<%BANNER%>

Volatile and Selected Non-Volatile Analysis of Juices From Huanglongbing Affected Hamlin and Valencia Oranges

Permanent Link: http://ufdc.ufl.edu/UFE0025084/00001

Material Information

Title: Volatile and Selected Non-Volatile Analysis of Juices From Huanglongbing Affected Hamlin and Valencia Oranges
Physical Description: 1 online resource (75 p.)
Language: english
Creator: Dagulo, Lilibeth
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: flavor, florida, hlb, huanglongbing, juice, orange
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Volatile and Selected Non-Volatile Analysis of Juices from Huanglongbing Affected Hamlin and Valencia Oranges Huanglongbing (HLB), or citrus greening, is a disease that produces multiple tree and fruit symptoms. Previous studies have reported that juice from HLB symptomatic fruit was of poor quality and bitter. The effects of HLB infection on volatile and non-volatile flavor compounds in Hamlin and Valencia orange juices were studied. Compounds which might produce bitterness, especially non-volatiles such as flavanone glycosides, polymethoxylated flavones, and limonin, were analyzed. HLB symptomatic, asymptomatic, and control Florida Hamlin and Valencia oranges at various maturity stages were harvested and juiced during the 2007-2008 season. Flavanone glycosides (FGs) and polymethoxylated flavones (PMFs) were analyzed using reversed phase HPLC. No bitter FGs (i.e. naringin) were detected. PMF concentrations were were all far below taste threshold levels reported in literature. An unknown compound with a PMF-like UV spectra was only found in HLB symptomatic juice. Limonin was analyzed using HPLC, and concentrations were 91-425% higher in symptomatic juice compared to control. However, concentrations were also below the reported average human detection threshold. Brix/acid ratios were 8-63% lower in symptomatic juice compared to control. Juice volatiles were identified by GC-MS using SPME headspace extractions. Terpenes, such as gamma-terpinene and alpha-terpinolene, were 1,320% and 62% higher in symptomatic juice than control. Esters, such as ethyl butanoate and ethyl hexanoate, important aroma volatiles, were 87% and 98% lower compared to control. The chemical composition of asymptomatic juices was very similar to control. The compound(s) responsible for producing flavor differences in juice from HLB infected and control fruit appear to be primarily associated with immaturity. The reported off-flavor associated with HLB symptomatic juices apparently stem from a lower concentration of sugars, a higher concentration of acid, and a different volatile profile.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lilibeth Dagulo.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Rouseff, Russell L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0025084:00001

Permanent Link: http://ufdc.ufl.edu/UFE0025084/00001

Material Information

Title: Volatile and Selected Non-Volatile Analysis of Juices From Huanglongbing Affected Hamlin and Valencia Oranges
Physical Description: 1 online resource (75 p.)
Language: english
Creator: Dagulo, Lilibeth
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: flavor, florida, hlb, huanglongbing, juice, orange
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Volatile and Selected Non-Volatile Analysis of Juices from Huanglongbing Affected Hamlin and Valencia Oranges Huanglongbing (HLB), or citrus greening, is a disease that produces multiple tree and fruit symptoms. Previous studies have reported that juice from HLB symptomatic fruit was of poor quality and bitter. The effects of HLB infection on volatile and non-volatile flavor compounds in Hamlin and Valencia orange juices were studied. Compounds which might produce bitterness, especially non-volatiles such as flavanone glycosides, polymethoxylated flavones, and limonin, were analyzed. HLB symptomatic, asymptomatic, and control Florida Hamlin and Valencia oranges at various maturity stages were harvested and juiced during the 2007-2008 season. Flavanone glycosides (FGs) and polymethoxylated flavones (PMFs) were analyzed using reversed phase HPLC. No bitter FGs (i.e. naringin) were detected. PMF concentrations were were all far below taste threshold levels reported in literature. An unknown compound with a PMF-like UV spectra was only found in HLB symptomatic juice. Limonin was analyzed using HPLC, and concentrations were 91-425% higher in symptomatic juice compared to control. However, concentrations were also below the reported average human detection threshold. Brix/acid ratios were 8-63% lower in symptomatic juice compared to control. Juice volatiles were identified by GC-MS using SPME headspace extractions. Terpenes, such as gamma-terpinene and alpha-terpinolene, were 1,320% and 62% higher in symptomatic juice than control. Esters, such as ethyl butanoate and ethyl hexanoate, important aroma volatiles, were 87% and 98% lower compared to control. The chemical composition of asymptomatic juices was very similar to control. The compound(s) responsible for producing flavor differences in juice from HLB infected and control fruit appear to be primarily associated with immaturity. The reported off-flavor associated with HLB symptomatic juices apparently stem from a lower concentration of sugars, a higher concentration of acid, and a different volatile profile.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lilibeth Dagulo.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Rouseff, Russell L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0025084:00001


This item has the following downloads:


Full Text

PAGE 1

1 VOLATILE AND SELECTED NON-VOLATILE ANALYSIS OF JUICES FROM HUANGLONGBING AFFECTED HAM LIN AND VALENCIA ORANGES By LILIBETH R. DAGULO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Lilibeth R. Dagulo

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS First and foremost, my sincere appreciation goes to Dr. Russell Rouseff for becoming my major advisor in my journey into higher educ ation. With his wisdom and guidance, I learned about citrus flavor and instrume ntal analysis. I became an expe rienced high performance liquid chromatography (HPLC) troubleshooter, handling various tools and disassembling solvent pump systems, a task that I thought I would never see myself doing. I greatly value the experience working with multiple analytical techniques and instruments. Dr. Rouseff was always helpful, patient, and kind, and I will always cons ider him a teacher, mentor, and friend. My gratitude also goes to my committee me mbers: Dr. Charles Sims, Dr. Rene GoodrichSchneider, and Dr. Edgardo Exteberria. Dr. Sims and Dr. Goodrich-Schneider are both exceptional professors, as I have taken their c ourses during both my undergraduate and graduate years at UF. I also appreciate that Dr. Sims let me stay in his lab my first year as I took classes, including me in lab meetings and sensory work. I never took a course by Dr. Exteberria, but he is an exceptional prof essor nonetheless. I would like to acknowledge Dr. Timothy Spa nn of the Citrus Research and Education Center for harvesting the oranges used for this study. Thanks also go to Dr. Michelle Danyluk and her lab at the Citrus Research and Educati on Center for juicing the oranges and determining juice acid levels. I would also like to thank Mr Carl Haun of the Florida De partment of Citrus for his technical support and for generous ly providing some of his PMF standards for the PMF analysis. I am very grateful to both Mr Allen Mitchell and Mr. Salvador Santos of Florida’s Natural for providing their lab and equipment in order to prepare and analyze samples for the limonin analysis. Lastly, I would like to thank Ms. Gwen Lundy of the Citrus Research and Citrus Center for providing a refractometer for the Brix analysis.

PAGE 5

5 To my lab group: Jack Smoot, Fatima Ja balpurwala, June Rouseff, Kanjana Mahattanatawee, Ozan Gurbuz, and Stephanie Ka sparian, I thank everyone for making every day at the lab interesting and memorable. I am in aw e with how everyone was always willing to offer a helping hand. I will always remember all the good memories and food we shared, and especially all the laughter and fun th at made everyone seem like a family. To all my friends back in Gainesville, I thank everyone for keeping me sane and motivated during the rough and stressful times. I would esp ecially like to acknowledge Rene and Yael, since they both helped me grow a lot in these two years. They were always there for me during good times and bad, and I will always cherish our friendship. Last, but definitely not least, I am who I am today because of my father, mother, brother, and boyfriend. Their infinite suppo rt throughout my undergraduate a nd graduate years also kept me motivated. I am eternally grateful to my pa rents, for having the motivation and courage to leave their home country for be tter opportunities. They guided me through a path in becoming a hard working student. Because of that, I am the first American born in both my parent’s families to receive a graduate degree at an American university.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ................................................................................................................ ...........8LIST OF FIGURES ............................................................................................................... ..........9ABSTRACT ...................................................................................................................... .............10 CHAPTER 1 INTRODUCTION ................................................................................................................ ..122 LITERATURE REVIEW .......................................................................................................14Overview of Huanglongbing ..................................................................................................14Casual Agents and Insect Vectors ...................................................................................14Symptoms and Economic Impact ....................................................................................15Volatile and Non-Volatile Aspects of Juice Flavor ................................................................16Juice Volatiles and Aroma Active Compounds ...............................................................16Juice Non-Volatiles and Bitterness .................................................................................18Flavanone glycosides ...............................................................................................18Polymethoxylated flavones ......................................................................................19Limonin ....................................................................................................................21Volatile and Non-Volatile Samp le Preparation and Analysis .........................................213 MATERIALS AND METHODS ...........................................................................................23Samples ....................................................................................................................... ............23Analysis of Volatiles ......................................................................................................... ......24Solid-Phase Micro Extrac tion (SPME) Procedure ..........................................................24Gas Chromatography – Mass Spectrometry ....................................................................25Peak Identification and Semi-Quantification ..................................................................25Analysis of Flavanone Glycosides ..........................................................................................25Sample Preparation ..........................................................................................................25High Performance Liquid Chromatography ....................................................................26Peak Identification and Quantification ............................................................................26Analysis of Polymethoxylated Flavones ................................................................................27Sample Preparation ..........................................................................................................27High Performance Liquid Chromatography ....................................................................27Peak Identification and Quantification ............................................................................28Analysis of Limonin ........................................................................................................... ....28Sample Preparation ..........................................................................................................28High Performance Liquid Chromatography ....................................................................29

PAGE 7

7 Peak Identification and Quantification ............................................................................29Brix and %Acid Analysis ...................................................................................................... 29Statistical Analysis .......................................................................................................... ........304 RESULTS AND DISCUSSION .............................................................................................31Non-Volatiles ................................................................................................................. .........31Flavanone Glycosides ......................................................................................................31Polymethoxylated Flavones .............................................................................................32Limonin ....................................................................................................................... ....34Brix and Acidity ............................................................................................................. 35A Possible Explanation for HL B Apparent Immaturity ..................................................37Volatiles ..................................................................................................................... .............38Volatile Profile .............................................................................................................. ..38Significantly Different Volatiles .....................................................................................39Biochemical Pathways of Volatiles and Maturity ...........................................................41Valencene and Maturity ..................................................................................................435 CONCLUSION .................................................................................................................. .....57APPENDIX: VOLATILE TABLES ..............................................................................................59REFERENCES .................................................................................................................... ..........71BIOGRAPHICAL SKETCH .........................................................................................................75

PAGE 8

8 LIST OF TABLES Table page 4-1 Polymethoxylated flavone concentrations ( g/mL) for Hamlin samples at various harvesting dates.. ............................................................................................................ ....494-2 Polymethoxylated flavone concentrations ( g/mL) for Valencia samples at various harvesting dates.. ............................................................................................................ ....50A-1 Concentration ( g/mL) of Hamlin 12/12/2007 volatiles confirmed by MS and LRI values.. ...................................................................................................................... .........59A-2 Concentration ( g/mL) of Hamlin 12/18/2007 volatiles confirmed by MS and LRI values.. ...................................................................................................................... .........61A-3 Concentration ( g/mL) of Hamlin 1/30/2008 volatil es confirmed by MS and LRI values.. ...................................................................................................................... .........63A-4 Concentration ( g/mL) of Valencia 4/4/2008 volat iles confirmed by MS and LRI values.. ...................................................................................................................... .........65A-5 Concentration ( g/mL) of Valencia 4/18/2008 vola tiles confirmed by MS and LRI values.. ...................................................................................................................... .........67A-6 Concentration ( g/mL) of Valencia 5/23/2008 vola tiles confirmed by MS and LRI values.. ...................................................................................................................... .........69

PAGE 9

9 LIST OF FIGURES Figure page 4-1 Reverse phase HPLC separation of citrus flavanone glycosides of symptomatic juice from 1/30/ 2008................................................................................................................ ..444-2 UV spectra of eluted chromat ographic peaks from Figure 4-1 ..........................................454-3 Summary of citrus flavanone glycosides ...........................................................................464-4 HPLC chromatogram and correspondi ng UV spectra of polymethoxylated flavone standards.. ................................................................................................................... .......474-5 HPLC chromatogram of polymethoxylated flavones in symptomatic juice from Valencia 4/4/2008 with UV spect ra from select compounds. ............................................484-6 Summary of limonin analysis results for both Hamlin and Valencia juice samples. ........514-7 Results of the sugar and acid analys is performed on the Hamlin samples ........................524-8 Results of the sugar and acid analys is performed on the Valencia samples ......................534-9 GC-MS chromatograms of control and sy mptomatic juice samples from 4/4/2008 .........544-10 Differences of select odor-active vola tiles in Valencia 4/18/2008 symptomatic, asymptomatic, and control juices. ......................................................................................554-11 Summary of Valencene concentrations ( g/mL) ...............................................................56

PAGE 10

10 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science VOLATILE AND SELECTED NON-VOLATILE ANALYSIS OF JUICES FROM HUANGLONGBING AFFECTED HAMLI N AND VALENCIA ORANGES By Lilibeth R. Dagulo August 2009 Chair: Russell L. Rouseff Major: Food Science and Human Nutrition Huanglongbing (HLB), or citrus greening, is a di sease that produces multiple tree and fruit symptoms. Previous studies have reported that juice from HLB symptomatic fruit was of poor quality and bitter. The effects of HLB infection on volatile and non-volatile flavor compounds in Hamlin and Valencia orange juices were st udied. Compounds which might produce bitterness, especially non-volatiles such as flavanone glycosides, polym ethoxylated flavones, and limonin, were analyzed. HLB symptomatic, asymptomatic and control Florida Hamlin and Valencia oranges at various maturity stages were harvested and juiced during the 2007-2008 season. Flavanone glycosides (FGs) and polymethoxylat ed flavones (PMFs) were analyzed using reversed phase HPLC. No bitter FGs (i.e. naringin) were dete cted. PMF concentrations were were all far below taste threshold levels repo rted in literature. An unknown compound with a PMF-like UV spectra was only found in HLB symp tomatic juice. Limonin was analyzed using HPLC, and concentrations were 91-425% higher in symptomatic juice compared to control. However, concentrations were also below th e reported average human detection threshold. Brix/acid ratios were 8-63% lower in symptomatic juice compared to control. Juice volatiles were identified by GC-MS using SPME headspace extractions. Terpenes, such as -terpinene and -terpinolene, were 1,320% and 62% higher in symptomatic juice th an control. Esters, such as

PAGE 11

11 ethyl butanoate and ethyl hexa noate, important aroma volatil es, were 87% and 98% lower compared to control. The chemical composition of asymptomatic juices was very similar to control. The compound(s) responsible for produc ing flavor differences in juice from HLB infected and control fruit appear to be primar ily associated with immaturity. The reported offflavor associated with HLB symptomatic juices apparently stem from a lower concentration of sugars, a higher concentration of aci d, and a different volatile profile.

