986b. Comparison of methionine metabolism in
symbiotic and aposymbiotic larvae of Sitophilus oryzae L. (Coloeptera:
Curculionidae)- n. Involvement of the symbiotic bacteria in the oxidation of
methionine. Comp. Biochem. Physiol. 58(1): 251-254.
Gauthier, V.L., R.N. Hoffinaster and M. Semel. 1981. History of Colorado potato beetle
control, pp. 13-34. IN J.H. Cashcomb and R. Casagrande (eds.), Advances in
Potato Pest Management. Hutchinson and Ross, Stroudsburg, PA. 672pp.
Geer, B.W. 1966. Utilization of D-amino acids for growth by Drosophila melanogaster
larvae. J. Nutr. 90: 31-39.
Giordana, B, M. Forcella, M.G. Leonardi, M. Casartelli, L. Fiandra, G.M. Hanozet and P.
Parenti. 2002. A novel regulatory mechanism for amino acid absorption in
lepidopteran larval midgut. J. Insect Physiol. 48: 585-592.
Giovanelli, J, S.H. Mudd and A.H. Datko. 1980. Sulfur amino acids in plants. IN: B.J.
Mifhn (ed.) The Biochemistry of Plants VoL 5, Academic Press, New York, pp.
453-505.
Giroux, S., R.M. Duchesne and D. Coderre. 1995. Predation of Leptinotarsa
decemlineata (Coleptera: Coccinellidae) by Coleomegilla maculata (Coleptera:
Coccinellidae): Comparative effectiveness of predator developmental stages and
effect of temperature. Environ. Entomol. 24: 748-754.
Glare, T.R, and M. O Callaghan, 1998. Environmental and Health Impacts of Bacillus
thuringiensis isrealensis. Report for the New Zealand Ministry of Health, 58pp.

107
Griffin, M.L. and K.V. Yeargan. 2002. Oviposition site selection by the spotted lady
beetle Coleomegilla maclala (Coleptera: Coccinellidae): Choices among plant
species. Environ. Entomol. 31(1): 107-111.
Groden, E., F.A. Drummond, R.A. Casagrande and D.H. Haynes. 1990. Coleomegilla
maculata (Coleptera: Coccinellidae): Its predation upon the Colorado potato
beetle (Coleptera: Chrysomelidae) and its incidence in potatoes and surrounding
crops. J. Econ. Entomol. 83:1306-1315.
Gubler, DJ. 1998. Resurgent vector-borne diseases as a global health problem. Emerg.
Infect. Dis. 4(3): 442-450.
Gubler, D.J. and G.G. Clark. 1995. Dengue/Dengue Hemorrhagic Fever: The emergence
of a global health problem. Emerg. Infect. Dis. 1(2): 55-57.
Haag, K.H. and D.G. Boucias. 1991. Infectivity of insect pathogens against Neochetina
eichhorniae, a biological control agent of water hyacinth. Florida Ent. 74(1): 128-
133.
Haag, K.H. and D.H. Habeck. 1991. Enhanced biological control of water hyacinth
following limited herbicide use. Aquat. Plant Management 29: 24-28.
Harrison, J. and R. Holliday. 1967. Senescence and the fidelity of protein synthesis in
Drosophila. Nature 214: 990-993.
Hazzard, R.V., D.N. Ferro, R.G. van Driesche and A.F. Tuttle. 1991. Mortality of eggs of
Colorado potato beetle (Coleptera: Chrysomelidae) from predation of
Coleomegilla maculata (Coleptera: Coccinellidae). Environ. Entomol. 20: 841
848.
Hegdekar, B.M. 1970. Amino acid analogues as inhibitors of insect reproduction. J.
Econ. Entomol. 63: 1950-1956.
Heim, D.C., G.G. Kennedy and J.W. Van Duyn. 1990. Survey of insecticide resistance
among North Carolina Colorado potato beetle (Coleptera: Chrysomelidae)
populations. J. Econ. Entomol. 83(4): 1229-1235.
Helena Chemicals, Inc. 2002. Silwett L-77. Technical Data Sheet SL77080596.
Helena Chemicals, Inc. Internet URL: http://www.helenachemicalwest.
com/data/TDS/Silwett.pdf. Accessed April 2004
Hilbeck, A. and G.G. Kennedy. 1996. Predators feeding on Colorado potato beetle
insecticide-free plots and insecticide-treated commercial potato fields in eastern
North Carolina. Biol. Control. 6: 273-282.
Hoffmann, M.P. and A.C. Frodsham. 1993. Natural Enemies of Vegetable Insect Pests.
Cooperative Extension, Cornell University, Ithaca, NY. 63pp.

108
Ito, T. and T. Inokuchi. 1981. Nutritive effects of D-amino acids on the silkworm,
Bombyx mor. J. Insect Physiol. 27(7): 447-453.
Jaffe, J.J. and L.R. Chrin. 1979. De novo synthesis of methionine in normal and Brugia-
infected Aedes aegypti. J. Parasitol. 65(4): 550-554.
Jones, D.C and R.M. MacPherson. 1997. Tobacco Insects: Summary of losses from
insect damage and costs of control in Georgia -1997. University of Georgia,
Waroell School of Forest Resources and College of Agricultural and
Environmental Sciences Internet URL: http://www.bugwood.org/tobbaco97.htm.
Accessed April 2004.
Kaldy, M.S. and A.M. Harper. 1979. Nutrient constituents of a grain aphid,
Metopolophium dirhodum (Homoptera: Aphididae), and its host, oats (Avena
sativa). Canadian Entomol. 111(7): 787-790.
Kammer, A.E., D. L. Dahlman and G.A. Rosenthal. 1978. Effects of the non-protein
aminoacids L-canavanine and L-canaline on the nervous system of the moth
Manduca sexta (L). J. Exp. Biol. 75: 123-132.
Kasting, R.. G.R.F. Davis amd A.J. McGinnis. 1962. Nutritionally essential and non-
essential amino acids for the prairie grain wireworm, Ctenicera desctructor
Brown, determined with Glucose-U-C. J. Insect Physiol. 8: 589-596.
Knutson, A., Boring IE, E.P., Michaels, Jr., G.J., and Gilstrap, F. 1993. Biological
Control of Insect Pests in Wheat Texas Agrie. Ext. Service Publ. B-5044 8pp.
Koo, S.I., T.A. Currin, M.G. Johnson, E.W. King and D.E. Turk. 1980. The nutritional
value and microbial content of dried face fly pupae (Musca autumnal is (DeGeer))
when fed to chicks. Poultry Sci. 59: 2514-2518.
Koyama, K. 1985. Nutritional physiology of the brown rice planthopper Nilaparvata
lugens StAl (Hemiptera: Delphacidae). II. Essential amino acids for nymphal
development Appl. Ent Zool. 20(4): 424-430.
Koyama, K. and J. Mitsuhashi. 1975. Essential amino acids for the growth of the smaller
brown planthopper, Laodelphax striatellus FallSn (Hemiptera: Delphacidae).
Appl. Ent Zool. 10(3): 208-215.
Lee, K., F.M. Horodyski, and M.E. Chamberlin. 1998. Inhibition of midgut ion transport
by allatotropin (Mas-AT) and Manduca FLRFamides in the tobacco homworm
Manduca sexta. J. Exp. Biol. 201:3067-3074.
Leonardi, M.G., M. Casartelli, L. Fiandra, P. Parenti and B. Giordana. 2001. Role of
specific activators of intestinal amino acid transport in Bombyx mor larval growth
and nutrition. Arch. Insect Biochem. Physiol. 48: 190-198.

109
Mallinckrodt Baker, Inc. 2001. L-methionine Material Safety Data Sheet MSDS
Number M2108. JT Baker Inc. Internet URL:httpV/www jtbakcr.com
/msds/englishhtml/M2108.htm. Accessed April 2004.
Marrone, P.G. and S.C. Macintosh. 1993. Resistance to Bacillus Ihuringiensis and
resistance management, pp. 221-236. IN, P.F. Entwistle, J.S. Cory, M.J. Bailey
and S. Higgs (eds.), An Environmental Biopesticide: Theory and Practice. John
Wiley and Sons New York. 330pp.
McPherson, R.M. and D.C. Jones. 2002. Tobacco Insects: Summary of losses from
insect damage and costs of control in Georgia-2001. University of Georgia
Integrated Pest Management. Internet URL: http://entomology.ent.uga.edu/IPM/
si 01/tobacco.htm. Accessed April 2004.
Melangeli, C., G.A. Rosenthal and D.L. Dahlman. 1997. The biochemical basis for L-
canavanine tolerance by the tobacco budworm Heliothis virescens (Noctuidae).
Proc. Natl. Acad. Sci. USA. 94: 2255-2260.
Mitchell, B.K. 1974. Behavioral and electrophysiological investigations on the
responses of larvae of the Colorado potato beetle (Leptinotarsa decent!ineata) to
amino acids. Ent. Exp. & Appl. 17:255-264.
Mitchell, B.K. and L.M. Schoonhoven. 1974. Taste receptors in Colorado potato beetle
larvae. J. Insect Physiol. 20:1787-1793.
Mittler, T.E. 1967a. Effect of amino acid and sugar concentrations on the food uptake of
the aphid Myzus persicae. Ent Exp. & Appl. 10: 39-51.
Mittler, T.E. 1967b. Gustation of dietary amino acids by the aphid Myzus persicae. Ent.
Exp. & Appl. 10: 87-96.
Munyaneza, J. and J.J. Obrycki. 1998. Development of three populations of
Coleomegilla maculata (Coleptera: Coccinellidae) feeding on eggs of Colorado
potato beetle (Coleptera: Chrysomelidae). Environ. Entomol. 27: 117-122.
Nakajima,N. S. Hiradate and Y. Fujii. 2001. Plant growth inhibitory activity of L-
canavanine and its mode of action. J. Chem. Ecol. 27(1): 19-31.
Nation, J. 2001. Insect Physiology and Biochemistry. CRC Press, Boca Raton. 496pp.
Neishman, O.N. and K. Vulinec. 2001. Florida Crop/Pest Management Profiles:
Eggplant CIR 1264. Pesticide Information Office, Food Science and Human
Nutrition Department, Florida Cooperative Extension Service, Institute of Food
and Agricultural Science, University of Florida. Internet URL:
http://www.edis.ifas.ufl.edu/ BODY_PI045Jitm. Accessed April 2004.

no
Nester, E.W., L.S. Thomashow, M. Metz and M. Gordon. 2002. One Hundred Years of
Bacillus thruingiensis: A Critica Scientific Assessment Report from the
American Academy of Microbiology. 16p.
Onifade, A.A., O.O. Oduguwa, A.O. Fanimo, A.O. Abu, T.O. Olutunde, A. Arije and G.
M. Babatunde. 2001. Effects of supplemental methionine and lysine on the
nutritional value of housefly larvae meal (Musca domestica) fed to rats. Biores.
Technol. 78:191-194.
Palumbo, R.E. and D.L. Dahlman. 1978. Reduction of Manduca sexta fecundity and
fertility by L-canavanine. J. Econ. Entomol, 71:674-676.
Pan, M.L. and W.H. Telfer. 1996. Methionine-rich hexamerin and arylphorin as
precursor reservoirs for reproduction and metamorphosis in female Luna moths.
Arch. Insect Biochem. Physiol. 32:149-162.
Patterson, K.D. 1992. Yellow fever epidemics and mortality in the United States, 1693-
1905. Soc. Sci. Med. 34(8): 855-865.
Perfect, T.J. 1992. IPM in 2000, pp.47-53. IN A.A.S.A. Kadir and H.S. Barlow (eds.),
Pest Management and the Environment in 2000. CAB International, Oxford, UK.
401pp.
Quick, M. and B.R. Stevens. 2001. Amino acid transporter CAATCH1 is also an amino
acid-gated cation channel. J. Bio. Chan. 276(36): 33143-33418.
Racioppi, J.V. and D.L. Dahlman. 1980. Effects of L-canavanine on Manduca sexta
(Sphingidae: Lepidoptera) larval hemolymph solutes. Comp. Biochem. Physiol.
67:35-39.
Ragsdale, D. and E.B. Radcliffe. 1999. Colorado potato beetle management. University
of Minnesota Cooperative Extension Service. Internet URL: http://ipmworld.umn.
edu/aphidalert/CPB~DWR.html. Accessed April 2004.
Robertson, J.L. and H.K. Priesler. 1991. Pesticide bioassays with arthropods. CRC
Press, Inc. Boca Raton, 127pp.
Rock, G.C. 1971. Utilization of D-isomers of the dietary, indispensable amino acids by
Argyrotaenia velutinana larvae. J. Insect Physiol. 17:2157-2168.
Rock, G.C. and E. Hodgson. 1971. Dietary amino requirements for Heliothis zea
determined by dietary deletion and radiometric techniques. J. Insect. Physiol. 17:
1087-1097.
Rock, G.C., BG. Lign, and E. Hogson. 1973. Utilization of methionine analoges by
Argyrotaenia velutinana larvae. Arm. Entol. Soc. Amer. 66(1): 177-179.

Ill
Rock, G.C., A. Khan, and E Hodgson. 1975. The nutritional value of seven D-amino
acids and a-keto acids for Argyrotaenia velutinana, Heliothis zea and Phormia
regina. J. Insect Physiol. 21:693-703.
Romoser, W.S. and J. G Stoffolano, Jr. 1998. The Science of Entomology, Sedition.
McGraw-Hill. Newyork, 605pp.
Rosen, D., F.D. Bennett, and J.L Capinera. 1996. Preface, pp. V-vi. IN D. Rosen, F.D.
Bennett and J.L. Capinaera (eds.), Pest Management in the Subtropics: Biological
Control- a Florida Perspective. Intercept Limited, Andover, UK. 737pp.
Rosenthal, G. A. 1977. The biological effects and mode of action of L-canavanine, a
structural analogue of L-arginine. Q. Rev. Biol. 52(2): 155-178.
Rosenthal, G.A. and D.L. Dahlman. 1975. Non-protein amino acid-insect interactions.
II. Effects of canaline-urea cycle amino acids growth and development of the
tobacco homworm, Manduca sexta (L.) (Sphingidae). Comp. Biochem. Physiol.
52:105-108.
Rosenthal, G.A. and D.L. Dahlman. 1988. Degradation of aberrant proteins by larval
tobacco homworm, Manduca sexta (L) (Sphingidae). Arch. Insect Biochem
Physiol. 8: 165-172.
Rosenthal, G. A. and D.L. Dahlman. 1991. Incorporation of L-canavanine into proteins
and the expression of its antimetabolic effects. J. Ag. and Food Chon. 39:987-
990.
Rosenthal, G.A., D.L. Dahlman, P.A. Crooks. S.N. Phuket, and L.S. Trifonov. 1995.
Insecticidal properties of some derivatives of L-canavanie. J. Agrie. Food Chem.
43:2728-2734.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1976. A novel means for dealing with
L-canavanine, a toxic metabolite. Science 192: 256-258.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1977. Degradation and detoxification
of canavanine by a specialized seed predator. Science 196:658-660.
Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1978. L-canaline detoxification: A
seed predators biochemical defense. Science 202: 528-529.
Rosenthal, G.A., P. Nkomo and D.L. Dahlman. 1998. Effect of long-chained esters on
the insecticidal properties of L-canavanine. J. Agrie. Food Chem. 46(1): 296-299.
Royer, T.A., K.L. Giles, S.D. Kindier and N.C. Elliott. 2001. Developmental response
of three geographic isolates of Lysiphlebus testaceipes (Hymenoptera: Aphididae)
to temperature. Environ. Entomol. 30(4): 637-641.

112
Schoonhoven, L.M. 1972. Plant recognition by lepidopterous larvae. Insect/Plant
Relationships, Symposium of the Roy. Entomol. Soc. London 6:83-93.
Schardt,J.D. 1987. 1987 Florida Aquatic Flora Survey Report. Florida. Department of
Natural Resources, Bureau of Aquatic Plant Management. Tallahassee, FL. 49 pp.
Schuster, D.J, J.E. Funderburk and P.A. Slansly. 1996. IPM in tomatoes, pp. 387-408.
IN D. Rosen, F.D. Bennett and J.L. Capinaera (eds.), Pest Management in the
Subtropics: Integrated Pest Management- A Florida Perspective. Intercept
Limited, Andover, UK. 737pp.
Singh, K.R.P. and A.W.A. Brown. 1957. Nutritional requirements of Aedes aegypti. J.
Insect Physiol. 1(1): 199-220.
Sorenson, C.E., R.L. Fery and G.G. Kennedy. 1989. Relationship between Colorado
potato beetle (Coleptera: Chrysomelidae) amd tobacco homworm (Lepidoptera::
Sphingidae) resistance in Lycopersicon hirsutum f. glagratum. J. Econ. Entomol.
82(4): 1743-1748.
Stansly, P.A., T.X. Liu, D.J. Schuster and D.E. Dean. 1996. Role of biorational
insecticides in management of Bemisia, pp. 605-615. IN D. Gerling and R. T.
Mayer, Jr. (eds.) Bemisia: 1995 Taxonomy, Biology, Damage Control and
Management. Andover, Hants, UKD. 702pp.
Stevens, B.R., D.H. Feldman, Z. Liu and W.R. Harvey. 2002. Conserved tyrosine-147
plays a critical role in the ligand-gated current of the epithelial cation/anion acid
transporter/channel CAATCH1. J. Exp. Biol. 205: 2545-2553.
Sugarman, D. and W. Jakinovitch, Jr. 1986. Behavioral gustatory responses of adult
cockroaches, Periplaneta americana to D and L amino acids. J. Insect Physiol.
32(1): 35-41.
Tobe, S.S. and N. Clarke. 1985. The effect of L-methionine concentration on juvenile
hormone biosynthesis by corpora allata of the cockroach Diploptera punctata.
Insect Biochem. 15(2): 175-179.
Tipping, P.W., C.A. Holko, A.A. Abdul-Baki, and J.R. Aldrich. 1999. Evaluating
Edovum puttleri Grissell and Podiscus maculiventris (Say) for augmentative
biological control of Colorado potato beetle in tomatoes. Biol. Control 16: 35-42.
Tzeng, D.D. 1988. Photodynamic action of methionine-riboflavin mixture and its
application in the control of plant diseases and other plant pests. Plant Protection
Bulletin 30(2): 87-100.
Tzeng, D.D. and J.E. Devay. 1989. Biocidal activity of mixtures of riboflavin against
plant pathogenic fungi and bacteria and possible modes of action. Mycologia
81(3): 404-412.

113
Tzeng, D.D., M.H. Lee, K.R. Chung and J.E. Devay. 1990. Products in light-mediated
reactions of free methionine-riboflavin mixtures that are biocidal to
microorganisms. Can. J. Microbiol. 36(7): 500-506.
Wadsworth, D.J. 1995. Animal health products, pp. 257-284. IN C.R.A. Godfrey (ed.)
Agrochemicals From Natural Products, Marcel Dekker, New York. 424pp.
Walker, T.J., J.J. Gaffney, A.W. Kidder and A.B. Ziffer. 1993. Florida Reach-Ins:
Environmental chambers for entomological research. American Entomologist
39(3): 177-182.
Weinzierl, R., T. Henn and P.G. Koehler. 1998. Microbial insecticides. ENY-275,
Cooperative Extension Service, Institute of Food and Agricultural Services,
University of Florida. 13pp.
Wheeler, D.E., I. Tuchinskaya, N.A. Buck and B.E. Tabahnik. 2000. Hexameric storage
proteins during metamorphosis and egg production in the diamondback moth,
Plutella xylostella (Lepidoptera). J. Insect. Physiol. 46: 951-958.
Womack, M. 1993. The yellow fever mosquito, Aedes aegypti. Wing Beats 5(4): 4.
Wright, R. 1995. Know Your Friends: Wasp Parasites of Greenbugs. Midwest
Biological Control News Online, 11:9.
Young, V.R. and A.E. El-Khoury. 1996. Human amino acid requirements: A re-
evaluation. Food and Nutrition Bulletin 17(3): 191-203.
Zar, J.H. 1999. Biostatistical Analysis, 4th ed. Prentice Hall. New Jersey. 663pp.
Zeh, M., A.P. Casazza, O. Kreft, U. Roessner, K. Bieberich, L. Willmitzer, R. Hoefgen
and H. Hesse. 2001. Antisense inhibition of threonine synthase leads to high
methionine content in transgenic potato plants. Plant Physiol. 127: 792-802.

BIOGRAPHICAL SKETCH
Lewis Scotty Long was bom in Calhoun, Georgia on August 20,1971. He
graduated from Madisonville High School (Madisonville, Tennessee) in May 1989. On a
biology scholarship, Lewis attended Middle Tennessee State University (MTSU), where
he earned his BS in May 1994. On graduation, he took a job as an aquatic biologist for
Aquatic Resources Center (Franklin, Tennessee). Lewis worked there specializing in
taxonomy of mayflies, stoneflies, caddisflies, and freshwater molluscs (snails and
mussels). While still employed at Aquatic Resources Center, he started his graduate
studies in 1996 at MTSU and continued the work he had started during his undergraduate
years. In May of 1999, Lewis graduated with his MS. After receiving his MS, Lewis
moved to Florida and entered the PhD program at the University of Florida, Department
of Entomology and Nematology. He worked with Dr. Bill Peters (Florida A&M
University) on the worldwide taxonomic revision of an understudied group of mayflies.
However, Dr. Peters unexpectedly passed away in 2000, and Lewis took this unfortunate
event as a chance to broaden his expertise in entomology. In 2000, he took a part-time
job with Drs. James Cuda and Bruce Stevens on research that was in the patent process.
This was the research that Lewis undertook for his dissertation. Lewis also served as a
teaching assistant for the department for classes such as Bugs and People, Life Sciences
for Education Majors, Principles of Entomology, and Medical and Veterinary
Entomology. He served as primary instructor for Insect Classification and Immature
114

115
Insects. Lewis, along with fellow graduate student Jim Dunford, were awarded the
Outstanding Teacher Award by the Entomology and Nematology Student Organization of
the University of Florida for outstanding teaching accomplishments in the department.
While at the University of Florida, Lewis joined the U.S. Army Reserve as a
medical entomologist. He was assigned to the local Medical Detachments, and served
there from 2000 to 2004. Originally he had planned on graduating in 2003, but was
called to active duty with the 1469th Medical Detachment as a part of Operation Enduring
Freedom (OEF). Lewis was the OEF Theater entomologist, and served as the Executive
Officer (responsible for the deployment of personnel and equipment to South West Asia).
He was stationed at Kandahar Airfield, where he performed his duty and was awarded an
Army Commendation Medal for his work in protecting soldiers from health hazards and
diseases associated with the area. Lewis returned and continued his work toward
graduation.
Lewis was married in August 1992 to Karen Abbott, and is the father of Emilia
Irene (1994) and Bryan Scott (1997). Lewis plans on having a career in the military as a
medical entomologist, and all look forward to seeing the world and the rest of their
future.

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in sqqpe and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jamefe P. Cuda, Chair
Assistant Professor of Entomology and
Nematology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Bruce R. Stevens, Cochair
Professor Physiology and Functional
Genomics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Professor of Physiology and Functional
Genomics
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Jane^ E. Maruniak
Associate Professor of Entomology and
Nematology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Simon S.J. Yu
Professor of Entomology and
Nematology

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Entomology and
Nematology
This dissertation was submitted to the Graduate Faculty of the College of
Agricultural and Life Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
May 2004
Dean, College of Agricultural
Sciences
Dean, Graduate School

IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE
USING TIME INTENSITY GC-0 AND GC-MS
By
KANJANA MAHATTANATAWEE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004

Copyright 2004
by
Kanjana Mahattanatawee

This dissertation is dedicated to my late father.

ACKNOWLEDGMENTS
I would especially like to thank to Dr. Russell L. Rouseff, my supervisor, for his
friendship and guidance throughout my graduate program. He provided me with
intellectual, thoughtful discussion; encouragement; and time. Without his help and
support I would not have come from Thailand and continued at the University of Florida.
His high ethical standards and philosophical views will never be forgotten. I would like
to thank Dr. R.M. Goodrich for all of her support, guidance and friendship. She was very
kind and helpful whenever I needed her help. I also thank the other members of my
supervisory committee, Dr. D.H. Powell and Dr. M.R. Marshall, Jr, for their kindness and
valuable, thoughtful discussion toward my research. Especially warm thanks go to Dr.
Kevin L. Goodner for his advice and support on instrumental. My special thanks go to
John Smoot and Dr. Filomina Valim who assisted me any time I needed them. My
appreciation goes to my dear friend Dr. Fahiem El-Borai Kora who assisted me whenever
I needed his help. I extend my appreciation to all my friends at Citrus Research and
Education Center (CREC) in Lake Alfred for their friendship and support. I would like
also to thank my professors back in Thailand expecially Dr. Twee Hormchong, who
taught me what good scientists and teachers are. My great thanks go to my beloved
family. My lovely wonderful mother, father, brothers and sisters always encouraged me
to follow my dream, and without their love and support, I could never be who I am today.
Finally, my appreciation goes to the Siam University for giving me the financial support
necessary to obtain my Ph.D from the University of Florida.
IV

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iv
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xii
CHAPTER
1 INTRODUCTION 1
2 LITERATURE REVIEW 3
Orange Juice Aroma 3
Carotenoids 3
Norisoprenoids 4
Norisoprenoids Formation from Carotenoids 5
Apple 7
Tomato 8
Saffron 9
Grape and Wine 10
Gas Chromatography-Olfactometry 11
Solid Phase Microextraction 13
Orange Juice Norisoprenoids 14
3 HPLC DETERMINATION OF CAROTENOID NORISOPRENOID
PRECURSORS IN ORANGE JUICE 15
Introduction 15
Objectives 17
Materials and Methods 17
Carotenoid Extraction 17
Carotenoid Saponification 17
HPLC Procedure 18
Results and Discussion 19
Carotenoids of Interest 19
Hydrolysis Conditions 20
v

HPLC Separation
Carotenoid Identification 23
Conclusions 25
4 IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING
TIME INTENSITY GC-0 AND GC-MS 26
Introduction 26
Objectives 28
Materials and Methods 28
Orange Juice Samples and Processing 28
Chemicals 29
Orange Juice Headspace Extraction 29
Gas Chromatography: GC-FID and GC-Olfactometer 30
Gas Chromatography-Mass Spectrometry 31
Aroma Peak Identification 32
Results and Discussion 32
Extraction and Concentration of Juice Norisoprenoids 32
GC-Olfactometry 35
Mass Spectrometry Norisoprenoid Identifications 36
Conclusion 41
5 QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT
OF NORISOPRENOIDS IN ORANGE JUICE 43
Introduction 43
Objectives 45
Materials and Methods 45
Quantification of Norisoprenoids in Orange Juice 45
Determination of the Relative Impact of Norisoprenoids in Orange Juice 47
Results and Discussion 47
Quantification of Norisoprenoids in Orange Juice 47
Norisoprenoid Quantification using Standard Additions 50
Determination of Relative Aroma Impact of Norisoprenoids 54
Norisoprenoid Contribution to Total Floral Aroma 58
Conclusion 61
6 THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL
SOLUTION 62
Introduction 62
Objective 63
Materials and Methods 63
Crystallization 63
Model Solutions 64
Analytical Methods 64
Results and discussion 65
vi

GC-0 Analysis of (3-Carotene Decomposition at 35C 68
MS Identification 70
Conclusion 72
7 CONCLUSIONS 73
LIST OF REFERENCES 75
BIOGRAPHICAL SKETCH 84
vii

LIST OF TABLES
Table
3-1 HPLC retention times, spectral characteristics of orange juice carotenoids 24
4-1 Identification, retention characteristics and aroma descriptions of aroma active
compounds in fresh orange juice 36
5-1 Reproducibility of SPME exposure time 45 min at 40C 49
5-2 Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique 53
5-3 Concentration of (3-damascenone in fresh, pasteurized and reconstituted
concentrate 53
5-4 Aroma active compounds in orange juice grouped by citrusy/minty 55
5-5 Aroma active compounds in orange juice grouped by metallic/mushroom/
geranium 56
5-6 Aroma active compounds in orange juice grouped by roasted/cooked/meaty/
spice 56
5-7 Aroma active compounds in orange juice grouped by fatty/soapy/green 56
5-8 Aroma active compounds in orange juice grouped by sulfury/solventy/medicine ..57
5-9 Aroma active compounds in orange juice grouped by floral 57
5-10 Aroma active compounds in orange juice grouped by sweet/fruity 57
5-11 Aroma active compounds in orange juice grouped by green/grassy 58
5-12 Norisoprenoids in orange juice and peel oil 60
6-1 Aroma active compounds from (3-carotene thermal degradation in model solution
pH 3.8, storage at 35C for 2 weeks 69
viii

LIST OF FIGURES
Figure page
2-1 Examples of carotene and xanthophyll carotenoid structures 4
2-2 Major fragment classes of carotenoid biodegradation 5
2-3 General steps for the conversion of carotenoids into flavor compounds, showing
the formation of P-ionone and P-damascenone from P-carotene and neoxanthin
respectively 6
2-4 Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters) 8
2-5 Stevens law, comparing two difference compounds: A= compound A, B=
compound B 12
3-1 Possible degradation pathways for the formation of P-cyclocitral and P-ionone
from P-carotene 15
3-2 Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of P-damascenone 19
3-3 Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH, MTBE
ternary solvent gradient 22
3-4 Absorbance spectra for P-carotene (a), leutoxanthin (b), and neoxanthin (c), peak
24, 12 and 4 respectively 23
4-1 GC-FID (top) and average time-intensity of four GC-0 runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and 23
correspond to norisoprenoids, all numbers refers to compounds in Table 4-1 34
4-2 Comparison between total ion chromatogram and selected ion chromatograms
(SIC) A: P-cyclocitral, B: P-ionone, C: a-ionone 37
4-3 Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190 38
IX

4-4 Upper spectra from orange juice MS at RT = 17.68 bottom spectra of (3-
cyclocitral from database NIST 2002 40
4-5 Upper spectra from orange juice MS at RT = 21.94, bottom spectra from
standard (3-ionone using identical ion trap MS at identical retention time 40
4-6 Upper spectra from orange juice MS at RT = 20.87, bottom spectra from
standard a-ionone using identical ion trap MS at identical retention time 41
5-1 Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 40C, = (3-cyclocitral, = p-damascenone, A= a-ionone,
= p-ionone 48
5-2 Standard addition data for [3-cyclocitral peak area vs. added concentration in fresh
orange juice. Regression line calculated from peak area at selected mass 137 50
5-3 Standard addition data for a-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
a-ionone concentration 51
5-4 Standard addition data for (3-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
(3-ionone concentration 52
5-5 Standard addition (3-damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190 52
5-6 Standard addition data of (3-damascenone peak area vs. added concentration in
reconstituted from concentrate orange juice 53
5-7 Aroma group profiles of fresh (), pasteurized (), and reconstituted from
concentrate () orange juice 58
5-8 Upper bar norisoprenoids contribute mainly to the total floral category, fresh =
78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent non-
norisoprenoids including linalool and unknown (LRI = 1255) 60
6-1 The standard (3-carotene (99% purity) as received (no purification) 65
6-2 Headspace volatiles from P-carotene in model solution pH 3.5 at 0 day 66
6-3 Headspace volatiles from P-carotene in model solution pH 3.5 after storage 1 day
at 35C: 1 = P-ionone, a = sweet/raspberry 66
6-4 Degradation of P-carotene in model solution at difference carbon bonds 67
x

6-5 Headspace volatiles from (3-carotene in model solution pH 3.5, after storage 2
weeks at 35C 69
6-6 Selected ion chromatogram (SIC) of model solution headspace volatiles after
storage 2 weeks at 35C 70
6-7 Upper spectra from model solution MS at RT 19.61, bottom spectra from
standard (3-cyclocitral using identical ion trap MS at identical retention time 70
6-8 Upper spectra from model solution MS at RT 20.46, bottom spectra from
standard P-homocyclocitral using identical ion trap MS at identical retention
time 71
6-9 Upper spectra from model solution MS at RT 25.13, bottom spectra from standard
P-ionone using identical ion trap MS at identical retention time 71
xi

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
IDENTIFICATION OF NEW CITRUS NORISOPRENOIDS IN ORANGE JUICE
USING TIME INTENSITY GC-0 AND GC-MS
By
Kanjana Mahattanatawee
May 2004
Chair: Russell L. Rouseff
Major Department: Food Science and Human Nutrition
Numerous analytical studies have quantified the major volatiles in orange juice in
an effort to duplicate this aroma. However, when combined, the resulting aroma does not
duplicate that of orange juice, suggesting that important aroma components were missing.
Citrus carotenoids have been studied primarily for their role as pigments and have
generally been ignored as a source of aroma compounds. Carotenoids can be degraded
into smaller (C9-C13), more volatile products called norisoprenoids. Norisoprenoids have
been identified as aroma impact in foods containing carotenoids (i.e. tea, grapes, tomato,
and saffron). Therefore, carotenoid-decomposition may be responsible for a portion of
orange juice aroma and also aroma changes associated with thermal processing or
elevated temperature storage.
Three norisoprenoids, (3-cyclocitral, (3-damascenone, a-ionone, were fully
identified in fresh and pasteurized orange juice for the first time. Beta-ionone was also
detected but had been previously identified. Identification was based upon SPME
xii

headspace, GC-O, and GC-MS data. Only two norisoprenoids (p-damascenone and
P-ionone) were detected in reconstituted juice that had been thermally concentrated. Peel
oil from the same fruit contained only P-damascenone and P-cyclocitral.
Concentrations of P-cyclocitral, P-damascenone, a-ionone, and P-ionone were
determined using standard addition SPME and GC-MS and found to be 145, 0.09, 47 and
83 /rg/L respectively. The concentration of P-damascenone increased from 0.09 to 0.85
fig/L after thermal concentration and reconstitution. Orange juice norisoprenoids
contribute approximately 8-10% of total aroma intensity as determined from combined
aromagram peak heights and 60-80% of the total floral-category.
Known norisoprenoids precursors (P-carotene, a-carotene, a-cryptoxanthin,
P-cryptoxanthin, and neoxanthin) were identified in Valencia orange juice using C30
reverse phase HPLC with photodiode array detection.
Thermal decomposition products of P-carotene in citric acid solutions buffered at
pH 3.8 were examined during 35C storage using GC-0 and GC-MS. Beta-cyclocitral,
p-homocyclocitral, P-damascone and P-ionone were detected after 2 weeks thus
demonstrating that P-carotene can produce norisoprenoids. Since half of the a-carotene,
a-cryptoxanthin, P-cryptoxanthin structures share the identical structure as P-carotene,
these carotenoids must be considered potential norisoprenoid sources as well.
xiii

CHAPTER 1
INTRODUCTION
The delicate aroma of fresh orange juice is the result of a complex mixture of
volatiles blended in specific proportions. Numerous analytical studies (1-5) have
identified and quantified the major volatiles in orange juice in an effort to duplicate this
aroma. However, when combined, the identified volatiles could not duplicate orange
juice aroma, suggesting that important aroma components were missing. Early orange
juice gaschromatography olfactometry (GC-O) studies (4, 6, 7) have shown that many of
the aroma-active compounds in orange juice exist as low-level volatiles that are difficult
to detect using typical flame ionization detector (FID) or mass spectrometer (MS)
detectors. Furthermore, these studies demonstrated that the major volatiles in orange
juice have little to no aroma activity. Recent orange juice GC-0 studies (5) quantified
the 25 most intense aroma-active compounds in fresh juice, using isotope dilution
analysis. Model solutions of the aroma components in orange juice based on GC-0
studies have come closer to duplicating the aroma of fresh orange juice than model
systems based on the composition of the volatiles found in highest concentration.
Carotenoids are too large (C40) to be volatile under normal conditions. Because
they contain a highly conjugated double bond structure, they can be degraded by enzyme,
chemical, and/or thermal reactions to form a wide range of structures, depending on
which double bond is broken. Some of their smaller (C9-C13), volatile, decomposition
products are called norisoprenoids. Norisoprenoids have been shown to have significant
aroma impact in fruits, vegetables and spices such as grapes, apples, lychee, starfruit,
1

2
mango, tomato, saffron, cured tobacco, and black tea (8-16). Only a single
norisoprenoid, (3-ionone, has been identified to date in fresh orange juice (4, 5)
More than 50 carotenoids have been separated and identified from the juice of
three varieties of Citrus sinensis (Shamouti, Valencia, and Washington Navel) using
column chromatography combined with thin layer chromatography (TLC) (17). Some of
these carotenoids (such as (3-carotene, a-carotene, neoxanthin, (3-crytoxanthin, lutein,
violaxanthin, and canthaxanthine) have the structural potential to form potent
norisoprenoid fragments (18-22). Furthermore, (3-carotene in tomato products has been
shown to produce (3-ionone and (3-cyclocitral (23). Beta-ionone and a-ionone have been
generated from (3-carotene and a-carotene respectively in carrots (24). Neoxanthin in
grapes has been shown to be a source of (3-damascenone (25). Prior orange juice
carotenoid studies were primarily directed toward the contribution of carotenoids to juice
color and for vitamin A content. They have been generally ignored as precursors of
aroma compounds. Since orange juice has so many carotenoids that could serve as
precursors for a wide range of norisopemoids, the objectives of this research were to:
1. Confirm the presence of possible carotenoid norisoprenoid precursors in orange juice
using HPLC and photodiode array detection. (Chapter 3)
2. Determine if additional norisoprenoid are present in orange juice. Characterize and
identify these new norisoprenoids. (Chapter 4)
3. Determine the relative aroma impact of carotenoid degradation products
(norisoprenoids) to the total aroma impact of orange juice in fresh, pasteurized, and
reconstituted from concentrate juice. (Chapter 5)
4. Develop quantitative procedures to isolate and quantify orange juice norisoprenoids
using static headspace SPME with GC-MS. (Chapter 5)
5. Determine if (3-carotene can form norisoprenoid degradation products at 35C storage
in model solutions. (Chapter 6).

CHAPTER 2
LITERATURE REVIEW
Orange Juice Aroma
The aroma of fresh orange juice is composed of a complex mixture of aldehydes,
esters, ketones, alcohols and terpenes blended in specific proportions. Numerous studies
(1-5) aimed at identifying the flavor volatiles in orange juice have led to the identification
of about 200 volatiles, but no single aroma character impact for orange flavor has ever
been reported. GC-olfactometry and orange juice volatile quantification have been used
to gain a more accurate understanding of their contribution to orange flavor (2, 5-7).
Early orange juice GC-0 studies have demonstrated that the orange juice volatiles present
in highest concentration have little to no aroma activity and many aroma active
compounds exist as low-level volatiles that are difficult to detect using typical
instrumental detectors.
Carotenoids
Carotenoids are primarily responsible for the colors of many plants, birds, and
insects; but also serve as plant photoprotection agents during photosynthesis, and as
essential human nutrients. However, the least-appreciated role of carotenoids is their
function as aroma precursors.
Carotenoids are tetraterpenes (C4o ) resulting from the joining together of eight molecules
of isoprene (C5) through tail-to-tail condensation. Most carotenoids have a C40 carbon
skeleton. The ends may or may not be cyclized into six membered rings. If the ends are
not cyclized, the molecule is termed acyclic. There are two main groups of carotenoids:
3

4
the hydrocarbon group, which contain only carbon and hydrogen; and the xanthophyll
group, which contain carbon, hydrogen, and oxygen. Oxygen in xanthophylls is usually
found as either hydroxyl-(monols, diols and polyols), epoxy- (5,6 and 5,8-epoxides),
methoxy, aldehyde, oxo, carboxy and/or esters. Hydroxyl substitution primarily occurs at
the C3 position in the ionone ring; and a carbonyl substitution usually occurs at the C4
position in the (3-ionone ring. In most of the cyclic carotenoids, the 5,6- and 5,6-double
bonds are the most susceptible to epoxidation. The unconjugated double bond in the a-
ionone 8 ring does not undergo epoxidation. Allenic carotenoids have a C=C=C
grouping at one end of the central chain, and acetylenic carotenoids have a -C=C- bond
in position 7,8 and/or 7,8 (26, 27). Figure 2-1 shows acyclic carotene (lycopene),
bicyclic carotene (P-carotene), the monol P-cryptoxanthin, and the diol zeaxanthin.
Figure 2-1. Examples of carotene and xanthophyll carotenoid structures.
Norisoprenoids
Norisoprenoids are volatile C9-CI3 fragments from the degradation of the C40
carotenoids. The formation of norisoprenoids from carotenoids is thought to proceed via

5
enzymatic and nonenzymatic pathways. Nonenzymatic cleavage reactions include photo
oxygenation (18), auto-oxidation (28, 29), and the thermal degradation processes (30,
31). The in vivo cleavage of the carotenoid chain is generally considered to be catalyzed
by dioxygenase (lipoxidase and peroxidase) systems and require molecular oxygen and
other cofactors for activity. The polyene chain of carotenes is readily oxidized, giving
rise to cyclic and acyclic compounds (often having an oxygen-containing functional
group on a trimethylcyclohexane ring, or an oxygen-containing functional group on the
allyllic side chain). Although all the in-chain double bonds seem to be vulnerable to
enzymatic attack, in actuality the formation of major fragment classes with 10, 13, 15 or
20 carbon atoms are most common (see Fig. 2-2). In fruit tissues, the bio-oxidative
cleavage of the 9,10 (9, 10) double bond seems to be the most preferred (15, 32, 33).
Figure 2-2. Major fragment classes of carotenoid biodegradation.
Norisoprenoids Formation from Carotenoids
Norisoprenoids can be generated from carotenoids via either direct cleavage or
cleavage followed by subsequent reactions. In the latter process, three steps are required
to generate an aroma compound from the parent carotenoid: 1) the initial dioxygenase
cleavage, 2) subsequent enzymatic transformations of the initial cleavage product giving

6
rise to polar intermediates (aroma precursors), and 3) acid-catalyzed conversions of the
nonvolatile precursors into the aroma active form (32). One example illustrating these
reaction is the formation of p-damascenone from neoxanthin (Fig. 2-3). The primary
oxidative cleavage product of neoxanthin, grasshopper ketone, must be enzymatically
reduced before finally being acid-catalyzed converted into the odoriferous ketone. In the
direct process, the target compound is immediately obtained after the initial cleavage
(i.e., formation of a- and P-ionone directly from a- and (3- carotene) (34).
P-carotene
P-ionone
Carotenoid
neoxanthin
Step 1
oxidative cleavage
Primary cleavage
product
Step 2
Enzymatic
transformation
Non-volatile metabolite
(aroma precursor)
Step 3
Acid catalyzed
conversions
Aroma compound
Figure 2-3. General steps for the conversion of carotenoids into flavor compounds,
showing the formation of P-ionone and P-damascenone from P-carotene and
neoxanthin respectively (Winterhalter, P., Rouseff, R. Carotenoids-Derived
Aroma Compounds: An Introduction. In Carotenoid-derived Aroma
Compounds; P. Winterhalter and R. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002, Fig. 4, page 12).
Recent studies have shown that some of the volatile Cn-compounds are not free,
uncomplexed plant constituents; but rather are derived from less or nonvolatile precursors
such as polyols, glycosides, and glucose esters. Carotenoid degradation is initiated by

7
oxidative cleavage of the intact carotenoid. After further enzymatic transformation steps,
the primary cleavage products are converted into reactive Cio to C13 fragments of the
initial carotenoid. These volatile fragments can be stabilized and made nonvolatile by
glycosylation (which involves glycosyltransferases of those norisoprenoid compounds
possessing a hydroxyl group) (35, 36).
Glycosilation stabilizes and solubilizes norisoprenoids in plant systems.
Degradation of the glycoconjugates librates the potent volatile and can produce profound
aroma changes. This process can be acid-catalyzed (e.g., during fruit processing) (9) or
enzymatic (e.g., during fermentation) (37). Another important class of precursors is the
polyols, which upon (allylic) elimination of water is transformed into volatile forms. An
example is the reactive allyl-l,6-diol that, under gentle reaction conditions (natural pH,
room temperature), is converted into isomeric theaspiranes (38). A third class of
carotenoid-derived aroma precursor is glucose esters (e.g., Cio-compounds derived from
the central part of the carotenoid chain, which is left after the cleavage of the endgroups)
that gives rise to isomeric marmelo lactones, key aroma constituents of quince fruit (see
Fig. 2-4) (39).
Norisoprenoids have been shown to have significant aroma impact in fruits (apple,
mango, grape, starfruit, lychee, passion fruit, nectarine, etc.) (9-12, 19, 36, 40) vegetables
(tomato (41)), spices (saffron (14) and paprika (42)), leaf products (tobacco (43) and tea
(44)) as well as roses (45), wine (46) and oak wood (47).
Apple
Beta-Damascenone is a potent aroma compound found in a variety of natural
products, with a threshold of 0.002 pg/L in water (48). Eight separate (3-damascenone

8
precursors have been detected in apples (Malus domestica Borkh. cv. Empire) (49). The
most abundant precursor, present at 4.6 ng/g, was the 9 (or 3) -a-L-arabinofuranosyl-
(l,6)-P-D-glucopyranoside of the acetylenic diol. The second most abundant precursor,
present at 3.1 ng/g, is a more polar glycoside of the acetylenic diol. (49). Beta-
Damascenone contributed 32% of the total aroma potency of heated apple juice, but only
1.6% of the total aroma of fresh apple juice as determined by GC-O. Thus, most of the
P-damascenone in heated apple juice was generated from nonvolatile precursors during
thermal processing (9).
OH
Allyl-1,6-diol Theaspiranes
OH
P-D-Gentiobioside of 3-Hydroxy-P-ionol
Marmelo Lactone
Figure 2-4. Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters).
Tomato
One of the most marked differences between the fresh tomato and the paste is the
almost complete loss of the major contributor to fresh tomato aroma, (Z)-3-hexenal. The
most notable increase is with the potent odorant, P-damascenone, which shows a 10 fold

9
increase in concentration in the paste (50). Beta-ionone in tomatoes seems to be formed
mainly by an oxidative mechanism. It was not detected among the glycoside hydrolysis
products. The compound P-damascenone was shown to be produced in fruits from
hydrolysis of glycosides via an intermediate acetylenic compound megastigm-5-en-7-
yne-3,9-diol. This appears to be the final step in tomato volatile norisoprenoid formation
(51). Three experimental lines of tomato: a high-P-carotene line; a high-lycopene line;
and a low-carotenoid line were examined for their norisoprenoid content. In fresh
tomato, the high (3-carotene line produced the highest concentrations of P-ionone (17
pg/L, versus 1 pg/L in the low-carotenoid line) and P-cyclociral produced 30 pg/L in the
high carotene line versus 0 pg/L in the low-carotene line). Both norisoprenoids are
known biological or chemical degradation products of P-carotene. The high lycopene
line, however, did not show any significantly higher concentration of the expected
lycopene degradation products, 6-methyl-5-hepten-2-one, 6-methyl-5-hepten-2-ol, or
geranylacetone. It did show a significantly higher value for geranial (21 pg/L) compared
to that of the common line (12 pg/L). Geranial could be considered a lycopene-
degradation product (41).
Saffron
Safranal (monoterpene aldehyde, C10H14O) is the characteristic impact compound
of saffron (dried stigmas of Crocus sativus), formed in saffron during drying and storage
by hydrolysis of picrocrocin. Picrocrocin was the colorless glycoside of the aglycone,
4-hydroxy-2,6,6-trimethyl-l-carboxaldehyde-l-cyclohexene (HTCC), which was the
main substance responsible for the bitter taste of saffron. Safranal was not present in the
fresh stigma. Its concentration in saffron depended strongly on both the drying and

10
storage conditions. Additional flavor compounds in saffron were formed upon cooking
of the spice (52). Aroma isolates of saffron have been prepared by simultaneous
distillation extraction (SDE) at pH 2.6 as well as liquid-liquid extraction using pentane:
diethyl ether (1:1) as solvent. Aroma activity and relative aroma strength was determined
using aroma extract dilution analysis (AEDA). Compounds with high FD-factors were
safranal and 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one as well as linalool and
isophorone. The 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one was only detected
in the SDE isolate and not in the liquid-liquid extract. This result shown the presence of
certain forms of precursors, which upon heat treatment are converted into the aroma
compound 2-hydroxy-4,4,6-trimethyl-2,5-cyclohexadien-l-one (55).
Grape and Wine
Norisoprenoids are important aroma constituents of grape and wine. They are
thought to arise from carotenoid breakdown; and occur in grapes as glycosidically bound
precursors. The major carotenoids in grape are P-carotene and lutein, representing nearly
85% of the total carotenoids. These are accompanied by minor carotenoids such as
neoxanthin, violaxanthin, lutein-5,6-epoxide, zeaxanthin, neochrome, flavoxanthin, and
luteoxanthin (54). Grape carotenoids decrease progressively during maturation, with a
concomitant increase of the volatile compounds. This degradation would occur during
berry metabolism either enzymatically or by chemical pathway in acid medium (54, 55).
This would account for the presence of volatile compounds, such as p-ionone and
p-damascenone, identified in grape (56) and possibly originating in carotenoids (36).
Many norisoprenoids occur in grapes as glycosidic precursors. Enzymatic and acid

11
cleavage during crushing, fermentation, and bottle-aging result in cleavage of the bound
sugar moiety releasing the free norisoprenoid aglycone (57, 58).
Gas Chromatography-Olfactometry
Gas chromatography-olfactometry (GC-O) is a technique that allows the effluent
from the GC column to be evaluated for aroma activity using the human nose. The
effluent from the GC column is usually split between an FID detector and sniff port. The
human being detects which of the volatiles eluting from a GC column are aroma active,
as well as to describe aroma quality, and to estimate aroma intensity. The FID detector is
used as a general mass detector. Some of the GC-0 techniques available are Charm
Analysis (59), Aroma Extraction Dilution Analysis (AEDA) (60), and OSME (61) which
is a time intensity method. Charm Analysis and ADEA are based on the determination
of odor detection thresholds of the compounds through a series of dilutions. Both define
aroma strength in terms of its dilution strength. OSME determines intensities based upon
magnitude estimation using a variable potentiometer to estimate intensity. Da Silva et al.
(61) suggested that dilution techniques might not give accurate values of aroma intensity,
since the odorants may have different dose-response functions above their thresholds.
Stevens law (62) establishes that the odor intensity (I) of a compound increases as a
power function (n, which varies from compound to compound) of the concentration
within a certain range of concentration (C) directly above the detection threshold (7).
The law is commonly expressed as:
I = k (C-T)n,
where k represents the proportionality constant. Response will increase once the
threshold concentration is exceeded. Even though not defined by the above equation, a
limit will be reached where the sensory response will no longer increase with increasing

12
concentration. This point is defined as saturation. When a sample is diluted below the
odor-detection threshold, there will be no sensory response. Stevens law suggests that
two different compounds (A and B) at the same concentration, with similar detection
thresholds but with different exponents (n values), will produce different dose-odor
intensity profiles (Fig. 2-5). Individual odors will contribute differently to the overall
food aroma intensity, depending on their concentration and n value (61).
Figure 2-5. Stevens law, comparing two difference compounds: A= compound A, B=
compound B.
The OSME is a time intensity procedure that determines the intensity of the
perceived odor without dilution. In this method, the trained assessors sniff the effluents
from GC mixed with humidified air, and directly record the odor intensity and duration of
each odor active component while describing its odor quality. Intensities of individual
components are plotted versus elution time; and the resultant graph is known as an
aromagram.

