Citation
Resistance to adult banded cucumber beetle, Diabrotica balteata LeConte, in romaine lettuce

Material Information

Title:
Resistance to adult banded cucumber beetle, Diabrotica balteata LeConte, in romaine lettuce
Creator:
Huang, Juan, 1967-
Publication Date:
Language:
English
Physical Description:
ix, 132 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Beetles ( jstor )
Chemicals ( jstor )
Eggs ( jstor )
Female animals ( jstor )
Infestation ( jstor )
Insects ( jstor )
Latex ( jstor )
Leaf area ( jstor )
Leaves ( jstor )
Lettuce ( jstor )
Diabrotica -- Host plants ( lcsh )
Dissertations, Academic -- Entomology and Nematology -- UF ( lcsh )
Entomology and Nematology thesis, Ph. D ( lcsh )
Lettuce -- Disease and pest resistance ( lcsh )
City of Layton ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2000.
Bibliography:
Includes bibliographical references (leaves 118-131).
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Juan Huang.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
025829789 ( ALEPH )
46714063 ( OCLC )

Downloads

This item has the following downloads:


Full Text











RESISTANCE TO ADULT BANDED CUCUMBER BEETLE, DIABROTICA
BALTEATA LECONTE, IN ROMAINE LETTUCE

















By

JUAN HUANG


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

UNIVERSITY OF FLORIDA


2000
































Copyright 2000



by


Juan Huang



























This effort is dedicated to my husband Honghong Zhang and to my parents Yuzhen Huang and Zhichang Mao.















ACKNOWLEDGMENTS

I express my respect and gratitude to Dr. Gregg S. Nuessly for serving as the

chairman of my committee and providing funding for this research. I would like to thank Wedgeworth family of Florida vegetable and sugar producers whose fellowship sponsored this research. My deepest gratitude also goes to Dr. Heather J. McAuslane for being my co-advisor, proving funding for this research and giving me moral and academic support, encouragement, and guidance throughout my Ph.D. program. I would also like to appreciate the advice and expertise of my other committee members, Dr. Frank Slansky, Dr. Russell Nagata, and Dr. Anson Moye. Special thanks are extended to Dr. Russell Nagata for providing lettuce seeds for this research. I am very grateful to Dr. Hans T. Alborn for providing me a micro spray device which made the research in Chapter 3 possible. I would not have been able to freeze dry lettuce leaves without the help of Mr. Rob Pellicit from Food Science and Human Nutrition Department. My sincere thanks go to Debbie Boyd for ordering research supplies and giving me unforgettable friendship. I also thank Yasmin Cardoza, Alonso Suazo, Ramazan Cetintas, and Jiang Chen for their friendship.

I wish to thank all of those I have failed to mention, your help and support made this journey complete.


iv
















TABLE OF CONTENTS

page

ACKN OW LED G M EN TS ................................................................................................. iv

ABSTRA CT..................................................................................................................... Viii

CHAPTERS

1 LITERATURE REVIEW AND RESEARCH GOALS ...............................................1

Introduction..................................................................................................................... I
D iabrotica balteata Biology and Pest Status............................................................. 2
Overview of Plant Resistance ......................................................................................... 4
H istory of Plant Resistance to Insects...................................................................... 4
Plant Resistance Term inology ................................................................................. 5
Constitutive Resistance ........................................................................................... 7
Induced Resistance ............................................................................................... 14
Host Plant Resistance to D iabrotica balteata .......................................................... 17
Host Plant Resistance in Lettuce ............................................................................... 19
Plant Secondary Chem icals .................................................................................... 19
Genetic Resistance ................................................................................................. 20
N utritional Quality ................................................................................................. 21
Possible Evidence of Inducible Resistance .......................................................... 23
Research Goals ............................................................................................................. 23

2 RESISTANCE IN ROMAINE LETTUCE CULTIVARS TO ADULT DIABROTICA BAL TEA TA ......................................................................................................................... 25

Introduction................................................................................................................... 25
M aterials and M ethods............................................................................................... 27
P la n ts ......................................................................................................................... 2 7
In se c ts ....................................................................................................................... 2 7
Binary-choice Test ................................................................................................. 30
No-choice Test ...................................................................................................... 31
Fitness Bioassay.................................................................................................... 31
Starvation Test ...................................................................................................... 32
R e su lts .......................................................................................................................... 3 3
Binary-choice Test ................................................................................................... 33
No-choice Test ............................................................................................................. 35


V









Fitness Bioassay ........................................................................................................ 35
Starvation Test.............................................................................................................. 38
Discussion .................................................................................................................... 41

3 ROLE OF THE LEAF SURFACE OF LETTUCE IN RESISTANCE TO DIABRO TICA BALTEATA............................................................................................. 47

Introduction................................................................................................................... 47
M aterials and M ethods............................................................................................... 50
P lan ts ......................................................................................................................... 5 0
In se c ts ........................................................................................................................ 5 1
Solvent Effects on Extraction.................................................................................. 51
Feeding Response to Solvent-extracted or Norm al Leaves.................................... 52
Feeding Response to Crude Extracts in Binary-choice Test .................................. 53
Feeding Response to Crude Extracts in No-choice Test......................................... 55
R e su lts .......................................................................................................................... 5 5
Solvent Extraction Com parisons ........................................................................... 55
Feeding Response to Solvent-extracted or Norm al Leaves.................................... 57
Feeding Response to Crude Extracts in Binary-choice Test .................................. 60
Feeding Response to Crude Extracts in No-choice Test ........................................ 64
Discussion .................................................................................................................... 64

4 INVESTIGATION OF PHYTOCHEMICAL NATURE OF RESISTANCE IN LETTUCE TO DIABROTICA BAL TEA TA USING ARTIFICIAL DIET .....................69

Introduction................................................................................................................... 69
M aterials and M ethods............................................................................................... 71
Plants and Insects ................................................................................................... 71
Preparation of Leaf Powder.................................................................................... 71
Artificial Diet ........................................................................................................ 72
Binary-choice Test ................................................................................................. 72
No-choice Test ........................................................................................................ 74
R e su lts .......................................................................................................................... 7 5
Binary-choice Test ................................................................................................. 75
No-choice Test ........................................................................................................ 78
Discussion .................................................................................................................... 83


5 POSSIBLE MECHANISMS INVOLVED IN LETTUCE RESISTANCE TO DIABRO TICA BA L TEA TA............................................................................................. 87

Introduction................................................................................................................... 87
M aterials and M ethods............................................................................................... 89
Plants and Insects ................................................................................................... 89
Screening M ethods ................................................................................................. 90
Latex Physical Properties ...................................................................................... 92
Latex Sm earing Test............................................................................................... 92


vi










Observation of Feeding Behavior ........................................................................ 93
Inducible Resistance ............................................................................................... 93
Results ..................................................................................... ...... -.....................- 94
Screening M ethods ................................................................................... .............- 94
Latex Physical Properties ...................................................................................... 98
Latex Sm earing Test................................................................................................ 98
Observation of Feeding Behavior............................................................................ 102
Inducible Resistance........................................................................................... 104
D iscussion .................................................................................................................. 104

6 SU M M ARY AND CON CLU SION S ...........................................................................113

APPEN D IX M ISCELLA N EA ................................................................................... 117

LIST O F REFEREN CES.................................................................................................118

BIOGRA PH ICA L SKETCH ...........................................................................................132




































vii















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

RESISTANCE TO ADULT BANDED CUCUMBER BEETLE, DIABROTICA
BAL TEA TA LECONTE, IN ROMAINE LETTUCE By

JUAN HUANG

December, 2000


Chairman: Gregg S. Nuessly
Major Department: Entomology and Nematology

Four cultivars of lettuce, Tall Guzmaine (TG), Parris White (PW), Short

Guzmaine (SG), and Valmaine (Val), were screened for resistance to adult Diabrotica balteata LeConte under laboratory conditions. Val had the highest level of resistance, followed by PW and SG; TG was particularly susceptible to adults when using leaf area consumption as the criterion for resistance. Female adults ate more foliage, gained more weight, and produced more eggs when feeding on TG than on Val for 10, 13 or 16 d. A starvation test confirmed that females on Val produced no mature eggs because of their inability to use Val as a food source.

The leaf surface of lettuce was investigated for its possible role in resistance. Even though D. balteata consumed a much greater amount from Val leaves with their surface chemicals removed, leaf surface extracts from Val were not deterrent to D. balteata feeding when applied to lima bean leaf surfaces at various concentrations. These


viii









results suggest that leaf surface chemicals in lettuce are not responsible for resistance in Val.

Freeze-dried leaf powder of TG or of Val was added to an artificial diet for D.

balteata at four concentrations (0, 10, 20, 30%) to determine the chemical basis of lettuce resistance. In choice experiments, there were no significant differences between TG diets and Val diets at each concentration. In no-choice experiments, no significant differences were found in the number of mature eggs produced, female's weight gain, or adult's mortality between TG and Val diets at four different concentrations. The results of this assay failed to detect any reason for the resistance of Val.

TG and Val differed in latex physical properties. Latex from Val turned brown faster than that from TG, and took more time to stop flowing. Both latex from TG and Val applied on lima bean leaves inhibited D. balteata feeding. Localized induced resistance to feeding was found in Val but not in TG after plants were previously damaged for 48 h.

Therefore, I propose that localized inducible resistance and physical and chemical defenses in latex may be two major reasons for resistance in Val to D. balteata.


ix














CHAPTER 1
LITERATURE REVIEW AND RESEARCH GOALS Introduction

Lettuce, Lactuca sativa L., is an important vegetable crop grown in many

countries in the world. By far the greatest commercial production of lettuce takes place in the USA (Ryder 1998). The majority of the insects which are economically important on lettuce belong to five orders: Homoptera, Lepidoptera, Diptera, Hymenoptera (Ryder 1998) and Coleoptera. These insect pests can cause damage directly by chewing on foliage, stems and roots or indirectly by transmitting disease. Moreover, they can also lower the value of the lettuce crop by causing unsightliness such as the presence of insect cast skins and detritus. To date, lettuce growers have relied heavily on insecticides to control insect pests to reduce economic losses. However the heavy use of nonselective insecticides has led to well-known problems, such as development of insect resistance to these chemicals, the resurgence of secondary insect pests, and the persistence of residues that are toxic to humans, animals and other nontarget organisms. The western corn rootworm, Diabrotica virgifera virgifera LeConte, provides a classic example of a manmade pest (Metcalf 1986). This insect was first described by LeConte in Kansas in 1868 and first found attacking corn in Colorado in 1909 before it slowly spread into the Nebraska in 1929 (see references cited by Metcalf 1986). Because of large-scale applications of insecticides, adult western corn rootworm resistance to insecticides was noted in theI960s. The resistant strain spread rapidly about an average of 120 miles per year from a single locus in southeastern Nebraska in 1961 to encompass much of the corn

1







2


growing area of many states. By 1980, this strain had spread throughout the U.S. corn belt (Metcalf 1986). Integrated pest management (IPM) programs have been developed to reduce the reliance on chemical insecticides. Host plant resistance is recognized as the most effective component of IPM (Panda and Khush 1995), because it has low impact on non-target organisms and the environment and is usually compatible with other control tactics, such as biological and cultural controls.

In nature, plants are under the constant threat of damage by insects, which

dominate the world's fauna in terms of number of species and individuals (Dicke 1999). However, plants defend themselves very effectively by using a variety of defensive strategies ranging from mechanical to chemical weapons (Louise 1999). Defenses can operate by either constitutive resistance, which is always active, or by induced resistance. In induced resistance, production and/or release of defensive chemicals or characteristics are induced by extrinsic physical or chemical stimuli, such as herbivory, pathogen infection, or mechanical injury (Kogan and Paxton 1983).

Diabrotica balteata Biology and Pest Status

Diabrotica balteata LeConte is commonly known as the banded cucumber beetle. It is distributed throughout the southern and western United States and south into central America (Krysan 1986). In the southern United States, D. balteata occurs as an adult during much of the year (Schalk 1986). Wolfenbarger (1963) reported D. balteata as a ubiquitous insect throughout Florida with fluctuation of its abundance and injuriousness from year to year. This insect is a pest of leafy vegetables and sweet corn in Florida. The adult D. balteata is marked with alternating green and yellow bands across the elytra, but the newly emerged adult has a soft body with pale color. The adult feeds for 4 to 8 d







3


before mating. The range from copulation to first egg deposition can be 9 to 28 d. Individual females lay approximately 350 to 849 eggs (depending on food source) over its lifetime of 5 to 11 wk under laboratory conditions. Oviposition usually occurs at intervals of 2 to 3 d beginning when the female is about 3 wk old and lasts from 2 to 8 wk. Eggs are deposited singly or in clusters in small cracks in the soil or under objects lying on the surface (Schalk 1986, Pitre and Kantack 1962). Three larval instars occur with a mean larval developmental period of 17 d at 270C to 23 d at 21 C, depending on food source and rearing environment. The third-instar larvae increase dramatically in size and do the most severe damage to plants by cutting and partially boring into young roots. Pupation takes place in the soil and the average time required for adult emergence is 7 to 9 d at temperatures of 21 to 27'C (Pitre and Kantack 1962).

The larvae of D. balteata severely damage the roots of crops, reducing yield in bean, melon and corn (Pitre and Kantack 1962). However, feeding on other plant parts such as potato tubers, and peanut pods has been reported also (Wolfenbarger 1963). Other host plants for the larvae include squash, sorghum, soybean, sweet potato, sweet corn and tomato (Table 2 in Teng 1983). No complete report of the larval host range has been reported in the literature. Adults feed on foliage, flowers or fruit (Elsey 1988) of more than 50 host plant species (Teng 1983) in 23 families (Saba 1970). These plants include broccoli, corn, cowpea, cucumber, lima bean, peanut, pepper, potato, spinach, squash, soybean, sweet potato, tomato and wheat (Table 3 in Teng 1983). Lettuce was also listed in Teng's table, even though there was controversy in the literature as to whether this was a host for D. balteata adults at that time (Saba 1970). In addition to feeding, adults also contribute to yield losses by transmitting plant viruses, such as







4

cowpea mosaic, bean rugose mosaic, squash mosaic, muskmelon necrotic spot and melon bacterial wilt (Da Costa and Jones 1971, Gergerich et al. 1986). Nuessly (personal communication) found that adult D. balteata caused serious damage to lettuce in southern Florida by feeding across the entire leaf surface, removing tissue and causing necrosis of surrounding tissues. Beetle feeding on older plants caused lettuce heads to be unmarketable due to feeding damage and fecal contamination. In addition, D. balteata caused yield reduction and uneven stand maturity by feeding on young and middle aged lettuce.

Overview of Plant Resistance

History of Plant Resistance to Insects

There are many historical accounts of the successful use of resistant cultivars to control insect pests. One of the most famous successes is that of the development of resistance in French wine grapes to attack by the grape phylloxera in 1873 by grafting on to American grape rootstock. This strategy saved the French wine industry (Ortman and Peters 1980). Resistant varieties of wheat and apples were first developed and cultivated in the USA during the eighteenth and early nineteenth centuries (Smith 1989). Highyielding, pest-resistant rice cultivars were used in tropical Asia during the "Green Revolution" of the 1960s. These cultivars helped several countries in southern and southeast Asia meet the food production needs of their populations (Smith 1989). Use of insect-resistant crop cultivars is still the principal method of control of some major pests today (Stoner 1996).

Many crops with resistance to insect pests have already been developed (Tables I and 2 in Stoner 1996). A survey on insect-resistant germplasm released in the USA from







5

1988 to early 1994 indicated that 34% of 117 releases were in grain crops, 27% in alfalfa and clover, 15% in cotton, 11% in vegetables, 4% in fruit and 8% in other crops (Stoner 1996). Genetic engineering techniques have made it possible to transfer genes directly into plants, thus opening new opportunities for host plant resistance (Panda and Khush 1995).

Plant Resistance Terminology

Painter (1951) defined host plant resistance as the relative amount of heritable

qualities possessed by the plant, which influence the ultimate degree of damage done by the insect. Three modalities of plant resistance, originally defined by Painter (1951) but modified by Horber (1980), are commonly referred to in plant resistance literature: antixenosis, antibiosis, and tolerance. However, not all resistance phenomena can be clearly classified into these three categories of resistance. Biophysical and biochemical plant defense, as well as nutritional factors are not only involved in antixenosis but also in antibiosis. In certain cases, antixenosis can not be clearly separated from antibiosis because the deterrent chemicals and toxins in the plant are sometimes difficult to distinguish (Panda and Khush 1995). Furthermore, these three main types of resistance can interact, complement and compensate with each other in different abiotic and biotic environments (Panda and Khush 1995). Overlaying the types of resistance are two categories of resistance, constitutive and inducible, that are distinguished based on whether the resistant characters are caused by extrinsic stimuli (Kogan and Paxton 1983).

Antixenosis, also referred to as nonpreference by Painter (1951), is the category of plant resistance to insects that describes the plant acting as a poor host to deter or reduce colonization by insects. Antixenotic characteristics include morphological,







6


physical, or structural qualities (such as plant pubescence, and frego bract), and biochemical factors (such as arrestants, attractants, repellents) which interfere with mating, oviposition, feeding, and food ingestion (Panda and Khush 1995). Biophysical and biochemical factors affect the behavior of an insect pest to force it to choose an alternate host plant. Plants exhibiting antixenosis should have a reduced initial number of colonizers early in the season and reduce the size of the insect population after each generation as compared with susceptible plants (Panda and Khush 1995).

Antibiosis is the resistance mechanism that describes the negative effects of a

resistant plant on the biology of an insect after this insect has colonized and started using the plant. Such resistance results in abnormal insect growth, development, reproduction and survival. Plants that exhibit antibiosis reduce the reproductive rate and survival of insects to slow the rate of population increase. The antibiotic properties of the host plant may be expressed as constitutive or induced resistance against herbivores. These properties may be biophysical (such as trichomes), biochemical (such as toxins, growth inhibitors), or nutritional in nature (Panda and Khush 1995).

Tolerance is the resistance modality in which a plant has the genetic ability to

withstand or recover from damage caused by an insect pest abundance equal to that on a susceptible cultivar. Tolerance does not affect the rate of population increase of the target pest, but raises the threshold level, thereby allowing plants to outgrow an insect infestation or recover and add new growth after the destruction or removal of damaged tissues (Smith 1989, Panda and Khush 1995).

Constitutive resistance, also called preformed resistance (Pollard 1992), is a broad term, referring to the presence of physical defenses as well as to the absence of some







7


required nutrients, presence of compounds with kairomonal action and non-induced accumulation of secondary metabolites with allomonal properties (Kogan and Paxton 1983).

Induced resistance is the qualitative or quantitative enhancement of the plant's defense against pest-related injury or extrinsic physical or chemical stimuli. These extrinsic stimuli are called inducers or elicitors (Kogan and Paxton 1983). Induced resistance indicates the ability of plants to reallocate resources and operates only after the plant is attacked by pests or stressed by other inducers (Thaler and Karban 1997). Constitutive Resistance

Constitutive defense against insects can be mediated by plant physical attributes (e.g., toughness of plant tissue, thorns and spines), by the presence of plant secondary metabolites (e.g., toxins, repellents, and digestibility reducers), or by other factors such as insufficient nutrition levels. Based on biosynthetic criteria, there are three principal groups of secondary metabolites which plants usually produce as constitutive resistance. One is a terpenes group (e.g., monoterpenes, sesquiterpenes, diterpenes, triterpenes), which is synthesized from acetyl CoA via the mevalonic acid pathway. The second group is phenolics (e.g., chlorogenic acid, caffeic acid, coumarins, tannins), which are formed via the shikimic acid pathway or malonic acid pathway. The third group is nitrogen-containing compounds (e.g., alkaloids, glucosinolates, cyanogenic glycosides), which are biosynthesized primarily from amino acids (Panda and Khush 1995).

Resistance at the plant surface. The plant surface is recognized as the first line of defense against insect attack. Insects depend on a combination of physical and chemical stimuli at the plant surface to assess whether the plant is suitable for oviposition, settling







8


or feeding (Woodhead and Chapman 1986). Hairs or trichomes have been shown to act as physical barriers, but they may also secrete behaviorally active chemicals. Other chemicals affecting the evaluation process are found within leaf waxes or other secretory structures such as essential oil glands.

The surface of virtually all plants is covered by an amorphous layer of waxes known as the epicuticular wax (Baker 1982), the primary role of which is to prevent water loss (Eigenbrode and Espelie 1995). The epicuticular wax is comprised of mainly aliphatic wax components, including n-alkanes, wax esters, free fatty alcohols and free fatty acids. Sugars, amino acids, nonprotein amino acids, sucrose and glucose esters, sesquiterpenes, diterpenes, phenolics, phenolic glycosides and glucosinolates are also components of epicuticular wax (Woodhead and Chapman 1986, Eigenbrode and Espelie 1995).

A growing number of reports show that waxes on many plant surfaces either enhance or deter oviposition, feeding and movement of various insects (Juniper and Southwood 1986, Eigenbrode and Espelie 1995). Leaf epicuticular wax in Brassicaceae crops appeared to be a major antixenotic factor that affects the feeding rate and feeding pattern of the crucifer flea beetle, Phyllotreta cruciferae (Goeze) (Bodnaryk 1992). The flea beetle fed on waxy leaves at much lower rates than on non-waxy leaves. It is hypothesized that edge feeding on waxy leaves, which is associated with a low feeding rate, may be determined by physical and chemical factors that render the leaf unpalatable or difficult to feed upon, or both. Other examples of leaf surfaces containing feeding deterrents include Colorado potato beetle, Leptinotarsa decemlineata Hawkes, on Solanurn berthaultii (Yencho et al. 1994), migratory grasshopper, Locusta migratoria







9


(L.), on sorghum (Woodhead 1983), and the brown planthopper, Nilaparvata lugens Stal., on rice varieties (Woodhead and Padgham 1988).

Fatty alcohols and a-tocopherylquinone isolated and identified from the leaf

surface of a poplar clone were found to stimulate feeding by adult cottonwood leaf beetle, Chrysomela scripta Fabr. (Lin et al. 1998). Fatty alcohols from mulberry stimulated feeding of larvae of the silkworm Bonbvx mori L. (Mori 1982). Fatty alcohols from several host plants stimulated feeding of several species of chrysomelid beetles (Adati and Matsuda 1993). Diamondback moth larvae, Plutella xylostella (L.), spent more time walking and searching on a resistant cabbage with glossy leaves than on a susceptible variety with normal wax bloom. This behavior resulted in increased larval mortality on glossy plants due to starvation and desiccation. Morphological characteristics of the leaf waxes were of primary importance in causing higher net movement rates and reduced feeding on glossy resistant lines (Eigenbrode and Shelton 1990). A similar behavioral pattern occurred on glass slides on which hexane or dichloromethane extracts of leaf surface wax of the two cabbage genotypes were deposited, which implicated strongly that leaf surface waxes played an important role in acceptance of cabbage varieties by diamondback moth larvae (Eigenbrode et al. 1991). Wax chemistry was proposed to be another reason causing these behavioral differences since the chemical compositions of the leaf extracts were markedly different. Several compounds, including the triterpenols x- and P-amyrin, were found only in the glossy waxes. Disruption of the leaf waxes on a susceptible cabbage plant resulted in high rates of larval movements, similar to those on glossy resistant plants (Eigenbrode and Shelton 1990). Leaf surface chemicals may affect insect oviposition. Chemicals on the leaf surface of Brassica oleracea stimulated







10


oviposition by the cabbage root fly, Delia radicum (L.) (Roessingh et al. 1992) and chemicals on carrot leaves stimulated oviposition by the carrot fly, Psila rosae (F.) (Stadler and Buser 1983).

Glandular and nonglandular trichomes have been found to be of importance to insect oviposition, movement and feeding through mechanical and/or chemical means. Physical factors such as trichome density, erectness, length and shape influence their effects on insects (Johnson 1975). Nonglandular trichomes impart general antixenosis resistance by providing an effective barrier that influences the attachment of insects onto the plant surface, as well as their movement and feeding (Southwood 1986). The high number of trichomes on the upper and lower surfaces of the leaves of a resistant maize cultivar was the major reason for deterring oviposition by the stem borer, Chilo partellus (Swinhoe) (Kumar 1992). Oghiakhe (1995) reported that pubescence in wild and cultivated cowpeas adversely affected oviposition, mobility, food consumption and utilization by the legume pod borer, Maruca testulalis (Geyer). They concluded that the greater length and density of nonglandular trichomes were responsible for reduced oviposition by the borer. Trichomes may also have an antibiotic effect on insects. For example, hooked trichomes on bean plants trapped or impaled adult leafminers, Liriomyza trifolii (Burgess) (Quiring et al. 1992), and aphids and whiteflies (Sengonca and Gerlach 1984).

Glandular trichomes effectively reduce herbivore feeding through the deployment of adhesives and toxins. Various compounds have been identified as the active components from trichomes (e.g., alkaloids, phenols, and aliphatic hydrocarbons) and are well known feeding deterrents to herbivores (Southwood 1986). An extract of Solanum







I1


berthaultii leaves containing exudates from type B trichomes inhibited Colorado potato beetle, L. decemlineata (Pelletier and Smilowitz 1990). 2-Tridecanone and 2-undecanone were the dominant compounds in the type VI trichomes of LYcopersicon species which were toxic to neonate larvae of the tomato pinworm, Keiferia lycopersicella (Walsingham), and the beet armyworm, Spodoptera exigua (Hubner) (Lin et al. 1986).

Furthermore, some plants can also rely on chemicals from other types of glands found on the surface of the plant to immobilize, repel and poison herbivores. These include glandular excretory structures, which discharge a variety of oils and resins, and oil glands, which contain various terpenoid oils (Stipanovic 1983). Ten essential oils from Labiate plants, such as spearmint, thyme, and rosemay, inhibit the settling of the green peach aphid, Myzus persicae (Sulzer), on them in choice tests (Hori 1999). In a nochoice test, aphids rarely settled on the sealing film which covered the diet containing spearmint or thyme oil and most of them died. Cotton pigment glands contain gossypol and other terpenoids that provide resistance to Heliothis sp. and other cotton insects (Stipanovic 1983, Parrott 1990, McAuslane et al. 1997).

Resistance due to internal factors. In addition to physical and chemical barriers at the epidermis, plants also contain various biochemicals acting as insect repellents, feeding inhibitors and toxins that may act constitutively or are induced by injury. These biochemicals protect plants at different phases of growth and include phenolic compounds (e.g., flavonoids and aromatic acids), terpenoids (e.g., sesquiterpene lactones and heliocides), nitrogenous compounds (e.g., amino acids and amides), proteinaceous compounds (e.g., protease inhibitors and lectin) and toxic seed lipids (e.g., unusual fatty acids) (Hedin 1986). Many such compounds have been isolated from many plants and







12


have been classified into two broad categories. General resistance is not tissue specific and chemical concentrations increase as the tissue matures (e.g., acids, phenols, terpenes, alcohols, chlorogenic acid, quercetin, tannin, etc.). Specific resistance factors reach their highest concentration in young leaves and fruits and decrease as the crop matures (e.g., sinigrin, tomatine, solanine, gossypol, 2,4-dihydroxy-7-methoxy- 1,4-benzoxazin-3-one (DIMBOA)) (Panda and Khush 1995). Alkaloids, such as nicotine and nornicotine extracted from tobacco plants, were used as early insecticides against insect pests. Pyrrolizidine and indole alkaloids, and other structurally unrelated alkaloids serve as feeding deterrents to many insects (Panda and Khush 1995). Plant lectins, which are carbohydrate-binding proteins of non-immune origin and cause cells to agglutinate and bind glycans of glycoproteins, glycolipids, or polysaccharides, are found to affect survival of lepidopteran, dipteran, coleopteran and homopteran insects (Carozzi and Koziel 1997). Many phenolic compounds are widely distributed among vascular plants and are thought to function as chemical defenses against herbivory because of their ability to interact with proteins and inhibit enzyme functions (Felton et al. 1989). Various phenolic compounds are feeding deterrents to greenbug aphid, Schizaphis graminumn (Rondani), and migratory locust, Locusta migratoria (Reiche and Fairmaire), on sorghum; stem borer, Chilo suppressalis (Walker), on rice; leaf beetle, Lochniaeae capreae cribrata (L.), on willow; and bird cherry oat aphid, Rhopalosiphum padi (L.), and English grain aphid, Sitobion avenae (F.), on wheat (see references cited by Smith 1989).

Canalicular defense. Latex is typically contained within specialized living cells or a series of fused cells called laticifers, which form a complex network of tubes







13


throughout the plant (Esau 1977). Latex occurs in 12,500 species belonging to 20 families such as Apocynaceae, Asclepiadaceae, Compositae, Euphorbiacae, and Moraceae (Fahn 1979). The laticifers are under positive pressure, which results in rapid emission of a viscous, often milky latex upon cutting (Data et a]. 1996). Several theories have been proposed for the biological role of latex in plants, including a nutritional reserve (Maksymowych and Ledbetter 1987), regulation of water balance (Sen and Chawan 1972), storage of nonfunctional metabolic by-products (Biesboer and Mahlberg 1978), and a defensive system containing adhesives or toxins to entrap or deter herbivores (Dussourd 1993, 1995, 1997, 1999). When injured, many plants exude viscous latex which frequently becomes sticky on exposure to air and may entrap or gum up the mouthparts of small herbivores such as aphids, whiteflies and ants (Dillon et al. 1983, Dussourd 1993, 1995).

Insects may avoid latex and circumvent mechanical stickiness and possible

toxicity of latex either by severing veins or leaf petioles or by cutting trenches before feeding on areas of the portion of the plant isolated by the cuts (Zalucki and Malcolm 1999, Dussourd 1997). A vein-cutting feeding behavior of the chrysomelid beetle, Labidomera clivicollis (Kby.), was first observed on milkweed, Asclepias syriaca, by Dussourd and Eisner (1987). Labidornera adults and larvae bit repeatedly into several adjacent branches of a leaf midrib to induce latex exudation before they moved distal to the cuts and fed on the edge of the leaf with no visible latex emission due to blockage of further latex flow to the sites. Similar counteradaptations also occur in diverse insect groups from I 1 families including various caterpillars, beetles, katydids, and sawflies







14


(Dussourd and Denno 1991, Dussourd 1993, McCloud et al. 1995). In each case, the behaviors induce latex drainage and blockage of latex flow to intended feeding sites.

Beside adhesives, latex from many plants contains secondary metabolites known to be toxic or deterrent to animals. These include cardiac glycosides in Asclepias latex (Zalucki and Brower 1992, Zalucki and Malcolm 1999); diterpenes (Evans and Schmidt 1976), triterpenoids (Spilatro and Mahlberg 1986) and nonprotein amino acids (Haupt 1976) in Euphorbia latex; sesquiterpene lactones in Cichorium latex (Rees and Harborne 1985) and morphine, berberine, and other alkaloids in Papaver and Chelidonium latex (Roberts 1987, Valle et al. 1987). Latex produced by sweetpotato served as a potential defense mechanism against the sweetpotato weevil, Cylasformicarius (F.). Less weevil feeding damage was found on young vine material with more latex production than on older more mature portions of the vine. Both feeding and oviposition were reduced when latex was applied to the surface of root cores and to a semi-artificial media (Data et al. 1996). First-instar larvae of the monarch butterfly Danaus plexippus L., grew faster and survived better on leaves of North American milkweed species after the latex flow was reduced by partially cutting the leaf petioles (Zalucki and Malcolm 1999). The active compound uscharin isolated from latex of Calotropis procer was highly toxic to land snails (Hussein et al. 1994). Plant latexes from several plants were found to be highly deleterious to many plant parasitic nematodes (Siddiqui and Alam 1990). Induced Resistance

Herbivory-induced responses by plants have been the subject of many excellent reviews (Baldwin 1989, Tallamy and Raupp 1991, Karban and Baldwin 1997). Induced defense against herbivores has been recorded in more than 100 plant species in 34







15


families since the 1970s (Karban and Baldwin 1997). Wound-induced resistance by herbivores or mechanical means can be correlated with the induction of activities of oxidative enzymes such as lipoxygenase (LOX), peroxidases (POD), and polyphenol oxidase (PPO) (Bi et al. 1994), synthesis of primary products such as protease inhibitors, and vitamins (Green and Ryan 1972, Stout et al. 1998, Bi et al. 1994, 1997), enhanced synthesis of secondary metabolites such as phenolic compounds, terpenoids, and nitrogen compounds (Bi et al. 1997, Tuomi et al. 1988, Hartley and Lawton 1991, McAuslane et al. 1997, Baldwin 1991, Baldwin and Ohnmeiss 1994), or decline in the nutritional quality such as protein and amino acids (Bi et al. 1994, 1997). Moreover, many of these induced responses are a complex phenomenon and associated with multiple components of induced biochemical pathways. Several plant species including tomato, potato, tobacco, and cotton have been studied in considerable depth. Growth of larval Helicoverpa zea Boddie was significantly decreased when they fed on previously damaged foliage or squares of cotton compared to the controls (Bi et al. 1994). This was due to a significant decline in host nutritional quality of protein and most amino acids in both foliage and squares after herbivory, increases in activities of oxidative enzymes such as POD, ascorbate oxidase, and diamine oxidase, and increases in the levels of chlorogenic acids and lipid peroxides. Because of the decline in the nutritional quality of foliar protein, and increases in LOX activity, lipid peroxidation products, and trypsin inhibitor content, the growth rate of fourth-instar H. zea was reduced significantly after they fed on wounded tomato foliage (Bi et al. 1994, Bi and Felton 1995). The enhanced production of phenolic compounds was indicated by an increase in the activity of phenylalanine ammonia lyase (PAL) in wounded tissues (Bi and Felton 1995).







16

There are three time scales proposed for the study of induced responses (Baldwin 1989, Edwards and Wratten 1983). First, highly localized chemical changes occur immediately upon damage and are restricted to the damaged tissues, usually resulting from the mixing of previously isolated enzymes and substrates. The mechanical damage of herbivory caused the cyanogenic glycoside dhurrin in sorghum to come into contact with an enzyme, resulting in the release of hydrogen cyanide and protecting sorghum from further attack by insects (Woodhead and Bernays 1977). Mustard oils, released from damaged Cruciferae crops due to the contact of glucosinolates with the enzyme myrosinase, are known to be toxic to many generalist herbivores (Louda and Mole 1991). On the second time scale, rapidly induced responses may occur within hours or days of the injury and can be systemic or localized to the damaged leaf. Phenolic compounds increased within one day in tissues of Chinese cabbage surrounding feeding sites of the bug Lygus disponsi Linnavuori (Hori and Atalay 1980). Chlorogenic acid and isochlorogenic acid are the most commonly formed substances, but others, such as pcoumaric acid, caffeic acid, and scopoletin are also commonly found (Edwards and Wratten 1983). On the third time scale, more widely-dispersed changes (systemic induced response) may affect an entire organ, branch or plant by de novo synthesis and activation of a suite of phytochemicals within days or years of injury. One example of the third category of induced resistance is the production of protease inhibitors in tomato (Green and Ryan 1972). Colorado potato beetle feeding or mechanical wounding of a tomato leaf induced the rapid accumulation of trypsin and chymotrypsin inhibitors throughout the above-ground plant tissues. Systemic induced resistance has received







17

much research attention from ecologists because of its potential importance in regulating herbivore populations (Karban and Myers 1989).

The patterns of systemic induced resistance may be correlated with the degree of vascular connectivity between damaged leaves and sampled leaves (Jones et al. 1993) based on the theory of the vasculature as the primary path of material transport between leaves (Geiger 1987). In eastern cottonwood, the pattern of systemically induced resistance was found to be directly related to the distribution of the plant vasculature (Jones et al. 1993). Mechanical damage to an individual leaf resulted in systemically induced response in non-adjacent, orthostichous leaves with direct vascular connections. However, the plant vascular connection was not found to be related with the spread and spatial extent of the induced resistance in Betula pendula Roth saplings. Instead, the resistance levels of all leaves increased following the damage (Mutikainen et al. 1996).

Host Plant Resistance to Diabrotica balteata

Cucurbitacins, some of the bitterest compounds and extremely toxic to most

invertebrate and vertebrate herbivores, are well known phagostimulants for diabroticite beetles (Peterson and Schalk 1985, Deheer and Tallamy 1991). It was suggested that selective breeding had either removed or reduced these extremely bitter cucurbitacins to very low levels in leaves and fruits of crop plants (Metcalf et al. 1982). Nugent et al. (1984) reported that bitter seedlings of muskmelon were observed to be more susceptible to feeding by D. balteata than the non-bitter seedlings. The genetic study of resistant materials showed that two recessive genes, bibi (non bitter genes) and cblcbl, were responsible for the muskmelon seedling resistance. The bi gene in cucumber may be the major reason for its resistance to D. balteata adults (Da Costa and Jones 1971). Six







18
varieties of cantaloupe were evaluated for their antixenosis to D. balteata (Overman and MacCarter 1972). One cultivar AC 67-59 was found to be less preferred by the adults. Soybean plant introductions were screened for resistance to two beetles, including D. balteata, in a no-choice laboratory bioassay. Several introductions were found to express a greater degree of resistance to this beetle than others (Layton et al. 1987). Resistance in sweet potato cultivars to larval D. balteata was confirmed by Schalk et al. (1986). Whole roots had the highest resistance level, the cortex was intermediate, and stellar tissue was the least. Antibiosis tended to be expressed more strongly in the early and late season than in the mid-season. In addition, methods to evaluate sweet potatoes for resistance to D. balteata in the field were developed (Schalk and Jones 1982).