PAGE 12

12 CHAPTER 1 INTRODUCTION Florida is the leading producer of sweet oranges in the United States. The majority of the oranges produced are used for juice processi ng. Huanglongbing (HLB), or citrus greening, is a fatal citrus disease that is threatening the worl d wide orange juice indu stry and has become a major problem in Florida. Apart from causing multiple vegetative symptoms, HLB also affects the fruit, and consequently, the juice. The main characteristics of symptomatic fruit are its size and color, for it is smaller and misshapened co mpared to fruit from non-infected trees, and it retains a green color. Early studies on HLB symp tomatic fruit in other parts of the world have reported that the juice was of poor quality and tasted bitter ( 1, 2 ). Another study noted the change in acidity and soluble solids between juices from symptomatic and control fruits ( 3 ). However, there are no studies which determine how (and if) the infec tion affects important flavor compounds. Generally, flavor can be defined as the in teraction of volatile and non-volatile compounds through smell and taste. A combination of odor activ e volatiles define fresh orange aroma. They typically include certain esters aldehydes, alcohols, and terpenes. Non-volatile compounds, such as limonin and neohesperidosides, are typically res ponsible for bitterness in citrus juices when found at levels exceeding their ta ste threshold. Therefore, the object ive of this research is to characterize the effects of HLB infection on flavor impact compounds of orange juice, specifically looking at volatile profile differences and non-vo latile compounds which might produce bitterness, including flavanone glyc osides, polymethoxylated flavones, and limonin. These flavor impact compounds can be analyz ed by employing various sample preparation techniques (i.e. solid-phase extractions) and an alytical instruments. Gas chromatography is typically used to study volatile compounds, wh ile high performance liquid chromatography is

PAGE 13

13 used to analyze non-volatile compounds. These resu lts will lay the foundation for more detailed studies (i.e. involving biochemistry, plant physiol ogy and genetics, etc.) and ultimately assist the orange juice industry in pres erving a high quality product.

PAGE 14

14 CHAPTER 2 LITERATURE REVIEW Overview of Huanglongbing The United States and Brazil ar e the major sweet orange ( Citrus sinensis ) juice producers in the world. During the 2007-08 season, Florida was responsible for 70% of orange production in the U.S. ( 4 ). The majority of the oranges produced in Florida are used for juice processing, with Hamlin and Valencia being the predominant cultivars. Hamlins are available from October to January, while Valencias are avai lable from February to June ( 5 ). Orange production in both countries has been threatened by the rapid spread of Huanglongbi ng (HLB), a devastating citrus disease. Citrus producing countries in Asia and Africa have been plagued by this disease since the early 1920s, but it has only recently appeared in the Amer icas. HLB was found in So Paulo, Brazil’s largest orange producing state, in March 2004 and in Florida in August 2005 ( 6 ). In June 2008, the disease was iden tified in Louisiana ( 7 ). As of February 16, 2009, there had been a positive confirmation of HLB in thirty-three counties in Florida ( 8 ). There are various names associated with th is disease, depending on the region. In the Philippines, it is called “mottle leaf ,” but in India and Indonesia, it is called “dieback” and “vein phloem degeneration,” respectively. In South Africa, the disease is known as “greening,” a term that has been frequently used until 1995, since one main symptom is fruit that retains a green color. The Chinese term, “huanglongbing,” meani ng “yellow dragon disease” or “yellow shoot disease,” is now the official term of this citrus disease ( 6 ). Casual Agents and Insect Vectors The causal agent of HLB is the bacteria Candidatus Liberibacter. There are three strains of this bacteria: ‘ Candidatus Liberibacter africanus’, ‘ Candidatus Liberibacter asiaticus’, and ‘ Candidatus Liberibacter americanus’( 6 ). Liberibacter are Gram-neg ative, sieve tube-restricted

PAGE 15

15 bacteria. Koch’s Postulates, which are criter ia to confirm a causal relationship between a microbe and a disease, has yet to be fulfilled fo r these HLB liberibacters, since they have not been successfully cultured outside of citrus leaves ( 9 ). The Asian Citrus Psyllid (ACP), Diaphorina citri is responsible for transmitting HLB. Psyllids obtain the bacteria when they feed o ff leaves from infected trees, inoculating other sections of the same tree or uninfected trees within a grove when they move around to feed. Younger trees are more susceptible to infection si nce they produce multiple flushes, or growth of new leaves, throughout the year, as psyllids pr efer feeding and breeding on younger leaves ( 10 ). In Asia and the Americas, the ACP species is prevalent and infect tr ees with Liberibacter asiaticus or Liberibacter americanus. Trioza erytreae however, has only been found in Africa and is known only to transmit Liberibacter africanus ( 11 ). Symptoms and Economic Impact HLB affects the entire tree, eventually i nducing death. Presently, there is no cure for HLB infected trees, but multiple research efforts are underway to try to control and minimize the spread of the disease, especially the psyllid vector. From an economic standpoint, the citrus industry has already seen an appr eciable rise in production costs. These additional costs include increased scouting for symptomatic trees and increas ed insecticide spraying to control the psyllid population ( 12 ). Since psyllids are more attracted to younger trees, because they have more flushes than older trees, insect icides must be applied more fr equently. Tree removal and/or replanting are also options for HLB management that also contribute to increased costs ( 10 ). Prior to twig dieback and decreased producti vity, there are noticeable symptoms in the root system, leaves, and fruit. In an infected tr ee, the root system starts to decay due to root starvation, and new growth is restrained ( 13 ). Therefore, the amount of fibrous roots that are present do not seem to ade quately support the tree ( 1 ). A blotchy mottle, or asymmetrical

PAGE 16

16 chlorosis, in which the inefficient production of chlorophyll causes yellowi ng or whitening of a leaf, is one of the main char acteristics of the disease ( 14 ). However, these chlorotic patterns may resemble symptoms of other citrus diseases (i.e. citrus tr isteza virus) or mineral deficiencies (i.e. iron or zinc). HLB also induces vein yellowi ng and the growth of small upright leaves ( 13 ). Fruit from a HLB infected tree tend to fall prematurely, and those that remain on the tree fail to mature correctly. The fruit is also sma ll and misshapen (i.e. l opsided) with a curved central core, often containing abor ted seeds. Another distinctive ch aracteristic of the disease is the failure of the fruit to color properly, remain ing green. The flesh and juice from an infected fruit is said to be of poor quality and have a bitter a nd unpleasant taste ( 1, 2 ). A study of HLB infected and non-infected Kinnow mandarins determined that there was a lower soluble solids content and higher acidity in HLB fruit than in control fruit ( 3 ). Volatile and Non-Volatile Aspects of Juice Flavor A food’s flavor is primarily due to the intera ction of aroma and tast e substances, which are generally volatile and non-vol atile compounds, respect ively. Volatile compounds interact with the olfactory epithelium in the nasal cavity ort honasally (through the nose) or retronasally (from the back of the throat via chewing, etc.). Flavor volatiles are considered secondary metabolites that are derived from primary metabolites, su ch as carbohydrates, lip ids, and proteins ( 15 ). Nonvolatile compounds in contrast, interact with tast e receptors on the tongue and palate generating perceptions of salty, sour, sweet bitter, or umami tastes ( 16 ). Juice Volatiles and Aroma Active Compounds The number of volatile compounds a food pr oduct contains depends on a number of factors, including the nature of the product and if it was subjecte d to any processing methods. In the case of fruits, for example, th e composition of volatile s can vary with maturity (development) and/or cultivar (i.e. Navel ora nge vs. Valencia orange). Foods produced via thermal processing

PAGE 17

17 or in conjunction with a fermentation process, su ch as coffee or tea, can differ greatly in the number of volatiles than their raw/original counterparts. Additional chemical reactions due to heat, enzymatic, or bacterial inte ractions can produce more than 700 volatile compounds in these products ( 16 ). These processes, however, may not have favorable effects in other food products. The volatile profile of processed or ange juice, for example, differs from that of freshly-squeezed orange juice. The formation of new compounds, lo ss of desirable compounds, and/or changes in the concentration ratio of cer tain compounds can produce off-fl avors, or aroma not normally present in the food ( 17 ). Although foods can contain as many as 700 vol atile compounds, only a small fraction is considered odor-active and actually contribute s to a food’s aroma ( 16 ). A volatile compound is odor-active only if its concentra tion in the food matrix exceeds its odor threshold (the lowest concentration in which the odor is recognized). Character im pact compounds, or key odorants, are odor-active compounds that provide the characte ristic aroma of a food. Methyl anthranilate, for instance, is the character impact compound fo r a Concord grape, while diacetyl is the key odorant for butter and citral for lemon ( 18 ). Early studies have identi fied more than 200 volatiles in freshly squeezed orange juice, but relatively fe w contribute to its aroma. Moreover, not one is considered an orange flavor character impact compound, so the aroma of fresh orange juice is due to a specific combinati on of odor-activ e volatiles ( 17 ). The aroma volatiles that prim arily contribute to fresh or ange juice aroma are aldehydes and esters ( 19 ). The majority of these compounds are produ cts from the oxidative degradation of fatty acids, such as linoleic and linolenic acids ( 18 ). Aldehydes (i.e. acetaldehyde, hexanal, octanal, (Z)-hex-3-enal, (E,E)-2,4-nonadienal, etc. ) provide fresh, citrus-like, green, grassy, and fatty odor notes to fresh orange aroma. Esters (i .e. ethyl acetate, ethyl butanoate, ethyl hexanoate,

PAGE 18

18 etc.) provide the fruity character of orange juice. Ethyl butanoate, especially, is one of the most potent aroma compounds and the most important ester. Its concentration in orange juice has been reported to increase with increasing fruit maturity ( 17 ). Other volatiles, including some alcohols and terpenoid hydrocarbons and their derivatives, contribute to fresh orange juice aroma. Aliphatic alcohols, such as 1-hexanol and (Z)-3-hexen-1ol, are responsible for woody, green and grassy notes. However, te rpene alcohols, which include linalool and geraniol, contribute floral and fruity notes. The terpene hydrocarbons constitute the majority of orange juice volatile s, and many are found in orange p eel oil. Only a few, though, are odor-active, and they include the pinenes ( and ) and -myrcene, which provide piney and musty odors to fresh orange juice ( 17 ). Juice Non-Volatiles and Bitterness Bitterness in citrus fruits can be caused by limonoids, flavanone glycosides and possibly polymethoxylated flavones. Limonoids and flavonoi ds are secondary plant metabolites that affect defense mechanisms and molecular signaling in plants, as well as provide health benefits for humans and animals via anticancer and antioxidant activity ( 20, 21 ). Flavanone glycosides There are several classes of flavonoids, but in fresh fruit and their juices, flavanones and flavones predominate. Flavanones are uncommon in pl ants and citrus is an unusually rich source for these compounds. The flavonoid molecular sk eleton consists of two aromatic rings (designated as A and B) connected by a dihydrop yrone ring, for flavanones, or a pyrone ring, for flavones (designated as C) ( 20 ). These compounds can influence the quality of ju ice primarily in terms of bitter taste but can in certain cases affect appearance. In Citrus flavanones are usually found with a disaccharide sugar attached at position 7, and ar e called flavanone glycosides. Not all citrus

PAGE 19

19 flavanone glycosides are bitter. Bitter flava none glycosides have a ne ohesperidose sugar (2O -L-rhamnosyl-D-glucose) whereas non-bitter flavanone glycosides have a rutinose sugar (6O -L-rhamnosyl-D-glucose) at position 7. Non sugar substitutions like hydroxyl and methoxyl groups at the 3’ and 4’ position on th e aromatic B ring differentiate them. The nonbitter rutionsides, such as hesperidin a nd narirutin, predominate in sweet orange ( C. sinensis ) varieties (i.e. Valencia, Hamlin, blo od, navel) and tangerine, or mandarin, ( C. reticulata ) varieties (i.e. Clementine, Satsuma). In sour oranges ( C. aurantium ) and grapefruit ( C. paradis ), neohesperidosides, such as nari ngin, are the dominate flavanon e glycosides. Other flavanone glycosides that can be found in Citrus juices include: didymin and eriocitrin, which are both rutinosides, and neoeriocitrin and poncir in, which are both neohesperidosides ( 22 ). Polymethoxylated flavones Polymethoxylated flavones (PMFs) are an other class of compounds that have been suggested to impart bitterness in juice ( 23, 24 ). Unlike the flavanone glycosides, which have a sugar constituent attached to the A ring, PMFs are found as aglycones (without a sugar moiety) with varying degrees of methoxylation on the ar omatic A and B rings. High concentrations can be found in sweet orange and tangerine cultivars, with lower levels found in other citrus (i.e. grapefruit) ( 25 ). They are present in every part of th e fruit, including the peel (flavedo with albedo), membranes, and juice, but the peel cont ains much higher concentrations compared to the juice ( 26 ). Studies conducted by Swift and others ( 23, 24 ) suggested that orange peel juice (liquid expressed from peel) was bitter. When extracted with benzene, it was found that the neutral fraction, among the acidic and lactone fractions was the most bitter, and PMFs were the predominant compounds found. They were later iden tified as sinensetin (S IN), nobiletin (NOB), tangeretin (TAN), 3,5,6,7,8,3’,4’-heptamethoxyflavone (HEP), and tetra-O-methylscutellarein

PAGE 20

20 (SCU). The taste thresholds for these PMFs we re determined in a synthetic medium that mimicked orange juice. Results from sensory ev aluation indicated that SC U had the lowest taste threshold (15 ppm), followed by HEP (28 ppm), SIN (30 ppm), and TAN (33 ppm), with NOB being the least potent (46 ppm) ( 27 ). For most PMFs, their averag e concentration in peel juice exceeded their taste thresholds, indicating that th ey contributed to the bitterness of peel juice ( 23 ). The presence of orange peel components in ju ice depends on the mode of extraction. For instance, hand squeezed oranges would result in little contact of the peel with the extracted juice. Thus, PMF concentration of the juice would rema in unchanged. However, during extraction with commercial juice machinery (i.e. Brown or FMC ju ice extractors), the p eel and its components can come in contact with the jui ce. The amount of orange oil that ends up in the juice depends on the desired juice quality, and it is controlled by setting certain parameters on the machine, such as peel clearance and extractor pressure. Higher ex tractor pressure, for example, would result in higher juice yield but also with elevated peel oil components in the juice. Lower extractor pressure, however, results in higher quality juice due to a lower concentration of peel components, but at the expe nse of lower juice yield ( 28 ). In the same study in which taste threshold for individual PMFs were established, commercial orange juice concentr ates obtained from commercial extractors (Brown and FMC) were also compared. The difference between the tw o brands is the extraction style, for Brown is a reamer type and FMC is a peel macerating type ( 5 ). The results were expressed as part per million (ppm) of PMFs of juice reconstituted to 12 Brix, or single-strength. In both cases, the levels of individual PMFs found in each sample were considerably less than their taste threshold, ranging from 0.20 to 2.05 ppm.