13
Solid Phase Microextraction
Solid phase microextraction (SPME) is a relative new technique whereby analytes
of interest partition from the sample matrix into a polymeric coated silica fiber,
developed by Pawliszyn and co-workers (63). It is a simple, rapid, solventless technique
to sample static headspace volatiles. A 1 or 2 cm length of fused silica fiber, coated with
a polymer, is bonded to a stainless steel plunger and installed in a holder that looks like a
modified microliter syringe. The plunger moves the fused silica fiber into and out of a
hollow needle. To use the unit, the analyst draws the fiber into the needle, passes the
needle through the septum that seals the sample vial, and depresses the plunger, exposing
the fiber to the sample or the headspace above the sample. Organic analytes adsorb to the
coating on the fiber. After adsorption equilibrium is attained, the fiber is drawn into the
needle, and the needle is withdrawn from the sample vial. Finally, the needle is
introduced into the gas chromatographic (GC) injector, where the adsorbed analytes are
thermally desorbed and delivered to the GC column.
The application of headspace SPME to flavor volatile compounds has been
employed in the study of flavor volatiles in orange juice using a PDMS coated fiber (64),
a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65), PDMS and polyacrylate
fiber (66). The partition coefficients of the polymeric coatings for the analyses differed
markedly. For example, terpenes such as a-pinene, P-myrcene, y-terpinenes, and
limonene are all nonpolar, and were extracted to a higher degree into the nonpolar PDMS
coating (66). Corresponding PDMS extracted the least amount of the more highly polar
volatiles, PDMS/DVB and Carbowax/DVB had partition coefficients higher than that of
PDMS for the most polar molecules (67). The Carboxen-PDMS fiber coating was more

14
selective for terpenes than early eluting alcohols and aldehydes (6). Polyacrylate was
more effective in extracting highly polar compounds such as methanol and ethanol (66).
Orange Juice Norisoprenoids
Only a single norisoprenoid (P-ionone) has been reported and completely identified
in fresh orange juice (4, 5). Recently, P-damescenone has been reported in heated orange
juice, but not completely identified (65). With so many carotenoid precursors present in
orange juice, it seems highly likely that additional norisoprenoids would also be present.
The primary objective of this study was to determine if these additional norisoprenoids
were present in orange.

CHAPTER 3
HPLC DETERMINATION OF CAROTENOID NORISOPRENOID PRECURSORS IN
ORANGE JUICE
Introduction
The color of orange juice is due to a complex mixture of plant pigments called
carotenoids. Over 50 carotenoids have been identified in orange juice including P-
carotene, a-carotene, P-cryptoxanthin and neoxanthin (17, 68). In addition to acting as
plant pigments and free radical scavengers (produced during photosynthesis), these large
highly conjugated molecules can break down forming smaller, highly potent aroma
volatiles called norisoprenoid (15, 39, 69, 70). The structure of P-carotene is shown
below. If this molecule hydrolyzes between carbon atoms 9 and 10, a C13 norisoprenoid,
P-ionone is formed. If the molecule hydrolyzes between carbon atoms 7 and 8, then a C10
norisoprenoid, P-cyclocitral is formed.
Figure 3-1. Possible degradation pathways for the formation of P-cyclocitral and P-
ionone from P-carotene.
Beta-cyclocitral, P-damascenone, a-ionone, and P-ionone are some of the volatiles
reported in tomato (23), wine (71), tobacco (22) and tea (16) as aroma active compounds.
15

16
In each case carotenoids have been shown to be their precursors. Citrus carotenoids have
been examined using a variety of chromatographic techniques such as column and thin
layer chromatography, TLC, and more recently high performance liquid chromatography,
HPLC (68, 72, 78). Once separated, individual carotenoids have been identified
primarily by their unique three fingered visible absorbance patterns. In orange juice,
most oxygen containing carotenoids are esterified with C12-C18 fatty acids (74) thus
increasing their size and structural complexity. The most common practice is to de-
esterify (hydrolyze) these esters so that each carotenoid will elute as a single peak rather
than several smaller peaks with various fatty acids attached. However, even with
hydrolysis, the large numbers and subtle structural differences of orange juice carotenoids
provide a severe separation challenge. To complicate matters further, carotenoids are
sensitive to heat, light and oxygen, thus artifacts are readily formed during sample
preparation and/or analysis steps. HPLC equipped with a photodiode array detector is the
preferred analytical technique of choice to separate and quantify carotenoids without
artifact formation. Both normal phase and reverse phase chromatography have been
employed to separate these plant pigments, but the reverse phase approach offer the most
advantages. The most common reverse phase column is C-18 and most of the early
carotenoid separations employed this column. However, in recent years a C-30 reverse
phase column has been developed especially for carotenoid separations. Several
investigators (68, 75, 76) have employed this column with ternary solvent gradient and
photodiode array detector to isolate and identify the complex mixture of carotenoids in
orange juice.

17
Objectives
The objective of this study was to confirm the presence of specific carotenoids in
Valencia orange juice which could serve as norisoprenoid precursors. The specific
carotenoids of interest include: a-cryptoxanthin, (3-cryptoxanthin, a-carotene, (3-carotene
and neoxanthin because they possess the structural features needed to serve as precursors
to the newly identified norisoprenoids. (See Objective #1)
Materials and Methods
Carotenoid Extraction
The carotenoid extraction method according to Lee et al. (75) was carried out with
slight modification. A 25 mL aliquot of Valencia juice was extracted with 50 mL of a
mixed solvent (hexane:acetone:ethanol, 50:25:25) using a Omni mixer homogenizer
(model no. 700, Lourdes, Vemitron Medical Products, Inc. Carlstadt, NJ). It was
extracted for 5 min at medium speed in ice bath, and centrifuged (CR412, Jouan, Inc.,
Winchester, VA) for 10 min at 4000 rpm and 10C. The top layer of hexane containing
pigments was collected and concentrated to dryness in rotary evaporator.
Carotenoid Saponification
Saponification was carried out according to Noga and Lenz (77) with slight
modification. The dried pigment was redissolved with 2 mL of methyl tert.-butyl ether
(MTBE), and placed in a 40 mL vial. Two mL of 10% methanolic potassium hydroxide
(KOH) was added to the sample and the headspace was blanketed with nitrogen before
closing. The sample was wrapped with aluminum foil to protect it from light, and placed
at room temperature for 1 hour. The sample was then transferred to separatory funnel to
which 5 mL of water was added and 2 mL of 0.1% butylated hydroxyl toluene (BHT) in
MTBE, and the aqueous layer removed. Additional water rinses were carried out until

18
free of alkali. The MTBE layer was then filtered through a small glass column filled with
deactivated glass wool (Restek Corporation, PA) and anhydrous sodium sulfate (Fisher
Scientific, NJ) to remove residue water from MTBE layer. Each sample was
concentrated by evaporation with nitrogen, and the volume adjusted with 0.1% BHT in
MTBE to 1 mL and placed in sealed amber vials under refrigeration (4C) until analyzed.
HPLC Procedure
Carotenoid pigments were analyzed according to Rouseff et al.(<58) by reverse
phase HPLC using ternary gradient of water, methanol, and MTBE with photo diode
array detection (PDA] by reverse phase HPLC using ternary gradient of water, methanol
(MeOH), and MTBE with photodiode array detection (PDA). The 4.6 mm i.d. x 250 mm
YMC Carotenoid 5 pm column (YMC, Inc., Waters Corporation, MA) was used. The
chromatographic system consisted of autosampler, LC pump, and PDA detector
(Surveyor, ThermoFinnigan, CA). The PDA was set to scan from 280 to 550 nm. Three
separate data channel were set to record the absorbances at 350, 430, and 486 nm with
spectral bandwidths of 1 nm. Data were collected, stored, and integrated, using the Atlas
software (Atlas 2003, Thermo Electron Corporation, Cheshire, UK). All reagents used
were HPLC grade (Fisher Scientific, NJ). One standard, (3-carotene, was purchased from
Acros (Acros, NJ). The initial ternary gradient composition consisted of 90% MeOH, 5%
water, and 5% MTBE. The solvent composition changed in a linear fashion to 95%
MeOH and 5% MTBE at 12 min. After the next 8 min (at 20 min) the solvent
composition was 86% MeOH and 14% MTBE. At this composition the solvent
composition was gradually changed to 75% MeOH and 25% MTBE at 30 min. The final
composition was 50% MeOH and 50% MTBE at 50 min. Intial conditions were

19
reestablished within 2 min and reequilibrated for 15 min before the next injection. Flow
rate was 1 mL/min and injection volume was 10 pL.
Results and Discussion
Carotenoids of Interest
Although over 50 carotenoids have been identified in orange juice, only a few
possess the structural requirements to produce potent norisoprenoids. The structures of
the carotenoids which have been shown to produce norisoprenoids of interests in other
food systems (15, 23, 31, 39, 69, 70, 78) are shown in Fig. 3-2. Hydrolysis points are
indicated with arrows and resulting norisoprenoid indicated as text.
Figure 3-2. Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of P-damascenone.
It is worth noting that the structures of the left half of the first four carotenoids are
identical. Each of these four carotenoids can produce either 3-cyclocitral or (3-ionone.

20
The right half of these four carotenoids differ considerably. The right half of (3-carotene
can produce both (3-cyclocitral and (3-ionone because it is identical to the left half of the
molecule. The right half of a-carotene can produce a-ionone and neither a- or (3-
cryptoxanthin produce norisoprenoids which were observed in orange juice. The final
carotenoid of interest, neoxanthin, has been shown to produce (3-damescenone in a three
step process (34).
Hydrolysis Conditions
As previously discussed, citrus carotenoids must be hydrolyzed to simplify the
separation due to the complexity from the multiple natural esters formed from C^-Cis
saturated fatty acids (74). Hydrolysis conditions must be optimized in order to free the
esterified carotenoids into a single form but not so long as to promote alkaline hydrolysis
of the carotenoids. Concentration of alkali, reaction time and temperature are the
variables of interest. In recent years, most carotenoid studies have employed 0.1 M KOH
and room temperature so only reaction time was optimized for this study.
Chromatograms with no saponification showed 80% of the total peak area eluting as an
unresolved band of peaks during the last quarter of the chromatogram. As saponification
time increased, the number of peaks at the end of the chromatogram diminished and the
peaks were more evenly distributed during the chromatographic run. Saponification
times in excess of one hour did not reduce the number of late eluting peaks and total
carotenoid peak area was lower at 4 hours and overnight saponification compared to the
one hour saponification. Therefore the one hour saponification was used for the
remainder of the study.

21
HPLC Separation
C-30 carotenoid columns with ternary gradient solvent systems and photodiode array
detectors have been employed to separate and identify carotenoids in citrus (68, 75, 76).
In this study saponified carotenoids were separated using a C-30 carotenoid column with
ternary solvent gradient system of water, methanol, and MTBE with photodiode array
detection. The resulting separation is shown in Figure 3-3. More than twenty-four
carotenoids were separated as distinct peaks and sixteen of these peaks were identified
based on their spectral characteristics (Table 3-1), relative elution order compared to
literature values and authentic standards. As seen from the chromatogram in Fig. 3-3,
peaks 11 and 20 corresponding to cis-violaxanthin and (3-cryptoxanthin (15.76 and 12.34
percentage of total peak area, respectively). They have been reported as the major
carotenoids in earlier studies. Beta-cryptoxanthin is well accepted as the major
contributor to the orange color of the juice (79) because of its relatively high
concentration an overall absorbance in the red/orange range of the spectrum. The last
four peaks (21-24) are due to a variety of carotenes which are not completely resolved.
Both a- and (3-carotene are of particular interest in this study because of their ability to
produce a range of norisoprenoids which have been observed in other food products. In
addition, peaks 18 and 20 were well resolved and corresponded a- and (3-cryptoxanthin
from the match of retention times and spectral characteristics compared to authentic
standards. The final peak of interest was neoxanthin and this compound corresponds to
peak 4 which is neither particularly well resolved nor large.
The large number and similarity of orange juice carotenoids make separation
difficult. Thus even under the best chromatographic systems, some peaks will not be

22
well resolved (i.e., peaks 8-10) and make accurate identification difficult. Lutein and
zeaxanthin (peaks 14 and 15, respectively) are usually difficult to resolve as they differ
only in the position of a single double bond in one of the terminal rings. These pigments
can be separated on a ZnC03-MgC03 column and the separation requires several hours
(80) but they are completely resolved in this chromatographic system. However lutein is
barely resolved from mutatoxanthin (peak 13) even though mutatoxanthin contains an
extra 5,8 epoxide group. Phytofluene and a-cryptoxanthin (peak 19 and 18, respectively)
were not well resolved chromatographically, but could easily be separately quantified
using different monitoring wavelengths as their respective absorbance maxima differ by
approximately 100 nm.
Retention Time (min)
Figure 3-3. Chromatogram of saponified carotenoid extract from orange juice separated
using a YMC C30 reverse phase carotenoid column and a water, MeOH,
MTBE ternary solvent gradient. Spectral characteristics for each numbered
peak are summarized in Table 3-1. See HPLC experimental section for
additional details.

23
Carotenoid Identification
Shown in Figure 3-4 is an overlay of peaks 4, 12 and 24. The height of their
spectra corresponds to their relative peak heights since the spectra were taken from the
apex of each peak. These were chosen to show the range and diversity of these spectra
which are not conveyed when just tabulated peak maxima are tabulated. The shape of the
absorbance band as well as the location of the absorbance maxima are all highly
characteristic of individual carotenoids. This information taken with retention time can
Figure 3-4. Absorbance spectra for (3-carotene (a), Ieutoxanthin (b), and neoxanthin (c),
peak 24, 12 and 4 respectively.
be used to identify specific carotenoids, especially if their spectral and chromatographic
characteristics have been reported elsewhere. The spectra and relative retention times of
a-, (3-carotene, a-, (3-cryptoxanthin and neoxanthin matched their published values and
were used as confirmation of the presence of these peaks in orange juice. It should be
pointed out that these five carotenoids have been previously reported in orange juice (77)

24
As shown in Table 3-1 the absorbance maxima observed exactly matched those published
in the literature or differed at most by 2 nm as in the case of the central peak for (3-
cryptoxanthin. Since the wavelength accuracy of most photodiode array detectors is only
+ 1 nm, the agreement is excellent. Since the carotenoids of interest have the same
elution and spectral characteristics as a-, (3-carotene, a-, (3-cryptoxanthin and neoxanthin,
it is reconfirmed that they are present in orange juice and could potentially serve as
norisoprenoid precursors.
Table 3-1. HPLC retention times, spectral characteristics of orange juice carotenoids
Peak
no.
Carotenoid
RTa
(min)
Observed (nm)
Literature (nm)
Ref.b
Peak
1
Peak
2
Peak
3
Peak
1
Peak
2
Peak
3
1
Valenciaxanthin
5.52
351
369
390
351
369
390
E
2
6.00
371
391
414
3
6.92
420
435
465
4
Neoxanthin
7.35
416
438
468
415
439
467
A
5
11.60
410
431
454
6
12.95
416
438
467
7
Neochrome
14.68
400
422
448
399.5
421.5
447.5
B
8
15.55
408
429
415
9
15.90
383
402
425
10
16.83
411
430
462
11
Cis-violaxanthin
17.60
415
437
464
414
437
464
C
12
Leutoxanthin
19.07
399
418
443
399.5
419.5
441.5
B
13
Mutatoxanthin
20.08
405
429
451
404
427
452
D
14
Lutein
20.92
420
445
471
424.5
445.5
471.5
B
15
Zeaxanthin
23.80
425
450
476
425
450
478
A
16
Isolutein
24.70
418
441
468
418
439.5
467.5
B
17
26.40
429
445
469
18
oc-cryptoxanthin
28.15
420
445
472
420
444
472
D
19
Phytofluene
28.83
331
348
367
331
348
367
A
20
3-cryptoxanthin
31.57
425
451
477
425
449
476
A
3-carotene,5,8:
21
5',8'-diepoxide
34.30
380
400
424
380
400
425
A
22
a-carotene
36.33
420
446
472
420
445
472
D
23
^-carotene
39.28
379
401
425
378
400
425
A
24
3-carotene
39.77
425
451
477
425
450
478
A
aRT = retention time, bA= Britton (81); B= Rouseff et al. (68); C= DeRitter and Prucell
(82); D= Farin et al. (83); E= Curl and Bailey (84).

25
Conclusions
Carotenoids in Valencia orange juice were extracted using mixed solvent
(hexane:acetone:ethanol, 50:25:25) and subsequently saponified. The saponified
carotenoids were separated using a C-30 carotenoid column with a ternary gradient
solvent system. Twenty-four carotenoids were separated as distinct peaks and sixteen of
these peaks were identified based on their spectral characteristics (Table 3-1), relative
elution order compared to literature values and authentic standards. Although they have
been reported previously, the presence of a-cryptoxanthin, (3-cryptoxanthin, a-carotene,
(3-carotene and neoxanthin in orange juice was confirmed by comparing retention and
spectral properties with standards or literature values. These specific carotenoids were of
interest because they possess the direct structural segments needed to serve as precursors
potent aroma norisoprenoids.

CHAPTER 4
IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING TIME
INTENSITY GC-0 AND GC-MS
Introduction
Early GC-0 studies (4, 6, 7) have shown that many aroma active compounds in
orange juice exist as low-level volatiles that are difficult to detect using typical FID or
MS detectors. Furthermore, these studies also demonstrated that the orange juice
volatiles present in highest concentration have little to no aroma activity. Recently, the
25 most intense aroma active compounds in fresh juice, as determined by dilution
analysis (5), were quantified using isotope dilution analysis (5). Beta-ionone is the only
orange juice norisoprenoid, which has been fully identified (4, 5). Even though it has a
moderately intense aroma, it was not one of the 25 odorants recently quantified using
isotope-dilution analysis (5). Norisoprenoids are volatile C9-C13 fragments with
extremely low aroma thresholds which can be formed from the degradation of C40
carotenoids. This degradation can be the result of in vivo enzymatic reaction, or post
harvest thermal degradation in foods containing carotenoids. They are also observed
from the release of glycosidically bound norisoprenoids which were originally from
carotenoid decompositions as in the case of wine (55). Norisoprenoids have been shown
to have significant aroma impact in fruits, vegetables and spices such as grapes (5),
apples (9), lychee (10) starfruit (11), mango (72), tomato (13), saffron (14) cured tobacco
(75) and black tea (16). During the ripening of red raspberries, a-ionone and (3-ionone
increased to produce the characteristic raspberry aroma (55). In heated apple juice,
26

27
p-damascenone contributed 32% of the total aroma potency of the juice, and only 1.6%
of the total aroma potency of fresh (unheated) apple juice (9). Safranal is a potent aroma
in saffron formed during drying and storage by hydrolysis from picrocrocin, a
monoterpene glycoside (86). Beta-cyclocitral, P-ionone and P-damascenone were
detected in fresh tomato. Only P-ionone and P-damascenone are the important to tomato
aroma because their concentrations (4 and 1 pg/L respectively) are higher than their odor
threshold (0.007 and 0.002 pg/L respectively). Beta-damascenone shows a ~ 10-fold
increase in concentration in heated tomato juice which was concentrated to tomato paste
(50). Buttery et al. (41) examined both low carotene and high P-carotene tomato lines for
norisoprenoids. They found that the high P-carotene line contained the highest
concentrations of P-ionone and P-cyclocitral. Both norisoprenoids are known biological
or chemical degradation products of P-carotene.
Carotenoids are widely distributed in the plant kingdom and orange juice is
particularly rich and a complex source of these compounds (87). Lutein, zeaxanthin,
P-crytoxanthin, a-carotene and P-carotene have been determined in unsaponified orange
juice carotenoids extracted by ethyl acetate (73). Thirtynine carotenoid pigments were
separated and identified in saponified orange juice carotenoids using HPLC (68). Among
these, P-carotene, a -carotene, neoxanthin P-crytoxanthin, lutein, violaxanthin, and
canthaxanthine have the structural potential to form potent norisoprenoid fragments
(18-22). These carotenoids have been confirmed to be present in the Valencia juice used
in this study based upon HPLC retention time and spectral characteristics data. (See
Chapter 3)

28
Objectives
Since orange juice has so many carotenoids that could serve as precursors for a
wide range of norisoprenoids, the objective of this study was to determine if more than
one aroma active norisoprenoid was present in fresh or heat-treated orange juice. If
additional norisoprenoids are found, they should be characterized and identified. (See
Objective #2)
Materials and Methods
Orange Juice Samples and Processing
Late-season Valencia oranges (from Haines City Citrus Growers Association,
Haines City Florida) were juiced using an FMC juice extractor at the Citrus Research and
Education Center (CREC), Lake Alfred, Florida. The oranges were juiced using a
commercial FMC juice extractor model 291 with standard juice settings. An FMC model
035 juice finisher (FMC Corp., Lakeland, FL) was used with a 0.02 inch screen. The
finished juice had a Brix value of 11.7, an acid content of 0.67% citric acid, a Brix/acid
ratio of 17.5 and an oil level of 0.0196%. The freshly squeezed juice was divided into
three groups. In Group 1, fresh orange juice was immediately chilled and NaCl (36 g/100
mL of juice) was added to inhibit enzymatic reactions. In Group 2, fresh orange juice
was pasteurized using UHT/HTST lab Microthermics tubular pasteurizer Model 25
(Microthermic Corp., Raleigh, NC) at 195F (90.5C), held for 12 seconds and filled at
41F (5C). The oil level of the pasteurized juice was 0.0168%. In Group 3, fresh
orange juice was concentrated to 65 Brix using a thermally accelerated short-time
evaporator (TASTE) built by Cook Machinery, Dunedin, Florida. The concentrate was
then reconstituted to 11.73 Brix by diluting with deionized water, but without restoring
volatiles.

29
Chemicals
Standard aroma compounds were obtained from the following sources: methional,
ethyl 2-methylpropanoate, ethyl 2-methylbutyrate, 1-octanol, 2-acetyl-2-thiazoline,
(E,Z)-2,6-nonadienal, (E)-2-nonenal, (E,E)-2,4-decadienal, l-octen-3-one, ethyl
hexanoate, (E,E)-2,4-nonadienal, L-carvone, E-2 octenal, terpinolene, a-terpinyl
acetate, a-terpineol, Z-4-decenal, neral, geranial, 4,5-epoxy-E-2-decenal, P-ionone and
P-cyclocitral were purchased from Aldrich (Milwaukee, WI). Octanal, limonene,
linalool, nonanal, hexanal, decanal, dodecanal, 1,8 cineole, citronella, terpinen-4-ol,
P-sinensal, P-myrcene, nootkatone ethyl butyrate, acetaldehyde, geraniol, and nerol were
obtained as gifts from SunPure (Lakeland, FL). Alpha-ionone was obtained as a gift
from Danisco (Lakeland, FL). The (Z)-2-nonenal was found in purchase of (E)-2-
nonenal at the 5-10% level. The (E,Z)-2,4-nonadienal and (E,Z)-2,4-decadienal were
found in the purchase of (E,E)-2,4-nonadienal and (E,E)-2,4-decadienal respectively.
Their identities were confirmed by mass spectra, retention indices and odor qualities.
Beta-damascenone and p-l-Menthen-8-thiol were obtained from Givaudan (Lakeland,
FL). The 4-mercapto-4-methyl-2-pentanone and 4-mercapto-4-methyl-2-pentanol were
synthesized in our laboratory. The 3-mercaptohexan-l-ol was bought from Interchim
(Montlucon, France).
Orange Juice Headspace Extraction
A 10 mL aliquot of orange juice was added to a 40 mL glass vial containing a
micro stirring bar and sealed with a screw-top cap that contained a Teflon-coated septa.
The bottle and contents were placed in a combination water bath and stirring plate set at
40C, and gently stirred. After equilibrating for 45 min a SPME fiber (50/30 mm

30
DVB/Carboxen/PDMS on a 2 cm StableFlex fiber, Supelco, Bellefonte, PA) was inserted
into the headspace of the sample bottle and exposed for 45 min. The fiber was then
removed from the headspace and inserted into the heated GC injector port at 220C
where the volatiles were thermally desorbed for 5 min.
Gas Chromatography: GC-FID and GC-Olfactometer
Separation was accomplished with a HP-5890 GC (Palo Alto, CA ) using either a
DB-wax column (30 m x 0.32 mm. i.d. x 0.5 mm, J&W Scientific; Folsom, CA) or
Zebron ZB-5 column (30 m x 0.32 mm. i.d. x 0.5 mm, Phenomenex, Torrance, CA).
Column oven temperature (for DB-wax) was programmed from 40 to 240C (or 40 to
265C for ZB-5) at 7 C/min with a 5 min hold. Helium was used as carrier gas at flow
rate of 1.55 mL/min. Injector and detector temperature were 220C and 290C,
respectively. A narrow-diameter injection port liner (0.75 mm.) was used to improve
peak shape and chromatographic efficiency for SPME thermal desorption. The entire
separation was conducted in the splitless mode. A GC splitter (Gerstel, Baltimore, MD)
split the column effluent between the FID and olfactometer (equipped with a high-
volume sniffing port, DATU, Geneva, NY) in a 1:2 ratio, respectively as described by
Bazemore et al. (6). A time-intensity approach was used to evaluate odor quality and
intensity at the sniffing port during the GC run. Assessors rated aroma intensity
continuously throughout the chromatographic separation process using a linear
potentiometer that supplied a continuous signal to the chromatographic software.
Retention times and verbal descriptors were recorded to permit aroma descriptors to be
coupled with computerized aroma time-intensity plots. Two olfactometry panelists were
trained in GC-sniffing with standard solution of 11 compounds typically found in orange
juice (ethyl butanoate, cis-3-hexenol, tran-2-hexenal, a-pinene, myrcene, linalool,

31
P-citronellol, carvone, terpin-4-ol, geranial, and neral). The panelists sniffed the effluent
of aroma standard from GC-0 with optimum positioning and breathing technique. The
intensity of each standard was recorded on a sliding scale (varying from none to strong
intensity) and panelists were provided verbal descriptors of aroma quality. For additional
experience, the extract of aroma volatiles from commercial orange juice was provided to
panelists under identical conditions. Panelists were accepted on they demonstrated an
ability to replicate aroma peak times for at least 80 % of the components in the test
mixture.
Two trained panalists evaluated the volatiles of orange juice (extracted by SPME)
in duplicate, thus producing four individual time-intensity aromagrams. Average
intensity from the four runs was calculated for each odorant. If no peak was detected in a
run, its value was treated as missing, not zero. An indication of aroma activity with
similar aroma descriptors, at the same retention was required from at least half the panel
results before a peak could be considered aroma-active. Averaged time-intensity
aromagrams were constructed by plotting average intensity versus retention time.
Chromatograms and aromagrams were recorded and integrated using Chromperfect
version 5.0, Justice Laboratory Software (Palo Alto, CA). Identification of the aroma-
active components was based on the combination of sensory descriptors, standardized
retention indices, and identification confirmed by comparison with standards and GC-MS
spectra.
Gas Chromatography-Mass Spectrometry
Orange juice headspace volatiles were extracted by SPME and introduced to the
GC-MS. Volatiles were separated and analyzed using a Finnigan GCQ ion trap mass

32
spectrometer (Finnigan, Palo Alto, CA) equipped with a DB5, 60M x 0.25 mm I.D.,
capillary column (J&W Scientific, Folsom, CA). The injector temperature and transfer
line temperature were 200 and 250 C, respectively. Helium was used as the carrier gas
at 1 ml/min. The oven temperature program consisted of a single thermal gradient from
40 to 275 C at 7C/min. The MS was set to scan from mass 40 to 300 at 2.0 scans/s in
the positive ion electron impact mode. The ionization energy was set at 70 eV.
Aroma Peak Identification
Initial identification was based on the combination of matches with standardized
alkane retention index values (Kovats Index) using two dissimilar column materials
(e.g., DB-wax and ZB-5) and aroma characteristics. If the aroma component was
sufficiently concentrated, fragmentation patterns were compared with library spectra
(NIST 2002 and Wiley (6th Edition) databases using the spectral fit criterion. Only those
compounds with spectral fit values equal to or greater than 800 were considered as
possible identification candidates. Whenever standards could be obtained, they were
used as a confirmation of identification, by comparing the resulting fragmentation
pattern, retention index value and aroma descriptor (88).
Results and Discussion
Extraction and Concentration of Juice Norisoprenoids
Solid phase microextraction (SPME) was used to extract and concentrate orange
juice volatiles because it is a rapid, solventless headspace sampling technique (<5). When
solvent extraction was used, early eluting peaks were obscured by the large quantities of
solvent. Early eluting aroma peaks such as acetaldehyde have been shown to be
important in orange juice flavor (89) but could not be examined using GC-0 in solvent-
extracted samples. Although solvent extraction would not have presented a problem in

33
determining norisoprenoids, as they elute fairly late, one of the secondary objectives in
the overall study was to determine the relative contribution of norisoprenoids to the total
aroma of orange juice. Solvent extracted juice samples would have been unsatisfactory
for this purpose for the above stated reason.
The application of headspace SPME to flavor volatile compounds has been
employed in the study of flavor volatiles in tomato and strawberry fruits using PDMS,
PDMS/DVB, and Carbowax/DVB coated fiber (67), in orange juice using a PDMS
coated fiber (64), a Carboxen-PDMS fiber (6), a DVB/Carboxen/PDMS fiber (65),
PDMS and polyacrylate fiber (66). The partition coefficients of the polymeric coatings
for the analyses differed markedly. For example, terpenes such as a-pinene, (3-myrcene,
y-terpinenes, and limonene are all nonpolar, and were extracted to a higher degree into
the nonpolar PDMS coating (66). Corresponding PDMS extracted the least amount of
the more highly polar volatiles, PDMS/DVB and Carbowax/DVB had partition
coefficients higher than that of PDMS for the most polar molecules (67). The Carboxen-
PDMS fiber coating was more selective for terpenes than early eluting alcohols and
aldehydes (6). Polyacrylate was more effective in extracting highly polar compounds
such as methanol and ethanol (66). Due to the wide range of volatile compounds from
orange juice and for the increased fiber capacity, the headspace volatiles in this study
were extracted and concentrated using the 50/30 mm DVB/Carboxen/PDMS coating on a
2 cm StableFlex fiber.
In examining adsorption curves for (3-cyclocitral, (3-damascenone and a- and
(3- ionone on the chosen fiber (see Fig. 5-1) it was concluded that 45 min. represents a
rough compromise for all four analytes between minimal exposure time and maximum

34
peak area. For example (Tdamascenone and (3-ionone reaching more than 80% of their
final equilibrium value within 45 min. It is a rare SPME analysis that employs true
equilibrium exposure time. If exposure time can be carefully controlled, then exposure
times of as little as 5 min. can be employed. These very short exposure times are usually
limited to analytes in relatively high concentration and even then the reproducibility is
not that good. In this study, very short exposure times were not an option as the analytes
of interest were present in very low concentration.
Time (min)
Figure 4-1. GC-FID (top) and average time-intensity of four GC-0 runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and
23 correspond to norisoprenoids, all numbers refers to compounds in Table
4-1

35
GC-Olfactometry
In this study, a total of 59 aroma active components were detected in SPME
headspace samples from fresh orange juice (orange juice group 1) Since the primary goal
of this study was to determine if additional aroma active norisoprenoids were present in
orange juice, GC-0 was employed primarily in the region where p-ionone and other
norisoprenoid standards eluted. Using standards of P-cyclocitral, a-ionone, P-ionone
and P-damascenone, the retention time region was established between 12 and 20 min,
and the resulting aromagram and concurrent chromatogram is shown in Fig. 4-1.
As noted in Fig. 4-1, four aroma peaks corresponding to peaks 5, 19, 21 and 23
were observed at the identical retention times as P-cyclocitral, P-damescenone, a-ionone
and P-ionone respectively. It is also apparent from the relative intensities shown in Table
3-1, that these potential norisoprenoid peaks were among the more intense aromas.
Beta-ionone was the most intense and P-cyclocitral was the weakest aroma compound of
all the four potential norisoprenoids observed. When the samples were rerun on a DB-
wax column the four aroma peaks also were found at retention index values that
corresponded with the four potential norisoprenoids. Furthermore, the aroma quality of
each juice norisoprenoid corresponded exactly with the aroma description of standards.
Since these compounds have the same retention behavior on two very dissimilar
chromatographic columns and also have the same aroma quality as standards, they are
probably P-cyclocitral, P-damescenone, a-ionone and P-ionone respectively. This
represents the first time that p-cyclocitral, and a-ionone have been reported in orange
juice. Beta-damescenone had recently been reported in heated orange juice but its
identity was not confirmed by supporting instrumental methods (65).

36
Table 4-1. Identification, retention characteristics and aroma descriptions of aroma
active compounds in fresh orange juice
No.
Compound
Aroma descriptor
Linear retention
Relativec
intensity
ZB-5
DB-wax
1
Terpinen-4-olb
Metallic, musty
1175
1619
5
2
Z-4-decenala
Green, metallic, soapy
1188
1542
7
3
Decanalb
Green, soapy
1198
1508
7
4
(E,E)-2,4-nonadienala
Fatty, green
1209
1702
7
5
3-cyclocitralb
Mild floral, sweet, hay-like
1214
1632
6
Nerol3
Lemongrass
1222
1798
5
7
Neralb
Lemongrass
1236
1692
7
8
L-carvoneb
Minty
1242
1747
8
9
Unknown
Metallic/ woody
1247
6
10
Geraniol3
Citrus, geranium
1265
1853
9
11
Unknown
Soapy, almond
1274
7
12
1 -p-menthene-8-thiola
Grapefruit
1281
1619
7
13
(E,Z)-2,4-decadienal3
Metallic, geranium
1293
1759
4
14
Geranial3
Green, minty
1310
1742
4
15
(E,E)-2,4-decadienal3
Fatty, green
1314
1819
4
16
a-teroinvl-acetate3
Sweet
1349
1663
6
17
4,5-epoxy-E-2-decenal3
Metallic, fatty
1375
2010
6
18
Unknown
Sweet nutty
1380
7
19
(3-rlflma Tobacco, apple, floral
1383
1829
7
20
Dodecanal3
Soapy
1403
1722
5
21
a-iononeb
Floral
1426
1863
8
22
Unknown3
Fermented, rancid butter
1459
5
23
3-iononeb
Floral, raspberry
1484
1951
24
Unknown
Nutty
1510
8
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as
compared with standard
b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as
compared with standard, and MS
c Averages of normalized intensities (10) evaluated by two trained panelists in four
replications
Mass Spectrometry Norisoprenoid Identifications
Headspace volatiles from fresh orange juice were analyzed using capillary GC with
an ion trap mass spectrometer. To achieve greater selectivity for the norisoprenoids of
interest, selected ion chromatograms were reconstructed in the retention region where
norisoprenoid standards were found to elute. The selectivity achieved is demonstrated in
Fig. 4-2. Specific m/z values were evaluated to provide the best peak height for each
norisoprenoid of interest as well as minimizing interference from non-norisoprenoid

37
components as well as noise. The following ions were monitored for the specific
norisoprenoids: P-cyclocitral, m/z = 137 and 152; P-damascenone, m/z = 175 and 190;
a-ionone, m/z = 177 and 192; P-ionone, m/z = 177 and 192.
Although only a single ion has been shown for each norisoprenoid, two or more
selective ions were employed to detect the presence of specific norisoprenoids. For
example, the selected ion chromatogram using m/z 137 was more intense than that from
m/z = 152 but not as specific for P-cyclocitral. The selected ions of m/z = 177 and 192
were extracted for the determination of a-ionone and P -ionone. Selected ion
chromatograms at m/z 177 provided excellent signal strength and selectivity for P -
ionone. The a-ionone was obviously present at much lower concentrations than P-
ionone. The SIC chromatogram at m/z = 192 (Fig. 4-2) provides more selectivity for a-
ionone but better signal and noise ratio was obtained at m/z 177.
*'\ TIC
m/z = 137
17.97 18.46 1&3E 19.3j19.70mm 20.56 21.2921.49 22.08
M 18.69 1911947 19.70 20.08 20.47
20.87
m/z = 177
21.85
m/z = 192
16.5 17.0 17.5 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0
Time (min)
Figure 4-2. Comparison between total ion chromatogram and selected ion
chromatograms (SIC) A: P-cyclocitral, B: P-ionone, C: a-ionone.

38
a>
u
c
9
"O
c
a
A
Â¥
a
a
P
Time (min)
Figure 4-3. Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190. Middle chromatogram P-damascenone
detected from orange juice, and lower overlay chromatogram of spiked (A)
and non-spiked (B) of orange juice with standard P-damascenone.
The P-damascenone, selected ion chromatograms using m/z = 175 and 190 (two
masses highly characteristic for P-damascenone) did not provide a clear signal at the
expected retention time of P-damascenone using the ion trap MS. Beta-damascenone had
been detected by GC-0 at the expected retention time with the characteristic aroma but
not detected by either FID or SIC ion-trap MS, suggesting that P-damascenone, if
present, was there at very low levels. Beta-damascenone has an extremely low odor
threshold, which is below the detection limits of most instrumental detectors (0.002
Hg/L). However, by employing quadrupole mass spectrometer in the single ion

39
monitoring mode at least a lOx greater sensitivity (lower detection level) can be achieved
because all the ions of a single mass are continuously measured rather than measured for
an instant before monitoring other masses. Using selected ion monitoring m/z values 175
and 190, P-damascenone was detected at the expected retention time (see Fig. 4-3 for the
case at m/z 190). The combined GC-0 and two SIM peaks at the exact retention time of
P-damascenone, confirm its presence in orange juice.
Although selected ion chromatograms strongly suggest the presence of the other
norisoprenoids of interest, they do not offer absolute proof. They only indicate that a
juice volatile elutes at the identical retention time as the norisoprenoid of interest, and
this volatile contains the same mass fragment. The combination of this information with
the GC-0 information provides three independent pieces of information strongly
suggesting the presence of specific norisoprenoids. However, to absolutely confirm the
presence of P-cyclocitral, a-ionone, and P-ionone, their spectra from the juice MS data at
the retention times of each respective norisoprenoid was obtained and compared with
reference spectra in standard libraries or compared with that obtained from authentic
standards. The resulting match for the case of P-cyclocitral is shown in Fig. 4-4. It is
readily apparent that although the relative ion abundances are not the same (usually a
function of instrument to instrument variation) an excellent spectral match has occurred
and that the presence of P-cyclocitral in orange juice is confirmed. Comparing the
relative abundances of ions m/z 137 vs 152 in the upper spectrum of Fig. 4-4, it can be
readily appreciated why examining the selective ion chromatogram at 137 provided a
better signal to noise ratio than the chromatogram at 152, the molecular ion for P-
cyclocitral.

40
<
o
X!
3
3
X
<
>
4>
3
0)
04
m/z
Figure 4-4. Upper spectra from orange juice MS at RT = 17.68 bottom spectra of |3-
cyclocitral from database NIST 2002.
O
C
a
T3
C
3
X
<
>

4
13
04
m/z
Figure 4-5. Upper spectra from orange juice MS at RT = 21.94, bottom spectra from
standard P-ionone using identical ion trap MS at identical retention time.

41
However, a-ionone has not been previously reported and is shown in Figure 4-6.
The spectral match in this case is good considering the very low levels of a-ionone
present, but not perfect. Even with careful background subtraction (which was done for
all the previous spectra as well), there will be a fair amount of extraneous peaks simply
due to random noise. However, the major fragment ions of m/z 192 (M+), 177, 163, 136,
121, 109, 93, 91, and 77 are all present, more than enough to confirm the presence of a-
ionone in orange juice.
m/z
Figure 4-6. Upper spectra from orange juice MS at RT = 20.87, bottom spectra from
standard a-ionone using identical ion trap MS at identical retention time.
Conclusion
Four norisoprenoids in fresh orange juice (p-cyclocitral, (3-damascenone, a-ionone,
and (3-ionone) have been conclusively identified through the combined information from

42
GC-0 retention index matches with standards on two dissimilar chromatographic column
materials, aroma descriptor matches and GC-MS matches of both retention time and
fragmentation spectra. Of these four norisoprenoids, P-ionone had been reported in two
previous orange juice GC-0 studies (4, 5). There is one previous mention of P-
damascenone in heated juice, but no MS or independent instrumental confirmation data
was presented (65). P-cyclocitral and a-ionone were detected in orange juice for the first
time in this study and confirming MS data for P-damascenone was presented for the first
time.

CHAPTER 5
QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT OF
NORISOPRENOIDS IN ORANGE JUICE
Introduction
Chromatographic data is often used to determine the relative concentrations of
components in a volatile mixture. Within the linear range of the detector, integrated peak
area is proportional to the amount of that component in the sample. The four techniques
commonly employed to quantify chromatographic components are normalization;
internal standards; external standards; and standard addition methods (90). Only a few of
these have been employed to determine the amounts of specific volatiles in orange juice
volatiles to better understand their contribution to orange flavor (2, 5). Buettner and
Schieberle (5) employed stable isotope dilution assay to quantify 25 volatiles from a
solvent extract of hand-squeezed Valencia orange juice. The juice was spiked with a
known amount of the labeled internal standard and the juice was extracted with diethyl
ether and subsequently analyzed by GC-MS. Standard curves of the labeled and
unlabeled reference odorants were used to establish a relationship between peak area and
concentration. Odor activity values (OAV, concentrations of the odorants divided by
their odor threshold) were determined to estimate their respective odor contributions.
The highest OAVs were calculated for (s)-ethyl 2-methylbutanoate, ethyl butanoate,
(Z)-3-hexenal, ethyl 2-methylpropanoate, acetaldehyde, and (R)-limonene.
Moshonas and Shaw (2) quantified the volatiles from orange juice using dynamic
headspace GC with a pressurized purge and trap apparatus. Concentrations for each
43

44
volatile were calculated using the standard addition procedure. Regression equations
were developed from peak area data from four different concentrations of each compound
added to a juice base. Odor activity values were calculated for each component measured
(although they were not identified as OAV values). Compounds which exceeded their
threshold by the greatest amounts (highest OAV values) and thus most likely to
contribute to fresh orange flavor included: limonene, myrcene, a-pinene, decanal,
octanal, ethyl butanoate, and linalool. The differences between these two studies which
both claim to determine the components most responsible for fresh orange juice flavor
are worth noting.
SPME is a rapid, solventless static headspace procedure. It can be used for the
quantitative analysis of flavor and fragrance compounds. The standard addition method
has been used primarily because the concentration in the headspace (volatility) will be
influenced by the sample matrix (66, 91, 92). Boa et al. (92) reported that reliability
problems of headspace SPME quantification is associated with the matrix and could be
reduced by employing the standard addition method or employing isotopically labeled
internal standards. Headspace SPME with standard additions were used in the present
study because SPME can extract and concentrate orange juice headspace volatiles which
transfer them directly into the injector of a GC in a simple, straightforward manner. Just
as important for this study, the nonvolatile carotenoids will not be extracted. If the
nonvolatile carotenoids were present they might degrade when exposed to the heat
(200C) of GC injection port and possibly produce artifact norisoprenoids. The major
problem with the standard addition approach is that several injections at each standard
addition level are required in order to obtain a single result. Thus, depending on the

45
number of levels and the number of injections per level, this procedure can be time
consuming. However, once a pseudocalibration curve produced, the calculated slope can
be used for other samples of similar matrix. Thus, it is not essential that the standard
addition be employed for each and every sample, but the slope of the pseudocalibration
curve should be checked from time to time. All solutions analyzed must fall within the
linear range of the detector response (90).
Objectives
The primary goal of this study was to quantify the norisoprenoids in orange juice
using static headspace SPME with standard additions and GC-MS. (Objective 4) A
secondary goal was to determine the relative aroma contribution of all four
norisoprenoids to the total orange juice aroma. (Objective 3)
Materials and Methods
Quantification of Norisoprenoids in Orange Juice
The adsorption (amount vs. exposure time) curves for the SPME fiber (50/30 mm
DVB/Carboxen/PDMS) employed in this study was determined by varying exposure time
from 5 to 150 minutes. Since native concentrations were so low, orange juice samples
were fortified with 8.6, 4, 5.4, and 5.27 ppm P-cyclocitral, [3-damascenone, a-ionone, and
P-ionone respectively so that adsorption characteristics could be more accurately
determined. Ten milliliters of the fortified juice were transferred to 40 ml vial with screw
cap coated with Teflon. After 45 min at 40C, the headspace volatiles were extracted
using SPME (as described in Chapter 4)
Headspace SPME and GC-MS were used to quantify norisoprenoids in orange juice
using the standard addition method. Each standard (P-cyclocitral, p-damascenone,

46
a-ionone, and (3-ionone) was added separately to the orange juice sample to obtain the
final concentration of each norisoprenoid from 0 to 2 ppm. Beta-damascenone was the
only exception; its added concentrations ranged from 0 to 0.02 ppm. Sampling was
accomplished by adding a 10 mL aliquot of the juice to a 40 ml glass vial containing a
micro-stirring bar sealed and a Teflon coated septa. Samples were equilibrated at 40C
for 45 minutes and gently stirred before a SPME fiber was inserted into the headspace of
the sample bottle and exposed for another 45 min. The fiber was then removed from the
headspace and inserted into the GC-MS. GC conditions were the same as in Chapter 4.
Each sample was prepared and injected at lease twice. Quantitative measurements were
made using integrated peak areas from selected ion chromatograms. The ions chosen to
reconstruct these single ion chromatograms were at m/z 137, 177, 177, and 190 and were
fairly unique for P-cyclocitral, P-ionone, a-ionone, and P-damascenone respectively.
The latter m/z values corresponded to the respective molecular ion of P-damascenone.
In order to quantify the low levels of P-damascenone, a quadrupole MS (Agilent
5973 Network Mass Selective Detector, Agilent Technologies, CA) was employed using
selected ion monitoring (SIM) mode at m/z 190. It was equipped with HP an Innowax 30
m x 0.25 pm x 0.25 pm capillary column (Agilent/J&W HP Innowax, Scientific
Instrument Services, Inc., NJ) and autosampler (Gerstel Multi Purpose Sampler MPS2,
Gerstel Inc., MD). The oven temperature program consisted of two ramps from 90 to
160C at 6C/min and from 160C to 250C at 120C/min (in order to shorter the GC
running time after the P-damascenone was eluted). Each sample was analyzed from the
response at m/z 190. A graph of SIM 190 peak area versus concentration was prepared

47
and the amount of (3-damascenone in the sample determined from the regression line
equation.
Determination of the Relative Impact of Norisoprenoids in Orange Juice
The aroma active compounds from 3 types of orange juice (fresh, pasteurized, and
reconstituted from concentrate) were separated and identified using GC-0 (chapter 4).
Intensities of aroma active compounds of each run were normalized so the highest
intensity was given a score of 10. The normalized intensities of all the runs were then
averaged, providing a similar aroma activity was detected at least half the time at that
retention time. If the compound was not detected in one run its value was treated as
missing, not zero. Aroma-active compounds from the entire GC-0 trial were categorized
into eight groups based on similar aroma description. These eight groups were 1)
citrusy/minty; 2) metallic/mushroom/geranium; 3) roasted/cooked/meaty/spice; 4)
fatty/soapy/green; 5) sulfury/solventy/medicine; 6) floral; 7) sweet/fruity; and 8)
green/grassy. The sums of the total olfactometry intensities for each aroma group was
determined and presented in spider web (radar graph) for each of the four juice types.
The contribution of norisoprenoids to orange juice was calculated from the total intensity
of norisoprenoids to the total intensity of all aroma active volatiles in the juice.
Results and Discussion
Quantification of Norisoprenoids in Orange Juice
The amount of volatile compounds found on the SPME coating depends on
exposure time, temperature, sample volume, headspace volume, and sample
concentration. In this study only exposure time was varied in order to determine the time
needed for equilibrium concentrations for each analyte to be established. All other
factors remained constant. Equilibrium time between SPME fiber and headspace of

48
fortified juices was indicated when little to no increase in peak area was observed with
additional exposure time. The equilibrium time for P-cyclocitral, P-damascenone,
a-ionone, and P-ionone were 75, 90, 115, and 120 min. respectively (Fig. 5-1). The
Figure 5-1. Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 40C, = P-cyclocitral, = P-damascenone, A= a-ionone,
= P-ionone.
results show that the time needed to reach equilibrium depends on the polarity and the
relative molecular mass of each norisoprenoid. Since 75-120 minutes to reach the
equilibrium would be too long to wait for practical purposes and may alter the volatile
profiles from thermally induced reactions, a shorter exposure time was chosen for routine
analyses. It can be seen from the adsorption curves for each compound that 45 min.
represents a rough compromise for all four analytes between minimal exposure time and
maximum peak area. For example P-damascenone and P-ionone reaching more than
80% of their final equilibrium value within 45 min. It is a rare SPME analysis that
employs true equilibrium exposure time. If exposure time can be carefully controlled,
then exposure times of as little as 5 min. can be employed. These very short exposure

49
times are usually limited to analytes in relatively high concentration and even then the
reproducibility is not that good. In this study, very short exposure times were not an
option as the analytes of interest were present in very low concentrations.
The reproducibility (analytical precision) of a fortified juice using SPME-GC-FTD
was determined in five replicates at 40C with 45 min exposure. The relative standard
deviations (RSD) obtained were 1.7, 1.7, 0.4, 1.4 % for P-cyclocitral, P-damascenone, a-
ionone, and P-ionone respectively (Table 5-1). It should be kept in mind that the orange
juice had been fortified with 8.6, 4, 5.4, and 5.27 ppm P-cyclocitral, P-damascenone, a-
ionone, and P-ionone respectively. The low RSD indicated that the SPME and GC
analytical conditions employed in this study could quantify norisoprenoids in orange
Table 5-1. Reproducibility of SPME exposure time 45 min at 40C
Replicate
P-cyclocitral
P-damascenone
a-ionone
P-ionone
1
84101
27181
46618
23072
2
84149
26797
46715
22861
3
84129
26534
46813
22589
4
81092
26382
46570
22337
5
82037
25961
46319
22335
Average
83101
26571
46607
22639
STD Va
1443
456
186
325
RSDb
1.7
1.7
0.4
1.4
Standard deviation,b relative standard deviation
juice in a highly reproducible manner. However, it should be pointed out the
concentrations used to fortify the sample were considerably higher than would ever be
found in an orange juice sample. Typical juice concentrations are 50 to 1000 times lower
so that typical RSDs for unfortified juice samples range from 20- 50% which might
seem high, but still very acceptable for analyses at the sub p,g/L level the complex matrix
of orange juice. The volatility of flavor compounds can be changed according to the

50
sample matrices. Boa et al.(92) reported that the combination of SPME with the standard
addition method reduce the problem of matrix effects and improved the precision of the
procedure.
Norisoprenoid Quantification using Standard Additions
The norisoprenoids P-cyclocitral, (3-damascenone, a-ionone, and p-ionone were
quantified in orange juice using the standard addition method (Fig. 5-2, 5-3, 5-4, 5-5, and
5-6). The integrate peak areas at specific m/z 137, 177, 177 and 190 for P-cyclocitral,
Figure 5-2. Standard addition data for P-cyclocitral peak area vs. added concentration in
fresh orange juice. Regression line calculated from peak area at selected mass
137.
P-ionone, a-ionone, and p-damascenone respectively were plotted versus the
concentration of the spiked standards. The amount of each norisoprenoid (Table 5-2 and
Table 5-3) was calculated from the regression equation where the calculated value was
determined at y = 0.
As seen from the plots of Fig. 5-2 through Fig. 5-6, the correlation coefficients for
the standard addition data was at least 0.99 in all cases except for P-cyclocitral (Fig. 5-2)

51
where it was 0.86. One way analysis of variance (ANOVA) show that there are a highly
significant differences (P<0.01) among orange juice spiked with different concentration
of standard a-ionone (Fig. 5-3) and P-ionone (Fig. 5-4). However there were no
significant differences within sample. In contrast for P-cyclocitral (Fig. 5-2), there were
no significant differences between the two samples spiked with standard P-cyclocitral at
concentration 0.54 and 1.1 ppm. This suggests that an error occurred during analysis
with at least one data point.
A quadrupole MS provides at least lOx greater sensitivity in the SIM mode than an
ion trap MS under the same conditions and was thus used to quantify P-damascenone in
fresh and pasteurized juice when the ion trap failed to detect this compound. The
calculated slope from the fresh juice data was also employed to determine the
concentration of P-damascenone in pasteurized juice as it was thought the matrix effects
would be the same for both samples.
Figure 5-3. Standard addition data for a-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. a-ionone concentration.