Different screening methods have been used to evaluate the resistance of plant genotypes against D. balteata. A rating scale was the most commonly used method for measuring resistance. A subjective rating of D. balteata damage was used in the field to study the resistance in soybean based on the following scales: low, no or very few feeding holes per leaflet on upper three fully expanded leaves; moderate, one to two feeding holes per leaflet on upper three fully expanded leaves; and high, three or more feeding holes per leaflet on upper three fully expanded leaves (Layton et al. 1987). A rating scale from 1 to 5 (1=100% consumed, 2=75% consumed, 3=50% consumed, 4=25% consumed, and 5=no feeding) was applied to screen for resistance in muskmelon to this beetle (Nugent et al. 1984). In order to minimize the effects of plant size on resistance ratings, a leaf disk technique was used and the proportions of the disks consumed by the beetles after 48 h at room temperature were rated visually. Resistance to three species of cucumber beetles including D. balteata was studied in cucumber







19


seedlings using a 1 to 3 rating scale (0=no feeding, 3=severe damage) (Da Costa and Jones 1971). The extent of feeding damage was measured on a scale of 0-9 to study the susceptibility in melon (Coudriet et al. 1980). Schalk and Jones (1982) developed a method to artificially infest sweet potatoes with eggs of D. balteata to evaluate resistance in this plant. Injury and damage ratings were obtained by counting the total number of holes per root and also by visually rating the damage to each root (a 1 to 5 scale). Layton et al. (1987) used leaf area consumption in addition to a rating scale to determine the resistance level in soybean. The leaf area consumed was then multiplied by the specific leaf weight (SLW=dry weight/leaf area) to obtain adjusted dry weight consumed, which was considered to give a more accurate measure of feeding rate in cases where leaf thickness may differ.

Host Plant Resistance in Lettuce

Plant Secondary Chemicals

Lettuce contains various identified chemical compounds such as terpenes

(sesquiterpene lactones), sterols (campesterol, stigmasterol, ect.), flavonoids and other phenolics (caffeic acid, dicaffeoyltartaric acid, rutin), and alkaloids (hyoscyamine) (Gonzalez 1977). Two triterpenes, the guaianolides lactucin and lactucopikrin, were isolated from dry latex of L. virosa, but lactucopikrin was rapidly degraded by enzymes in fresh latex (see references cited by Gonzalez 1977). Sharples (1964) found that chlorogenic acid, isochlorogenic acid, caffeic acid, glycosides of quercetin and kaempferol existed in lettuce leaves. A large concentration of chlorogenic acid was found in lettuce seeds (Butler 1960). The synthesis and accumulation of phenolic compounds are important aspects of secondary plant metabolism, and the oxidation of









phenolic acids to quinones by the action of PPO is responsible for the darkening of many fruits and vegetables when mechanically damaged (Cole 1984). In particular, chlorogenic acid is one of the substrates for PPO and POD enzymes (Matheis and Whitaker 1984). These enzymes can rapidly convert chlorogenic acid into chlorogenoquinone, a highly reactive molecule known to covalently bind to nucleophilic

-NH2 and -SH groups of molecules such as amino acids and proteins. This binding reduces leaf nutritional quality to the herbivore (Matheis and Whitaker 1984). Cole (1984) used high performance liquid chromatography, gas chromatography and UV absorbance to investigate the presence of certain compounds in certain lettuce cultivars and the lettuce root aphid, Pemphigus bursarius (L.). Isochlorogenic acid was found to be the only caffeoylquinic acid detected in quantity and there was a greater concentration in resistant than in susceptible cultivars. The first enzyme in the phenyl propanoid pathway, PPO, was more active in resistant cultivars. The higher level of browning in resistant cultivars was not related to the level of PPO activity but to the higher levels of isochlorogenic acid. The concentrations of caffeoylquinic, chlorogenic and isochlorogenic acids were used as criteria for screening lettuce breeding material for resistance to the lettuce root aphid, P. bursarius, through UV fluorescence (Cole 1987). Genetic Resistance

Genetic resistance of lettuce to the lettuce root aphid and several leaf aphids has been extensively investigated in the Netherlands. Almost complete resistance to the leaf aphid, Nasonovia ribisnigri (Mosley), was found in Lactuca virosa L., and transferred to L. sativa through a series of interspecific crosses (Eenink et al. 1982). This resistance is governed by the dominant Nr gene which always locates at the same locus (Eenink and







21


Dieleman 1982). Nr also confers partial resistance to the green peach aphid, Myzus persicae (Sulz), but has no effect on Macrosiphumn euphorbia (Thom.) (Reinink et al. 1989). Two lettuce cultivars were used to investigate the inheritance of resistance to lettuce root aphid P. bursarius (Ellis et al. 1994). Resistance was found to be controlled by a single dominant gene, which is linked to the downy mildew resistance gene Dm6. So far, no single source of resistance was found with resistance to all the aphid species. Isozyme electrophoresis is a potential tool for the identification of genetic markers in practical plant breeding programs (Tarksley and Orton 1983). Cole et al. (1991) surveyed allozyme pattern in wild populations of four Lactuca species by routine polyacrylamide gradient gel electrophoresis and found that the resistance of L. sativa to the lettuce root aphid was associated with specific allozyme bands. Nutritional Quality

There is some evidence to show that the availability of aromatic amino acids to herbivores is limited and that insects accumulate these substances for their growth (Bernays and Woodhead 1984, Rahbe et al. 1990, Mollema and Cole 1996). Phenylalanine was one of the limiting nutrients for the fifth instar nymphs of Schistocerca gregaria (Forskal) when feeding on lettuce. Supplementing this amino acid invariably increased the efficiency of ingested food conversion to body substance (Bernays and Woodhead 1984). Mollema and Cole (1996) analyzed four important crops including lettuce with unknown levels of resistance to thrips and found that a highly significant positive correlation existed between aromatic amino acid concentrations in leaf protein and thrips damage, regardless of crop species. Electrical penetration graphs of N. ribisnigri feeding on resistant and susceptible lettuces showed a large reduction in









the duration of food uptake on the resistant line, leading the authors to suggest that resistance was located in the phloem vessel. Both mechanical blocking of the sieve element after puncturing and a difference in composition of the phloem sap are possible resistant factors (van Helden and Tjallingii 1993). Investigation of phloem sap collections from L. sativa by three different methods (stylectomy, honeydew collection and EDTA chelation) found that the sugar yield from the resistant lettuce was around 30% lower than that on the susceptible plant (van Helden et al.1994a ). However, no difference in amino acid composition was found in the phloem samples from near isogenic susceptible and resistant lines of lettuce (van Helden et al. 1994b) except that the total concentration of amino acids was 50% lower in the resistant lines compared to the susceptible line. Sucrose was the main sugar in lettuce phloem, but fructose and glucose were also detected. HPLC chromatograms showed numerous unidentified secondary plant compounds in honeydew and EDTA samples, but these compounds were not identified and had no evidence to show that these compounds were involved in resistance. Nuessly and Nagata (1994) found that there were differential responses in the stipple rate by L. trifolii among four romaine lettuce cultivars (Floricos 83, Parris island Cos, Tall Guzmaine and Valmaine). Females of L. trifohii preferred Tall Guzmaine for oviposition, and lived significantly longer on this cultivar (Nagata et al. 1998). An investigation of leaf surface morphology did not find any obvious difference among the four lettuce cultivars, and no physical barriers were found to be responsible for the difference in stipples by adult leafminers (Nuessly, personal communication). It was suggested that a nutritional factor may be involved in host plant resistance in lettuce to the leafminer (Nagata et al. 1998).







2 ~


Possible Evidence of Inducible Resistance

Plant hypersensitivity is a term primarily used by plant pathologists to describe a response to infection by pathogens as well as to many nonpathogenic stimuli, which encompasses all morphological and histological changes that elicit the premature dying or necrosis of the infected tissue (Fernandes 1990). However, several hypersensitive reactions were also found in plants against insect herbivores such as galling insects, bark beetles, adelgids, and siricids (Fernandes 1990). Some evidence indicates that lettuce responds hypersensitively when challenged by the downy mildew fungus, Brenia lactucae Regel, resulting in highly localized accumulation of phenolics around penetration points. In addition to localized deposition of phenolics, phenolic esters identified as dicaffeoyltartaric and chlorogenic acids increase in concentration during incompatible interactions (Bennett et al. 1996). A second major component of the resistance response in lettuce to this fungus is the localized accumulation of the sesquiterpenoid lactone phytoalexin lettucenin A when a number of cells undergo the hypersensitive reaction because of B. lactucae infection (Bennett et al. 1994). Cells undergoing rapid hypersensitive reactions subsequently collapsed and became brown. Lettucenin A was significantly increased within one day after inoculation, but little change in concentration occurred between 1 and 7 days.

Research Goals

Host plant resistance to insect pests is a vital component of integrated pest management programs in agricultural systems. The development and use of plant cultivars with resistance to insect pests requires continuous identification of resistant and







24

susceptible germplasm, characterization of suitable resistant sources, and understanding the mechanisms of these resistances.

D. balteata is a polyphagous insect pest occurring throughout the growing season in Florida, which can cause serious problems in lettuce production. Since chemical control is costly and only partially effective, the development of plant resistance is important for controlling this herbivore. Use of cultivars resistant to adult D. balteata also can reduce the need for insecticide applications. Plant resistance to adult D. balteata has been found in muskmelon (Nugent et al. 1984), cucumber (Da Costa and Jones 1971), sweet potato (Schalk and Jones 1982, Schalk et al. 1986), and soybean (Layton et al. 1987). Preliminary studies showed that there were differences in feeding damage by D. balteata adults among four lettuce cultivars so that it is possible to screen cultivars for resistance against D. balteata (Nuessly, unpublished data).

The objectives of this research were to develop effective bioassay procedures to screen lettuce for resistance to D. balteata, compare different resistance levels among four romaine lettuce cultivars, examine effects of resistance on D. balteata fitness, and determine physical and chemical bases for this resistance.















CHAPTER 2
RESISTANCE IN ROMAINE LETTUCE CULTIVARS TO ADULT DIABROTICA BAL TEA TA


Introduction

Lettuce, Lactuca sativa L., is one of most important vegetable crops grown in the United States in terms of quantities produced and consumed (Ryder, 1998). As a cultivated crop, lettuce also serves as a favorite plant food to many insect pests including the banded cucumber beetle, Diabrotica balteata LeConte (Nuessly and Nagata, 1993), a polyphagous species with a host range of more than 50 plant species in 23 families (Saba 1970). Economic damage by D. balteata has been reported for other crops, such as corn and peanut (Elsey 1988), common bean (Cardona et al. 1982), soybean (Layton et al. 1987), sweet potato (Schalk et al. 1986), and cucurbits (Da Costa and Jones 1971). Nuessly and Nagata (1993) reported that adult D. balteata caused serious damage to lettuce plants, becoming a concern for lettuce growers in south Florida. In order to keep their crops insect free and to reduce economic losses, lettuce growers frequently spray with insecticides. However, the excessive dependence upon pesticides for insect control is associated with many economic, ecological, and environmental disadvantages. Moreover, general feeding habits and widespread dispersal movements of the adults make control of this insect more difficult than if they were more sedentary. Therefore, problems in controlling this pest have underlined the need to find alternative control


25







26


methods such as host plant resistance, which provides many benefits including its compatibility with other control methods in integrated pest management (Smith 1989).

Host plant resistance in lettuce to the lettuce root aphid, Penphigus bursarius L. (Ellis et al. 1994), the leaf aphid, Mvzus persicae Sulz and the green lettuce aphid, Nasonovia ribisnigri Mosley (Eenink and Dieleman 1982), and the potato aphid, Macrosiphun euphorbiae Thomas (Reinink et al. 1989) has been studied for many years. Almost complete resistance to N. ribisnigri was found in L. virosa, and this trait transferred to L. sativa (Eenink et al. 1982). Nuessly and Nagata (1994) found differential feeding response by the serpentine leafminer, Liriomyza trifolii (Burgess), among four romaine lettuce cultivars, including 'Tall Guzmaine' and 'Valmaine'. Significant differences in fecundity, in the total number of stipples (ovipositional puncture wounds in the leaf surface), and in adult L. trifo/ii longevity were observed under laboratory conditions among the four lettuce cultivars (Nagata et al. 1998). Plant resistance to D. balteata has been found in muskmelon (Nugent et al. 1984), cucumber (Da Costa and Jones 1971), cantaloupe (Overman and MacCarter 1972), sweet potato (Jones et al. 1976, Schalk and Jones 1982), and soybean (Layton et al. 1987), but no previous research has been done on resistance of lettuce to D. balteata.

Preliminary studies demonstrated that there were differences in feeding damage by D. balteata adults among four romaine lettuce cultivars, 'Tall Guzmaine' (TG), 'Parris White' (PW), 'Short Guzmaine' (SG), and 'Valmaine' (Val) in the field (Nuessly, unpublished data), and that resistance may be under genetic control. Therefore, the objectives of this study were to evaluate four lettuce cultivars for resistance against D. balteata using leaf consumption in choice and no choice assays as the criterion, to







27


examine the effect of a resistant cultivar on the fitness of adult D. balteata, and to determine the mechanisms of resistance in lettuce.

Materials and Methods

Plants

Four romaine lettuce cultivars were selected and screened for resistance to D. balteata based on field observations (Nuessly, unpublished data) and on their pedigree (Fig. 2-1) (Guzman and Zitter 1983, Guzman 1986). Seeds of TG, PW, SG and Val were provided by R. T. Nagata (Everglades Research and Education Center, University of Florida, FL). Seeds of each cultivar were kept overnight in the laboratory in a petri dish lined with a wet filter paper at the bottom for better germination. Germinated seeds were planted in a transplant tray filled with a commercial soil mix (MetroMix 220, Grace Sierra, Milpitas, CA) and grown for 2 wk in a greenhouse with natural light. Seedlings were transplanted to 10-cm diameter plastic pots filled with MetroMix 220. Each plant was watered daily and fertilized weekly with 10 ml of a lOg/L solution of a soluble fertilizer (Peters 20-20-20, N-P-K, W. R. Grace, Fogelsville, PA) from transplantation until the end of the experiment. Plants with seven to eight fully expanded leaves were selected and used for all experiments which were conducted in the laboratory at 26 1C under artificial illumination (4, 110-watt fluorescent bulbs, cool white) with a photoperiod of 14:10 h (L:D).

Insects

A colony of D. balteata was established from a wild population of adults

collected from the field in Belle Glade, Florida in June 1996. Rearing and manipulation







28


methods were based on those described by Schalk (1986) and Anonymous (1996) with some modifications.

Adult cucumber beetles were held in a screen cage (30.5 x 30.5 x 30.5 cm) with a cloth sleeve in an incubator with a photoperiod of 14:10 h (L:D) at 25 VC, R.H. 70 10%. Every other day, adults were fed leaves from lima bean (Fordhook 242, Illinois Foundation Seeds, Inc., IL) grown in trays in the greenhouse and sliced sweetpotato tubers purchased from a grocery store. Two egg-collecting devices (8.5 cm diam. x 6.5 cm high) were provided for their oviposition and shaded by placing them under 0.47 L strawberry baskets turned upside down with lima bean leaves placed on the top to stimulate egg deposition. The egg collecting device was a round Rubbermaid� plastic food container (8.4 cm diam. x 7 cm high) covered with a tight-fitting lid into which a 6cm-diam. hole was cut and then covered with a mesh screen (0.1 x 0.2 cm). Three layers of cheesecloth followed by three layers of paper towel (Kimberly-Clark Co., Roswell, GA) and one layer of moistened cotton balls were held tightly agaist the inside of the lid by a layer of upright small vials (20 ml).

Cheesecloth with eggs was collected every 2 d and put into a petri dish with a

screened lid. The cloth was covered with moisten paper towels cut to fit the petri dish to prevent dehydration. Eggs were allowed to develop in the incubator as above for 4 d before they were surface sterilized with 500 ml Clorox solution (5.25% sodium hypochloride, The Clorox Co., CA) (30 ml/L in water) for I min in a cylindrical container (18 cm diam. x 7.5 cm high). Sterilized eggs were then washed with distilled water three times in the container and covered with a damp paper towel. The container was then covered with a plastic screen lid and put back into the incubator. Eggs usually







29


hatched after 7 d from the day of egg collection. On day 6, a small amount of sprouted corn seeds (H93 x FB37, Illinois Foundation Seeds Inc., IL) was placed in the container so there would be food for the emerging larvae.

Corn kernels covered with the newly hatched larvae were placed into a soil-free rearing container (32.5 x 17.2 x 10 cm) with germinated corn seedlings (H93 x FB37, Illinois Foundation Seeds Inc., IL). These rearing containers were maintained in a room with a photoperiod of 14:10 h (L:D) at 25 I0C. After I wk, the larvae were transferred into another rearing container with fresh germinated corn seedlings to ensure adequate food supply. Corn seedlings were prepared as follows: corn seeds were soaked overnight in a Clorox solution (16 ml/L in water), rinsed with tap water the following morning, and stored in a plastic food container (17.2 cm diam. x 15.0 cm high) in the refrigerator until used. Soaked corn seeds were evenly placed on one sheet of moistened preassembled germination paper (14 x 29 cm) (Anchor Paper Co., St. Paul, MN). A wick at each end of the paper was placed in a tray filled with fresh water directly underneath the rearing container. The corn seeds were then covered with a double layer of paper towel that was also thoroughly moistened with water. The rearing container was covered with a lid that had a screened opening (7.5 x 13 cm) for ventilation. Seeds were allowed to germinate and grow for 3 d before their use as food for larvae.

After larvae grew for 15 d, they were almost ready to pupate and were transferred gently using forceps into a pupation container (18 cm diam. x 7.5 cm high) filled with moistened commercial soil mix (MetroMix 220). The soil mix was autoclaved at 20 psi and 121"C for 99 min. The potting soil was kept moist enough that it did not dry out for the entire pupal stage, but it was never saturated. The pupation container was covered







30


with a screened lid covered with dampened towel to retain moisture. Adults emerged around 12 d later and were transferred into the screen cage (30.5 x 30.5 x 30.5 cm) mentioned above for the colony use. Beetles to be used for bioassays were separated by sex during the pupal stage. Female pupae have two papillae below the anal opening and males lack such papillae (Schalk 1986). These adults were collected within 24 h of emergence, transferred to a plastic box (13.5 cm diam. x 13.1 cm high) with a ventilated top and sides containing a moistened paper towel, and held in the incubator for another 24 h without food prior to bioassays.

Binary-choice Test

Beetles were first given a choice between pairs of different romaine cultivars to evaluate feeding resistance. Beetles were exposed to the fully expanded attached leaves within feeding arenas made from plastic petri dishes (8.9 cm diam.). Two round holes (2.9 cm diam.) 65 mm apart were cut from the bottom of the dish, and a ventilation hole (5.8 cm diam.) was covered with gauze material at the top of the dish. The feeding arena was attached to the upper surface of two leaves of different cultivars using hair clips. These leaves were matched for position and age on each plant. Two females and one male were released into each arena. After 48 h, the section of leaf containing the damaged site was cut from the whole leaf and placed between two sheets of acetate transparencies. Leaf sections were scanned (JADE 2, Linotype-Hell, Taiwan) and imported into an imaging program (ImagePC beta version 1, Scion Corporation, Frederick, MD) where area eaten or leaf area remaining was determined. Feeding preferences were determined by testing all possible pair-wise combinations of the four lettuce cultivars. Each pair-wise combination was replicated 10 times, and all six pair-







31


wise combinations were tested simultaneously. Leaf areas consumed in feeding bioassays were expressed as area eaten in square mm. The difference of leaf area consumption between cultivars was analyzed by paired (-test using Proc MEANS (SAS Institute 1999).

No-choice Test

Resistance to beetle feeding was next tested in no-choice tests using fully

expanded attached leaves. One pair of beetles was confined on the seventh leaf (from the cotyledon) of a plant using a clip cage with a 4 cm diam. hole through which beetles accessed the upper leaf surface. The adults were allowed to feed for 48 h. Leaf area consumed was measured as described previously. This study was arranged as a randomized block design with each cultivar in each block. A block was the time when the four treatments (lettuce cultivars) were set up. Each block was replicated sixteen times. Leaf area consumed in the feeding bioassay was measured as above and analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD test with a significance level of a=0.05 (SAS Institute 1999) was used for post-hoc means separation. Fitness Bioassay

Beetle weight change and egg production were used to bioassay fitness effects of feeding on the four romaine varieties. Thirty, 48 h-old unfed females were randomly chosen from the colony and weighed individually before testing. These beetles were then released together with 15 males into a plastic cylindrical cage (60 cm high x 25 cm diam.) with ventilated top and sides containing two plants of the same cultivar. The shortest period from emergence to first egg deposition for D. balteata observed in the lab was 10 d. Therefore, adult beetles were allowed to feed for 10, 13 or 16 d without







32

providing an oviposition apparatus. New plants were constantly provided for the adults based on the level of damage in each cage except for the cage with Val, where Val plants were changed every 3 d because they were hardly damaged. At the end of the experiments (either 10, 13 or 16 d later), living beetles were counted and weighed individually. Beetles were stored in a freezer (-9'C) until they could be dissected under a dissecting microscope to determine the number of fully developed eggs in their ovaries. The experimental design was a 2 x 2 factorial with cultivar having four levels (TG, PW, SG, Val) and feeding time having three levels (10, 13, 16 d). The effects of cultivar on beetle egg production, weight change, feeding time and their interaction were analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD with a significance level of aX=0.05 (SAS Institute 1999) was used for post-hoc means separation. Starvation Test

Changes in egg production and adult weight may be associated with antibiotic or antixenotic mechanisms. Therefore, for purposes of comparison with feeding tests, starvation tests were conducted as an approximation of extreme antixenotic or poor food nutrition mechanisms on fitness parameters. The experiment was designed as a randomized complete block design. Four cylindrical cages were set up at one time as a block. Twenty, 48 h-old-unfed females along with 10 to 25 males were released into each plastic cylindrical cage (60 cm high x 25 cm diam.). Each block was replicated three times. Each cage received the same number of adults in each block. Each cage contained either one Val plant, one TG plant, one Val plant totally covered by a cloth, or one TG plant totally covered by a cloth. All cages contained four small water containers (4.5 cm high x 3.5 cm diam.). The cloth-covered plants were inaccessible to beetle









feeding but provided humidity equal to that in the other two treatments. The plant pots were also covered by plastic bags to prevent egg deposition into soil. Plants in all treatments were changed every 3 d including the cages with covered plants. After 7 d, living beetles were counted and females were dissected to check their ovary development and egg production. The total percentage of adult survival, the percentage of female survival and the percentage of females with yolk deposition in their ovaries in each treatment were subjected to arcsin(sqrt(x)) transformations before they were analyzed by Proc GLM. The number of mature eggs/female was also analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD with a significance level of a=0.05 (SAS Institute 1999) was used for post-hoc means separation.

Results

Binary-choice Tests

Adult D. balteata ate significantly less from Val leaves than from leaves of the other cultivars when they were given a choice between Val and SG, Val and PW, or Val and TG (Table 2-1). However, the beetles ate over two times as much surface area of Val when it was paired with PW as when it was paired with the other three varieties. D. balteata consumed almost 10 times more TG foliage than Val foliage when they were given a choice (Table 2-1). There was no significant difference (P >0.05) in feeding damage between TG and PW, TG and SG, or SG and PW. Beetles ate numerically more surface area of TG than of PW or SG in binary-choice tests, but the difference was not significant at P=0.05. D. balteata feeding damage among the four tested cultivars appeared to be in the order Val






34


Table 2-1. Lettuce leaf area consumed by three adult D. balteata in 48 h when presented a choice between two cultivars
Choice Cultivar' N Leaf area eaten (mm2) 2 SEM Pr> ITIb


Val SG


Val TG


Val PW TG PW TG SG SG PW


0.0019


0.0001


0.0001


10 10 10 10 10 10 10 10 10 10 10 10


43.2 5.7 131.7 19.9 23.8 6.0 235.0 33.9 122.6 26.6 393.7 47.8


409.2 49.8 293.6 65.8 190.8 35.9 120.8 12.7 128.9 37.0 174.5 18.1


0.2396


Val-SG


Val-TG


Val-PW


TG-PW


TG-SG


SG-PW


0.1908


0.1235


a TG=Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. b P value from paired t-test.









No-choice Test

Significant differences in leaf consumption were observed among the four

cultivars in no-choice feeding tests (F=52.02; df=3,56; P=0.0001) (Table 2-2). Adults consumed the greatest amount when confined on TG, followed by PW, then SG and Val. Therefore, the order of feeding damage observed in the binary-choice tests was confirmed by the results of the no-choice tests. There was no significant difference in area consumption between SG and Val at the 5% level, even though adult D. balteata ate just over half as much Val as SG. The beetles did some damage on Val, but the amount consumed was less than one-sixth of that on TG. Fitness Bioassay

The mean number of mature eggs produced per female significantly differed after feeding on the four cultivars for 10 d (F=66.02; df=3, 134; P=0.0001), 13 d (F=131.81; df=3, 134; P=0.0001), and 16 d (F=91.93; df=3, 134; P=0.0001) (Table 2-3). The length of feeding time (i.e., 10, 13, 16 d) also significantly affected egg production (F=9.18; df=2, 134; P=0.0002). There was a significant interaction between the lettuce cultivars and the length of feeding time on egg production (F=6.26; df=6, 134; P=0.0001). Egg production was significantly higher for adults feeding on PW than on the other varieties for 13 and 16 d. After 13 d, beetles that fed on PW produced a mean of 112 eggs compared to 93 on TG and SG and none on Val. No fully developed eggs were found in the ovaries of female beetles reared on Val for 10, 13 or 16 d. Furthermore, of the 42 surviving females out of 90 originally feeding on Val that were dissected, only three had some small undeveloped eggs after 13 or 16 d of feeding. In contrast, the egg production of beetles on the other three cultivars was very high. Egg number increased to the







36


Table 2-2. Mean leaf area consumed per pair of adult D. balteata in 48 h confined on four lettuce cultivars in a no-choice situation
Cultivara N Leaf area eaten (mm2)/pair SEM


TG 14 366.0 31.7 a

PW 15 304.21 24.0 b

SG 15 93.8 15.3 c

Val 16 53.5 8.3 c

a TG=Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.







37


Table 2-3. Number of fully developed eggs in ovaries of D. balteata females confined for different lengths of time on one of four lettuce cultivars Cultivarsa Time (Days) (Mean SEM)
10 13 16

TG 86 4a 93 6b 87 5b

PW 65 6 b 112 4a 115 l0 a

SG 56 11b 93 8b 76 12b

Val Oc Oc Oc

aTG-Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.







3S
highest on day 13 and decreased slightly on day 16. On day 13, all of the females on TG were full of mature eggs, but on day 16, the percentage of females with fully developed eggs had dropped to 70%. This decrease in the number of mature eggs in the ovaries of female beetles was also found on SG but not on PW after feeding for 16 d.

The weight of females reared on Val was significantly lower than that of females reared on TG and PW (F=7.15; df=3,337; P=0.0001) for 10 d (Table 2-4). The weight of females reared on Val was also significantly lower than that of females reared on TG, PW and SG for13 d (F=32.93; df=3, 227; P=0.0001) and 16 d (F=21.71; df=3, 227; P=0.0001). Females feeding on TG gained weight, ranging from 6.24 to 9.49 mg, while females feeding on Val gained only 1.31 to 2.61 mg. The highest weight gain occurred on day 13 for each cultivar.

Percentage of mortality for females on TG ranged from 7 to 33% when females were allowed to feed for 10 to 16 d. This range was lower than that on Val with a range from 40 to 73%. On PW and SG, the ranges of percentage of female mortality were from 27 to 47% and from 17 to 47%, respectively. Starvation Test

The percentage of D. balteata still alive on TG (mean of 79%) at the end of the 7d experiment was the highest among the four treatments (Table 2-5). When D. balteata had access to only water, its mortality was equally high on the TG-covered (11%) and the Val-covered treatments (7%). In each block, 3.3% to 20% of beetles survived when no food was available 7 d after the experiment was initiated. Therefore, all treatments were stopped in order to have some data in treatments with cloth-covered plants. Female







39


Table 2-4. Mean weight per female D. balteata feeding on four lettuce cultivars for different lengths of time
Cultivars Time (Days) (mg SEM)
0 10 13 16

TG 11.1 0.16 a 19.8 0.9 a 20.0 0.5 a 18.2 0.9 a

PW 11.5 0.2 a 17.0 0.5 b 20.5 0.5 a 20.3 0.6 a

SG 10.4 0.2 a 13.0 0.6 c 18.2 0.5 b 16.5 0.4 ab

Val 11.0 0.4 a 12.5 0.8 c 13.7 0.4 c 13.3 0.7 c

aTG-Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.







40


Table 2-5. Results of starvation test in which adult D. balteata were confined with food or no food for 7 d (Mean SEM)
Treatment Total Female Percentage of females Numbers of mature
survivorship % survivorship % with yolk deposition eggs/female

TG 79 2a 62 9a 97 3a 95 6a

Val 49 3b 40 3b 18 9b 0.0 b

TG-covered 11 2c 12 3b 0.0 b 0.0 b

Val-covered 7 3 c 10 3 b 0.0 b 0.0 b


TG- or Val-covered means plants of these varieties were covered by a cloth so that beetles could not access them as a food source. Data within a column followed by the same letter are not significantly different by Tukey's mean separation test at the 0.05 level.







41

survival was the highest on TG, followed by Val, TG-covered, and Val-covered (Table 25). Of all the females dissected in each treatment, all of them except one female on TG had at least yolk deposition in their oocytes, but only 18% of females surviving on Val had yolk deposition after 7 d. The ovaries of females without food had no yolk deposition and they were transparent, looking much like the ovaries of newly emerged females. Mature eggs were found in 20% of the females feeding on TG, averaging 95 eggs/female with a range from 74 to 119.

Discussion

There are several possible hypotheses that may explain the resistance in Val

lettuce. The four romaine cultivars tested are genetically related to each other so that it is possible that the resistance is under genetical control. This kind of resistance may be expressed by physical or chemical defenses.

Binary-choice and no-choice tests indicated that Val exhibits the highest degree of resistance among the four lettuce cultivars tested, and that TG is the most susceptible cultivar to D. balteata damage. The minimal feeding damage on Val indicates that it may contain deterrents or lack feeding stimulants, either at the leaf surface or internally. This feeding will be further investigated in following chapters. Val has been reported to be nonpreferred over TG by L. trifolii, both in choice and no-choice tests, when four lettuce cultivars including TG and Val were provided as the only food and oviposition source (Nuessly and Nagata 1994, Nagata et al. 1998). Female leafminers survived significantly longer and produced more pupae on TG than leafminers reared on the other three cultivars, including Val (Nagata et al. 1998).







42


That Val consistently and significantly exhibited resistance compared with TG was also seen in the fitness bioassay. Females that feed on Val had the lowest weight among beetles that fed on the four cultivars (Table 2-4). Moreover, females feeding on Val produced no mature eggs in three feeding periods when they had no choice of host plant (Table 2-3). D. balteata adults did not feed much on Val (Table 2-1, Table 2-2) due to the possible presence of lettuce physical or chemical defenses, causing beetles not to have enough nutrients to produce eggs. In contrast, D. balteata released on TG fed significantly more and generated more eggs. Egg production declined at 16 d in females fed on TG, PW and SG, which may have been due to egg reabsoption by females who were not provided an oviposition site, or the eggs may have been deposited before dissection. A few eggs were found in a slit on the plastic bag, which was used to cover soil.

Based on the starvation test (Table 2-3), most D. balteata could not survive

without food for 7 d. This was dramatically less than its normal longevity of at least I month based on our observations when supplied with sufficient food. Because their survival was in jeopardy, living females were not able to develop their reproductive systems. Almost 50% of adult D. balteata did survive on Val for 7 d by feeding on yellow leaves, major veins of fully expanded leaves, and on dead beetles, but most of them still did not develop eggs. This finding confirmed that females on Val have no mature egg development because of their inability to use the resistant cultivar Val as a food. It is possible that Val has some antixenotic characters which keep D. balteata from using it as a host, resulting in the size of its next generation dramatically decreasing as compared with that on the susceptible cultivar TG. When D. balteata was given a choice







.4 3


between TG and Val, some biophysical or biochemical factors from Val may force it to choose TG over Val.

D. balteata were not randomly distributed on lettuce plants; they liked to gather together and feed on a single plant before they moved onto another plant in a cage. Such feeding behavior contributed to the large standard deviations observed in many experiments. This may be due to the release of pheromones given off by the feeding insects (Regnier and Law 1968, Schalk et al. 1990). The same feeding phenomennon was also found when D. balteata fed on cucumber (Da Costa and Jones 1971). Feeding together may also help adults to break down physical defenses possibly involved in lettuce resistance.

Differences in feeding damage by D. balteata appear to closely follow the pedigrees of the tested cultivars (Fig. 2-1). Val and SG showed similar levels of resistance when beetles had no choice of food. SG is the result of a cross between Val and another cultivar so it is possible that SG inherited the resistance from Val. However, other modifying characters may be also involved because SG was significantly preferred over Val when D. balteata had a choice. Feeding on Val was significantly increased when paired with PW in binary-choice tests. This suggested that PW might contain D. balteata feeding stimulants. Although TG and Val are related, TG was produced from a direct cross between SG and PW. Therefore, the susceptibility of TG was probably inherited from PW. This hypothesis was also supported by Nuessly and Nagata (1994) who proposed that the susceptibility of TG to L. trifolii was probably introduced with the cross to PW.







44


This study indicates the importance of breeding and selection processes in

developing resistant cultivars. Val was the leading cultivar of romaine lettuce grown in Florida organic soils, but it was susceptible to thermodormancy, premature bolting, lettuce mosaic virus and corky root rot (Guzman 1986). Therefore, plant breeders began to develop and select TG with improved characteristics over Val. However, insect resistance was not included in the selection criteria (Nuessly and Nagata 1994), and unfortunately TG is a suitable host for L. infolii and D. balteata.

Foliar feeding by D. balteata causes problems in lettuce production such as

decreasing the photosynthetic capacity of the leaves, introducing frass into the heads, and opening the plants to pathogenic infection. These problems result in yield losses and increased production costs. In the southern United States, D. balteata adults can be found during much of the year (Schalk 1986). Moreover, D. balteata has a very high reproductive potential; oviposition can last from 2 to 8 wk, and one female can lay 849 eggs in her lifetime (Pitre and Kantack 1962). Because chemical control of the soilinhabiting larvae of this beetle is unreliable and incapable of reaching all the larvae and eggs in the soil (Schalk et al. 1986), an ideal way to reduce D. balteata populations may be by controlling adult beetles (Schalk et al. 1990). This study showed that Val has a dramatic potential ability to retard D. balteata population development by decreasing the weight and vigor of the adults, making them more susceptible to other abiotic and biotic mortality factors. The detailed mechanism of resistance in lettuce to the D. balteata needs to be further studied to help scientists develop new cultivars with desired characters, including insect resistance. Host plant resistance is of particular importance







45
to control D. balteata with its multiple generations per year in Florida. This strategy will reduce the impact of this insect on lettuce and reduce pesticide usage.







46


FL 746-Parris White

FL 1142 -- Tall Guzmaine
Short Guzmaine Valmaine

Valmaine _ Floricos 83


Fig. 2-1. Pedigree relationship of lettuce cultivars.














CHAPTER 3
ROLE OF THE LEAF SURFACE OF LETTUCE IN RESISTANCE TO DIABROTICA BALTEA TA

Introduction

The outer surface of all plants is coated with layers consisting of a lipid polymer and a mixture of extractable lipids, sometimes referred to as epicuticular waxes, which reduce water loss and help to block the entry of pathogenic fungi and bacteria (Esau 1976). The chemical compositions of these layers vary among species, among genotypes within a species, and among parts within a plant (Eigenbrode and Espelie 1995). These variations are unlikely to affect greatly the ability of the surface lipids to act as a barrier to moisture loss, but are likely to have a variety of ecological influences, including mediation of insect-plant interactions (Eigenbrode and Espelie 1995). The plant surface is the first physical contact between an insect and a plant when the insect lands or touches the plant (Schoonhoven et al. 1998). Important environmental cues, which lead insects to make decisions to feed or oviposit on a host plant, are associated with the leaf surface contact sensory cues (Derridi et al. 1996). Other than epicuticular waxes, various physical or chemical factors encountered at the plant surface sometimes are deployed in glandular and nonglandular trichomes or in resinous secretions (Southwood 1986, Oghiakhe 1995, Lin et al. 1986, Hori 1999, McAuslane et al. 1997).

The plant surface, especially its chemistry, constitutes the first line of resistance to insects (Schoonhoven et al. 1998). Contact with these surface chemicals often suffices to prevent insects from further investigation or usage of the plant (Schoonhoven et al.1998).