PAGE 21

21 Limonin While the presence of certain flavanone gl ycosides and polymet hoxylated flavones can cause bitterness immediate in orange juice, there is a group of compounds that can cause “delayed bitterness.” Limonoids are oxygenated triterpenoid compounds that are present in the seeds and flesh of the fruit. Limonin, the major limonoid found in orange juice, is extremely bitter. Its tasteless precursor, limonoate A-ring la ctone (LARL), is present in the flesh of the intact fruit ( 21 ). When the fruit is physically damaged by, for example, extractors for juice processing, the precursor interacts with the acid ic environment and is transformed into limonin. This biochemical transformation is catalyzed by the enzyme limonin D -ring lactone hydrolase and occurs at a pH of 6.5 or lower ( 29 ). However, multiple studies have shown that delayed bitterness by the conversion of LARL to limonin decreased throughout frui t growth and maturity, which pa rtly attributed to why juice from mature oranges were less bitter than im mature oranges. A study by Hasegawa and others ( 30 ) discovered that the concentration of LARL decreased as fruit matured, correlating with decreased bitterness in a process termed “natural debittering.” The mechanism of this process was not understood until the discovery of limonoid gl ucosides, which were tasteless compared to their bitter aglycones ( 31 ). As fruit matured, it was shown th at the concentration of limonin 17D-glucopyranoside (LG) increased as the concen tration of LARL decreased. It was concluded that LARL metabolized into LG instead of limonin at the late stages of maturity ( 30 ). Volatile and Non-Volatile Samp le Preparation and Analysis The extraction of non-volatiles and volatiles fr om orange juice has traditionally been accomplished using liquid-liquid extraction (LLE), which utilizes different solvents and the interactions between the solven ts and the compound of interest to isolate it. Solid-phase extraction (SPE) is a more rapid method that did not require the use of expensive glassware or

PAGE 22

22 large quantities of organic solvents ( 32 ). The main principle behind this extraction is that the compound of interest is absorbed onto a modified solid support from the sample, which is then desorbed, or eluted, by solvent or thermal means ( 33 ). The analysis of flavonoids that are found in orange juice, for example, may involve the use of C18-bonded silica, a common sorbent used for reversed-phased SPE ( 34 ). In reversed-phase SPE, the sample matrix is usually polar, but the compounds of interest are midto nonpolar. Theref ore, a modified nonpolar sorbent is used. As the sample is passed through the nonpolar statio nary phase, the nonpolar compounds are retained while the impurities are washed away. They are then eluted using a strong organic solvent, such as methanol or acetonitrile ( 32 ). Unlike traditional solid-phase ex traction, solid-phase micro ex traction (SPME) is a rapid and solvent-less technique that is commonly used for the analysis of volatiles. Th e volatiles can be directly extracted and concentrated from th e sample headspace onto a fiber coated with a modified sorbent (i.e. polydimethylsiloxane, etc. ). After a certain amount of exposure time, the fiber can go straight to the inje ction port of a gas chromatogra phy system where the analytes are thermally desorbed ( 33, 35 ). After the compounds of interest are extracted from the sample matrix, they are separated, identified, and quantified by analytical instrume nts. Volatile compounds are usually analyzed using gas chromatography, while non-volatiles are analyzed using high performance liquid chromatography.

PAGE 23

23 CHAPTER 3 MATERIALS AND METHODS The main objective of the study was to de termine how HLB affects flavor impact compounds in orange juice, so several samp le preparation techni ques and analytical instrumentation were used to study volatile and selected non-volatile compounds. A secondary objective was to determine whether changes (if any) from the disease differed during fruit development. Therefore, samples were harvested at different stages of maturity. However, since tree removal is a common practice when HLB is found in groves, it was difficult to sample from the same area at different harvesting dates. Because of this, the majority of the samples were obtained from different areas. Samples The Florida Hamlin oranges were obtaine d from commercial groves on December 12, 2007 (Sebring, FL), December 18, 2007 (Ft. Pierce, FL), and January 30, 2008 (Ft. Pierce, FL). The last two harvesting dates occurred from th e same grove. The Valencia oranges were also obtained from commercial grove s on April 4, 2008 (Dover, FL), April 18, 2008 (Clewiston, FL), and May 23, 2008 (Lake Placid, FL). Three types of fruit were harv ested from the same grove at each date. Fruit that showed physical symptoms of HLB that came from a HLB infected tree was designated as symptomatic (+ +). Fruit that di d not show physical symp toms of HLB but came from a HLB infected tree was designated as asym ptomatic (+ -). Fruit that was harvested from a non-infected tree served as control (-). Samp ling for each harvesting date was done from 3-5 trees. The number of trees sampled was determined by the severity of symptoms. For example, if a tree was heavily symptomatic, fewer trees were sampled. However, if symptoms were not very severe, more trees were sampled to obtain the ne eded sample size. The number of trees sampled for asymptomatic and control fruit was dete rmined by the number of trees sampled for

PAGE 24

24 symptomatic fruit. The fruits were picked ra ndomly from around the tree. However, because of the nature of the disease, HLB would be sector ed within the tree and a small section of the canopy or a branch would be symptomatic, so symptomatic fruit would be obtained from a limited area. During the harvesting period, the cas ual agent of HLB in Florida, Candidatus Liberibacter asiaticus, was under the Plant Protec tion and Quarantine (PPQ) Select Agents and Toxins list of the Animal and Plant Health Inspection Service (A PHIS) of the U.S. Department of Agriculture (USDA). Therefore, the oranges were juiced under a controlle d observation room All samples were juiced by hand using a Sunkist juice extractor until approxi mately 4.5 gallons were obtained. The juices were portioned into smalle r containers and then stored at -18C until analysis. All samples were thawed before any analysis. Analysis of Volatiles Solid-Phase Micro Extraction (SPME) Procedure Extraction of volatiles using headspace so lid-phase micro extraction (SPME) was done using a method described by Bazemore and others ( 36 ) and modified by Mahattanatawee and others ( 37 ). The extraction was accomplished using a 2 cm 50/30 m DVB/CarboxenTM/PDMS StableFlexTM fiber (Supelco, Bellefonte, PA). An aliquo t (10 mL) of whole ju ice was placed in a 40 mL glass vial with a silicone/PTFE septa sc rew cap. An internal standard was added (50 L of 2000 g/mL benzyl alcohol (Aldrich, St. Loui s, MO) in methanol) and thoroughly mixed before the vial headspace was purged with nitr ogen. The sample was gently stirred by a stirring bar and allowed to equilibrate in a 40C water bath for 30 minutes. After the equilibration period, the SPME fiber was inserted into the vial headspace and exposed at 40C for 45 minutes.

PAGE 25

25 Gas Chromatography – Mass Spectrometry A Perkin Elmer (Waltham, MA) Clarus 500 GC-MS system was used to analyze the volatiles. It contained a Stabilwax column (Restek 60 m, 0.25 mmID, 0.5 m df) with the mass spectrometer scanning 25 to 300 m/z. Helium wa s the carrier gas. The temperature program began at an initial temperature of 40C for 2 mi nutes, followed by a ramp of 7C/min to 240C, held for 9.50 minutes for a 40-minute total run time. Peak Identification a nd Semi-Quantification GC-MS chromatograms were analyzed usi ng TurboMass software (Perkin Elmer, Waltham, MA). Peak identifications were confir med with mass spectra obtained from libraries and linear retention index (LRI) values calculated by a standard curve generated by injecting low and high alkane series (C6-C9 and C8-C25). Since only one internal standard was used, the concentration of each volatile was only semi-qua ntified. The ratio of the internal standard’s concentration to its peak area response is propor tional to the ratio of a compound’s concentration to its peak area response. Since a known concentrat ion of the internal standard was added to the juice, the concentration of each compound can be calculated. If 50 L (0.050 mL) of a 2000 g/mL stock solution of benzyl alcohol is added to 10 mL of orange juice sample, there would be 10 g/mL of benzyl alcohol in the sample. The c oncentration of a compound in the sample, [C] = PAC (Peak Area of Compound) / PAIS (P eak Area of Intern al Standard) 10 g/mL. Analysis of Flavanone Glycosides Sample Preparation Extraction of flavanone glycosides was done using a modified version of methods described by Rouseff and others ( 38 ) and Bronner and Beecher ( 39 ). Rhoifolin (400 L of 1000 g/mL) (Indofine, Hillsborough, NJ), made up in meth anol, was added as an internal standard to 5 mL of juice sample and swirled. After centrif uging the sample at 4000 rpm for 15 minutes, the

PAGE 26

26 supernatant was carefully separated from the pell et on the bottom of the centrifuge tube. A C-18 SPE cartridge (Phenomenex Strata C18E 500 mg/6 mL, 17.5% carbon loading, 461 m2/g surface area, 76 pore size, 53 m particle size) was conditioned with 4 mL of methanol followed by 8 mL of deionized water. The supernatant was pass ed through the C-18 cartri dge at no faster than 1 drop/second. Afterwards, the cartridge was wa shed with 5 mL of deionized water. The flavanone glycosides were then slowly eluted wi th 3 mL of acetonitrile into a 10 mL volumetric flask. The pellet was sonicated with 3 mL of dimethylformamide and then centrifuged at 4000 rpm for 10 minutes. The resulting supernatant was added to the 10 mL volumetric flask with the C-18 eluant. Deionized water was used to bring solution to volume before mixing with a small magnetic stirring bar for up to 5 minutes. About 1.5 mL of the solution was filtered into a HPLC autosampler vial using a 0.45 nylon filter (Fisherbrand). High Performance Liquid Chromatography A Thermo (Waltham, MA) Finnigan Surveyor HPLC system was used to analyze the flavanone glycosides. It contained a revers ed phase C-18 column (Phenomenex Luna 5 C18, 250 x 4.60 mm 5 ) and PDA detector monitoring 240, 280, and 340 nm wavelengths. The mobile phase consisted of a gradient program that began at 18% acetonitrile and 82% aqueous acetic acid (1%) and ended at 60% acetonitrile and 40% aqueous acetic acid (1%) in 30 minutes. The flow rate was set at 1 mL /min and injection volume was 25 L. Injections were done in triplicate. Peak Identification and Quantification HPLC chromatograms were analyzed usi ng Xcalibur software (Thermo Electron Corporation, Waltham, MA). Peak identifications were confirmed by retention time by injecting naringin (Acros Organics NJ), hesperidin (A cros Organics, NJ), and rhoifolin standards. Confirmation was also done through maximum ab sorbance of certain wavelengths of the UV

PAGE 27

27 spectra as provided by the standards. These pa tterns were compared with those found in literature. Concentration of compounds ( g/mL) was calculated according to the peak area response of the internal standard, as desc ribed above with the GC-MS analysis. The concentration of rhoifolin in the final sample volume was 20 g/mL. Due to dilution during sample preparation, the concentr ation of a compound in the samp le, [C], = PAC (Peak Area of Compound) / PAIS (Peak Area of Internal Standard) 20 g/mL 2. Analysis of Polymethoxylated Flavones Sample Preparation Extraction of polymethoxylated flavones was done using a modified version of a method described by Mouly, Gaydou, and Arzouyan ( 40 ). After centrifuging a 10 mL juice sample at 4000 rpm for 15 minutes, the supernatant was carefully separated from the pellet at the bottom of the centrifuge tube. Flavone (30 L of 100 g/mL) (Acros Organics, NJ) was added to the supernatant as an internal standard. A C-18 SPE cartridge (Phenomenex Strata C18-E 500 mg/6 mL, 17.5% carbon loading, 461 m2/g surface area, 76 pore size, 53 m particle size) was conditioned with 5 mL of methanol followed by 10 mL of deionized water. The supernatant was passed through the C-18 cartridge at no faster than 1 drop/second. Afterwards, the cartridge was washed with 5 mL of deionized water followed by 3 mL of a purificati on solution (90% water and 10% methanol). The polymethoxylated flavones were then slowly eluted with 2 mL of methanol into an amber vial. About 1 mL of th e eluant was transferred to a HPLC autosampler vial prior to analysis. High Performance Liquid Chromatography A Thermo (Waltham, MA) Finnigan Surveyor HPLC system was used to analyze the polymethoxylated flavones. It contained a reve rsed phase C-18 column (Phenomenex Luna 5 C18, 250 x 4.60 mm 5 ) and PDA detector monitoring 240, 280, and 325 nm wavelengths. The

PAGE 28

28 mobile phase consisted of a gradient program that began at 45% acetonitrile (kept constant), 50% deionized water, and 5% methanol and ended at 20% deionized water and 35% methanol in 20 minutes. The flow rate was set at 1 ml/min and injection volume was 25 L. Injections were done in triplicate. Peak Identification and Quantification HPLC chromatograms were analyzed usi ng Xcalibur software (Thermo Electron Corporation, Waltham, MA). As with the flavanone glycosides analysis, peak identifications were confirmed by retention time by injecting sinensetin, nobiletin, fl avone, and tangeretin standards. Maximum absorbance of certain wa velengths of the UV spectra provided by the standards was also used for confirmation. Thes e patterns were compared with those found in literature. Concentration of compounds ( g/mL) was calculated according to the peak area response of the internal standar d, as described by the previous analyses. The concentration of flavone in the final sample volume was 1.5 g/mL. Due to concentrating during sample preparation, the concentrati on of a compound in the sample, [C], = PAC (Peak Area of Compound) / PAIS (Peak Area of Internal Standard) 1.5 g/mL 1/5. Analysis of Limonin Sample Preparation Juice samples were prepared according to a modified version of a procedure described by Widmer and Haun ( 41 ). 3 mL of whole juice was placed in a round-bottom centrifuge tube and heated in a 90C water bath for 10 minutes in order to convert remaining LARL to limonin and to dissolve any precipitated lim onin. The sample was diluted with 3 mL of 40% aqueous acetonitrile and thoroughly stirred using a Vortex for 5 seconds. The solution was filtered using a 0.45 m nylon filter with a glass microfiber before filling a HPLC autosampler vial.

PAGE 29

29 High Performance Liquid Chromatography A Perkin Elmer (Waltham, MA) Series 200 Auto sampler connected to two Acuflow Series III Pumps and a Perkin Elmer 785A UV/VIS Detect or was used to analyze limonin. The system was set up to perform an automated solid-phase extraction using a switching valve as described in Widmer and Haun (2000). One Acuflow pump pu mped a solution of 37% acetonitrile in water for limonin analysis, which remained isocratic throughout the analysis. The other pump pumped a solution of 19% acetonitrile in water for naringin analysis. The detector was set at 210 nm for limonin, and the flow rate was 1.10 ml/min. Injection volume was 20 L. Total run time was 20 minutes, with valve switches at 3 and 15 minutes. A Zorbax CN column (4.6 x 150 5 L) was used for limonin. Peak Identification and Quantification Chromatograms were analyzed using TotalChr om software (Perkin Elmer, Waltham, MA). Peak identification was confirmed with injections of limonin standa rd at different concentrations (20, 10, 5, and 1 g/mL). Concentration of compounds ( g/mL) was calculated according to a standard curve generated by the limonin standard injections. Brix and %Acid Analysis Procedures to determine juice quality, esp ecially total soluble solids (Brix) and total titratable acidity, were done according to FMC F oodTech’s Procedures for Analysis of Citrus Products ( 42 ). Brix values were determined by placi ng a couple drops of juice onto a digital abbe refractometer (Leica Mark II Plus) in trip licate. Acidity was determined by a titration method in which 25 mL of juice was brought up to 100 mL with deionized water and mixed thoroughly. A couple drops (4 to 5) of phenolphthalein was added as an indicator. The solution was titrated with 0.3123N sodium hydroxide (N aOH) until the color of the solution was constantly a light pink. Brix va lues were corrected using the %acid values obtained from the

PAGE 30

30 titrations, in which Brixc = Brix + (%acid*0.2). Brix/acid ratios were then calculated by Brixc/%acid. Statistical Analysis Concentration values of all compounds analyzed were averaged and standard deviations were calculated using Microsoft Excel software. All values were also subjected to one-way analysis of variance (ANOVA) and Tukey’s Hones tly Significant Differences (HSD) Test to determine if there were significant differen ces between the three juice types (control, asymptomatic, and asymptomatic). This was done using Statistica 7.1 from StatSoft (Tulsa, OK).