Peak area
52
Figure 5-4. Standard addition data for (3-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. P-ionone concentration
P-damascenone concentration (ppm)
Figure 5-5. Standard addition P-damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190.

53
Figure 5-6. Standard addition data of P-damascenone peak area vs. added concentration
in reconstituted from concentrate orange juice. GC- quadrupole mass
spectrometer in SIM mode at m/z 190.
The concentration of (3-damascenone in reconstituted from concentrate was calculated
from separate standard addition data (Fig. 5-6) as it was thought that the matrix would be
substantially different.
Table 5-2. Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique
Norisoprenoids
Concentration (pg/L)
Threshold (pg/L in water)3
OAV
(3-cyclocitral
145
5
25
P-damascenone
0.09
0.002
45
a-ionone
47
0.4
118
P-ionone
83
0.007
11857
a Buttery and Teranishi (50)
Table 5-3. Concentration of P-damascenone in fresh, pasteurized and reconstituted
concentrate
orange juice
concentration (pg/L)
OAV
Fresh
0.09
45
Pasteurized
0.18
90
Reconstituted
0.85
425

54
The calculated concentrations of P-cyclocitral, P-damascenone, a-ionone and
p-ionone in fresh orange juice were 145, 0.09, 47, and 83 pg/L respectively (Table 5-2).
The aroma active compounds in orange juice have been studied by GC-0 methods (4-7).
Only P-ionone was reported (4, 5) but has not been reported the concentration of this
volatile. The concentration of P-damascenone in 3 types of orange juice: fresh,
pasteurized, reconstituted from concentrate were 0.09, 0.18, and 0.85 pg/L respectively
(Table 5-3). This data suggest that there is precursors present in juice and generate P-
damascenone during thermal processing. These precursors are probably carotenoids like
neoxanthin, but could also be glycosided forms of P-damascenone. These precursors can
generate aroma volatiles in foods that have undergone thermal processing as reported for
tomato paste (23) and heated apple juice (9). It has been previously reported that citrus
juice pulp and cloud (insoluble solids) can retain considerable volatiles (93, 94).
Therefore P-damascenone may have been trapped in the pulp during thermal
concentration and might not be completely removed during evaporation. Its partial loss
may also been partially compensated by newly P-damascenone generated from thermally
unstable carotenoids during thermal concentration.
Determination of Relative Aroma Impact of Norisoprenoids
The odor activity value (OAV) is a rough way of determining relative aroma
contribution of various substances. It is determined by dividing the analytical
concentration by the aroma threshold. The OAV of P-cyclocitral, P-damascenone,
a-ionone and P-ionone were 25, 45, 118, and 11857 respectively (Table 5-2). The OAV
value shows that P-ionone is predicted to have the greatest contribution compared to the
other norisoprenoids. Hinterholzer and Schieberle, (4) analyzed the volatiles from orange

55
juice by solvent extraction and determined the aroma contribution by aroma extract
dilution analysis (AEDA). The value from AEDA was recorded as flavor dilution (FD)
factor (the highest dilution factor of the particular aroma active compounds which can be
perceived by human nose). The most odor active compound by this method was ethyl
butanoate (FD 1024) but the FD factor of P-ionone was onlyl. These same authors (5)
quantified twenty-five odor active compounds by stable isotope dilution assay and
estimated their respective odor contributions by OAV values. Unfortunately they did not
quantify (3-ionone. The OAV of ethyl butanoate (the compound with the highest dilution
value from AEDA (4)) was 1192 (concentration 1192 pg/L, odor threshold 1 pg/L). If
one compares this OAV value with the OAV of P-ionone in present study (11857),
P-ionone could be the most aroma active compound in orange juice. The apparent
conflict in the two sets of data suggests that P-ionone may not have been well extracted in
the AEDA study.
Table 5-4. Aroma active compounds in orange juice grouped by citrusy/minty
Compounds3
Description
L
RI
Relative
intensity
ZB-5
DB-wax
Unknown
Orange peel
963
8C, 8d
1,8 cineole
Minty, camphor
1026
1232
5C, 6d
Nonanal
Orange peel, soapy
1090
1398
6C, 6d
3-mercapto hexan-l-ol
Grapefruit
1121
7C, 7d
Citronellal
Minty,camphor
1160
1489
5C, 7d, 5e
Nerol
Lemomgrass
1222
1798
5C, 5d
Neralb
Lemomgrass
1236
1692
7C, 7d
L-carvoneb
Minty
1242
1747
8C, 8d
Geraniol
Citrusy,geranium
1265
1853
9C, 7d, 4e
1 -p-menthene-8-thiol
Grapefruit
1281
1619
7C, 7d, 5e
Geranial
Green,minty
1310
1742
4c,4d, 4e
Nootkatoneb
Sweet,sour, grapefruit
1824
7C, 5d

56
Table 5-5. Aroma active compounds in orange juice grouped by
metallic/mushroom/geranium i
Compounds3
Description
LI
RI
Relative
intensity
ZB-5
DB-
wax
l-octene-3-one
Metallic, mushroom
974
1308
6C, 7d, 5e
(3-myrcene
Geranium,plastic
979
1163
r, 7d, 8e
Octanalb
Metallic, orange peel
998
1299
8C, 8d, 8e
E-2-octenal
Metallic, fatty, green
1052
1449
4d
Terpinoleneb
Metallic, citrusy
1070
1296
6c,5d
Unknown
Green, metallic
1100
6d
Unknown
Metallic, pungent
1128
5d
Z-2-nonenal
Green, metallic
1141
1515
4C, 6d, 5e
Terpinen-4-olb
Metallic, musty
1175
1619
5C, 6d
Unknown
Metallic, woody
1247
6C, 6d
(E,Z)-2,4-decadienal
Metallic, geranium
1293
1759
4C, 7d
4,5-epoxy-E-2-decenal
Metallic, fatty
1375
2010
6c,6d, 4e
Unknown
Aquarium, metallic
1589
5C, 6d
(3-sinensal
Aquarium
1698
2244
o
OO
u
OO
Table 5-6. Aroma active compounds in orange juice grouped by
roasted/cooked/meaty/spice
Compounds3
Description
L
RI
Relative
intensity
ZB-5
DB-
wax
Methional
Cooked potato
904
1464
8e, 8d, T
2-acetyl-2-thiazoline
Cooked jusmine rice
1104
1766
6C, 6d, T
Unknown
Spice
1317
6d
Unknown
Sweet, nutty
1380
7C, 8d
Unknown
Fermented, rancid
1459
5C, 7d
Unknown
Green, overipe orange
1461
4d,4e
Unknown
Nutty
1510
8C, 9d
Unknown
spice
1718
7C, 7d
Table 5-7. Aroma active compounds in orange juice grouped by fatty/soapy/green
Compounds3
Description
LRI
Relative
intensity
ZB-5
DB-
wax
Hexanal6
Green, fatty
794
1083
7C, 6d, T
1-octanol
Green, soapy
1065
1565
8C, 7d, 5e
E-2-nonenal
Soapy
1153
1542
8C, 10d,6e
(Z)-4-decenal
Green, metallic, soapy
1188
1542
7C, 7d
Decanalb
Green, soapy
1198
1508
7C, 8d, 5e
(E,E)-2,4-nonadienal
Fatty, green
1209
1702
T, 7d
Unknown
Soapy, almond
1274
T, 7d
(E,E)-2,4-decadienal
Fatty, green
1314
1819
4c,6d
Dodecanal
Soapy
1403
1722
5C, 6d

57
Table 5-8. Aroma active compounds in orange juice grouped by
sulfury/solventy/medicine
Compounds2
Description
LI
ill
Relative
intensity
ZB-5
DB-
wax
Acetaldehyde
Fresh alcohol
445
732
6C, 6d
Carbon disulfide
Sulfur, fermented cabbage
678
6C, 7d, 6e
Dimethyl sulfide
Solventy, plastic
691
6C, 6d, 4e
Dimethyl disulfide
Plastic
772
1074
4C, 4d, T
Unknown
Fermented, sulfur
818
5C, 6d
2-methyl-3-furanthiol
Meaty, vitamin, medicine
865
1305
7C, 7d, 6e
4-mercapto-4-methyl-2-pentanone
Sulfury, grapefruit
944
1389
7C, 5d
Dimethyl trisulfide
Sulfur, sweaty
968
1392
3C, 5d,8e
4-mercapto-4-methyl-2-pentanol
Sweaty .grapefruit,guava
1039
7C, 7d, T
Unknown
Solventy
1167
5d
Dimethyl tetrasulfide
Sulfury, musty
1225
6e
Table 5-9. Aroma active compounds in orange juice grouped by floral
Compounds2
Description
LI
RI
Relative
intensity
ZB-5
DB-
wax
Linaloolb
Floral
1094
1551
8C, 8d, 5e
(3-cyclocitralb
Mild floral, hay-like
1214
1632
6C, 4d
Unknown
Tobacco,sweet, floral
1255
6e
(3-damascenoneb
Tabacco, apple, floral
1383
1829
7C, 8d, 8e
a-iononeb
Floral
1426
1863
OO
o
00
o.
P-iononeb
Floral, raspberry
1484
1951
8C, 9d, T
Table 5-10. Aroma active compounds in orange juice grouped by sweet/fruity
Compounds2
Description
L
RI
Relative
intensity
ZB-5
DB-
wax
Ethyl-2-methylpropanoate
Sweet, fruity
758
966
6C, 6d
Ethyl butyrateb
Sweet, fruity
795
1034
4C, 6d, 8e
Ethyl-2-methylbutyrate
Sweet, fruity
846
1051
5C, 6d
Ethyl hexanoateb
Sweet
994
1242
6C, 7d
a-terpinyl acetate
Sweet
1349
1663
6c,6d

58
Table 5-11. Aroma active compounds in orange juice grouped by green/grassy
Compounds11
Description
L
RI
Relative
intensity
ZB-5
DB-wax
3-(Z)-hexen-l-ol / (E)-2-hexenalb
(E,Z)-2,6-nonadienal
Green, grass
Green
854
1148
1226/1391
1593
6C, T
r, 8d, r
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard
b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard,
and MS
c fresh orange juice
d pasteurized orange juice
6 reconstituted from concentrate
1. Citrusy/minty
Figure 5-7. Aroma group profiles of fresh (), pasteurized (), and reconstituted from
concentrate () orange juice.
Norisoprenoid Contribution to Total Floral Aroma
All four of the identified norisoprenoids in orange juice have a general floral
aroma quality (Table 5-9). When all the relative intensities of these four norisoprenoids
were combined (in each type of orange juice) and compared with the total aroma

59
intensity, their contribution to total juice aroma were 7.8, 7.6, and 8.7 % in fresh,
pasteurized, and reconstituted from concentrate respectively. Fig. 5-7 illustrates aroma
group profiles from the total intensity in similar aroma quality groups from GC-0
aromagrams, comparing among fresh, pasteurized, and reconstituted from concentrate.
Comparing fresh and pasteurized orange juice, the major aroma category differences
were in the metallic/mushroom/geranium and in roasted/cooked/meaty/spice categories
which were higher in pasteurized juice than fresh juice. This difference come from three
addition aroma active compounds (E-2-octenal, and two unknown at LRI 1100 and LRI
1128) in metallic/mushroom/geranium category and two additional unknown ( LRI 1317
and LRI 1461) in roasted/cooked/meaty /spice category in pasteurized juice (Table 5-5
and Table 5-6). In fresh juice citrusy/minty category was higher than pasteurized juice.
The differences were the higher intensity of compounds in citrusy/minty category in fresh
juice than in pasteurized juice. The contribution of floral quality was the same in fresh
and pasteurized orange juice. From the aroma group profile of reconstituted from
concentrate there were many volatile compounds lost due to the thermal evaporation
process.
Fig. 5-8 shows the contribution of aroma active compounds in just the floral
category. This group includes linalool, p-cyclocitral, P-damascenone, a-ionone,
P-ionone and one unknown (LRI 1255) generated after thermal processing. The
norisoprenoids in orange juice contribute the majority of floral aroma in floral category,
specifically 78, 78, and 59% in fresh, pasteurized and reconstituted from concentrate
respectively.

60
40 -
35 -
30 -
£
o
Fresh Pasteurized Pumpout
Orange juice
Figure 5-8. Upper bar norisoprenoids contribute mainly to the total floral category, fresh
= 78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent non-
norisoprenoids including linalool and unknown (LRI = 1255) generated
during thermal processing.
Table 5-12. Norisoprenoids in orange juice and peel oil
Norisoprenoids
Fresh
Pasteurized
Reconstituted
concentrate
Hand
squeezed
Peel oil
P-cyclocitral
X
X
X
X
P-damascenone
X
X
X
X
X
a-ionone
X
X
X
P-ionone
X
X
X
X
X = indicates presence of norisoprenoids in the various samp
es.
Four norisoprenoids, P-cyclocitral, P-damascenone, a-ionone, and P-ionone were
detected in both fresh and pasteurized juice (Table 5-12). Only two norisoprenoids P-
damascenone and P-ionone were detected in reconstituted from concentrate, indicating
that these two compounds could be generated from precursors during thermal evaporation
and/or they were retained by the pulp during the evaporation process.

61
Conclusion
Concentrations of four orange juice norisoprenoids were determined using SPME
with the standard addition method. The concentrations of P-cyclocitral, P-damascenone,
a-ionone, and P-ionone in fresh orange juice were 145, 0.09, 47, and 83 p.g/L
respectively. The OAV (determined by dividing the analytical concentration by the
aroma threshold) of p-cyclocitral, p-damascenone, a-ionone, and p-ionone were 25, 45,
118, and 11857 respectively. The OAV values suggest that P-ionone provides the
greatest aroma contribution compared to the other norisoprenoids. The concentration of
p-damascenone increaded with thermal processing, indicating that there are precursors in
juice which generate P-damascenone during elevated temperatures. Combined, the four
norisoprenoids contribute 8-10% of the total aroma impact. The norisoprenoids have a
general floral character and contribute the majority (60-80%) of the floral character to
orange juice.

CHAPTER 6
THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL SOLUTION
Introduction
Carotenoids are unstable in both the presence of heat and/or light. The thermal
degradation of carotenoids produces a range of volatile products and norisoprenoids are
the most potent aroma compounds of all the volatiles produced. The formation of
specific norisoprenoids from the thermal degradation of carotenoids during heat treatment
of food products have been reported. Beta-ionone, a-ionone, and (3-damascenone have
been reported in tomato paste (23) and black tea (95). These norisoprenoids also were
detected from the thermal degradation of carotenoids in model systems such as (3-
carotene in water at 97C, 3 hrs (31), 1% solution of p-carotene heated at 188F for 72
hrs. (96), and thermal degradation of crystallize P-carotene at 240F in a vaccum (97).
However the norisoprenoids formed in those model systems were formed at high
temperature. At best these studies could be considered accelerated storage studies. In
order to have a model more representative of the conditions that a real world juice
might be exposed to, a 35C storage study was carried out. The model solution was
buffered to pH 3.8 (orange juice pH) using citric acid and tripotassium citrate. Sugars
and amino acids were not added to reduce the possibility that they could be possible
norisoprenoid sources. The concept that carotenoids could act as a source of
norisoprenoids is relatively new (15, 39, 69, 70). These studies have indicated that
specific carotenoids need be present in order to produce specific norisoprenoids.
62

63
Therefore juice carotenoids may be the source of some aromas due to thermal
degradation during processing and subsequence storage.
Twenty-four carotenoids from orange juice were isolated in present study (i.e.,
neoxanthin, P-carotene. a-carotene, and P-cryptoxanthin, see chapter 3). One of the
carotenoids isolated in orange juice, p-carotene, was studied in model aqueous solution to
determine which aroma active compounds could be produced via thermal degradation
and thus unequivocally demonstrate that a carotenoid found in orange juice can act as a
precursor of norisoprenoids. Beta-carotene was chosen because it was commercially
available, relatively inexpensive and should be the most unstable as it cannot be esterified
to improve thermal stability.
Objective
Determine if aroma active norisoprenoids are generated from P-carotene via
thermal degradation using model solutions adjusted to orange juice pH (3.8) and stored at
35C. (Objective 5)
Materials and Methods
Crystallization
Beta-carotene (99% purity, purchase from Acros) was recrystallized before using to
remove aroma active impurities. The method of recrystallization followed that of Schiedt
and Liaaen-Jensen {98) with minor modification. Beta-carotene was dissolved in the
smallest possible volume of petroleum ether (Fisher Scientific, NJ), filtered through glass
wool in a funnel, and ethanol (Fisher Scientific, NJ) was added drop-wise until turbidity
was observed. The mixture was left at room temperature for about an hour and the
temperature was then lowered gradually to 6C (refrigerator) and finally to -20C (deep
freeze) over night or until the crystals formed. The crystals were collected on a fine

64
sintered-glass, washed on the filter with cold ethanol and dried with the flow of nitrogen
gas. The headspace volatiles of recrystallized P-carotene were checked by SPME before
using. It was ready to use when no aroma active volatiles were detected.
Model Solutions
Acetone (Fisher Scientific, NJ) was chosen to dissolve the recrystallized P-carotene
because it was a polar solvent and would facilitate the transfer of P-carotene into the
model aqueous solution. One milligram of recrystallized P-carotene was dissolved in
acetone and diluted to citrate buffer pH 3.8 (citric acid 1.2 g., tripotassium citrate 0.6 g.
adjust pH to 3.8 by 1 N. NaOH, (Fisher Scientific, NJ)). Ten milliliters of the solution
were added into 40 ml vial with Teflon coated screw cap and wrapped with aluminum
foil kept in 35C for up to I month.
Analytical Methods
The headspace volatiles of the model solution were extracted by Solid Phase
Microextraction (SPME,50/30pm DVB/Carboxen/PDMS, Supelco). The solution was
equilibrated at 40C with gentle agitation (by stirring bar) for 45 min and then inserted
the SPME fiber to the headspace of the model solution in order to extract and concentrate
the headspace volatile by the fiber for another 45 min. The fiber was injected to GC (A
HP-5890 GC (Palo Alto, CA) with either a DB-Wax or ZB-5 column whose effluent was
split between an olfactometer or flame ionization detector (FID). Column oven
temperature was programmed from 40 to 240C at 7 C/min with a 5 min hold. The
aroma active compounds detected by GC-0 were identified from their aroma quality and
retention index by comparison with standards and confirmed by GC-MS (as describer in
chapter 4).

65
Results and discussion
Before beginning storage study, the high purity P-carotene (99% purity) was
evaluated for aroma active impurities using GC-0 of the material in the same manner as
the storage study. This demonstrated the potency of very minor impurities (less than 1%)
and the need to recrystallize the standard P-carotene to remove aroma active impurity
before beginning the storage study (Fig. 6-1). No effort was made to identify these
impurities only to remove them. Freshly recrystallized P-carotene was used in all model
solution storage studies. Before storage, a day 0 (control) was examined using GC-0 to
make certain no aroma active volatiles were detected (Fig. 6-2).
10 15 20 25
Time (min)
Figure 6-1. The standard P-carotene (99% purity) as received (no purification)

Olfactory Response FID Response S Olfactory Response FID Response
66
10 15 20 25
Time (min)
Headspace volatiles from 3-carotene in model solution pH 3.5 at 0 day
10 15 20 25
Time (min)
Figure 6-3. Headspace volatiles from p-carotene in model solution pH 3.5 after storage 1
day at 35C: 1 = P-ionone, a = sweet/raspberry

67
After one day of storage at 35C, only a single aroma active compound, P-ionone,
was detected (Fig. 6-3). This suggests that p-ionone is either one of the most common
thermal decomposition products and/or it has one of the lowest aroma thresholds. Beta-
ionone does have one of the lowest aroma thresholds (see Table 5-2), but is probably also
a common decomposition fragment and as shown in Fig. 6-3 might represent a certain
weakness in the C9-C10 double bond.
Figure 6-4. Degradation of P-carotene in model solution at difference carbon bonds
After two weeks storage, five distinct aroma active peaks were observed. Four of
these appeared to correspond to distinct FED peaks. It is interesting to note that P-ionone
is still the highest peak and that all of the predicted decomposition products shown in Fig.
6-4 were observed (e.g., peak 1 was due to P-cyclocitral, peak 2 was due to
P-homocyclocitral). It appears that oxidative degradation of p-carotene at double bond
C9-C10, is the most preferable and P-ionone was reported as the major product from
P-carotene degradation (32, 99). Beta-ionone was reported as an off-flavor of dehydrated
carrot stored in oxygen. When dehydrated carrot was stored in the presence of oxygen its

68
color, due to P-carotene, was destroyed and simultaneously an off-flavor (violet-like)
developed {24).
The polyene-carotenes are apparently oxidized at the first conjugated diene bonds.
Oxidation is more prevalent adjacent to the methyl group and it is possible that inductive
effects of the methyl group make the double bond adjacent to it more susceptible to
oxidation {100). When P-carotene in benzene or tetrachloromethane is allowed to react
with molecular oxygen in the absence of light at 30C, P-ionone can be formed within the
first few hours. As the oxidation progressed a number of shorter chain products are
formed, including P-cyclocitral {101). The aroma active volatile compounds formed
indicates that oxidative scission of P-carotene can occur at carbon bond C7-C8, C8-C9 and
C9-C10 of p-carotene to generate Co: P -cyclocitral, Cn: P-homocyclocitral, C13:
P-damascone and C13: P-ionone respectively (see figure 6-4). The same degradation
position and volatile compounds (except p-homocyclocitral) have been reported by
photo-oxygenation {102) autoxidation with molecular oxygen at 30C in the dark
conditions {101) and oxidation in water at 97C {31). At the scission Cs-Cg of
p-carotene, P-homocyclocitral (Cn) was formed in present model condition but at the
same scission position, dihydroactinodiolide (Cn) was formed and has been reported in
the different model conditions {31, 101, 102).
GC-0 Analysis of P-Carotene Decomposition at 35C
The GC-0 data for the two week storage sample is in Table 6-1. It should be noted
that the same retention and aroma descriptors observed earlier for standard and juices
were also observed for this storage sample. It is interesting to see just how similar the
values in Table 6-1 are with their corresponding components in Table 5-9. In making this

69
10 15 20 25
Time (min)
Figure 6-5. Headspace volatiles from (3-carotene in model solution pH 3.5, after storage
2 weeks at 35C : 1 = (3-cyclocitral, 2 = (3-homocyclocitral, 3 = P-damascone,
4 = unknown, 5 = (3-ionone, a = sweet/floral/hay-like, b = sweet/floral/hay
like, c = sweet/apple, d = sweet/raspberry, e = sweet.
Table 6-1. Aroma active compounds from (3-carotene thermal degradation in model
solution pH 3.8, storage at 35C for 2 weeks
Compounds
Aroma description
LI
RI
MS
ZB-5
DB-wax
. P-cyclocitral
Sweet, floral, hay-like
1228
1632
X
p-homocyclocitral
Sweet, floral, hay-like
1262
1780
X
, P-damascone
Sweet, floral
1425
1835
. P-ionone
Sweet, raspberry
1495
1960
X
comparison, it will be noted that neither (3-homocyclocitral or P-damascone was found in
orange juice. Their retention characteristics and aroma descriptors exactly matched that
of authentic standards, providing enough evidence for at least a tentative identification.
Positive identification of these compounds was achieved from the MS data. The MS
fragmentation patterns for the identified norisoprenoids are shown in the following
figures which should offer conclusive proof as to their identity.

Relative Abundance
70
MS Identification
100=
8CT
60
40
20J
2913
19.62
18,13 1B8.3 19.3 100a
ni/z = 137
500:
24.30
m/z = 151
25.13
m/z = 177
i. i????,.y?fiy
18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0
Time (min)
Figure 6-6. Selected ion chromatogram (SIC) of model solution headspace volatiles after
storage 2 weeks at 35C : A = (3-cyclocitral, B = P-homocyclocitral, C = (3-
ionone.
m/z
Figure 6-7. Upper spectra from model solution MS at RT 19.61, bottom spectra from
standard P-cyclocitral using identical ion trap MS at identical retention time.

Relative Abundance
71
m/z
Figure 6-8. Upper spectra from model solution MS at RT 20.46, bottom spectra from
standard P-homocyclocitral using identical ion trap MS at identical retention
time
Figure 6-9. Upper spectra from model solution MS at RT 25.13, bottom spectra from
standard (3-ionone using identical ion trap MS at identical retention time.

72
Conclusion
The norisoprenoids detected in this study were also detected in other foods that
have been thermally processed e.g.,tea (95), tomato paste (50). Therefore the results of
this study indicate that (3-carotene could be a precursor of (3-cyclocitral and P-ionone in
orange juice. Furthermore it could be a precursor of norisoprenoids during thermal
processing and subsequence storage at relative high temperature and it is reasonable to
assume that any appreciable change in carotenoids content of orange juice will have an
effect on flavor.

CHAPTER 7
CONCLUSIONS
Four norisoprenoids, P-cyclocitral, p-damascenone, a-ionone and P-ionone were
identified in orange juice using headspace SPME, GC-O, GC-FID and GC-MS. Three of
them, P-cyclocitral, P-damascenone, a-ionone were identified and confirmed by GC-MS
for the first time. Their concentrations in fresh orange juice were determined using
SPME with standard addition technique. Odor activity values (OAV) were calculated
using published threshold values. Calculated OAV values suggest that P-ionone provided
the greatest contribution to total floral aroma in orange juice compared to the other three
norisoprenoids. The concentration of P-damascenone increased almost 10 fold after
thermal processing, indicating there are thermally unstable precursors which generate
P-damascenone at elevated temperatures. All four of the norisoprenoids in orange juice
contribute 8-10% floral aroma to the total aroma quality of the orange juice and are the
major contributors (60-80%) in the floral category.
Several carotenoids were identified using HPLC with photodiode array detection.
Twenty-four carotenoids were separated as distinct peaks and sixteen of these peaks were
identified based on their spectral characteristics, relative elution order compared to
literature values and authentic standards. The identified carotenoids include: a-carotene,
P-carotene, a-cryptoxanthin, P-cryptoxanthin and neoxanthin, which are known as
norisoprenoid precursors. These specific carotenoids were of interest because they
73

74
possess the direct structural segments needed to serve as precursors to the newly
identified norisoprenoids.
To demonstrate that carotenoids could serve as norisoprenoids precursors,
(3-carotene was studied in a model system at 35C storage. GC-0 and GC-MS data
confirmed the presence of (3-cyclocitral and (3-ionone in these solutions in as little as two
weeks. This was direct proof that (3-carotene can degrade to form specific norisoprenoids
under conditions an orange juice might encounter.

LIST OF REFERENCES
1 Moshonas, M. G.; Shaw, P. E. Quantitative analysis of orange juice flavor
volatiles by direct-injection gas chromatography. J. Agrie. Food Chem. 1987, 35,
161-165. *
2 Moshonas, M. G.; Shaw, P. E. Quantitative determination of 46 volatile
constituents in fresh, unpasteurized orange juices using dynamic headspace gas
chromatography. J. Agrie. Food Chem. 1994, 42, 1525-1528.
3 Tonder, D.; Petersen, M. A.; Poll, L.; Olsen, C. E. Discrimination between freshly
made and stored reconstituted orange juice using GC odor profiling and aroma
values. Food Chem. 1998, 61, 223-229.
4 Hinterholzer, A.; Schieberle, P. Identification of the most odor-active volatiles in
fresh, hand-extracted juice of Valencia late oranges by odor dilution techniques.
Flavour Frag. J. 1998, 13, 49-55.
5 Buettner, A.; Schieberle, P. Evaluation of aroma differences between hand-
squeezed juices from Valencia late and Navel oranges by quantitation of key
odorants and flavor reconstitution experiments. J. Agrie. Food Chem. 2001, 49,
2387-2394.
6 Bazemore, R.; Goodner, K.; Rouseff, R. Volatiles from unpasteurized and
excessively heated orange juice analyzed with solid phase microextraction and
GC-olfactometry. J. Food Sci. 1999, 64, 800-803.
7 Marin, A. B.; Aeree, T. E.; Hotchkiss, J. H.; Nagy, S. Gas chromatography-
olfactometry of orange juice to assess the effects of plastic polymers on aroma
character. J. Agrie. Food Chem. 1992, 40, 650-654.
8 Kotseridis, Y.; Baumes, R. Identification of impact odorants in bordeaux red
grape juice, in the commercial yeast used for its fermentation, and in the produced
wine. J. Agrie. Food Chem. 2000, 48, 400-406.
9 Zhou, P. G.; Cox, J. A.; Roberts, D. D.; Aeree, T. E. Beta-Damascenone
precursors in apples. Prog. Flavour Precursor Stud. Proc. Int. Conf. 1993, 261-
273.
10 Ong, P. K. C.; Aeree, T. E. Gas chromatography/olfactory analysis of Lychee
(Litchi chinensis Sonn.). J. Agrie. Food Chem. 1998, 46, 2282-2286.
75

11
12
13
14
15
16
17
18
19
20
21
22
23
76
MacLeod, G.; Ames, J. M. Volatile components of starfruit. Phytochemistry 1990,
29, 165-172.
Olle, D.; Baron, A.; Lozano, Y. F.; Sznaper, C.; Baumes, R.; Bayonove, C.;
Brillouet, J. M. Microfiltration and reverse osmosis affect recovery of mango
puree flavor compounds. J. Food Sci. 1997, <52, 1116-1119.
Ishida, B. K.; Mahoney, N. E.; Ling, L. C. Increased lycopene and flavor volatile
production in tomato calyces and fruit cultured in vitro and the effect of 2-(4-
Chlorophenylthio)triethylamine. J. Agrie. Food Chem. 1998, 46, 4577-4582.
Tarantilis, P. A.; Polissiou, M. G. Isolation and identification of the aroma
components from saffron (Crocus sativus). J. Agrie. Food Chem. 1997, 45, 459-
462.
Enzell, C. Biodegradation of carotenoids an important route to aroma
compounds. Pure Appl. Chem. 1985, 57, 693-700.
Sanderson, G. W.; Co, H.; Gonzalez, J. G. Biochemistry of tea fermentation: role
of carotenes in black tea aroma formation. J. Food Sci. 1971, 36, 231-236.
Gross, J.; Gabai, M.; Lifshitz, A. Comparative study of the carotenoid pigments in
juice of Shamouti, Valencia, and Washington oranges, three varieties of Citrus
sinensis. Phytochemistry 1972, 11, 303-308.
Isoe, S.; Hyeon, S. B.; Sakan, T. Photo-oxygenation of carotenoids. I. Formation
of dihydroactinidiolide and (3-ionone from P-carotene. Tetrahedron Lett. 1969,
279-281.
Demole, E.; Enggist, P.; Winter, M.; Furrer, A.; Schulte-Elte, K. H.; Egger, B.;
Ohloff, G. Megastigma-5,8-dien-4-one, an aroma constituent of the yellow
passion fruit and Virginia tobacco. Helv. Chim. Acta 1979, 62, 67-75.
Strauss, C. R.; Wilson, B.; Anderson, R.; Williams, P. J. Development of
precursors of C13 nor-isoprenoid flavorants in Riesling grapes. Am. J. Enol. Vitic.
1987, 38, 23-27.
Williams, P. J.; Sefton, M. A.; Francis, I. L. Glycosidic Precursors of Varietal
Grape and Wine Flavor. In Flavor Precursors: Thermal and Enzymatic
Conversions', R. Teranishi; G. R. Takeoka and M. Guntert, Eds.; American
Chemical Society: Washington, DC, 1992; pp 74-86.
Enzell, C. R. Influence of Curing on the Formation of Tobacco Flavor. In Flavour
'81; P. Schreier, Ed.; de Gruyter: Berlin, New York, 1981; pp 449-478.
Buttery, R. G.; Teranishi, R.; Flath, R. A.; Ling, L. C. Identification of additional
tomato paste volatiles. J. Agrie. Food Chem. 1990, 38, 792-795.

77
24 Ayers, J. E.; Fishwick, M. J.; Land, D. G.; Swain, T. Off-flavor of dehydrated
carrot stored in oxygen. Nature 1964, 203, 81-82.
25 Vemin, G. Terpenoids and norisoprenoids in grape berry aromas and in wines. 1.
Origins and formation. Rivista Italiana EPPOS 1997, 8, 19-31.
26 Gross, J. Pigments in Fruits', Academic Press Inc.: London, 1987; 303 pp.
27 Weedon, B. C. L.; Moss, G. P. Structure and nomenclature. In Carotenoids', G.
Britton; S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel Boston
Berlin, 1995; pp 27-70.
28 Mordi, R. C. Mechanism of (3-carotene degradation. Biochem. J. 1993, 292, 310-
312.
29 Lutz-Roder, A.; Jezussek, M.; Winterhalter, P. Nickel peroxide induced oxidation
of canthaxanthin. J. Agrie. Food Chem. 1999, 47, 1887-1891.
30 Ouyang, J. M.; Daun, H.; Chang, S. S.; Ho, C. T. Formation of carbonyl
compounds from (3-carotene during palm oil deodorization. J. Food Sci. 1980, 45,
1214-1217.
31 Kanasawud, P.; Crouzet, J. C. Mechanism of formation of volatile compounds by
thermal degradation of carotenoids in aqueous medium. 1. (3-Carotene
degradation. J. Agrie. Food Chem. 1990, 38, 237-243.
32 Winterhalter, P. Carotenoid-Derived Aroma Compounds: Biogenetic and
Biotechnological Aspects. In Biotechnology for Improved Foods and Flavors; G.
R. Takeoka; R. Teranishi; P. J. Williams and A. Kobayashi, Eds.; American
Chemical Society: Washington, DC, 1996; pp 295-308.
33 Weeks, W. W. Carotenoids: A Source of Flavor and Aroma. In Biogeneration of
Aromas; H. P. Thomas and C. Rodney, Eds.; American Chemical Society:
Washington, D.C., 1986; pp 157-166.
34 Winterhalter, P.; Rouseff, R. Carotenoid-Derived Aroma Compounds: an
Introduction. In Carotenoid-Derived Aroma Compounds; P. Winterhalter and R.
Rouseff, Eds.; American Chemical Society: Washington, DC, 2002; pp 1-17.
35 Knapp, H.; Straubinger, M.; Stingl, C.; Winterhalter, P. Analysis of
Norisoprenoid Aroma Precursors. In Carotenoid-Derived Aroma Compounds; P.
Winterhalter and R. L. Rouseff, Eds.; American Chemical Society: Washington,
DC, 2002; pp 20-35.
36 Baumes, R.; Wirth, J.; Bureau, S.; Gunata, Y.; Razungles, A. Biogeneration of
C13-norisoprenoid compounds: experiments supportive for an apo-carotenoid
pathway in grapevines. Anal. Chim. Acta 2002, 458, 3-14.

78
37 Gunata, Z.; Dugelay, I.; Sapis, J. C.; Baumes, R.; Bayonove, C. Role of enzymes
in the use of the flavor potential from grape glycosides in winemaking. Prog.
Flavour Precursor Stud. Proc. lnt. Conf. 1993, 219-234.
38 Winterhalter, P. Oxygenated C13-Norisoprenoids: Important Flavor Precursors. In
Flavor Precursors: Thermal and Enzymatic Conversions', R. Teranishi; G. R.
Takeoka and M. Guntert, Eds.; American Chemical society: Washington, DC,
1992; pp 98-115.
39 Lutz, A.; Winterhalter, P. Isolation of additional carotenoid metabolites from
quince fruit (Cydonia oblonga Mill.). J. Agrie. Food Chem. 1992, 40, 1116-1120.
40 Aubert, C.; Ambid, C.; Baumes, R.; Guenata, Z. Investigation of Bound Aroma
Constituents of Yellow-Fleshed Nectarines (Prunus prsica L. Cv. Springbright).
Changes in Bound Aroma Profile during Maturation. J. Agrie. Food Chem. 2003,
51, 6280-6286.
41 Buttery, R. G.; Teranishi, R.; Ling, L. C.; Flath, R. A.; Stem, D. J. Quantitative
studies on origins of fresh tomato aroma volatiles. J. Agrie. Food Chem. 1988, 36,
1247-1250.
42 Wilkins, C. K. Paprika: relationships between aroma profile data and both GC and
HPLC data. Lebensm-Wiss.-Technol. 1992, 25, 212-218.
43 Enzell, C. R.; Wahlberg, I. Tobacco isoprenoids precursors of important aroma
constituents. Pure Appl. Chem. 1990, 62, 1353-1356.
44 Kawakami, M.; Ganguly, S. N.; Banerjee, J.; Kobayashi, A. Aroma composition
of oolong tea and black tea by brewed extraction method and characterizing
compounds of darjeeling tea aroma. J. Agrie. Food Chem. 1995, 43, 200-207.
45 Suzuki, M.; Matsumoto, S.; Fleischmann, H.-P.; Shimada, H.; Yamano, Y.; Ito,
M.; Watanabe, N. Identification of (3-damascenone Progenitors and Their
Biogenesis in Rose Flowers (Rosa damascena Mill.). In Carotenoid-Derived
Aroma Compounds', P. Winterhalter and R. L. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002; pp 89-101.
46 Kotseridis, Y.; Baumes, R. L.; Skouroumounis, G. K. Quantitative determination
of free and hydrolytically liberated .beta.-damascenone in red grapes and wines
using a stable isotope dilution assay. J. Chromatogr., A 1999, 849, 245-254.
47 Sefton, M. A.; Francis, I. L.; Williams, P. J. Volatile norisoprenoid compounds as
constituents of oak woods used in wine and spirit maturation. J. Agrie. Food
Chem. 1990, 38, 2045-2049.
48 Ohloff, G. Importance of minor components in flavors and fragrances. Perfum.
Flavor. 1978, 3, 11-12, 14-16, 18-22.

49
50
51
52
53
54
55
56
57
58
59
79
Roberts, D. D.; Mordehai, A. P.; Aeree, T. E. Detection and Partial
Characterization of Eight (3-Damascenone Precursors in Apples (Malus domestica
Borkh. Cv. Empire). J. Agrie. Food Chem. 1994, 42, 345-349.
Buttery, R. G.; Teranishi, R.; Ling, L. C.; Tumbaugh, J. G. Quantitative and
sensory studies on tomato paste volatiles. J. Agrie. Food Chem. 1990, 38, 336-
340.
Buttery, R. G.; Ling, L. C. Volatile Components of Tomato Fruit and Plant Parts:
Relationship and Biogenesis. In Bioactive Volatile Compounds from Plants', R.
Teranishi; R. G. Buttery and H. Sugisawa, Eds.; American Chemical Society:
Washington, DC, 1993; pp 23-34.
Lozano, P.; Delgado, D.; Gomez, D.; Rubio, M.; Iborra, J. L. A non-destructive
method to determine the safranal content of saffron (Crocus sativus L.) by
supercritical carbon dioxide extraction combined with high-performance liquid
chromatography and gas chromatography. J. Biochem. Biophys. Methods 2000,
43, 367-378.
Knapp, H.; Straubinger, M.; Witte, A.; Winterhalter, P. Aroma Formation in
Saffron. In Frontiers of Flavour Science', P. Winterhalter, Ed.: Freising, Germany,
2000; pp 440-444.
Razungles, A.; Bayonove, C. L.; Cordonnier, R. E.; Sapis, J. C. Grape
carotenoids: changes during the maturation period and localization in mature
berries. Am. J. Enol. Vitic. 1988, 39, 44-48.
Razungles, A.; Gunata, Z.; Pinatel, S.; Baumes, R.; Bayonove, C. Content of
terpenes, norisoprenoids and their precursors in several varieties of grapes. Sci.
Aliments 1993,13, 59-72.
Schreier, P.; Drawert, F.; Junker, A. Identification of volatile constituents from
grapes. J. Agrie. Food Chem. 1976, 24, 331-336.
Williams, P. J.; Sefton, M. A.; Wilson, B. Nonvolatile Conjugates of Secondary
Metabolites as Precursors of Varietal Grape Flavor Components. In Flavor
Chemistry: Trends and Development', R. Teranishi; R. G. Buttery and F. Shahidi,
Eds.; American Chemical Society: Washington, DC, 1989; pp 35-48.
Schneider, R.; Razungles, A.; Augier, C.; Baumes, R. Monoterpenic and
norisoprenoidic glycoconjugates of Vitis vinifera L. cv. Melon B. as precursors of
odorants in Muscadet wines. J. Chromatogr., A 2001, 936, 145-157.
Aeree, T. E.; Barnard, J.; Cunningham, D. G. A procedure for the sensory
analysis of gas chromatographic effluents. Food Chem. 1984, 14, 273-286.

60
61
62
63
64
65
66
67
68
69
70
71
80
Schieberle, P.; Grosch, W. Identificierung von Aromastoffen ausder Kruste von
Roggenbrot Verleich mit den Aromastoffen der Krume. Z. Lebensm.-Unters.
Forsch. 1984, 479-483.
da Silva, M. A. A. P.; Lundahl, D. S.; M.R., M. The Capabilities and
Psychophysics of Osme: a New GC-Olfactometry Technique. In Trends in Flavor
Research-, H. Maarse and V. D. Heij, Eds.; Elsevier: New York, 1994; pp 191-
210.
Stevens, S. S. Concerning the psychophysical power law. Quart. J. Exp. Psychol.
1964,16, 383-385.
Arthur, C. L.; Pawliszyn, J. Solid phase microextraction with thermal desorption
using fused silica optical fibers. Analytical Chemistry 1990, 62, 2145-2148.
Jia, M.; Zhang, Q. H.; Min, D. B. Optimization of Solid-Phase Microextraction
Analysis for Headspace Flavor Compounds of Orange Juice. J. Agrie. Food
Chem. 1998, 46, 2744-2747.
Bezman, Y.; Rouseff, R. L.; Naim, M. 2-Methyl-3-furanthiol and methional are
possible off-flavors in stored orange juice: aroma-similarity, NEF/SNIF GC-O,
and GC analyses. J. Agrie. Food Chem. 2001, 49, 5425-5432.
Steffen, A.; Pawliszyn, J. Analysis of Flavor Volatiles Using Headspace Solid-
Phase Microextraction. J. Agrie. Food Chem. 1996, 44, 2187-2193.
Song, J.; Fan, L.; Beaudry, R. M. Application of Solid Phase Microextraction and
Gas Chromatography/Time-of-Flight Mass Spectrometry for Rapid Analysis of
Flavor Volatiles in Tomato and Strawberry Fruits. J. Agrie. Food Chem. 1998, 46,
3721-3726.
Rouseff, R.; Raley, L.; Hofsommer, H.-J. Application of diode array detection
with a C-30 reversed phase column for the separation and identification of
saponified orange juice carotenoids. J. Agrie. Food Chem. 1996, 44, 2176-2181.
Winterhalter, P.; Lutz, A.; Schreier, P. Isolation of a glucosidic precursor of
isomeric marmelo lactones from quince fruit. Tetrahedron Lett. 1991, 32, 3669-
3670.
Lutz, A.; Winterhalter, P. Bio-oxidative cleavage of carotenoids: important route
to physiological active plant constituents. Tetrahedron Lett. 1992, 33, 5169-5172.
Kotseridis, Y.; Baumes, R. L.; Bertrand, A.; Skouroumounis, G. K. Quantitative
determination of b-ionone in red wines and grapes of bordeaux using a stable
isotope dilution assay. Journal of Chromatography, A 1999, 848, 317-325.

72
73
74
75
76
77
78
79
80
81
82
83
84
81
Lee, H. S.; Castle, W. S. Seasonal Changes of Carotenoid Pigments and Color in
Hamlin, Early gold, and Budd Blood Orange Juices. J. Agrie. Food Chem. 2001,
49, 877-882.
Pupin, A. M.; Dennis, M. J.; Toledo, M. C. F. HPLC analysis of carotenoids in
orange juice. Food Chem. 1999, 64, 269-275.
Philip, T.; Chen, T. S.; Nelson, D. B. Liquid chromatographic profiles of major
carotenoid esters in commercially processed California Navel and Valencia
orange juice concentrates. J. Chromatogr. 1988, 442, 249-265.
Lee, H. S.; Castle, W. S.; Coates, G. A. High-performance liquid chromatography
for the characterization of carotenoids in the new sweet orange (Earlygold) grown
in Florida, USA. J. Chromatogr., A 2001, 913, 371-377.
Mouly, P. P.; Gaydou, E. M.; Lapierre, L.; Corsetti, J. Differentiation of several
geographical origins in single-strength Valencia orange juices using quantitative
comparison of carotenoid profiles. J. Agrie. Food Chem. 1999, 47, 4038-4045.
Noga, G.; Lenz, F. Separation of citrus carotenoids by reversed-phase high-
performance liquid chromatography. Chromatographia 1983,17, 139-142.
Marais, J. l,L6-Trimethyl-l,2-dihydronaphthalene (TDN): a possible degradation
product of lutein and b-carotene. South African J. Enol. Viticulture 1992,13, 52-
55.
Chen, C. S.; Shaw, P. E.; Parish, M. E. Orange and Tangerine Juice. In Fruit
Juice Processing Technology, S. Nagy; C. S. Chen and P. E. Shaw, Eds.;
AGScience, Inc.: Aubumdale, 1993; pp 110-165.
Stewart, I.; Wheaton, T. A. Continuous flow separation of carotenoids by liquid
chromatography. J. Chromatogr. 1971, 55, 325-336.
Britton, G. UV/Visible Spectroscopy. In Carotenoids, Vol. 1A: Isolation and
Analysis; G. Britton; S. Liaaen-Jenson and H. Pfander, Eds.; Birkhauser Verlag:
Basel, Boston, Berlin, 1995; pp 13-62.
De Ritter, E.; Purcell, A. E. Carotenoid analytical Methods. In Carotenoids as
Colorants and Vitamin A Precursors; J. C. Bauemfeind, Ed.; Academic Press:
London, 1981; pp 815-923.
Farin, D.; Ikan, R.; Gross, J. The carotenoid pigments in the juice and flavedo of a
mandarin hybrid (Citrus reticulata) cv. Michal during ripening. Phytochemistry
1983, 22, 403-408.
Curl, A. L.; Bailey, G. F. The carotenoids of Navel oranges. J. Food Sci. 1961,
26, 442-447.

85
86
87
88
89
90
91
92
93
94
95
96
97
82
Robertson, G. W.; Griffiths, D. W.; Woodford, J. A. T.; Birch, A. N. E. Changes
in the chemical composition of volatiles released by the flowers and fruits of the
red raspberry (Rubus idaeus) cultivar glen Prosen. Phytochemistry 1995, 38,
1175-1179.
Castellar, M. R.; Montijano, H.; Manjon, A.; Iborra, J. L. Preparative high-
performance liquid chromatographic purification of saffron secondary
metabolites. 7. Chromatogr. 1993, 648, 187-190.
Gross, J.; Gabai, M; Lifshitz, A. Carotenoids in juice of Shamouti orange. 7.
Food Sci. 1971, 36, 466-473.
Valim, M. F.; Rouseff, R. L.; Lin, J. Gas Chromatographic-Olfactometric
Characterization of Aroma Compounds in Two Types of Cashew Apple Nectar. 7.
Agrie. Food Chem. 2003, 51, 1010-1015.
Ahmed, E. M.; Dennison, R. A.; Shaw, P. E. Effect of selected oil and essence
volatile components on flavor quality of pumpout orange juice. 7. Agrie. Food
Chem. 1978, 26, 368-372.
Poole, C. F. The essence of chromatography, Elsevier: Amsterdam, Boston,
London, New York, Oxford, Paris, san Diego, San Francisco, Singapore, Sidney,
Tokyo, 2003.
Zhang, Z.; Pawliszyn, J. Headspace solid-phase microextraction. Anal. Chem.
1993,65, 1843-1852.
Bao, M.; Mascini, M.; Griffini, O.; Burrini, D.; Santianni, D.; Barbieri, K.
Headspace solid-phase microextraction for the determination of trace levels of
taste and odor compounds in water samples. Analyst (Cambridge, United
Kingdom) 1999,124, 459-466.
Radford, T.; Kawashima, K.; Friedel, P. K.; Pope, L. E.; Gianturco, M. A.
Distribution of volatile compounds between the pulp and serum of some fruit
juices. 7. Agrie. Food Chem. 1974, 22, 1066-1070.
Lin, J.; Rouseff, R. L.; Barros, S.; Naim, M. Aroma Composition Changes in
Early Season Grapefruit Juice Produced from Thermal Concentration. 7. Agrie.
Food Chem. 2002, 50, 813-819.
Sanderson, G. W.; Grahamm, H. N. Formation of black tea aroma. 7. Agrie. Food
Chem. 1973, 21, 576-585.
Day, W. C.; Erdman, J. G. Ionene; a thermal degradation product of (3-carotene.
Science 1963, 141, 808.
Mader, I. P-Carotene: thermal degradation. Science 1964,144, 533-534.

83
98 Schiedt, K.; Liaaen-Jensen, S. Isolation and Analysis. In Carotenoids-, G. Britton;
S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel, Boston, Berlin,
1995; pp 81-108.
99 Ravichandran, R. Carotenoid composition, distribution and degradation to flavor
volatiles during black tea manufacture and the effect of carotenoid
supplementation on tea quality and aroma. Food Chem. 2002, 78, 23-28.
100 Stevens, M. A. Relationship Between Polyene-Carotene Content and Volatile
Compound Composition of tomatoes. J. Amer. Soc. Hort. Sci. 1970, 95, 461-464.
101 Mord, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Ingold, K. U.; Lindsay,
D. A. Exploratory study of (3-carotene autoxidation. Tetrahedron Lett. 1991, 32,
4203-4206.
102 de Heij, H. T.; van Dort, H. M. Autoxidation of carotenes. Prog. Flavour
Precursor Stud. Proc. Int. Conf. 1993, 409-413.

BIOGRAPHICAL SKETCH
Kanjana Mahattanatawee was bom in Bangkok, Thailand on 17 May, 1965. She
received a B.Sc. in biology with a major in Microbiology in 1988 from Sri-Nakharinwirot
University, Thailand. She continued to pursue her Master of Science degree in the area
of industrial microbiology at the Department of Microbiology, Chulalongkom University,
Bangkok Thailand from 1988-1991. From 1991-1992 she worked as a researcher, in the
Department of Microbiology, Chulalongkom University, Thailand. In 1992-1993
Kanjana was awarded a UNESCO scholarship to earn her Diploma in Microbiology and
Biotechnology from Osaka University, Japan. Kanjana was appointed to a position as
Lecturer, Department of Food Technology, Faculty of Science, Siam University from
1993-1999. From 1995-1997 she was an adjunct lecturer, Faculty of Environment and
Natural Resource, Mahidol University. Kanjana conducted research and taught two
microbiology courses (Industrial Microbiology and Fermentation Technology) for
undergraduate students at the Faculty of Science, Siam University, Bangkok, Thailand.
She was awarded a scholarship from Siam University to pursue her Ph.D. In
Spring 1999, she enrolled in the graduate program at the Department of Food Science and
Human Nutrition at the University of Florida under Dr. R.L. Rouseff s supervision. She
considers herself very fortunate to be enrolled in one of the greatest graduate programs in
flavor chemistry, with excellent scientists who are a pleasure to work with. She
completed her research for her Ph.D. degree at the Citrus Research and Education Center
(CREC) in Lake Alfred, Florida.
84

85
After completing her Ph.D. program, Kanjana plans to work as a postdoctoral
researcher to gain more experience in this subject area. Later, she will return to Thailand
to fulfill an appointed position as an associate professor at Siam University. Kanjana will
teach and conduct research. She hopes to deliver the excitement and enthusiasm to her
students in Thailand that she has experienced from her professors in the U.S.A.