47







48


Epicuticular waxes on the plant surface are known to influence insect feeding, oviposition and movement either by physical (Stork 1980, Bernays et al. 1983) or by chemical mechanisms (Adati and Matsuda 1993, Eigenbrode and Espelie 1995, Eigenbrode and Shelton 1990, Eigenbrode et al. 1991, Lin et al. 1998). Epicuticular waxes are usually extracted from a plant surface using an organic solvent such as chloroform, methylene chloride or hexane (Bakker et al. 1998). Such extracts often contain sugars, amino acids, and secondary plant substances such as glucobrassicin (a glucosinolate), furanocoumarins and alkaloids (Table 3.7 in Schoonhoven 1998). A resistant cultivar of Sorghum bicolor (L.) Moench was shown to become more acceptable to nymphs of Locusta nigratoria L. once waxes had been removed with chloroform (Woodhead 1983). Rinses of Solanum berthaultii Hawkes leaves with methylene chloride deterred Colorado potato beetle feeding when applied to S. tuberosum tuber and leaf disks (Yencho et al. 1994). Surface lipid extracts with hexane from a rice cultivar resistant to the brown planthopper, Nilaparvata lugens (Stal), deterred feeding and increased restlessness of this insect when the extracts were applied to the surface of susceptible plants (Woodhead and Padghm 1988).

Most plant waxes consist of a few major classes of aliphatic components

including n-alkanes, wax esters, free fatty alcohols and free fatty acids (Eigenbrode and Espelie 1995). Differences in epicuticular lipid composition were associated with resistance of cultivated plants to herbivores or with herbivore behavioral responses (Eigenbrode and Espelie 1995). For instance, y-hydroxybenzaldehyde was identified as an antifeedant present in the surface wax of seedling S. bicolor to nymphs of L. migratoria (Woodhead 1982). Reduced feeding by the spotted alfalfa aphid, Therioaphis







49


naculata (Buckton), was found in alfalfa genotypes with high levels of triacontanol (Bergman et al. 1991). Aphid resistant sorghums had higher levels of triterpenols in the surface wax than did susceptible sorghums (Heupel 1985).

On the other hand, plant surface chemicals may help some insects to recognize their specific host plants. Several species of chrysomelid beetles are stimulated to feed by leaf surface wax of their various host plants (Adati and Matsuda 1993, Lin et al. 1998).

The composition of the leaf surface of lettuce, Lactuca sativa L, has been reported recently (Pilipenko et al. 1994, Bakker et al. 1998). Pilipenko et al. (1994) reported that neutral lipids such as hydrocarbons, sterols and free fatty acids predominated on the lettuce leaf surface, followed by glycolipids and phospholipids. Esters of higher fatty acids and lower alcohols were also important components of the surface chemicals of lettuce. Leaf rinses, made with chloroform, methylene chloride, toluene, hexane, and a mixture of chloroform and methanol, all consisted of long-chain linear alcohols and minor amounts of fatty acids (Bakker et al. 1998).

In Chapter 2, both feeding choice and no-choice experiments showed that

Valmaine had the highest level of resistance to the banded cucumber beetle, Diabrotica balteata LeConte, among four lettuce cultivars (Tall Guzmaine, Short Guzmaine, Parris White, and Valmaine). Valmaine when fed upon had an ability to inhibit or delay normal development of the reproductive system of D. balteata. These results indicate that Valmaine may contain deterrents or lack feeding stimulants, either on the leaf surface and/or inside of leaves. The objective of this research was to determine the possible role of the leaf surface chemistry in lettuce resistance to D. halteata. Feeding bioassays were







50
conducted on solvent-treated (hexane, methylene chloride, and methanol) lettuce leaves and on palatable lima bean leaves to which leaf rinses were applied. The amounts of extractable lettuce leaf surface components were also quantified.

Materials and Methods

Plants

Based on the results in Chapter 2, Valmaine (Val, resistant) and Tall Guzmaine (TG, susceptible) were used to search for possible resistance factors on the leaf surface. Seeds of lettuce cultivars (provided by R. T. Nagata, Everglades Research and Education Center, University of Florida, FL) were germinated overnight in a petri dish lined with a moistened filter paper in the laboratory for improved and uniform germination. Germinated seeds were planted in a transplant tray filled with a commercial soil mix (MetroMix 200, Grace Sierra, Milpitas, CA) and grown for 2 wk in a greenhouse with natural light before transplanting into 10-cm diameter plastic pots filled with MetroMix 200. Each plant was watered daily and fertilized weekly with 10 ml of a IOg/L solution of a soluble fertilizer (Peters 20-20-20, N-P-K, W. R. Grace, Fogelsville, PA) from transplantation until the end of the experiment.

Fordhook 242 Lima bean seeds were purchased from a commercial company

(Illinois Foundation Seeds Inc., IL) and sown in seedling tray cells containing MetroMix 200 in the same greenhouse where lettuce was grown. Lima bean plants used for sustaining the beetle colony were watered daily and fertilized weekly with the same solution as used for lettuce plants after they reached the first-true-leaf stage. Lima bean plants used for experiments were watered daily but were not fertilized because only the first true leaves were used.







51


Insects

Adult banded cucumber beetles were obtained from a laboratory culture originally collected from the field in Belle Glade, FL in June 1996. Adults were reared in an incubator at 25 1C, 14:10 (L:D) photoperiod, and 70 10% humidity on lima bean leaves and sliced sweetpotato purchased from a grocery store. Larvae were reared on germinated corn seedlings in a laboratory at 25 IC, 14:10 (L:D) photoperiod as described in Chapter 2. Only unfed adults which had emerged within the past 48 h were used for all the assays.

Solvent Effects on Extraction

The sixth through eighth fully expanded leaves (1 = first true leaf) were cut from the plants of each lettuce cultivar in preparation for solvent extraction. Excised leaves were pooled and separated into groups of three leaves so that each group was treated as one replicate. The leaves from each group were immersed in hexane for 5 s, methylene chloride for 5 s, or methanol for 30 s at room temperature in a fume hood. The length of extraction depended on the time before chlorophyll was extracted into the solvent. After dipping, the leaves were flattened and placed between two sheets of acetate transparencies. The total leaf area extracted was determined by scanning the leaves (JADE 2, Linotype-Hell, Taiwan) and importing the scanned images into an imaging program (ImagePC beta version 1, Scion Corporation, Frederick, Maryland). The extracts were filtered through a Whatman No. 1 filter paper into a preweighed flask. The solvent was evaporated to dryness at room temperature in the fume hood to preserve low boiling compounds. The total amount of chemicals removed in each group was determined by weighing the residues and was expressed as the amount of material









removed per unit area. Data were analyzed as a 2 x 3 factorial design by Proc GLM (SAS Institute 1999), in which cultivar was treated as one factor with two levels, and solvent was treated as the other factor with three levels after data were checked to conform to assumptions of normality. Post-hoc means separation was conducted using Tukey's HSD test (cx=0.05, SAS Institute 1999). Feeding Response to Solvent-extracted and Normal Leaves

The seventh or eighth fully expanded leaves were selected from plants with eight fully expanded leaves for this assay. One pair of beetles was placed in a binary-choice feeding arena (described in Chapter 2) and was given a choice between a solvent-dipped leaf and a non-treated normal leaf, both still attached to the plants. After 48 h, the section of leaf containing the damaged site was cut from the leaf and placed between two sheets of acetate transparencies. Leaf sections were scanned and imported into an imaging program where area consumed was determined. Solvent-dipped leaves were immersed in methylene chloride for 5 s or in methanol for 30 s. Hexane was also tested, but phytotoxicity was so rapid and severe that the leaves were not suitable for the assay. Normal leaves were not dipped in any solvent. The feeding response of adult D. balteata was tested in all five possible pair-wise combinations between leaves with and without surface removal for each cultivar (Table 3-2). The control pair, TG vs. Val, was also conducted to monitor the resistance level. There were eight replicates for each combination. Area consumed on each leaf in each treatment combination was analyzed by paired -test using Proc MEANS (SAS Institute 1999).









Feeding Response to Crude Extracts in Binary-choice Test

Lima bean leaves were used as the substrate to test the effects of lettuce leaf wax crude extracts on D. balteata feeding. The first true leaves were cut from greenhousegrown lima bean plants and their petioles immediately immersed in beakers with tap water. A binary-choice test arena (described in Chapter 2) was used to expose a pair of adult D. balteata to two leaf areas on the same lima bean leaf for 24 h. Based on the results of experiments of feeding response to solvent-extracted and normal leaves (no significant difference in feeding between Val leaves and Val leaves with surface chemicals removed by methanol), only methylene chloride was chosen to extract lettuce leaf surface chemicals for the binary-choice test. The upper surfaces of two leaf areas were sprayed with either solvent or crude extract depending on the experiment design as described below. Extracts were applied at five different concentrations based on the natural concentration of methylene chloride-extractable chemicals on TG or Val leaf surfaces. The different concentrations were made from stock solution and diluted with methylene chloride so that 100 tl was applied to each confined area. Leaf area consumption was determined as described above and analyzed with paired t-test using Proc MEANS (SAS Institute 1999). A micro spray device (provided by H. T. Alborn, USDA, ARS, Gainesville, FL) was used to apply extracts. This device leads nitrogen gas through the horizontal tube to cause the solution to be drawn up through the vertical tube, resulting in formation of a fine spray. The solvent was allowed to evaporate completely in the fume hood before the beetles were confined on the leaves. The extraction process and concentrations used in the bioassays are described below.


5 3







54
Crude extracts. The sixth through eighth fully expanded lettuce leaves were cut from the plants of TG or Val. Extracts of leaf surface chemicals were prepared by dipping excised leaves in methylene chloride for 5 s. Extracts were filtered into a flask, concentrated to less than 5 ml under a nitrogen stream, then transferred into preweighed 5 ml glass vials. The solvent was continuously evaporated to dryness under a stream of nitrogen. The total amount of chemicals removed from Val or TG was determined by weighing the residues. The vials were then completely wrapped with aluminum foil and temporarily stored in a refrigerator freezer (-9'C) before bioassays. The stock solutions of crude extracts were made by redissolving the residues with methylene chloride in the vials.

Extracts from TG vs. solvent. The five concentrations chosen for TG were 13, 27, 41, 55, and 69 pig/cm2 . Among them, 27 [ig/cm2 was its natural concentration. One of two exposed areas confined by the binary-choice feeding arena was sprayed with 100 pl of an extract of designated concentration, the other was sprayed with only 100 Pl methylene chloride. Each combination was replicated 15 times.


22
Extracts from Val vs. solvent.- Five concentrations, 5, 11, 16, 22, and 27 pg/cm2, were also applied on lima bean leaf surfaces. Among them, 11 pg/cm2 was Val's natural concentration. One leaf area was treated with 100 pl extracts and the other with 100 pil solvent. Each combination was replicated 17 times.

Extracts from TG vs. from Val. Five pair-wise combinations between these

varieties were set up as follows: 11 pg/cm2 TG vs. 27 pg/cm2 Val, 11 pg/cm2 TG vs. 11
PgC2 2t/M I/M
ig/cm2 Val, 27 pg/cm2 TG vs. 11 jpg/cm2 Val, 27 g/cm2 TG vs. 27 pg/cm Val, and 19 jig/cm2 TG vs. 19 jig/cm2 Val. Two confined leaf areas on the same lima bean leaf were







55

sprayed with 100 pAl extracts of either Val or TG at the designated concentrations. Each combination had 16 replicates.

Feeding Response to Crude Extracts in No-choice Test

The same feeding arenas used in the binary-choice tests were used for this test,

but only one hole was clipped on a lima bean leaf sprayed with crude extracts from either TG or from Val. The other hole was covered to prevent beetles from escaping. Crude extraction, preparation and spraying strategy were done in the same manner as described above. Five different concentrations and 16 replicates for each concentration were used. The five concentrations applied on lima bean leaf surfaces for TG or Val were 6, 10, 17, 27, and 55 pg/cm 2. The experiment was set up and analyzed as a 2 x 5 factorial design with cultivar as one factor and concentration as the other factor using Proc GLM (SAS Institute, 1999). Post-hoc means separation was performed using Tukey's HSD test (cc=0.05, SAS Institute 1999).

Results

Solvent Extract Comparisons

The weight of surface components extracted per unit of leaf area was not

significantly different among the three solvents (F=2.97; df=2,18; P=0.0766) (Table 3-1). However, this statistic was strongly affected by the nearly identical values for TG crude extract weights among the tested solvents. Post-hoc tests (Tukey's HSD) run on Val data do indicate significant means separation between crude extract weights for methylene chloride and methanol. The weight of surface chemicals extracted was significantly different between the two cultivars (F=87.41; df=l, 18; P=0.0001). No interaction was observed between cultivar and solvent, which means that any differences in the weight of







56


Table 3-1. Total amount of crude material extracted from the leaf surface of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce using three solvents
Solvent Crude extracts (pg/cm2 SEM)
TG Val

Hexane 27.8 3.1 14.5 0.2

CH2Cl2 27.4 1.6 11.0 1.2

Methanol 27.8 0.4 18.8 1.7







57

components per unit of leaf area among three solvents are the same for both cultivars, and any differences between the cultivars are the same for each solvent (F=2.48; df=2,18; P=0.1 117). Therefore, data from three solvents on different cultivars were pooled to study the effect of cultivar on the weight of leaf surface chemicals removed. The weight of components extracted from TG, with the average of 28 Ipg/cm 2(mean SEM), was significantly higher than the weight of components extracted from Val with the average of 15 1 Ig/cm2 (mean SEM) per unit of leaf area (F=66.52, df=1, 22; P=0.0001). Feeding Response to Solvent-extracted or Normal Leaves

Solvent extraction of lettuce leaf samples led to increased leaf consumption by beetles on both TG and Val compared to untreated leaves. Beetles consumed at least twice as much lettuce leaf tissue from methylene chloride-extracted leaves of TG and Val (TG + S, Val + S) than from untreated leaves (TG, Val) of the same cultivars (Table 3-2). However, statistically significant differences (P < 0.05) existed only between Val and Val+S. Similar feeding patterns for TG versus TG +S and for Val versus Val + S were also found between methanol-dipped leaves and untreated leaves, but differences were not significant (Table 3-3). Removing leaf waxes and their constituents from Val led to large increases in feeding compared to untreated leaves. However, beetles still fed significantly more from treated and untreated TG leaves than from treated Val leaves. In the rest of the treatment combinations, regardless of whether leaf surface chemicals were removed by methylene chloride or methanol, adults still strongly preferred to feed on leaves of TG than on leaves of Val (Tables 3-2, 3-3). Methylene chloride extraction led to greater increases in beetle feeding than did methanol extraction, particularly for Val.







58


Table 3-2. Mean amounts of feeding by adult D. balteata on leaves of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce with and without surface layer removal by methylene chloride (CH2CI2)
Choice Cultivara N Leaf area eaten (mm2 SEM) Pr > TI


TG


TG-TG+S


TG+S


Val


Val-Val+S


Val+S


TG


TG-Val


TG+S-Val+S


Val TG+S

Val+S TG+S


TG+S-Val


Val


7

7


8

8


8

8


8

8


8

8


159.2 45.1 346.8 59.9 16.6 3.7 193.6 36.7 326.1 57.4 12.0 2.5 442.4 60.7 58.1 38.9 223.4 31.9 37.6 6.9


0.0938


0.0015


0.0009


0.0030


a TG+S = leaves from TG treated with CH2CI2. Val+S = leaves from Val treated with CH2Cl2.


0.0003







59


Table 3-3. Mean amounts of feeding by adult D. balteata on leaves of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce with and without surface layer removal by methanol
Choice Cultivar N Leaf area eaten (mm2 SEM) P > TI


TG


TG-TG+S


TG+S


Val


Val-Val+S


Val+S


TG


TG-Val


TG+S-Val+S


Val TG+S

Val+S


TG


TG-Val+S


Val+S TG+S


TG+S-Val


Val


12 12 12 12 12 12 12 12 12 12 12 12


254.7 29.3 317.7 26.4 54.5 11.1 95.0 27.6 310.9 38.7 22.1 5.7 372.0 35.5

40.3 9.4 343.5 25.0

96.4 41.2 401.0 32.3 34.7 9.7


0.0696


0.2111


0.0001


0.0001


0.0001


a TG+S = leaves from TG treated with methanol. Val+S = leaves from Val treated with methanol.


0.0001







60


Browning was observed on methylene chloride treated leaves more often than on methanol treated leaves.

Feeding Response to Crude Extracts in Binary-choice Test

There was no significant difference in adult feeding between control and extract treated leaves when TG leaf extracts in methylene chloride were applied to lima bean leaf surfaces, except at the natural concentration of 27 pg/cm 2(Fig. 3-1). At this concentration, less feeding occurred on TG extract-sprayed lima bean leaves compared to control leaves. TG extracts above 27 pg/cm2 led to reduced feeding, but it was significant (P=0.05) only at this concentration due to large variation in feeding rates between beetle pairs. TG below its natural concentration led to increased feeding, but it was not significant.

The only significant difference found between a lima bean leaf treated with Val surface extracts and a control leaf was at the level of 5 pig/cm2, which was lower than its natural concentration (Fig. 3-2). Spraying Val extracts at this level resulted in increased leaf area consumption. Extracts above this concentration did not result in any consistent effect on feeding on Val. Application of surface crude extracts of TG or Val to lima bean leaf surfaces at 1.5, 2, or 2.5 times their natural concentrations did not significantly stimulate or deter feeding by adult D. balteata (Fig. 3-1, 3-2), but caused sprayed lima bean leaves to be browned with greater frequency as the concentration applied increased.

No significant difference in feeding was found in pair-wise testing of methylene chloride extract of TG and Val on lima bean leaves at any concentration combination (Fig. 3-3). For example, when both of the natural concentrations of Val (11 pg/cm2) and TG (27 pIg/cm2) were sprayed on lima bean leaf surface, the amount of feeding on the







61


F6TG
250 00K


E 200
E

E
150 *
0

100


50


0
14 27 41 55 69

Crude leaf extracts applied (ptg/cm2 )

Fig. 3-1. Mean leaf area consumption by adult D. halteata exposed to
lima bean leaves treated with methylene chloride extracts of susceptible Tall Guzmaine (TG) or control solvent (CK) in a
binary-choice test. Vertical lines indicate 1 SEM. Asterisk
= significant difference between TG and control (CK) at
the 0.05 level by paired 1-test.







02















250 -_0 Val

s' 200 LICK


150


CI 0


0
S 0




5 11 16 22 27

Crude leaf extracts applied (pig/cm2 )

Fig. 3-2. Mean leaf area consumption by adult D. balteata exposed to lima
bean leaves with methylene chloride extracts of resistant Valmaine (Val) or control solvent (CK) in a binary-choice test. Vertical lines
indicated I SEM. Asterisk = significant difference between Val
and Ck at the 0.05 level by paired 1-test.







03


300

STG 250
2 _0 Val

200
E
150


100
C3
50
50


0
11-27 11-11 19-19 27-27 27-11
Pair-wise combinations of concentration

Fig. 3-3. Mean leaf area consumed by adult D. balteata on lima bean
leaves treated with methylene chloride extracts from either susceptible Tall Guzmaine (TG) or from resistant Valmaine (Val) in binary-choice tests. Vertical lines indicate I SEM.







64

lima bean leaf with TG extracts was 172.6 mm2, which was very close to the amount on the leaf with Val extracts (176.7 mm2).

Feeding Response to Crude Extracts in No-choice Test

No significant difference in feeding damage was observed between lima bean leaves applied with extracts from TG or Val in no-choice tests (F=0.36; df=l, 150; P=0.5471) (Fig. 3-4). However, extract concentration had a significant effect on adult feeding (F=3.30; df=4,150; P=0.0126). Adult feeding decreased linearly when methylene chloride leaf rinses were applied in increasing concentrations to lima bean leaf surfaces both for TG (F=6.78; df=1,150; P=0.0165) and for Val (F=5.73; df=1,150; P=0.0123). No significant interaction was found between concentration and cultivar (F=0.46; df=4,150; P=0.7671).

Discussion

The solvents, hexane and methylene chloride, used in this study were reported to be efficient for extraction of leaf surface chemicals of lettuce (Bakker et al. 1998). They used chloroform, methylene chloride, toluene, hexane, and a chloroform and methanol mixture to extract the leaf surface of L. sativa, and found that the distribution of wax compounds was very similar for the extracts made with various solvents, except for the chloroform and methanol mixture. Because the mixture of chloroform and methanol extracted chlorophyll and some internal cytoplasmic lipids, such as membrane sterols, they concluded that methanol or a combination of methanol with other solvents was not suitable for the extraction of leaf cuticular waxes of lettuce. However in this study, leaves treated with methanol had the least amount of browning and wilting compared to





















400 350 300 250

200 150 -


100


1 TG
EJVal


6


I


> j


C * a'- - ' ___


10


17


27


45


Crude leaf extracts applied (ug/cm2 )

Fig. 3-4. Mean leaf area consumed by adult D. balteata on lima bean
leaves treated with methylene chloride extracts of
susceptible Tall Guzmaine (TG) or resistant Valmaine (Val) in no-choice tests. Vertical lines indicate 1 SEM.


65


E

-o


0
0
S.

~-1


ILJ_







66

the other two solvents used. It seems that methanol alone may be another suitable solvent to extract polar leaf surface chemicals from lettuce leaves.

That leaves of lettuce with their surface chemicals removed were consumed much more, especially with the resistant Val, and that the total surface chemical removed by three different solvents was significantly different between resistant and susceptible cultivars, lead me to promote a hypothesis that leaf surface chemicals may have a role in expression of lettuce resistance. However, leaf surface extracts from Val were not deterrent to adult D. balteata when applied to lima bean leaf surfaces at various concentrations in binary-choice and no-choice tests. Leaf surface extracts from TG applied at increasing concentrations did not cause adults to feed more on lima bean leaves but actually decreased feeding similar to extracts of Val (Fig. 3-6). Therefore, these results suggest that leaf surface chemicals in lettuce do not explain resistance of Val to adult D. balteata. Reduced feeding with increased extract concentrations may be due to the increased chance of solvent browning on the leaf surface of lima bean, which resulted in the leaf losing moisture quickly and affecting adult consumption. High concentrations of extracts reduced solvent evaporation, causing solvent to remain in contact with the leaf surface longer than with lower concentration extracts.

One possible reason to explain the increased palatability of solvent-extracted leaves is that removal of surface chemicals with solvents disrupts the leaf surface, causing the leaves to lose water and internal pressure easily. Leaves with reduced internal pressure may be unable to deploy a physical response to beetle feeding, such as emission of latex or other vascularly associated chemicals. Methylene chloride wilted







67


lettuce leaves more easily than methanol, which may cause adults to feed more on methylene chloride-dipped leaves.

Another possible reason for increasing palatability of solvent-treated leaves is that disruption of cellular membranes by solvents may cause some plant secondary chemicals to come together with their specific degradation enzymes, resulting in the breakdown of deterrent compounds in solvent-dipped leaves. It is well known that many plant secondary chemicals against herbivore attack are also self-toxic (Schoonhoven et al. 1998). In order to prevent autotoxicity, plants usually store less toxic precursors of these highly toxic compounds or store toxic chemicals in cell compartments which are remote from metabolism (i.e. cell walls and vacuoles; Schoonhoven et al. 1998). Leaf browning occurred after leaves of TG and Val were dipped in methylene chloride for less than 5 seconds. Several oxidizing enzymes, such as polyphenol oxidase or peroxidases, are responsible for tissue browning by oxidation of phenolics to quinones at the surface of wounded plant tissue where cell compartmentation breaks down (Edwards and Wratten 1983, Dowd 1994). Therefore, it is possible that some secondary chemicals, possibly phenolics, are broken down by oxidative enzymes, resulting in increased feeding by D. balteata on leaves with surface removal by methylene chloride, although in some cases, oxidation of phenolics to quinones leads to greater deterrence of insect feeding (Beck and Reese 1976, Dowd 1994). A similar hypothesis was also proposed by other researchers for leaves that were dipped in non-polar solvents, such as chloroform or hexane (Keathley et al. 1999, de Boer and Hanson 1988). Keathley et al. (1999) found that chloroform dipping rendered leaves of Bradford callery pear palatable to the Japanese beetle, Popilliajaponica Newman, but extracts containing surface waxes were not







6S

deterrent. Dewaxed leaves of resistant Canna generates L. were more susceptible to the tobacco hornworm, Manduca sexta (L.), but reapplying leaf surface chemicals to the dewaxed leaves did not restore resistance (de Boer and Hanson 1988).

It is also possible that the solvent used for leaf surface extraction destroyed

chemical deterrents or deterrent physical characteristics of Val since reapplication of leaf surface chemicals on the lima bean leaves did not restore the resistance. However, leaves from Val treated with either methylene chloride or methanol still had strong resistance to feeding damage by D. halteata. Choice and no-choice experiments in which the leafsurface extracts of Val were put onto the lima bean leaf surface did not render the acceptable lima bean leaves unpalatable. Apparently, interior chemicals or other unidentified factors play a major role in resistance of Val. This possibility is investigated in Chapters 4 and 5.














CHAPTER 4
INVESTIGATION OF THE PHYTOCHEMICAL NATURE OF RESISTANCE IN
LETTUCE TO DIABROTICA BAL TEA TA USING ARTIFICIAL DIET Introduction

Plants are the richest source of organic chemicals on Earth, producing an

extraordinary array of secondary metabolites to protect themselves at different phases of growth (Schoonhoven et al. 1998). Secondary plant metabolites are not universally found in higher plants, but are restricted to certain plant taxa, or occur in certain plant taxa at much higher concentrations than in others, and are usually of little or no nutritional significance to insects (Schoonhoven 1972, Slansky 1992). The vital role of secondary plant metabolites in the defense system against insects and other enemies has been recognized as inhibition of food intake in the majority of plant-feeding insects by acting as feeding inhibitors (deterrents) and toxins (Panda and Khush 1995, Fraenkel 1959). A few specialized species, however, exploit these chemicals as stimulants (Fraenkel 1959). It has been estimated that the total number of secondary plant metabolites may reach 100,000 or more, but only about 10,000 of them have been chemically defined (Schoonhoven 1982).

Much of the research on plant-insect interactions has centered on the

identification of chemicals important in insect feeding by either applying plant chemicals to a plant or to an artificial substrate (Miller and Miller 1986). As a prerequisite to identifying the chemicals involved, it is essential to develop a reliable, quantitative and time-efficient bioassay for the response of insects to feeding stimulants and deterrents.


69







70


However, the screening of plant extracts for feeding deterrents and the isolation and purification of bioactive compounds are often very long, expensive, and tedious processes. Insect feeding bioassays especially require large numbers of insects to monitor all fractionation and purification steps, and only a small percentage of that effort produces a significant output because active substances can comprise only a few out of a large number of constituents (Escoubas et al. 1992). Several substrates, some natural and others artificial, have been used for testing feeding inhibitory activity of a chemical or chemical mixture (Miller and Miller 1986). For example, test chemicals can either be applied to natural food, such as whole plants (Erikson and Feeny, 1974), leaves (Lewis and van Emden 1986), leaf disks (Hsiao and Fraenkel 1968) or mixed with an insect's natural dried food (Bernays and Chapman 1977). Test chemicals can also be added to artificial substrates, such as sucrose-flavored agar (Alfaro et al. 1979), agar-cellulose (Ma 1972), filter paper, glass fiber disks (Blaney and Winstanley 1980) and artificial diet (Escoubas et al. 1992, Alonso-Amelot et al. 1994).

Artificial diets of known chemical composition have been shown to be an indispensable tool for studying the behavioral responses of insects to specific plant compounds or the nutritional role of certain plant components (Schoonhoven et al. 1998). An artificial diet has been reported to support successfully the adult stage of the banded cucumber beetle, Diabrotica balteata (Creighton and Cuthbert 1968). Furthermore, a commercial artificial diet for adult D. balteata is available. Studies in Chapter 2 and 3 indicated that the lettuce cultivar Valmaine was the most resistant cultivar against adult D. balteata feeding, and Tall Guzmaine was the most susceptible. Studies of the leaf surface chemicals indicated that surface wax on lettuce leaves was not responsible for







71
resistance in Valmaine. Experiments in this chapter were designed to determine whether plant chemicals inside lettuce leaves act as feeding deterrents to adult D. balteata when incorporated into an artificial diet. Experiments were also designed to determine whether the adults would exhibit a response varying with the concentration of leaf powder, and to determine whether lettuce leaf powder can be used as the starting material for the isolation of the compounds responsible for resistance. The quantitative effect of leaf powder on feeding responses was determined by measuring diet consumption and female mass in a binary-choice situation, and reproduction and adult mortality in a no-choice situation.

Materials and Methods

Plants and Insects

Two lettuce cultivars, Valmaine (Val, resistant) and Tall Guzmaine (TG,

susceptible) were used to search for possible resistance factors in the leaves. Lettuce was planted as described in Chapter 2.

Adult D. balteata for feeding bioassays were obtained from a laboratory culture originally collected from the field in Belle Glade, Florida in June 1996. Adults and larvae were reared in the same manner as described in Chapter 2. Preparation of Leaf Powder

Four to five fully expanded middle lettuce leaves were excised from plants of each cultivar when they had eight to nine fully expanded leaves. Excised leaves were immediately put into trays (30 x 43 cm) and transferred to a large freezer (-400C) for I d. The trays with frozen leaves were put into a freeze drier (model 25SRC, The Virtis Company, Gardiner, NY) for 4 d. Freeze dried leaves were immediately put into plastic







72


bags in a cooler containing dry ice, transported to the lab and ground in a Wiley mill (model 3383-L10, Thomas Scientific, Mexico) to a uniform powder (40 mesh). Leaf powder was stored in small airtight glass containers (4.4 cm diam., 4.5 cm high) in a freezer (-70'C) until used.

Artificial Diet

The components of artificial diet were ordered from a commercial company (BioServ, Frenchtown, NJ). 100 ml normal diet was prepared as follows: 1.74 g agar was boiled in 100 ml demonized water on a hot plate (model SPA1025B, Barnstead/Thermolyne, Dubuque, IA). KOH solution (1 ml) and diet dry mix (14.91 g) were added into the agar after it had cooled below 40'C, and thoroughly mixed to avoid the formation of lumps. The diet was poured into a glass petri dish (14 cm diam.), covered with the glass cover, wrapped completely in aluminum foil, and stored in a refrigerator (4-6"C). The test diet containing lettuce leaf powder was made by adding fixed amounts of the leaf powder into the normal diet by replacement of diet dry mix with the same amount of leaf powder. Three treatment diets for each cultivar were made containing 10, 20, or 30% leaf powder incorporated into the normal diet. The diet dry mix and leaf powder were thoroughly mixed before they were added to the agar. All diets were prepared 2 to 3 h before the following bioassays. Binary-choice Test

In order to test whether beetles still preferred TG to Val when leaf powder made from both cultivars was incorporated into an artificial diet, binary-choice tests were performed. Diet disks (3-4 mm thick) were made using a No. 13 cork borer (ca. 2.1 cm diam.) to punch out plugs from the cooled diet in petri dishes. The bottom and top of the







73

disk were covered with a circle of plastic, whose diameter was a little bit larger than that of the diet disk, in order to slow evaporation rate and keep diet free of frass. Only the edge of the diet disk was exposed to beetle feeding.

Each experimental unit consisted of six disks, three from a diet with TG leaf

powder and three from a diet with Val leaf powder at the same concentration, arranged alternately in a circle in a plastic container (27.5 x 20 x 10 cm). Twenty-five unsexed and unfed adults emerged within 48 h were released into this container and were allowed to feed on the diet for 3 d. Ten replicates were performed for each experimental unit. The experiments were carried out in an incubator with a photoperiod of 14:10 (L:D) at 25

0lC and 70 10% R. H.

Moisture calculation of the fresh diet (diet moisture). When the test diet disks were punched out from a diet, ten diet disks were weighed individually (disk fresh wt) before they were put into an oven at 50 5"C. After 2 d, these disks were reweighed individually (disk dry wt). The moisture content of each disk was then calculated as follows. Disk moisture = (disk fresh wt-disk dry wt)/disk fresh wt. The moisture content of a diet was the average moisture content of 10 disks.

Dry weight of a diet disk consumed (dry wt consumed). The test diet disk was

weighed (test disk fresh wt) before it was given to the adults. After 3 d, the diet disk was put into the oven at 50 5'C for 2 d and was weighed again to obtain its remaining dry weight (test disk dry wt afterwards). The dry weight of the diet consumed was calculated as follows.

Dry wt given to beetles = test disk fresh wt x (1 - diet moisture)

Dry wt consumed = dry wt given to beetles - test disk dry wt afterwards







74


The total dry weights of TG and Val diets consumed in 3 d were calculated by adding up the dry weights consumed of all the three disks from the same diet in each experimental unit. Dry diet consumed was analyzed as a split plot design with random blocks replicated through time. The three levels of leaf powder were assigned at random to whole plots within each block. Each whole plot was divided into two subplots to which TG and Val diets were assigned at random (SAS Institute 1999). Post-hoc means separation was performed using Tukey's HSD test (cx=0.05, SAS Institute 1999). No-choice Test

Two disks from the same diet were presented to five pairs of adult D. balteata on a plastic cover of a plastic container placed upside down (10.5 cm diam. upper, 7.2 cm diam. bottom, 17.5 cm high) in no-choice tests. Adults used in the trial were unfed and had emerged within 48 h of the trial. Diet disks with TG leaf powder at 10, 20 and 30% concentrations, and diet with Val leaf powder at 10, 20, and 30% concentrations were tested. Diet disks from full strength Diabrotica diets (normal diet) were also tested and considered as 0% leaf powder. The experiments were carried out in an insect rearing room with a photoperiod of 14:10 (L:D) at 25 IOC. Before releasing these adults, females were marked on their elytra with blue, green, pink, yellow or ivory paper correction fluid and weighed separately. After 13 d, marked surviving females were collected and reweighed so that their weight gain could be calculated. These females were stored in a refrigerator freezer (-9'C) until they were dissected to check their egg production. The Chi-square test was used to analyze female, male and total mortality of D. balteata on three different diets for each cultivar and on full strength Diabrotica diets. Numbers of eggs produced were analyzed as an 8 x 2 x 4 three factorial design using







75


Proc GLM (SAS Institute, 1999) in which the replicate (replicating through time) had eight levels, cultivars had two levels, and leaf powder concentration had four levels. Means were separated using Tukey's HSD test with a significance level of U=0.05 (SAS Institue 1999). Female final weight and weight gain after a 13-d feeding period were also analyzed as a three factorial design by Proc GLM (SAS Institue 1999). Female initial weight was used as a covariate when analyzing female final weight.

Results

Binary-choice Test

Significant differences in diet consumption were found among the concentrations of leaf powder added to the artificial diet (F=47.9; df=2, 19; P=0.0001) (Fig. 4-1). TG diet consumption increased from 118.5 mg to 172.0 mg with an increase of leaf powder concentration from 10% to 30%. Similarly, beetle feeding on Val diets increased from 104.5 mg to 166.7 mg when the concentration of leaf powder increased from 10% to 30%. No significant difference in feeding was found between TG diets and Val diets at any concentration (F=1.8717; df=l, 27; P=0.1826). Therefore, data of leaf area consumption on TG diets and Val diets at the same concentration were pooled to study its relationship with leaf powder concentrations. The regression analysis showed that diet consumption increased linearly with leaf powder concentration (R2= 0.99) (Fig. 4-2). The replications through time (time factor) had a significant effect on the amount of diet consumed by adult D. halteata (F= 17.23; df=9, 18; P=0.0001). Interactions between the time and concentration (F=0.7633; df=18, 27; P=0.7208) and between concentration and cultivar (F=0.8372; df=2, 27; P=0.4438) were not found to be significant.







76


200
180 C TGI 160 Val
140
120
E
100
0
S 80 a 60
40 20
0
10% 20% 30%

Leaf powder concentration

Fig. 4-1. Dry weight consumed by adult D. balteata feeding on diets
with lettuce leaf powder of susceptible Tall Guzmaine (TG)
and resistant Valmaine (Val) lettuce incorporated at different
concentrations for 3 d in binary-choice tests. Vertical lines
indicate I SEM.

























2
Y =2.89 X +81.16, R = 0.99


0 5


10


15


20 25


30


Leaf powder concentration (%)


Fig. 4-2. Relationship between leaf powder concentration
and diet consumption by adult D. balteata for 3 d
in binary-choice tests. Vertical lines indicate 1
SEM.


77


~i)
E
0

tj~
0 C.)


200 180 160


140 120 100


80


35











No-choice Test

Number of mature eggs produced by females feeding on TG diets for 13 d did not differ significantly from that of females feeding on Val diets (F=0.5883; df=l, 9; P=0.4634) (Fig. 4-3). Females feeding on TG diets with different concentrations of leaf powder had 65 to 75 mature eggs while females on Val diets had 63 to 70 mature eggs. Moreover, the increasing concentrations of leaf powder added did not significantly affect egg production of females, regardless of cultivars (F=1.8272; df=3, 24; P=0.1691). No replication effect was observed on egg production of females feeding on diets with different levels of leaf powder from TG and Val (F=2.0708; df=7, 12; P=0.1289). No significant effect of the interaction among replication, cultivar and concentration on female reproduction was found (F=0.5179; df=21, 125; P=0.9587). Neither were twofactor interactions between replication and cultivar (F=0.9350; df=7, 26; P=0.4967), nor between cultivar and concentration (F=0.8141; df=3,28; P=0.4969) significant. However, the interaction between time and concentration was significant (F=2.2539; df=21, 21; P=0.0347).

The percentage of females with mature eggs on TG diets was not significantly different from that on Val diets at 0, 10, 20 and 30% levels (Table 4-1). Eighty-five or ninety percent of females feeding on the diets without any TG (0% TG) or Val (0% Val) leaf powder added had mature eggs after 13 d. This was not significantly higher than any other diet with either TG leaf powder or Val powder.