PAGE 31

31 CHAPTER 4 RESULTS AND DISCUSSION Non-Volatiles Flavanone Glycosides The objective of this portion of the study was to determine if bitter flavanone glycosides (FGs) were present in the juice samples and might be responsible for the literature reported bitterness in HLB symptomatic juice. Since bitte r FGs, such as naringin and neohesperidin, are present only in sour orange a nd grapefruit varieties, there was a minimal, if any, chance that these compounds were present in Hamlin and Vale ncia oranges, which are sweet oranges. Sweet oranges do not contain bitter FGs but are abundant in non-bitter FGs, such as narirutin and hesperidin. It was not known, however, if HLB can cause bitter compounds to develop in symptomatic juice. Using HPLC, it was determined that all ju ice samples (control, asymptomatic, and symptomatic) from both Hamlin and Valencia sa mples did not contain any traces of bitter flavanone glycosides. However, three non-bitte r flavanone glycosides including narirutin, hesperidin, and didymin, were identified. Figure 4-1 shows the elution order of the FGs under reverse phase HPLC conditions, where narirutin is the most polar, followed by hesperidin and didymin. Indications (arrows) also show where naringin and neohesp eridin would have eluted if they were present in the juice sample. Although th ey eluted at different times, the UV spectra of narirutin, hesperidin, and didymin were very similar, having max at 330, 280, and 240 nm, as shown in Figure 4-1(A). Rhoifolin, the internal standard used in this analysis, is a flavone glycoside instead of a flavanone glycoside, so its UV spectra differed from the other compounds with max at 340, 270, and 240 nm.

PAGE 32

32 Except for narirutin in all samples within the April 4, 2008 harvesting date, there was a significant difference ( = 0.05) between control and symp tomatic juices. Changes in FG concentration from control ranged as low as 4% and as high as 343%, but there were no consistent trends, as seen in Figure 4-3. Hesperidin was a difficult compound to examine due to its limited solubility. It is almost insoluble in water, so standard solutions were made up in dimethylforamide (DMF) and acetonitrile. In this study, concen tration of hesperidin was reduced after freeze thaw cycles, even though sample preparation and HPLC conditions remained unchanged. During normal sample preparation, the juice wa s centrifuged and only the supernatant was analyzed while the pellet was discarded. Therefore, only dissolved hesperidin was analyzed. However, hesperidin can readily precipitate out of solution, and wo rk by Gil-Izquierdo and others ( 43 ) reported that freezing decreased the concentration of dissolved hesperid in, due to precipitation. In order to measure the amount of total hesperidin, extra steps were followed to solubili ze precipitated hesperidin. DMF was added to extract and solubilize hesperidin. The resulting solution was combined with that from the supernatant extraction and analyzed with HPLC so that total hesperidin (soluble and insoluble) was determined. Another symptom of HLB found in symptomatic fruit was found the presence of white crystals, varying in size, within juice segmen t membranes. Presently, the composition of the crystals is unknown, but given the nature of hesp eridin, it has been hypothesi zed that the crystals are precipitated and crystallized hesperidin. Polymethoxylated Flavones Polymethoxylated flavones (PMFs) are also reported to impart bitterness ( 24 ). Given that these juice samples were hand squeezed instead of mechanically squeezed using commercial extractors, the amount of PMF’s and other peel components found in the juice was expected to

PAGE 33

33 be lower than commercial juices. Highest levels of PMF’s are found in the peel particularly the flavedo and corresponding peel oil. A subsequent study ( 27 ) determined PMF concentrations in commercial and hand extracted juices as well as de termining bitterness taste threshold levels for five PMFs. In both juice types, the concentrations of all PMF’s were far below their respective taste thresholds. However, PMF’s in that study were determined using thin-layer chromatography (TLC) and not HPLC. An ensuing HPLC study ( 44 ) compared PMF values obtained using HPLC with those obtained from the TLC procedure. For almost all PMFs quantified, HPLC values were greatly lower than corresponding TLC values, possibly due to the greater chromatographic resolution of HPLC versus that of TLC. Six major citrus PMFs have been re ported, including sinensetin (SIN), hexamethoxyflavone (quercetogetin) (HEX), nobiletin (NOB), tetr amethyl-O-scutellarein (SCU), heptamethoxyflavone (HEP), and tangeretin (TAN). Figure 4-4 shows a chromatogram of the separation between SIN, NOB, fla vone (the internal standard), and TAN standards, along with their corresponding UV spectra. TAN was not quan tified as it coeluted with another compound of similar spectral prop erties and approximately equal concentrations. A unique peak was observed only in chromat ograms of symptomatic juice samples from Hamlin dates 12/12/2007 and 1/30/2008 and Vale ncia dates 4/4/2008 and 4/18/2008. It eluted between SIN and HEX at about 9.7 min and is labeled as peak B in Figure 4-5. The unknown peak’s UV spectra and those of HEX, SCU, and HEP are also shown. The unknown peak contains a strong response around 325 nm, which is characteristic of polymethoxylated flavones. However, coumarins, another class of compounds that are derived from the same biochemical pathway as flavonoids, possess similar UV spectra as polymethoxylated flavones ( 45 ). In plants, coumarins are associated with defense, es pecially against phyt opathogens and stress ( 46 ). This

PAGE 34

34 suggests that the unknown compound may be a coum arin instead of a polymethoxylated flavone, since it seems to appear only in symptomatic juice. Concentrations ( g/mL) of the polymethoxylated flavone s found in the juice samples are summarized in Table 4-1 for Hamlins and Table 4-2 for Valencias. There were significant differences between control and symptomatic ju ices for the majority of the PMFs for both Hamlin and Valencia cultivars. However, there were no consistent trends in terms of one type of juice having higher or lower concen trations of PMFs than the othe r, possibly because of different harvesting locations and cultural practices. PMF values in current study were similar to those reported for hand-squeezed juices and never exceede d reported bitterness thresholds. Therefore, bitterness due to PMFs is highl y unlikely. For example, the reporte d taste threshold for SCU, the most potent of the PMFs, is 15 g/mL, but its concentration f ound in the samples ranged from 0.056 – 1.084 g/mL (about 15x lower than th e bitterness threshold) ( 27 ). Limonin Limonin is responsible for delayed bitterness in juices in sweet orange cultivars. When juice is freshly extracted the bitter component exists as a ta steless precursor, limonate-A-ring lactone (LARL). It rapidly converts to the bitter limonin in the presence of acid and heat. Studies conducted by Hasegawa and others ( 30, 31, 47 ), followed by a study by Fong and others ( 48 ) determined that the concentration of LARL decreases in the flesh of an orange as it matures. As the fruit matures, the tasteless precursor LARL converts to another tasteless but more stable glucoside, limonin 17-D-glucopyranoside (LG). Therefore th ere is less of the LARL to be converted to bitter limonin. This “natural debitte ring” process might enable limonin to act as an indicator of maturity. Figure 4-6 is the summary of limonin concentr ation for all three ju ice types and all six harvesting dates. Significant differences ( = 0.05) were especially seen between control and

PAGE 35

35 symptomatic juices. Limonin concentrations were 91-425% higher in symptomatic juice compared to control. This sugge sts that the conversion LARL to the non bitter glucoside LG was inhibited or delayed in symptomatic juice. Additional HPLC studies to determine LARL and LG concentrations during normal fru it maturation would have to be done to confirm this hypothesis. Nonetheless, the limonin concentration values suggest that symptomatic fruit appears to be immature. For early season Hamlins (12/12/2007 and 12/18/2007) and Valencias (4/4/2008), the differences between control and asymptomatic jui ces were not significant. The differences were significant for the later harvesting dates, but were still overshadowed by higher limonin concentrations in symptomatic juice. Overall, it seems that the differences decrease as the fruit matures. According to Guadagni and others ( 49 ), the average human dete ction threshold of limonin in orange juice is approximately 6.5 g/mL. Although limonin levels were higher in symptomatic juice, and ranged from 2.405 – 5.137 g/mL, they never exceeded the average threshold. Although some bitter sensitive people might be able to detect bitterness in symptomatic juice, the average person would pr obably not detect it. Work with a trained descriptive analysis panel, which contained panelists with a wide range of bitterness sensitivities, would have to be done to determine if limonin bitterness would be of concern. Brix and Acidity Other non-volatile compounds that play an importa nt role in flavor, es pecially pertaining to citrus juice flavor, are sugars (i.e. monoand di -saccharides) and acids (i.e. citric, malic, etc.). These compounds stimulate taste receptors for sweetness and sourness, respectively. Soluble solids, or Brix, is mostly made up of sugars, incl uding sucrose, fructose, a nd glucose. Citric acid is the dominant acid in orange juice, so % acid valu es generally refers to th e percentage of citric acid. In orange juice processing, these two values are very impor tant quality markers since they

PAGE 36

36 are also maturity indicators. The USDA has set minimum and maximum values for these quality markers for different grades of orange juice. Generally, as fruit mature s, sugar concentration increases as acid concentration decreases. Oranges used for ju ice processing must mature long enough to contain a minimum Brix to % acid ratio in order to be sold as a juice from Florida. ( 28 ). This ratio denotes the balance between sweetness and acidity ( 5 ). Figure 4-7 (A) shows that there are significant differences betw een all three types of juice in Brix (with acid correction) concentrations for the Hamlins. Also, Brix values were significantly lower by 13-16% in sy mptomatic juices compared to c ontrol, with the exception of the 12/18/2007 harvesting date. The same general trend was seen with the Valencias in Figure 48 (A). Brix values were signi ficantly lower by 18-24% in symptomatic juices compared to control. Therefore, it can be speculated agai n that symptomatic juice appears to be from immature fruit even though the ju ices are from fruit of the same age. In other words, HLB may be delaying maturity. The Brix/acid ratios are summarized for the Hamlins in Figure 4-7 (B) and the Valencias in Figure 4-8 (B). Since sugar concentration incr eases as acidity decreases with maturity, an increase in the ratio should be seen. For the Ha mlins, the Brix/acid ratio was significantly lower in symptomatic juice compared to control by 8-22%. The same was seen with the Valencias. Because acid titrations for the Valencias could not be repeated because of a lack of sample, statistical analysis on Brix/acid ratios could not be carried out. However, the data seems to suggest that ratios were lower in symptomatic juice compared to control by 45-63%, again suggesting that symptomatic fruit appears to be less mature. The USDA has specific requirements for grades, or quality standards, of orange juice. For example, Grade A not from concentrate (NFC) or ange juice has a minimum Brix/acid ratio of

PAGE 37

37 12.5 ( 5 ). According to Figure 4-7 (B ) and Figure 4-8 (B), almost all symptomatic juice ratios do not meet this minimum requirement. A Possible Explanation for HLB Apparent Immaturity A recent study by Kim and others ( 50 ) on the response of sweet orange ( Citrus sinensis ) leaves to Candidatus Liberibacter asiasticus infection, the casual agent of HLB in Florida, indicated that the pathogen affected numer ous metabolism categories. 317 genes were downregulated (decreased gene and corresponding pr otein expression) while 307 genes were upregulated (increased gene and corresponding protein expression) in HLB infected leaves. Many of these genes were related to plant defense and stress, nutrient metabolism (i.e. protein, sugar, lipid), metabolite transport, and phytohormones, among others. Citrus fruit is considered a “sink organ,” in which nutrients from mature leaves are transported and stored. Photosynt hetic cells are located in the leaves, where light energy is converted to nutrients, especially sugar. Result s indicated that genes di rectly corre lated with photosynthesis were not affected by the HLB pathoge n. However, more than 4% of the affected genes were related to sugar metabolism, especia lly for starch synthesi s and degradation. Three genes responsible for starch synthesis (AGPas e, starch synthase, and granule-bound starch synthase) were up-regulated. There was a higher concentration of sucrose in the HLB leaves compared to the healthy control due to sieve tube blockage, whic h would also contribute to the accumulation of starch. It was suggested that the blockage would cause nutr ient deficiencies in sink organs, which would hinder seed developmen t and fruit maturation. This prior study might help explain why sugar levels were lowest in HLB fruit juice. Overall, juice from symptomatic fruit may not be bitter, but will probably be more sour and less sweet due to a lower B/A ratio. Also, the claim that symptomatic fruit in South Africa was bitter was from a personal observation and no t a trained sensory panel. Furthermore, HLB

PAGE 38

38 disease was already established in the area for mo re than 30 years by a vect or and agent that is different from those found in Florida, where HL B was found in 2005. Therefore, it can be said that the Florida oranges came from newly infected trees. Candidatus Liberibacter asiaticus, th e HLB pathogen, affects orange leaves at a genetic level, resulting in the inability of nutrients to be transported from the leaves to developing fruit. This would cause a delay in maturity and othe r developmental problems. Elevated limonin and reduced B/A ratios in juices from symptomatic fruit suggest that HLB delays maturity and secondary metabolite formation. Although limonin causes delayed bitterness, a “nat ural debittering” process occurs in the fruit as it matures, where limonin’s tasteless precursor, limonoate A-ring lactone (LARL), converts to a tasteless limonin gl ucoside instead of the bitter lim onin. Therefore, limonin may be used as a maturity marker. In that sense, si gnificantly higher lim onin concentrations in symptomatic juice compared to control suggests th at symptomatic fruit appears to be immature. Volatiles Volatile Profile The aroma of a food greatly contributes to its overall flavor, so examining the volatile profile of juices from HLB sy mptomatic and asymptomatic fru it would determine whether if there were compounds responsible for producing an off-flavor. The polar column (DB-Wax) on the GC-MS separated more than 90 com pounds from juice headspace solid phase microextraction (SPME). However, only 50 of th ese could be identified using the combination of mass spectra and LRI values. Various classe s of compounds were f ound, including alcohols, aldehydes, esters, ketones, and es pecially terpenes. The compounds that were identified, as well as their concentrations ( g/mL), are summarized in Tables A-1 through A-6. They included 15 terpenes, 13 alcohols (including te rpene alcohols), 10 es ters, 7 aldehydes, 4 ketones, and 1 oxide.