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Russell L. Rouseff, Chair
Professor of Food Science and Human Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Maurice R. Marshall, Jr.
Professor of Food Science and Human Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Reriee M./ioodrich
Assistant Professor of Food Science and Human
Nutrition
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
David H. Powell
Faculty Scientist of Chemistry
This dissertation was submitted to the Graduate Faculty of the College of Education and
to the Graduate School and was accepted as partial fulfillment of the requirements for the
degree of Doctor of Philosophy.
May 2004
Dean, Graduate School



Olfactory Response FID Response S Olfactory Response FID Response
66
10 15 20 25
Time (min)
Headspace volatiles from 3-carotene in model solution pH 3.5 at 0 day
10 15 20 25
Time (min)
Figure 6-3. Headspace volatiles from p-carotene in model solution pH 3.5 after storage 1
day at 35C: 1 = P-ionone, a = sweet/raspberry


42
GC-0 retention index matches with standards on two dissimilar chromatographic column
materials, aroma descriptor matches and GC-MS matches of both retention time and
fragmentation spectra. Of these four norisoprenoids, P-ionone had been reported in two
previous orange juice GC-0 studies (4, 5). There is one previous mention of P-
damascenone in heated juice, but no MS or independent instrumental confirmation data
was presented (65). P-cyclocitral and a-ionone were detected in orange juice for the first
time in this study and confirming MS data for P-damascenone was presented for the first
time.


TABLE OF CONTENTS
page
ACKNOWLEDGMENTS iv
LIST OF TABLES viii
LIST OF FIGURES ix
ABSTRACT xii
CHAPTER
1 INTRODUCTION 1
2 LITERATURE REVIEW 3
Orange Juice Aroma 3
Carotenoids 3
Norisoprenoids 4
Norisoprenoids Formation from Carotenoids 5
Apple 7
Tomato 8
Saffron 9
Grape and Wine 10
Gas Chromatography-Olfactometry 11
Solid Phase Microextraction 13
Orange Juice Norisoprenoids 14
3 HPLC DETERMINATION OF CAROTENOID NORISOPRENOID
PRECURSORS IN ORANGE JUICE 15
Introduction 15
Objectives 17
Materials and Methods 17
Carotenoid Extraction 17
Carotenoid Saponification 17
HPLC Procedure 18
Results and Discussion 19
Carotenoids of Interest 19
Hydrolysis Conditions 20
v


CHAPTER 2
LITERATURE REVIEW
Orange Juice Aroma
The aroma of fresh orange juice is composed of a complex mixture of aldehydes,
esters, ketones, alcohols and terpenes blended in specific proportions. Numerous studies
(1-5) aimed at identifying the flavor volatiles in orange juice have led to the identification
of about 200 volatiles, but no single aroma character impact for orange flavor has ever
been reported. GC-olfactometry and orange juice volatile quantification have been used
to gain a more accurate understanding of their contribution to orange flavor (2, 5-7).
Early orange juice GC-0 studies have demonstrated that the orange juice volatiles present
in highest concentration have little to no aroma activity and many aroma active
compounds exist as low-level volatiles that are difficult to detect using typical
instrumental detectors.
Carotenoids
Carotenoids are primarily responsible for the colors of many plants, birds, and
insects; but also serve as plant photoprotection agents during photosynthesis, and as
essential human nutrients. However, the least-appreciated role of carotenoids is their
function as aroma precursors.
Carotenoids are tetraterpenes (C4o ) resulting from the joining together of eight molecules
of isoprene (C5) through tail-to-tail condensation. Most carotenoids have a C40 carbon
skeleton. The ends may or may not be cyclized into six membered rings. If the ends are
not cyclized, the molecule is termed acyclic. There are two main groups of carotenoids:
3


6
rise to polar intermediates (aroma precursors), and 3) acid-catalyzed conversions of the
nonvolatile precursors into the aroma active form (32). One example illustrating these
reaction is the formation of p-damascenone from neoxanthin (Fig. 2-3). The primary
oxidative cleavage product of neoxanthin, grasshopper ketone, must be enzymatically
reduced before finally being acid-catalyzed converted into the odoriferous ketone. In the
direct process, the target compound is immediately obtained after the initial cleavage
(i.e., formation of a- and P-ionone directly from a- and (3- carotene) (34).
P-carotene
P-ionone
Carotenoid
neoxanthin
Step 1
oxidative cleavage
Primary cleavage
product
Step 2
Enzymatic
transformation
Non-volatile metabolite
(aroma precursor)
Step 3
Acid catalyzed
conversions
Aroma compound
Figure 2-3. General steps for the conversion of carotenoids into flavor compounds,
showing the formation of P-ionone and P-damascenone from P-carotene and
neoxanthin respectively (Winterhalter, P., Rouseff, R. Carotenoids-Derived
Aroma Compounds: An Introduction. In Carotenoid-derived Aroma
Compounds; P. Winterhalter and R. Rouseff, Eds.; American Chemical
Society: Washington, DC, 2002, Fig. 4, page 12).
Recent studies have shown that some of the volatile Cn-compounds are not free,
uncomplexed plant constituents; but rather are derived from less or nonvolatile precursors
such as polyols, glycosides, and glucose esters. Carotenoid degradation is initiated by


Mean Amount of Leaf Material Consumed (ci)i
49
(Error Bars @ 95%; F(0.05)i,i8=5.98, F=1.64; P =0.217)
Control 1.0% L-methionine
Figure 4-5. Mean leaf consumption by Colorado potato beetle in the preference
tests. Error bars denote 95% SE, and treatments were found not to
be statistically different. No correlation between either Control or
Treatment Diet consumed and mean head capsule width was found
(Pearson Correlation Coefficient 0.466, P=0.175 and 0.665,
P=0.036, respectively).


4
the hydrocarbon group, which contain only carbon and hydrogen; and the xanthophyll
group, which contain carbon, hydrogen, and oxygen. Oxygen in xanthophylls is usually
found as either hydroxyl-(monols, diols and polyols), epoxy- (5,6 and 5,8-epoxides),
methoxy, aldehyde, oxo, carboxy and/or esters. Hydroxyl substitution primarily occurs at
the C3 position in the ionone ring; and a carbonyl substitution usually occurs at the C4
position in the (3-ionone ring. In most of the cyclic carotenoids, the 5,6- and 5,6-double
bonds are the most susceptible to epoxidation. The unconjugated double bond in the a-
ionone 8 ring does not undergo epoxidation. Allenic carotenoids have a C=C=C
grouping at one end of the central chain, and acetylenic carotenoids have a -C=C- bond
in position 7,8 and/or 7,8 (26, 27). Figure 2-1 shows acyclic carotene (lycopene),
bicyclic carotene (P-carotene), the monol P-cryptoxanthin, and the diol zeaxanthin.
Figure 2-1. Examples of carotene and xanthophyll carotenoid structures.
Norisoprenoids
Norisoprenoids are volatile C9-CI3 fragments from the degradation of the C40
carotenoids. The formation of norisoprenoids from carotenoids is thought to proceed via


68
color, due to P-carotene, was destroyed and simultaneously an off-flavor (violet-like)
developed {24).
The polyene-carotenes are apparently oxidized at the first conjugated diene bonds.
Oxidation is more prevalent adjacent to the methyl group and it is possible that inductive
effects of the methyl group make the double bond adjacent to it more susceptible to
oxidation {100). When P-carotene in benzene or tetrachloromethane is allowed to react
with molecular oxygen in the absence of light at 30C, P-ionone can be formed within the
first few hours. As the oxidation progressed a number of shorter chain products are
formed, including P-cyclocitral {101). The aroma active volatile compounds formed
indicates that oxidative scission of P-carotene can occur at carbon bond C7-C8, C8-C9 and
C9-C10 of p-carotene to generate Co: P -cyclocitral, Cn: P-homocyclocitral, C13:
P-damascone and C13: P-ionone respectively (see figure 6-4). The same degradation
position and volatile compounds (except p-homocyclocitral) have been reported by
photo-oxygenation {102) autoxidation with molecular oxygen at 30C in the dark
conditions {101) and oxidation in water at 97C {31). At the scission Cs-Cg of
p-carotene, P-homocyclocitral (Cn) was formed in present model condition but at the
same scission position, dihydroactinodiolide (Cn) was formed and has been reported in
the different model conditions {31, 101, 102).
GC-0 Analysis of P-Carotene Decomposition at 35C
The GC-0 data for the two week storage sample is in Table 6-1. It should be noted
that the same retention and aroma descriptors observed earlier for standard and juices
were also observed for this storage sample. It is interesting to see just how similar the
values in Table 6-1 are with their corresponding components in Table 5-9. In making this


CHAPTER 6
FIELD EVALUATION OF L-METHIONINE AS AN INSECTICIDE
Introduction
The role of methionine in animal systems is well known and only recently
understood in plants. Methionine is required for protein synthesis; it is a precursor to
several important biochemical compounds including ethylene and polyamines, sulfate
uptake and assimilation, and also acts as an activator of threonine-synthase (Giovanelli et
al. 1980; Droux et at. 2000; Bourgis et at. 2000; Zeh et at. 2001). Recently, research has
focused on the transgenic modification of crop plants to overproduce methionine in order
to increase their nutritional quality without affecting other biochemical processes (Zeh et
al 2001). However, little work has been conducted on the effects of exogenous
methionine and it became important to understand the role of externally applied
methionine on plant health.
Furthermore, the application of L-methionine to plants exposed to natural
conditions presents additional problems in terms of how long the residue remains on the
plant. Observations of other experiments using L-methionine revealed the tendency of
this compound to crystallize after the aqueous portion evaporated forming a brittle, crusty
coating that is easily removed. This coating does not appear to interfere with respiration
and transpiration at the concentrations studied (1% and lower). To prevent the loss of
L-methionine from the plants in a natural setting, the adjuvant Silwett L-77 (Helena
Chemical; Collierville, TN) was included in this portion of the study in an effort to
increase residual activity on the plant. Silwet L-77 is a nonionic organosilicate
69


66
hours, and to 0.11% for 48-168 hours and remained constant since the trial lasted longer
because of the use of neonates instead of 3rd instars. The D-methionine treatments were
similar with 0.44% for 24 and 48 hours, 0.33% for 72 hours and 0.32% after 168 hours.
While not as striking as the others, Beta-alanine had a LC50 concentration of 1.1% after
24 hours, dropped to 0.5% after 48 hours and leveled off around at 0.35% after 72 and
168 hours. Probit analysis of the Proline and L-leucine treatments was not performed, as
the mortality associated with those treatments was too low (Figures 5-5 and 5-6).
Discussion
Although not commonly encountered, the D- form of methionine had virtually the
same effect as the L- form on larval mosquito mortality. The D-and L-methionine trials
showed that the D- form had lower mortality associated with it than the more reactive
L-counterpart. Insects do not commonly use the D- form of amino acids, although
D-methionine is metabolized by some orders to a limited extent (Ito and Inokuchi, 1981).
The YFM could be an example of this phenomenon.
Because of the nature of the CAATCH1 system in the alkaline midgut, buffering
may have acted to increase the effectiveness of the system. Buffering the solutions did
result in an increase in mortality, with even the lowest concentration of 0.1%
L-methionine exhibiting a two-fold increase with the buffered form (Figure 5-4).
Complete mortality was reached sooner with the buffered forms even for concentrations
that did not reach 100% in the unbuffered form. In a field setting, the addition of
L-methionine would be buffered naturally by the chemical properties of the bodies of
water to which it was applied and similar results would be expected.


67
Jaffe and Chrin (1979) found the adults of YFM females infected with Brugia, a
filiaral parasite, were depleted of free form methionine because of the infection and were
able to make up the difference by converting homocysteine to methionine with a special
synthetase. The ability of YFM adults to synthesize methionine from homocysteine may
be present in the larvae as well. This could be the result of the lack of methionine in the
diet and possible evidence of the CAATCH1 system being present in at least the adult
stage. The susceptibility of the larvae to L-methionine also could be the result of
overexposure to a compound that is normally not encountered in high concentrations
(>0.1%). However, the alkalinity of the particulate feeding larvae and the high mortality
to L-methionine suggests that the CAATCH1 system is present and could be exploited in
other species with similar midgut characteristics (Dadd, 1975).
The survival of YFM larvae exposed to both Beta-alanine and L-leucine was
unusual in that they each had the opposite effect on the YFM larvae. L-leucine was
expected to have similar blocking properties as L-methionine based on CAATCH1
research (Feldman et al., 2000). Instead, almost no mortality was observed indicating the
possibility of another system involved with the transport of this amino acid. Conversely,
beta-alanine was not found to be reactive with the CAATCH1 system based on the work
of Feldman et al. (2000). The unusually high larval mortality associated with this amino
acid may be the result of a yet to be discovered midgut property.
The similar mortalities observed for the higher concentrations of L-methionine
and Bti is encouraging considering the resistance to this compound that has been
documented in many insect species because of reduced receptor activity and binding
(Bills et al., 2004; Nester et al., 2002). Resistance in insects involves a variety of


49
times are usually limited to analytes in relatively high concentration and even then the
reproducibility is not that good. In this study, very short exposure times were not an
option as the analytes of interest were present in very low concentrations.
The reproducibility (analytical precision) of a fortified juice using SPME-GC-FTD
was determined in five replicates at 40C with 45 min exposure. The relative standard
deviations (RSD) obtained were 1.7, 1.7, 0.4, 1.4 % for P-cyclocitral, P-damascenone, a-
ionone, and P-ionone respectively (Table 5-1). It should be kept in mind that the orange
juice had been fortified with 8.6, 4, 5.4, and 5.27 ppm P-cyclocitral, P-damascenone, a-
ionone, and P-ionone respectively. The low RSD indicated that the SPME and GC
analytical conditions employed in this study could quantify norisoprenoids in orange
Table 5-1. Reproducibility of SPME exposure time 45 min at 40C
Replicate
P-cyclocitral
P-damascenone
a-ionone
P-ionone
1
84101
27181
46618
23072
2
84149
26797
46715
22861
3
84129
26534
46813
22589
4
81092
26382
46570
22337
5
82037
25961
46319
22335
Average
83101
26571
46607
22639
STD Va
1443
456
186
325
RSDb
1.7
1.7
0.4
1.4
Standard deviation,b relative standard deviation
juice in a highly reproducible manner. However, it should be pointed out the
concentrations used to fortify the sample were considerably higher than would ever be
found in an orange juice sample. Typical juice concentrations are 50 to 1000 times lower
so that typical RSDs for unfortified juice samples range from 20- 50% which might
seem high, but still very acceptable for analyses at the sub p,g/L level the complex matrix
of orange juice. The volatility of flavor compounds can be changed according to the


75
Figure 6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant
leaves treated with L-methionine and the adjuvant Silwett L-77
(nTotai=120). Data corrected for control mortality using Abbotts formula.
Note the overlap in trend lines for the Control treatments and 0.1%L-methk>nine
treatment.


2
mango, tomato, saffron, cured tobacco, and black tea (8-16). Only a single
norisoprenoid, (3-ionone, has been identified to date in fresh orange juice (4, 5)
More than 50 carotenoids have been separated and identified from the juice of
three varieties of Citrus sinensis (Shamouti, Valencia, and Washington Navel) using
column chromatography combined with thin layer chromatography (TLC) (17). Some of
these carotenoids (such as (3-carotene, a-carotene, neoxanthin, (3-crytoxanthin, lutein,
violaxanthin, and canthaxanthine) have the structural potential to form potent
norisoprenoid fragments (18-22). Furthermore, (3-carotene in tomato products has been
shown to produce (3-ionone and (3-cyclocitral (23). Beta-ionone and a-ionone have been
generated from (3-carotene and a-carotene respectively in carrots (24). Neoxanthin in
grapes has been shown to be a source of (3-damascenone (25). Prior orange juice
carotenoid studies were primarily directed toward the contribution of carotenoids to juice
color and for vitamin A content. They have been generally ignored as precursors of
aroma compounds. Since orange juice has so many carotenoids that could serve as
precursors for a wide range of norisopemoids, the objectives of this research were to:
1. Confirm the presence of possible carotenoid norisoprenoid precursors in orange juice
using HPLC and photodiode array detection. (Chapter 3)
2. Determine if additional norisoprenoid are present in orange juice. Characterize and
identify these new norisoprenoids. (Chapter 4)
3. Determine the relative aroma impact of carotenoid degradation products
(norisoprenoids) to the total aroma impact of orange juice in fresh, pasteurized, and
reconstituted from concentrate juice. (Chapter 5)
4. Develop quantitative procedures to isolate and quantify orange juice norisoprenoids
using static headspace SPME with GC-MS. (Chapter 5)
5. Determine if (3-carotene can form norisoprenoid degradation products at 35C storage
in model solutions. (Chapter 6).


85
86
87
88
89
90
91
92
93
94
95
96
97
82
Robertson, G. W.; Griffiths, D. W.; Woodford, J. A. T.; Birch, A. N. E. Changes
in the chemical composition of volatiles released by the flowers and fruits of the
red raspberry (Rubus idaeus) cultivar glen Prosen. Phytochemistry 1995, 38,
1175-1179.
Castellar, M. R.; Montijano, H.; Manjon, A.; Iborra, J. L. Preparative high-
performance liquid chromatographic purification of saffron secondary
metabolites. 7. Chromatogr. 1993, 648, 187-190.
Gross, J.; Gabai, M; Lifshitz, A. Carotenoids in juice of Shamouti orange. 7.
Food Sci. 1971, 36, 466-473.
Valim, M. F.; Rouseff, R. L.; Lin, J. Gas Chromatographic-Olfactometric
Characterization of Aroma Compounds in Two Types of Cashew Apple Nectar. 7.
Agrie. Food Chem. 2003, 51, 1010-1015.
Ahmed, E. M.; Dennison, R. A.; Shaw, P. E. Effect of selected oil and essence
volatile components on flavor quality of pumpout orange juice. 7. Agrie. Food
Chem. 1978, 26, 368-372.
Poole, C. F. The essence of chromatography, Elsevier: Amsterdam, Boston,
London, New York, Oxford, Paris, san Diego, San Francisco, Singapore, Sidney,
Tokyo, 2003.
Zhang, Z.; Pawliszyn, J. Headspace solid-phase microextraction. Anal. Chem.
1993,65, 1843-1852.
Bao, M.; Mascini, M.; Griffini, O.; Burrini, D.; Santianni, D.; Barbieri, K.
Headspace solid-phase microextraction for the determination of trace levels of
taste and odor compounds in water samples. Analyst (Cambridge, United
Kingdom) 1999,124, 459-466.
Radford, T.; Kawashima, K.; Friedel, P. K.; Pope, L. E.; Gianturco, M. A.
Distribution of volatile compounds between the pulp and serum of some fruit
juices. 7. Agrie. Food Chem. 1974, 22, 1066-1070.
Lin, J.; Rouseff, R. L.; Barros, S.; Naim, M. Aroma Composition Changes in
Early Season Grapefruit Juice Produced from Thermal Concentration. 7. Agrie.
Food Chem. 2002, 50, 813-819.
Sanderson, G. W.; Grahamm, H. N. Formation of black tea aroma. 7. Agrie. Food
Chem. 1973, 21, 576-585.
Day, W. C.; Erdman, J. G. Ionene; a thermal degradation product of (3-carotene.
Science 1963, 141, 808.
Mader, I. P-Carotene: thermal degradation. Science 1964,144, 533-534.


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Entomology and
Nematology
This dissertation was submitted to the Graduate Faculty of the College of
Agricultural and Life Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of Philosophy.
May 2004
Dean, College of Agricultural
Sciences
Dean, Graduate School


20
The right half of these four carotenoids differ considerably. The right half of (3-carotene
can produce both (3-cyclocitral and (3-ionone because it is identical to the left half of the
molecule. The right half of a-carotene can produce a-ionone and neither a- or (3-
cryptoxanthin produce norisoprenoids which were observed in orange juice. The final
carotenoid of interest, neoxanthin, has been shown to produce (3-damescenone in a three
step process (34).
Hydrolysis Conditions
As previously discussed, citrus carotenoids must be hydrolyzed to simplify the
separation due to the complexity from the multiple natural esters formed from C^-Cis
saturated fatty acids (74). Hydrolysis conditions must be optimized in order to free the
esterified carotenoids into a single form but not so long as to promote alkaline hydrolysis
of the carotenoids. Concentration of alkali, reaction time and temperature are the
variables of interest. In recent years, most carotenoid studies have employed 0.1 M KOH
and room temperature so only reaction time was optimized for this study.
Chromatograms with no saponification showed 80% of the total peak area eluting as an
unresolved band of peaks during the last quarter of the chromatogram. As saponification
time increased, the number of peaks at the end of the chromatogram diminished and the
peaks were more evenly distributed during the chromatographic run. Saponification
times in excess of one hour did not reduce the number of late eluting peaks and total
carotenoid peak area was lower at 4 hours and overnight saponification compared to the
one hour saponification. Therefore the one hour saponification was used for the
remainder of the study.


59
exposure (Figures 5-4). The 1.0%L.methionme treatment caused 100% mortality after 2 days
while the Bti treatment took 3 days to reach the same level of control. The proline
treatment caused less than 10% mortality.
In contrast to methionine, survival of YFM larvae exposed to proline and
L-leucine was higher, with only approximately 20% mortality for the higher 0.7% proline
and 1.0% proline concentrations (Figure 5-5) and less than 3% mortality with the highest
L-leucine concentration (Figure 5-6). Beta-alanine mortality was similar to the
L-methionine treatments with between 75% and 83% mortality for the 0.5% Beta-alanine
thru 1.0% Beta-alanine concentrations, respectively, greater than 40% mortality with the
0.3% Beta-alanine, and less than 5% mortality for the 0.1% Beta-alanine concentrations
(Figure 5-7).
Growth and Development
Developmental rates of YFM larvae resulted in three distinct groups, with the
control and proline treatments, producing virtually identical results; both were
statistically different from the 0.1%L-methionine treatment and the remaining
L-methionine treatments (Figure 5-8). The Bti treatment was statistically the same as the
0.3% L-methionine to 1.0% L-methionine treatments, with very little growth taking
place.
Probit analysis for unbuffered L-methionine (nrotai=40 for 5 treatments; 0.1%,
0.3%, 0.5%, 0.7% and 1.0%) revealed an overall LC50 of 0.19% concentration for the
YFM after 7 days of exposure (Figure 5-9). The LC50 of 1.2% for 24 hours dropped to
0.41% after 48 hours and to 0.24% after 72 hours. When the L-methionine treatments
(same concentrations) were buffered to a pH Of 7.0, the values dropped to 0.64% for 24


8
precursors have been detected in apples (Malus domestica Borkh. cv. Empire) (49). The
most abundant precursor, present at 4.6 ng/g, was the 9 (or 3) -a-L-arabinofuranosyl-
(l,6)-P-D-glucopyranoside of the acetylenic diol. The second most abundant precursor,
present at 3.1 ng/g, is a more polar glycoside of the acetylenic diol. (49). Beta-
Damascenone contributed 32% of the total aroma potency of heated apple juice, but only
1.6% of the total aroma of fresh apple juice as determined by GC-O. Thus, most of the
P-damascenone in heated apple juice was generated from nonvolatile precursors during
thermal processing (9).
OH
Allyl-1,6-diol Theaspiranes
OH
P-D-Gentiobioside of 3-Hydroxy-P-ionol
Marmelo Lactone
Figure 2-4. Formation of norisoprenoids aroma compounds from different classes of
precursors (i.e., polyols, glycosides, and glucose esters).
Tomato
One of the most marked differences between the fresh tomato and the paste is the
almost complete loss of the major contributor to fresh tomato aroma, (Z)-3-hexenal. The
most notable increase is with the potent odorant, P-damascenone, which shows a 10 fold


88
Figure 7-1. Mortality of Coleomegilla maculata adults after exposure to L-methionine
treated artificial diet. Data corrected for control mortality using Abbotts
formula.


35
(Bror Bars @ 95%; /r(0.05)i, 18=5.98, F-1.64; P =0.217)
Control 1.0% Treatment
Figure 3-12. Mean leaf consumption by tobacco homworm in the preference
tests. Error bars denote 95% SE, and treatments were found not to
be statistically different. However, there was correlation between
the control diet consumed and mean head capsule width (Pearson
Correlation Coefficient 0.885, P=0.001) while no correlation was
found between the Treatment diet consumed and mean head
capsule width (Pearson Correlation Coefficient 0.630, P=0.05).


44
volatile were calculated using the standard addition procedure. Regression equations
were developed from peak area data from four different concentrations of each compound
added to a juice base. Odor activity values were calculated for each component measured
(although they were not identified as OAV values). Compounds which exceeded their
threshold by the greatest amounts (highest OAV values) and thus most likely to
contribute to fresh orange flavor included: limonene, myrcene, a-pinene, decanal,
octanal, ethyl butanoate, and linalool. The differences between these two studies which
both claim to determine the components most responsible for fresh orange juice flavor
are worth noting.
SPME is a rapid, solventless static headspace procedure. It can be used for the
quantitative analysis of flavor and fragrance compounds. The standard addition method
has been used primarily because the concentration in the headspace (volatility) will be
influenced by the sample matrix (66, 91, 92). Boa et al. (92) reported that reliability
problems of headspace SPME quantification is associated with the matrix and could be
reduced by employing the standard addition method or employing isotopically labeled
internal standards. Headspace SPME with standard additions were used in the present
study because SPME can extract and concentrate orange juice headspace volatiles which
transfer them directly into the injector of a GC in a simple, straightforward manner. Just
as important for this study, the nonvolatile carotenoids will not be extracted. If the
nonvolatile carotenoids were present they might degrade when exposed to the heat
(200C) of GC injection port and possibly produce artifact norisoprenoids. The major
problem with the standard addition approach is that several injections at each standard
addition level are required in order to obtain a single result. Thus, depending on the


47
and the amount of (3-damascenone in the sample determined from the regression line
equation.
Determination of the Relative Impact of Norisoprenoids in Orange Juice
The aroma active compounds from 3 types of orange juice (fresh, pasteurized, and
reconstituted from concentrate) were separated and identified using GC-0 (chapter 4).
Intensities of aroma active compounds of each run were normalized so the highest
intensity was given a score of 10. The normalized intensities of all the runs were then
averaged, providing a similar aroma activity was detected at least half the time at that
retention time. If the compound was not detected in one run its value was treated as
missing, not zero. Aroma-active compounds from the entire GC-0 trial were categorized
into eight groups based on similar aroma description. These eight groups were 1)
citrusy/minty; 2) metallic/mushroom/geranium; 3) roasted/cooked/meaty/spice; 4)
fatty/soapy/green; 5) sulfury/solventy/medicine; 6) floral; 7) sweet/fruity; and 8)
green/grassy. The sums of the total olfactometry intensities for each aroma group was
determined and presented in spider web (radar graph) for each of the four juice types.
The contribution of norisoprenoids to orange juice was calculated from the total intensity
of norisoprenoids to the total intensity of all aroma active volatiles in the juice.
Results and Discussion
Quantification of Norisoprenoids in Orange Juice
The amount of volatile compounds found on the SPME coating depends on
exposure time, temperature, sample volume, headspace volume, and sample
concentration. In this study only exposure time was varied in order to determine the time
needed for equilibrium concentrations for each analyte to be established. All other
factors remained constant. Equilibrium time between SPME fiber and headspace of


CHAPTER 6
THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL SOLUTION
Introduction
Carotenoids are unstable in both the presence of heat and/or light. The thermal
degradation of carotenoids produces a range of volatile products and norisoprenoids are
the most potent aroma compounds of all the volatiles produced. The formation of
specific norisoprenoids from the thermal degradation of carotenoids during heat treatment
of food products have been reported. Beta-ionone, a-ionone, and (3-damascenone have
been reported in tomato paste (23) and black tea (95). These norisoprenoids also were
detected from the thermal degradation of carotenoids in model systems such as (3-
carotene in water at 97C, 3 hrs (31), 1% solution of p-carotene heated at 188F for 72
hrs. (96), and thermal degradation of crystallize P-carotene at 240F in a vaccum (97).
However the norisoprenoids formed in those model systems were formed at high
temperature. At best these studies could be considered accelerated storage studies. In
order to have a model more representative of the conditions that a real world juice
might be exposed to, a 35C storage study was carried out. The model solution was
buffered to pH 3.8 (orange juice pH) using citric acid and tripotassium citrate. Sugars
and amino acids were not added to reduce the possibility that they could be possible
norisoprenoid sources. The concept that carotenoids could act as a source of
norisoprenoids is relatively new (15, 39, 69, 70). These studies have indicated that
specific carotenoids need be present in order to produce specific norisoprenoids.
62


CHAPTER 1
INTRODUCTION
The delicate aroma of fresh orange juice is the result of a complex mixture of
volatiles blended in specific proportions. Numerous analytical studies (1-5) have
identified and quantified the major volatiles in orange juice in an effort to duplicate this
aroma. However, when combined, the identified volatiles could not duplicate orange
juice aroma, suggesting that important aroma components were missing. Early orange
juice gaschromatography olfactometry (GC-O) studies (4, 6, 7) have shown that many of
the aroma-active compounds in orange juice exist as low-level volatiles that are difficult
to detect using typical flame ionization detector (FID) or mass spectrometer (MS)
detectors. Furthermore, these studies demonstrated that the major volatiles in orange
juice have little to no aroma activity. Recent orange juice GC-0 studies (5) quantified
the 25 most intense aroma-active compounds in fresh juice, using isotope dilution
analysis. Model solutions of the aroma components in orange juice based on GC-0
studies have come closer to duplicating the aroma of fresh orange juice than model
systems based on the composition of the volatiles found in highest concentration.
Carotenoids are too large (C40) to be volatile under normal conditions. Because
they contain a highly conjugated double bond structure, they can be degraded by enzyme,
chemical, and/or thermal reactions to form a wide range of structures, depending on
which double bond is broken. Some of their smaller (C9-C13), volatile, decomposition
products are called norisoprenoids. Norisoprenoids have been shown to have significant
aroma impact in fruits, vegetables and spices such as grapes, apples, lychee, starfruit,
1


94
trials. One possible explanation for this observation could be that the excess
L-methionine increased the dietary quality of the artificial and natural diets for the PSLB
in the treatments. However, because only adults were available, further tests are needed
to determine if the larvae, also predaceous on the same pests as the adults, are sensitive to
this compound. It should be noted that the midgut properties (i.e., alkalinity) for this
species are not well known and may not even have the CAATCH1 proteins present in the
midgut.
The mottled water hyacinth weevil also appears not to be adversely affected by
exposure to excess amounts of L-methionine despite its herbivorous habit like the THW
and CPB. Another weevil within the same family {Anthonomus granis Boheman
(Coleptera: Curculionidae)) is known to have an acidic midgut and the same could apply
to the MWHW based on these results (Nation 2001). Therefore, this species and possibly
other weevils may not be affected by compounds like L-methionine because of the lack
of an alkaline midgut needed for the CAATCH1 protein to operate (Feldman et al. 2000;
Quick and Stevens 2001). Again, further research is necessary to determine if
CAATCH1 proteins are present in this weevil species.
The greenbug parasitoid also was unaffected by exposure to the excess
L-methionine found on treated leaves infested with aphids. Dadd and Krieger (1968)
found higher methionine requirements for the greenbug Myzus persicae Sulzer
(Hemiptera: Aphididae) when cysteine is scarce because of its ability to transform excess
methionine to much needed sulfur and could possibly explain the parasitoids tolerance to
high methionine concentrations. Because of the life cycle of the GBP, and many other
parasitoids, direct contact with compounds such as L-methionine would occur inside the


22
end of the trial (using an Olympus Tokyo Model 213598 stereomicroscope with a optical
micrometer) to monitor larval development
Trials to determine the total amount of L-methionine applied to excised leaves
also were included to quantify how much of the amino acid was physically present on
leaves at the different concentration levels. Leaves were weighed before dipping into the
control (0%) and L-methionine solutions (0.1%-10%), allowed to air dry for 30 min and
weighed again. The difference was assumed to be the actual amount of L-methionine
residue on the leaf. This value then was used to determine the total amount of
L-methionine on the leaf surface of the excised leaves and the amount of L-methionine
consumed per gram of leaf material, based on calculations of the physical amount of the
compound for each % concentration.
Preference Tests
It was unknown if the additional methionine acted to attract or repel larvae.
Neonate larvae were used in the choice tests to determine if there was a preference
between the control (deionized H2O) and the Treatments (1.0% L-methionine). Leaves
were obtained from potted plants maintained in the outdoor shade house. The tests
consisted of 4 leaf disks (30 mm diameter) dipped into the control solution and placed
into the chamber alternately with four leaf disks (30 mm diameter) dipped into the
treatment solution and replicated with a total of 10 chambers. Each chamber consisted of
a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The
filter disk was moistened routinely with deionized H2O to prevent the leaf disks from
desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental
constants described previously. The leaf disks also were scanned photometrically and


32
Figure 3-9. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (njota 160) on excised eggplant leaves for feeding and
development trials. Proline (1.0%) and Btk were included for comparison
as positive and negative controls. Data were adjusted using Abbotts
formula for control mortality. Note the overlap in the 0.7% L-methionine,
1.0% L-methionine and Btk treatments at Day 1.


85
After completing her Ph.D. program, Kanjana plans to work as a postdoctoral
researcher to gain more experience in this subject area. Later, she will return to Thailand
to fulfill an appointed position as an associate professor at Siam University. Kanjana will
teach and conduct research. She hopes to deliver the excitement and enthusiasm to her
students in Thailand that she has experienced from her professors in the U.S.A.


41
Nematology green and shade houses. Excised leaves were dipped in solutions of
deionized H2O containing different concentrations of methionine and held in the clear
plastic boxes and held at the aforementioned environmental conditions (Figure 3-1).
Additional treatments of proline (1.0%) and Bt-tenebrionis (Novodor FC @12.4 mL/L;
Valent Biosciences, Libertyville, IL) were included as positive and negative controls,
respectively. Survivorship data were pooled from several different trials for data
analysis.
Feeding and Development
To test L-methionine on the developmental rates of CPB, larvae were exposed to
excised eggplant leaves dipped in different concentrations of L-methionine under the
same conditions as the survivorship trials. Additional treatments of proline (1.0%) and
Btt were included as positive and negative controls, respectively. Leaves were scanned
photometrically using the Cl 203 Area Meter with conveyor attachment (CID, Inc.,
Camas, WA) before exposure to the larvae and measuring after leaf consumption. The
difference in leaf areas resulting from the missing leaf tissue was assumed to be the
amount eaten by the developing larvae. Larval head capsule widths were measured at the
time of death or the end of the trial (using an Olympus Tokyo Model 213598
stereomicroscope with an ocular micrometer) as an evaluation of larval development.
Preference Tests
It was unknown if the additional methionine acted to attract or repel larvae.
Neonate larvae were used in the choice tests to determine if there was a preference
between the Control (deionized H20) and the treatments (1.0% L-methionine). Leaves
were obtained from potted plants maintained in the outdoor shade house. The tests


100
synthesis of homocysteine to produce methionine to the presence of methionine rich
hexamerins and allophorins and protein synthesis, the role of methionine in plant-insect
interactions may be larger than originally theorized.
The production of methionine overproducing plants could also be used in future
IPM strategies. Preliminary results indicate that genetically modified plants do produce
enough methionine to affect the survivorship of caterpillars feeding on the plant
(unpublished data). This could be used in crops in which improved nutritional quality is
important as well as the insecticidal properties of the additional methionine. However,
there appears to be a sublethal level (0.1%) of L-methionine in which THW and CPB can
tolerate and survive with little mortality (Figures 3-9 and 4-1). Any system that makes
use of a crop that can overproduce compounds like L-methionine would have to be able
to express levels greater than this level to avoid any resistance/tolerance.
This research has also provided more possibilities for the use of compounds such
as L-methionine in the YFM portion of this study. The amino acid Beta-alanine provided
similar levels of control, as did the methionine trials (Figure 5-7). Although unexpected
(as discussed in Chapter 5), it shows that there are several other systems that can possibly
be exploited in controlling some insects.
Further research is necessary to determine if the combination of a compound like
methionine and a pesticide already in use would result in the increase in toxicity or the
decrease in the concentration of pesticide used. If compatibility between methionine and
Bacillus thuringiensis does exists, then it is possible that resistance could be broken in a
given population. For example, if a population of THW started to show resistance to
Bacillus thuringiensis kurstaki then methionine could be used to remove both susceptible


31
diet at 24 h and dropped to 0.4% (19.9 mg/g leaf material) at 48 h and 0.25% (12.8 mg/g
leaf material) after 72 h exposure. Overall, the LC50 at the end of the experiment for the
natural diet was well below the value for the artificial diet, with close to a 90% reduction.
Feeding and Development
Mortality of THW for the developmental tests ranged from approximately 30%
for the 0.1% L-methionine treatment and over 40% for the proline treatment (Figure 3-9).
Complete mortality for the 0.3% L-methionine occurred after 7 days while the 0.5%
L-methionine treatment took only 5 days. The Btk treatment mortality was similar to the
0.7% L-methionine and 1.0-%L-methionine treatment, resulting in 100% mortality after 1
day of exposure to the amino acid. Both the mean head capsule width and amount of leaf
material consumed showed significant differences between treatments, with the control,
0.1% L-methionine and proline treatments being different that the remaining treatments
(Figures 3-10 and 3-11).
Preference Tests
The amount of control and 1.0% L-methionine leaf tissue consumed during the
preference tests were found not to be statistically different (Figure 3-12). In addition to
the amount of leaf material consumed between treatments not being different, the mean
head capsule width (i.e., development) showed a correlation with the amount of control
diet consumed (Pearson Correlation Coefficient 0.885, P<0.001) while no correlation to
the Treatment diet consumed (Pearson Correlation Coefficient 0.630, P=0.051) (Figure
3-11).


103
Bills, P.S., D. Mota-Sanchez and M. Whalon. 2004. The Database of Arthropods
Resistant to Pesticides. Michigan State University Center for Integrated Plant
Systems. Internet URL: http://www.cips.msu.edu/resistance/nndb/. Accessed
April 2004.
Boucher, T. J. 1999. Using IPM on CPB saves money, insecticides. Yankee Grower
1(2): 7-9.
Bourgis, F., S. Roje, M.L. Nuccio, D.B. Fisher, M.C. Tarczynski, C. Li, C. Herschbach,
H. Rennenberg, M.J. Pimenta, T. Shen, D.A. Gage and A.D. Hanson. 2000. S-
methymethionine has a major role in pholem, sulfer transport and is synthesized
by a novel methyltransferase. Pp. 283-284. IN C. Brunold (ed.), Sulfiir Nutrition
and Sulfur Assimilation in Higher Plants. Paul Haupt, Bern, Switzerland. 427pp.
Bownes, M. and L. Partridge. 1987. Transfer of molecules from ejaculate to females in
Drosophila melanogaster and Drospohila pseudoobscura. J. Insect Physiol.
33(12): 941-947.
Brogdon, W.G and J.C. McAllister. 1998. Insecticide resistance and vector management.
Emerging Infect. Diseases 4(4): 605-613.
Capinera, J.L, F.D. Bennett and D. Rosen. 1994. Introduction: Why biological control
and IPM are important to Florida, pp.3-8. In D. Rosen, F.D. Bennett and J.L.
Capinera (eds.), Pest Management in the Subtropics: Biological Control- a Florida
Perspective. Intercept Limited, Andover, UK. 737pp.
Center, T.D. 1994. Biological control of weeds, Chapter 23. pp.481-521. IN: D. Rosen,
F.D. Bennett, J.L. Capinera, (eds.), Pest Management in the Subtropics:
Biological Control-The Florida Experience. Intercept, Ltd., Andover, Hampshire,
UK. 737pp.
Center, T.D., F.A. Dray and V.V. Vandriver, Jr. 1998. Biocontrol with insects: The
water hyacinth weevils. Florida Cooperative Extension Service, Institute of Food
and Agricultural Sciences, University of Florida. Internet URL:
http://edis.ifas.ufl.edu/scripts/htmlgen.exe?body&DOCUMUMENT_AG01.
Accessed April 2004.
Centers for Disease Control (CDC). 2003. Malaria: General Information. Centers for
Disease Control. Internet URL: http://www.cdc.gov/travel/malinfo.htm. Accessed
April 2004.
Chen, P.S. 1958. Studies on the protein metabolism of Culexpipens L.-I. Metabolic
changes of free amino acids during larval and pupal development. J. Ins. Physiol.
2:38-51.
Cibula, A.B., R.H. Davidson, F.W. Fisk and J.B. LaPidus. 1967. Relationship of free
amino acids of some Solanaceous plants to growth and development of


74
possess the direct structural segments needed to serve as precursors to the newly
identified norisoprenoids.
To demonstrate that carotenoids could serve as norisoprenoids precursors,
(3-carotene was studied in a model system at 35C storage. GC-0 and GC-MS data
confirmed the presence of (3-cyclocitral and (3-ionone in these solutions in as little as two
weeks. This was direct proof that (3-carotene can degrade to form specific norisoprenoids
under conditions an orange juice might encounter.


19
reestablished within 2 min and reequilibrated for 15 min before the next injection. Flow
rate was 1 mL/min and injection volume was 10 pL.
Results and Discussion
Carotenoids of Interest
Although over 50 carotenoids have been identified in orange juice, only a few
possess the structural requirements to produce potent norisoprenoids. The structures of
the carotenoids which have been shown to produce norisoprenoids of interests in other
food systems (15, 23, 31, 39, 69, 70, 78) are shown in Fig. 3-2. Hydrolysis points are
indicated with arrows and resulting norisoprenoid indicated as text.
Figure 3-2. Carotenoid precursors of selected norisoprenoids including neoxanthin, the
indirect precursor of P-damascenone.
It is worth noting that the structures of the left half of the first four carotenoids are
identical. Each of these four carotenoids can produce either 3-cyclocitral or (3-ionone.


4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to excised
eggplant leaves treated with various concentrations of L-methionine
(niotai=320) 46
4-4. Total leaf area consumed by Colorado potato beetle larvae exposed to excised
eggplant leaves treated with various concentrations of L-methionine
(nrotar=320) 48
4-5. Mean leaf consumption by Colorado potato beetle in the preference tests 49
5-1. Bioassay setup for yellow fever mosquito larvae exposed to various concentrations
of amino acids and Bti 55
5-2. Mortality of yellow fever mosquito larvae exposed to various concentrations of
L-methionine (nTOtai=240) 57
5-3. Mortality of yellow fever mosquito larvae exposed to various concentrations of
D-methionine (nrotai=240) 58
5-4. Mortality of yellow fever mosquito larvae exposed to various concentrations of Tris-
buffered L-methionine (nrotai~240) 60
5-5. Mortality of YFM larvae exposed to various concentrations of Proline (nrotai=240) 61
5-6. Mortality of yellow fever mosquito larvae exposed to various concentrations of
L-leucine (nTotai=240) 62
5-7. Mortality of YFM larvae exposed to various concentrations of Beta-alanine
(nTotai=240) 63
5-8. Mean head capsule widths of yellow fever mosquito larvae exposed to various Tris
buffered (7.0 pH) concentrations of L-methionine (niotai=320) 64
5-9. Concentrations (%) resulting in 50% mortality (LC50) of yellow fever mosquito
larvae exposed to various amino acids after 10 days (nrotar=240 for each amino
acid) 65
6-1. Overview of the design layout used to study the effects of L-methionine and Silwett
L-77 solutions on yield of eggplant 72
6-2. Weed Systems, Inc. KQ 3L CO2 backpack back sprayer used for application of
L-methionine and Silwett L-77 solutions 73
6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant leaves treated
with L-methionine and the adjuvant Silwett L-77 (nTotai=120) 75
6-4. Effects of L-methionine and Silwett L-77 on eggplant yield (A) and mean weight
in grams of fruit (B) from 09 June to 31 August 2001 76
IX


18
interaction in a variety of ways, including artificial diet, natural diet (excised leaves,
whole plant, and choice tests. The purpose of this portion of this study was to determine
whether L-methionine was detrimental to the survival and development of the THW and
to determine if L-methionine could be used to control this species.
Materials and Methods
Eggs of THW were obtained from the insectary of North Carolina State
University, and were held in 26.4L x 19.2W x 9.5H (cm) clear plastic rearing chambers
with a hardware cloth (to facilitate cleaning) (Figure 3-1). Florida Reach-In Units
(FRIUs) were used to control the environment for the rearing containers (Walker et al.
1993) Containers were held at 27 C, 60% relative humidity, and a 16L:8D photoperiod
in FRIUs with either artificial or natural diet (excised eggplant leaves or whole plants)
depending on the pending experiment. Neonates were allowed to feed for 2 days after
eclosin before being transferred to treatment groups. A camel hair brush was used for
transferring larvae, to minimize the risk of damage.
Diets and Survivorship
The artificial diet was prepared using the procedures outlined in Baumhover et al.
(1977) with the inclusion of L-methionine for the treatment concentrations of 0.1%,
0.3%, 0.5%, 1.0%, 3.0%, 5.0% and 10.0% (wt/wt). The artificial diet was changed on a
regular basis to prevent desiccation and fungal growth. Larvae were exposed to the
artificial diet in the clear plastic rearing chambers with a hardware cloth, and kept in the
FRIUs programmed with the aforementioned environmental constants.
Natural diets consisted of excised eggplant leaves (Solarium melongena
L.,Classic variety) of potted plants grown and maintained at the University of Florida,


14
selective for terpenes than early eluting alcohols and aldehydes (6). Polyacrylate was
more effective in extracting highly polar compounds such as methanol and ethanol (66).
Orange Juice Norisoprenoids
Only a single norisoprenoid (P-ionone) has been reported and completely identified
in fresh orange juice (4, 5). Recently, P-damescenone has been reported in heated orange
juice, but not completely identified (65). With so many carotenoid precursors present in
orange juice, it seems highly likely that additional norisoprenoids would also be present.
The primary objective of this study was to determine if these additional norisoprenoids
were present in orange.


59
intensity, their contribution to total juice aroma were 7.8, 7.6, and 8.7 % in fresh,
pasteurized, and reconstituted from concentrate respectively. Fig. 5-7 illustrates aroma
group profiles from the total intensity in similar aroma quality groups from GC-0
aromagrams, comparing among fresh, pasteurized, and reconstituted from concentrate.
Comparing fresh and pasteurized orange juice, the major aroma category differences
were in the metallic/mushroom/geranium and in roasted/cooked/meaty/spice categories
which were higher in pasteurized juice than fresh juice. This difference come from three
addition aroma active compounds (E-2-octenal, and two unknown at LRI 1100 and LRI
1128) in metallic/mushroom/geranium category and two additional unknown ( LRI 1317
and LRI 1461) in roasted/cooked/meaty /spice category in pasteurized juice (Table 5-5
and Table 5-6). In fresh juice citrusy/minty category was higher than pasteurized juice.
The differences were the higher intensity of compounds in citrusy/minty category in fresh
juice than in pasteurized juice. The contribution of floral quality was the same in fresh
and pasteurized orange juice. From the aroma group profile of reconstituted from
concentrate there were many volatile compounds lost due to the thermal evaporation
process.
Fig. 5-8 shows the contribution of aroma active compounds in just the floral
category. This group includes linalool, p-cyclocitral, P-damascenone, a-ionone,
P-ionone and one unknown (LRI 1255) generated after thermal processing. The
norisoprenoids in orange juice contribute the majority of floral aroma in floral category,
specifically 78, 78, and 59% in fresh, pasteurized and reconstituted from concentrate
respectively.


Copyright 2004
by
Kanjana Mahattanatawee


LIST OF FIGURES
Figure Pagi
2-1. The CAATCH1 system identified from the midgut of the tobacco homworm 15
3-1. Rearing chamber for tobacco homworm and Colorado potato beetle larvae used in
the artificial and excised leaf diet tests 19
3-2. Setup for whole plant studies involving tobacco homworm 21
3-3. Chambers used for tobacco homworm and Colorado potato beetle preference tests 23
3-4. Amount of L-methionine present on leaf surface after treatment 25
3-5. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (n-rotar^O) in artificial diet 26
3-6. Survivorship of THW larvae exposed to various concentrations of L-methionine
(nTota 1,540) on excised eggplant leaves 28
3-7. Mortality of tobacco homworm larvae exposed to various concentrations of
L-methionine (nrotai=256) on whole plants 29
3-8. Concentrations (%) of L-methionine required for the mortality of 50% of test
population of tobacco homworm after 9 days exposure (nT0tai=1,540; n=180
for 3.0% L-methionine 10.0% L-methionine, n=200 for remainder) 30
3-9. Mortality of tobacco homworm larvae exposed to various concentrations of L-
methionine (ntota 160) on excised eggplant leaves for feeding and
development trials 32
3-10. Mean head capsule widths of tobacco homworm larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nTOtai=320) 33
3-11. Total leaf area consumed by tobacco homworm larvae exposed to excised eggplant
leaves treated with various concentrations of L-methionine (nrotai=320) 34
3-12. Mean leaf consumption by tobacco homworm in the preference tests 35
4-1. Mortality of Colorado potato beetle larvae exposed to excised eggplant leaves treated
with various concentrations of L-methionine (nTOtai=560) 44
4-2. Concentrations (%) of L-methionine concentrations required for the mortality of
50% of the test population of Colorado potato beetle after 8 days exposure
(niotai=220) 45
vui


48
fortified juices was indicated when little to no increase in peak area was observed with
additional exposure time. The equilibrium time for P-cyclocitral, P-damascenone,
a-ionone, and P-ionone were 75, 90, 115, and 120 min. respectively (Fig. 5-1). The
Figure 5-1. Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 40C, = P-cyclocitral, = P-damascenone, A= a-ionone,
= P-ionone.
results show that the time needed to reach equilibrium depends on the polarity and the
relative molecular mass of each norisoprenoid. Since 75-120 minutes to reach the
equilibrium would be too long to wait for practical purposes and may alter the volatile
profiles from thermally induced reactions, a shorter exposure time was chosen for routine
analyses. It can be seen from the adsorption curves for each compound that 45 min.
represents a rough compromise for all four analytes between minimal exposure time and
maximum peak area. For example P-damascenone and P-ionone reaching more than
80% of their final equilibrium value within 45 min. It is a rare SPME analysis that
employs true equilibrium exposure time. If exposure time can be carefully controlled,
then exposure times of as little as 5 min. can be employed. These very short exposure


HPLC Separation
Carotenoid Identification 23
Conclusions 25
4 IDENTIFICATION OF NORISOPRENOIDS IN ORANGE JUICE USING
TIME INTENSITY GC-0 AND GC-MS 26
Introduction 26
Objectives 28
Materials and Methods 28
Orange Juice Samples and Processing 28
Chemicals 29
Orange Juice Headspace Extraction 29
Gas Chromatography: GC-FID and GC-Olfactometer 30
Gas Chromatography-Mass Spectrometry 31
Aroma Peak Identification 32
Results and Discussion 32
Extraction and Concentration of Juice Norisoprenoids 32
GC-Olfactometry 35
Mass Spectrometry Norisoprenoid Identifications 36
Conclusion 41
5 QUANTIFICATION AND DETERMINATION OF THE RELATIVE IMPACT
OF NORISOPRENOIDS IN ORANGE JUICE 43
Introduction 43
Objectives 45
Materials and Methods 45
Quantification of Norisoprenoids in Orange Juice 45
Determination of the Relative Impact of Norisoprenoids in Orange Juice 47
Results and Discussion 47
Quantification of Norisoprenoids in Orange Juice 47
Norisoprenoid Quantification using Standard Additions 50
Determination of Relative Aroma Impact of Norisoprenoids 54
Norisoprenoid Contribution to Total Floral Aroma 58
Conclusion 61
6 THERMAL DEGRADATION OF BETA-CAROTENE IN MODEL
SOLUTION 62
Introduction 62
Objective 63
Materials and Methods 63
Crystallization 63
Model Solutions 64
Analytical Methods 64
Results and discussion 65
vi


86
compared with a One-way ANOVA and mean separation was performed using Tukeys
Multiple Comparison test (Zar, 1999).
Lvsiphlebus testaceipes
To test the effects of methionine on the GBP, cotton plants (Gossypium sp.;
Family: Malvacae) were grown and maintained at the University of Florida, Department
of Entomology and Nematology green and shade houses from 07 October 2002 to 25
November 2002. Aphids (A. gossypii Glover) were supplied from other experiments
using this organism and kept on plants within a sealed greenhouse to prevent unwanted
parasitism. Plants were maintained in the sealed greenhouse, infested with aphids and
then placed in the open shadehouse area to encourage parasitation. In total, 20 plants
were used for 2 treatments, 1.0% L-methionine and 0% L-methionine (Control) mixed
with deionized H20. Plants were sprayed weekly (12 October 2002 through 17
November 2002) with approximately 10 ml of solution using a hand-held spray bottle.
Counts of parasitized aphids began approximately two weeks after placing plants outside
to ensure adequate time for parasitism (Royer et al. 2001). Counts were made using a
hand lens and counter; mummies with exit holes were enumerated and removed. A
few parasitized aphids were removed and held in glass vials to ensure correct
identification of the parasitoid.
Data Analysis
Data from the parasitoid experiments were analyzed using Minitab Version 12
(Minitab, Inc.; State College, PA). Control and experimental plants were compared
against one another with a One-way ANOVA and separation of significant means was
performed with Tukeys Multiple Comparison test (Zar, 1999).