There was no significant difference between cultivars in female survivorship at 0, 10, 20 and 30% leaf powder (Table 4-1). Similarly, male survivorship was not







79


90 85 -DTGI 80 VaI
75
70 65
60 55 50
45
40
0 10 20 30

Leaf powder concentration (%) Fig. 4-3. Average number of mature eggs produced per female after
feeding for 13 d on diets containing leaf powder of susceptible
Tall Guzmaine (TG) or resistant Valmaine (Val) lettuce at
different concentrations. Vertical lines indicated I SEM.







so


Table 4-1. Results of no-choice tests in which adult D. balteata were allowed to feed on diet with different concentrations of leaf powder from susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) lettuce for 13 d Category Cultivar Leaf powder concentration (Mean SEM)
0 10 20 30

TG 90.2 4.9 86.9 5.4 79.4 7.3 72.1 9.7

Percentage of females Val 85.2 6.0 80.6 5.5 61.3 7.4 75.6 8.8 with mature eggs
i2 1.902 0.893 4.075 5.679

P 0.965 0.996 0.771 0.578


TG 70.0 8.5 70.0 8.5 70.0 6.6 70.0 8.5

Val 80.0 7.5 80.0 7.6 75.0 7.3 67.5 10.0
Female survivorship
X2 1.038 1.537 3.722 3.651

P 0.99 0.981 0.811 0.819


TG 87.5 7.5 85.0 6.3 70.0 10.0 77.5 5.9

Val 90.0 5.3 87.5 5.3 80.0 8.5 87.5 6.5
Male survivorship
x2 1.772 0.573 4.094 1.485

P 0.971 0.999 0.769 0.983


Total survivorship


TG

Val

2
x


80.0 6.0 85.0 4.6

1.955


77.5 4.9 85.0 4.2

0.911


72.5 4.9 77.5 5.6

2.805


72.5 5.9 77.5 6.7

2.156


P 0.962 0.996


0.902 0.951







81

significantly different between the cultivars at 0, 10, 20 and 30 % leaf powder (Table 41). Finally, total survivorship was similar for the two cultivars at 0, 10, 20, and 30% leaf powder (Table 4-1).

Female final weight and weight gain during the experiment were significantly affected by replication through time (weight: F=3.0026; df=7, 18; P=0.0285) (weight gain: F=5.8003; df=7, 18; P=0.03) and by cultivar (weight: F=12.0303; df=1,10; P=0.006) (weight gain: F=1 1.8162; df=1, 10; P=0.0062) (Table 4-2). Females fed Val leaf powder diets gained significantly more weight than those fed TG leaf powder diets. No significant differences in pre-experiment weights were found among females assigned to the diets with leaf powder from different cultivars (F=0.000468; df=1, 9; P=0.9832) or to the diets with different concentrations (F=1.7745; df=3, 25; P=0.1781). On TG leaf powder diet, final weights and weight gains decreased slightly (16.3 to 14.3 mg and 7.4 to 6.3, respectively) as concentration increased. Similar results were recorded for beetles on Val leaf powder diet where final weights decreased from 16.9 to 15.3 mg and weight gain decreased from 8.5 to 7.2 mg. However, changes with increasing leaf powder concentration were not significant for final weight (F=2.1402; df=3,23; P=0.1226) or weight gain (F=2.1357; df=3,23; P=0.1234). Two-factor interactions were not significant between replication and cultivar for weight (F=0.8416; df=7, 24; P=0.5641) or for weight gain (F=0.8461; df=7, 24; P=0.5608). The interactions between cultivar and concentration for weight (F=1.6181; df=3, 32; P=0.2048) or for weight gain (F=1.5932; df=3, 32; P=0.2096) were also not significant. However, a significant interaction was found between replication and concentration for weight (F=5.1466; df=21, 21; P=0.0002) and for weight gain (F=5.1925; df=21, 21; P=0.0002). Further analysis on concentration







82


Table 4-2. Average female D. balteata weight gain and final weight after feeding on diets with different concentrations of leaf powder from susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) lettuce for 13 d in a no-choice situation Diet conc. (%) Weight TG (mg t SEM) Val (mg SEM)

0 Final weight 16.26 0.59 16.89 0.50

Weight gain 7.41 0.52 8.47 0.42

10 Final weight 16.11 0.50 16.84 0.50

Weight gain 7.66 0.49 8.39 0.42

20 Final weight 15.18 0.43 15.95 0.54

Weight gain 7.10 0.37 7.87 0.46

30 Final weight 14.34 0.50 15.26 0.50

Weight gain 6.34 0.45 7.21 0.44







83

by fixing replication through time factor showed a significant difference in final weight at the last replicate (F=6.88; df=3,24; P=0.0017). Significant difference in weight gain was also found at time 8 (F=5.57, df=3,24; P=0.0039). The three-factor interaction (replicate x cultivar x concentration) was not significant for beetle final weight (F=0.3785; df=21, 175; P=0.9943) or weight gain (F=0.3849; df=21, 176; P=0.9936).

In conclusion, adults feeding on Val leaf powder diets produced similar amount of eggs compared to those on TG leaf powder diets. There were no significant differences in female, male and total survivorships between these two diets. Females on Val leaf powder diets gained more weight than those on TG leaf powder diets.

Discussion

Adult D. balteata did not show a significant preference for TG over Val in binarychoice tests when leaves were freeze-dried, ground into powder and incorporated into the artificial diet at three different concentrations (Fig. 4-1). This result differs with what I found in Chapter 2, where fresh Val plants had significant resistance against D. balteata feeding, and in Chapter 3, where Val plants with leaf surface removed still had very strong resistance to D. balteata. Furthermore, increasing the amount of leaf powder of TG or Val added to the artificial diet resulted in increased diet consumption by D. balteata. This compensatory feeding may be due to suboptimal quality of the diet with lettuce leaf powder compared to adult's normal diet with the same amount of dry matter. Compensatory feeding responses by adjusting feeding rate to approach or realize maximal growth rate has been found in several insect species and is proposed to be general among herbivorous insects (Simpson and Simpson 1990). The increased diet consumption with increased leaf powder may be also due to the presence of feeding







84


stimulants or the level of feeding stimulants enhanced somehow compared to possible deterrents present in lettuce leaf powder. That Val loses its resistance when freeze-dried and powdered was also confirmed by no-choice tests, where females produced similar numbers of mature eggs regardless of the diet on which they fed for 13 d (i.e. control, or supplemented with TG or Val powder).

An artificial diet differs in many respects from natural food sources of herbivores. It may not be completely suitable for screening feeding stimulants or deterrents even though it offers uniformity, evenness of application, and ease of preparation and measurement (Lewis and van Emden 1986, Schoonhoven et al. 1998). An artificial diet usually contains feeding stimulants for designated insects, which can mask effects of feeding deterrents. It has been proposed that a change in the balance between phagostimulants and deterrents can make an unacceptable food acceptable and vice versa (Bernays and Chapman 2000). The feeding of L. migratoria on wheat flour wafers was almost totally inhibited by adding tomatine at 0.1% dry weight, but the effects of the tomatine were eliminated by adding increasing amounts of sucrose to the wafers (Bernays and Chapman 1978). Thresholds for the same deterrent may vary as much as 1000 times between natural and artificial substrates, resulting in differences in the effective concentrations of the chemicals presented to the insect (Schoonhoven 1982). Even small changes in diet composition can have drastic effects on insect performance, such as growth and reproduction (Schoonhoven et al. 1998). Moreover, the same chemicals may have neutral, phagostimulant, or deterrent effects at different concentrations depending upon existing background compounds, especially phagostimulants (Lewis and van Emden 1986). Therefore, the artificial diet used in this study may mask deterrents from







85

Val because of the difference in balance of feeding stimulants and deterrents between the artificial diet and lettuce plants.

It is also possible that feeding deterrents in Val may be broken down during the process of making lettuce leaf powder or during the storage period, even though it was stored at -70'C. Freeze-drying under high-vacuum conditions has been suggested to be inappropriate for low- and intermediate-molecular weight phenolics because many of the volatile phenolics are concentrated on the ice trap (van Sumere et al. 1983). Orians (1995) compared three methods of preservation, i.e. air-drying, freeze-drying, and vacuum-drying and found that freeze-drying was not appropriate for analysis of phenolic glycosides because of the degradation of these compounds. Air-drying was appropriate for phenolic glycosides but resulted in loss of tannins. Only vaccum-drying preserved both phenolic glycosides and tannins for analysis. Chlorogenic acid instead of isochlorogenic acid was detected in lettuce roots that had been vacuum-dried prior to extraction, and no caffeoylquinic acids were detected in root material, either stored in a deep-freezer at -20"C before extraction or freeze dried (Cole 1984). The reason for the presence of chlorogenic acid in dried root material and the loss of all caffeoylquinic acid after freezing or freeze-drying may be related to cell damage and the formation of insoluble quinones from the caffeoylquinic acids. Therefore, the treatment of plant material before chemical analysis is very important.

Other possible reasons for the loss of resistance in Val are that feeding deterrents are active only in fresh leaves attached on the plants and that any role that a physical barrier plays in resistance may act only when leaves are turgid. The chemicals in the leaf may be activated and released only after insect feeding. It is well known that some plants







86
can qualitatively or quantitatively enhance their defense against herbivory after they are attacked by pests or by other inducers (Tallamy and Raupp 1991, Karban and Baldwin 1997). Many studies show that wound-induced resistance by herbivores is related to the induction of enzymes activities and the release or synthesis of new secondary metabolites upon injury (Bi et al. 1994, Bi et al. 1997, Baldwin 1991, Louda and Mole 1991). Some physical barriers to host use by the insect, such as latex emission, may become ineffective when leaves are dried and powdered. The possible role of latex and inducible resistance in resistance of Valmaine lettuce to D. balteata will be studied in Chapter 5.














CHAPTER 5
POSSIBLE MECHANISMS INVOLVED IN LETTUCE RESISTANCE TO DIABROTICA BAL TEA TA

Introduction

Plant defense mechanisms against herbivores have been extensively studied over the past decade because of their potential application in integrated pest management programs, and in providing new leads for environmentally safer pesticides (Balandrin et al., 1985, Panda and Khush 1995). Understanding the mechanisms of resistance is also very important before the degree of resistance among plants can be ascertained (Panda and Khush 1995). Factors that determine host plant resistance to herbivores include the presence of morphological, physical, or structural barriers, allelochemicals, and nutritional imbalance (Panda and Khush 1995). These factors can be active whether or not herbivores are present. Host plant resistance can also be inducible after plant tissues are attacked by herbivores, resulting in induction of a wide array of chemical defenses against subsequent herbivory (Panda and Khush 1995).

Latex in some laticiferous plants has been reported as a natural defense system

against certain herbivores. In many laticiferous plants, including Lactuca sativa L., latex is stored under pressure within laticifers, which results in rapid release of latex upon cutting (Fahn 1979, Data et al. 1996, Dussourd 1995). The secretions often contain secondary metabolites known to be toxic or deterrent to animals (Farrell et al. 1991). These secretions also polymerize upon exposure to air, thus threatening small herbivores with the risk of entrapment in hardening exudates or gumming up their mouthparts


87







88

(Dussourd 1995, 1997). Data et al. (1996) found that young vine material of sweet potato produced more latex and had fewer sweetpotato weevils, Cylasformicarius (F.), than older and more mature portions of the vine. Feeding and oviposition were reduced when latex was applied to the surface of root cores. Several insects have been observed immobilized in exudates, such as caterpillars (Dussourd 1993), ants (Dillon et al. 1983), aphids and whiteflies (Dussourd 1995). Because of the rupture of latex canals, two aphid species, Uroleucon pseud/ambrosiae (Olive) and Rhopalosiphum maidis (Fitch), were physically glued to lettuce plants, primarily by their legs, after they alighted on inflorescences. Adult whitefly, Bemisia tabaci (Gennadius), likely became trapped in latex, primarily on the flower buds (Dussourd 1995).

In many crop plants, resistance can be induced through prior wounding by insects or mechanical means (Karban and Baldwin 1997). Induced resistance has been correlated with increased activities of oxidative enzymes (Bi et al. 1994), enhanced synthesis of secondary metabolites (Bi et al. 1997), or synthesis of primary gene products (Green and Ryan 1972). These induced resistance phenomena may involve decreases in both insect feeding preference and the nutritional value of induced foliage (Schoonhoven et al. 1998). Larval growth of Helicoverpa zea Boddie was decreased on previously damaged foliage or squares of cotton compared to growth on undamaged tissue because of a significant decline in host nutritional quality and increased activities of oxidative enzymes after herbivory (Bi et al. 1994). Second-instar larvae of the gypsy moth, Lymantria dispar L., preferred leaf disks from non-induced trees over those from induced trees (Havill and Raffa 1999). Inducible resistance in plants can also affect the ovipositional responses of adult females (Stout and Duffey 1996). Induced resistance







89
may be expressed either on rapid (hours to days), short-term (days to weeks) time scales, or long-term (years) time scales (Schoonhoven et al. 1998) through de novo synthesis of formerly absent compounds or an elevation of the normal level of resistance compounds. The short-term reaction may spread systemically over the whole plant, or it may be localized to the damaged leaf (Schoonhoven et al. 1998). Localized accumulation of a sesquiterpenoid lactone phytoalexin, phenolics and phenolic esters have been reported in lettuce after being attacked by a pathogen (Bennett et al. 1994, 1996).

In this chapter, the resistance of lettuce to D. balteata was further evaluated. Three screening methods were conducted to test the effect of methodology on the expression of resistance: leaf disks, intact leaves attached to plants and excised leaves. Leaf area consumed by beetles was evaluated in binary-choice and no-choice bioassays to compare results for both leaf disks and detached leaves against intact leaves. Other objectives of this chapter were to compare the physical properties of latex, investigate inducible resistance as a factor in lettuce resistance, and propose possible mechanisms involved in lettuce resistance to D. balteata.

Materials and Methods

Plants and Insects

Two lettuce cultivars, Valmaine (Val, resistant) and Tall Guzmaine (TG,

susceptible) were used for this experiment. Lettuce was planted in the same manner as described in Chapter 2. Fordhook 242 lima bean seeds were purchased from a commercial company (Illinois Foundation seeds Inc., IL) and planted in the same way as described in Chapter 2. Only the first true leaves were used for the latex smearing experiment.







90


Adult D. balteata for feeding bioassays were obtained from a laboratory culture originally collected from the field in Belle Glade, Florida in June 1996. Adults and larvae were reared as described in Chapter 2. Only unfed adults which had emerged within 48 h were used for the following assays. Screening Methods

Leaf disks. All leaves used in this assay were the seventh fully expanded leaf, counting from the first true leaf. Selected plants had six to eight fully expanded leaves.
2
Two disks of 380 mml were punched out from non-midrib areas of freshly excised leaves of Val or TG plants using a No. 15 cork-borer. A test unit consisted of a plastic petri dish (8.9 cm diam.) inside of which four leaf disks (two from Val and two from TG) were presented at an equal distance on two layers of moistened paper towel. One pair of adults was placed in each petri dish and allowed to feed for two days. The tests were conducted in a rearing room at 25 IC, 14:10 (L:D) photoperiod. Each test unit was replicated 15 times. To compare the resistance level of leaf disks with attached lettuce leaves, binarychoice tests were conducted simultaneously between intact leaves of TG and Val using the binary-choice feeding arenas as described in Chapter 2.

The extent of feeding for both screening methods was evaluated by measuring the area of the leaf remaining by scanning the leaf (JADE 2, Linotype-Hell, Taiwan) and importing the scanned images into an imaging program (ImagePC beta version 1, Scion Corporation, Frederick, Maryland) where remaining leaf area was determined. For the leaf disks method, the remaining leaf area was then subtracted from the mean disk area of 10 disks not offered to beetles for two days in order to account for shrinkage during the







91

assay. The difference in leaf area consumption between cultivars was analyzed by paired t-test using Proc MEANS (SAS Institute 1999).

Detached leaves. Resistance to beetle feeding was next compared between

detached and intact leaves. For the detached leaf tests, the seventh fully expanded leaf was excised from each plant and its petiole immediately immersed in a beaker filled with tap water. Binary-choice feeding arenas were used to expose single pairs of adults to both TG and Val detached leaves for 48 h. At the same time, binary-choice tests were carried out between two leaves still attached to TG and Val plants by using the same feeding arena. Each test unit was replicated 17 times in each bioassay. The difference in leaf area consumption between cultivars was measured as above and analyzed by paired t-test using Proc MEANS (SAS Institute 1999).

Since a significant difference was found in leaf area consumption between the two cultivars in binary-choice tests, the resistance level of detached leaves was further evaluated by using no-choice bioassays as follows. The seventh fully expanded leaf was excised from each plant, and its base was immediately put into a beaker with tap water. One pair of beetles was confined on the leaf by a no-choice feeding arena as described in Chapter 2. The adults were allowed to feed for 48 h. At the same time, a pair of adults was confined in the no-choice arena to the seventh leaf still attached to a plant. This study was arranged as a randomized complete block design with detached and intact leaves from each cultivar in each block. Each block was replicated 18 times. Leaf area consumed was measured as described above and was analyzed by Proc GLM (SAS Institute 1999). Means with significant ANOVA were separated using Tukey's HSD test with a significance level of u=0.05 (SAS Institute 1999).




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EUZJMKL1W_8ZZ4SE INGEST_TIME 2015-04-02T21:05:46Z PACKAGE AA00029911_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

RESISTANCE TO ADULT BANDED CUCUMBER BEETLE, DIABROTICA BALTEATA LECONTE, IN ROMAINE LETTUCE By JUAN HUANG A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2000

PAGE 2

Copyright 2000 by Juan Huang

PAGE 3

This effort is dedicated to my husband Honghong Zhang and to my parents Yuzhen Huang and Zhichang Mao.

PAGE 4

ACKNOWLEDGMENTS I express my respect and gratitude to Dr. Gregg S. Nuessly for serving as the chairman of my committee and providing funding for this research. I would hke to thank Wedgeworth family of Florida vegetable and sugar producers whose fellowship sponsored this research. My deepest gratitude also goes to Dr. Heather J. McAuslane for being my co-advisor, proving funding for this research and giving me moral and academic support, encouragement, and guidance throughout my Ph.D. program. I would also like to appreciate the advice and expertise of my other committee members. Dr. Frank Slansky, Dr. Russell Nagata, and Dr. Anson Moye. Special thanks are extended to Dr. Russell Nagata for providing lettuce seeds for this research. I am very grateful to Dr. Hans T. Albom for providing me a micro spray device which made the research in Chapter 3 possible. I would not have been able to freeze dry lettuce leaves without the help of Mr. Rob Pellicit from Food Science and Human Nutrition Department. My sincere thanks go to Debbie Boyd for ordering research supplies and giving me unforgettable friendship. I also thank Yasmin Cardoza, Alonso Suazo, Ramazan Cetintas, and Jiang Chen for their friendship. I wish to thank all of those I have failed to mention, your help and support made this journey complete. IV

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS iv ABSTRACT viii CHAPTERS 1 LITERATURE REVIEW AND RESEARCH GOALS 1 Introduction 1 Diabrotica balteata Biology and Pest Status 2 Overview of Plant Resistance 4 History of Plant Resistance to Insects 4 Plant Resistance Terminology 5 Constitutive Resistance 7 Induced Resistance 14 Host Plant Resistance to Diabrotica balteata 17 Host Plant Resistance in Lettuce 19 Plant Secondary Chemicals 19 Genetic Resistance 20 Nutritional Quality 21 Possible Evidence of Inducible Resistance 23 Research Goals 23 2 RESISTANCE IN ROMAINE LETTUCE CULTIVARS TO ADULT DIABROTICA BALTEATA 25 Introduction 25 Materials and Methods 27 Plants 27 Insects 27 Binary-choice Test 30 No-choice Test 31 Fitness Bioassay 31 Starvation Test 32 Results 33 Binary-choice Test 33 No-choice Test 35

PAGE 6

Fitness Bioassay 35 Starvation Test 38 Discussion 41 3 ROLE OF THE LEAF SURFACE OF LETTUCE IN RESISTANCE TO DIABROTICA BALTEATA 47 Introduction 47 Materials and Methods 50 Plants 50 Insects 5 1 Solvent Effects on Extraction 51 Feeding Response to Solvent-extracted or Normal Leaves 52 Feeding Response to Crude Extracts in Binary-choice Test 53 Feeding Response to Crude Extracts in No-choice Test 55 Results 55 Solvent Extraction Comparisons 55 Feeding Response to Solvent-extracted or Normal Leaves 57 Feeding Response to Crude Extracts in Binary-choice Test 60 Feeding Response to Crude Extracts in No-choice Test 64 Discussion 64 4 INVESTIGATION OF PHYTOCHEMICAL NATURE OF RESISTANCE IN LETTUCE TO DIABROTICA BALTEATA USING ARTIFICIAL DIET 69 Introduction 69 Materials and Methods 71 Plants and Insects 71 Preparation of Leaf Powder 71 Artificial Diet 72 Binary-choice Test 72 No-choice Test 74 Results 75 Binary-choice Test 75 No-choice Test 78 Discussion 83 5 POSSIBLE MECHANISMS INVOLVED IN LETTUCE RESISTANCE TO DIABROTICA BALTEATA 87 Introduction 87 Materials and Methods 89 Plants and Insects 89 Screening Methods 90 Latex Physical Properties 92 Latex Smearing Test 92 vi

PAGE 7

Observation of Feeding Behavior 93 Inducible Resistance 93 Results 94 Screening Methods 94 Latex Physical Properties 98 Latex Smearing Test 98 Observation of Feeding Behavior 102 Inducible Resistance 104 Discussion 104 6 SUMMARY AND CONCLUSIONS 113 APPENDIX MISCELLANEA 1 1 7 LIST OF REFERENCES 118 BIOGRAPHICAL SKETCH 132 vii

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RESISTANCE TO ADULT BANDED CUCUMBER BEETLE, DIABROTICA BALTEATA LECONTE, IN ROMAINE LETTUCE By JUAN HUANG December, 2000 Chairman: Gregg S. Nuessly Major Department: Entomology and Nematology Four cultivars of lettuce, Tall Guzmaine (TG), Parris White (PW), Short Guzmaine (SG), and Valmaine (Val), were screened for resistance to adult Diahrotica balteata LeConte under laboratory conditions. Val had the highest level of resistance, followed by PW and SG; TG was particularly susceptible to adults when using leaf area consumption as the criterion for resistance. Female adults ate more foliage, gained more weight, and produced more eggs when feeding on TG than on Val for 1 0, 1 3 or 1 6 d. A starvation test confirmed that females on Val produced no mature eggs because of their inability to use Val as a food source. The leaf surface of lettuce was investigated for its possible role in resistance. Even though D. balteata consumed a much greater amount from Val leaves with their surface chemicals removed, leaf surface extracts from Val were not deterrent to D. balteata feeding when applied to lima bean leaf surfaces at various concentrations. These viii

PAGE 9

results suggest that leaf surface chemicals in lettuce are not responsible for resistance in Val. Freeze-dried leaf powder of TG or of Val was added to an artificial diet for D. balteata at four concentrations (0, 10, 20, 30%) to determine the chemical basis of lettuce resistance. In choice experiments, there were no significant differences between TG diets and Val diets at each concentration. In no-choice experiments, no significant differences were found in the number of mature eggs produced, female's weight gain, or adult's mortality between TG and Val diets at four different concentrations. The resuhs of this assay failed to detect any reason for the resistance of Val. TG and Val differed in latex physical properties. Latex from Val turned brown faster than that from TG, and took more time to stop flowing. Both latex from TG and Val applied on lima bean leaves inhibited D. balteata feeding. Localized induced resistance to feeding was found in Val but not in TG after plants were previously damaged for 48 h. Therefore, I propose that localized inducible resistance and physical and chemical defenses in latex may be two major reasons for resistance in Val to D. balteata. ix

PAGE 10

CHAPTER 1 LITERATURE REVIEW AND RESEARCH GOALS Introduction Lettuce, Lactuca sativa L., is an important vegetable crop grown in many countries in the world. By far the greatest commercial production of lettuce takes place in the USA (Ryder 1998). The majority of the insects which are economically important on lettuce belong to five orders: Homoptera, Lepidoptera, Diptera, Hymenoptera (Ryder 1998) and Coleoptera. These insect pests can cause damage directly by chewing on foliage, stems and roots or indirectly by transmitting disease. Moreover, they can also lower the value of the lettuce crop by causing unsightliness such as the presence of insect cast skins and detritus. To date, lettuce glowers have relied heavily on insecticides to control insect pests to reduce economic losses. However the heavy use of nonselective insecticides has led to well-known problems, such as development of insect resistance to these chemicals, the resurgence of secondary insect pests, and the persistence of residues that are toxic to humans, animals and other nontarget organisms. The western com rootworm, Diabrotica virgifera virgifera LeConte, provides a classic example of a manmade pest (Metcalf 1986). This insect was first described by LeConte in Kansas in 1868 and first found attacking com in Colorado in 1 909 before it slowly spread into the Nebraska in 1929 (see references cited by Metcalf 1986). Because of large-scale applications of insecticides, adult westem com rootworm resistance to insecticides was noted in thel960s. The resistant strain spread rapidly about an average of 120 miles per year from a single locus in southeastern Nebraska in 1961 to encompass much of the com 1

PAGE 11

2 growing area of many states. By 1980, this strain had spread throughout the U.S. com belt (Metcalf 1986). Integrated pest management (IPM) programs have been developed to reduce the reliance on chemical insecticides. Host plant resistance is recognized as the most effective component of 1PM (Panda and Khush 1995), because it has low impact on non-target organisms and the environment and is usually compatible with other control tactics, such as biological and cultural controls. In nature, plants are under the constant threat of damage by insects, which dominate the world's fauna in terms of number of species and individuals (Dicke 1999). However, plants defend themselves very effectively by using a variety of defensive strategies ranging from mechanical to chemical weapons (Louise 1999). Defenses can operate by either constitutive resistance, which is always active, or by induced resistance. In induced resistance, production and/or release of defensive chemicals or characteristics are induced by extrinsic physical or chemical stimuli, such as herbivory, pathogen infection, or mechanical injury (Kogan and Paxton 1983). Diabrotica balteata Biology and Pest Status Diabrotica balteata LeConte is commonly known as the banded cucumber beetle. It is distributed throughout the southern and western United States and south into central America (Krysan 1 986). In the southern United States, D. balteata occurs as an adult during much of the year (Schalk 1986). Wolfenbarger (1963) reported D. balteata as a ubiquitous insect throughout Florida with fluctuation of its abundance and injuriousness from year to year. This insect is a pest of leafy vegetables and sweet com in Florida. The adult D. balteata is marked with alternating green and yellow bands across the elytra, but the newly emerged adult has a soft body with pale color. The adult feeds for 4 to 8 d

PAGE 12

3 before mating. The range from copulation to first egg deposition can be 9 to 28 d. Individual females lay approximately 350 to 849 eggs (depending on food source) over its lifetime of 5 to 11 wk under laboratory conditions. Oviposition usually occurs at intervals of 2 to 3 d beginning when the female is about 3 wk old and lasts from 2 to 8 wk. Eggs are deposited singly or in clusters in small cracks in the soil or under objects lying on the surface (Schalk 1986, Pitre and Kantack 1962). Three larval instars occur with a mean larval developmental period of 17 d at 27°C to 23 d at 21 "C, depending on food source and rearing environment. The third-instar larvae increase dramatically in size and do the most severe damage to plants by cutting and partially boring into young roots. Pupation takes place in the soil and the average time required for adult emergence is 7 to 9 d at temperatures of 21 to 27T (Pitre and Kantack 1962). The larvae of D. balteata severely damage the roots of crops, reducing yield in bean, melon and com (Pitre and Kantack 1962). However, feeding on other plant parts such as potato tubers, and peanut pods has been reported also (Wolfenbarger 1963). Other host plants for the larvae include squash, sorghum, soybean, sweet potato, sweet com and tomato (Table 2 in Teng 1983). No complete report of the larval host range has been reported in the literature. Adults feed on foliage, flowers or fmit (Elsey 1988) of more than 50 host plant species (Teng 1983) in 23 families (Saba 1970). These plants include broccoli, com, cowpea, cucumber, lima bean, peanut, pepper, potato, spinach, squash, soybean, sweet potato, tomato and wheat (Table 3 in Teng 1983). Lettuce was also listed in Teng's table, even though there was controversy in the literature as to whether this was a host for D. balteata adults at that time (Saba 1970). In addition to feeding, adults also contribute to yield losses by transmitting plant vimses, such as

PAGE 13

4 cowpea mosaic, bean mgose mosaic, squash mosaic, muskmelon necrotic spot and melon bacterial wilt (Da Costa and Jones 1971, Gergerich et al. 1986). Nuessly (personal communication) found that adult D. balteata caused serious damage to lettuce in southern Florida by feeding across the entire leaf surface, removing tissue and causing necrosis of surrounding tissues. Beetle feeding on older plants caused lettuce heads to be unmarketable due to feeding damage and fecal contamination. In addition, D. balteata caused yield reduction and uneven stand maturity by feeding on young and middle aged lettuce. Overview of Plant Resistance History of Plant Resistance to Insects There are many historical accounts of the successful use of resistant cultivars to control insect pests. One of the most famous successes is that of the development of resistance in French wine grapes to attack by the grape phylloxera in 1873 by grafting on to American grape rootstock. This strategy saved the French wine industry (Ortman and Peters 1 980). Resistant varieties of wheat and apples were first developed and cultivated in the USA during the eighteenth and early nineteenth centuries (Smith 1989). Highyielding, pest-resistant rice cultivars were used in tropical Asia during the "Green Revolution" of the 1960s. These cultivars helped several countries in southern and southeast Asia meet the food production needs of their populations (Smith 1989). Use of insect-resistant crop cultivars is still the principal method of control of some major pests today (Stoner 1996). Many crops with resistance to insect pests have already been developed (Tables 1 and 2 in Stoner 1996). A survey on insect-resistant germplasm released in the USA from

PAGE 14

1988 to early 1994 indicated that 34% of 1 17 releases were in grain crops, 27% in alfalfa and clover, 15% in cotton, 1 1% in vegetables, 4% in fruit and 8% in other crops (Stoner 1996). Genetic engineering techniques have made it possible to transfer genes directly into plants, thus opening new opportunities for host plant resistance (Panda and Khush 1995). Plant Resistance Terminology Painter (1951) defined host plant resistance as the relative amount of heritable qualities possessed by the plant, which influence the ultimate degree of damage done by the insect. Three modalities of plant resistance, originally defined by Painter (1951) but modified by Horber (1980), are commonly referred to in plant resistance literature: antixenosis, antibiosis, and tolerance. However, not all resistance phenomena can be clearly classified into these three categories of resistance. Biophysical and biochemical plant defense, as well as nutritional factors are not only involved in antixenosis but also in antibiosis. In certain cases, antixenosis can not be clearly separated from antibiosis because the deterrent chemicals and toxins in the plant are sometimes difficult to distinguish (Panda and Khush 1995). Furthermore, these three main types of resistance can interact, complement and compensate with each other in different abiotic and biotic environments (Panda and Khush 1995). Overlaying the types of resistance are two categories of resistance, constitutive and inducible, that are distinguished based on whether the resistant characters are caused by extrinsic stimuli (Kogan and Paxton 1983). Antixenosis, also referred to as nonpreference by Painter (1951), is the category of plant resistance to insects that describes the plant acting as a poor host to deter or reduce colonization by insects. Antixenotic characteristics include morphological.

PAGE 15

6 physical, or structural qualities (such as plant pubescence, and frego bract), and biochemical factors (such as arrestants, attractants, repellents) which interfere with mating, oviposition, feeding, and food ingestion (Panda and Khush 1995). Biophysical and biochemical factors affect the behavior of an insect pest to force it to choose an alternate host plant. Plants exhibiting antixenosis should have a reduced initial number of colonizers early in the season and reduce the size of the insect population after each generation as compared with susceptible plants (Panda and Khush 1995). Antibiosis is the resistance mechanism that describes the negative effects of a resistant plant on the biology of an insect after this insect has colonized and started using the plant. Such resistance results in abnormal insect growth, development, reproduction and survival. Plants that exhibit antibiosis reduce the reproductive rate and survival of insects to slow the rate of population increase. The antibiotic properties of the host plant may be expressed as constitutive or induced resistance against herbivores. These properties may be biophysical (such as trichomes), biochemical (such as toxins, growth inhibitors), or nutritional in nature (Panda and Khush 1995). Tolerance is the resistance modality in which a plant has the genetic ability to withstand or recover from damage caused by an insect pest abundance equal to that on a susceptible cultivar. Tolerance does not affect the rate of population increase of the target pest, but raises the threshold level, thereby allowing plants to outgrow an insect infestation or recover and add new growth after the destruction or removal of damaged tissues (Smith 1989, Panda and Khush 1995). Constitutive resistance, also called preformed resistance (Pollard 1992), is a broad term, referring to the presence of physical defenses as well as to the absence of some

PAGE 16

7 required nutrients, presence of compounds with kairomonal action and non-induced accumulation of secondary metabolites with allomonal properties (Kogan and Paxton 1983). Induced resistance is the qualitative or quantitative enhancement of the plant's defense against pest-related injury or extrinsic physical or chemical stimuli. These extrinsic stimuli are called inducers or elicitors (Kogan and Paxton 1983). Induced resistance indicates the ability of plants to reallocate resources and operates only after the plant is attacked by pests or stressed by other inducers (Thaler and Karban 1997). Constitutive Resistance Constitutive defense against insects can be mediated by plant physical attributes (e.g., toughness of plant tissue, thorns and spines), by the presence of plant secondary metabolites (e.g., toxins, repellents, and digestibility reducers), or by other factors such as insufficient nutrition levels. Based on biosynthetic criteria, there are three principal groups of secondary metabolites which plants usually produce as constitutive resistance. One is a terpenes group (e.g., monoterpenes, sesquiterpenes, diterpenes, triterpenes), which is synthesized from acetyl CoA via the mevalonic acid pathway. The second group is phenolics (e.g., chlorogenic acid, caffeic acid, coumarins, tannins), which are formed via the shikimic acid pathway or malonic acid pathway. The third group is nitrogen-containing compounds (e.g., alkaloids, glucosinolates, cyanogenic glycosides), which are biosynthesized primarily from amino acids (Panda and Khush 1995). Resistance at the plant surface. The plant surface is recognized as the first line of defense against insect attack. Insects depend on a combination of physical and chemical stimuli at the plant surface to assess whether the plant is suitable for oviposition, settling

PAGE 17

8 or feeding (Woodhead and Chapman 1986). Hairs or trichomes have been shown to act as physical barriers, but they may also secrete behaviorally active chemicals. Other chemicals affecting the evaluation process are found within leaf waxes or other secretory structures such as essential oil glands. The surface of virtually all plants is covered by an amorphous layer of waxes known as the epicuticular wax (Baker 1982), the primary role of which is to prevent water loss (Eigenbrode and Espelie 1995). The epicuticular wax is comprised of mainly aliphatic wax components, including n-alkanes, wax esters, free fatty alcohols and free fatty acids. Sugars, amino acids, nonprotein amino acids, sucrose and glucose esters, sesquiterpenes, diterpenes, phenolics, phenolic glycosides and glucosinolates are also components of epicuticular wax (Woodhead and Chapman 1986, Eigenbrode and Espelie 1995). A growing number of reports show that waxes on many plant surfaces either enhance or deter oviposition, feeding and movement of various insects (Juniper and Southwood 1986, Eigenbrode and Espelie 1995). Leaf epicuticular wax in Brassicaceae crops appeared to be a major antixenotic factor that affects the feeding rate and feeding pattern of the crucifer flea beetle, Phyllotreta cruciferae (Goeze) (Bodnaryk 1992). The flea beetle fed on waxy leaves at much lower rates than on non-waxy leaves. It is hypothesized that edge feeding on waxy leaves, which is associated with a low feeding rate, may be determined by physical and chemical factors that render the leaf unpalatable or difficult to feed upon, or both. Other examples of leaf surfaces containing feeding deterrents include Colorado potato beetle, Leptinotarsa decemlineata Hawkes, on Solarium berthaultii (Yencho et al. 1994), migratory grasshopper, Locusta migratoria

PAGE 18

9 (L.), on sorghum (Woodhead 1983), and the brown planthopper, Nilaparvata lugens Stal., on rice varieties (Woodhead and Padgham 1988). Fatty alcohols and a-tocopherylquinone isolated and identified from the leaf surface of a poplar clone were found to stimulate feeding by adult cottonwood leaf beetle, Chrysomela scripta Fabr. (Lin et al. 1998). Fatty alcohols from mulberry stimulated feeding of larvae of the silkworm Bombyx mori L. (Mori 1982). Fatty alcohols from several host plants stimulated feeding of several species of chrysomelid beetles (Adati and Matsuda 1993). Diamondback moth larvae, Plutella xylostella (L.), spent more time walking and searching on a resistant cabbage with glossy leaves than on a susceptible variety with normal wax bloom. This behavior resulted in increased larval mortality on glossy plants due to starvation and desiccation. Morphological characteristics of the leaf waxes were of primary importance in causing higher net movement rates and reduced feeding on glossy resistant lines (Eigenbrode and Shelton 1990). A similar behavioral pattern occurred on glass slides on which hexane or dichloromethane extracts of leaf surface wax of the two cabbage genotypes were deposited, which implicated strongly that leaf surface waxes played an important role in acceptance of cabbage varieties by diamondback moth larvae (Eigenbrode et al. 1991). Wax chemistry was proposed to be another reason causing these behavioral differences since the chemical compositions of the leaf extracts were markedly different. Several compounds, including the triterpenols aand p-amyrin, were found only in the glossy waxes. Disruption of the leaf waxes on a susceptible cabbage plant resulted in high rates of larval movements, similar to those on glossy resistant plants (Eigenbrode and Shelton 1990). Leaf surface chemicals may affect insect oviposition. Chemicals on the leaf surface of Brassica oleracea stimulated