PAGE 39

39 New compounds that might have been responsible for an off-flavor we re not detected in symptomatic and asymptomatic juice samples compared to control. There were, however, obvious differences in the volatile profile, in terms of concentrations of certain compounds, between control and symptomatic juices, as illu strated in Figure 4-10. Two that stand out are ethyl butanoate, an ester, (B) and valencene, a sesquiterpene, (I). Both happen to be compounds of interest, since they have been associated with fruit maturity. A more detailed discussion on both will follow in the next sections. Significantly Different Volatiles Depending on the cultivar, there were a number of compounds out of the 50 identified that were significantly different ( = 0.05) between the juice type s, Control vs. Symptomatic and Asymptomatic vs. Symptomatic. For Hamlins, the compounds that were consistently significantly different between control and symp tomatic juices were ethyl acetate, ethanol, pinene, ethyl butanoate, hexanal, -pinene, myrcene, methyl hexanoate, -terpinene, terpinolene, hexanol, nonanal, 1-octen-3-ol, linalool, 4-terp ineol, ethyl-3-hydroxyhexanoate, terpineol, and perillaldehyde. For the comparison of asymptomatic with symptomatic juices, only ethyl octanoate was consistently significantly different. Compounds th at were consistent with the comparison of control with symptomatic and as ymptomatic with symptomatic included ethyl butanoate, hexanal, -pinene, linalool, and 4-terpineol. For the Valencia juices, the compounds that were consistently significantly different between control and symptomatic juices were ethyl hexanoate, -terpinene, octanal, ethyl octanoate, 1-octen-3-ol, linalool, 4-te rpineol, ethyl-3hydroxyhexanoate, -terpineol, and caryophyllene oxide. For the comparison of asym ptomatic with symptomatic juices, decyl acetate and nootkatone were consis tently significantly different. Compounds that were consistent

PAGE 40

40 with both comparisons included ethyl hexanoate, -terpinene, ethyl octa noate, 1-octen-3-ol, linalool, ethyl-3-hydroxyhexanoate and caryophyllene oxide. The majority of the data suggests that HLB affects classes of compounds differently. For example, with Hamlin juices, the esters (ethyl acetate, ethyl butanoate, methyl hexanoate, and ethyl-3-hydroxyhexanoate) that were consistently significantly different between control and symptomatic juices were 33 – 87% lower in symptomatic compared to control. However, the terpenes and alcohols derived from terpenes ( -pinene, -pinene, myrcene, -terpinene, terpinolene, linalool 4-terpineol, and -terpineol) were 79 – 773% hi gher in symptomatic juices compared to control. No clear trend was seen with the aliphatic alcohol s (ethanol, hexanol, and 1-octen-3-ol). Surprisingly, the aldehydes (h exanal, nonanal, and pe rillaldehyde) were 81 – 337% higher in symptomatic jui ces compared to control. Except for the aldehydes, the same trend was seen with the esters and terpenes deemed as consistently significantly different for the compar ison between control and symptomatic juices in the Valencias. The esters (eth yl hexanoate, ethyl octanoate, and ethyl-3-hydroxyhexanoate) were 29 – 84% lower in symptomatic compared to cont rol juices. The terpenes and alcohols derived from terpenes ( -terpinene, linalool, 4-terpineol, and -terpineol) were 80 – 1,320% higher in symptomatic juices compared to control. Figure 4-11 illustrates the magnitude of the di fferences for select compounds that are, according to literature, important odor-active compounds. Esters and aldehydes, for example, contribute the most to orange juice odor. A lthough most of the compounds detected were terpenes, most have very limited aroma activit y. Gas chromatography – olfactometry studies would have to be done to confirm the aroma activ ity of these compounds from the juice samples. All the esters that were found to be significan tly different possess aroma activity according to

PAGE 41

41 literature, but at different degr ees, with ethyl butanoate being th e most potent. Overall, they provide a fruity aroma to orange jui ces. Of all the terpenes listed, only -pinene, -pinene, myrcene, and -terpinolene are cited to possess aroma activity, and they offer green, woody, and piney aromas. Linalool, a terpene al cohol, has been cited to be the most potent alcohols in juice, providing a sweet, floral and fruity smell ( 17 ). Biochemical Pathways of Volatiles and Maturity It has been determined that the casual ag ent of HLB in Florida has affected multiple genes in the leaves to have an effect on fr uit development and maturity. Changes in sugar metabolism and the restriction of nutrient flow from leaves to fruits probably accounts for the lower level of sugars in juice from HLB inf ected oranges. Sucrose can be broken down into glucose, which is a common precursor to va rious other biochemical pathways for other metabolites (i.e. lipids, amino acids, etc.) and inte rmediates (i.e. acetyl-CoA) that are important flavor precursors. A number of odor-active volatile compounds, especially aliphatic aldehydes, alcohols, and esters, are derived from the oxidation of lipids. In fruit, the oxidation of linoleic (C18:2) and linolenic (C18:3) acids yield these com pounds. Pathways for this process include and oxidation, where removal of two-carbon units, in the form of acetyl-CoA, yield shorter chain acids. These would then react with alcohols to produce esters. For instance, a series of oxidations of linoleic acid can produce hexanoic acid (C6), which could then react with an alcohol to yield hexanoate esters. Another pa thway involves the catalyzed insertion of oxygen into the fatty acid by th e enzyme lipoxygenase to yield hydro peroxides. The hydroperoxides can then be further degraded (with or without en zymes) to produce volatile aldehydes and alcohols. The lipoxygenase catalyzed oxidatio n of linoleic acid, for example, can produce hexanal, an odor-active compound, which can be converted to hexanol ( 18 ).

PAGE 42

42 For oranges, one indication of fruit maturati on is the color change of the rind, or flavedo, from green to a combination of red, orange, and/or yellow pigments. This occurs when chlorophyll in the flavedo degr ades, causing an increase of car otenoid expression that these colors are derived from ( 51 ). Carotenoids are also precursors to many aroma active compounds, particularly damasecnones, damascones, and ionone s, that provide sweet, fr uity, and floral notes to ripe fruit ( 52 ). For example, -ionone, which has a woody, violet and/or raspberry-like odor, is one of the main carotenoid-deri ved aroma compounds found in oranges ( 17 ). Concentration of carotenoids is fairly constant in mature fruit until the chloroplasts in the cholorophyll begin to disappear and chromoplasts surface due to ripening ( 52 ). The degradation of chloroplasts also release membrane lipids derived fr om linoleic and linolenic acids ( 53 ). A number of enzymes from the degradation process then act on these fla vor precursors to yield th e characteristic aroma compounds found in mature fruit ( 52 ). The flavedo of an orange is also rich in terpenoid hydrocarbons that make up more than 90% of total volatiles ( 17 ). This diverse class of compounds is generated via the mevalonic acid pathway. The production of terpenes is limited to the action of the HMG-CoA reductase enzyme. Studies on this enzyme report that it is important in the early stages of fruit development, with enzyme activity associating with cell division. However, HMG-CoA reductase does not seem to be active when fruit is matured ( 54 ). Therefore, it can be speculated that there would typically be more terpenes in an immature fruit than a matu re fruit. The GC-MS results, shown in Tables 4-3 to 4-8, indicated that terpene concentrations we re significantly higher in symptomatic juices compared to control, which suggests that sy mptomatic fruit may appear immature. Another indication that symptomatic fruit may appear to be immature is the significantly higher concentration of esters in cont rol juices compared to symptomatic juices. As described above,

PAGE 43

43 the precursors that esters are derived from are released when chloroplasts degrade during fruit maturation, so a ripe and mature orange whose p eel colors are orange, red, and yellow, would have a higher concentration of esters than an immature orange whose peel colors are not developed. Valencene and Maturity The concentration of valencene is the second highest terpene after limonene. It has been shown to lack aroma activity ( 55 ). However, unlike the majority of the terpenes, valencene concentrations were higher in control juices compared to symptomatic. In some cases, the difference was significant, as seen in Figure 4-13. For the signif icant differences, valencene was 50 – 67% lower in symptomatic juice compared to control. Sharon-Asa and others ( 56 ) studied the Cstps1 gene that encoded for the valencene synthase enzyme that produces valencene. They reported that gene expr ession occurred as the fruit matured, causing an accumulation of valencene. A higher concentration of valencene in control juices compared to symptomatic juices again suggests that symptomatic fruit appears to be less mature.

PAGE 44

44 Figure 4-1. Reverse phase HPLC separation of citr us flavanone glycosides of symptomatic juice from 1/30/ 2008. The flavanone glycosides found are labeled: (A) narirutin, (D) hesperidin, and (F) didymin. A rrows indicate where the bi tter flavanone glycosides (B) naringin and (E) neohespe ridin would have eluted. (C) Rhoifolin is the internal standard used in this analysis.

PAGE 45

45 A B Figure 4-2. UV spectra of eluted chromatographic peaks from Figure 4-1. A) UV spectra of narirutin (solid), hesperidin (dash), and didymin (dash-dot). B) UV spectra of the internal standard, rhoifolin.

PAGE 46

46 A B Figure 4-3. Summary of citrus flavanone glycosides. Signifi cant mean separations are represented by letters a, b, c. A) Ha mlin samples. B) Valencia samples.

PAGE 47

47 A B C Figure 4-4. HPLC chromatogram and correspo nding UV spectra of polymethoxylated flavone standards A) Reverse phase separation of (A) sinense tin, (B) nobiletin, (C) flavone (internal standard), and (D) tangeretin. B) UV spectra of si nensetin (solid), nobiletin (dash), and tangeretin (dash-dot). C) UV sp ectra of the internal standard, flavone.

PAGE 48

48 A B C Figure 4-5. HPLC chromatogram of polymethoxyl ated flavones in symptomatic juice from Valencia 4/4/2008 with UV spectra from select compounds. A) Reverse phase separation of (A) sinens etin, (B) unknown PMF, (C ) hexamethoxyflavone, (D) nobiletin, (E) scutellarein, (F) heptamet hoxyflavone, (G) flavone, and (H) TAN + coeluted unknown. B) UV spectra of HEX (s olid), SCU (dash), and HEP (dash-dot). C) UV spectra of unknown PMF

PAGE 49

49 Table 4-1. Polymethoxylated flavone concentrations ( g/mL) for Hamlin samples at various harvesting dates. Significant mean separa tions are represented by letters a, b, c. 12/12/2007 Harvesting Date. Compound Control Asymptomatic Symptomatic Sinensetin 0.289 0.00931 a 0.271 0.0153 a 0.233 0.0400 a Unknown N/A N/A N/A Hexamethoxyflavone 0.135 0.0185 a 0.131 0.0135 a 0.0366 0.00789 b Nobiletin 0.550 0.0518 a 0.510 0.0149 a 0.535 0.0268 a Tetramethyl-O-Scutellarein 0.0509 0.0046 b 0.118 0.0142 a 0.153 0.0247 a Heptamethoxyflavone 0.383 0.00353 a 0.241 0.00982 b 0.224 0.0244 b 12/18/2007 Harvesting Date Compound Control Asymptomatic Symptomatic Sinensetin 0.135 0.00348 c 0.329 0.0457 b 1.10 0.0570 a Unknown N/A N/A 0.256 0.0164 Hexamethoxyflavone N/A 0.0518 0.0140 b 0.286 0.0233 a Nobiletin 0.392 0.0118 b 0.409 0.100 b 1.50 0.0895 a Tetramethyl-O-Scutellarein 0.138 0.00230 b 0.113 0.0382 b 0.431 0.0682 a Heptamethoxyflavone 0.170 0.00299 c 0.314 0.0725 b 1.08 0.0635 a 1/30/2008 Harvesting Date Compound Control Asymptomatic Symptomatic Sinensetin 0.301 0.0265 a 0.225 0.0193 b 0.342 0.0289 a Unknown N/A N/A 0.128 0.0250 Hexamethoxyflavone 0.0996 0.0930 a 0.0471 0.00445 a 0.0881 0.0179 a Nobiletin 0.200 0.0103 b 0.222 0.0291 b 0.389 0.0437 a Tetramethyl-O-Scutellarein 0.0183 0.00262 c 0.0466 0.00679 b 0.128 0.00671 a Heptamethoxyflavone 0.259 0.0285 a 0.181 0.0150 b 0.286 0.0133 a

PAGE 50

50 Table 4-2. Polymethoxylated flavone concentrations ( g/mL) for Valencia samples at various harvesting dates. Significant mean separa tions are represented by letters a, b, c. 4/4/2008 Harvesting Date Compound Control Asymptomatic Symptomatic Sinensetin 0.594 0.0344 b 0.771 0.0126 b 1.37 0.134 a Unknown N/A N/A 0.489 0.113 Hexamethoxyflavone 0.0940 0.00873 a 0.0982 0.00702 a 0.137 0.0328 a Nobiletin 0.399 0.0320 b 0.609 0.0406 ab 1.12 0.360 a Tetramethyl-O-Scutellarein 0.138 0.0142 b 0.183 0.0243 ab 0.426 0.179 a Heptamethoxyflavone 0.192 0.0285 b 0.204 0.0116 b 0.413 0.0469 a 4/18/2008 Harvesting Date Compound Control Asymptomatic Symptomatic Sinensetin 0.161 0.00759 c 0.602 0.0362 a 0.258 0.0200 b Unknown N/A N/A 0.0216 0.000817 Hexamethoxyflavone 0.116 0.0521 ab 0.173 0.0215 a 0.0391 0.0201 b Nobiletin 1.06 0.0203 a 0.687 0.0574 b 0.351 0.0314 c Tetramethyl-O-Scutellarein 0.544 0.00415 a 0.302 0.0437 b 0.180 0.0140 c Heptamethoxyflavone 0.177 0.0137 b 0.334 0.0113 a 0.134 0.0162 c 5/23/2008 Harvesting Date Compound Control Asymptomatic Symptomatic Sinensetin 0.159 0.0284 b 0.167 0.0193 b 0.446 0.0345 a Unknown N/A N/A N/A Hexamethoxyflavone 0.0832 0.0297 ab 0.0564 0.00876 b 0.104 0.00839 a Nobiletin 0.155 0.0283 c 0.329 0.0138 b 0.452 0.0319 a Tetramethyl-O-Scutellarein 0.0996 0.0145 c 0.246 0.00705 b 0.332 0.00904 a Heptamethoxyflavone 0.108 0.0174 b 0.0564 0.00876 c 0.247 0.0255 a

PAGE 51

51 Figure 4-6. Summary of limonin anal ysis results for both Hamlin and Valencia juice samples. Bars with a weighted outline are Valencia samples. Significant mean separations are represented by letters a, b, c.

PAGE 52

52 A B Figure 4-7. Results of the sugar and acid analysis performed on the Hamlin samples. Signficiant mean separations are represented by letters a, b, c. A) Brix results. B) Brix/Acid ratio results.

PAGE 53

53 A B Figure 4-8. Results of the sugar and acid analysis performed on the Valencia samples. Signficiant mean separations are represented by letters a, b, c. A) Brix results. B) Brix/Acid ratio results.

PAGE 54

54 Figure 4-9. GC-MS chromatograms of control and symptomatic juice samples from 4/4/2008. The black outlined chromatogram from the symptomatic sample is superimposed onto the grey outlined chromatogram from the c ontrol sample. Select vol atiles are labeled: (A) acetaldehyde, (B) ethyl but anoate, (C) hexanal, (D) myrcene, (E) ethyl hexanoate, (F) -terpinene, (G) (Z)-3-hexen-1-ol, (H) li nalool, (I) valencene, and (J) benzyl alcohol, the internal standard.

PAGE 55

55 Figure 4-10. Differences of select odor-activ e volatiles in Valencia 4/18/2008 symptomatic, asymptomatic, and control juices.

PAGE 56

56 A B Figure 4-11. Summary of Va lencene concentrations ( g/mL). Significant mean separations are represented by letters a, b. A) Hamlins. B) Valencias.

PAGE 57

57 CHAPTER 5 CONCLUSION The objective of this study wa s to determine the effects of Huanglongbing on both volatile and selected non-volatile flavor compounds in Fl orida orange juices. Special attention was directed at compounds whic h may produce bitterness. Overall, there were differences between HLB symptomatic and control juices for sugars, acids, and limonin. However, ther e was little to no difference in juices from fruit which did not exhibit external HLB symptoms (asymptomatic fru it) to control juices. HLB delays maturity in symptomatic fruit. Sugars (measured as Brix) were lower and acid le vels were higher in symptomatic fruit. Esters, especially ethyl butanoate, were 29-87% lower in symptomatic juices compared to control, while some terpenes (i.e. -pinene, myrcene, -terpinene, linalool) were 791,320% higher compared to control. The con centration of valencene, a commonly accepted maturity marker, was also 50-67% lower in symptomatic juice compared to control Although symptomatic juice contained elevated levels of limonin (91-425% higher than control), no limonin level exceeded the bitterness thre shold in orange juice. Therefore, bitterness would not be detected by most people who taste the juice. Flavanone glycosides and polymethoxylated flavones were also eliminated as possible sources of bitterness. Therefore, reported off-flavor associated with HLB symptoma tic juices is not bitterness, but apparently stem from a lower concentration of sugars, a high er concentration of acid, and an imbalance of certain volatile compounds. Results from this study can be used to genera te a database on HLB or ange juice and flavor. The biochemical pathways that produce th e primary (sugars and acids) and secondary metabolites (flavanone glycosides, esters, terpenes, etc.) that were shown to be affected by HLB will have to be closely examined. The idea that sy mptomatic fruit appears to be immature can be

PAGE 58

58 explored through a comparative study between HLB symptomatic fruit and non-infected immature fruit. Differences in juice peel oi l content should be explored as this study only examined hand squeezed juices where oil levels were much lower than commercially extracted juice.