106
Fogarty International Center and the U.S. National Institutes of Health (FIC-NIH). 2003.
Multilateral Initiative on Malaria. U.S. National Institutes of Health. Internet
URL: http://mim.nih.gov/english/index.html. Accessed April 2004.
Forgash, AJ. 1985. Insecticide resistance in the Colorado potato beetle, pp. 33-52. IN
D.N. Ferro and R.H. Voss (eds.) Proceedings of the Symposium on Colorado
Potato Beetle. XVII International Congress of Entomology, Massachusetts
Agricultural Experiment Station Bulletin 704. Amherst Massachusetts.
Friend, W.G., R.H. Backs and L.M. Cass. 1957. Studies on amino acid requirements of
larvae of the onion maggot, Hylema antiqua (MG.), under aseptic conditions.
Can. J. Zool. 35: 535-543.
Gasnier-Fauchet, F. and P. Nardon. 1986a. Comparison of sarcosine and methionine
sulfoxide levels in symbiotic and aposymbiotic larvae of two sibling species,
Sitophilus oryzae and Sitophilus zeamais (Coleptera: Curculionidae). Insect
Biochemistry 17(1): 17-20.
Gasnier-Fauchet, F. and P. Nardon. 1986b. Comparison of methionine metabolism in
symbiotic and aposymbiotic larvae of Sitophilus oryzae L. (Coloeptera:
Curculionidae)- n. Involvement of the symbiotic bacteria in the oxidation of
methionine. Comp. Biochem. Physiol. 58(1): 251-254.
Gauthier, V.L., R.N. Hoffinaster and M. Semel. 1981. History of Colorado potato beetle
control, pp. 13-34. IN J.H. Cashcomb and R. Casagrande (eds.), Advances in
Potato Pest Management. Hutchinson and Ross, Stroudsburg, PA. 672pp.
Geer, B.W. 1966. Utilization of D-amino acids for growth by Drosophila melanogaster
larvae. J. Nutr. 90: 31-39.
Giordana, B, M. Forcella, M.G. Leonardi, M. Casartelli, L. Fiandra, G.M. Hanozet and P.
Parenti. 2002. A novel regulatory mechanism for amino acid absorption in
lepidopteran larval midgut. J. Insect Physiol. 48: 585-592.
Giovanelli, J, S.H. Mudd and A.H. Datko. 1980. Sulfur amino acids in plants. IN: B.J.
Mifhn (ed.) The Biochemistry of Plants VoL 5, Academic Press, New York, pp.
453-505.
Giroux, S., R.M. Duchesne and D. Coderre. 1995. Predation of Leptinotarsa
decemlineata (Coleptera: Coccinellidae) by Coleomegilla maculata (Coleptera:
Coccinellidae): Comparative effectiveness of predator developmental stages and
effect of temperature. Environ. Entomol. 24: 748-754.
Glare, T.R, and M. O Callaghan, 1998. Environmental and Health Impacts of Bacillus
thuringiensis isrealensis. Report for the New Zealand Ministry of Health, 58pp.


51
The higher concentrations of L-methionine that produced mortality similar to the
Btt is encouraging considering the occurrence of resistance to this compound seen in
many pest insect species because of reduced receptor activity and binding (Bills et al.
2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and
many are the result of exposure to a combination of different pesticide classes. The
Methionine-CAATCHl system could be exploited in cases where the only alternative is
applying different pesticides or using higher rates to break resistance. Further research is
needed to determine compatibility with Bt and L-methionine for cases in which resistance
is observed in natural populations. Given the safety of L-methionine and the shorter time
required for 100% mortality (when compared to Btt), this compound could represent a
new biorational tool for the management of the CPB.


Copyright 2004
by
Lewis Scotty Long


Relative Abundance
70
MS Identification
100=
8CT
60
40
20J
2913
19.62
18,13 1B8.3 19.3 100a
ni/z = 137
500:
24.30
m/z = 151
25.13
m/z = 177
i. i????,.y?fiy
18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5 26.0
Time (min)
Figure 6-6. Selected ion chromatogram (SIC) of model solution headspace volatiles after
storage 2 weeks at 35C : A = (3-cyclocitral, B = P-homocyclocitral, C = (3-
ionone.
m/z
Figure 6-7. Upper spectra from model solution MS at RT 19.61, bottom spectra from
standard P-cyclocitral using identical ion trap MS at identical retention time.


36
Table 4-1. Identification, retention characteristics and aroma descriptions of aroma
active compounds in fresh orange juice
No.
Compound
Aroma descriptor
Linear retention
Relativec
intensity
ZB-5
DB-wax
1
Terpinen-4-olb
Metallic, musty
1175
1619
5
2
Z-4-decenala
Green, metallic, soapy
1188
1542
7
3
Decanalb
Green, soapy
1198
1508
7
4
(E,E)-2,4-nonadienala
Fatty, green
1209
1702
7
5
3-cyclocitralb
Mild floral, sweet, hay-like
1214
1632
6
Nerol3
Lemongrass
1222
1798
5
7
Neralb
Lemongrass
1236
1692
7
8
L-carvoneb
Minty
1242
1747
8
9
Unknown
Metallic/ woody
1247
6
10
Geraniol3
Citrus, geranium
1265
1853
9
11
Unknown
Soapy, almond
1274
7
12
1 -p-menthene-8-thiola
Grapefruit
1281
1619
7
13
(E,Z)-2,4-decadienal3
Metallic, geranium
1293
1759
4
14
Geranial3
Green, minty
1310
1742
4
15
(E,E)-2,4-decadienal3
Fatty, green
1314
1819
4
16
a-teroinvl-acetate3
Sweet
1349
1663
6
17
4,5-epoxy-E-2-decenal3
Metallic, fatty
1375
2010
6
18
Unknown
Sweet nutty
1380
7
19
(3-rlflma Tobacco, apple, floral
1383
1829
7
20
Dodecanal3
Soapy
1403
1722
5
21
a-iononeb
Floral
1426
1863
8
22
Unknown3
Fermented, rancid butter
1459
5
23
3-iononeb
Floral, raspberry
1484
1951
24
Unknown
Nutty
1510
8
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as
compared with standard
b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as
compared with standard, and MS
c Averages of normalized intensities (10) evaluated by two trained panelists in four
replications
Mass Spectrometry Norisoprenoid Identifications
Headspace volatiles from fresh orange juice were analyzed using capillary GC with
an ion trap mass spectrometer. To achieve greater selectivity for the norisoprenoids of
interest, selected ion chromatograms were reconstructed in the retention region where
norisoprenoid standards were found to elute. The selectivity achieved is demonstrated in
Fig. 4-2. Specific m/z values were evaluated to provide the best peak height for each
norisoprenoid of interest as well as minimizing interference from non-norisoprenoid


42
consisted of 4 leaf disks (30 mm diameter) dipped into the Control solution and placed
into the chamber alternately with four leaf disks (30 mm diameter) dipped into the
treatment solution and replicated with a total of 10 chambers. Each chamber consisted of
a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The
filter disk was moistened routinely with deionized H2O to prevent the leaf disks from
desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental
constants described previously. The leaf disks also were scanned photometrically and
larval head capsule measurements made using the same procedures described in the
Feeding and Development section.
Data Analysis
Sample sizes of all experiments were chosen according to the guidelines
recommended by Robertson and Preisler (1991) for optimal sample size and data
analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were
performed using PROBIT Version 1.5 (Ecological Monitoring Research Division,
USEPA) after Abbotts correction for control mortality (Abbott 1925). Survival data
were normalized to the previous value when control mortality was greater than the
treatment mortality, to produce a smoother trend line. Statistical analysis was performed
on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the
data included One-way ANOVA and separation of significant means using Tukeys
Multiple Comparison and Pearson Correlation was performed on the choice trial data to
examine possible relationships between development and consumption of treated leaf
material (Zar 1999).


pests. The tobacco horn worm (THW), Colorado potato beetle (CPB) and the yellow
fever mosquito (YFM) were tested and found to be susceptible to concentrations greater
than 0.1%. Diets, both natural and artificial, containing this compound resulted in the
complete mortality of THW and also in the natural diet for CPB. Development and
feeding rates were also affected by the addition of L-methionine to diets for THW and
CPB. Survivorship and developmental rates of YFM were also affected by the addition
of this amino acid to the larval habitat.
In Chapter 6 it was found that the field application of L-methionine under natural
conditions was able to control CPB. It was also determined that L-methionine was
compatible with Silwett L-77, a commonly used adjuvant, and showed no detrimental
effects on crop yield of eggplant.
Finally, the application of a compound such as L-methionine has to be able to
control the pests that it is used against and not have an effect on beneficial organisms that
may come into contact with this compound. Chapter 7 detailed the results of tests that
involved various beneficial insects from different feeding guilds (herbivore, predator and
parasitoid) showed that L-methionine does not appear to pose a threat to nontarget
organisms.
One aspect of the use of a compound like L-methionine that is very important is
the relative safety. The health hazards related to the contamination of the environment
with pesticides are well documented and in the recent years have resulted of the review
and removal of several insecticides from commercial and private use. The use of
L-methionine as an insecticide would alleviate the dangers associated with other
pesticides. The approved use as a nutritional supplement for livestock feed is a testament


115
Insects. Lewis, along with fellow graduate student Jim Dunford, were awarded the
Outstanding Teacher Award by the Entomology and Nematology Student Organization of
the University of Florida for outstanding teaching accomplishments in the department.
While at the University of Florida, Lewis joined the U.S. Army Reserve as a
medical entomologist. He was assigned to the local Medical Detachments, and served
there from 2000 to 2004. Originally he had planned on graduating in 2003, but was
called to active duty with the 1469th Medical Detachment as a part of Operation Enduring
Freedom (OEF). Lewis was the OEF Theater entomologist, and served as the Executive
Officer (responsible for the deployment of personnel and equipment to South West Asia).
He was stationed at Kandahar Airfield, where he performed his duty and was awarded an
Army Commendation Medal for his work in protecting soldiers from health hazards and
diseases associated with the area. Lewis returned and continued his work toward
graduation.
Lewis was married in August 1992 to Karen Abbott, and is the father of Emilia
Irene (1994) and Bryan Scott (1997). Lewis plans on having a career in the military as a
medical entomologist, and all look forward to seeing the world and the rest of their
future.


34
peak area. For example (Tdamascenone and (3-ionone reaching more than 80% of their
final equilibrium value within 45 min. It is a rare SPME analysis that employs true
equilibrium exposure time. If exposure time can be carefully controlled, then exposure
times of as little as 5 min. can be employed. These very short exposure times are usually
limited to analytes in relatively high concentration and even then the reproducibility is
not that good. In this study, very short exposure times were not an option as the analytes
of interest were present in very low concentration.
Time (min)
Figure 4-1. GC-FID (top) and average time-intensity of four GC-0 runs by two panelists
(inverted, bottom) of fresh orange juice on ZB-5 column. Peaks 5, 19, 21 and
23 correspond to norisoprenoids, all numbers refers to compounds in Table
4-1


58
Table 5-11. Aroma active compounds in orange juice grouped by green/grassy
Compounds11
Description
L
RI
Relative
intensity
ZB-5
DB-wax
3-(Z)-hexen-l-ol / (E)-2-hexenalb
(E,Z)-2,6-nonadienal
Green, grass
Green
854
1148
1226/1391
1593
6C, T
r, 8d, r
a Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard
b Identified by linear retention index on ZB-5 and/or DB-wax, aroma description as compared with standard,
and MS
c fresh orange juice
d pasteurized orange juice
6 reconstituted from concentrate
1. Citrusy/minty
Figure 5-7. Aroma group profiles of fresh (), pasteurized (), and reconstituted from
concentrate () orange juice.
Norisoprenoid Contribution to Total Floral Aroma
All four of the identified norisoprenoids in orange juice have a general floral
aroma quality (Table 5-9). When all the relative intensities of these four norisoprenoids
were combined (in each type of orange juice) and compared with the total aroma


Peak area
52
Figure 5-4. Standard addition data for (3-ionone peak area vs. added concentration in
fresh orange juice. The regression line created by peak area at selected mass
177 vs. P-ionone concentration
P-damascenone concentration (ppm)
Figure 5-5. Standard addition P-damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190.


29
Chemicals
Standard aroma compounds were obtained from the following sources: methional,
ethyl 2-methylpropanoate, ethyl 2-methylbutyrate, 1-octanol, 2-acetyl-2-thiazoline,
(E,Z)-2,6-nonadienal, (E)-2-nonenal, (E,E)-2,4-decadienal, l-octen-3-one, ethyl
hexanoate, (E,E)-2,4-nonadienal, L-carvone, E-2 octenal, terpinolene, a-terpinyl
acetate, a-terpineol, Z-4-decenal, neral, geranial, 4,5-epoxy-E-2-decenal, P-ionone and
P-cyclocitral were purchased from Aldrich (Milwaukee, WI). Octanal, limonene,
linalool, nonanal, hexanal, decanal, dodecanal, 1,8 cineole, citronella, terpinen-4-ol,
P-sinensal, P-myrcene, nootkatone ethyl butyrate, acetaldehyde, geraniol, and nerol were
obtained as gifts from SunPure (Lakeland, FL). Alpha-ionone was obtained as a gift
from Danisco (Lakeland, FL). The (Z)-2-nonenal was found in purchase of (E)-2-
nonenal at the 5-10% level. The (E,Z)-2,4-nonadienal and (E,Z)-2,4-decadienal were
found in the purchase of (E,E)-2,4-nonadienal and (E,E)-2,4-decadienal respectively.
Their identities were confirmed by mass spectra, retention indices and odor qualities.
Beta-damascenone and p-l-Menthen-8-thiol were obtained from Givaudan (Lakeland,
FL). The 4-mercapto-4-methyl-2-pentanone and 4-mercapto-4-methyl-2-pentanol were
synthesized in our laboratory. The 3-mercaptohexan-l-ol was bought from Interchim
(Montlucon, France).
Orange Juice Headspace Extraction
A 10 mL aliquot of orange juice was added to a 40 mL glass vial containing a
micro stirring bar and sealed with a screw-top cap that contained a Teflon-coated septa.
The bottle and contents were placed in a combination water bath and stirring plate set at
40C, and gently stirred. After equilibrating for 45 min a SPME fiber (50/30 mm


53
Figure 5-6. Standard addition data of P-damascenone peak area vs. added concentration
in reconstituted from concentrate orange juice. GC- quadrupole mass
spectrometer in SIM mode at m/z 190.
The concentration of (3-damascenone in reconstituted from concentrate was calculated
from separate standard addition data (Fig. 5-6) as it was thought that the matrix would be
substantially different.
Table 5-2. Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique
Norisoprenoids
Concentration (pg/L)
Threshold (pg/L in water)3
OAV
(3-cyclocitral
145
5
25
P-damascenone
0.09
0.002
45
a-ionone
47
0.4
118
P-ionone
83
0.007
11857
a Buttery and Teranishi (50)
Table 5-3. Concentration of P-damascenone in fresh, pasteurized and reconstituted
concentrate
orange juice
concentration (pg/L)
OAV
Fresh
0.09
45
Pasteurized
0.18
90
Reconstituted
0.85
425


69
10 15 20 25
Time (min)
Figure 6-5. Headspace volatiles from (3-carotene in model solution pH 3.5, after storage
2 weeks at 35C : 1 = (3-cyclocitral, 2 = (3-homocyclocitral, 3 = P-damascone,
4 = unknown, 5 = (3-ionone, a = sweet/floral/hay-like, b = sweet/floral/hay
like, c = sweet/apple, d = sweet/raspberry, e = sweet.
Table 6-1. Aroma active compounds from (3-carotene thermal degradation in model
solution pH 3.8, storage at 35C for 2 weeks
Compounds
Aroma description
LI
RI
MS
ZB-5
DB-wax
. P-cyclocitral
Sweet, floral, hay-like
1228
1632
X
p-homocyclocitral
Sweet, floral, hay-like
1262
1780
X
, P-damascone
Sweet, floral
1425
1835
. P-ionone
Sweet, raspberry
1495
1960
X
comparison, it will be noted that neither (3-homocyclocitral or P-damascone was found in
orange juice. Their retention characteristics and aroma descriptors exactly matched that
of authentic standards, providing enough evidence for at least a tentative identification.
Positive identification of these compounds was achieved from the MS data. The MS
fragmentation patterns for the identified norisoprenoids are shown in the following
figures which should offer conclusive proof as to their identity.


8
L-canavanine exhibits a range of insecticidal effects in artificial diets when
exposed to the THW. Dahlman (1977) demonstrated a reduction in consumption of
artificial diet containing less than 1% canavanine (w/v) which resulted in a lower body
mass and increased developmental time to the adult stage. Fecundity and fertility also
was affected by L-canavanine. Rosenthal and Dahlman (1975) showed that
concentrations as low as 0.5 mM L-canavanine in the diets of the THW resulted in the
reduction of ovarial mass of adults, while Palumbo and Dahlman (1978) showed that
concentrations of L-canavanine in agar-based diets resulted in the reduction of
chorionated oocyte production in concentrations between 1.0 and 2.0 mM.
Under natural conditions, L-canavanine was found to retard development, and
increased the susceptibility of exposed larvae to biotic and abiotic mortality factors
(Dahlman 1980). However, field applications of L-canavanine were shown to be
impractical because of the expense involved in synthesizing L-canavanine from its
source, the jack bean (Canavalia ensiformis (L.) DC. (Family: Fabaceae)).
Other sources of L-canavanine (i.e., analogues and homologues) were sought in
an attempt to find a more practical source of the amino acid. Structural homologues of
canavanine were examined and found to contribute to pupal deformities (and to a lesser
degree, to mortality) (Rosenthal et al. 1998). Long-chain esters of L-canavanine were
found to be more toxic than the parent compound when injected or added to an artificial
diet exposed to last instar of THW specimens (Rosenthal et al. 1998). Adding amino
acids other than arginine (the parent compound to L-canavanine) to diets containing L-
canavanine increased deformities and mortality of THW larvae and was attributed to the
structure and position of the functional groups on the added compounds (Dahlman and
Rosenthal 1982). Although the THW has an effective means of degrading aberrant


56
Table 5-5. Aroma active compounds in orange juice grouped by
metallic/mushroom/geranium i
Compounds3
Description
LI
RI
Relative
intensity
ZB-5
DB-
wax
l-octene-3-one
Metallic, mushroom
974
1308
6C, 7d, 5e
(3-myrcene
Geranium,plastic
979
1163
r, 7d, 8e
Octanalb
Metallic, orange peel
998
1299
8C, 8d, 8e
E-2-octenal
Metallic, fatty, green
1052
1449
4d
Terpinoleneb
Metallic, citrusy
1070
1296
6c,5d
Unknown
Green, metallic
1100
6d
Unknown
Metallic, pungent
1128
5d
Z-2-nonenal
Green, metallic
1141
1515
4C, 6d, 5e
Terpinen-4-olb
Metallic, musty
1175
1619
5C, 6d
Unknown
Metallic, woody
1247
6C, 6d
(E,Z)-2,4-decadienal
Metallic, geranium
1293
1759
4C, 7d
4,5-epoxy-E-2-decenal
Metallic, fatty
1375
2010
6c,6d, 4e
Unknown
Aquarium, metallic
1589
5C, 6d
(3-sinensal
Aquarium
1698
2244
o
OO
u
OO
Table 5-6. Aroma active compounds in orange juice grouped by
roasted/cooked/meaty/spice
Compounds3
Description
L
RI
Relative
intensity
ZB-5
DB-
wax
Methional
Cooked potato
904
1464
8e, 8d, T
2-acetyl-2-thiazoline
Cooked jusmine rice
1104
1766
6C, 6d, T
Unknown
Spice
1317
6d
Unknown
Sweet, nutty
1380
7C, 8d
Unknown
Fermented, rancid
1459
5C, 7d
Unknown
Green, overipe orange
1461
4d,4e
Unknown
Nutty
1510
8C, 9d
Unknown
spice
1718
7C, 7d
Table 5-7. Aroma active compounds in orange juice grouped by fatty/soapy/green
Compounds3
Description
LRI
Relative
intensity
ZB-5
DB-
wax
Hexanal6
Green, fatty
794
1083
7C, 6d, T
1-octanol
Green, soapy
1065
1565
8C, 7d, 5e
E-2-nonenal
Soapy
1153
1542
8C, 10d,6e
(Z)-4-decenal
Green, metallic, soapy
1188
1542
7C, 7d
Decanalb
Green, soapy
1198
1508
7C, 8d, 5e
(E,E)-2,4-nonadienal
Fatty, green
1209
1702
T, 7d
Unknown
Soapy, almond
1274
T, 7d
(E,E)-2,4-decadienal
Fatty, green
1314
1819
4c,6d
Dodecanal
Soapy
1403
1722
5C, 6d


9
increase in concentration in the paste (50). Beta-ionone in tomatoes seems to be formed
mainly by an oxidative mechanism. It was not detected among the glycoside hydrolysis
products. The compound P-damascenone was shown to be produced in fruits from
hydrolysis of glycosides via an intermediate acetylenic compound megastigm-5-en-7-
yne-3,9-diol. This appears to be the final step in tomato volatile norisoprenoid formation
(51). Three experimental lines of tomato: a high-P-carotene line; a high-lycopene line;
and a low-carotenoid line were examined for their norisoprenoid content. In fresh
tomato, the high (3-carotene line produced the highest concentrations of P-ionone (17
pg/L, versus 1 pg/L in the low-carotenoid line) and P-cyclociral produced 30 pg/L in the
high carotene line versus 0 pg/L in the low-carotene line). Both norisoprenoids are
known biological or chemical degradation products of P-carotene. The high lycopene
line, however, did not show any significantly higher concentration of the expected
lycopene degradation products, 6-methyl-5-hepten-2-one, 6-methyl-5-hepten-2-ol, or
geranylacetone. It did show a significantly higher value for geranial (21 pg/L) compared
to that of the common line (12 pg/L). Geranial could be considered a lycopene-
degradation product (41).
Saffron
Safranal (monoterpene aldehyde, C10H14O) is the characteristic impact compound
of saffron (dried stigmas of Crocus sativus), formed in saffron during drying and storage
by hydrolysis of picrocrocin. Picrocrocin was the colorless glycoside of the aglycone,
4-hydroxy-2,6,6-trimethyl-l-carboxaldehyde-l-cyclohexene (HTCC), which was the
main substance responsible for the bitter taste of saffron. Safranal was not present in the
fresh stigma. Its concentration in saffron depended strongly on both the drying and


5
enzymatic and nonenzymatic pathways. Nonenzymatic cleavage reactions include photo
oxygenation (18), auto-oxidation (28, 29), and the thermal degradation processes (30,
31). The in vivo cleavage of the carotenoid chain is generally considered to be catalyzed
by dioxygenase (lipoxidase and peroxidase) systems and require molecular oxygen and
other cofactors for activity. The polyene chain of carotenes is readily oxidized, giving
rise to cyclic and acyclic compounds (often having an oxygen-containing functional
group on a trimethylcyclohexane ring, or an oxygen-containing functional group on the
allyllic side chain). Although all the in-chain double bonds seem to be vulnerable to
enzymatic attack, in actuality the formation of major fragment classes with 10, 13, 15 or
20 carbon atoms are most common (see Fig. 2-2). In fruit tissues, the bio-oxidative
cleavage of the 9,10 (9, 10) double bond seems to be the most preferred (15, 32, 33).
Figure 2-2. Major fragment classes of carotenoid biodegradation.
Norisoprenoids Formation from Carotenoids
Norisoprenoids can be generated from carotenoids via either direct cleavage or
cleavage followed by subsequent reactions. In the latter process, three steps are required
to generate an aroma compound from the parent carotenoid: 1) the initial dioxygenase
cleavage, 2) subsequent enzymatic transformations of the initial cleavage product giving


CHAPTER 8
SUMMARY AND DISCUSSION
The creation and implementation of Integrated Pest Management (IPM) strategies
to combat pest species were developed as a response to the economic losses associated
with the overuse of chemical control. However. IPM strategies are not widely used
because of the lack of alternatives and the ease of use of pesticides. This has resulted in
the resistance to pesticides in many insect species, including economic and medical pests.
In an effort to provide alternatives to traditional chemical control, biorational methods
have been investigated and one such avenue is the use of non-protein amino acids.
Chapter 2 covered the history of the use of non-protein amino acids as a pesticide,
and discussed the CAATCH1 system and the safety of L-methionine. Only a handful of
these amino acids have been investigated as a means of controlling insect pests but still
lack the practicality and cost effectiveness as current chemical control methods. Recent
discovery of a new midgut membrane protein, CAATCH1, has revealed a new possibility
in insect control. The CAATCH1 system works in alkaline conditions and responds to
different amino acids, mainly the reduction in ion flow after exposure to methionine, an
essential amino acid required for normal development and metabolism of many species
including humans. The use of a compound such as methionine would be an excellent
addition to the IPM arsenal because of its relative safety to vertebrates and warrants
further study as a pesticide.
Chapters 3,4, and 5 were dedicated to examining the effects of L-methionine, a
common analog of methionine, on three different economic and medically important
96


54
temperature (23C) to permit the amino acid to fully dissolve before the addition of the
larvae. An additional trial of L-methionine buffered with Tris to a pH of 7.0 using a
Fisher Scientific Accumet pH 900 was conducted to determine if mortality was attributed
to a change in pH or exposure to the L-methionine.
Larvae of YFM (third instar) were obtained from the mosquito colony maintained
at the Department of Entomology and Nematology, University of Florida. Larvae were
transferred to the treatment jars using a camel hair, with 10 larvae per replicate for a total
of 40 Iarvae/treatment and niotai=240 for each amino acid bioassay experiment (Figure
5-1). Approximately 0.5g of finely ground fish food was added to serve as a larval food
source and nylon window screen was used to cover the tops of the jar to prevent the
escape of any emerged adults. Jars were held at 23C on a dedicated laboratory bench
top for approximately one week. The numbers of larvae surviving were recorded each
day.
Growth and Development
This experiment used the same Materials and Methods as the bioassay portion
with the exception of neonate larvae instead of 3rd instars. Eggs were placed in a tray of
water and held at 23 C for 2 days after eclosin. Neonates were removed using a camel
hair paintbrush and placed into each jar, with 10 larvae per replicate for a total of 40
larvae/treatment (nTOtai~240). Larval exuviae or dead larvae were removed and used to
examine growth rates by measuring the head capsules. Larvae head capsule widths were
measured (using an Olympus Tokyo Model 213598 stereomicroscope with an ocular
micrometer) as an evaluation of larval development.


CHAPTER 3
EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE
TOBACCO HORNWORM, Manduca sexta, UNDER LABORATORY CONDITIONS
Introduction
Manduca sexta (L.) (Lepidoptera: Sphingidae), the tobacco homworm (THW), is
a widespread species considered an economic pest throughout North and South America.
The caterpillar is known for its voracious appetite. In Georgia, the THW was responsible
for between approximately $1.2 to $1.5 million in losses and costs for control annually in
tobacco from 1997 to 2001 (Jones and McPherson 1997; McPherson and Jones 2002). In
addition to its well-earned reputation as an agricultural pest of solanaceous crops, the
THW has shown to be resistant to common pesticides (such as endrin and endosulfan),
with the possibility of cross-resistance (Bills et al. 2004).
The THW also is very important to scientific research outside the arena of
economic entomology, with studies ranging from molecular-based research to ecological
and physiological research, mainly because of its availability and ease in culturing
(Dwyer 1999). One research area of interest to scientists involves the chemistry and
physiology of the midgut. Insect control (or the development of new insecticides) was
probably not the main purpose of the research that resulted in identifying the CAATCH1
protein, yet it became the basis of our research project.
Because little information is available on the insecticidal properties of
methionine, several baseline experiments were necessary to determine that concentrations
of this amino acid to test It also was necessary to test I .-methionine and THW
17



PAGE 1

EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL IMPORTANCE By LEWIS SCOTTY LONG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Lewis Scotty Long

PAGE 3

ACKNOWLEDGMENTS I thank Jim Cuda and Bruce Stevens for giving me the financial and intellectual freedom that made this work possible. I want to thank Jim for housing me in his lab and providing the facilities to perform this work, and Bruce for allowing me to take his initial work and elaborate on it as well as including me as a co-inventor of the research presented. Most of all, I would like to express my sincere appreciation to Judy Gillmore. Without her support and help this research would not have been completed. Judy was integral in every aspect of this endeavor and put up with more than her fair share of my "research". I extend heartfelt thanks to George Gerencser, James Maruniak, Simon Yu, and Susan Webb for serving as members of my supervisory committee. I would like to also thank Jim Lloyd, Jerry Butler, and Carl Barfield for all the experiences and knowledge shared. Finally, I want to express my deepest, eternal gratitude to my fellow graduate students Jim Dunford and Heather Smith, for providing support and guidance that only colleagues, intellectual equals, and close friends can give. I can only hope to repay them for their help by providing the same amount of support for their endeavors as they did mine. iii

PAGE 4

I dedicate this work to Karen, my wife and best friend. I thank her for putting up with living as a "graduate student" for the last 5 years in fulfillment of my childhood dream of being a "Doctor". She has been my pillar of support, and I would not have made it this far without her love and understanding.

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS iii LIST OF FIGURES viii ABSTRACT xi CHAPTERS 1 THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE CONVENTIONAL PESTICIDES THE ONLY ANSWER? 1 Introduction 1 Importance of IPM in Florida and Surrounding States 2 Problems Associated with Pesticide Misuse 4 Biorational CompoundsAn Alternative to Chemical Pesticides 5 2 HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL INSECT PESTS 7 Non-Protein Amino Acids 7 Essential Amino Acids 10 The Cation-Anion Modulated Amino Acid Transporter with Channel Properties (CAATCH1 ) System 9 Methionine 13 Research Objectives 16 3 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE TOBACCO HORNWORM, Manduca sexta, UNDER LABORATORY CONDITIONS 17 Introduction 17 Materials and Methods 18 Diets and Survivorship 18 Feeding and Development 20 Preference Tests 22 Data Analysis 24 Results 24 v

PAGE 6

Diets and Survivorship 24 Feeding and Development 31 Choice Tests 31 Discussion 36 4 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER LABORATORY CONDITIONS 39 Introduction 39 Materials and Methods 40 Survivorship 40 Feeding and Development 41 Preference Tests 41 Data Analysis 42 Results 43 Survivorship 43 Feeding and Development 43 Preference Tests 47 Discussion 47 5 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY CONDITIONS 52 Introduction 52 Materials and Methods 53 Bioassay 53 Growth and Development 54 Data Analysis 56 Results 56 Bioassay 56 Growth and Development 59 Discussion 66 6 EVALUATION OF L-METHIONINE UNDER NATURAL FIELD CONDITIONS69 Introduction 69 Materials and Methods 70 Preliminary Investigation of Silwet L-77 and L-methionine 70 Plot Design 70 Fruit Yield 71 Pest Introduction 71 Data Analysis 74 vi

PAGE 7

Results 74 Effects of L-methionine and Silwett L-77 on Colorado Potato Beetle Adults Under Laboratory Conditions 74 Effects of L-methionine and Silwett L-77 on yield 74 Survival of CPB larvae 74 Discussion 78 7 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE NON-TARGET SPECIES 82 Introduction 82 Materials and Methods 84 Coleomegilla maculata 84 Neochetina eichhorniae 85 Lysiphlebus testaceipes 86 Data Analysis 86 Results 87 Coleomegilla maculata 87 Neochetina eichhorniae 87 Lysiphlebus testaceipes 87 Discussion 87 8 SUMMARY AND DISCUSSION 96 LIST OF REFERENCES 102 BIOGRAPHICAL SKETCH 114 vii

PAGE 8

LIST OF FIGURES Figure Page 21. The CAATCH1 system identified from the midgut of the tobacco hornworm 15 31. Rearing chamber for tobacco hornworm and Colorado potato beetle larvae used in the artificial and excised leaf diet tests 19 3-2. Setup for whole plant studies involving tobacco hornworm 21 3-3. Chambers used for tobacco hornworm and Colorado potato beetle preference tests 23 3-4. Amount of L-methionine present on leaf surface after treatment 25 3-5. Mortality of tobacco hornworm larvae exposed to various concentrations of L-methionine (nTotar^O) in artificial diet 26 3-6. Survivorship of THW larvae exposed to various concentrations of L-methionine (nTotai = 1,540) on excised eggplant leaves 28 3-7. Mortality of tobacco hornworm larvae exposed to various concentrations of L-methionine (nrotai=256) on whole plants 29 3-8. Concentrations (%) of L-methionine required for the mortality of 50% of test population of tobacco hornworm after 9 days exposure (nrotai^ 1,540; n=180 for 3.0% L-methionine 10.0% L-methionine, n=200 for remainder) 30 3-9. Mortality of tobacco hornworm larvae exposed to various concentrations of Lmethionine (nr 0 tai = 160) on excised eggplant leaves for feeding and development trials 32 3-10. Mean head capsule widths of tobacco hornworm larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (n To tai=320) 33 3-11. Total leaf area consumed by tobacco hornworm larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (nT O tai=320) 34 312. Mean leaf consumption by tobacco hornworm in the preference tests 35 41. Mortality of Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (n To tai=560) 44 4-2. Concentrations (%) of L-methionine concentrations required for the mortality of 50% of the test population of Colorado potato beetle after 8 days exposure (n To tai=220) 45 vin

PAGE 9

4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (n To tai = 320) 46 4-4. Total leaf area consumed by Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (moai=320) 48 45. Mean leaf consumption by Colorado potato beetle in the preference tests 49 51. Bioassay setup for yellow fever mosquito larvae exposed to various concentrations of amino acids and Bti 55 5-2. Mortality of yellow fever mosquito larvae exposed to various concentrations of L-methionine (nr o tai = 240) 57 5-3. Mortality of yellow fever mosquito larvae exposed to various concentrations of D-methionine (nTotai = 240) 58 5-4. Mortality of yellow fever mosquito larvae exposed to various concentrations of Trisbuffered L-methionine (nT O taf = 240) 60 5-5. Mortality of YFM larvae exposed to various concentrations of Proline (n T otai = 240) 61 5-6. Mortality of yellow fever mosquito larvae exposed to various concentrations of L-leucine (nTotai=240) 62 5-7. Mortality of YFM larvae exposed to various concentrations of Beta-alanine (n To tai=240) 63 5-8. Mean head capsule widths of yellow fever mosquito larvae exposed to various Tris buffered (7.0 pH) concentrations of L-methionine (n To tai = 320) 64 59. Concentrations (%) resulting in 50% mortality (LC 50 ) of yellow fever mosquito larvae exposed to various amino acids after 10 days (nxotaf^O for each amino acid) 65 61. Overview of the design layout used to study the effects of L-methionine and Silwett L-77 solutions on yield of eggplant 72 6-2. Weed Systems, Inc. KQ 3L C0 2 backpack back sprayer used for application of L-methionine and Silwett L-77 solutions 73 6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant leaves treated with L-methionine and the adjuvant Silwett L-77 (n To tai = 120) 75 6-4. Effects of L-methionine and Silwett L-77 on eggplant yield (A) and mean weight in grams of fruit (B) from 09 June to 3 1 August 2001 76 ix

PAGE 10

65. Mortality of Colorado potato beetle larvae on eggplants treated with L-methionine and Silwett L-77 87 71. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated artificial diet 88 7-2. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated cotton plant leaves infested with aphids 89 7-3. Feeding scars on water hyacinth (Eichhornia crassipes) leaf after exposure to Neochetina eichhorniae adults 90 7-4. Mortality of Neochetina eichhorniae on treated water hyacinth leaves 91 7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated with L-methionine and Proline 92 7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with L-methionine 93 x

PAGE 11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL IMPORTANCE By Lewis Scotty Long May, 2004 Chair: James P. Cuda Cochair: Bruce R. Stevens Major Department: Department of Entomology and Nematology Integrated pest management (IPM) strategies were developed in an effort to control pests with fewer pesticides. However, because of the misuse of pesticides and the failure to adopt IPM practices pesticide use is higher than ever. An alternative to conventional broad-spectrum pesticides is the use of biorational compounds; those that pose minimal risk to the environment and are specific to the target pests. The recent discovery of the CAATCH1 system in the midgut of the tobacco hornworm (THW), Manduca sexta, has revealed a novel means to control certain insect pests. This membrane protein works in alkaline conditions as both an amino acid transporter and also independently as a cation channel. However, the amino acid L-merhionine blocks amino acid transport thus altering the ion flow. xi

PAGE 12

Bioassays involving the tobacco hornworm, Colorado potato beetle (CPB), Leptinotarsa decemlineata, and the yellow fever mosquito (YFM), Aedes aegypti, were conducted to determine the insecticidal properties of this compound. L-methionine in artificial and natural diets resulted in the mortality of 50 to 100% in concentrations of 0.3% and higher for THW and CPB. Feeding rates and larval development also were affected with reduced levels (>0.1%) of L-methionine. Bioassay trials involving YFM larvae were similar, concentrations greater than 0.1% L-methionine produced mortality rates of 70 to 100%. All three species responded better to higher concentrations of Lmethionine than to Bacillus thuringiensis, the most commonly used and commercially available biorational pesticide. Field trials and non-target tests also were performed to determine L-methionine effectiveness under natural settings and safety to other organisms. Eggplant yield was not reduced by the application of L-methionine, which effectively controlled CPB larvae on the plants. Furthermore, several beneficial insects that were tested (a predator, a herbivore, and a parasitoid) were not affected by the addition of L-methionine to their diets. Based on these results, L-methionine was found to be effective in controlling selective agriculturally and medically important insect pest species, yet posed little threat to the crop plants applied to or to non-target organisms. The use of L-methionine as a pesticide, its relationship with insects and its possible use in delaying insecticide resistance were also examined. xu

PAGE 13

CHAPTER 1 THE INTEGRATED PEST MANAGEMENT DILEMMA: ARE CONVENTIONAL PESTICIDES THE ONLY ANSWER? Introduction Integrated Pest Management (IPM), the sustainable approach to the management of pest species using a combination of biological, chemical and cultural methods to reduce economic, environmental, and public health risk, was a result of economic losses associated with years of overuse of chemical control leading to resistance problems. The use of IPM strategies have certainly decreased pesticide usage and encouraged the use of methods that ensure a safer environment but many feel that it is not enough. After three decades of research efforts in the United States, IPM as it was envisioned in the 1970s was practiced on less than 8% of U.S. crop acreage based on Consumers Union estimates — well short of the national commitment to implement IPM on 75% of the total U.S. acreage by the end of the 1990s (Ehler and Bottrell 2000). This means that farm practices have changed little since the national IPM initiative was established in 1994 to implement biologically based alternatives to pesticides for controlling arthropod pests. It should be noted that the low percentage of IPM practices on commercial U.S. farmland may possibly be related to the lack of sufficient reporting means and actually may be higher than believed when the local growers and homeowners are included. However, the United States is considered the worlds' largest user of chemical pesticides, accounting for nearly 50% of total worldwide production and shows no sign of slowing (Deedat 1994). Pesticides remain the primary tool of pest consultants and farmers, because of the lack of economic incentives to adopt alternative strategies that require more effort to 1

PAGE 14

implement, produce unpredictable results, and require new knowledge (Barfield and Swisher 1994; Ehler and Bottrell 2000). Importance of IPM in Florida and Surrounding States Considerable effort has been devoted to developing IPM programs in Florida because of its unique pest problems and crop production systems, sensitivity to chemical pollutants, and increased urbanization (Capinera et al. 1994; Rosen et al. 1996). The necessity for developing IPM protocols for Florida's major plant and animal pests was underscored in a new statewide initiative. In November 1999, the Institute of Food and Agricultural Sciences (IF AS) at the University of Florida launched Putting Florida FIRST — Focusing IF AS Resources on Solutions for Tomorrow (Florida FIRST 1999). The Florida FIRST initiative was created (with input from stakeholders) to define the role of IF AS in shaping Florida's future in the 21 st century. Increasing concerns (expressed repeatedly by Florida's scientific community and the general public) about environmental contamination, food safety issues, and human and animal health problems resulting from the indiscriminate use (and often misuse) of pesticides are making existing methods for pest management obsolete. Successful implementation of "true" IPM, as it was envisioned by those who envisioned the original concept, will have the added benefit of helping Florida ". . enhance natural resources, provide consumers with a wide variety of safe and affordable foods, . provide enhanced environments for homes, work places and vacations, maintain a sustainable food and fiber system, and improve the quality of life. ." (Florida FIRST 1999). This effort to promote IPM programs in the state of Florida also benefits the surrounding states. For example, solanaceous crops produced in the southeastern U.S. (such as tomato, tobacco, eggplant, peppers and potato) are subjected to the same

PAGE 15

3 defoliation and fruit damage from various lepidopteran and coleopteran pests that also threaten Florida. The tomato pinworm [Keiferia lycopersicella (Walshingham) (Lepidoptera: Gelechidae)], armyworms [Spodoptera spp. (Lepidoptera: Noctuidae)], the Colorado potato beetle [Leptinotarsa decemlineata (Say) (Coleoptera: Chrysomelidae)], and hornworms [Manduca spp. (Lepidoptera: Sphingidae)] are some examples of pests that threaten both conventional producers and homeowners alike. For example, the estimated loss from and the cost of control of the tobacco hornworm, the number-one pest in tobacco crops in Georgia, reached $1 .5 (and $2.3 million), respectively, for the years 1996-1997 (Jones and McPherson 1997). From 1992-1998, tomato, eggplant, and pepper producing areas in the Southeast had a total of 1 ,247,000 pounds of endosulfan applied over 270,000 acres (Aerts and Neshiem 1999; Neshiem and Vulinec 2001). The cost of insecticides applied in Florida tomato production alone for 1993-1994 amounted to approximately $l,052/hectare for a total of $2.1 million; and rose to $2550/acre, totaling $103M for the 1996-1997 season (Aerts and Neshiem 1999; Schuster et al. 1996). The use of pesticides in Florida tomato production is high because tomatoes account for 30% of the total vegetable-crop value and 13% of the total vegetable acreage for the state, with 99% of production aimed toward the fresh market (Schuster et al. 1 996). For Florida potato producers, the cost of applying pesticides from 1 9951 996 was $1 1 .5M, and 96% of total Florida eggplant-crop acreage was treated with chemical insecticides (mainly methomyl and endosulfan) (Neshiem and Vulinec 2001). In addition to the monetary cost of pesticide use, commonly used insecticides such as endosulfan and fenvalerate show a high degree of toxicity to parasitoids of the tomato pinworm, thus negating the benefits of predation by natural enemies (Schuster et al. 1996). These figures may be the result of the "more is better" attitude of producers who want to avoid

PAGE 16

all risk of insect damage by using more applications and stronger pesticides (Schuster et al. 1996). Problems Associated with Pesticide Misuse The use of pesticides is not completely ruled out under IPM strategies, but rather IPM encourages responsible use to minimize environmental harm and to protect human safety and health (Deedat, 1994). However, the misuse (both intentional, in terms of "more is better;" and unintentional, as in agricultural runoff) also has resulted in resistance in some of the target pests. For example, surveys in North Carolina have shown that the Colorado potato beetle has become resistance to fenvalerate, carbofuran, and azinphosmethyl as a result of control failures in the field (Heim et al. 1990). Resistance to insecticides has also been observed in more than 450 arthropod pests (Romoser and Stoffolano 1998). Bills et al. (2004) found a 38% increase in the number of registered compounds used as pesticides from 1989-2000, and also a 16% increase in pesticide resistance of arthropod species worldwide. Losses are not limited to agricultural systems alone. Across Africa for example, populations of insecticide-resistant mosquitoes are the result of a variety of mechanisms, including exposure to pesticide residues in agricultural runoff, mutation of target sites, and migration of resistant populations into areas where there were no previous problem (FIC-NIH 2003). Parts of southwest Asia have seen a resurgence of malaria in some areas where it was considered eradicated (due to a combination of resistance and the economics associated with control of mosquito vectors) (Deedat 1994). The importance of this example becomes even more relevant when one considers that one million individuals die every year as a result of malaria, with upwards of 500 million cases per year (Centers for Disease Control 2003). The existence of other mosquito-borne diseases

PAGE 17

5 such as Dengue fever, yellow fever, and West Nile virus to name just a few, put countless millions more at risk. It would be dangerous to think that these diseases only occur in underdeveloped countries and not the United States. Integrated Pest Management practices also should be adopted for controlling the medical and veterinarian important insect vectors of these and other diseases. Biorational Compounds: An Alternative to Traditional Chemical Insecticides One way to reduce this reliance on traditional chemical pesticides and delay resistance is by increasing the variety and use biorational compounds. Biorational compounds are effective against selected pest species but are innocuous to nontarget or beneficial organisms; and have limited affect (if any) on biological control agents (Stansly et al. 1996). Biorational compounds include detergents, oils, pheromones, botanical products, microbes, and systemic and insect growth regulators (Perfect 1992; Wienzierl et al. 1998). Their safety lies in the low toxicity of the compound to nontarget organisms and the compound's short residual activity in the field. For example, Bacillus thuringiensis isrealensis (Bti) currently is one of the most widely used microbial pesticides for controlling aquatic dipteran pests (i.e., mosquitoes and black flies) because of its selectivity to this group and minimal nontarget effects (Glare and O'Callaghan 1998). However, resistance to Bt products has occurred in many species of lepidoptera from overuse of Bacillus thuringiensis kurstaki, and in some mosquito species to Bti, thus showing the need for alternatives to these compounds that are still effective (Brogdon and McAllister 1998; Marrone and Macintosh 1993). In addition to resistance, other problems are associated with the use of microbial control agents. Cook et al. (1996) discussed potential hazards, not properly identified in the planning stages, of displacement of native microorganisms, allergic responses in susceptible humans and

PAGE 18

6 animals, and eventual toxicity to nontarget organisms. Because of these problems, alternatives are needed to prevent another crisis like the one from which IPM originally arose.

PAGE 19

CHAPTER 2 HISTORY OF THE USE OF AMINO ACIDS AS A MEANS TO CONTROL INSECT PESTS Non-Protein Amino Acids One avenue of pest management explored in the field of biorational pesticides is the use nonprotein amino acids. Secondary plant materials such as these serve many functions in insect-plant relationships from attractants and repellents to crude insecticides (Dahlman 1980). Only a few nonprotein amino acids have been examined as a potential means to control insect pests. L-canavanine and its by-product of detoxification, Lcanaline, have been studied extensively, with a variety of effects ranging from developmental deformities to aberrant adult behavior (Dahlman and Rosenthal 1975; 1976; Rosenthal et al. 1995). L-canavanine is found mainly in leguminous plants, including several economic species (Bell 1978; Felton and Dahlman 1984). It is believed that plants produce this allelochemical for protection against feeding by phytophagous insects and herbivores (Rosenthal 1977). The mode of action for canavanine can be traced to several metabolic processes, including disruption of DNA/RNA and protein synthesis, arginine metabolism, uptake, anomalous canavanyl protein formation, and the reduction of active transport of K + in the midgut epithelium (Kammer et al. 1978; Racioppi and Dahlman 1980; Rosenthal 1977; Rosenthal et al. 1977; Rosenthal and Dahlman 1991). In contrast, canaline possesses neurotoxic characteristics with an unknown mode of action (Kammer et al. 1978). The species of choice for studies involving nonprotein amino acids has been the tobacco hornworm (THW), Manduca sexta (L.) (Lepidoptera: Sphingidae). 7

PAGE 20

8 L-canavanine exhibits a range of insecticidal effects in artificial diets when exposed to the THW. Dahlman (1977) demonstrated a reduction in consumption of artificial diet containing less than 1% canavanine (w/v) which resulted in a lower body mass and increased developmental time to the adult stage. Fecundity and fertility also was affected by L-canavanine. Rosenthal and Dahlman (1975) showed that concentrations as low as 0.5 mM L-canavanine in the diets of the THW resulted in the reduction of ovarial mass of adults, while Palumbo and Dahlman (1978) showed that concentrations of L-canavanine in agar-based diets resulted in the reduction of chorionated oocyte production in concentrations between 1 .0 and 2.0 mM. Under natural conditions, L-canavanine was found to retard development, and increased the susceptibility of exposed larvae to biotic and abiotic mortality factors (Dahlman 1980). However, field applications of L-canavanine were shown to be impractical because of the expense involved in synthesizing L-canavanine from its source, the jack bean (Canavalia ensiformis (L.) DC. (Family: Fabaceae)). Other sources of L-canavanine (i.e., analogues and homologues) were sought in an attempt to find a more practical source of the amino acid. Structural homologues of canavanine were examined and found to contribute to pupal deformities (and to a lesser degree, to mortality) (Rosenthal et al. 1998). Long-chain esters of L-canavanine were found to be more toxic than the parent compound when injected or added to an artificial diet exposed to last instar of THW specimens (Rosenthal et al 1998). Adding amino acids other than arginine (the parent compound to L-canavanine) to diets containing Lcanavanine increased deformities and mortality of THW larvae and was attributed to the structure and position of the functional groups on the added compounds (Dahlman and Rosenthal 1982). Although the THW has an effective means of degrading aberrant

PAGE 21

9 proteins (produced by the assimilation of L-canavanine) into newly synthesized proteins; the proteases involved do not efficiently degrade enough to prevent some damage from occurring in the insect (Rosenthal and Dahlman 1986; 1988). Surprisingly, L-canavanine also was shown to increase the effectiveness of Bacillus thuringiensis in vivo by altering membrane properties, mainly gut permeability, and active transport in the midgut of the THW (Felton and Dahlman 1984). However, despite the possible synergistic relationship between the relatively safe Bt product and this amino acid, no further research has been conducted to evaluate the combination for future commercial use. Other species of insects have also been tested for susceptibility to canavanine with a variety of results. Larvae of Drosophilia melanogaster Meigen (Diptera: Drosophilidae) showed no deleterious response to lower concentrations of canavanine, but showed mortality increased at concentrations over 1,000 ppm (Harrison and Holiday 1967). Lower concentrations also were ineffective against adult Pseudosarcophaga affinis (Fallen) (Diptera: Calliphoridae), with no effect on oocyte development (Hegdekar 1970). Dahlman et al. (1979) examined four species of muscoid flies and found greater than 70% mortality at the higher concentration (800 ppm) and decreased pupal weights as concentrations of canavanine increased. Despite the toxicity of canavanine to some insects, others have evolved detoxifying mechanisms to deal with high concentrations of this compound. Rosenthal et al. (1978) attributed the detoxification of canavanine in the bruchid Caryedes brasiliensis Thunberg (Coleoptera: Bruchidae) to the beetle's ability to convert canavanine to canaline, another toxic amino acid. Canaline is metabolized through reductive deamination to homoserine and ammonia, with the overall result being the detoxification

PAGE 22

10 of the two antimetabolites. This process actually increases the nitrogen intake from the foodstuff (from the increase of ammonia) (Rosenthal et al. 1976; Rosenthal et al. 1977). Another insect, the tobacco budworm {Heliothis virescens (Fab.) (Lepidoptera: Noctuidae)) was able to metabolize far more canavanine then the bruchid beetle larva ever takes in during its development, suggesting that generalists may have more than a single detoxification mechanism for compounds they may encounter (Berge et al. 1986). Metabolism of L-canavanine by the tobacco budworm was attributed to the gut enzyme canavanine hydrolase, and may have been the result of feeding on canavanine-containing plants of the Fabaceae (Melangeli et al. 1997). Essential Amino Acids In despite of the extensive toxicological research conducted on nonprotein amino acids, another group of amino acids, the essential ones, has been overlooked. One reason this avenue for research has not been pursued is that we do not want to give pests convenient access to an integral part of their diet. The fear of creating a "super" insect (that has been provided with compounds that actually aid in its development) is a rational one. Mittler (1967a; 1967b) found an increase in gustation in Myzus persicae (Sulzer) (Hemiptera: Aphididae), with amino acid levels as low as 0.2% concentration in a sucrose solution. Likewise, Sugarman and Jakinovich (1986) found increased gustatory response to both D-and L-methionine by Periplaneta americana (L) (Blattodea: Blattidae) adults. Another reason that essential amino acids have not been examined for use as a pesticide is the knowledge regarding the limited mode of action these compounds could be involved with (i.e., an active site or systemic response). Recent studies on the membrane proteins of insects show the possibility of a biophysiological system that can be exploited for insect control with certain essential amino acids.