PAGE 19

10 oviposition by the cabbage root fly, Delia radicum (L.) (Roessingh et al. 1992) and chemicals on carrot leaves stimulated oviposition by the carrot fly, Psila rosae (F.) (Stadlerand Buser 1983). Glandular and nonglandular trichomes have been found to be of importance to insect oviposition, movement and feeding through mechanical and/or chemical means. Physical factors such as trichome density, erectness, length and shape influence their effects on insects (Johnson 1975). Nonglandular trichomes impart general antixenosis resistance by providing an effective barrier that influences the attachment of insects onto the plant surface, as well as their movement and feeding (Southwood 1986). The high number of trichomes on the upper and lower surfaces of the leaves of a resistant maize cultivar was the major reason for deterring oviposition by the stem borer, Chilo partellus (Swinhoe) (Kumar 1992). Oghiakhe (1995) reported that pubescence in wild and cultivated cowpeas adversely affected oviposition, mobility, food consumption and utilization by the legume pod borer, Maruca testulalis (Geyer). They concluded that the greater length and density of nonglandular trichomes were responsible for reduced oviposition by the borer. Trichomes may also have an antibiotic effect on insects. For example, hooked trichomes on bean plants trapped or impaled adult leafminers, Liriomyza trifolii (Burgess) (Quiring et al. 1992), and aphids and whiteflies (Sengonca and Gerlach 1984). Glandular trichomes effectively reduce herbivore feeding through the deployment of adhesives and toxins. Various compounds have been identified as the active components from trichomes (e.g., alkaloids, phenols, and aliphatic hydrocarbons) and are well known feeding deterrents to herbivores (Southwood 1986). An extract oi Solatium

PAGE 20

11 berthaultii leaves containing exudates from type B trichomes inhibited Colorado potato beetle, L. decemlineata (Pelletier and Smilowitz 1 990). 2-Tridecanone and 2-undecanone were the dominant compounds in the type VI trichomes of Lycopersicon species which were toxic to neonate larvae of the tomato pinworm, Keiferia lycopersicella (Walsingham), and the beet armyworm, Spodoptera exigua (Hubner) (Lin et al. 1986). Furthermore, some plants can also rely on chemicals from other types of glands found on the surface of the plant to immobilize, repel and poison herbivores. These include glandular excretory structures, which discharge a variety of oils and resins, and oil glands, which contain various terpenoid oils (Stipanovic 1983). Ten essential oils from Labiate plants, such as spearmint, thyme, and rosemay, inhibit the settling of the green peach aphid, Myzus persicae (Sulzer), on them in choice tests (Hori 1999). In a nochoice test, aphids rarely settled on the sealing film which covered the diet containing spearmint or thyme oil and most of them died. Cotton pigment glands contain gossypol and other terpenoids that provide resistance to Heliothis sp. and other cotton insects (Stipanovic 1983, Parrott 1990, McAuslane et al. 1997). Resistance due to internal factors. In addition to physical and chemical barriers at the epidermis, plants also contain various biochemicals acting as insect repellents, feeding inhibitors and toxins that may act constitutively or are induced by injury. These biochemicals protect plants at different phases of growth and include phenolic compounds (e.g., flavonoids and aromatic acids), terpenoids (e.g., sesquiterpene lactones and heliocides), nitrogenous compounds (e.g., amino acids and amides), proteinaceous compounds (e.g., protease inhibitors and lectin) and toxic seed lipids (e.g., unusual fatty acids) (Hedin 1986). Many such compounds have been isolated from many plants and

PAGE 21

12 have been classified into two broad categories. General resistance is not tissue specific and chemical concentrations increase as the tissue matures (e.g., acids, phenols, terpenes, alcohols, chlorogenic acid, quercetin, tannin, etc.). Specific resistance factors reach their highest concentration in young leaves and fruits and decrease as the crop matures (e.g., sinigrin, tomatine, solanine, gossypol, 2,4-dihydroxy-7-methoxy-l,4-benzoxazin-3-one (DIMBOA)) (Panda and Khush 1 995). Alkaloids, such as nicotine and nomicotine extracted from tobacco plants, were used as early insecticides against insect pests. Pyrrolizidine and indole alkaloids, and other structurally unrelated alkaloids serve as feeding deterrents to many insects (Panda and Khush 1995). Plant lectins, which are carbohydrate-binding proteins of non-immune origin and cause cells to agglutinate and bind glycans of glycoproteins, glycolipids, or polysaccharides, are found to affect survival of lepidopteran, dipteran, coleopteran and homopteran insects (Carozzi and Koziel 1997). Many phenolic compounds are widely distributed among vascular plants and are thought to function as chemical defenses against herbivory because of their ability to interact with proteins and inhibit enzyme functions (Felton et al. 1989). Various phenolic compounds are feeding deterrents to greenbug aphid, Schizaphis graminum (Rondani), and migratory locust, Locusta migratoria (Reiche and Fairmaire), on sorghum; stem borer, Chilo suppressalis (Walker), on rice; leaf beetle, Lochmaeae capreae cribrata (L.), on willow; and bird cherry oat aphid, Rhopalosiphum padi (L.), and English grain aphid, Sitobion avenae (F.), on wheat (see references cited by Smith 1989). Canalicular defense. Latex is typically contained within specialized living cells or a series of fused cells called laticifers, which form a complex network of tubes

PAGE 22

13 throughout the plant (Esau 1977). Latex occurs in 12,500 species belonging to 20 families such as Apocynaceae, Asclepiadaceae, Compositae, Euphorbiacae, and Moraceae (Fahn 1979). The laticifers are under positive pressure, which results in rapid emission of a viscous, often milky latex upon cutting (Data et al. 1996). Several theories have been proposed for the biological role of latex in plants, including a nutritional reserve (Maksymowych and Ledbetter 1987), regulation of water balance (Sen and Chawan 1 972), storage of nonfunctional metabolic by-products (Biesboer and Mahlberg 1 978), and a defensive system containing adhesives or toxins to entrap or deter herbivores (Dussourd 1993, 1995, 1997, 1999). When injured, many plants exude viscous latex which frequently becomes sticky on exposure to air and may entrap or gum up the mouthparts of small herbivores such as aphids, whiteflies and ants (Dillon et al. 1983, Dussourd 1993, 1995). Insects may avoid latex and circumvent mechanical stickiness and possible toxicity of latex either by severing veins or leaf petioles or by cutting trenches before feeding on areas of the portion of the plant isolated by the cuts (Zalucki and Malcolm 1999, Dussourd 1997). A vein-cutting feeding behavior of the chrysomelid beetle, Labidomera clivicollis (Kby.), was first observed on milkweed, Asclepias syriaca, by Dussourd and Eisner (1987). Labidomera adults and larvae bit repeatedly into several adjacent branches of a leaf midrib to induce latex exudation before they moved distal to the cuts and fed on the edge of the leaf with no visible latex emission due to blockage of further latex flow to the sites. Similar counteradaptations also occur in diverse insect groups from 1 1 families including various caterpillars, beetles, katydids, and sawflies

PAGE 23

14 (Dussourd and Denno 1991, Dussourd 1993, McCloud et al. 1995). In each case, the behaviors induce latex drainage and blockage of latex flow to intended feeding sites. Beside adhesives, latex from many plants contains secondary metabolites known to be toxic or deterrent to animals. These include cardiac glycosides in Asclepias latex (Zalucki and Brower 1992, Zalucki and Malcolm 1999); diterpenes (Evans and Schmidt 1976), triterpenoids (Spilatro and Mahlberg 1986) and nonprotein amino acids (Haupt 1976) in Euphorbia latex; sesquiterpene lactones in Cichorium latex (Rees and Harbome 1985) and morphine, berberine, and other alkaloids in Papaver and Chelidonium latex (Roberts 1987, Valle et al. 1987). Latex produced by sweetpotato served as a potential defense mechanism against the sweetpotato weevil, Cylas formicarius (F.). Less weevil feeding damage was found on young vine material with more latex production than on older more mature portions of the vine. Both feeding and oviposition were reduced when latex was applied to the surface of root cores and to a semi-artificial media (Data et al. 1996). First-instar larvae of the monarch butterfly Danaus plexippus L., grew faster and survived better on leaves of North American milkweed species after the latex flow was reduced by partially cutting the leaf petioles (Zalucki and Malcolm 1999). The active compound uscharin isolated from latex of Calotropis procer was highly toxic to land snails (Hussein et al. 1994). Plant latexes from several plants were found to be highly deleterious to many plant parasitic nematodes (Siddiqui and Alam 1990). Induced Resistance Herbivory-induced responses by plants have been the subject of many excellent reviews (Baldwin 1989, Tallamy and Raupp 1991, Karban and Baldwin 1997). Induced defense against herbivores has been recorded in more than 100 plant species in 34

PAGE 24

15 families since the 1970s (Karban and Baldwin 1997). Wound-induced resistance by herbivores or mechanical means can be correlated with the induction of activities of oxidative enzymes such as lipoxygenase (LOX), peroxidases (POD), and polyphenol oxidase (PPO) (Bi et al. 1994), synthesis of primary products such as protease inhibitors, and vitamins (Green and Ryan 1972, Stout et al. 1998, Bi et al. 1994, 1997), enhanced synthesis of secondary metabolites such as phenolic compounds, terpenoids, and nitrogen compounds (Bi et al. 1997, Tuomi et al. 1988, Hartley and Lawton 1991, McAuslane et al. 1997, Baldwin 1991, Baldwin and Ohnmeiss 1994), or decline in the nutritional quality such as protein and amino acids (Bi et al. 1994, 1997). Moreover, many of these induced responses are a complex phenomenon and associated with multiple components of induced biochemical pathways. Several plant species including tomato, potato, tobacco, and cotton have been studied in considerable depth. Growth of larval Helicoverpa zea Boddie was significantly decreased when they fed on previously damaged foliage or squares of cotton compared to the controls (Bi et al. 1994). This was due to a significant decline in host nutritional quality of protein and most amino acids in both foliage and squares after herbivory, increases in activities of oxidative enzymes such as POD, ascorbate oxidase, and diamine oxidase, and increases in the levels of chlorogenic acids and lipid peroxides. Because of the decline in the nutritional quality of foliar protein, and increases in LOX activity, lipid peroxidation products, and trypsin inhibitor content, the growth rate of fourth-instar H. zea was reduced significantly after they fed on wounded tomato foliage (Bi et al. 1994, Bi and Felton 1995). The enhanced production of phenolic compounds was indicated by an increase in the activity of phenylalanine ammonia lyase (PAL) in wounded tissues (Bi and Felton 1995).

PAGE 25

16 There are three time scales proposed for the study of induced responses (Baldwin 1989, Edwards and Wratten 1983). First, highly localized chemical changes occur immediately upon damage and are restricted to the damaged tissues, usually resulting from the mixing of previously isolated enzymes and substrates. The mechanical damage of herbivory caused the cyanogenic glycoside dhurrin in sorghum to come into contact with an enzyme, resulting in the release of hydrogen cyanide and protecting sorghum from further attack by insects (Woodhead and Bemays 1977). Mustard oils, released from damaged Cruciferae crops due to the contact of glucosinolates with the enzyme myrosinase, are known to be toxic to many generalist herbivores (Louda and Mole 1991). On the second time scale, rapidly induced responses may occur within hours or days of the injury and can be systemic or localized to the damaged leaf Phenolic compounds increased within one day in tissues of Chinese cabbage surrounding feeding sites of the bug Lygus disponsi Linnavuori (Hori and Atalay 1 980). Chlorogenic acid and isochlorogenic acid are the most commonly formed substances, but others, such as pcoumaric acid, caffeic acid, and scopoletin are also commonly found (Edwards and Wratten 1983). On the third time scale, more widely-dispersed changes (systemic induced response) may affect an entire organ, branch or plant by de novo synthesis and activation of a suite of phytochemicals within days or years of injury. One example of the third category of induced resistance is the production of protease inhibitors in tomato (Green and Ryan 1972). Colorado potato beetle feeding or mechanical wounding of a tomato leaf induced the rapid accumulation of trypsin and chymotrypsin inhibitors throughout the above-ground plant tissues. Systemic induced resistance has received

PAGE 26

17 much research attention from ecologists because of its potential importance in regulating herbivore populations (Karban and Myers 1989). The patterns of systemic induced resistance may be correlated with the degree of vascular connectivity between damaged leaves and sampled leaves (Jones et al. 1993) based on the theory of the vasculature as the primary path of material transport between leaves (Geiger 1987). In eastern cottonwood, the pattern of systemically induced resistance was found to be directly related to the distribution of the plant vasculature (Jones et al. 1993). Mechanical damage to an individual leaf resulted in systemically induced response in nonadjacent, orthostichous leaves with direct vascular connections. However, the plant vascular connection was not found to be related with the spread and spatial extent of the induced resistance in Betula pendula Roth saplings. Instead, the resistance levels of all leaves increased following the damage (Mutikainen et al. 1996). Host Plant Resistance to Diabrotica balteata Cucurbitacins, some of the bitterest compounds and extremely toxic to most invertebrate and vertebrate herbivores, are well known phagostimulants for diabroticite beetles (Peterson and Schalk 1985, Deheer and Tallamy 1991). It was suggested that selective breeding had either removed or reduced these extremely bitter cucurbitacins to very low levels in leaves and fruits of crop plants (Metcalf et al. 1982). Nugent et al. (1984) reported that bitter seedlings of muskmelon were observed to be more susceptible to feeding by A balteata than the non-bitter seedlings. The genetic study of resistant materials showed that two recessive genes, bibi (non bitter genes) and cblcbl, were responsible for the muskmelon seedling resistance. The bi gene in cucumber may be the major reason for its resistance to D. balteata adults (Da Costa and Jones 1971). Six

PAGE 27

18 varieties of cantaloupe were evaluated for their antixenosis to D. balteata (Overman and MacCarter 1972). One cultivar AC 67-59 was found to be less preferred by the adults. Soybean plant introductions were screened for resistance to two beetles, including D. balteata, in a no-choice laboratory bioassay. Several introductions were found to express a greater degree of resistance to this beetle than others (Layton et al. 1987). Resistance in sweet potato cultivars to larval D. balteata was confirmed by Schalk et al. (1986). Whole roots had the highest resistance level, the cortex was intermediate, and stellar tissue was the least. Antibiosis tended to be expressed more strongly in the early and late season than in the mid-season. In addition, methods to evaluate sweet potatoes for resistance to D. balteata in the field were developed (Schalk and Jones 1 982). Different screening methods have been used to evaluate the resistance of plant genotypes against D. balteata. A rating scale was the most commonly used method for measuring resistance. A subjective rating of D. balteata damage was used in the field to study the resistance in soybean based on the following scales: low, no or very few feeding holes per leaflet on upper three fully expanded leaves; moderate, one to two feeding holes per leaflet on upper three fully expanded leaves; and high, three or more feeding holes per leaflet on upper three fully expanded leaves (Layton et al. 1987). A rating scale from 1 to 5 (1=100% consumed, 2=75% consumed, 3=50% consumed, 4=25% consumed, and 5=no feeding) was applied to screen for resistance in muskmelon to this beefle (Nugent et al. 1984). In order to minimize the effects of plant size on resistance ratings, a leaf disk technique was used and the proportions of the disks consumed by the beetles after 48 h at room temperature were rated visually. Resistance to three species of cucumber beetles including D. balteata was studied in cucumber

PAGE 28

19 seedlings using a 1 to 3 rating scale (0=no feeding, 3=severe damage) (Da Costa and Jones 1971). The extent of feeding damage was measured on a scale of 0-9 to study the susceptibility in melon (Coudriet et al. 1980). Schalk and Jones (1982) developed a method to artificially infest sweet potatoes with eggs of D. balteata to evaluate resistance in this plant. Injury and damage ratings were obtained by counting the total number of holes per root and also by visually rating the damage to each root (a 1 to 5 scale). Layton et al. (1987) used leaf area consumption in addition to a rating scale to determine the resistance level in soybean. The leaf area consumed was then multiplied by the specific leaf weight (SLW=dry weight/leaf area) to obtain adjusted dry weight consumed, which was considered to give a more accurate measure of feeding rate in cases where leaf thickness may differ. Host Plant Resistance in Lettuce Plant Secondary Chemicals Lettuce contains various identified chemical compounds such as terpenes (sesquiterpene lactones), sterols (campesterol, stigmasterol, ect.), flavonoids and other phenolics (caffeic acid, dicaffeoyltartaric acid, rutin), and alkaloids (hyoscyamine) (Gonzalez 1977). Two triterpenes, the guaianolides lactucin and lactucopikrin, were isolated from dry latex of L. virosa, but lactucopikrin was rapidly degraded by enzymes in fresh latex (see references cited by Gonzalez 1977). Sharpies (1964) found that chlorogenic acid, isochlorogenic acid, caffeic acid, glycosides of quercetin and kaempferol existed in lettuce leaves. A large concentration of chlorogenic acid was found in lettuce seeds (Butler 1960). The synthesis and accumulation of phenolic compounds are important aspects of secondary plant metabolism, and the oxidation of

PAGE 29

20 phenolic acids to quinones by the action of PPO is responsible for the darkening of many fruits and vegetables when mechanically damaged (Cole 1984). In particular, chlorogenic acid is one of the substrates for PPO and POD enzymes (Matheis and Whitaker 1984). These enzymes can rapidly convert chlorogenic acid into chlorogenoquinone, a highly reactive molecule known to covalently bind to nucleophilic -NH2 and -SH groups of molecules such as amino acids and proteins. This binding reduces leaf nutritional quality to the herbivore (Matheis and Whitaker 1984). Cole (1984) used high performance liquid chromatography, gas chromatography and UV absorbance to investigate the presence of certain compounds in certain lettuce cultivars and the lettuce root aphid. Pemphigus bursarius (L.). Isochlorogenic acid was found to be the only caffeoylquinic acid detected in quantity and there was a greater concentration in resistant than in susceptible cultivars. The first enzyme in the phenyl propanoid pathway, PPO, was more active in resistant cultivars. The higher level of browning in resistant cultivars was not related to the level of PPO activity but to the higher levels of isochlorogenic acid. The concentrations of caffeoylquinic, chlorogenic and isochlorogenic acids were used as criteria for screening lettuce breeding material for resistance to the lettuce root aphid, P. bursarius, through UV fluorescence (Cole 1987). Genetic Resistance Genetic resistance of lettuce to the lettuce root aphid and several leaf aphids has been extensively investigated in the Netherlands. Almost complete resistance to the leaf aphid, Nasonovia ribisnigri (Mosley), was found in Lactuca virosa L., and transferred to L. saliva through a series of interspecific crosses (Eenink et al. 1982). This resistance is governed by the dominant Nr gene which always locates at the same locus (Eenink and

PAGE 30

21 Dieleman 1982). Nr also confers partial resistance to the green peach aphid, Myziis persicae (Sulz), but has no effect on Macrosiphum euphorbia (Thorn.) (Reinink et al. 1989). Two lettuce cultivars were used to investigate the inheritance of resistance to lettuce root aphid P. bursariiis (Elhs et al. 1994). Resistance was found to be controlled by a single dominant gene, which is linked to the downy mildew resistance gene Dm6. So far, no single source of resistance was found with resistance to all the aphid species. Isozyme electrophoresis is a potential tool for the identification of genetic markers in practical plant breeding programs (Tarksley and Orton 1983). Cole et al. (1991) surveyed allozyme pattern in wild populations of four Lactuca species by routine polyacrylamide gradient gel electrophoresis and found that the resistance of L. sativa to the lettuce root aphid was associated with specific allozyme bands. Nutritional Quality There is some evidence to show that the availability of aromatic amino acids to herbivores is limited and that insects accumulate these substances for their growth (Bemays and Woodhead 1984, Rahbe et al. 1990, Mollema and Cole 1996). Phenylalanine was one of the limiting nutrients for the fifth instar nymphs of Schistocerca gregaria (Forskal) when feeding on lettuce. Supplementing this amino acid invariably increased the efficiency of ingested food conversion to body substance (Bemays and Woodhead 1984). Mollema and Cole (1996) analyzed four important crops including lettuce with unknown levels of resistance to thrips and found that a highly significant positive correlation existed between aromatic amino acid concentrations in leaf protein and thrips damage, regardless of crop species. Electrical penetration graphs of iV. ribisnigri feeding on resistant and susceptible lettuces showed a large reduction in

PAGE 31

22 the duration of food uptake on the resistant hne, leading the authors to suggest that resistance was located in the phloem vessel. Both mechanical blocking of the sieve element after puncturing and a difference in composition of the phloem sap are possible resistant factors (van Helden and Tjallingii 1993). Investigation of phloem sap collections from L. sativa by three different methods (stylectomy, honeydew collection and EDTA chelation) found that the sugar yield from the resistant lettuce was around 30% lower than that on the susceptible plant (van Helden et al. 1994a ). However, no difference in amino acid composition was found in the phloem samples from near isogenic susceptible and resistant lines of lettuce (van Helden et al. 1994b) except that the total concentration of amino acids was 50% lower in the resistant lines compared to the susceptible line. Sucrose was the main sugar in lettuce phloem, but fructose and glucose were also detected. HPLC chromatograms showed numerous unidentified secondary plant compounds in honeydew and EDTA samples, but these compounds were not identified and had no evidence to show that these compounds were involved in resistance. Nuessly and Nagata (1994) found that there were differential responses in the stipple rate by Z. trifolii among four romaine lettuce cultivars (Floricos 83, Parris island Cos, Tall Guzmaine and Valmaine). Females of L. trifolii preferred Tall Guzmaine for oviposition, and lived significantly longer on this cultivar (Nagata et al. 1998). An investigation of leaf surface morphology did not find any obvious difference among the four lettuce cultivars, and no physical barriers were found to be responsible for the difference in stipples by adult leafminers (Nuessly, personal communication). It was suggested that a nutritional factor may be involved in host plant resistance in lettuce to the leafminer (Nagata etal. 1998).

PAGE 32

23 Possible Evidence of Inducible Resistance Plant hypersensitivity is a term primarily used by plant pathologists to describe a response to infection by pathogens as well as to many nonpathogenic stimuli, which encompasses all morphological and histological changes that elicit the premature dying or necrosis of the infected tissue (Femandes 1 990). However, several hypersensitive reactions were also found in plants against insect herbivores such as galling insects, bark beetles, adelgids, and siricids (Femandes 1990). Some evidence indicates that lettuce responds hypersensitively when challenged by the downy mildew fungus, Bremia lactucae Kegel, resulting in highly localized accumulation of phenolics around penetration points. In addition to localized deposition of phenolics, phenolic esters identified as dicaffeoyltartaric and chlorogenic acids increase in concentration during incompatible interactions (Bennett et al. 1996). A second major component of the resistance response in lettuce to this fungus is the localized accumulation of the sesquiterpenoid lactone phytoalexin lettucenin A when a number of cells undergo the hypersensitive reaction because of B. lactucae infection (Bennett et al. 1994). Cells undergoing rapid hypersensitive reactions subsequently collapsed and became brown. Lettucenin A was significantly increased within one day after inoculation, but little change in concentration occurred between 1 and 7 days. Research Goals Host plant resistance to insect pests is a vital component of integrated pest management programs in agricultural systems. The development and use of plant cultivars with resistance to insect pests requires continuous identification of resistant and

PAGE 33

24 susceptible germplasm, characterization of suitable resistant sources, and understanding the mechanisms of these resistances. D. balteata is a polyphagous insect pest occurring throughout the growing season in Florida, which can cause serious problems in lettuce production. Since chemical control is costly and only partially effective, the development of plant resistance is important for controlling this herbivore. Use of cultivars resistant to adult D. balteata also can reduce the need for insecticide applications. Plant resistance to adult D. balteata has been found in muskmelon (Nugent et al. 1984), cucumber (Da Costa and Jones 1971), sweet potato (Schalk and Jones 1982, Schalk et al. 1986), and soybean (Layton et al. 1987). Preliminary studies showed that there were differences in feeding damage by D. balteata adults among four lettuce cultivars so that it is possible to screen cultivars for resistance against D. balteata (Nuessly, unpublished data). The objectives of this research were to develop effective bioassay procedures to screen lettuce for resistance to D. balteata, compare different resistance levels among four romaine lettuce cultivars, examine effects of resistance on D. balteata fitness, and determine physical and chemical bases for this resistance.

PAGE 34

CHAPTER 2 RESISTANCE IN ROMAINE LETTUCE CULTIVARS TO ADULT DIABROTICA BALTEATA Introduction Lettuce, Lactuca sativa L., is one of most important vegetable crops grown in the United States in terms of quantities produced and consumed (Ryder, 1998). As a cultivated crop, lettuce also serves as a favorite plant food to many insect pests including the banded cucumber beetle, Diabrotica balteata LeConte (Nuessly and Nagata, 1993), a polyphagous species with a host range of more than 50 plant species in 23 families (Saba 1970). Economic damage by D. balteata has been reported for other crops, such as com and peanut (Elsey 1988), common bean (Cardona et al. 1982), soybean (Layton et al. 1987), sweet potato (Schalk et al. 1986), and cucurbits (Da Costa and Jones 1971). Nuessly and Nagata (1993) reported that adult D. balteata caused serious damage to lettuce plants, becoming a concern for lettuce growers in south Florida. In order to keep their crops insect free and to reduce economic losses, lettuce growers frequently spray with insecticides. However, the excessive dependence upon pesticides for insect control is associated with many economic, ecological, and environmental disadvantages. Moreover, general feeding habits and widespread dispersal movements of the adults make control of this insect more difficult than if they were more sedentary. Therefore, problems in controlling this pest have underlined the need to find alternative control 25

PAGE 35

26 methods such as host plant resistance, which provides many benefits including its compatibility with other control methods in integrated pest management (Smith 1989). Host plant resistance in lettuce to the lettuce root aphid, Pemphigus bursarius L. (Ellis et al. 1994), the leaf aphid, Myziis persicae Sulz and the green lettuce aphid, Nasonovia ribisnigri Mosley (Eenink and Dieleman 1982), and the potato aphid, Macrosiphum euphorbiae Thomas (Reinink et al. 1989) has been studied for many years. Almost complete resistance to N. ribisnigri was found in L. virosa, and this trait transferred to L. sativa (Eenink et al. 1982). Nuessly and Nagata (1994) found differential feeding response by the serpentine leafminer, Liriomyza trifolii (Burgess), among four romaine lettuce cultivars, including 'Tall Guzmaine' and 'Valmaine'. Significant differences in fecundity, in the total number of stipples (ovipositional puncture wounds in the leaf surface), and in adult L. trifolii longevity were observed under laboratory conditions among the four lettuce cultivars (Nagata et al. 1998). Plant resistance to D. balteata has been found in muskmelon (Nugent et al. 1984), cucumber (Da Costa and Jones 1971), cantaloupe (Overman and MacCarter 1972), sweet potato (Jones et al. 1976, Schalk and Jones 1982), and soybean (Layton et al. 1987), but no previous research has been done on resistance of lettuce to D. balteata. Preliminary studies demonstrated that there were differences in feeding damage hy D. balteata adults among four romaine lettuce cultivars, 'Tall Guzmaine' (TG), 'Parris White' (PW), 'Short Guzmaine' (SG), and 'Valmaine' (Val) in the field (Nuessly, unpublished data), and that resistance may be under genetic control. Therefore, the objectives of this study were to evaluate four lettuce cultivars for resistance against D. balteata using leaf consumption in choice and no choice assays as the criterion, to

PAGE 36

27 examine the effect of a resistant cultivar on the fitness of adult D. halteata, and to determine the mechanisms of resistance in lettuce. Materials and Methods Plants Four romaine lettuce cultivars were selected and screened for resistance to D. balteata based on field observations (Nuessly, unpublished data) and on their pedigree (Fig. 2-1) (Guzman and Zitter 1983, Guzman 1986). Seeds of TG, PW, SG and Val were provided by R. T. Nagata (Everglades Research and Education Center, University of Florida, FL). Seeds of each cultivar were kept overnight in the laboratory in a petri dish lined with a wet filter paper at the bottom for better germination. Germinated seeds were planted in a transplant tray filled with a commercial soil mix (MetroMix 220, Grace Sierra, Milpitas, CA) and grown for 2 wk in a greenhouse with natural light. Seedlings were transplanted to 10-cm diameter plastic pots filled with MetroMix 220. Each plant was watered daily and fertilized weekly with 10 ml of a lOg/L solution of a soluble fertilizer (Peters 20-20-20, N-P-K, W. R. Grace, Fogelsville, PA) from transplantation until the end of the experiment. Plants with seven to eight fully expanded leaves were selected and used for all experiments which were conducted in the laboratory at 26 ± 1°C under artificial illumination (4, 1 10-watt fluorescent bulbs, cool white) with a photoperiod of 14:10 h (L:D). Insects A colony of D. balteata was established from a wild population of adults collected from the field in Belle Glade, Florida in June 1996. Rearing and manipulation

PAGE 37

28 methods were based on those described by Schalk (1986) and Anonymous (1996) with some modifications. Adult cucumber beetles were held in a screen cage (30.5 x 30.5 x 30.5 cm) with a cloth sleeve in an incubator with a photoperiod of 14:10 h (L:D) at 25 ± TC, R.H. 70 ± 10%. Every other day, adults were fed leaves from lima bean (Fordhook 242, Illinois Foundation Seeds, Inc., IL) grown in trays in the greenhouse and sliced sweetpotato tubers purchased from a grocery store. Two egg-collecting devices (8.5 cm diam. x 6.5 cm high) were provided for their oviposition and shaded by placing them under 0.47 L strawberry baskets turned upside down with lima bean leaves placed on the top to stimulate egg deposition. The egg collecting device was a round Rubbermaid® plastic food container (8.4 cm diam. x 7 cm high) covered with a tight-fitting lid into which a 6cm-diam. hole was cut and then covered with a mesh screen (0.1 x 0.2 cm). Three layers of cheesecloth followed by three layers of paper towel (Kimberly-Clark Co., Roswell, GA) and one layer of moistened cotton balls were held tightly agaist the inside of the lid by a layer of upright small vials (20 ml). Cheesecloth with eggs was collected every 2 d and put into a petri dish with a screened lid. The cloth was covered with moisten paper towels cut to fit the petri dish to prevent dehydration. Eggs were allowed to develop in the incubator as above for 4 d before they were surface sterilized with 500 ml Clorox solution (5.25% sodium hypochloride. The Clorox Co., CA) (30 ml/L in water) for 1 min in a cylindrical container (18 cm diam. x 7.5 cm high). Sterilized eggs were then washed with distilled water three times in the container and covered with a damp paper towel. The container was then covered with a plastic screen lid and put back into the incubator. Eggs usually

PAGE 38

hatched after 7 d from the day of egg collection. On day 6, a small amount of sprouted com seeds (H93 x FB37, Illinois Foundation Seeds Inc., IL) was placed in the container so there would be food for the emerging larvae. Com kemels covered with the newly hatched larvae were placed into a soil-free rearing container (32.5 x 17.2 x 10 cm) with germinated com seedlings (H93 x FB37, Illinois Foundation Seeds Inc., IL). These rearing containers were maintained in a room with a photoperiod of 14:10 h (L:D) at 25 ± l^C. After 1 wk, the larvae were transferred into another rearing container with fresh germinated com seedlings to ensure adequate food supply. Com seedlings were prepared as follows: com seeds were soaked ovemight in a Clorox solution (16 ml/L in water), rinsed with tap water the following moming, and stored in a plastic food corltainer (17.2 cm diam. x 15.0 cm high) in the refrigerator until used. Soaked com seeds were evenly placed on one sheet of moistened preassembled germination paper (14 x 29 cm) (Anchor Paper Co., St. Paul, MN). A wick at each end of the paper was placed in a tray filled with fresh water directly undemeath the rearing container. The com seeds were then covered with a double layer of paper towel that was also thoroughly moistened with water. The rearing container was covered with a lid that had a screened opening (7.5 x 13 cm) for ventilation. Seeds were allowed to germinate and grow for 3 d before their use as food for larvae. After larvae grew for 1 5 d, they were almost ready to pupate and were transferred gently using forceps into a pupation container (18 cm diam. x 7.5 cm high) filled with moistened commercial soil mix (MetroMix 220). The soil mix was autoclaved at 20 psi and 12rc for 99 min. The potting soil was kept moist enough that it did not dry out for the entire pupal stage, but it was never saturated. The pupation container was covered

PAGE 39

30 with a screened lid covered with dampened towel to retain moisture. Adults emerged around 12 d later and were transferred into the screen cage (30.5 x 30.5 x 30.5 cm) mentioned above for the colony use. Beetles to be used for bioassays were separated by sex during the pupal stage. Female pupae have two papillae below the anal opening and males lack such papillae (Schalk 1 986). These adults were collected within 24 h of emergence, transferred to a plastic box (13.5 cm diam. x 13.1 cm high) with a ventilated top and sides containing a moistened paper towel, and held in the incubator for another 24 h without food prior to bioassays. Binary-choice Test Beetles were first given a choice between pairs of different romaine cultivars to evaluate feeding resistance. Beetles were exposed to the fully expanded attached leaves within feeding arenas made from plastic petri dishes (8.9 cm diam.). Two round holes (2.9 cm diam.) 65 mm apart were cut from the bottom of the dish, and a ventilation hole (5.8 cm diam.) was covered with gauze material at the top of the dish. The feeding arena was attached to the upper surface of two leaves of different cultivars using hair clips. These leaves were matched for position and age on each plant. Two females and one male were released into each arena. After 48 h, the section of leaf containing the damaged site was cut from the whole leaf and placed between two sheets of acetate transparencies. Leaf sections were scanned (JADE 2, Linotype-Hell, Taiwan) and imported into an imaging program (ImagePC beta version 1 , Scion Corporation, Frederick, MD) where area eaten or leaf area remaining was determined. Feeding preferences were determined by testing all possible pair-wise combinations of the four lettuce cultivars. Each pair-wise combination was replicated 10 times, and all six pair-

PAGE 40

31 wise combinations were tested simultaneously. Leaf areas consumed in feeding bioassays were expressed as area eaten in square mm. The difference of leaf area consumption between cultivars was analyzed by paired Mest using Proc MEANS (SAS Institute 1999). No-choice Test Resistance to beetle feeding was next tested in no-choice tests using fully expanded attached leaves. One pair of beetles was confined on the seventh leaf (from the cotyledon) of a plant using a clip cage with a 4 cm diam. hole through which beetles accessed the upper leaf surface. The adults were allowed to feed for 48 h. Leaf area consumed was measured as described previously. This study was arranged as a randomized block design with each cultivar in each block. A block was the time when the four treatments (lettuce cultivars) were set up. Each block was replicated sixteen times. Leaf area consumed in the feeding bioassay was measured as above and analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD test with a significance level of a=0.05 (SAS Institute 1999) was used for post-hoc means separation. Fitness Bioassay Beetle weight change and egg production were used to bioassay fitness effects of feeding on the four romaine varieties. Thirty, 48 h-old unfed females were randomly chosen from the colony and weighed individually before testing. These beetles were then released together with 1 5 males into a plastic cylindrical cage (60 cm high x 25 cm diam.) with ventilated top and sides containing two plants of the same cultivar. The shortest period from emergence to first egg deposition for D. balteata observed in the lab was 10 d. Therefore, adult beetles were allowed to feed for 10, 13 or 16 d without

PAGE 41

32 providing an oviposition apparatus. New plants were constantly provided for the adults based on the level of damage in each cage except for the cage with Val, where Val plants were changed every 3 d because they were hardly damaged. At the end of the experiments (either 10, 13 or 16 d later), living beetles were counted and weighed individually. Beetles were stored in a freezer (-9°C) until they could be dissected under a dissecting microscope to determine the number of fully developed eggs in their ovaries. The experimental design was a 2 x 2 factorial with cultivar having four levels (TG, PW, SG, Val) and feeding time having three levels (10, 13, 16 d). The effects of cultivar on beetle egg production, weight change, feeding time and their interaction were analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD with a significance level of a=0.05 (SAS Institute 1999) was used for post -hoc means separation. Starvation Test Changes in egg production and adult weight may be associated with antibiotic or antixenotic mechanisms. Therefore, for purposes of comparison with feeding tests, starvation tests were conducted as an approximation of extreme antixenotic or poor food nutrition mechanisms on fitness parameters. The experiment was designed as a randomized complete block design. Four cylindrical cages were set up at one time as a block. Twenty, 48 h-old-unfed females along with 10 to 25 males were released into each plastic cylindrical cage (60 cm high x 25 cm diam.). Each block was replicated three times. Each cage received the same number of adults in each block. Each cage contained either one Val plant, one TG plant, one Val plant totally covered by a cloth, or one TG plant totally covered by a cloth. All cages contained four small water containers (4.5 cm high x 3.5 cm diam.). The cloth-covered plants were inaccessible to beetle

PAGE 42

33 feeding but provided humidity equal to that in the other two treatments. The plant pots were also covered by plastic bags to prevent egg deposition into soil. Plants in all treatments were changed every 3 d including the cages with covered plants. After 7 d, living beetles were counted and females were dissected to check their ovary development and egg production. The total percentage of adult survival, the percentage of female survival and the percentage of females with yolk deposition in their ovaries in each treatment were subjected to arcsin(sqrt(x)) transformations before they were analyzed by Proc GLM. The number of mature eggs/female was also analyzed by Proc GLM (SAS Institute 1999). Tukey's HSD with a significance level of a=0.05 (SAS Institute 1999) was used for post-hoc means separation. Results Binarv-choice Tests Adult D. balteata ate significantly less from Val leaves than from leaves of the other cultivars when they were given a choice between Val and SG, Val and PW, or Val and TG (Table 2-1). However, the beetles ate over two times as much surface area of Val when it was paired with PW as when it was paired with the other three varieties. D. balteata consumed almost 10 times more TG foliage than Val foliage when they were given a choice (Table 2-1). There was no significant difference (P >0.05) in feeding damage between TG and PW, TG and SG, or SG and PW. Beetles ate numerically more surface area of TG than of PW or SG in binary-choice tests, but the difference was not significant at P=0.05. D. balteata feeding damage among the four tested cultivars appeared to be in the order Val
PAGE 43

34 Table 2-1. Lettuce leaf area consumed by three adult D. balteata in 48 h when presented a choice between two cultivars Choice Cultivar' N Leaf area eaten (mm^) ± SEM Pr > ITI" Val 10 43.2 + 5.7 Val-SG 0.0019 SG 10 131. 7± 19.9 Val 10 23.816.0 Val-TG 0.0001 TG 10 235.0133.9 Val 10 122.6126.6 Val-PW 0.0001 PW 10 393.7147.8 TG 10 409.2 1 49.8 TG-PW 0.1908 PW 10 293.6 1 65.8 TG 10 190.8 1 35.9 TG-SG 0.1235 SG 10 120.81 12.7 SG 10 128.9137.0 SG-PW 0.2396 PW 10 174.5 1 18.1 " TG=Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. P value from paired Mest.