PAGE 59

59 APPENDIX: VOLATILE TABLES Table A-1. Concentration ( g/mL) of Hamlin 12/12/2007 volatiles confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.167 0.0228 a 0.232 0.0491 a 0.246 0.0638 a 901 Ethyl Acetate 0.915 0.171 a 0.453 0.0414 b 0.556 0.0387 b 912 Methanol 13.1 2.28 a 13.4 1.74 a 16.6 2.24 a 951 Ethanol 14.4 1.99 a 6.65 0.753 b 9.27 1.69 b 1002 Methyl Butanoate 0.387 0.0664 ab0.260 0.0314 b 0.404 0.0571 a 1034 -Pinene 1.45 0.178 b2.67 0.419 a 2.18 0.0470 a 1051 Ethyl Butanoate 0.191 0.0339 c 7.19 0.888 b 14.5 2.06 a 1100 Hexanal 0.0884 0.0283 c 2.27 0.318 b 4.57 0.701 a 1120 -Pinene 0.0681 0.00761 b0.190 0.0489 a 0.295 0.0596 a 1162 -3-Carene 0.115 0.0133 a 0.119 0.0124 a 0.0770 0.00993 b 1173 -Myrcene 6.39 0.683 b17.6 1.25 a 16.3 0.616 a 1195 -Terpinene 0.0747 0.00796 b0.173 0.0236 a 0.172 0.0119 a 1207 Methyl Hexanoate 0.0933 0.0182 b0.273 0.0217 ab 0.562 0.321 a 1228 Limonene 166 16.8 b439 45.7 a 385 24.5 a 1248 Ethyl Hexanoate 0.581 0.0473 c 7.77 0.874 b 11.1 0.984 a 1262 -Terpinene 0.426 0.0525 c 0.999 0.101 b 1.34 0.0324 a 1293 p -Cymene 0.236 0.200 b0.270 0.00196 b 0.548 0.107 a 1304 -Terpinolene 0.182 0.0206 b0.442 0.0198 a 0.523 0.0524 a 1311 Octanal 0.0829 0.0232 a 0.212 0.00736 a 1.15 0.744 a 1366 1-Hexanol 0.631 0.0573 b0.764 0.0783 b 2.48 0.265 a 1402 (Z)-3-Hexen-1-ol 2.48 0.235 a 1.45 0.157 b 3.10 0.353 a 1418 Nonanal 0.141 0.0955 b0.615 0.140 ab 3.76 2.39 a 1452 Ethyl Octanoate 0.0545 0.0356 c 1.45 0.134 a 1.09 0.134 b 1463 1-Octen-3-ol 0.0367 0.00690 b0.0771 0.00791 b 0.168 0.0514 a 1494 Octyl Acetate 0.115 0.0556 a 0.112 0.0122 a 0.182 0.0349 a 1525 Decanal 0.889 0.431 a 1.26 0.220 a 3.88 2.29 a 1561 Linalool 3.45 0.272 b4.43 0.645 b 6.43 0.773 a 1572 1-Octanol 0.457 0.164 b2.12 0.194 a 2.29 0.167 a 1621 -Elemene 0.162 0.0301 b0.546 0.124 a 0.599 0.0259 a 1635 4-Terpineol 2.54 0.437 b1.99 0.268 b 5.08 0.456 a 1639 -Caryophyllene 0.504 0.0651 c 0.840 0.214 b 1.21 0.0425 a 1673 1-Nonanol 0.188 0.127 ab0.0641 0.00799 b 0.371 0.160 a 1681 Citronellyl Acetate 0.556 0.0554 b0.735 0.215 ab 0.991 0.0413 a 1690 -Selinene 0.188 0.0430 a 0.0981 0.0197 b 0.1378 0.00146 ab 1697 Decyl Acetate 0.152 0.0335 a 0.143 0.0230 a 0.152 0.0218 a 1706 Ethyl-3-Hydroxyhexanoate 0.490 0.0422 c 0.669 0.0544 b 1.37 0.0658 a 1726 -Terpineol 41.1 2.90 a 0.567 0.221 b 0.439 0.114 b 1766 Valencene 20.2 3.04 a 19.6 3.69 a 26.5 0.593 a

PAGE 60

60 Table A-1. Continued. LRI Compound Control Asymptomatic Symptomatic 1770 -Selinene 1.73 0.261 b1.91 0.293 b 2.55 0.183 a 1779 -Citronellol 0.496 0.0792 c 0.942 0.0824 b 1.23 0.0660 a 1785 Carvone 0.0394 0.0244 b0.400 0.0580 a 0.437 0.0222 a 1791 -Cadinene 1.02 0.203 a 1.25 0.355 a 0.839 0.0112 a 1820 Nerol 0.0636 0.0108 b0.196 0.0240 a 0.207 0.00991 a 1841 Perillaldehyde 0.0547 0.0168 b0.261 0.0300 a 0.223 0.0181 a 1863 Geraniol 0.817 0.0682 a 0.534 0.0345 b 0.627 0.00937 b 1894 -Ionone 0.141 0.0202 a 0.137 0.0394 a 0.204 0.0196 a 1986 -Ionone 0.0533 0.0146 a 0.0657 0.00562 a 0.100 0.0326 a 2050 Caryophyllene Oxide 0.0463 0.00542 b0.0702 0.00607 a 0.0786 0.00490 a 2272 -Sinensal 0.0972 0.0364 a 0.110 0.0368 a 0.0911 0.0323 a 2576 Nootkatone 0.707 0.0157 a 0.580 0.138 a 0.769 0.692 a

PAGE 61

61 Table A-2. Concentration ( g/mL) of Hamlin 12/18/2007 volatiles confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.312 0.0294 a 0.330 0.0102 a 0.179 0.109 a 901 Ethyl Acetate 0.286 0.0435 a 0.283 0.0582 a 0.144 0.0286 b 912 Methanol 14.7 1.57 a 15.7 1.24 a 18.4 1.72 a 951 Ethanol 5.02 0.534 a 5.49 0.408 a 1.47 0.231 b 1002 Methyl Butanoate 0.566 0.137 a 0.483 0.0197 a 0.377 0.0624 a 1034 -Pinene 1.58 0.126 b 3.44 0.0556 a 4.35 0.724 a 1051 Ethyl Butanoate 16.8 4.11 a 9.48 1.90 b 2.10 0.176 c 1100 Hexanal 5.82 1.20 b 5.98 0.402 b 12.4 0.913 a 1120 -Pinene 0.184 0.0420 b 0.411 0.0966 b 1.32 0.125 a 1162 -3-Carene 0.0990 0.00223 b 0.111 0.0122 b 0.134 0.00513 a 1173 -Myrcene 9.18 0.638 b 21.1 1.77 a 22.8 1.48 a 1195 -Terpinene 0.423 0.516 a 0.125 0.0203 a 0.201 0.0398 a 1207 Methyl Hexanoate 0.955 0.316 a 0.548 0.149 b 0.380 0.162 b 1228 Limonene 237 5.92 b 605 32.8 a 620 72.4 a 1248 Ethyl Hexanoate 7.56 0.897 a 8.88 0.843 a 1.89 0.122 b 1262 -Terpinene 0.405 0.0488 b 1.01 0.0621 a 1.24 0.248 a 1293 p -Cymene 0.182 0.0218 a 0.232 0.0204 a 0.238 0.0718 a 1304 -Terpinolene 0.174 0.0178 b 0.628 0.0980 a 0.586 0.110 a 1311 Octanal 0.464 0.0322 a 0.296 0.0616 b 0.426 0.0782 ab 1366 1-Hexanol 1.71 0.179 b 2.37 0.232 b 3.32 0.359 a 1402 (Z)-3-Hexen-1-ol 1.94 0.221 b 2.68 0.208 a 2.90 0.285 a 1418 Nonanal 0.597 0.137 b 0.767 0.0781 ab 1.02 0.142 a 1452 Ethyl Octanoate 0.387 0.0695 b 1.05 0.0782 a 0.0937 0.0168 c 1463 1-Octen-3-ol 0.235 0.0210 a 0.133 0.0158 b 0.134 0.0197 b 1494 Octyl Acetate 0.136 0.0295 c 0.524 0.0410 b 0.941 0.0888 a 1525 Decanal 0.752 0.0704 b 0.784 0.353 ab 1.45 0.312 a 1561 Linalool 4.75 0.776 a 2.37 0.268 b 4.81 0.629 a 1572 1-Octanol 3.62 0.495 a 3.22 0.205 a 3.49 0.324 a 1621 -Elemene 0.541 0.0418 a 0.670 0.302 a 0.320 0.162 a 1635 4-Terpineol 2.44 0.459 b 2.58 0.263 b 4.37 0.802 a 1639 -Caryophyllene 0.778 0.0744 a 0.757 0.216 a 0.589 0.155 a 1673 1-Nonanol 0.355 0.533 a 0.416 0.0253 a 0.460 0.0536 a 1681 Citronellyl Acetate 0.797 0.0267 a 0.876 0.217 a 1.14 0.296 a 1690 -Selinene 0.111 0.0128 a 0.1194 0.0368 a 0.0744 0.0450 a 1697 Decyl Acetate 0.181 0.0329 a 0.287 0.0649 a 0.284 0.0423 a 1706 Ethyl-3-Hydroxyhexanoate 1.22 0.254 a 1.07 0.607 a 0.161 0.00410 b 1726 -Terpineol 0.408 0.0617 b 0.372 0.0273 b 0.778 0.0772 a 1766 Valencene 24.3 2.42 a 29.9 9.53 a 15.9 5.74 a 1770 -Selinene 2.18 0.241 a 3.06 1.22 a 1.57 0.608 a 1779 -Citronellol 0.974 0.149 a 1.21 0.175 a 0.891 0.118 a

PAGE 62

62 Table A-2. Continued. LRI Compound Control Asymptomatic Symptomatic 1785 Carvone 1.02 0.230 a 0.860 0.0690 a 0.357 0.0445 b 1791 -Cadinene 0.651 0.159 a 0.786 0.149 a 0.940 0.332 a 1820 Nerol 0.229 0.0352 a 0.234 0.0175 a 0.225 0.0198 a 1841 Perillaldehyde 0.274 0.0277 b 0.339 0.0168 b 0.496 0.101 a 1863 Geraniol 0.624 0.0767 a 0.722 0.0441 a 0.671 0.00685 a 1894 -Ionone 0.132 0.0150 b 0.143 0.0236 b 0.407 0.125 a 1986 -Ionone 0.0734 0.00839 a 0.0826 0.0152 a 0.0570 0.0227 a 2050 Caryophyllene Oxide 0.0788 0.00440 b 0.131 0.00882 a 0.0751 0.0186 b 2272 -Sinensal 0.119 0.00411 a 0.128 0.0448 a 0.0911 0.0331 a 2576 Nootkatone 0.708 0.117 ab 0.908 0.123 a 0.474 0.214 b

PAGE 63

63 Table A-3. Concentration ( g/mL) of Hamlin 1/30/2008 volatil es confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.299 0.0565 a 0.415 0.0774 a 0.309 0.0773 a 901 Ethyl Acetate 0.176 0.0123 b 0.397 0.0509 a 0.363 0.0780 a 912 Methanol 12.2 0.891 a 13.7 1.51 a 14.8 1.64 a 951 Ethanol 3.48 0.340 c 10.1 1.30 a 6.32 0.952 b 1002 Methyl Butanoate 0.409 0.0219 a 0.506 0.0889 a 0.376 0.0138 a 1034 -Pinene 0.879 0.0668 c 2.12 0.461 b 3.05 0.200 a 1051 Ethyl Butanoate 9.61 0.0849 b 16.4 1.94 a 4.20 0.611 c 1100 Hexanal 8.27 0.155 a 3.87 0.429 b 2.89 0.400 c 1120 -Pinene 0.400 0.0767 a 0.202 0.0310 b 0.222 0.0639 b 1162 -3-Carene 0.271 0.0211 b 0.615 0.972 a 0.383 0.0128 b 1173 -Myrcene 5.78 1.13 b 16.5 1.31 a 17.8 0.286 a 1195 -Terpinene 0.0640 0.00591 b 0.165 0.0392 a 0.154 0.0116 a 1207 Methyl Hexanoate 1.67 0.3112 a 0.483 0.177 b 0.366 0.0831 b 1228 Limonene 164 8.15 c 399 74.2 b 527 44.0 a 1248 Ethyl Hexanoate 6.52 0.699 b 9.77 1.47 a 0.123 0.0215 c 1262 -Terpinene 0.162 0.0158 c 0.782 0.238 b 1.41 0.144 a 1293 p -Cymene 0.209 0.0262 b 0.319 0.618 a 0.209 0.0214 b 1304 -Terpinolene 0.127 0.0221 b 0.455 0.113 a 0.611 0.0646 a 1311 Octanal 0.807 0.126 a 0.987 0.132 a 0.441 0.131 b 1366 1-Hexanol 3.44 0.134 a 2.338 0.252 b 2.51 0.568 b 1402 (Z)-3-Hexen-1-ol 4.64 0.108 a 2.80 0.281 b 1.36 0.128 c 1418 Nonanal 2.46 0.273 a 1.15 0.158 b 1.44 0.139 b 1452 Ethyl Octanoate 0.260 0.0458 b 1.252 0.279 a 0.395 0.0160 b 1463 1-Octen-3-ol 0.442 0.0346 a 0.235 0.0297 b 0.127 0.0169 c 1494 Octyl Acetate 0.205 0.010 b 0.150 0.0476 b 0.765 0.0406 a 1525 Decanal 0.676 0.205 a 1.14 0.158 a 1.65 0.759 a 1561 Linalool 5.81 0.378 c 10.1 1.05 b 17.1 2.18 a 1572 1-Octanol 2.80 0.233 b 4.11 0.385 a 3.85 0.455 a 1621 -Elemene 1.486 0.272 a 1.71 0.376 a 0.987 0.286 b 1635 4-Terpineol 1.31 0.235 b 2.02 0.221 b 3.37 0.461 a 1639 -Caryophyllene 1.32 0.228 a 1.51 0.419 a 1.34 0.394 a 1673 1-Nonanol 0.396 0.0413 a 0.365 0.0374 a 0.351 0.0236 a 1681 Citronellyl Acetate 2.36 0.244 a 2.25 0.605 a 0.551 0.0310 b 1690 -Selinene 0.405 0.0492 b 0.361 0.109 b 0.986 0.293 a 1697 Decyl Acetate 0.331 0.0171 b 0.330 0.0366 b 0.812 0.0116 a 1706 Ethyl-3-Hydroxyhexanoate 1.07 0.0978 b 2.27 0.0793 a 0.511 0.0384 c 1726 -Terpineol 0.503 0.0667 b 0.626 0.0532 ab 0.708 0.0753 a 1766 Valencene 77.34 9.56 a 72.9 17.0 a 38.5 8.86 b 1770 -Selinene 8.12 1.19 a 7.73 2.09 a 3.70 0.979 b 1779 -Citronellol 3.97 0.625 a 3.70 1.50 a 1.57 0.318 a