PAGE 23

The CationAnion Modulated Amino Acid Transporter With Channel Properties (CAATCHH System Cation-Anion modulated Aminoacid Transporter with Channel properties (CAATCH1) is a recently cloned insect-membrane protein isolated from larval midgut/hindgut nutritive absorptive epithelium. This membrane protein exhibits a unique polypeptide and nucleotide sequence related to, but different from, mammalian Na + -, Clcoupled neurotransmitter transporters (Feldman et al. 2000). Using a unique PCR-based strategy, the gene encoding CAATCH1 was cloned from the digestive midgut of THW larvae. The unique biochemical, physiological, and molecular properties of CAATCH1 indicate that it is a multifunction protein that mediates thermodynamically uncoupled amino acid uptake, functions as an amino acid-modulated gated alkali cation channel, and is likely a key protein in electrolyte and organic-solute homeostasis of pest insects (Quick and Stevens 2001). In the presence of no amino acids, the cations K + and Na + are transported through the membrane via the channel (Figure 2A). When exposed to proline, the amino acid is transported through the membrane with an increase in cation flow, especially Na + (Figure 2B). However, when exposed to methionine, the amino acid transport is stopped and cation flow is altered, mainly the increased flow of K + and the decreased flow of Na + (Figure 2C). The CAATCH1 system works in alkaline conditions, at a pH optimum ~ 9.5. This alkaline condition is found in the midgut of several species (Nation 2001) and has been attributed to a variety of causes, from the detoxification of plant allelochemicals to amino acid uptake (Giordana et al., 2002; Leonardi et al. 2001).

PAGE 24

12 No Amino Acid Proline Methionine Na Na K* Figure 2. The CAATCH1 system identified from the midgut of the tobacco hornworm (modified from Quick and Stevens 2001). In the presence of no amino acids, ion flow is similar for both K + and Na + (A). With the addition of an amino acid, flows are changed. When proline is added (B), the transport occurs but the binding of the amino acid increases the ion flow, notably Na + However, when methionine is added (C) transport occurs and the binding of the amino acid greatly decreases the flow of Na + while K + is increased

PAGE 25

13 Several amino acids were found to initiate the blocking action of ion flow through the CAATCH1 protein, including threonine, leucine, and methionine, with the latter producing the greater response, based on CAATCH1 research (Feldman et al. 2000; Stevens et al. 2002; Quick and Stevens 2001). Methionine The amino acid methionine is considered essential in the diets of many organisms. Methionine is considered an indispensable amino acid in humans. Because the body does not synthesize it, uptake of methionine must occur in the diet. The recommended daily allowance of methionine for a healthy lifestyle ranges from 13 to 27 mg/kg/day for infants to full-grown adults (Young and El-Khoury 1996). This amino acid is linked to a decrease in histamine levels, increased brain function, and is found in a variety of sources; with the highest concentration in various seeds, greens, beef, eggs, chicken, and fish (Dietary Supplement Information Bureau 2000). Recently, research has centered on the genetic modification of crop plants to overproduce methionine to increase its nutritional quality (Zeh et al. 2001). Wadsworth (1995) discussed using methionine as a feed supplement, as an aid in the therapy of ketosis in livestock, and as a treatment for urinary infections in domestic pets. Onifade et al. (2001) examined the use of housefly larvae as protein foodstuffs, and found an increase in body weight gain and erythrocyte counts in rats whose diets were supplemented with fly larvae and methionine. Likewise, Koo et al. (1980) suggested dry face fly pupae could be used as a dietary supplement and foodstuff extender for poultry because of the high concentration of methionine. The environmental safety of methionine is well known, as it poses no risk to vertebrates due to a rather high oral LD 50 of 36g/kg _1 observed in rats (Mallinckrodt Baker 2001) and also

PAGE 26

in its use as a feed supplement for livestock under the trade name of Alimet (Novus, Inc., St. Louis, MO). In addition to vertebrates, methionine also is considered an essential amino acid for insects (Nation 2001). Based on research on nutritional requirements for insects, the amount of methionine needed in a diet for survival ranged from as little as 0.0007 mg/mL (for Aedes aegypti (L.) (Diptera: Culicidae) to as high as 100 mg/mL (for Heliothis zea (Broddie) (Lepidoptera: Noctuidae)) (Dadd and Krieger 1968; Eymann and Friend 1985; Friend et al. 1957; Kaldy and Harper 1979; Kasting et al. 1962; Koyama 1985; Koyama and Mitsuhashi 1975; Rock and Hodgson 1971; Singh and Brown 1957). Methionine occurs naturally as the L-isomer while the D-isomer (an optical enantiomer) is toxic to many insects and is not found in nature (Anand and Anand 1990). A few exceptions are known, (mainly Diptera and Lepidoptera) that actually are capable of metabolizing the normally unusable D-isomer (Dimond et al. 1958; Geer 1966; Rock 1971; Rock et al. 1973; Rock et al. 1975). The requirement for small amounts of this amino acid (as compared to other amino acids) may be a result of the ability for some insects to synthesize methionine from cysteine (a common sulfur containing amino acid) thus reducing the need to take in exogenous sources of methionine. Jaffe and Chrin (1979) found that A. aegypti adults were able to synthesize methionine from homocysteine with the aid of a methionine synthetase. They found this enzyme similar to those common in other metazoans, and found that the levels of methionine synthetase increased with the presence of filarial parasites. They hypothesized that this increase in methionine synthetase was a result of the parasite depleting the host of methionine. Fertility and fecundity also have been associated with methionine in some insects (mainly D. melanogaster,) with the possibility if it being a limiting factor during egg

PAGE 27

15 production (Sang and King 1961). Lack of methionine in the diet of the female may also explain the transfer of methionine in the ejaculate of the male during fertilization (Bownes and Partridge 1987). Methionine plays another role in insect biochemistry, especially in juvenile hormone biosynthesis, inhibitory allatostatins, and storage proteins known as hexamerins. Audsley et al. (1999) found that in vitro rates of juvenile hormone synthesis in females of the tomato moth {Mamestra oleracea (L.) (Lepidoptera: Noctuidae)) were dependent on the concentration of methionine present in the incubation medium. Tobe and Clarke (1985) found a direct relationship between methionine concentration and juvenile hormone biosynthesis in the cockroach, Diploptera punctata (Eschscholtz) (Blattodea: Blaberidae), further supporting the idea that methionine plays an important role in insect biochemistry. Storage proteins, or hexamerins, act as a storehouse for amino acids that can be sequestered for later use in the developmental cycle (Pan and Telfer 1996). Many Lepidoptera have been identified with hexamerins containing high concentrations of methionine and are metabolized during the last larval stage, and presumably used for egg production (Wheeler et al. 2000). Methionine as a potential pesticide has not been overlooked entirely. Tzeng (1988) tested a methionine and riboflavin mixture and found it successful in controlling various pests, including the larvae of Culex spp. (Diptera: Culicidae). The mode of action for this mixture was attributed to a photodynamic reaction and the production of oxygen rich radicals (Tzeng et al. 1990). Their research led to the use of this methionine compound as a control agent for sooty mold of strawberry (Tzeng and Devay 1989; Tzeng et al. 1990) but not as an insecticide.

PAGE 28

16 Discovery of novel means for controlling various insect pests is one tenant of IPM. The amino acid methionine, an environmentally safe organic compound, appears to be a candidate for further study. Before it can be considered for use in controlling insects pests, several issues must be addressed, including the determination of concentrations needed to provide effective control, compatibility with current application systems, safety to nontarget organisms (i.e., beneficial or biological-control agents), and to phytotoxicity. Research Objectives Our overall goal was to evaluate the effects of L-methionine, and its amino acid analogues, on the CAATCH1 system putatively in the midgut/hindgut as a means to control different insect pests. The working hypothesis is that the L-methionine only affects the CAATCH1 system and no other system, especially those involving Na+ channels or pumps (i.e., nervous tissue). The L-isomer of methionine was chosen because of the inability of most insect species to utilize the D-isomer. Ideal targets for this research are those pests that cause severe damage to agricultural systems and to human health. Specific objectives were to • Examine the effects of L-methionine as an insecticide on the larvae of M. sexta (Tobacco hornworm), L. decemlineata (Colorado potato beetle) and A. aegypti (Yellow-fever mosquito) under various conditions • Determine any adverse effects of L-methionine on plant health to ensure its safe use in a cropping system • Examine the effects of L-methionine on various nontarget insect species to ensure the environmental safety of L-methionine and thus its compatibility with natural enemies in the context of IPM.

PAGE 29

CHAPTER 3 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE TOBACCO HORNWORM, Manduca sexta, UNDER LABORATORY CONDITIONS Introduction Manduca sexta (L.) (Lepidoptera: Sphingidae), the tobacco hornworm (THW), is a widespread species considered an economic pest throughout North and South America. The caterpillar is known for its voracious appetite. In Georgia, the THW was responsible for between approximately $1 .2 to $1 .5 million in losses and costs for control annually in tobacco from 1997 to 2001 (Jones and McPherson 1997; McPherson and Jones 2002). In addition to its well-earned reputation as an agricultural pest of solanaceous crops, the THW has shown to be resistant to common pesticides (such as endrin and endosulfan), with the possibility of cross-resistance (Bills et al. 2004). The THW also is very important to scientific research outside the arena of economic entomology, with studies ranging from molecular-based research to ecological and physiological research, mainly because of its availability and ease in culturing (Dwyer 1999). One research area of interest to scientists involves the chemistry and physiology of the midgut Insect control (or the development of new insecticides) was probably not the main purpose of the research that resulted in identifying the CAATCH1 protein, yet it became the basis of our research project. Because little information is available on the insecticidal properties of methionine, several baseline experiments were necessary to determine that concentrations of this amino acid to test It also was necessary to test L-methionine and THW 17

PAGE 30

interaction in a variety of ways, including artificial diet, natural diet (excised leaves, whole plant, and choice tests. The purpose of this portion of this study was to determine whether L-methionine was detrimental to the survival and development of the THW and to determine if L-methionine could be used to control this species. Materials and Methods Eggs of THW were obtained from the insectary of North Carolina State University, and were held in 26.4L x 19.2W x 9.5H (cm) clear plastic rearing chambers with a hardware cloth (to facilitate cleaning) (Figure 3-1). Florida Reach-In Units (FRIUs) were used to control the environment for the rearing containers (Walker et al. 1993) Containers were held at 27 C, 60% relative humidity, and a 16L.8D photoperiod in FRR7s with either artificial or natural diet (excised eggplant leaves or whole plants) depending on the pending experiment. Neonates were allowed to feed for 2 days after eclosion before being transferred to treatment groups. A camel hair brush was used for transferring larvae, to minimize the risk of damage. Diets and Survivorship The artificial diet was prepared using the procedures outlined in Baumhover et al. (1977) with the inclusion of L-methionine for the treatment concentrations of 0.1%, 0.3%, 0.5%, 1.0%, 3.0%, 5.0% and 10.0% (wt/wt). The artificial diet was changed on a regular basis to prevent desiccation and fungal growth. Larvae were exposed to the artificial diet in the clear plastic rearing chambers with a hardware cloth, and kept in the FRKJs programmed with the aforementioned environmental constants. Natural diets consisted of excised eggplant leaves (Solarium melongena L.,"Classic" variety) of potted plants grown and maintained at the University of Florida,

PAGE 31

Figure 3-1. Rearing chamber for tobacco hornworm and Colorado potato beetle larvae used in the artificial and excised leaf diet tests. Hardware cloth stage supporting the leaf allowed for easy clean up and minimized disease problems by preventing larvae from coming in contact with fecal material (paper liner not shown).

PAGE 32

20 Department of Entomology and Nematology green and shade houses. Excised leaves were dipped in solutions of deionized H2O containing different concentrations of methionine; depending on the experiment and exposed to larvae in the same rearing chambers as the artificial diet trials under the same conditions. Survivorship data were pooled from several different trials for data analysis. In total, 64 potted eggplants were used for the whole-plant portion of the study. Plants were held in FRIUs under the same conditions as the artificial and excised leaf trials, in 38H x 15D (cm) plexiglas cylinders (Figure 3-2). Four THW neonates were placed on each plant for a total of 64 larvae (16 replicates) per treatment (nT 0 tai = 256 larvae). The treatment of L-methionine was applied to the test plants (using a hand-held sprayer calibrated to deliver approximately 10 mL of solution to each plant) before the addition of larvae. Feeding and Development To test L-methionine on the developmental rates of THW, larvae were exposed to excised eggplant leaves dipped in solutions containing the same concentrations of Lmethionine used in the artificial diet trials. Additional treatments of proline (1 .0%) and Bt-kurstaki (Dipel 86% WP at 3.5 grams/liter; Bonide, Oriskany, NY) were included as positive and negative controls, respectively. Leaves were scanned photometrically using the CI 203 Area Meter with conveyor attachment (CID, Inc.; Camas, WA) to measure leaf consumption before and after exposure to larvae. The difference in leaf areas resulting from the missing leaf tissue was assumed to be the amount eaten by the developing larvae. Larval head capsule widths were measured at the time of death or the

PAGE 33

Figure 3-2. Setup for whole plant studies involving tobacco hornworm. Top and portions of the sides were replaced with fine mesh to allow for airflow and to reduce condensation.

PAGE 34

22 end of the trial (using an Olympus Tokyo Model 213598 stereomicroscope with a optical micrometer) to monitor larval development Trials to determine the total amount of L-methionine applied to excised leaves also were included to quantify how much of the amino acid was physically present on leaves at the different concentration levels. Leaves were weighed before dipping into the control (0%) and L-methionine solutions (0.1%10%), allowed to air dry for 30 min and weighed again. The difference was assumed to be the actual amount of L-methionine residue on the leaf. This value then was used to determine the total amount of L-methionine on the leaf surface of the excised leaves and the amount of L-methionine consumed per gram of leaf material, based on calculations of the physical amount of the compound for each % concentration. Preference Tests It was unknown if the additional methionine acted to attract or repel larvae. Neonate larvae were used in the choice tests to determine if there was a preference between the control (deionized H 2 0) and the Treatments (1 .0% L-methionine). Leaves were obtained from potted plants maintained in the outdoor shade house. The tests consisted of 4 leaf disks (30 mm diameter) dipped into the control solution and placed into the chamber alternately with four leaf disks (30 mm diameter) dipped into the treatment solution and replicated with a total of 1 0 chambers. Each chamber consisted of a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The filter disk was moistened routinely with deionized H 2 0 to prevent the leaf disks from desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental constants described previously. The leaf disks also were scanned photometrically and

PAGE 35

Figure 3-3. Chambers used for tobacco hornworm and Colorado potato beetle preference tests. Two treatments (control and 1.0% Lmethionine) were used to determine if any larvae exhibited any preference or avoidance to L-methionine. Treatments were alternated in the chamber and neonates were released in the center of the dish and allowed to search for food. The filter paper in the bottom of the dish was moistened to prevent desiccation of the leaf disks and the test specimens.

PAGE 36

24 larval head capsule measurements made using the same procedures described in the Feeding and Development section. Data Analysis Sample sizes of all experiments were chosen according to the guidelines recommended by Robertson and Preisler (1991) for optimal sample size and data analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were performed using PROBIT Version 1 .5 (Ecological Monitoring Research Division, USEPA) after Abbott's correction for control mortality (Abbott 1925). Survival data were normalized to the previous value when control mortality was greater than the treatment mortality, to produce a smoother trend line. Statistical analysis was performed on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the data included One-way ANOVA and separation of significant means using Tukey's Multiple Comparison and Pearson Correlation was performed on the choice trial data to examine possible relationships between development and consumption of treated leaf material (Zar 1999). Regression analysis using lest squares were performed on the leaf weights before and after the L-methionine treatment for the equation used to convert % concentration to mg/g plant material (Figure 3-4). Results Diets and Survivorship The artificial diet resulted in 100% mortality of THW larvae for the 3.0% L-methionine to 10.0% L-methionine treatment after only one day of exposure (Figure 3-5). Approximately 80% mortality was observed in the 1.0% L-methionine treatment after 4 days, and 50% mortality for both the 0.3% L-methionine and 0.5% L-methionine

PAGE 37

L-methionine Concentration (%) Figure 3-4. Amount of L-methionine present on leaf surface after treatment. Excised leaves were weighed, dipped into various concentrations of L-methionine, allowed to dry, and then re-weighed. Difference assumed to be the amount of L-methionine remaining on leaf surface (T=22.43, and PO.001).

PAGE 38

26 Days of Exposure Figure 3-5. Mortality of tobacco horn worm larvae exposed to various concentrations of L-methionine (njotai^SO) in artificial diet. Data were adjusted using Abbott's formula to account for control mortality. Note the overlap in trend lines for the 3.0% L-methionine10.0% L-methionine concentrations after Day 1 and the 0.3% L-methionine and 0.5% L-methionine treatments from Day 1 to Day 10.

PAGE 39

27 treatment after 10 days of exposure. The 0.1% L-methionine concentration had lowest larval mortality with approximately 30% observed for the triaL The excised leaf trials exhibited higher mortality rates associated with the treatments than did the artificial diet trials. Again, complete mortality was observed with the 3.0% L-methionine thru 10.0% L-methionine concentrations after 1 day of exposure (Figure 3-6). Greater than 90% in the 0.5% L-methionine and 1 .0% L-methionine treatments, followed by 80% mortality in the 0.3% L-methionine treatment, and greater than 60% mortality occurred in the 0.1% L-methionine treatment after 8 days. Whole plant trials produced results similar to the excised leaf trials with greater than 90% larval mortality observed with the 1 .0% L-methionine treated plants after 14 days (Figure 3-7). Mortalities exceeding 20% and 60% were observed for the 0.1% L-methionine and 0.5% L-methionine treatments, respectively, during the same time interval. PROBIT analysis of a sample size of n To tai= 1,540 for 7 treatments (0.1% L-methionine, 0.3% L-methionine, 0.5% L-methionine, 1.0% L-methionine, 3.0% L-methionine, 5.0% L-methionine and 10.0% L-methionine) revealed an overall LC 50 of 0.66% (32.3 mg/g leaf material) concentration for the artificial diet and 0.074% (4.39 mg/g leaf material) concentration for the natural diet after 9 days of exposure (Figure 3-8). The LC 50 for the THW exposed to artificial diet was approximately half the value of that for the natural diet for the 24 to 72 hour exposure period. The LC 50 for the artificial diet of 1.08% (52.3 mg/g leaf material) for 24 h dropped to 1.0% (48.5 mg/g leaf material) after 48 h and to 0.57% (28.0 mg/g leaf material) after 72 h. As for the natural diet, the LC 50 of 0.53% (26.1 mg/g leaf material) was found to be lower than the artificial

PAGE 40

28 Days of Exposure Figure 3-6. Mortality of tobacco hornworm larvae exposed to various concentrations of L-methionine (niotaP 1,540) on excised eggplant leaves. Data were adjusted using Abbott's formula for control mortality. Note the overlap in trend lines for the 3.0% L-methionine10.0% L-methionine concentrations after Day 1

PAGE 41

29 Figure 3-7. Survivorship of tobacco hornworm larvae exposed to various concentrations of L-methionine (nT 0 tai = 256) on whole plants. Lmethionine was applied using a hand-held sprayer in the amount of 10 mL/treatment. Data were adjusted using Abbott's formula for control mortality.

PAGE 42

30 Figure 3-8. Concentrations (%) of L-methionine required for the mortality of 50% of test population of tobacco hornworm after 9 days exposure (nTotai = 1,540; n=180 for 3.0% L-methionine 10.0% L-methionine, n=200 for remainder). Number range following value is the 95% confidence limits. Determination of LC50 was performed using PROBIT Version 1.5 (Ecological Monitoring Research Division,

PAGE 43

31 diet at 24 h and dropped to 0.4% (19.9 mg/g leaf material) at 48 h and 0.25% (12.8 mg/g leaf material) after 72 h exposure. Overall, the LC50 at the end of the experiment for the natural diet was well below the value for the artificial diet, with close to a 90% reduction. Feeding and Development Mortality of THW for the developmental tests ranged from approximately 30% for the 0.1% L-methionine treatment and over 40% for the proline treatment (Figure 3-9). Complete mortality for the 0.3% L-methionine occurred after 7 days while the 0.5% L-methionine treatment took only 5 days. The Btk treatment mortality was similar to the 0.7% L-methionine and 1 .0-%L-methionine treatment, resulting in 100% mortality after 1 day of exposure to the amino acid. Both the mean head capsule width and amount of leaf material consumed showed significant differences between treatments, with the control, 0.1% L-methionine and proline treatments being different that the remaining treatments (Figures 3-10 and 3-1 1). Preference Tests The amount of control and 1.0% L-methionine leaf tissue consumed during the preference tests were found not to be statistically different (Figure 3-12). In addition to the amount of leaf material consumed between treatments not being different, the mean head capsule width {i.e., development) showed a correlation with the amount of control diet consumed (Pearson Correlation Coefficient 0.885, PO.001) while no correlation to the Treatment diet consumed (Pearson Correlation Coefficient 0.630, P=0.051) (Figure 3-11).

PAGE 44

32 Days of Exposure Figure 3-9. Mortality of tobacco hornworm larvae exposed to various concentrations of L-methionine (nTotai = 160) on excised eggplant leaves for feeding and development trials. Proline (1.0%) and Btk were included for comparison as positive and negative controls. Data were adjusted using Abbott's formula for control mortality. Note the overlap in the 0.7% L-methionine, 1 .0% L-methionine and Btk treatments at Day 1

PAGE 45

33 (Error Bars @ 95%; F (0 0 5)7,i 52=2.37, F=18.2; P<0.001) Control 0.1% 0.3% 0.5% 0.7% 1.0% Proline Btk Figure 3-10. Mean head capsule widths of tobacco hornworm larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (n To ta] = 320). Proline (1.0%) and Btk were included for comparison as positive and negative controls. Error bars denote 2 SE. Bars within treatments having the same letter are not statistically different (Tukey's MST, PO.001).

PAGE 46

34 (Error Bars @ 95% ; F(0.05)7,i52^2 37, F=18J; P<0.001) 300 -i Figure 3-11. Total leaf area consumed by tobacco homworm larvae exposed to excised eggplant leaves treated with various concentrations of Lmethionine (n To ta] = 320). Proline (1.0%) and Btk were included for comparison as positive and negative controls. Error bars denote 2 SE. Bars within treatments having the same letter are not statistically different (Tukey's MSTP, P<0.001).

PAGE 47

35 (&ror Bars @ 95% ; F(0.os)u 8=5.98, F=1.64; P=0.217) Control 1.0% Treatment Figure 3-12. Mean leaf consumption by tobacco hornworm in the preference tests. Error bars denote 95% SE, and treatments were found not to be statistically different. However, there was correlation between the control diet consumed and mean head capsule width (Pearson Correlation Coefficient 0.885, P=0.001) while no correlation was found between the Treatment diet consumed and mean head capsule width (Pearson Correlation Coefficient 0.630, P=0.05).

PAGE 48

36 Discussion The initial studies involving the high concentrations of L-methionine (i.e., 3.0-10.0%, which are outside the range normally encountered in nature) showed that a concentration of 1 .0% L-methionine was sufficient enough to provide good control of THW larvae reared on both artificial and natural diets. The 0.1%L-methionine concentration remained similar to that of the control for developmental and feeding trials (Figure 3-9), indicating a level of methionine that can be tolerated to some extent, as seen in the low mortality of this treatment. This is in stark contrast to the mortality seen in the excised leaf trials in which the same concentration had over 60% mortality (Figure 3-6). One possible explanation could be the amount of L-methionine present on the leaf disk being low enough and ingested at a slower rate than that of the whole leaf, which was left in the chamber with the larvae until the leaf was either completely consumed or too wilted for the larvae to ingest. The preference tests did show some preference towards control leaf disks over the 1 .0%L-methionine treated disks as seen in the correlation analysis of the diet consumed and the mean head capsule width of the larvae. Despite the lack of a statistical difference between the amount consumed, the larvae could have fed on the treated disks and then switched to the control disks based on a physiological cue. It is unclear if THW larvae possess specialized sensory structures to detect amino acids like those found in other Lepidoptera (Beck and Henec 1958; Dethier and Kuch 1971; Schoonhoven 1972), but the possible switch from the methionine rich treatment to the control leaf disks does indicate some sort of mechanism for detection. Del Campo and Renwick (2000) found THW larvae were induced to feeding on plants outside of their normal diet when the plants

PAGE 49

were treated with an extract from potato foliage suggesting induced host preference, attraction, and dependence on this compound in the extent of sustained feeding and development. A combination of sensory structures may be involved for the detection of specific amino acids and host plant compounds, which may explain the selection of methionine depleted host plants to avoid problems with the CAATCH1 system present in the midgut of the THW. The difference in the LC50 for the artificial and natural diets was striking considering the concentrations were the same. One possible explanation is the L-methionine on the natural diet was more readily available than that found in the artificial diet. With the artificial diet, the L-methionine is presumably spread throughout the diet and would therefore take longer for the THW to ingest enough to adversely affect the CAATCH1 system. In contrast, the L-methionine was found on the surface of the leaf in higher concentrations than that of the artificial diet and was also freely available once ingested. Thus, larvae were exposed to a higher concentration of L-methionine with less work to digest, resulting in lower survivorship in the same period of time. The 1.0%L-methionine concentration had the same mortality, feeding and developmental rates for THW, as did the Btk treatments (Figure 3-9). The 0.3% L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments were virtually the same for mortality (Figure 3-9), developmental rate (Figure 3-10) and total leaf material consumed (Figure 3-1 1) and statistically the same as the 1 .0% L-methionine concentration and the Btk treatment. The similar mortality rate observed for the higher concentrations of L-methionine and Btk is encouraging considering the resistance to Bt seen in many insect species because of reduced receptor activity and binding (Bills et al.

PAGE 50

2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and many are the result of a combination of different pesticide classes. The CAATCH1 system is one that could be used in cases where the only alternative is by adding more pesticides or at higher rates to break resistance. Further research is needed to determine compatibility of the different Bt insecticides and L-methionine with each other for cases in which Bt resistance is observed in natural populations. Given the safety of L-methionine and the shorter time required for 100% mortality (when compared to Btk results of this study), this compound could represent a viable alternative for pesticides currently used in the management of the THW.

PAGE 51

CHAPTER 4 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE COLORADO POTATO BEETLE, Leptinotarsa decemlineata, UNDER LABORATORY CONDITIONS Introduction Leptinotarsa decemlineta (Say) (Coleoptera: Chrysomelidae), the Colorado potato beetle (CPB), is considered an economic pest throughout North America. The larvae and adults of the CPB feed on a wide variety of solanaceous crop plants and are responsible for $150 million in losses and control related costs (Durham 2000). To further complicate matters, the CPB is resistant to numerous pesticides, including various pyrethroids and carbamates (Bills et al. 2004). Historically, CPB management relied heavily on chemical control methods that led to the development of resistance to different pesticides in several areas of the eastern United States (Forgash 1985; Gauthier et al. 1 98 1 ). Control of CPB without the use of chemicals is further complicated given the species ability to develop resistance and the limitations on the use of resistant varieties of potato (Ragsdale and Radcliffe 1999). The use of plant varieties that are resistant to CPB and other pests also run the risk of developing tolerance to chemical pesticides in other pest species (Sorenson et al. 1989). Despite the success of Bacillus thuringiensistenebrionis (Btt) and the biocontrol agents Podisus maculiventris Say (Hemiptera: Pentatomidae) and Edovum puttleri Grissel (Hymenoptera: Eulophidae), more biorational alternatives are necessary for controlling CPB to prevent yet another devastating threat to the potato industry because of this insect's ability develop resistance and overcome control methods (Boucher 1999; Ferro 1985; Tipping et al. 1999). This makes the CPB 39

PAGE 52

40 an excellent candidate for the evaluation of L-methionine as a possible means of controlling this devastating pest. Because little information is available on the insecticidal properties of L-methionine, several baseline experiments were necessary to determine what concentrations of this amino acid to test. Therefore, it was necessary to test L-methionine and CPB interaction in a variety of ways including survivorship of both larvae and adults, development of larvae when exposed to different concentrations of the amino acid, and preference tests. The purpose of this portion of this study was to conduct bioassays to determine if exposure to L-methionine was detrimental to the survival and development of the CPB and to determine if L-methionine could be used to control this species. Materials and Methods Eggs of CPB were obtained under UDSA permit from the insectary of the New Jersey Department of Agriculture and held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth (to facilitate cleaning) and held at 27 C, 60% relative humidity and 16L/8D photoperiod in FRIUs (Figure 3-1). Excised eggplant leafs were placed in the chambers with the neonates and they were allowed to feed for 2 days after eclosion before being transferred to experiments. A camel hair brush was used for transferring the neonates to minimize the risk of damaging the larvae. Survivorship Larvae and adults of the CPB were tested in preliminary experiments with the highest concentration (1 .0% L-methionine (wt/wt)) observed in tests done on the THW in the previous chapter. The diet for the larvae and adults consisted of excised eggplant leaves {Solarium melongena L.,"Classic" variety (Family: Solanaceae)) from plants grown and maintained at the University of Florida, Department of Entomology and

PAGE 53

41 Nematology green and shade houses. Excised leaves were dipped in solutions of deionized H2O containing different concentrations of methionine and held in the clear plastic boxes and held at the aforementioned environmental conditions (Figure 3-1). Additional treatments of proline (1.0%) and Bt-tenebrionis (Novodor FC @12.4 mL/L; Valent Biosciences, Libertyville, IL) were included as positive and negative controls, respectively. Survivorship data were pooled from several different trials for data analysis. Feeding and Development To test L-methionine on the developmental rates of CPB, larvae were exposed to excised eggplant leaves dipped in different concentrations of L-methionine under the same conditions as the survivorship trials. Additional treatments of proline (1.0%) and Btt were included as positive and negative controls, respectively. Leaves were scanned photometrically using the CI 203 Area Meter with conveyor attachment (CID, Inc., Camas, WA) before exposure to the larvae and measuring after leaf consumption. The difference in leaf areas resulting from the missing leaf tissue was assumed to be the amount eaten by the developing larvae. Larval head capsule widths were measured at the time of death or the end of the trial (using an Olympus Tokyo Model 213598 stereomicroscope with an ocular micrometer) as an evaluation of larval development. Preference Tests It was unknown if the additional methionine acted to attract or repel larvae. Neonate larvae were used in the choice tests to determine if there was a preference between the Control (deionized H 2 0) and the treatments (1.0% L-methionine). Leaves were obtained from potted plants maintained in the outdoor shade house. The tests

PAGE 54

42 consisted of 4 leaf disks (30 mm diameter) dipped into the Control solution and placed into the chamber alternately with four leaf disks (30 mm diameter) dipped into the treatment solution and replicated with a total of 1 0 chambers. Each chamber consisted of a large petri dish (25.0 cm diameter x 9.0 cm depth) lined with a Seitz filter disk. The filter disk was moistened routinely with deionized H 2 0 to prevent the leaf disks from desiccation (Figure 3-3). Chambers were held in FRIUs at the same environmental constants described previously. The leaf disks also were scanned photometrically and larval head capsule measurements made using the same procedures described in the Feeding and Development section. Data Analysis Sample sizes of all experiments were chosen according to the guidelines recommended by Robertson and Preisler (1991) for optimal sample size and data analysis. Probit analysis and determination of mean Lethal Concentration (LC 50 ) were performed using PROBIT Version 1.5 (Ecological Monitoring Research Division, USEPA) after Abbott's correction for control mortality (Abbott 1925). Survival data were normalized to the previous value when control mortality was greater than the treatment mortality, to produce a smoother trend line. Statistical analysis was performed on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the data included One-way ANOVA and separation of significant means using Tukey's Multiple Comparison and Pearson Correlation was performed on the choice trial data to examine possible relationships between development and consumption of treated leaf material (Zar 1999).

PAGE 55

43 Results Survivorship Mortality of CPB larvae on treated excised eggplant leaves ranged from approximately 20% for the 0.1% L-methionine treatment after 4 days, 80% mortality for the 0.3% L-methionine treatment after 8 days of exposure and 100% for the remaining concentrations with the highest dose of 1 .0% L-methionine exhibiting complete control of CPB in 3 days post treatment (Figure 4-1). Some mortality (50%) was observed for the proline (1 .0%) treatment while the Bit larval treatment mortality was similar to the 1.0% L-methionine treatment, resulting in 100% mortality after 5 days. PROBIT analysis of a sample size of n to tai = l,320 for 6 treatments (Control), 0.1% L-methionine, 0.3% L-methionine, 0.5% L-methionine, 0.7% L-methionine and 1.0% L-methionine) revealed an overall LC50 of 0.21 8% concentration for the CPB after 8 days of exposure (Figure 4-2). The LC50 of 2.9% for 24 hours dropped to 1.1% after 48 hours and to 0.22% after 72 hours. Feeding and Development Mean head capsule widths between treatments were found to be statistically different (Figure 4-3). Four distinct groups were observed, with the Control, 0.1% L-methionine and proline treatments forming the first group. The second group of proline and 0.5% L-methionine were statistically the same and likewise the third group of the 0.3% L-methionine, 0.5% L-methionine, and 0.7% L-methionine treatments. The final group of Btt and 1.0% L-methionine treatments was statistically different from all other treatments.

PAGE 56

44 Days of Exposure Figure 4-1. Mortality of Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of L-methionine (n T otai =: 560). Proline (1.0%) and Btt were included for comparison as positive and negative controls. Data were adjusted using Abbott's formula for control mortality.

PAGE 57

45 Figure 4-2. Concentrations (%) of L-methionine concentrations required for the mortality of 50% of the test population of Colorado potato beetle after 8 days exposure (nT O tai = 220). Number range following value is the 95% confidence limits. Determination of LC50 was performed using PROBIT Version 1.5 (Ecological Monitoring Research Division, USEPA), including Abbott's

PAGE 58

46 (Error Bars @ 95%; F(o.o5)7,3i2=1.14;F=576.71; P<0.001) Figure 4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of Lmethionine (n To tai = 320). Proline (1.0%) and Bt were included for comparison as positive and negative controls. Error bars denote 2 SE. Bars within treatments having the same letter are not statistically different (Tukey's MST, PO.001).

PAGE 59

47 Feeding rates of CPB also were found to be statistically different among treatments (Figure 4-4). Three distinct groups were observed with the first group containing the Control and 0.1% L-methionine treatments while the second group of the 0.1% Lmethionine and 0.3% L-methionine, treatments were found to be statistically the same. The 0.5% L-methionine, 0.7% L-methionine, 1.0% L-methionine and Btt treatments were statistically different from the other groups. Overlap occurred with the proline treatment across all groups indicating no statistical difference with the rest of the treatments. Preference Tests The amount of Control and 1 .0% L-methionine leaf tissue consumed during the preference tests was found not to be statistically different (Figure 4-5). In addition, the mean head capsule width (/'.e., development) showed no relationship with either treatment based upon the low correlation coefficients. The 1.0% L-methionine concentration produced the same larval mortality, feeding and developmental rates for CPB, as did the Btt treatments (Figures 4-1, 4-3, and 4-4). The 0.3% L-methionine, 0.5% L-methionine and 0.7% L-methionine treatments took 4 days longer for complete control (Figure 4-1), but were statistically different for the developmental rates for the same treatments (Figure 4-3). As was the case with the THW survivorship, the 0.1% L-methionine concentration was not different from that of the Control. This may indicate a threshold of methionine that can be tolerated by the THW, and CPB to some extent, evidenced by the low mortality observed for this treatment. The Preference tests did not indicate any preference of leaf disks with or without L-methionine. The high mortality (90%) of the CPB larvae could be explained by a

PAGE 60

48 Figure 4-4. Total leaf area consumed by Colorado potato beetle larvae exposed to excised eggplant leaves treated with various concentrations of Lmethionine (n To tai=320). Proline (1.0%) and Btt were included for comparison as positive and negative controls. Error bars denote 2 SE. Bars within treatments having the same letter are not statistically different (Tukey'sMST,P<0.001).

PAGE 61

49 (Error Bars @ 95%; F (0 .o 5 )U8=5.98, F=1.64; P =0.217) Control 1 .0% L-methionine Figure 4-5. Mean leaf consumption by Colorado potato beetle in the preference tests. Error bars denote 95% SE, and treatments were found not to be statistically different. No correlation between either Control or Treatment Diet consumed and mean head capsule width was found (Pearson Correlation Coefficient 0.466, P=0.175 and 0.665, P=0.036, respectively).

PAGE 62

50 combination of the early consumption of the treated disks and mortality occurring after 48 hours, when a lower concentration is required for mortality. The larvae could have fed on the treated disks and then switched to the Control based on a physiological cue. Mitchell (1974) and Mitchell and Schoonhoven (1974) examined the taste receptors of CPB and found physiological and behavioral responses to some amino acids, mainly gamma aminobutyric acid (GABA) and alanine. They discussed the possibility that host selection in solanaceous plants may have been the result of these chemosensory structures and the concentration of amino acids in the leaves. It should be noted that both studies excluded methionine and no electrophysiological data were collected on the response of CPB to this amino acid. This is not surprising considering the fact that the diet of the CPB is low in methionine and therefore would not be a candidate for the inclusion in feeding stimulatory studies (Cibula et al. 1967). It is unknown if these sensory structures can detect methionine and possibly act as a means to avoid plant material high in this amino acid. This appears to be contradicted by the data in Figure 4-5, in which there was no difference between the treatments. The larvae feeding on the Control treatment, consuming the majority and then moving to the 1 .0% L-methionine treatment, could explain the lack of difference. There are some differences between some of the Feeding and Development treatments should be noted. The mean head capsule of the larvae in the 0.5% L-methionine treatment was higher than the 0.3% L-methionine treatment while the amount of leaf material consumed for the same treatment were the same indicating another factor involved with the greater head capsule width. The differences could be the result of the larger size of females and possibly could have included more females.

PAGE 63

51 The higher concentrations of L-methionine that produced mortality similar to the Btt is encouraging considering the occurrence of resistance to this compound seen in many pest insect species because of reduced receptor activity and binding (Bills et al. 2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and many are the result of exposure to a combination of different pesticide classes. The Methionine-CAATCHl system could be exploited in cases where the only alternative is applying different pesticides or using higher rates to break resistance. Further research is needed to determine compatibility with Bt and L-methionine for cases in which resistance is observed in natural populations. Given the safety of L-methionine and the shorter time required for 100% mortality (when compared to Btt), this compound could represent a new biorational tool for the management of the CPB.

PAGE 64

CHAPTER 5 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF THE YELLOW FEVER MOSQUITO, Aedes aegypti, UNDER LABORATORY CONDITIONS Introduction Integrated Pest Management practices are not restricted to agricultural pests. Medically important insect pests are responsible for epidemics that have changed the course of human existence, from bubonic plague spread by the Oriental rat flea (Xenopsylla cheops Rothschild (Siphonaptera: Pulicidae)), to malaria carried by anopheline mosquitoes. One medically important species that has had a significant impact on human existence is the yellow fever mosquito (YFM), Aedes aegypti (L.) (Diptera: Culicide). This cosmopolitan species is found worldwide and is the primary vector for human dengue and yellow fever despite concerted efforts at eradication in the United States (Womack, 1993). In the United States alone, upwards of 150,000 lives were lost to yellow fever in the period starting in the late 1 8 th century and into the early 20 th century (Patterson, 1992). However, because of the development of a vaccine, yellow fever has been replaced by Dengue which is now second only to malaria as a worldwide threat (Gubler, 1998). Because Dengue fever is also vectored by the YFM, it poses a risk by affecting tens of millions of people worldwide (Gubler and Clark, 1 995). The inclusion of the YFM in this study was an effort borne of curiosity because of the lack of knowledge of the CAATCH1 system in other insects and the availability of specimens for study. Mosquito larvae are particulate feeders and have dietary 52

PAGE 65

53 requirements of methionine in the amounts of 0.0007mg/ml for the YFM. This amino acid also is considered essential for other species of mosquito in untraceable (in those studies) amounts (Chen, 1958; Singh and Brown, 1957). Given the high alkalinity found in the midgut of the YFM as well as other mosquito species, this physiological condition indicates the possibility of the presence of the CAATCH1 system in larval mosquitoes (Dadd, 1975). The purpose of this portion of the study was to examine the survival and development of YFM larvae exposed to water treated with excess L-methionine (adults were not tested given the feeding nature). In addition to L-methionine, other amino acids were tested in an effort to see if their response (i.e., survivorship) was similar CAATCH1 responses to methionine found by Feldman et al. (2000). Materials and Methods Bioassay The bioassay experiments consisted of six treatments (control, 0.1%, 0.3%, 0.5%, 0.7% and 1 .0%) each with four replicates. Both L-methionine and D-methionine were tested along with proline, Beta-alanine and L-leucine to examine the other amino acids that were found to be reactive to the CAATCH-1 system (Feldman et al., 2000). Bt-isrealiensis (Aquabac @ a rate of 2.3 mL/m 2 ; Biocontrol Network, Brentwood, TN) and proline also were included in some trials of L-methionine to allow for comparison of both positive and negative effects. Amino acids were weighed using a Denver Instruments Co. XD2-2KD digital scale and added to glass quart jars containing 500ml of deionized H 2 0. Concentrations were based on the proportion of lg/ 100ml for a 1% solution and for corresponding concentrations. Solutions were allowed to sit at room

PAGE 66

54 temperature (23C) to permit the amino acid to fully dissolve before the addition of the larvae. An additional trial of L-methionine buffered with Tris to a pH of 7.0 using a Fisher Scientific Accumet pH 900 was conducted to determine if mortality was attributed to a change in pH or exposure to the L-methionine. Larvae of YFM (third instar) were obtained from the mosquito colony maintained at the Department of Entomology and Nematology, University of Florida. Larvae were transferred to the treatment jars using a camel hair, with 10 larvae per replicate for a total of 40 larvae/treatment and nT O tai = 240 for each amino acid bioassay experiment (Figure 5-1). Approximately 0.5g of finely ground fish food was added to serve as a larval food source and nylon window screen was used to cover the tops of the jar to prevent the escape of any emerged adults. Jars were held at 23 C on a dedicated laboratory bench top for approximately one week. The numbers of larvae surviving were recorded each day. Growth and Development This experiment used the same Materials and Methods as the bioassay portion with the exception of neonate larvae instead of 3 rd instars. Eggs were placed in a tray of water and held at 23 C for 2 days after eclosion. Neonates were removed using a camel hair paintbrush and placed into each jar, with 10 larvae per replicate for a total of 40 larvae/treatment (nTot a r = 240). Larval exuviae or dead larvae were removed and used to examine growth rates by measuring the head capsules. Larvae head capsule widths were measured (using an Olympus Tokyo Model 213598 stereomicroscope with an ocular micrometer) as an evaluation of larval development.

PAGE 67

55 Figure 5-1. Bioassay setup for yellow fever mosquito larvae exposed to various concentrations of amino acids and Bti. Jars contained 500mL of solution and were covered with screen to prevent the escape of emerging adults.

PAGE 68

56 Data Analysis Sample sizes of all experiments were selected according to the guidelines of Robertson and Preisler (1991) for optimal sample size and data analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were performed using PROBIT Version 1.5 (Ecological Monitoring Research Division, USEPA) after Abbott's correction for control mortality (Abbott 1925). Probit analysis was performed on different concentrations (0.1%, 0.3%, 0.5%, 0.7% and 1.0%) of L-methionine, Trisbuffered L-methionine, D-methionine, Beta-alanine, proline and L-leucine for 24, 48, 72 and 168 hours (the end of the trials). Survival data were normalized to the previous value when control mortality was greater than the treatment mortality, to produce a smoother trend line. Statistical analyses were performed on the data using Minitab Version 12. Analysis (Minitab, Inc; State College, PA) of the data included One-way ANOVA and separation of means using Tukey's Multiple Comparison test (Zar 1999). Results Bioassav Mortality of YFM larvae in both the unbuffered L-and D-methionine trials was similar with low or no mortality, at the 0.1% concentrations (Figures 5-2 and 5-3). The 0.3% concentration had lower mortality with D-methionine (45%) than L-methionine (75%) and greater than 80% mortality for the 0.5% concentration for both isomers. Higher concentrations of both D-and L-methionine forms produced 100% mortality of the larvae within 2 days after treatment. Greater than 40% mortality was observed for the buffered 0.1% L-methionine concentration with complete mortality for the remaining treatments within 5 days of

PAGE 69

57 Exposure (Days) Figure 5-2. Mortality of yellow fever mosquito larvae exposed to various concentrations of L-methionine (n To tai=240). Data were adjusted using Abbott's formula for control mortality.

PAGE 70

58 Days of Exposure Figure 5-3. Mortality of yellow fever mosquito larvae exposed to various concentrations of D-methionine (nx o tai = 240). Data were adjusted using Abbott's formula for control mortality.

PAGE 71

59 exposure (Figures 5-4). The 1 .0%L-methionine treatment caused 100% mortality after 2 days while the Bti treatment took 3 days to reach the same level of control. The proline treatment caused less than 10% mortality. In contrast to methionine, survival of YFM larvae exposed to proline and L-leucine was higher, with only approximately 20% mortality for the higher 0.7% proline and 1.0% proline concentrations (Figure 5-5) and less than 3% mortality with the highest L-leucine concentration (Figure 5-6). Beta-alanine mortality was similar to the L-methionine treatments with between 75% and 83% mortality for the 0.5% Beta-alanine thru 1.0% Beta-alanine concentrations, respectively, greater than 40% mortality with the 0.3% Beta-alanine, and less than 5% mortality for the 0.1% Beta-alanine concentrations (Figure 5-7). Growth and Development Developmental rates of YFM larvae resulted in three distinct groups, with the control and proline treatments, producing virtually identical results; both were statistically different from the 0.1%L-methionine treatment and the remaining L-methionine treatments (Figure 5-8). The Bti treatment was statistically the same as the 03% L-methionine to 1 .0% L-methionine treatments, with very little growth taking place. Probit analysis for unbuffered L-methionine (nT O tai = 40 for 5 treatments; 0.1%, 0.3%, 0.5%, 0.7% and 1.0%) revealed an overall LC 50 of 0.19% concentration for the YFM after 7 days of exposure (Figure 5-9). The LC 50 of 1 .2% for 24 hours dropped to 0.41% after 48 hours and to 024% after 72 hours. When the L-methionine treatments (same concentrations) were buffered to a pH Of 7.0, the values dropped to 0.64% for 24

PAGE 72

60 Days of Exposure Figure 5-4. Mortality of yellow fever mosquito larvae exposed to various concentrations of Tris-buffered L-methionine (nxotai = 240). Data were adjusted using Abbott's formula for control mortality. Note the longer exposure because of the bioassay involving neonates instead of 3 rd instars. Note the overlap in some of the trend lines on Day 1 with the 0.3% L-methionine and 0.5% Lmethionine treatments.

PAGE 73

61 100 n 80 1 60 Days of Exposure Figure 5-5. Mortality of yellow fever mosquito larvae exposed to various concentrations of Proline (nT O tai = 240). Data were adjusted using Abbott's formula for control mortality. Note the overlap of trend lines for all treatments except the 0.7% L-methionine and 1 .0% L-methionine treatments. ^hControl *-0.10% A 0.30% -—0.50% ^^0.70% •—1.00%

PAGE 74

62 100 80 i 60 o 40 20 Control 0.10% 0.30% 0.50% 0.70% 1.00% 0 1 2 3 4 5 6 7 Days of Exposure Figure 5-6. Mortality of yellow fever mosquito larvae exposed to various concentrations of L-leucine (nTotai = 240). Data were adjusted using Abbott's formula for control mortality. Note the overlap in trend lines for all treatments.

PAGE 75

63 Figure 5-7. Mortality of yellow fever mosquito larvae exposed to various concentrations of Beta-alanine (nTotai = 240). Data were adjusted using Abbott's formula for control mortality.

PAGE 76

64 (Error Bars @ 95% ; F
PAGE 77

65 1.4 n 24h 48h 72h Overall (240h) Figure 5-9. Concentrations (%) resulting in 50% mortality (LC 50 ) of yellow fever mosquito larvae exposed to various amino acids after 10 days (nToti = 240 for each amino acid). Number range following value is the 95% confidence limits. Proline and L-leucine were also tested but did not exhibit sufficient mortality to allow for Probit Analysis.