PAGE 44

No-choice Test Significant differences in leaf consumption were observed among the four cultivars in no-choice feeding tests (F=52.02; df=3,56; P=0.0001) (Table 2-2). Adults consumed the greatest amount when confined on TG, followed by PW, then SG and Val. Therefore, the order of feeding damage observed in the binary-choice tests was confirmed by the results of the no-choice tests. There was no significant difference in area consumption between SG and Val at the 5% level, even though adult D. balteata ate just over half as much Val as SG. The beetles did some damage on Val, but the amount consumed was less than one-sixth of that on TG. Fitness Bioassay The mean number of mature eggs produced per female significantly differed after feeding on the four cultivars for 10 d (F=66.02; df=3, 134; P=0.0001), 13 d (F=131.81; df=3, 134; P=0.0001), and 16 d (F=91.93; df=3, 134; P=0.0001) (Table 2-3). The length of feeding time (i.e., 10, 13, 16 d) also significantly affected egg production (F=9.18; df=2, 134; P=0.0002). There was a significant interaction between the lettuce cultivars and the length of feeding time on egg production (F=6.26; df=6, 134; P=0.0001). Egg production was significantly higher for adults feeding on PW than on the other varieties for 13 and 16 d. After 13 d, beetles that fed on PW produced a mean of 1 12 eggs compared to 93 on TG and SG and none on Val. No fully developed eggs were found in the ovaries of female beetles reared on Val for 10, 13 or 16 d. Furthermore, of the 42 surviving females out of 90 originally feeding on Val that were dissected, only three had some small undeveloped eggs after 13 or 16 d of feeding. In contrast, the egg production of beetles on the other three cultivars was very high. Egg number increased to the

PAGE 45

Table 2-2. Mean leaf area consumed per pair of adult D. balteata in 48 h confined on four lettuce cultivars in a no-choice situation Cultivar' N Leaf area eaten (mm^)/pair ± SEM TG 14 366.0 ±31.7 a PW 15 304.21 ± 24.0 b SG 15 93.8 ± 15.3 c Val 16 53.5 ± 8.3 c ' TG=Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val=Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.

PAGE 46

37 Table 2-3. Number of fully developed eggs in ovaries of D. balteata females confined for different lengths of time on one of four lettuce cultivars Cultivars' Time (Days) (Mean ± SEM) ~ 10 13 16 TG 86 ±4 a 93 ±6b 87 ±5 b PW 65±6b 112±4a 115± 10a SG 56± 11 b 93±8b 76±12b Val Oc Oc Oc ^TG-Tall Guzmaine, PW=Parris White, SG=Short Guzmaine, and Val^Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.

PAGE 47

38 highest on day 13 and decreased slightly on day 1 6. On day 13, all of the females on TG were full of mature eggs, but on day 16, the percentage of females with fully developed eggs had dropped to 70%. This decrease in the number of mature eggs in the ovaries of female beetles was also found on SG but not on PW after feeding for 16 d. The weight of females reared on Val was significantly lower than that of females reared on TG and PW (F=7.15; df=3,337; P=0.0001) for 10 d (Table 2-4). The weight of females reared on Val was also significantly lower than that of females reared on TG, PW and SG forl3 d (F=32.93; df^3, 227; P=0.0001) and 16 d (F=21.71; df=3, 227; P^O.OOOl). Females feeding on TG gained weight, ranging from 6.24 to 9.49 mg, while females feeding on Val gained only 1.31 to 2.61 mg. The highest weight gain occurred on day 1 3 for each cultivar. Percentage of mortality for females on TG ranged from 7 to 33% when females were allowed to feed for 10 to 16 d. This range was lower than that on Val with a range from 40 to 73%. On PW and SG, the ranges of percentage of female mortality were from 27 to 47% and from 17 to 47%, respectively. Starvation Test The percentage of D. balteata still alive on TG (mean of 79%) at the end of the 7d experiment was the highest among the four treatments (Table 2-5). When D. balteata had access to only water, its mortality was equally high on the TG-covered (1 1%) and the Val-covered treatments (7%). In each block, 3.3% to 20% of beetles survived when no food was available 7 d after the experiment was initiated. Therefore, all treatments were stopped in order to have some data in treatments with cloth-covered plants. Female

PAGE 48

39 Table 2-4. Mean weight per female D. balteata feeding on four lettuce cultivars for different lengths of time Cultivars" Time (Days) (mg ± SEM) 0 10 13 16 TG 11.1 ±0.16 a 19.8 ±0.9 a 20.0 ± 0.5 a 18.2 ±0.9 a PW 11.5 ±0.2 a 17.0 ± 0.5 b 20.5 ± 0.5 a 20.3 ± 0.6 a SG 10.4 ±0.2 a 13.0 ± 0.6 c 18.2 ± 0.5 b 16.5 ± 0.4 ab Val 11.0 ±0.4 a 12.5 ± 0.8 c 13.7 ± 0.4 c 13.3 ± 0.7 c ''TG-Tall Guzmaine, PW^Parris White, SG=Short Guzmaine, and Val=Valmaine. Data within a column followed by the same letter are not significantly different by Tukey's HSD test at the 0.05 level.

PAGE 49

40 Table 2-5. Results of starvation test in which adult D. balteatci were confined with food or no food for 7 d (Mean ± SEM) Treatment Total Female Percentage of females Numbers of mature survivorship % survivorship % with yolk deposition eggs/female TG 79±2a 62±9a 97 ±3 a 95 ±6 a Val 49±3b 40±3b 18±9b O.Ob TG-covered 11 ±2c 12±3b O.Ob O.Ob Val-covered 7±3 c 10±3b O.Ob O.Ob TGor Val-covered means plants of these varieties were covered by a cloth so that beetles could not access them as a food source. Data within a column followed by the same letter are not significantly different by Tukey's mean separation test at the 0.05 level. k.

PAGE 50

41 survival was the highest on TG, followed by Val, TG-covered, and Val-covered (Table 25). Of all the females dissected in each treatment, all of them except one female on TG had at least yolk deposition in their oocytes, but only 1 8% of females surviving on Val had yolk deposition after 7 d. The ovaries of females without food had no yolk deposition and they were transparent, looking much like the ovaries of newly emerged females. Mature eggs were found in 20% of the females feeding on TG, averaging 95 eggs/female with a range from 74 to 11 9. Discussion There are several possible hypotheses that may explain the resistance in Val lettuce. The four romaine cultivars tested are genetically related to each other so that it is possible that the resistance is under genetical control. This kind of resistance may be expressed by physical or chemical defenses. Binary-choice and no-choice tests indicated that Val exhibits the highest degree of resistance among the four lettuce cultivars tested, and that TG is the most susceptible cultivar to D. balteata damage. The minimal feeding damage on Val indicates that it may contain deterrents or lack feeding stimulants, either at the leaf surface or internally. This feeding will be further investigated in following chapters. Val has been reported to be nonpreferred over TG by L. trifolii, both in choice and no-choice tests, when four lettuce cultivars including TG and Val were provided as the only food and oviposition source (Nuessly and Nagata 1994, Nagata et al. 1998). Female leafminers survived significantly longer and produced more pupae on TG than leafminers reared on the other three cultivars, including Val (Nagata et al. 1998).

PAGE 51

42 That Val consistently and significantly exhibited resistance compared with TG was also seen in the fitness bioassay. Females that feed on Val had the lowest weight among beetles that fed on the four cultivars (Table 2-4). Moreover, females feeding on Val produced no mature eggs in three feeding periods when they had no choice of host plant (Table 2-3). D. halteata adults did not feed much on Val (Table 2-1, Table 2-2) due to the possible presence of lettuce physical or chemical defenses, causing beetles not to have enough nutrients to produce eggs. In contrast, D. balteata released on TG fed significantly more and generated more eggs. Egg production declined at 16 d in females fed on TG, PW and SG, which may have been due to egg reabsoption by females who were not provided an oviposition site, or the eggs may have been deposited before dissection. A few eggs were found in a slit on the plastic bag, which was used to cover soil. Based on the starvation test (Table 2-3), most D. balteata could not survive without food for 7 d. This was dramatically less than its normal longevity of at least 1 month based on our observations when supplied with sufficient food. Because their survival was in jeopardy, living females were not able to develop their reproductive systems. Almost 50% of adult D. balteata did survive on Val for 7 d by feeding on yellow leaves, major veins of fully expanded leaves, and on dead beetles, but most of them still did not develop eggs. This finding confirmed that females on Val have no mature egg development because of their inability to use the resistant cultivar Val as a food. It is possible that Val has some antixenotic characters which keep D. balteata from using it as a host, resulting in the size of its next generation dramatically decreasing as compared with that on the susceptible cultivar TG. When D. balteata was given a choice

PAGE 52

43 between TG and Val, some biophysical or biochemical factors from Val may force it to choose TG over Val. D. halteata were not randomly distributed on lettuce plants; they liked to gather together and feed on a single plant before they moved onto another plant in a cage. Such feeding behavior contributed to the large standard deviations observed in many experiments. This may be due to the release of pheromones given off by the feeding insects (Regnier and Law 1968, Schalk et al. 1990). The same feeding phenomennon was also found when D. balteata fed on cucumber (Da Costa and Jones 1971). Feeding together may also help adults to break down physical defenses possibly involved in lettuce resistance. Differences in feeding damage by D. balteata appear to closely follow the pedigrees of the tested cultivars (Fig. 2-1). Val and SG showed similar levels of resistance when beetles had no choice of food. SG is the result of a cross between Val and another cultivar so it is possible that SG inherited the resistance from Val. However, other modifying characters may be also involved because SG was significantly preferred over Val when D. balteata had a choice. Feeding on Val was significantly increased when paired with PW in binary-choice tests. This suggested that PW might contain D. balteata feeding stimulants. Although TG and Val are related, TG was produced from a direct cross between SG and PW. Therefore, the susceptibility of TG was probably inherited from PW. This hypothesis was also supported by Nuessly and Nagata (1994) who proposed that the susceptibility of TG to L. trifolii was probably introduced with the cross to PW.

PAGE 53

This study indicates the importance of breeding and selection processes in developing resistant cultivars. Val was the leading cultivar of romaine lettuce grown in Florida organic soils, but it was susceptible to thermodormancy, premature bolting, lettuce mosaic virus and corky root rot (Guzman 1 986). Therefore, plant breeders began to develop and select TG with improved characteristics over Val. However, insect resistance was not included in the selection criteria (Nuessly and Nagata 1994), and unfortunately TG is a suitable host for L. trifolii and D. balteata. Foliar feeding by D. balteata causes problems in lettuce production such as decreasing the photosynthetic capacity of the leaves, introducing frass into the heads, and opening the plants to pathogenic infection. These problems result in yield losses and increased production costs. In the southern United States, D. balteata adults can be found during much of the year (Schalk 1 986). Moreover, D. balteata has a very high reproductive potential; oviposition can last from 2 to 8 wk, and one female can lay 849 eggs in her lifetime (Pitre and Kantack 1962). Because chemical control of the soilinhabiting larvae of this beetle is unreliable and incapable of reaching all the larvae and eggs in the soil (Schalk et al. 1986), an ideal way to reduce D. balteata populations may be by controlling adult beetles (Schalk et al. 1990). This study showed that Val has a dramatic potential ability to retard D. balteata population development by decreasing the weight and vigor of the adults, making them more susceptible to other abiotic and biotic mortality factors. The detailed mechanism of resistance in lettuce to the D. balteata needs to be further studied to help scientists develop new cultivars with desired characters, including insect resistance. Host plant resistance is of particular importance

PAGE 54

45 to control D. balteata with its multiple generations per year in Florida. This strategy will reduce the impact of this insect on lettuce and reduce pesticide usage.

PAGE 55

46 FL 746 . Valmaine _ Parris White > FL 1142 Valmaine Tall Guzmaine Short Guzmaine Floricos 83 Fig. 2-1. Pedigree relationship of lettuce cultivars.

PAGE 56

CHAPTER 3 ROLE OF THE LEAF SURFACE OF LETTUCE IN RESISTANCE TO DIABROTICA BALTEATA Introduction The outer surface of all plants is coated with layers consisting of a lipid polymer and a mixture of extractable lipids, sometimes referred to as epicuticular waxes, which reduce water loss and help to block the entry of pathogenic fungi and bacteria (Esau 1976). The chemical compositions of these layers vary among species, among genotypes within a species, and among parts within a plant (Eigenbrode and Espelie 1995). These variations are unlikely to affect greatly the ability of the surface lipids to act as a barrier to moisture loss, but are likely to have a variety of ecological influences, including mediation of insect-plant interactions (Eigenbrode and Espelie 1995). The plant surface is the first physical contact between an insect and a plant when the insect lands or touches the plant (Schoonhoven et al. 1998). Important environmental cues, which lead insects to make decisions to feed or oviposit on a host plant, are associated with the leaf surface contact sensory cues (Derridi et al. 1996). Other than epicuticular waxes, various physical or chemical factors encountered at the plant surface sometimes are deployed in glandular and nonglandular trichomes or in resinous secretions (Southwood 1986, Oghiakhe 1995, Lin et al. 1986, Hori 1999, McAuslane et al. 1997). The plant surface, especially its chemistry, constitutes the first line of resistance to insects (Schoonhoven et al. 1998). Contact with these surface chemicals often suffices to prevent insects from further investigation or usage of the plant (Schoonhoven et al.l998). 47

PAGE 57

48 Epicuticular waxes on the plant surface are known to influence insect feeding, oviposition and movement either by physical (Stork 1980, Bemays et al. 1983) or by chemical mechanisms (Adati and Matsuda 1993, Eigenbrode and Espelie 1995, Eigenbrode and Shelton 1990, Eigenbrode et al. 1991, Lin et al. 1998). Epicuticular waxes are usually extracted from a plant surface using an organic solvent such as chloroform, methylene chloride or hexane (Bakker et al. 1998). Such extracts often contain sugars, amino acids, and secondary plant substances such as glucobrassicin (a glucosinolate), furanocoumarins and alkaloids (Table 3.7 in Schoonhoven 1998). A resistant cultivar of Sorghum bicolor (L.) Moench was shown to become more acceptable to nymphs of Locusta migratoria L. once waxes had been removed with chloroform (Woodhead 1983). Rinses of Solarium berthaultii Hawkes leaves with methylene chloride deterred Colorado potato beetle feeding when applied to S. tuberosum tuber and leaf disks (Yencho et al. 1994). Surface lipid extracts with hexane from a rice cultivar resistant to the brown planthopper, Nilaparvata lugens (Stal), deterred feeding and increased restlessness of this insect when the extracts were applied to the surface of susceptible plants (Woodhead and Padghm 1988). Most plant waxes consist of a few major classes of aliphatic components including n-alkanes, wax esters, free fatty alcohols and free fatty acids (Eigenbrode and Espelie 1995). Differences in epicuticular lipid composition were associated with resistance of cultivated plants to herbivores or with herbivore behavioral responses (Eigenbrode and Espelie 1995). For instance, y-hydroxybenzaldehyde was identified as an antifeedant present in the surface wax of seedling S. bicolor to nymphs of L. migratoria (Woodhead 1982). Reduced feeding by the spotted alfalfa aphid, Therioaphis

PAGE 58

49 maculata (Buckton), was found in alfalfa genotypes with high levels of triacontanol (Bergman et al. 1991 ). Aphid resistant sorghums had higher levels of triterpenols in the surface wax than did susceptible sorghums (Heupel 1985). On the other hand, plant surface chemicals may help some insects to recognize their specific host plants. Several species of chrysomelid beetles are stimulated to feed by leaf surface wax of their various host plants (Adati and Matsuda 1993, Lin et al. 1998). The composition of the leaf surface of lettuce, Lactuca saliva L, has been reported recently (Pilipenko et al. 1994, Bakker et al. 1998). Pilipenko et al. (1994) reported that neutral lipids such as hydrocarbons, sterols and free fatty acids predominated on the lettuce leaf surface, followed by glycolipids and phospholipids. Esters of higher fatty acids and lower alcohols were also important components of the surface chemicals of lettuce. Leaf rinses, made with chloroform, methylene chloride, toluene, hexane, and a mixture of chloroform and methanol, all consisted of long-chain linear alcohols and minor amounts of fatty acids (Bakker et al. 1998). In Chapter 2, both feeding choice and no-choice experiments showed that Valmaine had the highest level of resistance to the banded cucumber beetle, Diabrotica balteata LeConte, among four lettuce cultivars (Tall Guzmaine, Short Guzmaine, Parris White, and Valmaine). Valmaine when fed upon had an ability to inhibit or delay normal development of the reproductive system of D. balteata. These results indicate that Valmaine may contain deterrents or lack feeding stimulants, either on the leaf surface and/or inside of leaves. The objective of this research was to determine the possible role of the leaf surface chemistry in lettuce resistance to D. balteata. Feeding bioassays were

PAGE 59

50 conducted on solvent-treated (hexane, methylene chloride, and methanol) lettuce leaves and on palatable lima bean leaves to which leaf rinses were applied. The amounts of extractable lettuce leaf surface components were also quantified. Materials and Methods Plants Based on the results in Chapter 2, Valmaine (Val, resistant) and Tall Guzmaine (TG, susceptible) were used to search for possible resistance factors on the leaf surface. Seeds of lettuce cultivars (provided by R. T. Nagata, Everglades Research and Education Center, University of Florida, PL) were germinated overnight in a petri dish lined with a moistened filter paper in the laboratory for improved and uniform germination. Germinated seeds were planted in a transplant tray filled with a commercial soil mix (MetroMix 200, Grace Sierra, Milpitas, CA) and grown for 2 wk in a greenhouse with natural light before transplanting into 10-cm diameter plastic pots filled with MetroMix 200. Each plant was watered daily and fertilized weekly with 10 ml of a lOg/L solution of a soluble fertilizer (Peters 20-20-20, N-P-K, W. R. Grace, Fogelsville, PA) from transplantation until the end of the experiment. Fordhook 242 Lima bean seeds were purchased from a commercial company (Illinois Foundafion Seeds hic, IL) and sown in seedling tray cells containing MetroMix 200 in the same greenhouse where lettuce was grown. Lima bean plants used for sustaining the beetle colony were watered daily and fertilized weekly with the same solution as used for lettuce plants after they reached the first-true-leaf stage. Lima bean plants used for experiments were watered daily but were not fertilized because only the first true leaves were used.

PAGE 60

5 ;.JJB;"BB
PAGE 61

52 removed per unit area. Data were analyzed as a 2 x 3 factorial design by Proc GLM (SAS Institute 1999), in which cultivar was treated as one factor with two levels, and solvent was treated as the other factor with three levels after data were checked to conform to assumptions of normality. Post-hoc means separation was conducted using Tukey's HSD test (a=0.05, SAS Institute 1999). Feeding Response to Solvent-extracted and Normal Leaves The seventh or eighth fully expanded leaves were selected from plants with eight fully expanded leaves for this assay. One pair of beetles was placed in a binary-choice feeding arena (described in Chapter 2) and was given a choice between a solvent-dipped leaf and a non-treated normal leaf, both still attached to the plants. After 48 h, the section of leaf containing the damaged site was cut from the leaf and placed between two sheets of acetate transparencies. Leaf sections were scanned and imported into an imaging program where area consumed was determined. Solvent-dipped leaves were immersed in methylene chloride for 5 s or in methanol for 30 s. Hexane was also tested, but phytotoxicity was so rapid and severe that the leaves were not suitable for the assay. Normal leaves were not dipped in any solvent. The feeding response of adult D. balteata was tested in all five possible pair-wise combinations between leaves with and without surface removal for each cultivar (Table 3-2). The control pair, TG vs. Val, was also conducted to monitor the resistance level. There were eight replicates for each combination. Area consumed on each leaf in each treatment combination was analyzed by paired /-test using Proc MEANS (SAS Institute 1999).

PAGE 62

53 Feeding Response to Crude Extracts in Binary-choice Test Lima bean leaves were used as the substrate to test the effects of lettuce leaf wax crude extracts on D. balteata feeding. The first true leaves were cut from greenhousegrown lima bean plants and their petioles immediately immersed in beakers with tap water. A binary-choice test arena (described in Chapter 2) was used to expose a pair of adult D. balteata to two leaf areas on the same lima bean leaf for 24 h. Based on the results of experiments of feeding response to solvent-extracted and normal leaves (no significant difference in feeding between Val leaves and Val leaves with surface chemicals removed by methanol), only methylene chloride was chosen to extract lettuce leaf surface chemicals for the binary-choice test. The upper surfaces of two leaf areas were sprayed with either solvent or crude extract depending on the experiment design as described below. Extracts were applied at five different concentrations based on the natural concentration of methylene chloride-extractable chemicals on TG or Val leaf surfaces. The different concentrations were made from stock solution and diluted with methylene chloride so that 100 |..il was applied to each confined area. Leaf area consumption was determined as described above and analyzed with paired /-test using Proc MEANS (SAS Institute 1999). A micro spray device (provided by H. T. Albom, USDA, ARS, Gainesville, FL) was used to apply extracts. This device leads nitrogen gas through the horizontal tube to cause the solution to be drawn up through the vertical tube, resulting in formation of a fine spray. The solvent was allowed to evaporate completely in the fume hood before the beetles were confined on the leaves. The extraction process and concentrations used in the bioassays are described below. i

PAGE 63

Crude extracts. The sixth through eighth fully expanded lettuce leaves were cut from the plants of TG or Val. Extracts of leaf surface chemicals were prepared by dipping excised leaves in methylene chloride for 5 s. Extracts were filtered into a flask, concentrated to less than 5 ml under a nitrogen stream, then transferred into preweighed 5 ml glass vials. The solvent was continuously evaporated to dryness under a stream of nitrogen. The total amount of chemicals removed from Val or TG was determined by weighing the residues. The vials were then completely wrapped with aluminum foil and temporarily stored in a refrigerator freezer (-9°C) before bioassays. The stock solutions of crude extracts were made by redissolving the residues with methylene chloride in the vials. Extracts from TG vs. solvent. The five concentrations chosen for TG were 13, 27, 41, 55, and 69 (xg/cm^. Among them, 27 (xg/cm^ was its natural concentration. One of two exposed areas confined by the binary-choice feeding arena was sprayed with 1 00 i^l of an extract of designated concentration, the other was sprayed with only 1 00 |il methylene chloride. Each combination was replicated 1 5 times. Extracts from Val vs. solvent. Five concentrations, 5, 11, 16, 22, and 27 \xg/cm^, were also applied on lima bean leaf surfaces. Among them, 1 1 ^ig/cm was Val's natural concentration. One leaf area was treated with 100 yil extracts and the other with 100 )al solvent. Each combination was replicated 1 7 times. Extracts from TG vs. from Val. Five pair-wise combinations between these variefies were set up as follows: 1 1 \ig/cm^ TG vs. 27 ]xg/cm^ Val, 1 1 ^g/cm^ TG vs. 1 1 Hg/cm^ Val, 27 ^ig/cm^ TG vs. 1 1 ^g/cm^ Val, 27 ^g/cm^ TG vs. 27 ^ig/cm^ Val, and 19 (j,g/cm^ TG vs. 19 ^g/cm^ Val. Two confined leaf areas on the same lima bean leaf were

PAGE 64

55 sprayed with 100 |li1 extracts of either Val or TG at the designated concentrations. Each combination had 1 6 rephcates. Feeding Response to Crude Extracts in No-choice Test The same feeding arenas used in the binary-choice tests were used for this test, but only one hole was clipped on a lima bean leaf sprayed with crude extracts from either TG or from Val. The other hole was covered to prevent beetles from escaping. Crude extraction, preparation and spraying strategy were done in the same manner as described above. Five different concentrations and 16 replicates for each concentration were used. The five concentrations applied on lima bean leaf surfaces for TG or Val were 6, 10, 17, 27, and 55 [ig/crn^. The experiment was set up and analyzed as a 2 x 5 factorial design with cultivar as one factor and concentration as the other factor using Proc GLM (SAS Institute, 1999). Post-hoc means separation was performed using Tukey's HSD test (a=0.05, SAS Institute 1999). Resuhs Solvent Extract Comparisons The weight of surface components extracted per unit of leaf area was not significantly different among the three solvents (F=2.97; df=2,18; P=0.0766) (Table 3-1). However, this statistic was strongly affected by the nearly identical values for TG crude extract weights among the tested solvents. Post-hoc tests (Tukey's HSD) run on Val data do indicate significant means separation between crude extract weights for methylene chloride and methanol. The weight of surface chemicals extracted was significantly different between the two cultivars (F=87.41; df=l, 18; P=0.0001). No interaction was observed between cultivar and solvent, which means that any differences in the weight of

PAGE 65

56 Table 3-1 . Total amount of crude material extracted from the leaf surface of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce using three solvents Solvent Crude extracts (ng/cm^ ± SEM) TG Val Hexane 27.8 ±3.1 14.5 ±0.2 CH2CI2 27.4 ±1.6 11.0 ±1.2 Methanol 27.8 ± 0.4 18.8 ±1.7

PAGE 66

57 components per unit of leaf area among three solvents are the same for both cultivars, and any differences between the cultivars are the same for each solvent (F=2.48; df==^2,18; P=0. 1117). Therefore, data from three solvents on different cultivars were pooled to study the effect of cultivar on the weight of leaf surface chemicals removed. The weight of components extracted from TG, with the average of 28 ± 1 )ag/cm^ (mean ± SEM), was significantly higher than the weight of components extracted from Val with the average of 15 ± 1 |ag/cm^ (mean ± SEM) per unit of leaf area (F=66.52, df=l, 22; P=O.OOOI). Feeding Response to Solvent-extracted or Normal Leaves Solvent extraction of lettuce leaf samples led to increased leaf consumption by beetles on both TG and Val compared to untreated leaves. Beetles consumed at least twice as much lettuce leaf tissue from methylene chloride-extracted leaves of TG and Val (TG + S, Val + S) than from untreated leaves (TG, Val) of the same cultivars (Table 3-2). However, statistically significant differences (P < 0.05) existed only between Val and Val-HS. Similar feeding patterns for TG versus TG +S and for Val versus Val + S were also found between methanol-dipped leaves and untreated leaves, but differences were not significant (Table 3-3). Removing leaf waxes and their constituents from Val led to large increases in feeding compared to untreated leaves. However, beetles still fed significantly more from treated and untreated TG leaves than from treated Val leaves. In the rest of the treatment combinations, regardless of whether leaf surface chemicals were removed by methylene chloride or methanol, adults still strongly preferred to feed on leaves of TG than on leaves of Val (Tables 3-2, 3-3). Methylene chloride extraction led to greater increases in beetle feeding than did methanol extraction, particularly for Val.

PAGE 67

58 Table 3-2. Mean amounts of feeding by adult D. halteata on leaves of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce with and without surface layer removal by methylene chloride (CH2CI2) Choice Cultivar^ N Leaf area eaten (mm" ± SEM) Pr > |T| TG 7 159.2 ±45.1 TG-TG+S 0.0938 TG+S 7 346.8 ± 59.9 Val 8 16.6 + 3.7 Val-Val+S 0.0015 Val+S 8 193.6 ±36.7 TG 8 326.1 ±57.4 TG-Val 0.0009 Val 8 12.0 ±2.5 TG+S 8 442.4 ± 60.7 TG+S-Val+S 0.0030 Val+S 8 58.1 ±38.9 TG+S 8 223.4 ±31.9 TG+S-Val 0.0003 Val 8 37.6 ±6.9 "TG+S = leaves from TG treated with CH2CI2. Val+S = leaves from Val treated with CH2CI2.

PAGE 68

59 Table 3-3. Mean amounts of feeding by adult D. balteata on leaves of resistant Valmaine (Val) and susceptible Tall Guzmaine (TG) lettuce with and without surface layer removal by methanol Choice TG-TG+S Cultivar TG TG+S N Leaf area eaten (mm^ ± SEM) P > |T| 12 12 254.7 ±29.3 317.7 ±26.4 0.0696 Val-Val+S Val Val+S 12 12 54.5 ± 11.1 95.0 ±27.6 0.2111 TG-Val TG Val 12 12 310.9 ±38.7 22.1 ±5.7 0.0001 TG+S-Val+S TG+S Val+S 12 12 372.0 ±35.5 40.3 ± 9.4 0.0001 TG-Val+S TG Val+S 12 12 343.5 + 25.0 96.4 ±41.2 0.0001 TG+S-Val TG+S Val 12 12 401.0 ±32.3 34.7 ±9.7 0.0001 TG+S leaves from TG treated with methanol. Val+S = leaves from Val treated with methanol.

PAGE 69

Browning was observed on methylene chloride treated leaves more often than on methanol treated leaves. Feeding Response to Crude Extracts in Binary-choice Test There was no significant difference in adult feeding between control and extract treated leaves when TG leaf extracts in methylene chloride were applied to lima bean leaf surfaces, except at the natural concentration of 27 ).ig/cm^ (Fig. 3-1). At this concentration, less feeding occurred on TG extract-sprayed lima bean leaves compared to control leaves. TG extracts above 27 }ig/cm^ led to reduced feeding, but it was significant (P=0.05) only at this concentration due to large variation in feeding rates between beetle pairs. TG below its natural concentration led to increased feeding, but it was not significant. The only significant difference found between a lima bean leaf treated with Val surface extracts and a control leaf was at the level of 5 |.ig/cm^ which was lower than its natural concentration (Fig. 3-2). Spraying Val extracts at this level resulted in increased leaf area consumption. Extracts above this concentration did not result in any consistent effect on feeding on Val. Application of surface crude extracts of TG or Val to lima bean leaf surfaces at 1.5, 2, or 2.5 times their natural concentrations did not significantly stimulate or deter feeding by adult D. halteata (Fig. 3-1, 3-2), but caused sprayed lima bean leaves to be browned with greater frequency as the concentration applied increased. No significant difference in feeding was found in pair-wise testing of methylene chloride extract of TG and Val on lima bean leaves at any concentration combination (Fig. 3-3). For example, when both of the natural concentrations of Val (11 ^g/cm^) and TG (27 ).ig/cm^) were sprayed on lima bean leaf surface, the amount of feeding on the

PAGE 70

61 14 27 41 55 69 Crude leaf extracts applied (|ag/cm^) Fig. 3-1. Mean leaf area consumption by adult D. halteata exposed to lima bean leaves treated with methylene chloride extracts of susceptible Tall Guzmaine (TG) or control solvent (CK) in a binary-choice test. Vertical lines indicate 1 SEM. Asterisk = significant difference between TG and control (CK) at the 0.05 level by paired /-test.

PAGE 71

250 ^Val 5 11 16 22 27 Cmde leaf extracts applied ((ig/cm^) . 3-2. Mean leaf area consumption by adult D. balteata exposed to lima bean leaves with methylene chloride extracts of resistant Valmaine (Val) or control solvent (CK) in a binary-choice test. Vertical lines indicated 1 SEM. Asterisk = significant difference between Val and Ck at the 0.05 level by paired /-test.

PAGE 72

300 ^TG Val 11-27 11-11 19-19 27-27 27-11 Pair-wise combinations of concentration Fig. 3-3. Mean leaf area consumed by adult D. balteata on lima bean leaves treated with methylene chloride extracts from either susceptible Tall Guzmaine (TG) or from resistant Valmaine (Val) in binary-choice tests. Vertical lines indicate 1 SEM.

PAGE 73

64 lima bean leaf with TG extracts was 1 72.6 mm^, which was very close to the amount on the leaf with Val extracts (176.7 mm^). Feeding Response to Crude Extracts in No-choice Test No significant difference in feeding damage was observed between lima bean leaves applied with extracts from TG or Val in no-choice tests (F=0.36; df=l, 150; P=0.5471) (Fig. 3-4). However, extract concentration had a significant effect on adult feeding (F=3.30; df=4,150; P=0.0126). Adult feeding decreased linearly when methylene chloride leaf rinses were applied in increasing concentrations to lima bean leaf surfaces both for TG (F=6.78; df-1,150; P=0.0165) and for Val (F=5.73; df=l,150; P=0.0123). No significant interaction was found between concentration and cultivar (F=0.46; df=4,150; P=0.7671). Discussion The solvents, hexane and methylene chloride, used in this study were reported to be efficient for extraction of leaf surface chemicals of lettuce (Bakker et al. 1998). They used chloroform, methylene chloride, toluene, hexane, and a chloroform and methanol mixture to extract the leaf surface of L. saliva, and found that the distribution of wax compounds was very similar for the extracts made with various solvents, except for the chloroform and methanol mixture. Because the mixture of chloroform and methanol extracted chlorophyll and some internal cytoplasmic lipids, such as membrane sterols, they concluded that methanol or a combination of methanol with other solvents was not suitable for the extraction of leaf cuticular waxes of lettuce. However in this study, leaves treated with methanol had the least amount of browning and wilting compared to

PAGE 74

65 ^TG Val Crude leaf extracts applied (ug/cm^) Fig. 3-4. Mean leaf area consumed by adult D. halteata on lima bean leaves treated with methylene chloride extracts of susceptible Tall Guzmaine (TG) or resistant Valmaine (Val) in no-choice tests. Vertical lines indicate 1 SEM.

PAGE 75

66 the other two solvents used. It seems that methanol alone may be another suitable solvent to extract polar leaf surface chemicals from lettuce leaves. That leaves of lettuce with their surface chemicals removed were consumed much more, especially with the resistant Val, and that the total surface chemical removed by three different solvents was significantly different between resistant and susceptible cultivars, lead me to promote a hypothesis that leaf surface chemicals may have a role in expression of lettuce resistance. However, leaf surface extracts from Val were not deterrent to adult D. balteata when applied to lima bean leaf surfaces at various concentrations in binary-choice and no-choice tests. Leaf surface extracts from TG applied at increasing concentrations did not cause aduhs to feed more on lima bean leaves but actually decreased feeding similar to extracts of Val (Fig. 3-6). Therefore, these results suggest that leaf surface chemicals in lettuce do not explain resistance of Val to adult D. balteata. Reduced feeding with increased extract concentrations may be due to the increased chance of solvent browning on the leaf surface of lima bean, which resulted in the leaf losing moisture quickly and affecting adult consumption. High concentrations of extracts reduced solvent evaporation, causing solvent to remain in contact with the leaf surface longer than with lower concentration extracts. One possible reason to explain the increased palatability of solvent-extracted leaves is that removal of surface chemicals with solvents disrupts the leaf surface, causing the leaves to lose water and internal pressure easily. Leaves with reduced internal pressure may be unable to deploy a physical response to beetle feeding, such as emission of latex or other vascularly associated chemicals. Methylene chloride wilted

PAGE 76

67 lettuce leaves more easily than methanol, which may cause adults to feed more on methylene chloride-dipped leaves. Another possible reason for increasing palatability of solvent-treated leaves is that disruption of cellular membranes by solvents may cause some plant secondary chemicals to come together with their specific degradation enzymes, resulting in the breakdown of deterrent compounds in solvent-dipped leaves. It is well known that many plant secondary chemicals against herbivore attack are also self-toxic (Schoonhoven et al. 1998). In order to prevent autotoxicity, plants usually store less toxic precursors of these highly toxic compounds or store toxic chemicals in cell compartments which are remote from metabolism (i.e. cell walls and vacuoles; Schoonhoven et al. 1998). Leaf browning occurred after leaves of TG and Val were dipped in methylene chloride for less than 5 seconds. Several oxidizing enzymes, such as polyphenol oxidase or peroxidases, are responsible for tissue browning by oxidation of phenolics to quinones at the surface of wounded plant tissue where cell compartmentation breaks down (Edwards and Wratten 1983, Dowd 1994). Therefore, it is possible that some secondary chemicals, possibly phenolics, are broken down by oxidative enzymes, resulting in increased feeding by D. balteata on leaves with surface removal by methylene chloride, although in some cases, oxidation of phenolics to quinones leads to greater deterrence of insect feeding (Beck and Reese 1976, Dowd 1994). A similar hypothesis was also proposed by other researchers for leaves that were dipped in non-polar solvents, such as chloroform or hexane (Keathley et al. 1999, de Boer and Hanson 1988). Keathley et al. (1999) found that chloroform dipping rendered leaves of Bradford callery pear palatable to the Japanese beetle, Popilliajaponica Newman, but extracts containing surface waxes were not

PAGE 77

68 deterrent. Dewaxed leaves of resistant Carina generales L. were more susceptible to the tobacco homworm, Manduca sexta (L.), but reapplying leaf surface chemicals to the dewaxed leaves did not restore resistance (de Boer and Hanson 1988). It is also possible that the solvent used for leaf surface extraction destroyed chemical deterrents or deterrent physical characteristics of Val since reapplication of leaf surface chemicals on the lima bean leaves did not restore the resistance. However, leaves from Val treated with either methylene chloride or methanol still had strong resistance to feeding damage by D. balteata. Choice and no-choice experiments in which the leafsurface extracts of Val were put onto the lima bean leaf surface did not render the acceptable lima bean leaves unpalatable. Apparently, interior chemicals or other unidentified factors play a major role in resistance of Val. This possibility is investigated in Chapters 4 and 5.