PAGE 64

64 Table A-3. Continued. LRI Compound Control Asymptomatic Symptomatic 1785 Carvone 1.28 0.149 a 0.890 0.0986 b 0.339 0.111 c 1791 -Cadinene 0.896 0.222 a 0.866 0.241 a 0.983 0.195 a 1820 Nerol 0.191 0.0282 b 0.246 0.0327 ab 0.296 0.0299 a 1841 Perillaldehyde 0.310 0.0112 b 0.285 0.0469 b 0.6801 0.0633 a 1863 Geraniol 0.974 0.110 ab 1.080 0.808 a 0.850 0.243 b 1894 -Ionone 0.365 0.0296 a 0.449 0.0618 a 0.480 0.0549 a 1986 -Ionone 0.0931 0.0140 a 0.0873 0.0103 a 0.0670 0.0159 a 2050 Caryophyllene Oxide 0.238 0.0336 a 0.291 0.104 a 0.229 0.0182 a 2272 -Sinensal 0.106 0.00442 a 0.110 0.0324 a 0.131 0.0133 a 2576 Nootkatone 0.672 0.175 a 0.781 0.305 a 1.22 0.335 a

PAGE 65

65 Table A-4. Concentration ( g/mL) of Valencia 4/4/2008 volat iles confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.329 0.0529 a 0.349 0.449 a 0.224 0.100 a 901 Ethyl Acetate 0.594 0.0387 a 0.548 0.0400 a 0.210 0.0316 b 912 Methanol 13.4 1.10 a 14.9 1.27 a 14.8 0.846 a 951 Ethanol 12.6 0.212 a 12.7 0.706 a 10.3 0.868 b 1002 Methyl Butanoate 0.392 0.340 a 0.361 0.0329 a 0.393 0.0516 a 1034 -Pinene 2.44 0.428 a 3.36 0.859 a 2.66 0.0230 a 1051 Ethyl Butanoate 13.2 0.328 a 12.4 0.779 a 5.15 0.454 b 1100 Hexanal 4.45 0.278 a 3.06 0.4111 b 1.35 0.1888 c 1120 -Pinene 0.283 0.0301 a 0.278 0.0544 a 0.246 0.0185 a 1162 -3-Carene 0.196 0.0131 a 0.110 0.00788 b 0.139 0.0180 b 1173 -Myrcene 15.8 1.36 a 17.9 1.24 a 16.9 0.151 a 1195 -Terpinene 0.119 0.00883 a 0.204 0.0321 a 0.254 0.165 a 1207 Methyl Hexanoate 0.641 0.487 a 0.887 0.487 a 0.123 0.0354 a 1228 Limonene 514 40.7 b 621 57.0 a 433 8.16 b 1248 Ethyl Hexanoate 11.2 0.802 a 11.7 0.821 a 3.09 0.570 b 1262 -Terpinene 0.792 0.113 c 2.49 0.272 b 3.16 0.0706 a 1293 p -Cymene 0.367 0.0713 b 0.581 0.0471 a 0.422 0.0399 b 1304 -Terpinolene 0.535 0.0569 b 0.814 0.0639 a 0.629 0.0189 b 1311 Octanal 1.29 0.0580 a 0.697 0.0563 b 0.281 0.0394 c 1366 1-Hexanol 1.33 0.0742 a 1.30 0.0229 ab 1.13 0.116 b 1402 (Z)-3-Hexen-1-ol 1.44 0.103 a 1.44 0.0889 a 1.48 0.108 a 1418 Nonanal 0.624 0.0662 ab 0.507 0.0286 b 0.782 0.164 a 1452 Ethyl Octanoate 2.04 0.165 b 2.72 0.189 a 0.590 0.0378 c 1463 1-Octen-3-ol 0.166 0.00375 a 0.129 0.00777 b 0.0888 0.0167 c 1494 Octyl Acetate 2.31 0.149 a 2.52 0.265 a 1.12 0.0967 b 1525 Decanal 2.10 0.307 a 1.71 0.694 ab 0.989 0.113 b 1561 Linalool 9.15 0.685 b 7.37 0.430 b 19.2 1.49 a 1572 1-Octanol 8.41 0.408 a 6.10 0.231 b 3.82 0.207 c 1621 -Elemene 1.34 0.428 a 1.21 0.396 a 0.762 0.0664 a 1635 4-Terpineol 1.96 0.465 c 4.02 0.261 b 6.17 0.544 a 1639 -Caryophyllene 1.26 0.189 a 1.08 0.227 a 0.616 0.0332 b 1673 1-Nonanol 0.871 0.0379 a 0.658 0.0452 b 0.336 0.0180 c 1681 Citronellyl Acetate 1.10 0.131 b 3.55 0.624 a 1.98 0.136 b 1690 -Selinene 0.312 0.113 a 0.256 0.0673 ab 0.102 0.0136 b 1697 Decyl Acetate 0.864 0.135 b 1.845 0.544 a 0.754 0.218 b 1706 Ethyl-3-Hydroxyhexanoate 1.46 0.080 a 1.08 0.0921 b 0.222 0.0183 c 1726 -Terpineol 0.571 0.0169 b 0.543 0.0643 b 1.11 0.0880 a 1766 Valencene 62.8 16.0 a 56.4 10.2 a 20.4 0.900 b 1770 -Selinene 7.02 2.61 a 6.09 1.84 a 2.68 0.351 a 1779 -Citronellol 3.41 1.39 a 2.22 0.358 a 1.84 0.868 a

PAGE 66

66 Table A-4. Continued. LRI Compound Control Asymptomatic Symptomatic 1785 Carvone 0.994 0.105 a 0.878 0.172 a 0.808 0.0536 a 1791 -Cadinene 0.441 0.262 a 0.890 0.639 a 0.473 0.0690 a 1820 Nerol 0.588 0.0343 a 0.444 0.0379 b 0.578 0.0413 a 1841 Perillaldehyde 0.666 0.0160 a 0.704 0.0415 a 0.681 0.0672 a 1863 Geraniol 1.21 0.0137 a 1.17 0.0888 a 0.751 0.0376 b 1894 -Ionone 0.443 0.01061 a 0.481 0.0566 a 0.295 0.126 a 1986 -Ionone 0.114 0.00517 a 0.126 0.00993 a 0.0549 0.00460 b 2050 Caryophyllene Oxide 0.243 0.00969 a 0.244 0.0151 a 0.0802 0.00127 b 2272 -Sinensal 0.223 0.0249 a 0.243 0.0174 a 0.298 0.0702 a 2576 Nootkatone 1.47 0.0488 a 1.69 0.184 a 0.639 0.308 b

PAGE 67

67 Table A-5. Concentration ( g/mL) of Valencia 4/18/2008 vola tiles confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.545 0.0610 a 0.452 0.056 a 0.339 0.123 a 901 Ethyl Acetate 1.14 0.110 a 1.10 0.0301 a 0.758 0.429 a 912 Methanol 14.8 1.30 a 15.8 0.305 a 17.8 2.59 a 951 Ethanol 16.9 1.94 a 18.1 1.64 a 17.6 4.11 a 1002 Methyl Butanoate 0.538 0.0498 b 0.532 0.0151 b 0.770 0.112 a 1034 -Pinene 2.02 0.113 b 4.51 0.215 a 4.53 1.02 a 1051 Ethyl Butanoate 24.2 2.65 a 14.4 1.11 b 6.19 0.599 c 1100 Hexanal 2.65 0.650 b 6.30 0.617 a 3.40 0.598 b 1120 -Pinene 0.177 0.0351 b 0.503 0.0141 a 0.448 0.0714 a 1162 -3-Carene 0.176 0.0139 c 0.335 0.0107 b 0.507 0.0815 a 1173 -Myrcene 13.4 0.761 b 19.7 0.666 a 22.3 2.80 a 1195 -Terpinene 0.109 0.00679 b 0.202 0.0133 a 0.225 0.0209 a 1207 Methyl Hexanoate 0.260 0.0736 a 0.247 0.0753 a 0.712 0.306 a 1228 Limonene 376 31.0 b 741 32.1 a 674 107 a 1248 Ethyl Hexanoate 8.69 0.837 a 5.10 0.727 b 2.89 0.303 c 1262 -Terpinene 0.500 0.0293 b 0.120 0.0454 b 7.10 1.29 a 1293 p -Cymene 0.250 0.0205 b 0.543 0.00733 ab 0.661 0.267 a 1304 -Terpinolene 0.336 0.0150 b 1.11 0.937 a 1.29 0.326 a 1311 Octanal 0.795 0.140 c 10.9 0.874 a 5.40 0.737 b 1366 1-Hexanol 1.22 0.141 a 1.14 0.264 a 0.869 0.122 a 1402 (Z)-3-Hexen-1-ol 0.856 0.963 b 0.804 0.0631 b 1.23 0.184 a 1418 Nonanal 0.697 0.136 b 2.89 0.0486 a 2.33 1.19 ab 1452 Ethyl Octanoate 1.18 0.0299 a 1.30 0.0894 a 0.511 0.0885 b 1463 1-Octen-3-ol 0.165 0.0103 a 0.158 0.0101 a 0.0972 0.0136 b 1494 Octyl Acetate 0.342 0.0239 c 1.64 0.0120 a 0.976 0.154 b 1525 Decanal 1.40 0.297 c 12.6 0.209 a 5.68 1.68 b 1561 Linalool 11.6 1.59 b 19.4 3.07 b 35.9 4.39 a 1572 1-Octanol 5.03 0.372 b 9.23 1.13 a 9.14 0.947 a 1621 -Elemene 0.774 0.0626 ab 0.998 0.130 a 0.678 0.157 b 1635 4-Terpineol 1.75 0.232 c 7.27 1.08 b 11.0 1.24 a 1639 -Caryophyllene 0.888 0.0909 ab1.15 0.277 a 0.608 0.146 b 1673 1-Nonanol 0.359 0.01530 b 0.971 0.0986 a 1.17 0.167 a 1681 Citronellyl Acetate 0.606 0.0172 b 1.77 0.0651 a 1.63 0.204 a 1690 -Selinene 0.220 0.0233 a 0.205 0.0523 a 0.0848 0.00970 b 1697 Decyl Acetate 0.404 0.0177 c 0.970 0.0397 a 0.597 0.0829 b 1706 Ethyl-3-Hydroxyhexanoate 1.37 0.0780 a 1.33 0.115 a 0.309 0.0168 b 1726 -Terpineol 0.611 0.0461 c 1.23 0.0692 b 2.20 0.279 a 1766 Valencene 45.0 3.13 a 54.4 9.18 a 18.7 3.24 b 1770 -Selinene 4.42 0.339 a 5.34 0.913 a 2.62 0.384 b 1779 -Citronellol 1.96 0.0604 a 2.46 0.379 a 2.66 0.371 a

PAGE 68

68 Table A-5. Continued. LRI Compound Control Asymptomatic Symptomatic 1785 Carvone 0.662 0.0586 b 1.51 0.186 a 0.460 0.0567 b 1791 -Cadinene 0.418 0.0665 a 0.929 0.302 a 0.930 0.200 a 1820 Nerol 0.312 0.0166 b 0.535 0.0666 a 0.649 0.0831 a 1841 Perillaldehyde 0.609 0.0768 c 1.31 0.119 b 1.93 0.151 a 1863 Geraniol 0.839 0.0710 a 0.736 0.0499 a 0.790 0.102 a 1894 -Ionone 0.294 0.0220 c 0.424 0.00457 b 0.559 0.0690 a 1986 -Ionone 0.0971 0.0104 b 0.124 0.00672 a 0.0533 0.00632 c 2050 Caryophyllene Oxide 0.138 0.00553 c 0.219 0.00933 b 0.249 0.0153 a 2272 -Sinensal 0.161 0.0167 b 0.288 0.00990 a 0.352 0.0824 a 2576 Nootkatone 0.995 0.0563 ab 1.35 0.129 a 0.577 0.296 b

PAGE 69

69 Table A-6. Concentration ( g/mL) of Valencia 5/23/2008 vola tiles confirmed by MS and LRI values. Significant mean separati ons are represented by a, b, c. LRI Compound Control Asymptomatic Symptomatic 721 Acetaldehyde 0.655 0.137 a 0.793 0.132 a 0.750 0.282 a 901 Ethyl Acetate 1.67 0.0774 a 1.84 0.0730 a 2.44 0.709 a 912 Methanol 14.7 0.561 a 15.2 1.04 a 16.9 3.25 a 951 Ethanol 19.0 0.480 a 18.7 2.08 a 21.8 6.37 a 1002 Methyl Butanoate 0.655 0.0179 a 0.530 0.00521 a 0.713 0.182 a 1034 -Pinene 1.63 0.232 b 2.61 0.497 a 1.70 0.287 b 1051 Ethyl Butanoate 25.1 0.248 a 18.8 0.284 a 19.9 5.74 a 1100 Hexanal 8.68 0.113 a 9.14 0.293 a 8.31 2.44 a 1120 -Pinene 0.374 0.0100 a 0.552 0.173 a 0.522 0.294 a 1162 -3-Carene 0.164 0.0200 b 0.239 0.0366 b 0.332 0.0487 a 1173 -Myrcene 12.3 0.475 a 17.5 2.08 a 14.8 3.56 a 1195 -Terpinene 0.159 0.0148 b 0.144 0.0508 b 0.340 0.0513 a 1207 Methyl Hexanoate 1.11 0.0331 a 0.824 0.407 a 0.839 0.172 a 1228 Limonene 305 33.5 b 464 51.3 a 325 55.5 b 1248 Ethyl Hexanoate 11.3 0.951 a 11.5 0.600 a 7.94 1.76 b 1262 -Terpinene 0.118 0.00664 b 0.232 0.00360 b 1.67 0.276 a 1293 p -Cymene 0.309 0.0571 a 0.432 0.0581 a 0.531 0.180 a 1304 -Terpinolene 0.300 0.0389 b 0.534 0.0570 a 0.489 0.126 ab 1311 Octanal 2.88 0.230 a 0.876 0.0255 b 1.12 0.304 b 1366 1-Hexanol 1.11 0.0557 b 2.32 0.883 a 1.74 0.483 ab 1402 (Z)-3-Hexen-1-ol 0.679 0.0260 b 1.27 0.0246 a 1.07 0.279 ab 1418 Nonanal 0.927 0.231 a 0.645 0.0503 a 1.07 0.461 a 1452 Ethyl Octanoate 0.823 0.117 b 1.12 0.0499 a 0.486 0.0670 c 1463 1-Octen-3-ol 0.239 0.00864 a 0.271 0.00890 a 0.154 0.0342 b 1494 Octyl Acetate 0.599 0.576 a 0.241 0.0368 b 0.186 0.0756 b 1525 Decanal 2.00 0.267 a 1.29 0.468 a 1.45 0.331 a 1561 Linalool 14.0 0.219 b 13.6 0.283 b 26.7 6.63 a 1572 1-Octanol 6.59 0.105 a 4.38 0.0514 b 4.96 0.780 b 1621 -Elemene 1.06 0.285 a 1.73 0.341 a 1.10 0.188 a 1635 4-Terpineol 1.48 0.0353 b 3.27 0.0222 b 5.77 1.36 a 1639 -Caryophyllene 0.977 0.327 a 1.39 0.360 a 0.784 0.114 a 1673 1-Nonanol 0.467 0.329 a 0.384 0.0128 a 0.398 0.0626 a 1681 Citronellyl Acetate 0.317 0.405 b 1.13 0.0777 a 0.426 0.135 b 1690 -Selinene 0.237 0.0777 a 0.304 0.0638 a 0.189 0.0233 a 1697 Decyl Acetate 0.203 0.0246 b 0.443 0.0595 a 0.208 0.0317 b 1706 Ethyl-3-Hydroxyhexanoate 1.67 0.0229 b 1.97 0.0164 a 0.929 0.114 b 1726 -Terpineol 0.450 0.0806 b 0.730 0.0325 a 0.813 0.168 a 1766 Valencene 50.5 13.8 a 64.6 10.9 a 38.8 5.03 a 1770 -Selinene 5.43 1.68 bc 7.63 1.65 ac 4.03 0.403 b 1779 -Citronellol 1.70 0.490 b 3.67 0.959 a 2.10 0.297 ab