PAGE 78

66 hours, and to 0.1 1% for 48-168 hours and remained constant since the trial lasted longer because of the use of neonates instead of 3 rd instars. The D-methionine treatments were similar with 0.44% for 24 and 48 hours, 0.33% for 72 hours and 0.32% after 168 hours. While not as striking as the others, Beta-alanine had a LC50 concentration of 1 .1% after 24 hours, dropped to 0.5% after 48 hours and leveled off around at 0.35% after 72 and 168 hours. Probit analysis of the Proline and L-leucine treatments was not performed, as the mortality associated with those treatments was too low (Figures 5-5 and 5-6). Discussion Although not commonly encountered, the Dform of methionine had virtually the same effect as the Lform on larval mosquito mortality. The D-and L-methionine trials showed that the Dform had lower mortality associated with it than the more reactive L-counterpart. Insects do not commonly use the Dform of amino acids, although D-methionine is metabolized by some orders to a limited extent (Ito and Inokuchi, 1981). The YFM could be an example of this phenomenon. Because of the nature of the CAATCH1 system in the alkaline midgut, buffering may have acted to increase the effectiveness of the system. Buffering the solutions did result in an increase in mortality, with even the lowest concentration of 0.1% L-methionine exhibiting a two-fold increase with the buffered form (Figure 5-4). Complete mortality was reached sooner with the buffered forms even for concentrations that did not reach 100% in the unbuffered form. In a field setting, the addition of L-methionine would be buffered naturally by the chemical properties of the bodies of water to which it was applied and similar results would be expected.

PAGE 79

67 Jaffe and Chrin (1979) found the adults of YFM females infected with Brugia, a filiaral parasite, were depleted of free form methionine because of the infection and were able to make up the difference by converting homocysteine to methionine with a special synthetase. The ability of YFM adults to synthesize methionine from homocysteine may be present in the larvae as well. This could be the result of the lack of methionine in the diet and possible evidence of the CAATCH1 system being present in at least the adult stage. The susceptibility of the larvae to L-methionine also could be the result of overexposure to a compound that is normally not encountered in high concentrations (>0. 1 %). However, the alkalinity of the particulate feeding larvae and the high mortality to L-methionine suggests that the CAATCH1 system is present and could be exploited in other species with similar midgut characteristics (Dadd, 1975). The survival of YFM larvae exposed to both Beta-alanine and L-leucine was unusual in that they each had the opposite effect on the YFM larvae. L-leucine was expected to have similar blocking properties as L-methionine based on CAATCH1 research (Feldman et al, 2000). Instead, almost no mortality was observed indicating the possibility of another system involved with the transport of this amino acid. Conversely, beta-alanine was not found to be reactive with the CAATCH1 system based on the work of Feldman et al. (2000). The unusually high larval mortality associated with this amino acid may be the result of a yet to be discovered midgut property. The similar mortalities observed for the higher concentrations of L-methionine and Bti is encouraging considering the resistance to this compound that has been documented in many insect species because of reduced receptor activity and binding (Bills et al., 2004; Nester et al., 2002). Resistance in insects involves a variety of

PAGE 80

68 mechanisms and many are the result of a combination of different pesticide classes. The CAATCH1 system is one that could be exploited in cases where the only alternative is applying different or higher rates of pesticides to break resistance. Further research is needed to determine compatibility of Bti and L-methionine for cases in which resistance is observed in natural populations. Given the safety of L-methionine and the similar time required for 100% mortality (when compared to Bti\ this compound could represent a viable alternative to traditional biorational compounds used in the management of the YFM or other susceptible pest mosquito species.

PAGE 81

CHAPTER 6 FIELD EVALUATION OF L-METHIONINE AS AN INSECTICIDE Introduction The role of methionine in animal systems is well known and only recently understood in plants. Methionine is required for protein synthesis; it is a precursor to several important biochemical compounds including ethylene and polyamines, sulfate uptake and assimilation, and also acts as an activator of threonine-synthase (Giovanelli et al. 1980; Droux et at. 2000; Bourgis et al. 2000; Zeh et al. 2001). Recently, research has focused on the transgenic modification of crop plants to overproduce methionine in order to increase their nutritional quality without affecting other biochemical processes (Zeh et al 2001). However, little work has been conducted on the effects of exogenous methionine and it became important to understand the role of externally applied methionine on plant health. Furthermore, the application of L-methionine to plants exposed to natural conditions presents additional problems in terms of how long the residue remains on the plant. Observations of other experiments using L-methionine revealed the tendency of this compound to crystallize after the aqueous portion evaporated forming a brittle, crusty coating that is easily removed. This coating does not appear to interfere with respiration and transpiration at the concentrations studied (1% and lower). To prevent the loss of L-methionine from the plants in a natural setting, the adjuvant Silwett L-77 (Helena Chemical; Collierville, TN) was included in this portion of the study in an effort to increase residual activity on the plant. Silwet L-77 is a nonionic organosilicate 69

PAGE 82

70 surfactant that has wetting and spreading properties (Helena Chemicals 2002) and was found to be compatible with solutions of L-methionine. The objectives for this portion of the study were to examine the effects of a methionine and Silwet L-77 mixture on a crop plant (eggplant) in terms of yield (both fruit weight and total yield) and to evaluate this mixture as an insecticide under natural conditions. Materials and Methods Preliminary Investigation of Silwet L-77 and L-methionine Adult CPBs were obtained from the University of Florida Horticultural Unit, Gainesville and held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth (to facilitate cleaning) and held at 27C, 60% relative humidity and 16L/8D photoperiod in FRIUs. Twenty-four adults were exposed used in each of the 5 treatments, with 4 replicates per treatment (niotar^O). Adults were used because of the lack of sufficient numbers of larvae to test. Excised leaves were dipped in solutions of deionized H2O containing different concentrations of methionine and Silwett L-77 (0.5% concentration), 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine and controls of deionized H 2 0 and deionized H 2 0 +Silwet L-77. The additional control was to determine the possible insecticidal properties of Silwet L-77 alone and to make sure the addition of this adjuvant did not affect mortality or deter feeding. Plot Design Eggplants {Solarium melongena L.,"Classic" variety) were grown and maintained at the University of Florida Horticultural Unit, Gainesville, from 18 June to 04 November 2001 Eight, one hundred ft. rows of plants were used for this study, with two rows on each side consisting of buffer rows and four rows in the middle used for the experiments.

PAGE 83

71 Each row contained the 4 treatment plots of 10 plants (control (0% L-methionine), 0.1% L-methionine, 0.5% L-methionine and 1 .0% L-methionine in deionized water solutions) in a Latin square design. Plants within treatment plots were spaced 3 feet apart while treatment plots were 9 feet apart. Figure 6-1 shows the diagrammatic representation of the field plot. Plant Yield Before beginning the experiment, all developing eggplants were removed from the plants in an effort to standardize the treatments and ensure all eggplant development occurred after the exposure of methionine. Treatments were administered using a KQ 3L CO2 (Weed Systems, Inc.; Hawthorne, FL) backpack sprayer charged to 30 lbs PSI and a 3-nozzle boom to ensure complete coverage of the plant (Figure 6-2). Each treatment consisted of a 3L application over the 4 representative groups. The adjuvant Silwett L-77 (0.5% concentration) was included to improve the residual effect of the methionine under the field conditions. Plants were sprayed a total of nine times at approximately two-week intervals. Fruits were harvested at various times during the study and were weighed in the field using a Tokyo Electronics hand-held digital scale. Pest Introduction Neonate CPB larvae were reared on excised eggplant leaves for two days at 27C, 60% relative humidity and 16L/8D photoperiod in FRIUs to ensure healthy individuals for the test. Larvae were transferred to the field plants using a camel hairbrush and the branch marked with flagging tape. Introduction was made after the last spray treatment in November. Ten larvae were placed on each plant for a total sample size of 1 ,600 individuals. Plants were inspected for the next 5 days and larvae encountered noted.

PAGE 84

Barrier Barrier Rows Rows Control (A), 0.1% (C), 0.5% (B) and 1.0% (D) Figure 6-1 Overview of the design layout used to study the effects of L-methionine and Silwett L-77 solutions on yield of eggplant. Rows were four feet apart with individual plants three feet apart and treatments nine feet apart. Each letter represents a group of ten eggplants.

PAGE 85

73 Figure 6-2. Weed Systems, Inc. KQ 3L CO2 backpack back sprayer used for application of L-methionine and Silwett L-77 solutions. Boom consisted of three nozzles (middle top and end of each arm). In total, 3L were applied per treatment every two weeks from 09 July to 3 1 August 2001

PAGE 86

74 Data Analysis Data from the fruit and the CPB experiments were analyzed with ANOVA using Mini tab Version 12. Survivorship of CPB was corrected using Abbott's formula (Abbott 1925) to account for control mortality, mean separation was performed using Tukey's multiple comparison procedure (Zar 1999). Data for both the eggplant weight mean per treatment and also mean number of eggplants per treatment were analyzed using paired t-test. Results Effects of L-methionine and Silwett L-77 on CPB Adults Under Laboratory Conditions Little mortality was observed with the adult CPB at the 1 .0% L-methionine concentration (Figure 6-3). The 0.5% L-methionine concentration had the highest mortality of all the treatments at approximately 20% with the other treatments showing no adverse effects after correction for control mortality. Effects of L-methionine and Silwett L-77 on yield In total, 735 eggplants were collected during the course of this study from 09 June to 3 1 August 2001 Mean weight and yield of eggplants between the treatments were not statistically different from each other (Figures 6-4). Control plants produced 195 fruits with a mean weight of 276.9 grams, followed by the 0.1% treatment with 191 fruits at 281.2 grams. The 0.5% and 1.0% treatments yielded 175 and 174 fruits with mean weights of 295.7 grams and 283.6 grams, respectively. Survival of CPB larvae No statistical difference in survivorship of CPB larvae was observed between the three treatments for the first day after exposure (Figure 6-5) but treatment differences

PAGE 87

75 100 80 | to r 5 5 40 ^•Control Water A 0.10% -—0.50% •-1.00% -—Control Silwet 3 4 5 6 Days of Exposure Figure 6-3. Mortality of Colorado potato beetle adults exposed to excised eggplant leaves treated with L-methionine and the adjuvant Silwett L-77 (protaT^O). Data corrected for control mortality using Abbott's formula. Note the overlap in trend lines for the Control treatments and 0.1% L -methk>nine treatment.

PAGE 88

76 (Error Bars @ 95%; F (I)3> 1S6 =2.6626, F =030963; F= 0.81840) B Control 0.1% 0.5% 1.0% (n=195) (n=191) (n=175) (n=174) Figure 6-4. Effects of L-methionine and Silwett L-77 on eggplant yield (A) and mean weight in grams of fruit (B) from 09 June to 31 August 2001. Error bars denote 2 SE. There was no statistical difference for either eggplant yield or mean eggplant weight (Tukey's MST, P=0.05).

PAGE 89

77 80 Days after treatment Figure 6-5. Mortality of Colorado potato beetle larvae on eggplants treated with L-methionine and Silwett L-77. Mortality of larvae corrected using Abbott's formula (Abbott, 1925). Analysis performed on arcsin transformed data. Error bars denote 2 SE. Data points having by the same letter are not statistically different (Tukey's MST, P=0.05)

PAGE 90

78 were observed thereafter. By Day 4 the 1 .0% and 0.5% treatment were the only treatments that were statistically different from the control. There was substantial unexplained attrition of CPB larvae in the field for all treatments, which leveled off by Day 3. Data from day 5 was discounted because of the onset of a severe cold front that made it difficult to separate the effects of the weather from the treatments affects. Discussion The results of the field studies show that, using conventional application techniques, a mixture of methionine and Silwett L-77 did not appear to affect eggplant yield. Furthermore, the same combination produced substantial control of CPB larvae under natural field conditions after four days. Dahlman (1980) found that L-canavanine, a non-protein amino acid, could be used in the same manner for control of THW on tobacco, but the widespread use of this compound was limited by the cost ($107.85 for lg L-canavanine versus $3.35 for lg of L-methionine (Fisher Scientific International 2004)), adverse effect on plant development (Nakajima et al. 2001), and toxicity to vertebrates (Rosenthal 1977). Although complete coverage of the plant was not feasible, approximately 2.5 grams to 7.5 grams of L-methionine was applied to the plants in each of the treatment plots. Each plant, based on the amount applied, received approximately 7.5xl0 6 ug for the 1.0% L-methionine treatment, 3.8xl0 5 ug for the 0.5% L-methionine treatment and 2.5x1 0 4 ug for the 0.1% L-methionine treatment. This compares to only 4ug of L-canavanine, which resulted in decreased size, fecundity, and mortality of THW under field conditions (Dahlman 1980). It should be noted that the toxicity of L-canavanine is well documented and has a different mode of action than L-methionine and cannot be

PAGE 91

79 compared directly. However, the cost for the amount of L-canavanine required for largescale application far exceeds that of the largest amount of L-methionine needed. Despite the lack of a statistical difference between treatments for both mean weight and mean yield of eggplant, there were some interesting disparities within the data. First, there was an observable difference in mean weight of the eggplants between the treatments and the control. All eggplant weights were greater for the treatments than the control, with the 0.5% L-methionine concentration treatment producing the highest mean eggplant weight. It would appear that excess methionine decreases the number of fruit produced, but those fewer eggplants weighed more. Further research is needed to better understand the differences observed during this study. The addition of Silwet L-77 did not appear to adversely affect survival of CPB as seen in the preliminary tests on the adults and on the larvae during the field release (Figures 6-3 and 6-5). The low adult mortality observed could be attributed to the ability of this species to stop feeding and fly to a more suitable food source. Because the adults were unable to move to an untreated leaf, they were observed sitting motionless on the underside of the leaves. This was not observed in either of the controls as they were seen actively feeding the majority of the time. One aspect of this research that was not examined is that of fertility and fecundity of adults exposed to excess amounts of L-methionine. Despite the fact that methionine is used for egg production in many insect species, excess concentrations may act as a deterrent to feeding causing the adults to stop feeding and to seek other food sources. The lag time from the cessation in feeding to finding another food source may be long

PAGE 92

enough to significantly lower the fecundity of the females and possibly interfere with other behaviors such as mating. During the course of this portion of the study, some anecdotal data were collected based on personal observations. Predators (mainly arachnids) were observed on the plants until the end of the experiment. Other insects also were observed feeding on plants after treatments including piercing-sucking insects (i.e., aphids, coreids and cicadellids) with foliage feeders such as caterpillars rarely encountered except found only on control plants. Attempts to control predators via manual removal were unsuccessful, and predation may have contributed to the observed decrease in CPB. Because predators were present on all treatments, loss from predation was corrected with the use of Abbott's formula. The presence of natural enemies indicates the selectivity of the L-methionine in the field. The amount of methionine ingested by the predators was probably very small because they fed on other insects not plant material. Another set of observations on the safety of L-methionine was the exposure of potted eggplants to high (1 .0% methionine in distilled H 2 0 solution). In total, five plants were sprayed daily with the methionine solution and compared to five plants sprayed with water alone for 14 days. The only difference in the plants was the browning of the leaf tips and edges of the methionine sprayed plants. This also was seen in the excised leaf experiments with THW and CPB. A possible reason for this occurrence was the excess sulfur in the methionine might have burned the leaves. As mentioned earlier, the concentration was very high and also applied daily. Applications of the same concentration did not affect the plants in the field plots, indicating that treatments conducted at 2-week intervals would be safe for the plant.

PAGE 93

81 Overall, it appears that L-methionine can be used in a natural setting to control CPB larvae without affecting crop production. The adjuvant Silwett L-77 worked well with L-methionine in controlling CPB larvae but not the adults. The lack of effectiveness on the adults may be attributed to their ability to stop feeding and living off of reserves acquired during the larval stage until suitable food sources can be found. It is unknown if L-methionine, alone or in combination with Silwett L-77 adversely affects fecundity of the adults.

PAGE 94

CHAPTER 7 EFFECTS OF L-METHIONINE ON SURVIVAL AND DEVELOPMENT OF SELECTED NONTARGET SPECIES Introduction A biorational pesticide is defined as one that is effective against pest species but innocuous to non-target organisms and not disruptive to biological control agents and beneficial species (Stansly et al. 1996). To test L-methionine as a potential pesticide and determine if it could be considered biorational, it was necessary to examine the effects of this compound on selected nontarget species that could possibly come into contact with it, either directly while on the plant or indirectly via incidental contact or as a host that has come into direct contact with this compound. The species chosen reflect a variety of non-target organisms, mainly those that were shown to be important in controlling some pest species. The pink spotted ladybird beetle, Coleomegilla maculata (DeGeer), the mottled water hyacinth weevil, Neochetina eichhorniae Warner, and the greenbug parasitoid, Lysiphlebus testaceipes (Cresson) all are beneficial insects that have been effective against pests in the state of Florida and also are common and readily available. Each species also represents a different feeding guild (predator, herbivore and parasitoid, respectively) to ensure a thorough examination of the possible effects of methionine as it might be encountered in under natural conditions. The pink spotted ladybird beetle (PSLB) is an abundant polyphagus species that is known to feed on many lepidopteran and coleopteran pests, including the Colorado potato beetle, in which it was responsible for over 50% of the predation on eggs and early 82

PAGE 95

83 instars (Andow and Risch 1985; Giroux et al. 1995; Griffin and Yeargan 2002; Groden et al. 1990; Hazzard et al. 1991; Hilbeck and Kennedy 1996; Munyaneza and Obrycki 1998). This species is widespread throughout North America, and has been shown to provide effective biological control in several crop species, including corn, crucifers, tomato and potato (Hoffman and Frodsham 1993). However, the PSLB was found to be susceptible to carbaryl and menthamidophos, the same pesticides used for the control of many aphid species (Hoffman and Frodsham 1993), Since its introduction into the United States in 1884, water hyacinth {Eichhornia crassipes (Mart.) Solms-Laubach) has infested waterways of the southeast that has cost upwards of $2 million to control in Florida alone (Schardt 1987). The mottled water hyacinth weevil (MWHW), native to Argentina, was first released in Florida in 1972 and subsequently to other states and countries in an effort to control water hyacinth (Center 1994). The genus is restricted to feeding on members of Pontederiaceae, with the MWHW feeding mainly on the introduced water hyacinth; it can be found virtually everywhere the host plant is present (Haag and Habeck 1991 ; Center et al. 1998). The greenbug parasitoid (GBP) is an important natural enemy of many cereal aphids. This species is known for the production of "mummies", the bodies of parasitized aphids that act as a protective case for the developing wasp pupa, and is considered by many to be tolerant to cold temperatures (Elliott et al. 1999; Knutson et al. 1993; Wright 1995). However, this greenbug parasitoid is an insect and is just as susceptible to pesticides despite the protective case of the immature form (Knutson et al. 1993).

PAGE 96

84 The purpose of this portion of the study was to examine the effects of Lmethionine on selected nontarget species that are both important in terms of being beneficial in controlling other pest species and also represent different feeding guilds that would come into contact with this compound in different ways (e.g., on prey items, on plant surfaces, hosts of parasitoids). Materials and Methods Coleomegilla maculata Adults were obtained from ENTOMOS, LLC (Gainesville, Florida), and were held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth stage inserted (to facilitate cleaning) at 27C, 60% relative humidity and 16L/8D photoperiod in FRIUs. Natural diet consisted of excised cotton leafs infested with aphids (Aphis gossypii Glover (Hemiptera: Aphididae)). Leaves were then dipped into either a 1 .0% L-methionine solution or 0% L-methionine (control) mixed with deionized H 2 0. Five adults were used in each replicate for a total n=30 for each treatment. Leaves were replaced every other day from 27 October 2002 to 07 November 2002. Artificial diet was obtained from ENTOMOS and prepared according to their guidelines with the exception of the inclusion of methionine for the 1.0% L-methionine treatment (wt/wt). Diets were replaced every other day from 27 October 2002 to 07 November 2002. Ten adults were used for each replicate for a total n=60 for each treatment. Data was normalized to 0% mortality when the treatments were corrected for control mortality (i.e., when the control mortality was greater than that of the treatment).

PAGE 97

85 Neochetina eichhorniae Adults of the MWHW were used in this study since the larvae and pupae are buried deep in plant tissue and therefore not likely to come into contact with methionine that could be present in a body of water. Specimens were supplied by Hydromentia, Inc. (Ocala, FL), from areas around South Florida. Weevils were maintained following the procedures outlined by Haag and Boucias (1991), with small petri dishes fitted with moistened filter paper and freshly cut water hyacinth leaves. Water hyacinth plants were collected from Lake Alice on the campus of the University of Florida and maintained in the University of Florida, Department of Entomology and Nematology greenhouse. Treatments consisted of cut leaves dipped in deionized H2O (control) or solutions containing 0.1% L-methionine, 0.5% L-methionine, 1.0% L-methionine or 1.0% proline. Prior to weevil exposures, each leaf was inspected for feeding scars or damage and noted to ensure the counts were based on current feeding. Each treatment consisted of 4 replicates with n=5 per replicate (n=20 per treatment and total n=100). Weevils and hyacinth leaves were held in 26.4L x 19.2W x 9.5H (cm) clear plastic boxes with a hardware cloth (to facilitate cleaning) and maintained at 27 C, 60% relative humidity and 16L/8D photoperiod in FRIUs. Fresh leaves were provided every 4 days; exposed leaves were preserved in sealed plastic bags and placed in a refrigerator until scars could be counted. Feeding damage was determined (with the use of an Olympus Tokyo Model 213598 stereo microscope) by the total number of scars present with each counted scar marked with a fine tipped permanent marker (Figure 7-3). Statistical analyses of the weevil data were performed using Minitab Version 12 (Minitab, Inc.; State College, PA). Feeding scars on control and treatment leafs were

PAGE 98

86 compared with a One-way ANOVA and mean separation was performed using Tukey's Multiple Comparison test (Zar, 1999). Lysiphlebus testaceipes To test the effects of methionine on the GBP, cotton plants (Gossypium sp; Family: Malvacae) were grown and maintained at the University of Florida, Department of Entomology and Nematology green and shade houses from 07 October 2002 to 25 November 2002. Aphids (A. gossypii Glover) were supplied from other experiments using this organism and kept on plants within a sealed greenhouse to prevent unwanted parasitism. Plants were maintained in the sealed greenhouse, infested with aphids and then placed in the open shadehouse area to encourage parasitation. In total, 20 plants were used for 2 treatments, 1.0% L-methionine and 0% L-methionine (Control) mixed with deionized H 2 0. Plants were sprayed weekly (12 October 2002 through 17 November 2002) with approximately 10 ml of solution using a hand-held spray bottle. Counts of parasitized aphids began approximately two weeks after placing plants outside to ensure adequate time for parasitism (Royer et al. 2001). Counts were made using a hand lens and counter; "mummies" with exit holes were enumerated and removed. A few parasitized aphids were removed and held in glass vials to ensure correct identification of the parasitoid. Data Analysis Data from the parasitoid experiments were analyzed using Minitab Version 12 (Minitab, Inc.; State College, PA). Control and experimental plants were compared against one another with a One-way ANOVA and separation of significant means was performed with Tukey's Multiple Comparison test (Zar, 1999).

PAGE 99

87 Results Coleomegilla maculata There was virtually no difference between the control and treatment groups for either the artificial or natural diet tests after correction for control mortality. Mortality was slightly higher for the control groups than the 1.0% L-methionine treatment (Figures 7-1 and 7-2). Further analysis was not necessary because of the identical numbers. Neochetina eichhorniae Total mortality for the treatments was less than 20% for all treatments, with the individual treatments having similar results (Figure 7-4). Feeding damage ranged between 2,000 and 4,000 scars per treatment and an average of 10.7 to 16.9 scars per survivor during the course of the experiment (Figure 7-5). No statistical differences were observed between the treatment and control groups Lvsiphlebus testaceipes In total, 188 and 232 aphid mummies with exit holes were found on treatment and control plants, respectively. Means for each treatment were not statistically different for each collection period or overall based on One-way ANOVA (Figure 7-6) with the only exception being the second and last collection period Discussion In general, L-methionine did not have the same toxic effect on the non-target organisms tested when compared to the pest species exposed to the compound in previous chapters. The pink spotted ladybird beetle adults actually showed the least amount of susceptibility to L-methionine. Survival of the adult beetles was higher in the 1 .0% L-methionine treatments than the control for both the artificial and natural diet

PAGE 100

88 100 80 60 40 20 0 "ControlAD ^H 1 .0% L-methionineAD Survivorship of 1 .0%L-methionine Grp> Control Grp 23456789 Days After Exposure 10 11 12 Figure 7-1. Mortality of Coleomegilla maculata adults after exposure to L-methionine treated artificial diet. Data corrected for control mortality using Abbott's formula.

PAGE 101

89 C8 t 100 80 60 40 •ControlND • 1 .0% L-methionine ND Survivorship of 1 .0%L-methionine Grp> Control Grp 4 5 6 7 8 Days After Exposure 10 11 12 Figure 7-2. Mortality of Coleomegilla maculata adults after exposure to Lmethionine treated cotton plant leaves infested with aphids. Data corrected for control mortality using Abbott's formula.

PAGE 102

90 Figure 7-3. Feeding scars on water hyacinth (Eichhornia crassipes) leaf after exposure to Neochetina eichhorniae adults. Black marks represent feeding scars marked with a fine tip marker to aid in counting (other side counted but not shown).

PAGE 103

91 100 80 5? | 60 r o 40 6 8 Days of Exposure 10 Control 0.10% 0.50% 1.00% •Proline 12 14 Figure 7-4. Mortality of Neochetina eichhorniae on treated water hyacinth leaves. Data corrected for control mortality using Abbott's formula

PAGE 104

Figure 7-5. Feeding rate of Neochetina eichhorniae on water hyacinth leaves treated with L-methionine and Proline. No statistical differences were observed between treatments (Tukey's MST, P=0.038).

PAGE 105

(Error bars @ 95%; F (00S)I>18 =4.41; F=3.25; P=0.005) Figure 7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with L-methionine. Ten plants were used for each treatment and held in the shade house at the University of Florida, Department of Entomology and Nematology from 22 October to 25 November 2002. No statistical differences were observed except for the second and final collection date.

PAGE 106

trials. One possible explanation for this observation could be that the excess L-methionine increased the dietary quality of the artificial and natural diets for the PSLB in the treatments. However, because only adults were available, further tests are needed to determine if the larvae, also predaceous on the same pests as the adults, are sensitive to this compound. It should be noted that the midgut properties (i.e., alkalinity) for this species are not well known and may not even have the CAATCH1 proteins present in the midgut. The mottled water hyacinth weevil also appears not to be adversely affected by exposure to excess amounts of L-methionine despite its herbivorous habit like the THW and CPB. Another weevil within the same family (Anthonomus grandis Boheman (Coleoptera: Curculionidae)) is known to have an acidic midgut and the same could apply to the MWHW based on these results (Nation 2001). Therefore, this species and possibly other weevils may not be affected by compounds like L-methionine because of the lack of an alkaline midgut needed for the CAATCH1 protein to operate (Feldman et al. 2000; Quick and Stevens 2001). Again, further research is necessary to determine if CAATCH1 proteins are present in this weevil species. The greenbug parasitoid also was unaffected by exposure to the excess L-methionine found on treated leaves infested with aphids. Dadd and Krieger (1968) found higher methionine requirements for the greenbug Myzus persicae Sulzer (Hemiptera: Aphididae) when cysteine is scarce because of its ability to transform excess methionine to much needed sulfur and could possibly explain the parasitoid's tolerance to high methionine concentrations. Because of the life cycle of the GBP, and many other parasitoids, direct contact with compounds such as L-methionine would occur inside the

PAGE 107

95 body of the host, and not through direct contact with the foliage where the compound was applied. There is a possibility for the parasitoid having higher methionine requirements; based on filarial worm infected Aedes aegypti (L.) (Diptera; Culicidae) females and the associated drop in methionine levels in the haemolymph (Jaffe and Chrin 1979). This makes alternatives such as L-methionine safe for use around beneficial insects like the greenbug parasitoid. Overall, the results indicate that the PSLB (C. maculata), the MWHW (N. eichhorniae) and the GBP, (L. testaceipes) were not adversely affected by exposure to L-methionine in excess concentrations in a variety of artificial and natural diets. Survivorship and feeding rates were not statistically different between control and treatment groups for each species. From these data, it can be concluded that L-methionine is safe for use with beneficial insects and could be considered "biorational" in that it showed no adverse effects on non-target species. It also should be stressed that additional testing on other beneficial insects would be, on a case by case basis, necessary to examine the safety and "biorational" qualities of L-methionine.

PAGE 108

CHAPTER 8 SUMMARY AND DISCUSSION The creation and implementation of Integrated Pest Management (IPM) strategies to combat pest species were developed as a response to the economic losses associated with the overuse of chemical control. However. IPM strategies are not widely used because of the lack of alternatives and the ease of use of pesticides. This has resulted in the resistance to pesticides in many insect species, including economic and medical pests. In an effort to provide alternatives to traditional chemical control, biorational methods have been investigated and one such avenue is the use of non-protein amino acids. Chapter 2 covered the history of the use of non-protein amino acids as a pesticide, and discussed the CAATCH1 system and the safety of L-methionine. Only a handful of these amino acids have been investigated as a means of controlling insect pests but still lack the practicality and cost effectiveness as current chemical control methods. Recent discovery of a new midgut membrane protein, CAATCH1, has revealed a new possibility in insect control. The CAATCH1 system works in alkaline conditions and responds to different amino acids, mainly the reduction in ion flow after exposure to methionine, an essential amino acid required for normal development and metabolism of many species including humans. The use of a compound such as methionine would be an excellent addition to the IPM arsenal because of its relative safety to vertebrates and warrants further study as a pesticide. Chapters 3,4, and 5 were dedicated to examining the effects of L-methionine, a common analog of methionine, on three different economic and medically important 96

PAGE 109

97 pests. The tobacco hom worm (THW), Colorado potato beetle (CPB) and the yellow fever mosquito (YFM) were tested and found to be susceptible to concentrations greater than 0.1%. Diets, both natural and artificial, containing this compound resulted in the complete mortality of THW and also in the natural diet for CPB. Development and feeding rates were also affected by the addition of L-methionine to diets for THW and CPB. Survivorship and developmental rates of YFM were also affected by the addition of this amino acid to the larval habitat. In Chapter 6 it was found that the field application of L-methionine under natural conditions was able to control CPB. It was also determined that L-methionine was compatible with Silwett L-77, a commonly used adjuvant, and showed no detrimental effects on crop yield of eggplant. Finally, the application of a compound such as L-methionine has to be able to control the pests that it is used against and not have an effect on beneficial organisms that may come into contact with this compound. Chapter 7 detailed the results of tests that involved various beneficial insects from different feeding guilds (herbivore, predator and parasitoid) showed that L-methionine does not appear to pose a threat to nontarget organisms. One aspect of the use of a compound like L-methionine that is very important is the relative safety. The health hazards related to the contamination of the environment with pesticides are well documented and in the recent years have resulted of the review and removal of several insecticides from commercial and private use. The use of L-methionine as an insecticide would alleviate the dangers associated with other pesticides. The approved use as a nutritional supplement for livestock feed is a testament

PAGE 110

98 to the safety of this compound and residual found on the plant does not pose the same risk to the human population. It is difficult to understand how a compound such as methionine can be considered essential and deadly within the same organism. To understand this dichotomy, an examination of the role of this compound and how it relates to metabolism, development and reproduction is necessary. Although the diet of the THW is lacking high concentrations of methionine, the use of hexamerins may account for the levels needed for the biosynthesis of JH. The larvae take in methionine, metabolizing what is needed and storing the rest for later on during metamorphosis. In contrast, the larvae of the diamondback moth (Plutella xylostella (Lepidoptera: Plutellidae)), feeding mainly on methionine-rich crucifers, lack hexamerins with high methionine concentrations (Wheeler et tiL 2000). The levels of methionine encountered in a normal diet are below what the CAATCH1 proteins are capable of processing and may also be affected by the presence of symbiotic bacteria that is responsible for methionine oxidation in some insects (Gasnier-Fauchet and Nardon 1986a; 1986b). It is when the concentration exceeds the handling capacity of the midgut that problems occur. The time it takes to digest material containing natural amounts of methionine could be long enough for the CAATCH1 system to recover from exposure. The difference between the artificial and natural diet LC 50 for the THW (Figure 3-8) appears to support the idea that bound methionine (i.e., incorporated into the diet and not applied topically) takes longer to cause problems for the organism (if any) versus the relatively quick kill associated with the free methionine present on the leaf surface. The

PAGE 111

99 target ingests the methionine first as it feeds ensuring the overload CAATCH1 system and eventually death. As for the stored methionine, it is released from the storage proteins as needed to synthesize juvenile hormone and allow for transformation in addition to other functions. The remaining methionine is then used for protein synthesis in the tissues around the ovaries to boost yolk production, as seen in the transfer of methionine from male to female Drosophila species (Bownes and Partridge 1987). In the THW, the presence of hexamerins with high methionine content may be an alternative to the male contribution possibly found in its ejaculate. Methionine-rich hexamerins are common in Lepidoptera and have been shown to provide the larvae a source of amino acids during the synthesis of these proteins during the last stage of larval development (Wheeler et al. 2000). In addition to the need for methionine for metabolism and reproduction, the release of methionine may also in part account for the decrease in ion transport of the posterior region of the midgut during larval molts and the wandering stage present before pupation. Currently, little is known regarding the mechanisms involved with the decrease of ion transport during these developmental stages (Lee et al. 1998). Clearly there appears to be more to the role methionine plays in the development of some insects other than the vague designation of "essential" amino acid. Insects have evolved to deal with limiting resources, such as methionine, and have successfully found effective strategies like hexamerin storage or alternate pathways to deal with such problems. No attempt to link together all the aspects of the role of methionine in a whole organism or system context It appears that methionine actually may play a role far more important than that of just an essential amino acid. From the

PAGE 112

100 synthesis of homocysteine to produce methionine to the presence of methionine rich hexamerins and allophorins and protein synthesis, the role of methionine in plant-insect interactions may be larger than originally theorized. The production of methionine overproducing plants could also be used in future IPM strategies. Preliminary results indicate that genetically modified plants do produce enough methionine to affect the survivorship of caterpillars feeding on the plant (unpublished data). This could be used in crops in which improved nutritional quality is important as well as the insecticidal properties of the additional methionine. However, there appears to be a sublethal level (0.1%) of L-methionine in which THW and CPB can "tolerate" and survive with little mortality (Figures 3-9 and 4-1). Any system that makes use of a crop that can overproduce compounds like L-methionine would have to be able to express levels greater than this level to avoid any resistance/tolerance. This research has also provided more possibilities for the use of compounds such as L-methionine in the YFM portion of this study. The amino acid Beta-alanine provided similar levels of control, as did the methionine trials (Figure 5-7). Although unexpected (as discussed in Chapter 5), it shows that there are several other systems that can possibly be exploited in controlling some insects. Further research is necessary to determine if the combination of a compound like methionine and a pesticide already in use would result in the increase in toxicity or the decrease in the concentration of pesticide used. If compatibility between methionine and Bacillus thuringiensis does exists, then it is possible that resistance could be broken in a given population. For example, if a population of THW started to show resistance to Bacillus thuringiensis kurstaki then methionine could be used to remove both susceptible

PAGE 113

101 and resistant alike because of the difference in mode of action. Once the population was reduced, and the corresponding resistant genotype, Btk could be used once more at a lower concentration, closer to that of the susceptible population. This system could also be used for the reduction of Bt toxin resistance in the CPB and YFM if the compounds are compatible. In conclusion, it appears that L-methionine can be used as an insecticide to control insect pests of economic and medical importance. The target site (CAATCH1) is known and found in the midgut/hindgut (presumed) in at least three pest species (tobacco hornworm, Colorado potato beetle and the yellow fever mosquito) and possibly more. The compound (L-methionine) is a safe compound that is already used for livestock feed supplements, has very low mammalian toxicity, and is compatible with insecticide application systems. Non-target organisms were not affected with the application of Lmethionine, further supporting its use as a biorational insecticide. With increasing resistance to current insecticides in the study organisms, alternatives such as Lmethionine are needed now more than ever to further support of Integrated Pest Management strategies.

PAGE 114

LIST OF REFERENCES Abbott, W.S. 1925. A method for computing the effectiveness of an insecticide. J. Econ. Entomol. 18: 265-267. Aerts, M.J. and O.N. Neishiem. 1999. Florida Crop/Pest Management Profiles: Tomatoes. CIR 1238. Pesticide Information Office, Food Science and Human Nutrition Department, Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Internet URL: http://www.edis.ifas.ufl.edu/ BODY_PI039.htm. Accessed April 2004. Anand, R. and M. Anand. 1990. Nutritive effect of the D isomers of the essential amino acids in casein diet on Dacus cucurbitae (Coquillett) maggots. Indian J. Entomol 52(4): 525-528. Andow, D.A., and S.J. Risch. 1985. Predation in diversified agroecosystems: Relations between a coccinellid predator Coleomegilla maculata and its food. J. Appl Ecol. 22: 57-372. Audsley, N., R.J. Weaver and J.P. Edwards. 1999. Juvenile hormone synthesis by corpora allata of tomato moth, Lacanobia oleracea (Lepidoptera: Noctuidae), and the effects of allatostatins and allatotropin in vitro. Eur. J. Entomol. 96: 287-293. Barfield, C.S. and M.E. Swisher. 1994. Integrated pest management: ready for export? Historical context and internationalization of IPM. Food Reviews Internat 10(2V 215-267. Baumhover, A.H., W.W. Cantelo, J.M. Hobgod, Jr., CM. Knott, and J.J. Lam, Jr. 1977. An improved method for mass rearing the tobacco hornworm. Agricultural Research Service, Unites States Department of Agriculture ARS-S-167, 13 pp. Beck, S.D. and W. Hanec. 1958. Effects of amino acids on feeding behavior of the European corn borer, Pyraustra nubilialis (Hiibn.). J. Insect Physiol. 2:85-96. Bell,E.A. 1978. Toxins in seeds, pp. 143-161. IN J. Harbourne (ed.), Biochemical Aspects of Plant and Animal Coevolution. Academic Press, New York. 435pp. Berge, M.A., G.A. Rosenthal and D.L. Dahlman. 1986. Tobacco budworm, Heliothis virescens (Noctuidae) resistance to L-canavanine, a protective allelochemical Pest. Biochem and Physiol. 25: 319-326. 102

PAGE 115

103 Bills, P.S., D. Mota-Sanchez and M. Whalon. 2004. The Database of Arthropods Resistant to Pesticides. Michigan State University Center for Integrated Plant Systems. Internet URL: http:/Avww.cips.msu.edu/resistance/rmdb/. Accessed April 2004. Boucher, T. J. 1999. Using IPM on CPB saves money, insecticides. Yankee Grower 1(2): 7-9. Bourgis, F., S. Roje, M.L. Nuccio, D.B. Fisher, M.C. Tarczynski, C. Li, C. Herschbach, H. Rermenberg, M.J. Pimenta, T. Shen, D.A. Gage and A.D. Hanson. 2000. Smethymethionine has a major role in pholem, sulfer transport and is synthesized by a novel methyltransferase. Pp. 283-284. IN C. Brunold (ed), Sulfur Nutrition and Sulfur Assimilation in Higher Plants. Paul Haupt, Bern, Switzerland. 427pp. Bownes, M. and L. Partridge. 1987. Transfer of molecules from ejaculate to females in Drosophila melanogaster and Drospohila pseudoobscura. J. Insect Physiol 33(12): 941-947. Brogdon, W.G and J.C. McAllister. 1998. Insecticide resistance and vector management. Emerging Infect Diseases 4(4): 605-613. Capinera, J.L, F.D. Bennett and D. Rosen. 1994. Introduction: Why biological control and IPM are important to Florida, pp.3-8. In D. Rosen, F.D. Bennett and J.L. Capinera (eds.), Pest Management in the Subtropics: Biological Controla Florida Perspective. Intercept Limited, Andover, UK. 737pp. Center, T.D. 1994. Biological control of weeds, Chapter 23. pp.481 -521. IN: D. Rosen, F.D. Bennett, J.L. Capinera, (eds.), Pest Management in the Subtropics: Biological Control-The Florida Experience. Intercept, Ltd., Andover, Hampshire, UK. 737pp. Center, T.D., F.A. Dray and V.V. Vandriver, Jr. 1998. Biocontrol with insects: The water hyacinth weevils. Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences, University of Florida. Internet URL: http:// edisjfas.ufl.edu/scripts/htalgen.exe 1 Accessed April 2004. Centers for Disease Control (CDC). 2003. Malaria: General Information. Centers for Disease Control. Internet URL: hta://www.cdc.gov/travel/malinfo.htm. Accessed April 2004. Chen, P.S. 1958. Studies on the protein metabolism of Culex pipens L.-L Metabolic changes of free amino acids during larval and pupal development. J. Ins. Physiol. Cibula, A.B., R.H. Davidson, F.W. Fisk and J.B. LaPidus. 1967. Relationship of free amino acids of some Solanaceous plants to growth and development of

PAGE 116

104 Leptinotarsa decemlineata (Coleoptera: Chrysomelidae). Annals Ent. Soc Am 60(3): 626-631. Cook, R.J., W.L. Bruckart, J.R Coulson, M.S. Goettel, R.A. Humber, R.D. Lumsden, J.V. Maddox, MX. McManus, L. Moore, S.F. Meyer, P.C. Quimby, Jr., J.P. Stack, and J.L. Vaughn. 1996. Safety of microorganisms intended for pest and plant disease control: A framework for scientific evaluation. Biological Control 7: 333-351. Dadd, R.H. 1975. Alkalinity within the midgut of mosquito larvae with alkaline-active digestive enzymes. J. Insect Physiol. 21: 1847-1853. Dadd, R.H. and D.L. Krieger. 1968. Dietary amino acid requirements of the aphid, Myzus persicae. J. Insect Physiol. 14: 741-774. Dahlman, D.L. 1980. Field tests of L-canavanine for control of tobacco horn worm. J. Econ. Entomol, 73: 279-281. Dahlman, D.L., F. Herald and F.W. Knapp. 1979. L-canavanine effects on growth and development of four species of Muscidae. J. Econ. Entomol. 72: 678-679. Dahlman, D.L. and G.A. Rosenthal. 1975. Non-protein arnino acid-insect interations (1) Growth effects and symptomology of L-canavanine consumption by tobacco homworm, Manduca sexta (L). Comp. Biochem Physiol. 5 1 : 33-36. Dahlman, D.L. and GA. Rosenthal. 1976. Further studies on the effect of L-canavanine on the tobacco homworm, Manduca sexta. Insect Physiol. 22: 265-271 Dahlman, D.L. and G.A. Rosenthal. 1982. Potentiation of L-canavanine-induced developmental anomalies in the tobacco homworm, Manduca sexta, by some amino acids. J. Insect Physiol. 28(10): 829-833. Deedat,Y.D. 1994. Problems associated with the use of pesticides: An overview Insect Sci. Applic. 15(3): 247-251. Del Campo, M., and J.A.A. Renwick. 2000. Induction of host specificity in larvae of Manduca sexta: Chemical dependence controlling host recognition and developmental rate. Chemecol. 10: 115-121. Dethier,V.G.andJ.H.Kuch. 1971. Electrophysiological studies of gustation in lepidopterous larvae 1 Comparative sensitivity to sugars, arnino acids and glycosides. Z. Vergl. Phys. 72: 343-363. Dietary Supplement Information Bureau. 2000. Methionine. Dietary Supplement Education Alliance. Internet URL: htto://ww.supplementinfo.org/index htm Accessed April 2004.

PAGE 117

105 Dimond, J.B., A.O. Lea and D.M. Delong. 1958. Nutritional requirements for reproduction in insects. Proc. 10 Int Congr. Entomol. 2: 135-137. Droux, M, B.Gakiere, L. Denis, S. Ravanel, L. Tabe, A.G. Lappartient, D. Job. 2000. Methionine biosynthesis in plants: biochemical and regulatory aspects. Pp. 73-92. ZV Brunold, C, Rennenberg, H., De Kok, L.J., Stulen, L, Davidian, J.C. (eds.): Sulfur Nutrition and Sulfur Assimilation in Higher Plants. Molecular, Biochemical and Physiological Aspects. Paul Haupt Publishers. 447pp. Durham, S. 2000. Hairy vetch thwarts Colorado potato beetle. Agricultural Research Service, United States Department of Agriculture. Internet URL: http://ww.ars.usda.gov/is/pr/2000/000413.htm. Accessed April 2004. Dwyer.J. 1999. Research Links 2000 Tobacco Hornworm. Carleton College, Department of Biology. Internet URL: http://www.acad.carietori.edu/curricular /BIOL/resources/riink. Accessed April 2004. Ehler L.E. and D.G. Bottrell. 2000. The illusion of integrated pest management. Issues in Science and Technology Online. National Academies and the University of Texas (Dallas). Internet URL: rmp://www.nap.edu/issues/l 6.3/ehler.htm. Accessed April 2004. Elliott, N.C., J A. Webster, and S.D. Kindler. 1999. Developmental response of Lysiphlebus testaceipes to temperature. Southwest Entomol. 24: 1-4. Eymann, M. and W.G. Friend. 1985. Development of onion maggots (Diptera: Anthomyiidae) on bacteria-free onion agar supplemented with vitamins and amino acids. Ann. Entomol. Soc. Am. 78: 182-185. Feldman, D.H., W.R. Harvey and B.R. Stevens. 2000. A novel electrogenic amino acid transporter is activated by K + or Na + is alkaline pH-dependent, and is Cl"independent. J. Biol. Chem. 275: 24518-24526 Felton, G.W. and D.L. Dahlman. 1984. Allelochemical induces stress: Effects of Lcanavanine on the pathogenicity of Bacillus thuringiensis in Manduca sexta J Invert. Path. 44: 187-191. Ferro, D.N. 1985. Pest status and control strategies of the Colorado potato beetle. ZV Ferro, D.N. and R.H. Voss (eds.) Proceedings of the Symposium on the Colorado potato beetle, XVTI Inernational Congress of Entomology. Fisher Scientific International. 2004. Online Catalog. Fisher Science International. Internet URL: https://wwl.fishersci.com/index.jsp. Accessed April 2004. Florida FIRST. 1999. Putting Florida FIRST: Focusing ffAS resources on solutions for tomorrow. UniveisityofFlorioX Institute of Food and Agricultural Sciences 16pp.

PAGE 118

106 Fogarty International Center and the U.S. National Institutes of Health (FIC-NIH). 2003. Multilateral Initiative on Malaria. U.S. National Institutes of Health. Internet URL: http://mim.nih.gov/englishyindex.html. Accessed April 2004. Forgash, AJ. 1985. Insecticide resistance in the Colorado potato beetle, pp. 33-52. IN D.N. Fcrro and R.H. Voss (eds.) Proceedings of the Symposium on Colorado Potato Beetle. XVII International Congress of Entomology, Massachusetts Agricultural Experiment Station Bulletin 704. Amherst Massachusetts. Friend, W.G., R.H. Backs and L.M. Cass. 1957. Studies on amino acid requirements of larvae of the onion maggot, Hylema antiqua (MG.), under aseptic conditions Can. J. Zool. 35: 535-543. Gasnier-Fauchet, F. and P. Nardon. 1986a. Comparison of sarcosine and methionine sulfoxide levels in symbiotic and aposymbiotic larvae of two sibling species, Sitophilus oryzae and Sitophilus zeamais (Coleoptera: Curculionidae). Insect Biochemistry 17(1): 17-20. Gasnier-Fauchet, F. and P. Nardon. 1986b. Comparison of methionine metabolism in symbiotic and aposymbiotic larvae of Sitophilus oryzae L. (Coloeptera: Curculionidae)II. Involvement of the symbiotic bacteria in the oxidation of methionine. Comp. Biochem. Physiol. 58(1): 251-254. Gauthier, V.L., R.N. Hoffmaster and M. Semel. 1981 History of Colorado potato beetle control, pp. 13-34. IN J.H. Cashcomb and R. Casagrande (eds.), Advances in Potato Pest Management. Hutchinson and Ross, Stroudsburg, PA. 672pp. Geer.B.W. 1966. Utilization ofD-amino acids for growth by Drosopbila melanocaster larvae. J. Nutr. 90: 3 1 -39. Giordana, B, M. Forcella, M.G. Leonardi, M. Casartelli, L. Fiandra, G.M. Hanozet and P. Parenti. 2002. A novel regulatory mechanism for amino acid absorption in lepidopteran larval midgut. J. Insect Physiol. 48: 585-592. Giovanelli,J,S.H.MuddandA.H.Datko. 1980. Sulfur amino acids in plants. Z2V BJ Miflin (ed.) The Bioclieinistry of Plants VoL 5, Academic Press, New York, pp. Giroux, S., R.M. Duchesne and D. Coderre. 1995. Predation of Leptmotarsa decemlimata (Coleoptera: Coccinellidae) by Coleomegilla maculata (Coleoptera: Coccmelhdae): Comparative effectiveness of predator developmental stages and effect of temperature. Environ. Entomol. 24: 748-754. Glare, TJR, and M. O'Callaghan, 1 998. Environmental and Health Impacts of Bacillus thurmgiensis isrealensis. Report for the New Zealand Ministry of Health, 58pp.

PAGE 119

107 Griffin, MX. and K.V. Yeargan. 2002. Opposition site selection by the spotted lady beetle Coleomegilla metadata (Coleoptera: Coccinellidae): Choices among plant species. Environ. Entomol. 31(1): 107-111. Groden, E., F.A. Drummond, FLA. Casagrande and D.H. Haynes. 1990. Coleomegilla maculata (Coleoptera: Coccinellidae): Its predation upon the Colorado potato beetle (Coleoptera: Chrysomelidae) and its incidence in potatoes and surrounding crops. J. Econ. Entomol. 83: 1306-1315. Gubler, D.J. 1998. Resurgent vector-borne diseases as a global health problem Emerg Infect. Dis. 4(3): 442-450. Gubler, D.J. and G.G. Clark. 1995. Dengue/Dengue Hemorrhagic Fever: The emergence of a global health problem. Emerg. Infect. Dis. 1(2): 55-57. Haag, K.H. and D.G. Boucias. 1991 Infectivity of insect pathogens against Neochetina eichhorniae, a biological control agent of water hyacinth. Florida Ent 74(1)128133. Haag, K.H. and D.H. Habeck. 1991. Enhanced biological control of water hyacinth following limited herbicide use. Aquat. Plant Management 29: 24-28. Harrison, J. and R. Holliday. 1967. Senescence and me fidelity of protein synthesis in Drosophila. Nature 214: 990-993. Hazzard, R.V., D.N. Ferro, R.G. van Driesche and AJ\ Turtle. 1991. Mortality of eggs of Colorado potato beetle (Coleoptera: Chrysomelidae) from predation of Coleomegilla maculata (Coleoptera: Coccinellidae). Environ. Entomol. 20841848. Hegdekar, B.M. 1970. Amino acid analogues as inhibitors of insect reproduction J Econ. Entomol. 63: 1950-1956. Heim, D.C., G.G. Kennedy and J. W. Van Duyn. 1990. Survey of insecticide resistance among North Carolina Colorado potato beetle (Coleoptera: Chrysomelidae) populations. J. Econ. Entomol. 83(4): 1229-1235. Helena Chemicals, Inc. 2002. Silwett L-77. Technical Data Sheet SL77080596. Helena Chemicals, Inc. Internet URL: hrtp://www.helenachemicalwest com/datayTDS/Silwett.pdf. Accessed April 2004 Hilbeck, A. and G.G. Kennedy. 1996. Predators feeding on Colorado potato beetle msectjcide-rree plots and insecticide-treated commercial potato fields in eastern North Carolina. Biol. Control. 6: 273-282. Horrmarm,M.P.andA.C.Frodsham. 1993. Natural Enemies of Vegetable Insect Pests Cooperative Extension, Cornell University, Ithaca, NY. 63pp.