PAGE 78

CHAPTER 4 INVESTIGATION OF THE PHYTOCHEMICAL NATURE OF RESISTANCE IN LETTUCE TO DIABROTICA BALTEATA USING ARTIFICIAL DIET Introduction Plants are the richest source of organic chemicals on Earth, producing an extraordinary array of secondary metabohtes to protect themselves at different phases of growth (Schoonhoven et al. 1998). Secondary plant metabolites are not universally found in higher plants, but are restricted to certain plant taxa, or occur in certain plant taxa at much higher concentrations than in others, and are usually of little or no nutritional significance to insects (Schoonhoven 1972, Slansky 1992). The vital role of secondary plant metabolites in the defense system against insects and other enemies has been recognized as inhibition of food intake in the majority of plant-feeding insects by acting as feeding inhibitors (deterrents) and toxins (Panda and Khush 1995, Fraenkel 1959). A few specialized species, however, exploit these chemicals as stimulants (Fraenkel 1959). It has been estimated that the total number of secondary plant metabolites may reach 100,000 or more, but only about 10,000 of them have been chemically defined (Schoonhoven 1982). Much of the research on plant-insect interactions has centered on the identification of chemicals important in insect feeding by either applying plant chemicals to a plant or to an artificial substrate (Miller and Miller 1986). As a prerequisite to identifying the chemicals involved, it is essential to develop a reliable, quantitative and time-efficient bioassay for the response of insects to feeding stimulants and deterrents. 69

PAGE 79

70 However, the screening of plant extracts for feeding deterrents and the isolation and purification of bioactive compounds are often very long, expensive, and tedious processes. Insect feeding bioassays especially require large numbers of insects to monitor all fractionation and purification steps, and only a small percentage of that effort produces a significant output because active substances can comprise only a few out of a large number of constituents (Escoubas et al. 1992). Several substrates, some natural and others artificial, have been used for testing feeding inhibitory activity of a chemical or chemical mixture (Miller and Miller 1986). For example, test chemicals can either be applied to natural food, such as whole plants (Erikson and Feeny, 1974), leaves (Lewis and van Emden 1986), leaf disks (Hsiao and Fraenkel 1968) or mixed with an insect's natural dried food (Bemays and Chapman 1977). Test chemicals can also be added to artificial substrates, such as sucrose-flavored agar (Alfaro et al. 1979), agar-cellulose (Ma 1972), filter paper, glass fiber disks (Blaney and Winstanley 1980) and artificial diet (Escoubas et al. 1992, Alonso-Amelot et al. 1994). Artificial diets of known chemical composition have been shown to be an indispensable tool for studying the behavioral responses of insects to specific plant compounds or the nutritional role of certain plant components (Schoonhoven et al. 1998). An artificial diet has been reported to support successfully the adult stage of the banded cucumber beetle, Diabrotica balteata (Creighton and Cuthbert 1968). Furthermore, a commercial artificial diet for adult D. balteata is available. Studies in Chapter 2 and 3 indicated that the lettuce cultivar Valmaine was the most resistant cultivar against adult D. balteata feeding, and Tall Guzmaine was the most susceptible. Studies of the leaf surface chemicals indicated that surface wax on lettuce leaves was not responsible for

PAGE 80

71 resistance in Valmaine. Experiments in this chapter were designed to determine whether plant chemicals inside lettuce leaves act as feeding deterrents to adult D. balteata when incorporated into an artificial diet. Experiments were also designed to determine whether the adults would exhibit a response varying with the concentration of leaf powder, and to determine whether lettuce leaf powder can be used as the starting material for the isolation of the compounds responsible for resistance. The quantitative effect of leaf powder on feeding responses was determined by measuring diet consumption and female mass in a binary-choice situation, and reproduction and adult mortality in a no-choice situation. Materials and Methods Plants and Insects Two lettuce cultivars, Valmaine (Val, resistant) and Tall Guzmaine (TG, susceptible) were used to search for possible resistance factors in the leaves. Lettuce was planted as described in Chapter 2. Adult D. balteata for feeding bioassays were obtained from a laboratory culture originally collected from the field in Belle Glade, Florida in June 1996. Adults and larvae were reared in the same manner as described in Chapter 2. Preparation of Leaf Powder Four to five fully expanded middle lettuce leaves were excised from plants of each cultivar when they had eight to nine fully expanded leaves. Excised leaves were immediately put into trays (30 x 43 cm) and transferred to a large freezer (-40''C) for 1 d. The trays with frozen leaves were put into a freeze drier (model 25SRC, The Virtis Company, Gardiner, NY) for 4 d. Freeze dried leaves were immediately put into plastic

PAGE 81

72 bags in a cooler containing dry ice, transported to the lab and ground in a Wiley mill (model 3383-LlO, Thomas Scientific, Mexico) to a uniform powder (40 mesh). Leaf powder was stored in small airtight glass containers (4.4 cm diam., 4.5 cm high) in a freezer (-70°C) until used. Artificial Diet The components of artificial diet were ordered from a commercial company (BioServ, Frenchtown, NJ). 100 ml normal diet was prepared as follows: 1.74 g agar was boiled in 100 ml deionized water on a hot plate (model SPA1025B, Bamstead/Thermolyne, Dubuque, lA). KOH solution (1 ml) and diet dry mix (14.91 g) were added into the agar after it had cooled below 40°C, and thoroughly mixed to avoid the formation of lumps. The diet was poured into a glass petri dish (14 cm diam.), covered with the glass cover, wrapped completely in aluminum foil, and stored in a refrigerator (4-6°C). The test diet containing lettuce leaf powder was made by adding fixed amounts of the leaf powder into the normal diet by replacement of diet dry mix with the same amount of leaf powder. Three treatment diets for each cultivar were made containing 10, 20, or 30% leaf powder incorporated into the normal diet. The diet dry mix and leaf powder were thoroughly mixed before they were added to the agar. All diets were prepared 2 to 3 h before the following bioassays. Binary-choice Test In order to test whether beetles still preferred TG to Val when leaf powder made from both cultivars was incorporated into an artificial diet, binary-choice tests were performed. Diet disks (3-4 mm thick) were made using a No. 13 cork borer (ca. 2.1 cm diam.) to punch out plugs from the cooled diet in petri dishes. The bottom and top of the

PAGE 82

73 disk were covered with a circle of plastic, whose diameter was a little bit larger than that of the diet disk, in order to slow evaporation rate and keep diet free of frass. Only the edge of the diet disk was exposed to beetle feeding. Each experimental unit consisted of six disks, three from a diet with TG leaf ' powder and three from a diet with Val leaf powder at the same concentration, arranged alternately in a circle in a plastic container (27.5 x 20 x 10 cm). Twenty-five unsexed and unfed adults emerged within 48 h were released into this container and were allowed to feed on the diet for 3 d. Ten replicates were performed for each experimental unit. The experiments were carried out in an incubator with a photoperiod of 14:10 (L:D) at 25 ± rCand 70 ± 10% R. H. Moisture calculation of the fresh diet (diet moisture). When the test diet disks were punched out from a diet, ten diet disks were weighed individually (disk fresh wt) before they were put into an oven at 50 ± S^'C. After 2 d, these disks were reweighed individually (disk dry wt). The moisture content of each disk was then calculated as follows. Disk moisture = (disk fresh wt-disk dry wt)/disk fresh wt. The moisture content of a diet was the average moisture content of 10 disks. Dry weight of a diet disk consumed (dry wt consumed). The test diet disk was weighed (test disk fresh wt) before it was given to the adults. After 3 d, the diet disk was put mto the oven at 50 ± 5°C for 2 d and was weighed again to obtain its remaining dry weight (test disk dry wt afterwards). The dry weight of the diet consumed was calculated as follows. Dry wt given to beetles = test disk fresh wt x (1 diet moisture) Dry wt consumed = dry wt given to beetles test disk dry wt afterwards

PAGE 83

74 The total dry weights of TG and Val diets consumed in 3 d were calculated by adding up the dry weights consumed of all the three disks from the same diet in each experimental unit. Dry diet consumed was analyzed as a split plot design with random blocks replicated through time. The three levels of leaf powder were assigned at random to whole plots within each block. Each whole plot was divided into two subplots to which TG and Val diets were assigned at random (SAS Institute 1999). Post-hoc means separation was performed using Tukey's HSD test (a=0.05, SAS histitute 1999). No-choice Test Two disks from the same diet were presented to five pairs of adult D. balteata on a plastic cover of a plastic container placed upside down (10.5 cm diam. upper, 7.2 cm diam. bottom, 17.5 cm high) in no-choice tests. Adults used in the trial were unfed and had emerged within 48 h of the trial. Diet disks with TG leaf powder at 10, 20 and 30% concentrations, and diet with Val leaf powder at 10, 20, and 30% concentrations were tested. Diet disks from full strength Diabrotica diets (normal diet) were also tested and considered as 0% leaf powder. The experiments were carried out in an insect rearing room with a photoperiod of 14:10 (L:D) at 25 ± TC. Before releasing these adults, females were marked on their elytra with blue, green, pink, yellow or ivory paper correction fluid and weighed separately. After 13 d, marked surviving females were collected and reweighed so that their weight gain could be calculated. These females were stored in a refrigerator freezer (-9''C) until they were dissected to check their egg production. The Chi-square test was used to analyze female, male and total mortality of D. balteata on three different diets for each cultivar and on full strength Diabrotica diets. Numbers of eggs produced were analyzed as an 8 x 2 x 4 three factorial design using

PAGE 84

75 Proc GLM (SAS Institute, 1 999) in which the replicate (repHcating through time) had eight levels, cultivars had two levels, and leaf powder concentration had four levels. Means were separated using Tukey's HSD test with a significance level of a=0.05 (SAS Institue 1999). Female final weight and weight gain after a 13-d feeding period were also analyzed as a three factorial design by Proc GLM (SAS Institue 1999). Female initial weight was used as a covariate when analyzing female final weight. Results Binary-choice Test Significant differences in diet consumption were found among the concentrations of leaf powder added to the artificial diet (F=47.9; df=2, 19; P=0.0001) (Fig. 4-1). TO diet consumption increased from 11 8.5 mg to 172.0 mg with an increase of leaf powder concentration from 10% to 30%. Similarly, beetle feeding on Val diets increased from 104.5 mg to 166.7 mg when the concentration of leaf powder increased fi-om 10% to 30%. No significant difference in feeding was found between TG diets and Val diets at any concentrafion (F=1.8717; df=l, 27; P=0.1826). Therefore, data of leaf area consumption on TG diets and Val diets at the same concentration were pooled to study its relationship with leaf powder concentrations. The regression analysis showed that diet consumpdon increased linearly with leaf powder concentration (R^= 0.99) (Fig. 4-2). The replications through time (time factor) had a significant effect on the amount of diet consumed by adult D. halteata (F=17.23; df=9, 18; P=0.0001). Interactions between the time and concentration (F=0.7633; df=18, 27; P=0.7208) and between concentration and cultivar (F=0.8372; df=2, 27; P=0.4438) were not found to be significant.

PAGE 85

76 200 Leaf powder concentration Fig. 4-1 . Dry weight consumed by adult D. balteata feeding on diets with lettuce leaf powder of susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) lettuce incorporated at different concentrations for 3 d in binary-choice tests. Vertical lines indicate 1 SEM.

PAGE 86

1 Of\ 180 160 ' Y = 2.89X + 81.16, = 0.99 140 / .^^^ > 120 100 80 1 1 1 T 0 5 10 15 20 25 30 35 Leaf powder concentration (%) Fig. 4-2. Relationship between leaf powder concentration and diet consumption by adult D. balteata for 3 d in binary-choice tests. Vertical lines indicate ± 1 SEM.

PAGE 87

78 No-choice Test Number of mature eggs produced by females feeding on TG diets for 1 3 d did not differ significantly from that of females feeding on Val diets (F=0.5883; df=l, 9; P=0.4634) (Fig. 4-3). Females feeding on TG diets with different concentrations of leaf powder had 65 to 75 mature eggs while females on Val diets had 63 to 70 mature eggs. Moreover, the increasing concentrations of leaf powder added did not significantly affect egg production of females, regardless of cultivars (F=1.8272; df=3, 24; P=0.1691). No replication effect was observed on egg production of females feeding on diets with different levels of leaf powder from TG and Val (F=2.0708; df-7, 12; P=0.1289). No significant effect of the interaction among replication, cultivar and concentration on female reproduction was found (F=0.5179; df=21, 125; P=0.9587). Neither were twofactor interactions between replication and cultivar (F=0.9350; df=7, 26; P=0.4967), nor between cultivar and concentration (F=0.8141; df=3,28; P=0.4969) significant. However, the interaction between time and concentration was significant (F=2.2539; df=21,21;P=0.0347). The percentage of females with mature eggs on TG diets was not significantly different from that on Val diets at 0, 10, 20 and 30% levels (Table 4-1). Eightyfive or ninety percent of females feeding on the diets without any TG (0% TG) or Val (0% Val) leaf powder added had mature eggs after 13 d. This was not significantly higher than any other diet with either TG leaf powder or Val powder. There was no significant difference between cultivars in female survivorship at 0, 10, 20 and 30% leaf powder (Table 4-1). Similarly, male survivorship was not

PAGE 88

90 85 80 75 70 65 -] 60 55 H 50 45 40 0 10 20 Leaf powder concentration (%) 30 Fig. 4-3. Average number of mature eggs produced per female after feeding for 13 d on diets containing leaf powder of susceptible Tall Guzmaine (TG) or resistant Valmaine (Val) lettuce at different concentrations. Vertical lines indicated 1 SEM.

PAGE 89

80 Table 4-1 . Results of no-choice tests in which adult D. balteata were allowed to feed on diet with different concentrations of leaf powder from susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) lettuce for 13 d Category Cultivar Leaf powder concentration (Mean ± SEM) 0 10 20 30 TG 90.2 ± 4.9 86.9 ±5.4 79.417.3 72.1 19.7 Percentage of females with mature eggs Val 85.216.0 80.6 ±5.5 61.3 17.4 75.618.8 1.902 0.893 4.075 5.679 P J. V/. yyj^ n 771 U. / / I A <7C U.J /O TG 70.0 ± 8.5 70.0 ±8.5 70.0 1 6.6 70.018.5 Female survivorship Val 80.0 ±7.5 80.0 ±7.6 75.017.3 67.5 1 10.0 1.038 1.537 3.722 3.651 p 0.99 0 981 0 811 u.o 1 y TG 87.5 ±7.5 85.016.3 70.01 10.0 77.515.9 Male survivorship Val 90.0 ± 5.3 87.5 ±5.3 80.018.5 87.516.5 1.772 0.573 4.094 1.485 P 0.971 0.999 0.769 0.983 TG 80.0 ±6.0 77.514.9 72.5 14.9 72.515.9 Total survivorship Val 85.0 ±4.6 85.014.2 77.5 15.6 77.516.7 1.955 0.911 2.805 2.156 P 0.962 0.996 0.902 0.951

PAGE 90

81 significantly different between the cultivars at 0, 10, 20 and 30 % leaf powder (Table 41). Finally, total survivorship was similar for the two cultivars at 0, 10, 20, and 30% leaf powder (Table 4-1). Female final weight and weight gain during the experiment were significantly affected by rephcafion through fime (weight: F=3.0026; df=7, 18; P=0.0285) (weight gain: F=5.8003; df=7, 18; P=0.03) and by cultivar (weight: F=12.0303; df=l,10; P=0.006) (weight gain: F=l 1.8162; df=l, 10; P=0.0062) (Table 4-2). Females fed Val leaf powder diets gained significantly more weight than those fed TG leaf powder diets. No significant differences in pre-experiment weights were found among females assigned to the diets with leaf powder fi-om different cultivars (F=0.000468; df=l, 9; P=0.9832) or to the diets with different concentrations (F= 1.7745; df=3, 25; P=0.1781). On TG leaf powder diet, final weights and weight gains decreased slightly (16.3 to 14.3 mg and 7.4 to 6.3, respectively) as concentration increased. Similar results were recorded for beetles on Val leaf powder diet where final weights decreased from 16.9 to 15.3 mg and weight gain decreased from 8.5 to 7.2 mg. However, changes with increasing leaf powder concentration were not significant for final weight (F=2.1402; df=3,23; P=0.1226) or weight gain (F=2.1357; df=3,23; P=0.1234). Two-factor interactions were not significant between rephcation and cultivar for weight (F=0.8416; df=7, 24; P=0.5641) or for weight gain (F=0.8461 ; df=7, 24; P=0.5608). The interactions between cultivar and concentration for weight (F=1.6181; df=3, 32; P=0.2048) or for weight gain (F=1.5932; df=3, 32; P=0.2096) were also not significant. However, a significant interaction was found between rephcafion and concentration for weight (F=5.1466; df=21, 21; P=0.0002) and for weight gain (F=5.1925; df=21, 21; P=0.0002). Further analysis on concentration

PAGE 91

82 Table 4-2. Average female D. halteata weight gain and final weight after feeding on diets with different concentrations of leaf powder from susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) lettuce for 13 d in a no-choice situation Diet cone. (%) Weight TG (mg + SEM) Val (mg ± SEM) 0 Final weight 16.26 ±0.59 16.89 ±0.50 Weight gain 7.41 ±0.52 8.47 ± 0.42 10 Final weight 16.11 ±0.50 16.84 ±0.50 Weight gain 7.66 ± 0.49 8.39 ±0.42 20 Final weight 15.18 ±0.43 15.95 ±0.54 Weight gain 7.10 ±0.37 7.87 ± 0.46 30 Final weight 14.34 ±0.50 15.26 ±0.50 Weight gain 6.34 ± 0.45 7.21 ±0.44

PAGE 92

83 by fixing replication through time factor showed a significant difference in final weight at the last replicate (F=6.88; df=3,24; P=0.001 7). Significant difference in weight gain was also found at time 8 (F=5.57, df=3,24; P=0.0039). The three-factor interaction (replicate X cultivar x concentration) was not significant for beetle final weight (F=0.3785; df=21, 175; P=0.9943) or weight gain (F=0.3849; df=21, 176; P=0.9936). In conclusion, adults feeding on Val leaf powder diets produced similar amount of eggs compared to those on TG leaf powder diets. There were no significant differences in female, male and total survivorships between these two diets. Females on Val leaf powder diets gained more weight than those on TG leaf powder diets. Discussion Adult D. balteata did not show a significant preference for TG over Val in binarychoice tests when leaves were freeze-dried, ground into powder and incorporated into the artificial diet at three different concentrafions (Fig. 4-1). This result differs with what I found in Chapter 2, where fresh Val plants had significant resistance against D. balteata feeding, and in Chapter 3, where Val plants with leaf surface removed still had very strong resistance to D. balteata. Furthermore, increasing the amount of leaf powder of TG or Val added to the artificial diet resulted in increased diet consumption by D. balteata. This compensatory feeding may be due to suboptimal quality of the diet with lettuce leaf powder compared to adult's normal diet with the same amount of dry matter. Compensatory feeding responses by adjusting feeding rate to approach or realize maximal growth rate has been found in several insect species and is proposed to be general among herbivorous insects (Simpson and Simpson 1990). The increased diet consumption with increased leaf powder may be also due to the presence of feeding

PAGE 93

84 stimulants or the level of feeding stimulants enhanced somehow compared to possible deterrents present in lettuce leaf powder. That Val loses its resistance when freeze-dried and powdered was also confirmed by no-choice tests, where females produced similar numbers of mature eggs regardless of the diet on which they fed for 13 d (i.e. control, or supplemented with TG or Val powder). An artificial diet differs in many respects from natural food sources of herbivores. It may not be completely suitable for screening feeding stimulants or deterrents even though it offers uniformity, evenness of application, and ease of preparation and measurement (Lewis and van Emden 1986, Schoonhoven et al. 1998). An artificial diet usually contains feeding stimulants for designated insects, which can mask effects of feeding deterrents. It has been proposed that a change in the balance between phagostimulants and deterrents can make an unacceptable food acceptable and vice versa (Bemays and Chapman 2000). The feeding of L. migratoria on wheat flour wafers was almost totally inhibited by adding tomatine at 0.1% dry weight, but the effects of the tomatine were eliminated by adding increasing amounts of sucrose to the wafers (Bemays and Chapman 1978). Thresholds for the same deterrent may vary as much as 1000 times between natural and artificial substrates, resulting in differences in the effective concentrations of the chemicals presented to the insect (Schoonhoven 1982). Even small changes in diet composition can have drastic effects on insect performance, such as growth and reproduction (Schoonhoven et al. 1998). Moreover, the same chemicals may have neutral, phagostimulant, or deterrent effects at different concentrations depending upon existing background compounds, especially phagostimulants (Lewis and van Emden 1986). Therefore, the artificial diet used in this study may mask deterrents from

PAGE 94

85 Val because of the difference in balance of feeding stimulants and deterrents between the artificial diet and lettuce plants. It is also possible that feeding deterrents in Val may be broken down during the process of making lettuce leaf powder or during the storage period, even though it was stored at -70°C. Freeze-drying under high-vacuum conditions has been suggested to be inappropriate for lowand intermediate-molecular weight phenolics because many of the volatile phenolics are concentrated on the ice trap (van Sumere et al. 1983). Orians (1995) compared three methods of preservation, i.e. air-drying, freeze-drying, and vacuum-drying and found that freeze-drying was not appropriate for analysis of phenolic glycosides because of the degradation of these compounds. Air-drying was appropriate for phenolic glycosides but resulted in loss of tannins. Only vaccum-drying preserved both phenolic glycosides and tannins for analysis. Chlorogenic acid instead of isochlorogenic acid was detected in lettuce roots that had been vacuum-dried prior to extraction, and no caffeoylquinic acids were detected in root material, either stored in a deep-freezer at -20°C before extraction or freeze dried (Cole 1984). The reason for the presence of chlorogenic acid in dried root material and the loss of all caffeoylquinic acid after freezing or freeze-drying may be related to cell damage and the formation of insoluble quinones from the caffeoylquinic acids. Therefore, the treatment of plant material before chemical analysis is very important. Other possible reasons for the loss of resistance in Val are that feeding deterrents are active only in fresh leaves attached on the plants and that any role that a physical barrier plays in resistance may act only when leaves are turgid. The chemicals in the leaf may be activated and released only after insect feeding. It is well known that some plants

PAGE 95

86 can qualitatively or quantitatively enhance their defense against herbivory after they are attacked by pests or by other inducers (Tallamy and Raupp 1991, Karban and Baldwin 1997). Many studies show that wound-induced resistance by herbivores is related to the induction of enzymes activities and the release or synthesis of new secondary metabolites upon injury (Bi et al. 1994, Bi et al. 1997, Baldwin 1991, Louda and Mole 1991). Some physical barriers to host use by the insect, such as latex emission, may become ineffective when leaves are dried and powdered. The possible role of latex and inducible resistance in resistance of Valmaine lettuce to D. balteata will be studied in Chapter 5.

PAGE 96

CHAPTER 5 POSSIBLE MECHANISMS INVOLVED IN LETTUCE RESISTANCE TO DIABROTICA BALTEATA Introduction Plant defense mechanisms against herbivores have been extensively studied over the past decade because of their potential application in integrated pest management programs, and in providing new leads for environmentally safer pesticides (Balandrin et al., 1985, Panda and Khush 1995). Understanding the mechanisms of resistance is also very important before the degree of resistance among plants can be ascertained (Panda and Khush 1995). Factors that determine host plant resistance to herbivores include the presence of morphological, physical, or structural barriers, allelochemicals, and nutritional imbalance (Panda and Khush 1 995). These factors can be active whether or not herbivores are present. Host plant resistance can also be inducible after plant tissues are attacked by herbivores, resulting in induction of a wide array of chemical defenses against subsequent herbivory (Panda and Khush 1995). Latex in some laticiferous plants has been reported as a natural defense system against certain herbivores. In many laticiferous plants, including Lactuca saliva L., latex is stored underpressure within laticifers, which results in rapid release of latex upon cutting (Fahn 1979, Data et al. 1996, Dussourd 1995). The secretions often contain secondary metabolites known to be toxic or deterrent to animals (Farrell et al. 1991). These secretions also polymerize upon exposure to air, thus threatening small herbivores with the risk of entrapment in hardening exudates or gumming up their mouthparts 87

PAGE 97

88 (Dussourd 1995, 1997). Data et al. (1996) found that young vine material of sweet potato produced more latex and had fewer sweetpotato weevils, Cylas formicarius (F.), than older and more mature portions of the vine. Feeding and oviposition were reduced when latex was applied to the surface of root cores. Several insects have been observed immobilized in exudates, such as caterpillars (Dussourd 1993), ants (Dillon et al. 1983), aphids and whiteflies (Dussourd 1995). Because of the rupture of latex canals, two aphid species, Uroleucon pseudambrosiae (Olive) and Rhopalosiphum maidis (Fitch), were physically glued to lettuce plants, primarily by their legs, after they alighted on inflorescences. Adult whitefly, Bemisia tabaci (Gennadius), likely became trapped in latex, primarily on the flower buds (Dussourd 1995). In many crop plants, resistance can be induced through prior wounding by insects or mechanical means (Karban and Baldwin 1 997). Induced resistance has been correlated with increased activities of oxidative enzymes (Bi et al. 1994), enhanced synthesis of secondary metabolites (Bi et al. 1997), or synthesis of primary gene products (Green and Ryan 1972). These induced resistance phenomena may involve decreases in both insect feeding preference and the nutritional value of induced foliage (Schoonhoven et al. 1998). Larval growth of Helicoverpa zea Boddie was decreased on previously damaged foliage or squares of cotton compared to growth on undamaged tissue because of a significant decline in host nutritional quality and increased activities of oxidative enzymes after herbivory (Bi et al. 1994). Second-instar larvae of the gypsy moth, Lymantria dispar L., preferred leaf disks from non-induced trees over those from induced trees (Havill and Raffa 1999). Inducible resistance in plants can also affect the ovipositional responses of adult females (Stout and Duffey 1996). Induced resistance

PAGE 98

may be expressed either on rapid (hours to days), short-term (days to weeks) time scales, or long-term (years) time scales (Schoonhoven et al. 1998) through de novo synthesis of formerly absent compounds or an elevation of the normal level of resistance compounds. The short-term reaction may spread systemically over the whole plant, or it may be localized to the damaged leaf (Schoonhoven et al. 1998). Localized accumulation of a sesquiterpenoid lactone phytoalexin, phenolics and phenolic esters have been reported in lettuce after being attacked by a pathogen (Bennett et al. 1994, 1996). In this chapter, the resistance of lettuce to D. balteata was further evaluated. Three screening methods were conducted to test the effect of methodology on the expression of resistance: leaf disks, intact leaves attached to plants and excised leaves. Leaf area consumed by beetles was evaluated in binary-choice and no-choice bioassays to compare results for both leaf disks and detached leaves against intact leaves. Other objectives of this chapter were to compare the physical properties of latex, investigate inducible resistance as a factor in lettuce resistance, and propose possible mechanisms involved in lettuce resistance to D. balteata. Materials and Methods Plants and Insects Two lettuce cultivars, Valmaine (Val, resistant) and Tall Guzmaine (TG, susceptible) were used for this experiment. Lettuce was planted in the same manner as described in Chapter 2. Fordhook 242 lima bean seeds were purchased from a commercial company (Illinois Foundation seeds Inc., IL) and planted in the same way as described in Chapter 2. Only the first true leaves were used for the latex smearing experiment.

PAGE 99

90 Adult D. balteata for feeding bioassays were obtained from a laboratory culture originally collected from the field in Belle Glade, Florida in June 1996. Adults and larvae were reared as described in Chapter 2. Only unfed adults which had emerged within 48 h were used for the following assays. Screening Methods Leaf disks. All leaves used in this assay were the seventh fully expanded leaf, counting from the first true leaf Selected plants had six to eight fully expanded leaves. Two disks of 380 mm^ were punched out from non-midrib areas of freshly excised leaves of Val or TG plants using a No. 15 cork-borer. A test unit consisted of a plastic petri dish (8.9 cm diam.) inside of which four leaf disks (two from Val and two from TG) were presented at an equal distance on two layers of moistened paper towel. One pair of adults was placed in each petri dish and allowed to feed for two days. The tests were conducted in a rearing room at 25 ± 14:10 (L:D) photoperiod. Each test unit was replicated 15 times. To compare the resistance level of leaf disks with attached lettuce leaves, binarychoice tests were conducted simultaneously between intact leaves of TG and Val using the binary-choice feeding arenas as described in Chapter 2. The extent of feeding for both screening methods was evaluated by measuring the area of the leaf remaining by scanning the leaf (JADE 2, Linotype-Hell, Taiwan) and importing the scanned images into an imaging program (ImagePC beta version 1 , Scion Corporation, Frederick, Maryland) where remaining leaf area was determined. For the leaf disks method, the remaining leaf area was then subtracted from the mean disk area of 10 disks not offered to beetles for two days in order to account for shrinkage during the

PAGE 100

91 assay. The difference in leaf area consumption between cultivars was analyzed by paired r-test using Proc MEANS (SAS Institute 1999). Detached leaves. Resistance to beetle feeding was next compared between detached and intact leaves. For the detached leaf tests, the seventh fully expanded leaf was excised from each plant and its petiole immediately immersed in a beaker filled with tap water. Binary-choice feeding arenas were used to expose single pairs of adults to both TG and Val detached leaves for 48 h. At the same time, binary-choice tests were carried out between two leaves still attached to TG and Val plants by using the same feeding arena. Each test unit was replicated 1 7 times in each bioassay. The difference in leaf area consumption between cultivars was measured as above and analyzed by paired Mest using Proc MEANS (SAS histitute 1999). Since a significant difference was found in leaf area consumption between the two cultivars in binary-choice tests, the resistance level of detached leaves was further evaluated by using no-choice bioassays as follows. The seventh fully expanded leaf was excised from each plant, and its base was immediately put into a beaker with tap water. One pair of beetles was confined on the leaf by a no-choice feeding arena as described in Chapter 2. The adults were allowed to feed for 48 h. At the same time, a pair of adults was confined in the no-choice arena to the seventh leaf still attached to a plant. This study was arranged as a randomized complete block design with detached and intact leaves from each cuUivar in each block. Each block was replicated 18 times. Leaf area consumed was measured as described above and was analyzed by Proc GLM (SAS Institute 1999). Means with significant ANOVA were separated using Tukey's HSD test with a significance level of a=0.05 (SAS Insfitute 1999).

PAGE 101

92 Latex Physical Properties Latex browning test. One leaf (the seventh or eighth leaf) per plant was chosen for this test. Leaves used were matched for size and position within the plant, and were still attached to the plants. Small insect pins were used to pierce the leaf surface several times until latex flowed out from the holes. Time from latex exudation to latex browning was measured in seconds. Measurements were replicated 20 times for each cultivar. Data were analyzed as a randomized complete design by Proc GLM (SAS Institute 1999). Latex weight and latex flow cessation. A whole plant was cut off at the base of its stem. Fresh latex was collected continuously onto a preweighed filter paper until latex stopped flowing. The filter paper was reweighed and then placed in an oven at 55°C for 24 h to determine latex dry weight. In addition, time between the beginning of the cut and the end of latex flow was also measured. Measurements were replicated 28 times for each cultivar. Latex weight and time to flow cessation were analyzed as randomized complete designs by Proc GLM (SAS Institute 1999). Latex Smearing Test Plants used for this study were at the six to eight fully expanded leaves stage and were cut off at the base of their stem to allow latex to ooze out. Latex was collected from the stem of a plant onto a knife blade, and then immediately smeared into one of two confined areas on a lima bean leaf by a binary-choice feeding arena as described in Chapter 2. The other confined area was either untreated or treated with latex from the other cultivar so that three treatment combinations were run (latex from Val vs. latex from TO, latex from Val vs. control, latex from TO vs. control). Latex from one plant was spread on one confined leaf area of a lima bean leaf and the amount spread was not

PAGE 102

93 quantified (control vs. control). Each treatment combination was replicated fourteen times. Two confined areas on a lima bean leaf were left free of latex to record the total amount of leaf consumption per pair of adults in a control situation. One pair of adults was released into the feeding arena and allowed to feed for 48 h after the latex on the lima bean leaf had dried. Leaf area consumed was measured as described in Screening Methods and analyzed by paired f-test using Proc MEANS (SAS Institute 1999). Total leaf area consumed per pair of adults for 48 h was calculated by adding the area of the two exposed areas in each feeding arena and analyzing as a randomized complete block design by Proc GLM (SAS Insfitute 1999). Means with significant ANOVA were separated using Tukey's HSD test with a significant level of a=0.05 (SAS Institute 1999). Observafion of Feeding Behavior A brief behavioral observation was made of beetle feeding on TG plants. Four to five beetles were confined on a fully expanded leaf still attached to the plant by a nochoice feeding arena as described in Chapter 2. Beetle behavior was observed under a dissecting scope. Inducible Resistance Localized induction. Plants used in this bioassay had six to eight fully developed leaves. Leaves used for assays were matched for position within the plant and size. The same feeding arenas used in the binary-choice tests were used for prior wounding of a leaf, but there was only one hole clipped on the leaf so that it represented a no-choice feeding arena. One pair of unfed adults was caged on a leaf in this feeding arena. Control leaves were treated the same way but no adults were present. This pair of adults

PAGE 103

94 was then removed from the plants after 48 h. Previously damaged leaves of Val or of TG were considered as the treatment, and non-damaged leaves of Val or TG treated as the control for the following binary-choice assay. Previously damaged Val was paired with undamaged Val and caged together by a binary-choice feeding arena placed 1 cm below the previously caged leaf area (Fig. 5-1). Another pair of unfed adults was released into the cage immediately after the initial feeding damage was finished, and they were allowed to feed for another 48h. The same experiment was done with TG. Leaf area consumption in these two experiments was recorded as described previously. The difference in leaf area consumption between treatment and control leaves was analyzed by paired r-test using Proc MEANS (SAS Institute 1999). Systemic induction. Infliction of initial feeding damage was done the same way as described in the localized induction experiment. In the feeding bioassay, one pair of adults was given a choice between the leaf directly above the previously damaged leaf, and the corresponding leaf from the undamaged control plant. An undamaged leaf from damaged Val was paired with an undamaged leaf from undamaged Val (Fig. 5-2). The same experiment was done with TG. Leaf area consumption was recorded as described above. The difference in leaf area consumption between treatment and control leaves was analyzed by paired r-test using Proc MEANS (SAS Institute 1999). Results Screening Methods Val was strongly resistant to beetle feeding when intact or detached leaves were presented in binary-choice tests with susceptible TG (Table 5-1). However, when leaf disks were used as test materials, there was no significant difference in feeding between

PAGE 104

95 Fig. 5-1. Experimental set-up to detect localized inducible resistance. Feeding per pair of adult D. balteata for 48 h was monitored at a site 1 cm below the site previously damaged by another pair of adults for 48 h on the same leaf

PAGE 105

1 96 Binary-choice feeding arena Fig. 5-2. Experimental set-up to detect systemic induction. Feeding per pair of adult D. balteata for 48 h was monitored on a leaf directly above the leaf previously damaged by another pair of adults for 48 h.

PAGE 106

97 Table 5-1. Amount of lettuce leaf area consumed per pair of adult D. balteata in 48 h when presented a choice between two cultivars using different screening methods Methods Cultivar^ N Leaf area (mm^± SEM) Pr > |T| TG 17 350.5 ±28.7 Intact leaf (expt 1) 0.0001 Val 17 28.8 ±4.8 TG 17 532.2 ±57.8 Detached leaf 0.0004 Val 17 125.9 ±53.8 TG 11 318.1 ±29.3 Intact leaf (expt 2) 0.0001 Val 11 14.4 ±2.1 TG 15 382.8 ±36.3 Leaf disk 0.0941 Val 15 334.6 ±37.4 TG=Tall Guzmaine, Val=Valmaine.