PAGE 70

70 Table A-6. Continued. LRI Compound Control Asymptomatic Symptomatic 1785 Carvone 1.25 0.0931 a 1.25 0.0355 a 0.669 0.0389 b 1791 -Cadinene 0.252 0.154 a 0.460 0.131 a 0.295 0.0223 a 1820 Nerol 0.273 0.00644 b 0.451 0.0231 a 0.420 0.0486 a 1841 Perillaldehyde 0.476 0.0269 b 0.513 0.0255 b 0.654 0.0742 a 1863 Geraniol 0.733 0.0655 a 0.164 0.0114 b 0.739 0.0944 a 1894 -Ionone 0.244 0.0150 b 0.386 0.0412 a 0.395 0.0396 a 1986 -Ionone 0.0724 0.00580 b 0.124 0.00592 a 0.0670 0.00230 b 2050 Caryophyllene Oxide 0.224 0.0137 b 0.339 0.0158 a 0.114 0.0217 c 2272 -Sinensal 0.153 0.0414 a 0.186 0.0103 a 0.159 0.00672 a 2576 Nootkatone 0.819 0.0557 ab 1.15 0.0682 a 0.606 0.220 b

PAGE 71

71 REFERENCES 1. McClean, A. P. D.; Oberholzer, P. C. J., Gr eening disease of the sweet orange: evidence that it is caused by a transmisable virus. S. Afr. J. Agric. Sci. 1965, 8, 253-276. 2. McClean, A. P. D.; Schwarz, R. E., Green ing or blotchy-mottle disease of citrus. Phytophylactica 1970, 2, 177-194. 3. Kapur, S. P.; Kapoor, S. K.; Cheema, S. S.; Dhillon, R. S., Effect of greening disease on tree and fruit character s of kinnow mandarin. The Punjab Hort. J. 1978, 18, 176-179. 4. USDA; NASS, Citrus Summary 2007-08. In Service, F. A. S., Ed. 2009; p 1. 5. Anon, The orange book Tetra Pak Processing System s, AB: Lund, Sweden, 1997; p 180. 6. Bove, J. M., Huanglongbing: a destructive, ne wly-emerging, century-old disease of citrus. J. Plant Path. 2006, 88, (1), 7-37. 7. USDA National Invasive Species Information Center Microbe s: Species Profiles: Citrus Greening. 2009. http://www.invasivespeciesinf o.gov/microbes/citrusgreen.shtml Date last accessed: April 2009. 8. Yates, J. D.; Dewdney, M. M. Citrus Greening (Huanglongbing) Florida History. 2008. CREC, Lake Alfred, FL. http://www.crec.ifas.ufl.edu/ex tension/greening/history.htm Date last accessed: March 2009. 9. Halbert, S. E.; Manjunath, K. L., Asian ci trus psyllids (Sternorr hyncha : Psyllidae) and greening disease of citrus: A literature revi ew and assessment of risk in Florida. Florida Entomologist 2004, 87, (3), 330-353. 10. Morris, A.; Muraro, R., Economic evaluation of citrus greening management and control strategies. In IFAS, Ed. EDIS: Gainesville, 2008; Vol. FE712, pp 2-3. 11. Teixeira, D. D.; Saillard, C.; Eveillard, S.; Danet, J. L.; da Costa, P. I.; Ayres, A. J.; Bove, J., 'Candidatus Liberibacter americanus', a ssociated with citrus huanglongbing (greening disease) in Sao Paulo State, Brazil. Int. J. Sys. Evol. Micro. 2005, 55, 1857-1862. 12. Yates, J. D.; Futch, S. H.; Spann, T. M., Sc outing for Citrus Greeni ng. In IFAS, Ed. EDIS: Gainesville, 2008; Vol. HS1147, pp 1-2. 13. da Graca, J. V., Citrus Greening Disease. Ann. Rev. Phytopath. 1991, 29, 109-136. 14. Brlansky, R. H.; Chung, K. R.; Rogers, M. E., 2008 Florida Citrus Pest Management Guide: Huanglongbing (Citrus Greening). In IF AS, Ed. EDIS: Gainesville, 2007; Vol. PP225, pp 1-2.

PAGE 72

72 15. Fisher, C.; Scott, T. R., Food Flavours Biology and Chemistry The Royal Society of Chemistry: Cambridge, 1997; p 165. 16. Belitz, H. D.; Grosch, W.; Schieberle, P., Food Chemistry 3rd ed.; Springer: Berlin, 2004; p 1071. 17. Perez-Cacho, P. R.; Rouseff, R. L., Fr esh squeezed orange juice odor: A review. Crit. Rev. Food Sci. Nutr. 2008, 48, (7), 681-695. 18. Reineccius, G., Flavor Chemistry and Technology 2nd ed.; Taylor and Francis Group, LLC: 2006; p 489. 19. Nisperos-Carriedo, M. O.; Shaw, P. E., Comp arison of volatile flavor components in fresh and processed orange juices. J. Agric. Food Chem. 1990, 38, (4), 1048-1052. 20. Gattuso, G.; Barreca, D.; Gargiulli, C.; Leu zzi, U.; Caristi, C., Flavonoid composition of citrus juices. Molecules 2007, 12, (8), 1641-1673. 21. Manners, G. D., Citrus limonoids: analysis bioactivity, and biomedical prospects. J. Agric. Food Chem. 2007, 55, (21), 8285-8294. 22. Peterson, J. J.; Dwyer, J. T.; Beecher, G. R.; Bhagwat, S. A.; Gebhardt, S. E.; Haytowitz, D. B.; Holden, J. M., Flavanones in oranges, ta ngerines (mandarins), tangors, and tangelos: a compilation and review of the data from the analytical literature. J. Food Comp. Anal. 2006, 19, S66-S73. 23. Swift, L. J., Peel juice flavor proximate an alyses of Florida orange peel juice extract for 1962-63 and 1963-64 seasons. J. Agric. Food Chem. 1965, 13, (3), 282-284. 24. Swift, L. J., Flavones of neutral fraction of benzene extractables of an orange peel juice J. Agric. Food Chem. 1965, 13, (5), 431-433. 25. Kryger, R. A., Role of polymethoxylat ed flavones in citrus flavor. In Natural Flavors and Fragrances: Chemistry, Analysis, and Production Frey, C.; Rouseff, R., Eds. American Chemical Society: Washington, DC, 2005; pp 161-172. 26. Ooghe, W. C.; Ooghe, S. J.; Detavernier, C. M.; Huyghebaert, A., Characterization of orange juice (citrus-sinensis ) by polymethoxylated flavones. J. Agric. Food Chem. 1994, 42, (10), 2191-2195. 27. Veldhuis, M. K.; Swift, L. J.; Scott, W. C., Fully-methoxylated flavones in Florida orange juices. J. Agric. Food Chem. 1970, 18, (4), 590-592. 28. Braddock, R. J., Handbook of Citrus By-Produc ts and Processing Technology John Wiley & Sons, Inc.: New York, 1999; p 247. 29. Maier, V. P.; Hasegawa, S.; Hera, E., Li monin d-ring-lactone hydrolase a new enzyme from citrus seeds. Phytochemistry 1969, 8, (2), 405-407.

PAGE 73

73 30. Hasegawa, S.; Ou, P.; Fong, C. H.; Herman, Z. ; Coggins, C. W.; Atkin, D. R., Changes in the limonoate a-ring lactone and limonin 17beta-d-glucopyranoside content of navel oranges during fruit-growth and maturation. J. Agric. Food Chem. 1991, 39, (2), 262-265. 31. Hasegawa, S.; Bennett, R. D.; Herman, Z.; Fong, C. H.; Ou, P., Limonoid glucosides in citrus. Phytochemistry 1989, 28, (6), 1717-1720. 32. Supelco, Guide to Solid Phase Extracti on. In 910 ed.; Sigma-Aldrich Co.: 1998. 33. Arthur, C. L.; Pawliszyn, J., Solid-phase microextraction with th ermal-desorption using fused-silica optical fibers. Anal. Chem. 1990, 62, (19), 2145-2148. 34. de Rijke, E.; Out, P.; Niessen, W. M. A.; Ariese, F.; Gooijer, C.; Brinkman, U. A. T., Analytical separation and dete ction methods for flavonoids. J. Chromatogr. A 2006, 1112, (1-2), 31-63. 35. Steffen, A.; Pawliszyn, J., Analysis of fl avor volatiles using headspace solid-phase microextraction. J. Agric. Food Chem. 1996, 44, (8), 2187-2193. 36. Bazemore, R.; Goodner, K.; Rouseff, R. In Volatiles from unpasteurized and excessively heated orange juice analyzed with solid phase microextraction and GC-olfactometry 1999; Inst Food Technologists: 1999; pp 800-803. 37. Mahattanatawee, K.; Rouseff, R.; Valim, M. F.; Naim, M., Identification and aroma impact of norisoprenoids in orange juice. J. Agric. Food Chem. 2005, 53, (2), 393-397. 38. Rouseff, R. L.; Martin, S. F.; Youtsey, C. O ., Quantitative survey of narirutin, naringin, hesperidin, and neohesperidin in citrus. J. Agric. Food Chem. 1987, 35, (6), 1027-1030. 39. Bronner, W. E.; Beecher, G. R., Extraction and measurement of prominent flavonoids in orange and grapefruit juice concentrates. J. Chromatogr. A 1995, 705, (2), 247-256. 40. Mouly, P. P.; Gaydou, E. M.; Arzouyan, C., Se paration and quantitation of orange juices using liquid chromatography of polymethoxylated flavones. Analusis 1999, 27, (3), 284288. 41. Widmer, W. W.; Haun, C. A., Analysis of limonin and flavonoids in citrus juices and byproduct extracts by direct injectio n and in-line sample clean-up. In Citrus Limonoids: Functional Chemicals in Agriculture and Food Berhow, M. A.; Hasegawa, S.; Manners, G. D., Eds. American Chem ical Society: 2000; pp 60-72. 42. Cheng, G., Procedures for Analysis of Citrus Products. In 4th ed.; FMC Technologies, Inc. FMC FoodTech, Citrus Syst ems: Lakeland, 2002; p 192. 43. Gil-Izquierdo, A.; Gil, M. I. ; Tomas-Barberan, F. A.; Ferreres, F., Influence of industrial processing on orange juice flavanone solubi lity and transformati on to chalcones under gastrointestinal conditions. J. Agric. Food Chem. 2003, 51, (10), 3024-3028.

PAGE 74

74 44. Rouseff, R. L.; Ting, S. V., Quantitation of polymethoxylated flavones in orange juice by high-performance liquid-chromatography. J. Chromatogr. 1979, 176, (1), 75-87. 45. Stanley, W. L.; Jurd, L., Citrus Coumarins. J. Agric. Food Chem. 1971, 19, (6), 11061110. 46. Bourgaud, F.; Hehn, A.; Larbat, R.; Doerper, S.; Gontier, E.; Kelln er, S.; Matern, U., Biosynthesis of coumarins in plants: a major pa thway still to be unravelled for cytochrome P450 enzymes. Phytochemistry Reviews 2006, 5, (2-3), 293-308. 47. Hasegawa, S.; Hoagland, J. E., Bi osynthesis of limonoids in citrus. Phytochemistry 1977, 16, (4), 469-471. 48. Fong, C. H.; Hasegawa, S.; Coggins, C. W. ; Atkin, D. R.; Miyake, M., Contents of limonoids and limonin 17-beta-d-glu copyranoside in fruit tissue of valencia orange during fruit-growth and maturation. J. Agric. Food Chem. 1992, 40, (7), 1178-1181. 49. Guadagni, D. G.; Maier, V. P.; Turnbaug.Jg, E ffect of some citrus juice constituents on taste thresholds for limonin and naringin bitterness. J. Sci. Food Agric.. 1973, 24, (10), 1277-1288. 50. Kim, J. S.; Sagaram, U. S.; Burns, J. K. ; Li, J. L.; Wang, N., Response of sweet orange (citrus sinensis) to 'candidatu s liberibacter asiaticus' inf ection: microscopy and microarray analyses. Phytopathology 2009, 99, (1), 50-57. 51. Ortiz, J. M., Botany: taxonomy, morphology an d physiology of fruits, leaves and flowers. In Citrus: The genus Citrus Dugo, G.; Di Giacomo, A., Eds. Taylor and Francis: London, 2002; pp 16-35. 52. Weeks, W. W., Carotenoids: A S ource of Flavor and Aroma. In Biogeneration of Aromas Parliment, T. H.; Croteau, R., Eds. Ameri can Chemical Society: Washington, D.C., 1986; pp 157-166. 53. Fellman, J. K.; Miller, T. W.; Mattinson, D. S.; Mattheis, J. P. In Factors that influence biosynthesis of volatile fla vor compounds in apple fruits 2000; Amer Soc Horticultural Science: 2000; pp 1026-1033. 54. Takita, M. A.; Berger, I. J.; Basilio-Palmieri, A. C.; Borges, K. M.; de Souza, J. M.; Targon, M., Terpene production in th e peel of sweet orange fruits. Gen. Molec. Bio. 2007, 30, (3), 841-847. 55. Elston, A.; Lin, J. M.; Rouseff, R., Determinati on of the role of valencene in orange oil as a direct contributor to aroma quality. Flav. Fragr. J. 2005, 20, (4), 381-386. 56. Sharon-Asa, L.; Shalit, M.; Frydman, A. ; Bar, E.; Holland, D.; Or, E.; Lavi, U.; Lewinsohn, E.; Eyal, Y., Citrus fruit flavor and aroma biosynthesis: isolation, functional characterization, and developmental regulation of Cstps1, a key gene in the production of the sesquiterpene aroma compound valencene. Plant Journal 2003, 36, (5), 664-674.

PAGE 75

75 BIOGRAPHICAL SKETCH Lilibeth Rubio Dagulo graduated cum laude with a Bachelor of Science in food science and human nutrition at the University of Florid a in May 2007. She is the first American born child of Filipino immigrants to receive an unde rgraduate degree from an American institution. Lilibeth wanted to pursue a specialization in flavor chemisty. Therefore, in August 2007, she entered the Master of Science Program in the Food Science and Human Nutrition Department at the University of Florida. Under the supervision of Dr. Russell L. Rouseff, Lilibeth received training in citrus flavor analysis. Lilibeth plans to pursue a career as a flavor chemist and/or flavor analyst at a fla vor or beverage company.