PAGE 120

108 Ito, T. and T. Inokuchi. 1 98 1 Nutritive effects of D-amino acids on the silkworm, Bombyx mori. J. Insect Physiol. 27(7): 447-453. Jaffe, J.J. and L.R. Chrin. 1979. De novo synthesis of methionine in normal and Brugiainfected Aedes aegypti. J. Parasitol. 65(4): 550-554. Jones, D.C and R.M. MacPherson. 1997. Tobacco Insects: Summary of losses from insect damage and costs of control in Georgia 1997. University of Georgia, Warnell School of Forest Resources and College of Agricultural and Environmental Sciences Internet URL: http ://www. bugwood org/tobbaco97 htm Accessed April 2004. Kaldy.MS. and A.M. Harper. 1979. Nutrient constituents of a gram aphid, Metopolophium dirhodum (Homoptera: Aphididae), and its host, oats (Avena sativa). Canadian Entomol. 1 1 1(7): 787-790. Kammer, A.E., D. L. Dahlman and G.A. Rosenthal. 1978. Effects of the non-protein aminoacids L-canavanine and L-canaline on the nervous system of the moth Manduca sexta (L). J. Exp. Biol. 75: 123-132. Kasting, R.. G.R.F. Davis amd A.J. McGinnis. 1962. Nutritionally essential and nonessential amino acids for the prairie grain wireworm, Ctenicera desctructor Brown, determined with Glucose-U-C. J. Insect Physiol. 8: 589-596. Knutson, A., Boring ED, E.P., Michaels, Jr., G.J., and Gilstrap, F. 1993. Biological Control of Insect Pests in Wheat Texas Agric. Ext Service Publ. B-5044 8pp. Koo,S.I.,T.A.Curriii,M.G.Johi^^ 1980. The nutritional value and microbial content of dried face fly pupae (Musca autumnal is (DeGeer)) when fed to chicks. Poultry Sci. 59: 25 14-25 1 8. Koyama, K. 1985. Nutritional physiology of the brown rice planthopper Nilaparvata lugens StAl (Hemiptera: Delphacidae). U. Essential amino acids for nymphal development Appl. Ent Zool. 20(4): 424-430. Koyama, K and J. Mitsuhashi. 1975. Essential amino acids for the growth of the smaller brown planthopper, Laodelphax striatellus FallSn (Hemiptera: Delphacidae) Appl. Ent Zool. 1 0(3): 208-2 15. Lee,K. F.M. Horwrysld, and M.E. Chamberlin. 1998. Inhibition of midgut ion transport by allatotropin (Mas-AT) and Manduca FLRFamides in the tobacco hornworm Manduca sexta. J. Exp. Biol. 201 : 3067-3074. Leonardi, M.G., M. Casartelli, L. Fiandra, P. Parenti and B. Giordana. 2001. Role of specific activators of intestinal amino acid transport in Bombyx mori larval growth and nutrition. Arch. Insect Biochem. Physiol. 48: 190-198.

PAGE 121

109 Mallinckrodt Baker, Inc. 2001. L-mcthioninc Material Safety Data Sheet MSDS Number M2108. JT Baker Inc. Internet URL.http7Avww.jtbakcr.com /msds/englishhtml/M2 1 08.htm. Accessed April 2004. Marrone, P.G. and S.C. Macintosh. 1993. Resistance to Bacillus thwingiensis and resistance management, pp. 221-236. IN, P.F. Entwistle, J.S. Cory, M.J. Bailey and S. Higgs (eds.), An Environmental Biopesticide: Theory and Practice John Wiley and Sons New York. 330pp. McPherson, RM. and D.C. Jones. 2002. Tobacco Insects: Summary of losses from insect damage and costs of control in Georgia-2001. University of Georgia Integrated Pest Management Internet URL: http://cntomology.cnt.uga.edu/IPM/ slOl/tobacco.htm. Accessed April 2004. Melangeli, C, GA. Rosenthal and D.L. Dahlman. 1997. The biochemical basis for Lcanavanine tolerance by the tobacco budworm Heliothis virescens (Noctuidae) Proc. Natl. Acad Sci. USA. 94: 2255-2260. Mitchell, B.K. 1974. Behavioral and electrophysiological investigations on the responses of larvae of the Colorado potato beetle (Leptinotarsa decemlineata) to ammo acids. Ent. Exp. & Appl. 1 7: 255-264. Mitchell, B.K. and L.M. Schoonhoven. 1974. Taste receptors in Colorado potato beetle larvae. J. Insect Physiol. 20: 1 787-1 793. Mitder, T.E. 1 967a. Effect of amino acid and sugar concentrations on the food uptake of the aphid Myzus persicae. Ent Exp. A Appl. 10: 39-51. ^^p.'&Ap^l ^^ fd ^ a ^ acids ^ the aphid Myzus persicae. Ent Muiryar^J.aiidJ.J.Obrycki.1998. Development of three populations of M'omegilla maadata (Coleoptera: CoccineUidae) feeding on eggs of Colorado potato beetle (Coleoptera: Chrysomelidae). Environ. Entomol. 27: 1 17-122. Nakajima,N. S. Hiradate and Y. Fujii. 2001. Plant growth inhibitory activity of Lcanavamne and its mode of action. J. Chem. Ecol. 27(1): 19-31. Nation, J. 2001. Insect Physiology and Biochemistry. CRC Press, Boca Raton. 496pp. Neishman, O N. and K. Vulincc. 2001. Florida CrorVPest Management ProfilesE^bnt Ctt 1264. Pesticide Information Office, F^SderJe^ Hum^ NttoUon Department Florida Cooperative Extension Service, Institute of Food and Agricultural Science, University of Florida. Internet URLhttoy/W.edis.iras.ufl.edu/ BODY_PI045 Jitm. Accessed April 2004

PAGE 122

no Nester, E.W., L.S. Thomashow, M. Metz and M. Gordon. 2002. One Hundred Years of Bacillus thruingiensis: A Critica Scientific Assessment Report from the American Academy of Microbiology. 16p. Onifade, A.A., O.O. Oduguwa, A.O. Fanimo, A.O. Abu, T.O. Olutunde, A. Arije and G. M. Babatunde. 2001. Effects of supplemental methionine and lysine on the nutritional value of housefly larvae meal (Musca domestica) fed to rats. Biores. Technol. 78: 191-194. Palumbo, R.E. and D.L. Dahhnan. 1978. Reduction of Manduca sexta fecundity and fertility by L-canavanine. J. Econ. Entomol. 71 : 674-676. Pan, M.L. and W.H. Telfer. 1996. Methionine-rich hexamerin and arylphorin as precursor reservoirs for reproduction and metamorphosis in female Luna moths. Arch. Insect Biochem. Physiol. 32: 149-162. Patterson, K.D. 1992. Yellow fever epidemics and mortality in the United States, 16931905. Soc. Sci. Med. 34(8): 855-865. Perfect, T. J. 1992. IPM in 2000, pp.47-53. IN A.A.S A. Kadir and H.S. Barlow (eds.), Pest Management and the Environment in 2000. CAB International, Oxford, UK. 401pp. Quick, M. and B.R. Stevens. 2001. Amino acid transporter CAATCH1 is also an amino acid-gated cation channel. J. Bio. Chem. 276(36): 33143-33418. Racioppi, J.V. and D.L. Dahhnan. 1980. Effects of L-canavanine on Manduca sexta (Sphingidae: Lepidoptera) larval hemolymph solutes. Comp. Biochem. Physiol 67: 35-39. Ragsdale, D. and E.B. Radcliffe. 1999. Colorado potato beetle management. University of Minnesota Cooperative Extension Service. Internet URL: http://ipmworld.umn. edu/arjhidalert/CPB~DWR.html. Accessed April 2004. Robertson, J.L. and H.K. Priesler. 1991. Pesticide bioassays with arthropods. CRC Press, Inc. Boca Raton, 127pp. Rock, G.C. 1971. Utilization of D-isomers of the dietary, mdispensable armno acids by Argyrotaenia velutinana larvae. J. Insect Physiol. 17: 2157-2168. Rock, G.C. and E. Hodgson. 1 971 Dietary amino requirements for Heliothis zea aetermined by dietary deletion and radiometric techniques. J. Insect Phvsiol 171087-1097. Rock, G.C, BG.Ligon, and E. Hogson. 1973. Utilization of methionine analoges by Argyrotaenia velutinana larvae. Ann. Entol. Soc. Amer. 66(1): 177-179.

PAGE 123

Ill Rock, G.C., A. Khan, and E Hodgson. 1975. The nutritional value of seven D-amino acids and a-keto acids for Argyrotaenia velutinana, He Hot his zea and Phormia regina. J. Insect Physiol. 21:693-703. Romoser, W.S. and J. G Stoffolano, Jr. 1998. The Science of Entomology, 4* edition. McGraw-Hill. Newyork, 605pp. Rosen, D., F.D. Bennett, and J.L Capinera. 1996. Prefece, pp. V-vi. IN D. Rosen, F.D. Bennett and J.L. Capinaera (eds.), Pest Management in the Subtropics: Biological Controla Florida Perspective. Intercept Limited, Andover, UK. 737pp. Rosenthal, G. A. 1977. The biological effects and mode of action of L-canavanine, a structural analogue of L-arginine. Q. Rev. Biol. 52(2): 155-178. Rosenthal, G.A. and D.L. Dahlman. 1975. Non-protein amino acid-insect interactions, n. Effects of canaline-urea cycle amino acids growth and development of the tobacco hornworm, Manduca sexta (L.) (Sphingidae). Comp. Biochem. Physiol 52: 105-108. Rosenthal, G.A. and D.L. Dahlman. 1988. Degradation of aberrant proteins by larval tobacco hornworm, Manduca sexta (L) (Sphingidae). Arch. Insect Biochem Physiol. 8: 165-172. Rosenthal, G.A.andD.L. Dahlman. 1991. Incorporation of L-canavanine into proteins and the expression of its antimetabolic effects. J. Ag. and Food Chem 39987990. Rosenthal, G.A., D.L. Dahlman, P.A. Crooks. S.N. Phuket, and L.S. Trifonov. 1995. Insecticidal properties of some derivatives of L-canavanie. J. Agric. Food Chem 43: 2728-2734. Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1 976. A novel means for dealing with L-canavanine, a toxic metabolite. Science 192: 256-258. Rosenthal, G.A..D.L. Dahlman and D.H. Janzen. 1977. Degradation and detoxification of canavanine by a specialized seed predator. Science 196: 658-660. Rosenthal, G.A., D.L. Dahlman and D.H. Janzen. 1978. L-canaline detoxification: A seed predator's biochemical defense. Science 202: 528-529. Rosenthal, G.A., P. Nkomo and D.L. Dahlman. 1998. Effect of long-chained esters on the insecticidal properties of L-canavanine. J. Agric. Food Chem. 46(1): 296-299. Royer, T.A., K.L. Giles, S.D. Kindler and N.C. Elliott. 2001. Developmental response of three geographic isolates of Lysiphlebus testaceipes (Hymenoptera: Aphididae) to temperature. Environ. Entomol. 30(4): 637-641.

PAGE 124

112 Schoonhoven, L.M. 1972. Plant recognition by lepidopterous larvae. Insect/Plant Relationships, Symposium of the Roy. Entomol. Soc. London 6:83-93. Schardt, J.D. 1987. 1987 Florida Aquatic Flora Survey Report. Florida. Department of Natural Resources, Bureau of Aquatic Plant Management. Tallahassee, FL. 49 pp. Schuster, D.J, J.E. Funderburk and P.A. Slansly. 1996. IPM in tomatoes, pp. 387-408. IN D. Rosen, F.D. Bennett and J.L. Capinaera (eds.), Pest Management in the Subtropics: Integrated Pest ManagementA Florida Perspective. Intercept Limited, Andover, UK. 737pp. Singh, K.RP. and A.W.A. Brown. 1957. Nutritional requirements of Aedes aegypti. J. Insect Physiol. 1(1): 199-220. Sorenson, C.E., RL. Fery and G.G.Kennedy. 1989. Relationship between Colorado potato beetle (Coleoptera: Chrysomelidae) amd tobacco hornworm (Lepidoptera:: Sphingidae) resistance in Lycopersicon hirsutism f. glagratum. J. Econ. Entomol 82(4): 1743-1748. Stansly, P.A., IX. Liu, D.J. Schuster and D.E. Dean. 19%. Role of biorational insecticides in management of Bemisia, pp. 605-615. IND. Gerling and R T. Mayer, Jr. (eds.) Bemisia: 1995 Taxonomy, Biology, Damage Control and Management. Andover, Hants, UKD. 702pp. Stevens, B.R., D.H. Feldman, Z. Liu and W.R Harvey. 2002. Conserved tyrosine147 plays a critical role in the ligand-gated current of the epithelial cation/anion acid transporter/channel CAATCH1. J. Exp. Biol. 205: 2545-2553. Sugarman, D. and W. Jakinovitch, Jr. 1986. Behavioral gustatory responses of adult cockroaches, Periplaneta americana to D and L amino acids. J Insect Phvsiol 32(1): 35-41. Tobe,S.S. and N.Clarke. 1985. The effect of L-methionine concentration on juvenile hormone biosynthesis by corpora allata of the cockroach Diploptera punctata Insect Biochem. 1 5(2): 1 751 79. Tipping, P.W., C.A. Holko, A.A. Abdul-Baki, and J.R Aldrich. 1999. Evaluating Edovum puttier i Grissell and Podiscus maculfventris (Say) for augmentative biological control of Colorado potato beetle in tomatoes. Biol. Control 16: 35-42. Tzeng,D.D. 1988. Photodynamic action of methionine-riboflavin mixture and its application in the control of plant diseases and other plant pests. Plant Protection Bulletin 30(2): 87-100. Tzeng,DD.andJ.E.Devay. 1989. Biocidal activity of mixtures of riboflavin against tlQ^SSSu 311(1 baCteria 311(1 P SSible m deS f aCti n MycoI 8 ia

PAGE 125

113 Tzeng,D.D., M.H. Lee, K.R. Chung and J.E.Devay. 1990. Products in light-mediated reactions of free methionine-riboflavin mixtures that are biocidal to microorganisms. Can. J. Microbiol. 36(7): 500-506. Wadsworth, D.J. 1995. Animal health products, pp. 257-284. Z7VC.R.A. Godfrey (ed.) Agrochemicals From Natural Products, Marcel Dekker, New York. 424pp. Walker, T.J., J J. Gaffhey, A.W. Kidder and A.B. ZifTer. 1993. Florida Reach-Ins: Environmental chambers for entomological research. American Entomologist 39(3): 177-182. Weinzierl, R., T. Henn and P.G. Koehler. 1998. Microbial insecticides. ENY-275 Cooperative Extension Service, Institute of Food and Agricultural Services University of Florida. 13pp. Wheeler, D.E., L Tuchinskaya, N.A. Buck and B.E. Tabahnik. 2000. Hexameric storage proteins during metamorphosis and egg production in the diamondback moth, Plutellaxylostella (Lepidoptera). J. Insect. Physiol. 46: 951-958. Womack, M. 1 993. The yellow fever mosquito, Aedes aegypti. Wing Beats 5(4): 4. Wright, R. 1995. Know Your Friends: Wasp Parasites of Greenbugs. Midwest Biological Control News Online, 11:9. Young, V.R. and A.E. El-Khoury. 1996. Human amino acid requirementsA reevaluation. Food and Nutrition Bulletin 17(3): 191-203. Zar,J.H. 1999. Biostansrical Analysis, 4th ed. Prentice Hall. New Jersey. 663pp. Zeh, M., A.P. Casazza, G Kreft U. Roessner, K. Bieberich, L. Willmitzer, R. Hoefgen and H. Hesse. 2001. Antisense inhibition of tlireonine synthase leads to high methionine content in transgenic potato plants. Plant Physiol. 127792-802

PAGE 126

BIOGRAPHICAL SKETCH Lewis Scotty Long was born in Calhoun, Georgia on August 20, 1971 He graduated from Madisonville High School (Madisonville, Tennessee) in May 1989. On a biology scholarship, Lewis attended Middle Tennessee State University (MTSU), where he earned his BS in May 1994. On graduation, he took a job as an aquatic biologist for Aquatic Resources Center (Franklin, Tennessee). Lewis worked there specializing in taxonomy of mayflies, stoneflies, caddisflies, and freshwater molluscs (snails and mussels). While still employed at Aquatic Resources Center, he started his graduate studies in 1996 at MTSU and continued the work he had started during his undergraduate years. In May of 1 999, Lewis graduated with his MS. After receiving his MS, Lewis moved to Florida and entered the PhD program at the University of Florida, Department of Entomology and Nematology. He worked with Dr. Bill Peters (Florida A&M University) on the worldwide taxonomic revision of an understudied group of mayflies. However, Dr. Peters unexpectedly passed away in 2000, and Lewis took this unfortunate event as a chance to broaden his expertise in entomology. In 2000, he took a part-time job with Drs. James Cuda and Bruce Stevens on research that was in the patent process. This was the research that Lewis undertook for his dissertation. Lewis also served as a teaching assistant for the department for classes such as Bugs and People, Life Sciences for Education Majors, Principles of Entomology, and Medical and Veterinary Entomology. He served as primary instructor for Insect Classification and Immature 114

PAGE 127

Insects. Lewis, along with fellow graduate student Jim Dunford, were awarded the Outstanding Teacher Award by the Entomology and Nematology Student Organization of the University of Florida for outstanding teaching accomplishments in the department. While at the University of Florida, Lewis joined the U.S. Army Reserve as a medical entomologist. He was assigned to the local Medical Detachments, and served there from 2000 to 2004. Originally he had planned on graduating in 2003, but was called to active duty with the 1469 th Medical Detachment as a part of Operation Enduring Freedom (OEF). Lewis was the OEF Theater entomologist, and served as the Executive Officer (responsible for the deployment of personnel and equipment to South West Asia). He was stationed at Kandahar Airfield, where he performed his duty and was awarded an Army Commendation Medal for his work in protecting soldiers from health hazards and diseases associated with the area. Lewis returned and continued his work toward graduation. Lewis was married in August 1992 to Karen Abbott, and is the father of Emilia Irene (1994) and Bryan Scott (1997). Lewis plans on having a career in the military as a medical entomologist, and all look forward to seeing the world and the rest of their future.

PAGE 128

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jamas P. Cuda, Chair Assisrant Professor of Entomology and Nematology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. j Bruce R. Stevens, Cochair Professor Physiology and Functional Genomics I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. George ASJ§erencser Professor of Physiology and Functional Genomics I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jarnies E. Maruniak Associate Professor of Entomology and Nematology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Simon S.J. Yu Professor of Entomology and Nematology

PAGE 129

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Susan E. Webb Associate Professor of Entomology and Nematology This dissertation was submitted to the Graduate Faculty of the College of Agricultural and Life Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophv. May 2004 Dean, College of Agricultural Sciences Dean, Graduate School


50
combination of the early consumption of the treated disks and mortality occurring after
48 hours, when a lower concentration is required for mortality. The larvae could have
fed on the treated disks and then switched to the Control based on a physiological cue.
Mitchell (1974) and Mitchell and Schoonhoven (1974) examined the taste receptors of
CPB and found physiological and behavioral responses to some amino acids, mainly
gamma aminobutyric acid (GABA) and alanine. They discussed the possibility that host
selection in solanaceous plants may have been the result of these chemosensory structures
and the concentration of amino acids in the leaves. It should be noted that both studies
excluded methionine and no electrophysiological data were collected on the response of
CPB to this amino acid. This is not surprising considering the fact that the diet of the
CPB is low in methionine and therefore would not be a candidate for the inclusion in
feeding stimulatory studies (Cibula et al. 1967). It is unknown if these sensory structures
can detect methionine and possibly act as a means to avoid plant material high in this
amino acid. This appears to be contradicted by the data in Figure 4-5, in which there was
no difference between the treatments. The larvae feeding on the Control treatment,
consuming the majority and then moving to the 1.0% L-methionine treatment, could
explain the lack of difference.
There are some differences between some of the Feeding and Development
treatments should be noted. The mean head capsule of the larvae in the 0.5%
L-methionine treatment was higher than the 0.3% L-methionine treatment while the
amount of leaf material consumed for the same treatment were the same indicating
another factor involved with the greater head capsule width. The differences could be the
result of the larger size of females and possibly could have included more females.


12
concentration. This point is defined as saturation. When a sample is diluted below the
odor-detection threshold, there will be no sensory response. Stevens law suggests that
two different compounds (A and B) at the same concentration, with similar detection
thresholds but with different exponents (n values), will produce different dose-odor
intensity profiles (Fig. 2-5). Individual odors will contribute differently to the overall
food aroma intensity, depending on their concentration and n value (61).
Figure 2-5. Stevens law, comparing two difference compounds: A= compound A, B=
compound B.
The OSME is a time intensity procedure that determines the intensity of the
perceived odor without dilution. In this method, the trained assessors sniff the effluents
from GC mixed with humidified air, and directly record the odor intensity and duration of
each odor active component while describing its odor quality. Intensities of individual
components are plotted versus elution time; and the resultant graph is known as an
aromagram.


Mean Number of Parasitized
Mummies/Plant
93
(Error bars @ 95%; 1=4.41: F=3.25; P =0.005)
16
10/22/02 10/29/02 11/5/02 11/12/02 11/19/02 11/26/02
Figure 7-6. Lysephlebius testiceipes parasitized aphids on cotton plants treated with
L-methionine. Ten plants were used for each treatment and held in the
shade house at the University of Florida, Department of Entomology and
Nematology from 22 October to 25 November 2002. No statistical
differences were observed except for the second and final collection date.


24
larval head capsule measurements made using the same procedures described in the
Feeding and Development section.
Data Analysis
Sample sizes of all experiments were chosen according to the guidelines
recommended by Robertson and Preisler (1991) for optimal sample size and data
analysis. Probit analysis and determination of mean Lethal Concentration (LC50) were
performed using PROBIT Version 1.5 (Ecological Monitoring Research Division,
USEPA) after Abbotts correction for control mortality (Abbott 1925). Survival data
were normalized to the previous value when control mortality was greater than the
treatment mortality, to produce a smoother trend line. Statistical analysis was performed
on the data using Minitab Version 14 (Minitab, Inc.; State College, PA). Analysis of the
data included One-way ANOVA and separation of significant means using Tukeys
Multiple Comparison and Pearson Correlation was performed on the choice trial data to
examine possible relationships between development and consumption of treated leaf
material (Zar 1999). Regression analysis using lest squares were performed on the leaf
weights before and after the L-methionine treatment for the equation used to convert %
concentration to mg/g plant material (Figure 3-4).
Results
Diets and Survivorship
The artificial diet resulted in 100% mortality of THW larvae for the 3.0%
L-methionine to 10.0% L-methionine treatment after only one day of exposure (Figure
3-5). Approximately 80% mortality was observed in the 1.0% L-methionine treatment
after 4 days, and 50% mortality for both the 0.3% L-methionine and 0.5% L-methionine


83
98 Schiedt, K.; Liaaen-Jensen, S. Isolation and Analysis. In Carotenoids-, G. Britton;
S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel, Boston, Berlin,
1995; pp 81-108.
99 Ravichandran, R. Carotenoid composition, distribution and degradation to flavor
volatiles during black tea manufacture and the effect of carotenoid
supplementation on tea quality and aroma. Food Chem. 2002, 78, 23-28.
100 Stevens, M. A. Relationship Between Polyene-Carotene Content and Volatile
Compound Composition of tomatoes. J. Amer. Soc. Hort. Sci. 1970, 95, 461-464.
101 Mord, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Ingold, K. U.; Lindsay,
D. A. Exploratory study of (3-carotene autoxidation. Tetrahedron Lett. 1991, 32,
4203-4206.
102 de Heij, H. T.; van Dort, H. M. Autoxidation of carotenes. Prog. Flavour
Precursor Stud. Proc. Int. Conf. 1993, 409-413.


4-4 Upper spectra from orange juice MS at RT = 17.68 bottom spectra of (3-
cyclocitral from database NIST 2002 40
4-5 Upper spectra from orange juice MS at RT = 21.94, bottom spectra from
standard (3-ionone using identical ion trap MS at identical retention time 40
4-6 Upper spectra from orange juice MS at RT = 20.87, bottom spectra from
standard a-ionone using identical ion trap MS at identical retention time 41
5-1 Exposure time between SPME fiber and the headspace of orange juice spiked
with standards at 40C, = (3-cyclocitral, = p-damascenone, A= a-ionone,
= p-ionone 48
5-2 Standard addition data for [3-cyclocitral peak area vs. added concentration in fresh
orange juice. Regression line calculated from peak area at selected mass 137 50
5-3 Standard addition data for a-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
a-ionone concentration 51
5-4 Standard addition data for (3-ionone peak area vs. added concentration in fresh
orange juice. The regression line created by peak area at selected mass 177 vs.
(3-ionone concentration 52
5-5 Standard addition (3-damascenone peak area vs. added concentration in fresh
orange juice. GC-quadrupole mass spectrometer in SIM mode at m/z 190 52
5-6 Standard addition data of (3-damascenone peak area vs. added concentration in
reconstituted from concentrate orange juice 53
5-7 Aroma group profiles of fresh (), pasteurized (), and reconstituted from
concentrate () orange juice 58
5-8 Upper bar norisoprenoids contribute mainly to the total floral category, fresh =
78%, pasteurized = 78%, and reconstituted = 59%, lower bar represent non-
norisoprenoids including linalool and unknown (LRI = 1255) 60
6-1 The standard (3-carotene (99% purity) as received (no purification) 65
6-2 Headspace volatiles from P-carotene in model solution pH 3.5 at 0 day 66
6-3 Headspace volatiles from P-carotene in model solution pH 3.5 after storage 1 day
at 35C: 1 = P-ionone, a = sweet/raspberry 66
6-4 Degradation of P-carotene in model solution at difference carbon bonds 67
x


77
24 Ayers, J. E.; Fishwick, M. J.; Land, D. G.; Swain, T. Off-flavor of dehydrated
carrot stored in oxygen. Nature 1964, 203, 81-82.
25 Vemin, G. Terpenoids and norisoprenoids in grape berry aromas and in wines. 1.
Origins and formation. Rivista Italiana EPPOS 1997, 8, 19-31.
26 Gross, J. Pigments in Fruits', Academic Press Inc.: London, 1987; 303 pp.
27 Weedon, B. C. L.; Moss, G. P. Structure and nomenclature. In Carotenoids', G.
Britton; S. Liaaen-Jensen and H. Pfander, Eds.; Birkhauser Verlag: Basel Boston
Berlin, 1995; pp 27-70.
28 Mordi, R. C. Mechanism of (3-carotene degradation. Biochem. J. 1993, 292, 310-
312.
29 Lutz-Roder, A.; Jezussek, M.; Winterhalter, P. Nickel peroxide induced oxidation
of canthaxanthin. J. Agrie. Food Chem. 1999, 47, 1887-1891.
30 Ouyang, J. M.; Daun, H.; Chang, S. S.; Ho, C. T. Formation of carbonyl
compounds from (3-carotene during palm oil deodorization. J. Food Sci. 1980, 45,
1214-1217.
31 Kanasawud, P.; Crouzet, J. C. Mechanism of formation of volatile compounds by
thermal degradation of carotenoids in aqueous medium. 1. (3-Carotene
degradation. J. Agrie. Food Chem. 1990, 38, 237-243.
32 Winterhalter, P. Carotenoid-Derived Aroma Compounds: Biogenetic and
Biotechnological Aspects. In Biotechnology for Improved Foods and Flavors; G.
R. Takeoka; R. Teranishi; P. J. Williams and A. Kobayashi, Eds.; American
Chemical Society: Washington, DC, 1996; pp 295-308.
33 Weeks, W. W. Carotenoids: A Source of Flavor and Aroma. In Biogeneration of
Aromas; H. P. Thomas and C. Rodney, Eds.; American Chemical Society:
Washington, D.C., 1986; pp 157-166.
34 Winterhalter, P.; Rouseff, R. Carotenoid-Derived Aroma Compounds: an
Introduction. In Carotenoid-Derived Aroma Compounds; P. Winterhalter and R.
Rouseff, Eds.; American Chemical Society: Washington, DC, 2002; pp 1-17.
35 Knapp, H.; Straubinger, M.; Stingl, C.; Winterhalter, P. Analysis of
Norisoprenoid Aroma Precursors. In Carotenoid-Derived Aroma Compounds; P.
Winterhalter and R. L. Rouseff, Eds.; American Chemical Society: Washington,
DC, 2002; pp 20-35.
36 Baumes, R.; Wirth, J.; Bureau, S.; Gunata, Y.; Razungles, A. Biogeneration of
C13-norisoprenoid compounds: experiments supportive for an apo-carotenoid
pathway in grapevines. Anal. Chim. Acta 2002, 458, 3-14.


17
Objectives
The objective of this study was to confirm the presence of specific carotenoids in
Valencia orange juice which could serve as norisoprenoid precursors. The specific
carotenoids of interest include: a-cryptoxanthin, (3-cryptoxanthin, a-carotene, (3-carotene
and neoxanthin because they possess the structural features needed to serve as precursors
to the newly identified norisoprenoids. (See Objective #1)
Materials and Methods
Carotenoid Extraction
The carotenoid extraction method according to Lee et al. (75) was carried out with
slight modification. A 25 mL aliquot of Valencia juice was extracted with 50 mL of a
mixed solvent (hexane:acetone:ethanol, 50:25:25) using a Omni mixer homogenizer
(model no. 700, Lourdes, Vemitron Medical Products, Inc. Carlstadt, NJ). It was
extracted for 5 min at medium speed in ice bath, and centrifuged (CR412, Jouan, Inc.,
Winchester, VA) for 10 min at 4000 rpm and 10C. The top layer of hexane containing
pigments was collected and concentrated to dryness in rotary evaporator.
Carotenoid Saponification
Saponification was carried out according to Noga and Lenz (77) with slight
modification. The dried pigment was redissolved with 2 mL of methyl tert.-butyl ether
(MTBE), and placed in a 40 mL vial. Two mL of 10% methanolic potassium hydroxide
(KOH) was added to the sample and the headspace was blanketed with nitrogen before
closing. The sample was wrapped with aluminum foil to protect it from light, and placed
at room temperature for 1 hour. The sample was then transferred to separatory funnel to
which 5 mL of water was added and 2 mL of 0.1% butylated hydroxyl toluene (BHT) in
MTBE, and the aqueous layer removed. Additional water rinses were carried out until


74
Data Analysis
Data from the fruit and the CPB experiments were analyzed with ANOVA using
Minitab Version 12. Survivorship of CPB was corrected using Abbotts formula (Abbott
1925) to account for control mortality, mean separation was performed using Tukeys
multiple comparison procedure (Zar 1999). Data for both the eggplant weight mean per
treatment and also mean number of eggplants per treatment were analyzed using paired
t-test.
Results
Effects of L-methionine and Silwett L-77 on CPB Adults Under Laboratory Conditions
Little mortality was observed with the adult CPB at the 1.0% L-methionine
concentration (Figure 6-3). The 0.5% L-methionine concentration had the highest
mortality of all the treatments at approximately 20% with the other treatments showing
no adverse effects after correction for control mortality.
Effects of L-methionine and Silwett L-77 on yield
In total, 735 eggplants were collected during the course of this study from 09 June
to 31 August 2001. Mean weight and yield of eggplants between the treatments were not
statistically different from each other (Figures 6-4). Control plants produced 195 fruits
with a mean weight of 276.9 grams, followed by the 0.1% treatment with 191 fruits at
281.2 grams. The 0.5% and 1.0% treatments yielded 175 and 174 fruits with mean
weights of 295.7 grams and 283.6 grams, respectively.
Survival of CPB larvae
No statistical difference in survivorship of CPB larvae was observed between the
three treatments for the first day after exposure (Figure 6-5) but treatment differences


38
2004; Nester et al. 2002). Resistance in insects involves a variety of mechanisms and
many are the result of a combination of different pesticide classes. The CAATCH1
system is one that could be used in cases where the only alternative is by adding more
pesticides or at higher rates to break resistance. Further research is needed to determine
compatibility of the different Bt insecticides and L-methionine with each other for cases
in which Bt resistance is observed in natural populations. Given the safety of
L-methionine and the shorter time required for 100% mortality (when compared to Btk
results of this study), this compound could represent a viable alternative for pesticides
currently used in the management of the THW.


EVALUATION OF THE AMINO ACID METHIONINE FOR BIORATIONAL
CONTROL OF SELECTED INSECT PESTS OF ECONOMIC AND MEDICAL
IMPORTANCE
By
LEWIS SCOTTY LONG
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2004


6
animals, and eventual toxicity to nontarget organisms. Because of these problems,
alternatives are needed to prevent another crisis like the one from which IPM originally
arose.


11
12
13
14
15
16
17
18
19
20
21
22
23
76
MacLeod, G.; Ames, J. M. Volatile components of starfruit. Phytochemistry 1990,
29, 165-172.
Olle, D.; Baron, A.; Lozano, Y. F.; Sznaper, C.; Baumes, R.; Bayonove, C.;
Brillouet, J. M. Microfiltration and reverse osmosis affect recovery of mango
puree flavor compounds. J. Food Sci. 1997, <52, 1116-1119.
Ishida, B. K.; Mahoney, N. E.; Ling, L. C. Increased lycopene and flavor volatile
production in tomato calyces and fruit cultured in vitro and the effect of 2-(4-
Chlorophenylthio)triethylamine. J. Agrie. Food Chem. 1998, 46, 4577-4582.
Tarantilis, P. A.; Polissiou, M. G. Isolation and identification of the aroma
components from saffron (Crocus sativus). J. Agrie. Food Chem. 1997, 45, 459-
462.
Enzell, C. Biodegradation of carotenoids an important route to aroma
compounds. Pure Appl. Chem. 1985, 57, 693-700.
Sanderson, G. W.; Co, H.; Gonzalez, J. G. Biochemistry of tea fermentation: role
of carotenes in black tea aroma formation. J. Food Sci. 1971, 36, 231-236.
Gross, J.; Gabai, M.; Lifshitz, A. Comparative study of the carotenoid pigments in
juice of Shamouti, Valencia, and Washington oranges, three varieties of Citrus
sinensis. Phytochemistry 1972, 11, 303-308.
Isoe, S.; Hyeon, S. B.; Sakan, T. Photo-oxygenation of carotenoids. I. Formation
of dihydroactinidiolide and (3-ionone from P-carotene. Tetrahedron Lett. 1969,
279-281.
Demole, E.; Enggist, P.; Winter, M.; Furrer, A.; Schulte-Elte, K. H.; Egger, B.;
Ohloff, G. Megastigma-5,8-dien-4-one, an aroma constituent of the yellow
passion fruit and Virginia tobacco. Helv. Chim. Acta 1979, 62, 67-75.
Strauss, C. R.; Wilson, B.; Anderson, R.; Williams, P. J. Development of
precursors of C13 nor-isoprenoid flavorants in Riesling grapes. Am. J. Enol. Vitic.
1987, 38, 23-27.
Williams, P. J.; Sefton, M. A.; Francis, I. L. Glycosidic Precursors of Varietal
Grape and Wine Flavor. In Flavor Precursors: Thermal and Enzymatic
Conversions', R. Teranishi; G. R. Takeoka and M. Guntert, Eds.; American
Chemical Society: Washington, DC, 1992; pp 74-86.
Enzell, C. R. Influence of Curing on the Formation of Tobacco Flavor. In Flavour
'81; P. Schreier, Ed.; de Gruyter: Berlin, New York, 1981; pp 449-478.
Buttery, R. G.; Teranishi, R.; Flath, R. A.; Ling, L. C. Identification of additional
tomato paste volatiles. J. Agrie. Food Chem. 1990, 38, 792-795.


6-5 Headspace volatiles from (3-carotene in model solution pH 3.5, after storage 2
weeks at 35C 69
6-6 Selected ion chromatogram (SIC) of model solution headspace volatiles after
storage 2 weeks at 35C 70
6-7 Upper spectra from model solution MS at RT 19.61, bottom spectra from
standard (3-cyclocitral using identical ion trap MS at identical retention time 70
6-8 Upper spectra from model solution MS at RT 20.46, bottom spectra from
standard P-homocyclocitral using identical ion trap MS at identical retention
time 71
6-9 Upper spectra from model solution MS at RT 25.13, bottom spectra from standard
P-ionone using identical ion trap MS at identical retention time 71
xi


99
target ingests the methionine first as it feeds ensuring the overload CAATCH1 system
and eventually death.
As for the stored methionine, it is released from the storage proteins as needed to
synthesize juvenile hormone and allow for transformation in addition to other functions.
The remaining methionine is then used for protein synthesis in the tissues around the
ovaries to boost yolk production, as seen in the transfer of methionine from male to
female Drosophila species (Bownes and Partridge 1987). In the THW, the presence of
hexamerins with high methionine content may be an alternative to the male contribution
possibly found in its ejaculate. Methionine-rich hexamerins are common in Lepidoptera
and have been shown to provide the larvae a source of amino acids during the synthesis
of these proteins during the last stage of larval development (Wheeler et al. 2000). In
addition to the need for methionine for metabolism and reproduction, the release of
methionine may also in part account for the decrease in ion transport of the posterior
region of the midgut during larval molts and the wandering stage present before pupation.
Currently, little is known regarding the mechanisms involved with the decrease of ion
transport during these developmental stages (Lee et al. 1998). Clearly there appears to be
more to the role methionine plays in the development of some insects other than the
vague designation of essential amino acid.
Insects have evolved to deal with limiting resources, such as methionine, and have
successfully found effective strategies like hexamerin storage or alternate pathways to
deal with such problems. No attempt to link together all the aspects of the role of
methionine in a whole organism or system context. It appears that methionine actually
may play a role far more important than that of just an essential amino acid. From the


108
Ito, T. and T. Inokuchi. 1981. Nutritive effects of D-amino acids on the silkworm,
Bombyx mor. J. Insect Physiol. 27(7): 447-453.
Jaffe, J.J. and L.R. Chrin. 1979. De novo synthesis of methionine in normal and Brugia-
infected Aedes aegypti. J. Parasitol. 65(4): 550-554.
Jones, D.C and R.M. MacPherson. 1997. Tobacco Insects: Summary of losses from
insect damage and costs of control in Georgia -1997. University of Georgia,
Waroell School of Forest Resources and College of Agricultural and
Environmental Sciences Internet URL: http://www.bugwood.org/tobbaco97.htm.
Accessed April 2004.
Kaldy, M.S. and A.M. Harper. 1979. Nutrient constituents of a grain aphid,
Metopolophium dirhodum (Homoptera: Aphididae), and its host, oats (Avena
sativa). Canadian Entomol. 111(7): 787-790.
Kammer, A.E., D. L. Dahlman and G.A. Rosenthal. 1978. Effects of the non-protein
aminoacids L-canavanine and L-canaline on the nervous system of the moth
Manduca sexta (L). J. Exp. Biol. 75: 123-132.
Kasting, R.. G.R.F. Davis amd A.J. McGinnis. 1962. Nutritionally essential and non-
essential amino acids for the prairie grain wireworm, Ctenicera desctructor
Brown, determined with Glucose-U-C. J. Insect Physiol. 8: 589-596.
Knutson, A., Boring IE, E.P., Michaels, Jr., G.J., and Gilstrap, F. 1993. Biological
Control of Insect Pests in Wheat Texas Agrie. Ext. Service Publ. B-5044 8pp.
Koo, S.I., T.A. Currin, M.G. Johnson, E.W. King and D.E. Turk. 1980. The nutritional
value and microbial content of dried face fly pupae (Musca autumnal is (DeGeer))
when fed to chicks. Poultry Sci. 59: 2514-2518.
Koyama, K. 1985. Nutritional physiology of the brown rice planthopper Nilaparvata
lugens StAl (Hemiptera: Delphacidae). II. Essential amino acids for nymphal
development Appl. Ent Zool. 20(4): 424-430.
Koyama, K. and J. Mitsuhashi. 1975. Essential amino acids for the growth of the smaller
brown planthopper, Laodelphax striatellus FallSn (Hemiptera: Delphacidae).
Appl. Ent Zool. 10(3): 208-215.
Lee, K., F.M. Horodyski, and M.E. Chamberlin. 1998. Inhibition of midgut ion transport
by allatotropin (Mas-AT) and Manduca FLRFamides in the tobacco homworm
Manduca sexta. J. Exp. Biol. 201:3067-3074.
Leonardi, M.G., M. Casartelli, L. Fiandra, P. Parenti and B. Giordana. 2001. Role of
specific activators of intestinal amino acid transport in Bombyx mor larval growth
and nutrition. Arch. Insect Biochem. Physiol. 48: 190-198.


Barrier Barrier
Rows Rows
Figure 6-1. Overview of the design layout used to study the effects of
L-methionine and Silwett L-77 solutions on yield of
eggplant. Rows were four feet apart with individual
plants three feet apart and treatments nine feet apart.
Each letter represents a group of ten eggplants.


LIST OF TABLES
Table
3-1 HPLC retention times, spectral characteristics of orange juice carotenoids 24
4-1 Identification, retention characteristics and aroma descriptions of aroma active
compounds in fresh orange juice 36
5-1 Reproducibility of SPME exposure time 45 min at 40C 49
5-2 Concentration of norisoprenoids in fresh orange juice as determinded by
standard addition technique 53
5-3 Concentration of (3-damascenone in fresh, pasteurized and reconstituted
concentrate 53
5-4 Aroma active compounds in orange juice grouped by citrusy/minty 55
5-5 Aroma active compounds in orange juice grouped by metallic/mushroom/
geranium 56
5-6 Aroma active compounds in orange juice grouped by roasted/cooked/meaty/
spice 56
5-7 Aroma active compounds in orange juice grouped by fatty/soapy/green 56
5-8 Aroma active compounds in orange juice grouped by sulfury/solventy/medicine ..57
5-9 Aroma active compounds in orange juice grouped by floral 57
5-10 Aroma active compounds in orange juice grouped by sweet/fruity 57
5-11 Aroma active compounds in orange juice grouped by green/grassy 58
5-12 Norisoprenoids in orange juice and peel oil 60
6-1 Aroma active compounds from (3-carotene thermal degradation in model solution
pH 3.8, storage at 35C for 2 weeks 69
viii


38
a>
u
c
9
"O
c
a
A
Â¥
a
a
P
Time (min)
Figure 4-3. Upper, total ion current chromatogram from orange juice headspace, other
chromatograms using SIM at m/z 190. Middle chromatogram P-damascenone
detected from orange juice, and lower overlay chromatogram of spiked (A)
and non-spiked (B) of orange juice with standard P-damascenone.
The P-damascenone, selected ion chromatograms using m/z = 175 and 190 (two
masses highly characteristic for P-damascenone) did not provide a clear signal at the
expected retention time of P-damascenone using the ion trap MS. Beta-damascenone had
been detected by GC-0 at the expected retention time with the characteristic aroma but
not detected by either FID or SIC ion-trap MS, suggesting that P-damascenone, if
present, was there at very low levels. Beta-damascenone has an extremely low odor
threshold, which is below the detection limits of most instrumental detectors (0.002
Hg/L). However, by employing quadrupole mass spectrometer in the single ion


18
free of alkali. The MTBE layer was then filtered through a small glass column filled with
deactivated glass wool (Restek Corporation, PA) and anhydrous sodium sulfate (Fisher
Scientific, NJ) to remove residue water from MTBE layer. Each sample was
concentrated by evaporation with nitrogen, and the volume adjusted with 0.1% BHT in
MTBE to 1 mL and placed in sealed amber vials under refrigeration (4C) until analyzed.
HPLC Procedure
Carotenoid pigments were analyzed according to Rouseff et al.(<58) by reverse
phase HPLC using ternary gradient of water, methanol, and MTBE with photo diode
array detection (PDA] by reverse phase HPLC using ternary gradient of water, methanol
(MeOH), and MTBE with photodiode array detection (PDA). The 4.6 mm i.d. x 250 mm
YMC Carotenoid 5 pm column (YMC, Inc., Waters Corporation, MA) was used. The
chromatographic system consisted of autosampler, LC pump, and PDA detector
(Surveyor, ThermoFinnigan, CA). The PDA was set to scan from 280 to 550 nm. Three
separate data channel were set to record the absorbances at 350, 430, and 486 nm with
spectral bandwidths of 1 nm. Data were collected, stored, and integrated, using the Atlas
software (Atlas 2003, Thermo Electron Corporation, Cheshire, UK). All reagents used
were HPLC grade (Fisher Scientific, NJ). One standard, (3-carotene, was purchased from
Acros (Acros, NJ). The initial ternary gradient composition consisted of 90% MeOH, 5%
water, and 5% MTBE. The solvent composition changed in a linear fashion to 95%
MeOH and 5% MTBE at 12 min. After the next 8 min (at 20 min) the solvent
composition was 86% MeOH and 14% MTBE. At this composition the solvent
composition was gradually changed to 75% MeOH and 25% MTBE at 30 min. The final
composition was 50% MeOH and 50% MTBE at 50 min. Intial conditions were


Head Capsule Width (mm
46
(Error Bars @ 95%; F(o.o5)7,3i2=1.14;F=576.71; P<0.001)
Figure 4-3. Mean head capsule widths of Colorado potato beetle larvae exposed to
excised eggplant leaves treated with various concentrations of L-
methionine (nTOtai=320). Proline (1.0%) and Bt were included for
comparison as positive and negative controls. Error bars denote 2 SE.
Bars within treatments having the same letter are not statistically
different (Tukeys MST, PO.OOl).


79
compared directly. However, the cost for the amount of L-canavanine required for large-
scale application far exceeds that of the largest amount of L-methionine needed.
Despite the lack of a statistical difference between treatments for both mean
weight and mean yield of eggplant, there were some interesting disparities within the
data. First, there was an observable difference in mean weight of the eggplants between
the treatments and the control. All eggplant weights were greater for the treatments than
the control, with the 0.5% L-methionine concentration treatment producing the highest
mean eggplant weight. It would appear that excess methionine decreases the number of
fruit produced, but those fewer eggplants weighed more. Further research is needed to
better understand the differences observed during this study.
The addition of Silwet L-77 did not appear to adversely affect survival of CPB
as seen in the preliminary tests on the adults and on the larvae during the field release
(Figures 6-3 and 6-5). The low adult mortality observed could be attributed to the ability
of this species to stop feeding and fly to a more suitable food source. Because the adults
were unable to move to an untreated leaf, they were observed sitting motionless on the
underside of the leaves. This was not observed in either of the controls as they were seen
actively feeding the majority of the time.
One aspect of this research that was not examined is that of fertility and fecundity
of adults exposed to excess amounts of L-methionine. Despite the fact that methionine is
used for egg production in many insect species, excess concentrations may act as a
deterrent to feeding causing the adults to stop feeding and to seek other food sources.
The lag time from the cessation in feeding to finding another food source may be long


107
Griffin, M.L. and K.V. Yeargan. 2002. Oviposition site selection by the spotted lady
beetle Coleomegilla maclala (Coleptera: Coccinellidae): Choices among plant
species. Environ. Entomol. 31(1): 107-111.
Groden, E., F.A. Drummond, R.A. Casagrande and D.H. Haynes. 1990. Coleomegilla
maculata (Coleptera: Coccinellidae): Its predation upon the Colorado potato
beetle (Coleptera: Chrysomelidae) and its incidence in potatoes and surrounding
crops. J. Econ. Entomol. 83:1306-1315.
Gubler, DJ. 1998. Resurgent vector-borne diseases as a global health problem. Emerg.
Infect. Dis. 4(3): 442-450.
Gubler, D.J. and G.G. Clark. 1995. Dengue/Dengue Hemorrhagic Fever: The emergence
of a global health problem. Emerg. Infect. Dis. 1(2): 55-57.
Haag, K.H. and D.G. Boucias. 1991. Infectivity of insect pathogens against Neochetina
eichhorniae, a biological control agent of water hyacinth. Florida Ent. 74(1): 128-
133.
Haag, K.H. and D.H. Habeck. 1991. Enhanced biological control of water hyacinth
following limited herbicide use. Aquat. Plant Management 29: 24-28.
Harrison, J. and R. Holliday. 1967. Senescence and the fidelity of protein synthesis in
Drosophila. Nature 214: 990-993.
Hazzard, R.V., D.N. Ferro, R.G. van Driesche and A.F. Tuttle. 1991. Mortality of eggs of
Colorado potato beetle (Coleptera: Chrysomelidae) from predation of
Coleomegilla maculata (Coleptera: Coccinellidae). Environ. Entomol. 20: 841
848.
Hegdekar, B.M. 1970. Amino acid analogues as inhibitors of insect reproduction. J.
Econ. Entomol. 63: 1950-1956.
Heim, D.C., G.G. Kennedy and J.W. Van Duyn. 1990. Survey of insecticide resistance
among North Carolina Colorado potato beetle (Coleptera: Chrysomelidae)
populations. J. Econ. Entomol. 83(4): 1229-1235.
Helena Chemicals, Inc. 2002. Silwett L-77. Technical Data Sheet SL77080596.
Helena Chemicals, Inc. Internet URL: http://www.helenachemicalwest.
com/data/TDS/Silwett.pdf. Accessed April 2004
Hilbeck, A. and G.G. Kennedy. 1996. Predators feeding on Colorado potato beetle
insecticide-free plots and insecticide-treated commercial potato fields in eastern
North Carolina. Biol. Control. 6: 273-282.
Hoffmann, M.P. and A.C. Frodsham. 1993. Natural Enemies of Vegetable Insect Pests.
Cooperative Extension, Cornell University, Ithaca, NY. 63pp.


104
Leptinotarsa decemlineata (Coleptera: Chrysomelidae). Annals Ent. Soc Am.
60(3): 626-631.
Cook, R.J., W.L. Bruckart, J.R. Coulson, M.S. Goettel, R.A. Humber, R.D. Lumsden,
J.V. Maddox, M.L. McManus, L. Moore, S.F. Meyer, P.C. Quimby, Jr., J.P.
Stack, and J.L. Vaughn. 1996. Safety of microorganisms intended for pest and
plant disease control: A framework for scientific evaluation. Biological Control
7: 333-351.
Dadd, R.H. 1975. Alkalinity within die midgut of mosquito larvae with alkaline-active
digestive enzymes. J. Insect Physiol. 21:1847-1853.
Dadd, R.H. and D.L. Krieger. 1968. Dietary amino acid requirements of the aphid,
Myzus persicae. J. Insect Physiol. 14: 741-774.
Dahlman, D.L. 1980. Field tests ofL-canavanine for control of tobacco homworm. J.
Econ. Entomol, 73:279-281.
Dahlman, D.L., F. Herald and F.W. Knapp. 1979. L-canavanine effects on growth and
development of four species of Muscidae. J. Econ. Entomol. 72: 678-679.
Dahlman, D.L. and G.A. Rosenthal. 1975. Non-protein amino acid-insect alterations (1)
Growth effects and symptomology of L-canavanine consumption by tobacco
homworm, Manduca sexta (L). Comp. Biochem Physiol. 51: 33-36.
Dahlman, D.L. and G A. Rosenthal. 1976. Further studies on the effect of L-canavanine
on the tobacco homworm, Manduca sexta. Insect Physiol. 22:265-271.
Dahlman, D.L. and G.A. Rosenthal. 1982. Potentiation of L-canavanine-induced
developmental anomalies in the tobacco homworm, Manduca sexta, by some
amino acids. J. Insect Physiol. 28(10): 829-833.
Deedat,Y.D. 1994. Problems associated with the use of pesticides: An overview. Insect
Sci. Applic. 15(3): 247-251.
Del Campo, M., and J.A.A. Renwick. 2000. Induction of host specificity in larvae of
Manduca sexta: Chemical dependence controlling host recognition and
developmental rate. Chemecol. 10: 115-121.
Dethier, V.G. and J.H. Kuch. 1971. Electrophysiological studies of gustation in
lepidopterous larvae 1. Comparative sensitivity to sugars, amino acids and
glycosides. Z. Vergl. Phys. 72: 343-363.
Dietary Supplement Information Bureau. 2000. Methionine Dietary Supplement
Education Alliance. Internet URL: http://www.supplementinfo.org/index.htm
Accessed April 2004.