PAGE 107

98 TG and Val. Adults ate almost 24 times as much Val on leaf disks as on intact leaves. Mean leaf area consumed on leaf disks and intact leaves of TG was similar, and both were over 300 mm^. A significant difference among treatments in leaf area consumed per pair of adults in no-choice tests was found (F=21.78; df=3, 51; P=0.0001) (Fig. 5-3). Resistance was again noted in Val when intact leaves were presented in no-choice tests, but excised leaves of Val did not show significant resistance against adult feeding. Feeding was significantly increased on detached leaves compared to intact leaves, irrespective of cultivar. Intact Val leaves were the least damaged by adult D. balteata with leaf area consumption of 66.3 ±9.7 mm^ which is only 12% of that on detached Val leaves. Latex Physical Properties Latex from Val plants took 473 ± 15 s on average to turn brown after it was exposed to the air, which was 60 s faster than latex from TG (Table 5-2). No significant difference in latex fresh weight was observed between TG and Val. Similarly, dry weight of TG latex was not significantly different from that of Val. However, Val flowed for 41 s longer after cutting than did TG, an increase of 27%. Latex Smearing Test Significantly less feeding damage was found on the lima bean leaf tissue spread with latex either from Val or TG (Table 5-3). Lima bean leaf smeared with latex was barely damaged by adult D. balteata, but they ate control lima bean leaf tissue 24 to 59 times more compared to leaf tissue with latex. Most of the leaf area consumed on latex treated leaves could be attributed to experimental error in spreading latex evenly on the leaves. Most damaged areas on the latex-treated lima bean leaves were found along the

PAGE 108

99 S S 3 c o D-TG D-Val I-TG I-Val Fig. 5-3. Leaf area consumed per pair of adult D. balteata for 48 h in nochoice test using detached and intact leaves from susceptible Tall Guzmaine (D-TG, I-TG) and resistant Valmaine (D-Val, I-Val). Bars topped with the same letter are not significantly different by Tukey's HSD test at the 0.05 level. Vertical lines indicate +1 SEM.

PAGE 109

100 Table 5-2. Physical properties of latex of susceptible Tall Guzmaine (TG) and resistant Valmaine (Val) Physical properties N TG Val F P value Time to latex browning (s ± SEM) 20 533 ±23 473 ± 15 8.38 0.0063 Latex fresh weight * (mg ± SEM) 28 57.6 ±2.8 58.1 ±2.2 0.02 0.8886 Latex dry weight * (mg ± SEM) 28 4.6 ± 0.5 5.0 ± 0.5 0.35 0.5551 Time to flow stopping (s ± SEM) 28 152 ±5 193 ±7 23.5 0.0001 * Amount produced from time of stem cutting until stoppage of flow.

PAGE 110

101 Table 5-3. Amount of lima bean leaf area consumed per pair of adult D. halteata in 48 h when given an untreated area or an area smeared with latex either from susceptible Tall Guzmaine (TG) or resistant Valmaine (Val) Choice Lima bean leaf N Leaf area eaten (mm^ ± SEM) Pr>|T| Control area vs. Control area 14 546.3 ±43.0 TG latex area 0.0001 TG latex area 14 23.2 ±5.9 Control area vs. Control area 14 476.1 ±44.1 Val latex area 0.0001 Val latex area 14 8.0 ±6.2 137.5 ±22.2 0.0797 80.5 ± 14.6 TG latex area vs TG latex area 14 Val latex area Val latex area 14

PAGE 111

102 perimeter of binary-choice feeding arenas where no latex or less latex was spread. Some beetles gained access to the untreated undersurface of leaves by chewing through these non-latex treated spots. Once beneath the leaves, they consumed the entire lower leaf surfaces and left the upper surfaces intact. This happened more often in the choice test between leaf areas with latex fi-om TG and from Val. Total amount of leaf area consumed per pair of adults in 48 h was the least in the choice test between leaf tissues spread with latex from TG or fi-om Val (F=16.09; df-3,39; P=0.0001). Pairs of adults exposed only to untreated lima bean leaves ate all available leaf tissue (average of 576.1 ± 47.3 mm^) in 48 h (Table 5-4), which is not significantly different from that in the choice tests between control leaf tissues and leaf tissues with TG or Val latex. Significantly less total leaf surface was eaten by beetles in the choice tests between TG and Val than in those given a choice between control and either TG or Val. Observation of Feedina Behavior Lettuce plants store their latex within secretory cells (laticifers), which may have a role in lettuce resistance. When these cells were ruptured by adult feeding, the white milky fluids were immediately secreted and became adhesive. In addition to any deterrent chemicals in the latex, latex may act as a physical defense for lettuce to affect normal beetle feeding behavior. When adults came in contact with fresh latex fi-om injured TG tissue, their continuous feeding behavior was suddenly disrupted. Due to such encounters, latex often adhered to the mouthparts and tarsi. This generally resulted in beetles engaging in cleaning behaviors before they moved onto other areas of the same leaf or other leaves. After several experiences with fresh latex, beetles tried to avoid it by

PAGE 112

103 Table 5-4. Mean total leaf area consumed per pair of adult D. balteata in 48 h on two exposed areas of a lima bean leaf either untreated or smeared with latex from one of two lettuce cultivars Choice N Leaf area eaten (mm^ ± SEM) Control area vs. control area 14 576.1 ±47.3 a Control area vs. TG latex area 14 569.5 ±45.5 a Control area vs. Val latex area 14 484.2 ±45.1 a TG latex area vs. Val latex area 14 218.0 ± 22.6 b Data followed by the same letter are not significantly different by Tukey's mean separation test at the 0.05 level.

PAGE 113

104 chewing on the leaf surface first before they ate through the leaf or coming back to the previously damaged spots made by either themselves or other beetles. Inducible Resistance Localized response. A localized resistance response was observed on Val following 48 h of previous feeding by other beetles. Adults ate 20.0 ± 4.7 mm^ of the available leaf area on previously damaged leaves of Val, which was only 26% of that on undamaged leaves from the same cuUivar (t=3.7137; df=ll; P=0.0034) (Fig. 5-4). Ten percent less feeding damage was observed on previously damaged than on undamaged TG leaves, but the difference was not significant at P=0.05 (t=1.6161; df=l 1; P=0.1343). Systemic response. No systemic response to previous feeding was observed after a 48 h test period (Fig. 5-5). The leaf area consumed on undamaged leaves of plants previously damaged for 48 h did not differ significantly from that on undamaged leaves of undamaged TG plants (t=0.6168; df=l 1; P=0.5499). Adult beetles ate 220.7 ± 28.1 mm^ of leaf area on undamaged leaves of undamaged TG plants, which was very similar to the amount of leaf area consumed on undamaged leaves of damaged TG plants. The same lack of preference by adult D. balteata was found between undamaged leaves of damaged Val plants (40.2 ± 7.2 mm^) and undamaged leaves of undamaged plants of the same cultivar (37.7 ± 7.3 mm^) (t=0.2284; df=l 1 ; P=0.8235). Discussion Although excised leaves or leaf disks are often used for evaluating plants for resistance to leaf feeding insects, biochemical and physiological changes in such excised leaves or leaf disks may affect the feeding of the test insects (Raina et al. 1980, Risch 1985, van Emden and Bashford 1976). In our case, Val expressed a very high degree of

PAGE 114

105 300 n 250 200 150 100 50 TG-control TG-damaged Val-control Val-damaged Fig. 5-4. Leaf area consumed per pair of adult D. balteata for 48 h when given a choice between undamaged and damaged leaves of susceptible Tall Guzmaine (TG-control vs. TG-damaged) and between undamaged and damaged leaves of resistant Valmaine (Val-control vs. Val-damaged). Vertical lines indicate +1 SEM. Asterisk means significant difference at the 0.05 level by Tukey's HSD test.

PAGE 115

106 TG-control TG-damaged Val-control Val-damaged Fig. 5-5. Leaf area consumed per pair of adult D. balteata for 48 h when given a choice between undamaged leaves of damaged susceptible Tall Guzmaine (TG-damaged) and undamaged leaves of undamaged TG (TG-control) and between undamaged leaves of damaged resistant Valmaine (Val-damaged) and undamaged leaves of undamaged Val (Val-control). Vertical lines indicate +1 SEM.

PAGE 116

107 resistance to adult feeding when intact leaves and excised leaves were used in choice tests, but it failed to show resistance in a choice test when leaf disks were used and in a no-choice test when excised leaves were used. Therefore, methods of testing for resistance had a significant effect on the feeding preferences of adult D. balteata between TG and Val. The intact leaf method, not the excised leaf and leaf disk methods, is suitable and reliable to evaluate lettuce cultivars for resistance to D. balteata. Risch (1985) also reported that method of testing (whole plants, excised leaves, and leaf disks) had a very significant effect on preferences of a group of specialist and generalist chrysomeHd beetles, including D. balteata, for com, bean and squash. Test methodology affected the level of statistical significance for a particular preference in some cases. In other cases, the direction of preference was reversed altogether. A much greater effect of test method was found between the disks test and the whole plant tests than between the excised leaves and whole plant tests. Furthermore, the feeding preferences of the two specialist species, Acalymona thiemei (Baly) and Ceratoma ruficornis (Olivier) were less affected by test method than were the more generalist species, D. balteata and D. adelpha (Harold). Lettuce cultivars resistant to the lettuce aphid, Nasonovia ribisnigri, became fully acceptable when leaf fragments were given in a leaf disk test (see references cited by Schoonhoven et al. 1998). Leaf disk size was found to be another variable affecting the outcome of insect feeding preferences because the ratio of cut edge to overall leaf disk surface area influenced the chance of encountering internal attractants/stimulants (Jones and Coleman 1988). In contrast, there are some cases in which excised leaf and leaf disk methods are quite reliable for screening resistant cultivars (Raina et al. 1980, Sams et al. 1975, East et al. 1992).

PAGE 117

108 Localized inducible resistance was found in Val but not in TG after lettuce was previously damaged for 48 h. Wound-induced resistance by herbivores is correlated with the induction of activities of oxidative enzymes, de novo or enhanced synthesis of secondary metabolites, and decline in nutritional quality (Bi et al. 1994; 1997, McAuslane et al. 1997, Tuomi et al. 1988). After a leaf of Val has been attacked by the beetles, changes such as induction of activities of lipoxygenase, peroxidases, or polyphenol oxidase may occur, which in turn affects those beetles attempting to use this leaf at a later time. It has been reported that phenylalanine ammonia lyase (PAL), the first enzyme in the phenyl propanoid pathway (Fig. 5-6), is more active in aphid resistant cultivars of Z. saliva, and that resistant cultivars have a greater tendency to brown when plants were damaged (Cole 1984). This kind of resistance was due to polyphenol oxidase activity and its substrate isochlorogenic acid in the leaves. Induced resistance may also be correlated with enhanced synthesis of secondary metabolites in Val. It has been observed that browning reactions occur at the site of damage on lettuce (Cole 1984). This browning is reported to be due to oxidation of hydroxylated phenolic compounds to quinones that are more toxic than their precursors to herbivores (Cole 1984). The enhanced production of phenolic compounds is indicated by an increase in the activity of PAL in wounded tissues (Bi and Felton 1995). This localized inducible resistance may not be expressed strongly enough or exist in excised leaves so that beetles may show significantly different responses from intact leaves. The presence of localized induced resistance in whole plants but not in leaf disks or excised leaves may be one of the reasons why Val was less resistant when tested as excised

PAGE 118

109 leaves and leaf disks. How quickly this localized response occurs after initial feeding damage and for how long it lasts needs to be further investigated. Systemic resistance was not found in Val plants after they were previously damaged for 48 h. Val plants may save energy and nutrients for growth and development by changing chemical status only in those damaged leaves. Furthermore, systemic resistance may be not an efficient and fast way to protect small leafy plants fi-om insect attack because systemic resistance usually takes longer to mobilize than localized inducible resistance (Schoonhoven et al. 1998). The induction times for systemic resistance are relatively slow and vary from 12 h to as long as one year or more, while localized resistance is usually more rapid and can occur as a consequence of disruption of cell compartmentation (Edwards and Wratten 1983). Alternatively, systemic resistance may take longer than 48 h to appear in a neighboring leaf Latex may act as a physical defense for lettuce to affect beetle normal feeding behavior based on the brief behavioral observation of beetle feeding on TG plants. It was mentioned in chapter 2 that beetles liked to gather together to feed on a single plant because of an aggregation pheromone. ft is also possible that gathering together to feed may help to break down this latex physical defense system through reducing latex volume or pressure. Val partially or totally lost its resistance when excised leaf tissues was used, suggesting that reduced or ceased latex emission may be partially responsible for the loss of resistance and that latex flow may act as a physical barrier to this insect feeding. When both Val and TG leaves were cut off the plants, adults ate more leaf tissue than from attached leaves. This may be due to decreased latex flow and water potential

PAGE 119

no caused by cutting of the petioles. Furthermore, when leaves were cut into disks, Val was no longer resistant to adult feeding, possibly because of depressurization of laticifers. Latex may also provide chemical defenses for lettuce. Many different organic compounds have been identified in latex of Lactuca sp, including organic acids, phenolics, and a triterpene alcohol (Crosby 1963, Gonzalez 1977, Cole 1984). Like many plant secondary compounds these organic compounds may act as deterrents or toxins to potential herbivores. It has been reported that three sesquiterpene lactones (8deoxylactucin, lactupicrin and cichoriin) secreted in the latex in chicory plants reduce feeding of Schistocerca gregaria Forskal and provide a significant barrier to herbivory (Rees and Harbome 1985). Results from the latex smearing test reported here showed that latex from both TG and Val were very deterrent to adult feeding. The deterrents may be the original chemicals produced by the plants but still active in oxidized latex or may be chemicals produced from oxidation. However, since both cultivars produce latex once a laticifer is ruptured, latex physical and chemical defenses may also be present in TG. Adults ate more foliage on detached leaves of TG and Val than on intact leaves of TG in no-choice tests, suggesting that intact leaves of TG also show some resistance (Fig. 5-3). This kind of resistance may be due to latex flow but not localized inducible resistance so that the level of resistance was significantly lower than that in intact leaves of Val. How the adults circumvent these defenses on TG is unknown. It was also not clear whether latex per se or something else in the latex was the cause of the resistance in Val but not in TG. One possibility is that the volume of latex produced and the internal pressure are different between TG and Val. Because of quick coagulation of the latex upon contact with air, latex volume was hard to measure. Another possibility is that the potential

PAGE 120

Ill antifeedants in lettuce latex may not be enough to keep the beetles away from TG, but in symphony with localized induced resistance or other plant secondary chemicals in Val, they result in low foliar damage and reproductive failure. Dussourd (1995) proposed that latex might be useful in pest management and that young leaves that released latex would be invulnerable to most herbivores. Based on our results, two hypotheses are proposed to explain resistance in the Valmaine cultivar of romaine lettuce against D. balteata. One is localized inducible resistance, and another is the presence of a physical barrier or chemical defenses in latex. Thus, beetles may have to contend with both latex flow or chemical defense in latex and localized inducible resistance in response to feeding when they try to eat Val plants. Each plant species has a unique set of defense traits, ranging from morphological to phytochemical parameters, against a potential herbivore. These defense traits are often inseparable, and can synergize with each other as well. Therefore, plant resistance mechanisms against herbivores are multidimensional. It is possible that localized inducible resistance and latex defenses in Val lettuce synergize with each other and enhance Val resistance to D. balteata. Resistance that only relies on a single defense trait will be less compelling because this kind of resistance can be overcome quickly and easily by certain insects. Resistant plant varieties can be an important component of an integrated pest management program. Determining the components involved in resistance mechanisms can allow for more rapid and precise introduction of these traits into commercial varieties through conventional plant breeding.

PAGE 121

i 112 Phenylalanine Phenylalanine ammonia lyase (PAL) ^ Cinnamic acid Ferulic acid Caffeic acid Quinic acid p-Coumaric acid Polyphenyloxidase (PPO) Chlorogenic acid PPO o-Benzoquinone Isochlorogenic acid ^ PPO g. 5-6. Relationship and biosynthetic derivation of various phenolic acids of the phenylpropanoid pathway (Cole 1984).

PAGE 122

CHAPTER 6 SUMMARY AND CONCLUSIONS Four cultivars of lettuce, Lactuca sativa L., were evaluated for their resistance to the banded cucumber beetle, Diabrotica balteata LeConte, under laboratory conditions. Binary-choice and no-choice tests showed Valmaine (Val) had the highest level of resistance, and Tall Guzmaine (TG) was susceptible to the adults when using leaf area 'consumption as the criterion for resistance. Parris White (PW) and Short Guzmaine (SG) had an intermediate level of resistance compared to Val. Female adults ate more foliage, gained more weight, and produced more eggs when they fed on TG for 10, 13 or 16 d, compared with females fed on Val. No mature eggs were found in any females on Val at day 10, 13 and 16. Three out of forty-two females from Val had a few undeveloped eggs. PW and SG also had some effects on female weight and number of eggs produced, but not as significantly as Val. A starvation test showed that most D. balteata could not survive without food after 7 d. This lends credence to the hypothesis that females on Val had no mature egg development because of their inability to use Val as a food source. The reduced amount of feeding damage on Val and lack of mature eggs indicate that Val may have some antixenotic characters, such as biophysical and biochemical factors, to keep D. balteata from using it as a host, resulting in high mortality of the beetles and no mature egg production by living females, as compared with that on the susceptible cultivar TG. The leaf surface of a plant, and especially its chemistry, constitutes the first line of resistance to herbivores and other pests. Adult D. balteata consumed much more leaf 113

PAGE 123

114 tissue of Val and TG with their surface chemicals removed, which led me to think that leaf surface chemicals may have a role in expression of lettuce resistance. However, leaf surface extracts from Val were not deterrent to adult D. halteata when applied to lima bean leaf surfaces at various concentrations in binary-choice and no-choice tests. Leaf surface extracts from TG applied at increasing concentrations did not cause adults to feed more on lima bean leaves but actually decreased feeding similar to extracts of Val. Therefore, these results suggest that leaf surface chemicals in lettuce do not explain resistance of Val to adult D. balteata and that chemicals inside the leaf may play a role in resistance. It is also possible that disrupting the leaf surface with solvents changed the plant's ability to respond to feeding damage. Freeze-dried leaf powder from susceptible TG and from resistant Val was incorporated into artificial diet for/), balteata to evaluate whether the chemicals within the lettuce leaves were responsible for lettuce resistance by measuring diet consumed, egg production, female weight gain and adult's mortality. In choice experiments, there were no significant differences in measured responses between the diet with TG leaf powder and the diet with Val leaf powder at any concentration. However, as the level of leaf powder of TG or Val increased, adult consumption of the diet also increased. In nochoice experiments, no great differences were found in number of mature eggs produced, female weight gain, and adult mortality between TG and Val diets at the four different levels. Using this assay I again failed to detect any reason for the resistance in the lettuce cultivar Val, which had the highest resistance against adult's feeding among four lettuce cultivars when fresh whole plants were used. I concluded that resistance to adult D.

PAGE 124

115 balteata in Val lettuce may be due to physical properties of fresh plant material or due to other chemicals which are only active in fresh leaves. The resistance in lettuce to adult D. balteata was further evaluated using three screening methods: leaf disk, excised leaves and intact leaves attached to the plant. Presentation of different kinds of leaf material had a significant effect on the level of feeding damage by adults on TG and Val. Val was strongly resistant to beetle feeding when intact or excised leaves were presented in binary tests with susceptible TG. Resistance was also noted in Val when intact leaves were presented in no-choice tests. However, no resistance to beetle feeding was observed when binary-choice and no-choice tests were performed using excised leaf disks. The results indicate that it is most suitable and reliable to use intact leaves and egg production to evaluate lettuce cultivars for resistance to adults. I further investigated physical properties of latex of TG and Val, possible chemical defenses in latex, and possible inducible resistance as mechanisms for lettuce resistance. The two cultivars differed in latex physical properties. Latex from Val turned brown faster than that from TG, and flowed for a longer period of time before stopping. No differences were found in latex fresh and dry weights between Val and TG. Latex from both TG and Val was strongly deterrent to the beetle feeding. Localized induced resistance to feeding was found in Val but not in TG after plants were previously damaged for 48 h. No systemic resistance was found either in Val or TG after plants were previously damaged for 48 h. Based on studies done so far, I propose that localized inducible resistance and a physical barrier and chemical defenses in latex may be two major reasons for resistance

PAGE 125

116 in Val. After a leaf of Val is attacked by adult beetles, enzymatic changes or enhanced synthesis of secondary metabolites may occur to keep this leaf from further damage. Latex fi-om lettuce becomes adhesive and turns brown upon exposure to air, which acts as a physical barrier to keep the adults from feeding continuously. The loss of turgor pressure and associated reduction or cessation of latex flow from damaged tissue on excised leaves and leaf disks may be responsible for the loss of resistance expression in Val tissue in these tests. Localized inducible resistance and latex physical barrier may also explain the reason why Val decreases or loses its resistance in detached leaves and leaf disks. In the southern United States, D. balteata occurs as an adult during much of the year. This insect is a pest of leafy vegetables and com in Florida because of its direct feeding damage on foliage, its capacity to transmit plant pathogens, and contamination of lettuce with frass. Since chemical control is costly and only partially effective against this insect, the use of host plant resistance is an alternative way to control this herbivore. Moreover, use of cultivars resistant to adult D. balteata also can reduce the need for insecticide application. Understanding the mechanisms of resistance is a prerequisite for application and development of host plant resistance in pest management programs. The mechanisms of resistance in lettuce against D. balteata have still not been completely elucidated. How quickly the localized inducible resistance occurs after initial feeding damage and how long it lasts still need to be investigated. Moreover, how enzymatic activities change after induction will also need to be studied. Whether latex chemistry contributes to the resistance in Val is also very interesting and presents a challenging area for further research.

PAGE 126

APPENDIX MISCELLANEA Leaf disks were cut from the seventh leaves of plants by a No. 15 cork borer. Two leaf disks from TG and Val were presented to a female or male in a petri dish for 16 h. Total leaf area consumed by female or male beetles on the two disks was calculated. Feeding differences between females and males were not found to be significant (F=1.85; df=l, 24; P=0.1869). Females ate 533.4 ± 24.5 mm^ (Mean ± SEM) of leaf surface area, while males ate almost the same amount of 589.7 ± 27.6 mm^. Before using adult artificial diet to test for possible feeding preferences between TG and Val leaf powder, plain agar and agar with sucrose (3%) were offered to newly emerged beetles to see whether the plain agar or agar with sucrose could be used as a testing substrate. Adults were not interested in the plain agar or the agar with sucrose. In order to study possible differences in laticifer structure between TG and Val plants, leaves were cleared by ethanol and stained with basic Fuchsin. However, this staining procedure resulted in staining only the xylem tissue with pink color and laticifers were still not visible. 117

PAGE 127

LIST OF REFERENCES Anonymous. 1 996. Standard operating procedure-rearing, Bayer Corporation, Vero Beach, FL. Adati, T., and K. Matsuda. 1993. Feeding stimulants for various leaf beetles (Coleoptera: Chrysomelidae) in the leaf surface wax of their host plants. App. Entomol. Zool. 28: 319-324. Alfaro, R. 1., H. D. Pierce, Jr., J. H. Borden, and A. C. Oehlschlager. 1979. A quantitative feeding bioassay for Pissodes strobi Peck (Coleoptera: Curculionidae). J. Chem. Ecol. 5: 663-671. Alonso-Amelot, M. E., J. L. Avila, L. D. Otero, F. Mora, and B. Wolff 1994. A new bioassay for testing plant extracts and pure compounds using red flour beetle Tribolium castaneum Herbst. J. Chem. Ecol. 20: 1 161-1 177. Baker, E. A. 1982. Chemistry and morphology of plant epicuticular waxes, pp. 139-165. In D. F. Cutler, K. L. Alvin, and C. E. Price [eds.]. The Plant Cuticle. Linnean Society of London, London. Bakker, M. 1., W. J. Baas, D. T. H. M. Sum, and D. Kolloffel. 1998. Leaf wax of Lactuca sativa and Plantago major. Phytochem. 47: 1489-1493. Balandrin, M. F., J. A. Kloche, E. S. Wurtele and W. H. Bollinger. 1985. Natural plant chemicals: sources of industrial and medicinal materials. Science 228: 1 154-1 160. Baldwin, I. T. 1989. Chemical changes rapidly induced by folivory, pp. 1-24. /// E. A. Bemays [ed.], Insect-Plant Interactions, vol. 5. CRC Press, Boca Raton, Florida. Baldwin, I. T. 1991. Damage-induced alkaloids in wild tobacco, pp. 47-69. In D. W. Tallamy and M. J. Raupp [eds.], Phytochemcial Induction by Herbivores. John Wiley & Sons, Inc., New York. Baldwin, L. T., and T. E. Ohnmeiss. 1994. Coordination of photosynthetic and alkaloidal responses to damage in uninducible and inducible Nicotiana sylvestris Ecology 751003-1014. Beck, S. D., and J. C. Reese. 1976. Insect-plant interactions: nutrition and metabolism. Recent. Advan. Phytochem. 10: 41-92. 118

PAGE 128

119 Bennett, M. H., M. Gallagher, C. Bestwich, J. Rossiter, and J. Mansfield. 1994. The phytoalexin response of lettuce to challenge by Botrytis cinerea, Bremia lactucae and Pseudomonas syringae pv. phaseolicola. Physiol. Mol. Plant Pathol. 44: 321333. Bennett, M., M. Gallagher, J. Fagg, C. Bestwich, T. Paul, M. Beale, and J. Mansfield. 1 996. The hypersensitive reaction, membrane damage and accumulation of autofluorescent phenolics in lettuce cells challenged by Bremia lactuca. The Plant J. 9: 851-865. Bergman, D. K., J. W. Dillwith, A. A. Zarrabi, J. L. Caddel, and R. C. Berberet. 1991. Epicuticular lipids of alfalfa relative to its susceptibility to spotted alfalfa aphids (Homoptera: Aphididae). Environ. Entomol. 20: 781-785. Bemays, E. A., and R. F. Chapman. 1977. Deterrent chemicals as a basis of oligophagy in Locusta migratoria. Ecol. Entomol. 2: 1-18. Bemays, E. A., and R. F. Chapman. 1978. Plant chemistry and acridoid feeding behaviour, pp. 99-141. /« J. B. Harbme [ed.], Coevolution of Plants and Animals. Academic Press. London. Bemays, E. A., and R. F. Chapman. 2000. Plant secondary compounds and grasshoppers: beyond plant defenses. J. Chem. Ecol. 26: 1773-1794. Bemays, E. A., R. F. Chapman, and S. Woodhead. 1983. Behaviour of newly hatched larvae of Chilo partellus (Swinhoe) (Lepidoptera: Pyralidae) associated with their establishment in the host-plant, sorghum. Bull. Entomol. Res. 73: 75-83. Bemays, E. A., and S. Woodhead. 1984. The need for high levels of phenylalanine in the diet of Schistocerca gregaria nymphs. J. Insect Physiol. 30: 489-493. Bi, J. L., and G. W. Felton. 1995. Foliar oxidative stress and insect herbivory: primary compounds, secondary metabolites, and reactive oxygen species as components of induced resistance. J. Chem. Ecol. 21: 1511-1530. Bi, J. L., G. W. Felton, and A. J. Mueller. 1994. Induced resistance in soybean to Helicoverpa zea: role of plant protein quality. J. Chem. Ecol. 20: 183-198. Bi, J. L., J. B. Murphy, and G. W. Felton. 1997. Antinutritive and oxidative components as mechanisms of induced resistance in cotton to Helicoverpa zea. J Chem Ecol 23:97-117. Biesboer, D. D., and P. G. Mahlberg. 1978. Accumulation of nonutilizable starch i laticifers of Euphorbia heterophylla and E. myrsinites. Planta 143: 5-10.

PAGE 129

120 Blaney, W. M., and C. Winstanley. 1980. Insect Neurobiology and Pesticide Action. Society of Chemical Industry. London, U.K. Bodnaryk, R. P. 1992. Leaf epicuticular wax, an antixenotic factor in Brassicaeae that affects the rate and pattern of feeding of flea beetles, Phyllotreta cruciferae (Goeze). Can. J. Plant Sci. 72: 1295-1303. de Boer , G., and F. E. Hanson. 1988. The role of leaf lipids in food selection by larvae of the tobacco homworm, Manduca sexta. J. Chem. Ecology 14: 669-682. Butler, W. L. 1960. Chlorogenic acid of lettuce seeds. Nature 185: 856-857. Cardona, C, R. Gonzalez, and A. V. Schoonhoven. 1982. Evaluation of damage to common beans by larvae and adults of Diabrotica balteata and Cerotoma facialis. J. Econ. Entomol. 75: 324-327. Carozzi, N. and M. Koziel. 1997. Advances in Insect Control: the Role of Transgenic Plants. Taylor & Francis Publishers, London. Cole, R. A. 1984. Phenolic acids associated with the resistance of lettuce cultivars to the lettuce root aphid. Ann. Appl. Biol. 105: 129-145. Cole, R. A. 1987. Intensity of radicle fluorescence as related to the resistance of seedlings of lettuce to the lettuce root aphid and carrot to the carrot fly. Ann. Appl Biol 1 1 1 • 629-639. Cole, R. A., R. A. Sutherland, and W. E. Riggall. 1991. The use of polyacrylamide gradient gel electrophoresis to identify variation in isozymes as markers for Lactuca species and resistance to the lettuce root aphid Pemphigus bursarius. Euphytica 36237-242. Coudriet, D. L., A. N. Kishaba, J. D. McCreight, and G. W. Bohn. 1980. Susceptibility in 'Deserta Naja' casaba melon to a cucumber beetle. HortSci. 15: 660-662. Creighton, C. S., and E. R. Cuthbert, Jr. 1968. A semisyntheitc diet for adult banded cucumber beetles. J. Econ. Entomol. 61: 337-338. Crosby, D. G. 1963. The organic constituents of food. 1. Lettuce. J. Food Science 28347-355. Da Costa, D. P., and C. M. Jones. 1 971 . Resistance in cucumber, Cucumis sativits L. to three species of cucumber beetles. HortScience 6: 430-442. Data, E. S., S. F. Nottingham, and S. J. Kays. 1996. Effect of sweetpotato latex on sweetpotato weevil (Coleoptera: Curculionidae) feeding and oviposition J Econ Entomol 89: 544-549.

PAGE 130

121 Deheer, C. J., and D. W. Tallamy. 1991 . Affinity of spotted cucumber beetle (Coleoptera: Chrysomelidae) larvae to cucurbitacins. Environ. Entomol. 1 173-1 175. Derridi, S., B. R., Wu, L. Stammitti, J. P. Garrec, and A. Derriea. 1996. Chemicals on the leaf surface, information about the plant available to insects. Entomol. Exp. App. 80: 197-201. Dillon, P. M., S. Lowrie, and D. McKey. 1983. Disarming the "Evil woman": petile constriction by a sphingid larvae circumvents mechanical defenses of its host plant, Cnidoscolus urens (Euphorbiaceae). Biotropica 15: 112-116. Dicke, M. 1999. Specificity of herbivore-induced plant defences, pp 43-59. In InsectPlant Interactions and Induced Plant Defence, Novartis Foundation Symposium 223. John Wiley & Sons. Ltd, Chichester, U.K. Dowd, P. F. 1994. Enhanced maize {Zea mays L.) pericarp browning: associations with insect resistance and involvement of oxidizing enzymes. J. Chem. Ecol. 20: 27772803. Dussourd, D. E. 1993. Foraging with fitness: caterpillar adaptations for circumventing plant defenses, pp. 92-131. /« N. E. Stamp and R. M. Casey [eds.]. Caterpillars: Ecological and Evolutionary Constraints on Foraging. Chapman and Hall, New York. Dussourd, D. E. 1995. Entrapment of aphids and whiteflies in lettuce latex. Ann. Entomol. Soc. Am. 88: 163-172. Dussourd, D. E. 1997. Plant exudates trigger leaf-trenching by cabbage loopers, Trichoplusia ni (Noctuidae). Oecologia 1 12: 362-369. Dussourd, D. E. 1999. Behavioral sabotage of plant defense: Do vein cuts and trenches reduce insect exposure to exudates? J. Insect Behav. 17: 501-515. Dussourd, D. E., and R. F. Denno. 1991. Deactivation of plant defense: correspondence between insect behavior and secretory canal architecture. Ecology 72: 1381-1396. Dussourd, D. E., and T. Eisner. 1987. Vein-cutting behavior: insect counterploy to the latex defense of plants. Science 237: 898-901. East, D. A., J. V. Edelson, E. L. Cox, and M. K. Harris. 1992. Evaluation of screening methods and search for resistance in muskmelon, Cucumis melo L., to the twospotted spider mite, Tetranychus urticae Koch. Crop Protection 1 1 : 39-44. Edwards, P. J., and S. D. Wratten. 1983. Wound induced defences in plants and their consequences for patterns of insect grazing. Oecologia 59: 88-93.

PAGE 131

122 Eenink, A. H., and F. L. Dieleman. 1982. Resistance of Lactuca assessions to leaf aphids: components of resistance and exploitation of wild Lactuca species as sources of resistance, pp. 349-355. In J. H. Visser and A. K. Minks [eds.], Insect-Plant Relationships, Proceedings of the 5th International Symposium. Wageningen, the Netherlands. Eenink, A. H., R. Groenwold, and F. L. Dieleman. 1982. Resistance of lettuce {Lactuca) to the leaf aphid Nasonovia ribisttigri. 1. Transer of resistance from L. virosa to L. sativa by interspecific crosses and selection of resistant breeding lines. Euphytica 31:291-300. Eigenbrode, S. D., and K. E., Espelie. 1995. Effects of plant epicuticular lipids on insect herbivores. Annu. Rev. Entomol. 40: 171-194. Eigenbrode, S. D., and A. M. Shelton. 1990. Behavior of neonate diamondback moth larvae (Lepidoptera: Plutellidae) on glossy-leaved resistant genotypes of Brassica olemcea L.. Environ. Entomol. 19: 1566-1571. Eigenbrode, S. D., K. E., Espelie, and A. M. Shelton. 1991. Behavior of neonate dimaonback moth larvae {Plutella xylostella L.) on leaves and on extracted leaf waxes of resistant and susceptible cabbages. J. Chem. Ecol. 17: 1691-1704. Ellis, P. R., D. A. C. Pink, and A. D. Ramsey. 1994. Inheritance of resistance to lettuce root aphid in the lettuce cultivars 'Avoncrisp' and 'Lakeland'. Ann. Appl. Biol. 124: 141-151. Elsey, K. D. 1988. Cucumber beetle seasonality in coastal South Carolina. Environ. Entomol. 17: 496-502. Erikson, J. M., and P. Feeny. 1974. Sinigrin: a chemical barrier to the black swallowtail butterfly, Papilio polyxenes. Ecol. 55: 103-1 1 1 . Esau, K. 1976. Anatomy of Seed Plants, 2nd ed. John Wiley & Sons, New York. Escoubas, P., Y. Fukushi, L. Lajide, and J. Mizutani. 1992. A new method for fast isolation of insect antifeedant compounds from complex mixtures. J. Chem. Ecol 18: 1819-1832. Evans, F. J., and R. J. Schmidt. 1976. Two new toxins from the latex of Euphorbia poissonii. Phytochem. 15: 333-335. Fahn, A. 1979. Secretory Tissues in Plants. Academic Press, London. Farrell, B. D., D. E. Dussourd, and C. Mitter. 1991. Escalation of plant defense: do latex/resin canals spur plant diversification? Am. Nat. 138: 891-900.

PAGE 132

123 Felton, G. W., K. Donate, R. J. Vecchio, and S. S. Duffey. 1989. Activation of plant foliar oxidases by insect feeding reduced nutritive quality of foliage for noctuid herbivores. J. Chem. Ecol. 15: 2667-2694. Femandes, G. W. 1990. Hypersensitivity: a neglected plant resistance mechanism against insect herbivores. Environ. Entomol. 19: 1 173-1 182. Fraenkel, G. S. 1959. The raison d'etre of secondary plant substances. Science 129: 14661470. Geiger, D. R. 1987. Understanding interactions of source and sink regions of plants. Plant Physiol. Biochem. 25: 659-666. Gergerich, R. C, H. A. Scott and J. P. Fulton. 1986. Evaluation of Diabrotica beetles as vectors of plant viruses, pp. 227-249. In J. L. Krysan and T. A. Miller [eds.], Methods for the Study of Pest Diabrotica, Springer, New York. Gonzalez, A. G. 1977. Lactuceae Chemical review, pp 1081-1095. In V.H. Heywood and J. B. Harbome [eds.]. The Biology and Chemistry of the Compositae. Academic Press, New York. Green, T. R., and C. A. Ryan. 1972. Wound induced proteinase inhibitors in plant leaves. Science 175: 776-777. Guzman, V. L. 1986. Short Guzmaine, Tall Guzmaine and Floriglade, three cos lettuce cultivarsresistanttolettucemosaicvirus. IFAS, Univ. ofFla, Agric Exp Stn Circ S-326. lip. Guzman, V. L., and T. A. Zitter. 1983. Floricos 83: a cos lettuce cultivar resistant to two viruses, for Florida organic soils. IFAS, Univ. of Fla, Agric. Exp. Stn. Circ S-326 9p. Haupt, 1. 1976. Separation of the sites of synthesis and accumulation of 3,4dihydroxyphenylalanine in Euphorbia lathyris L. Nova Acta Leopoldina Supplementum 7: 129-132. Hartley, S. E., and J. H. Lawton. 1991. Biochemcial aspects and significance of the rapidly induced accumulation of phenolics in birch foliage, pp. 105-132. /// D. W. Tallamy and M. J. Raupp [eds.], Phytochemcial hiduction by Herbivores. John Wiley & Sons, Inc., New York. Havill, N. P., and K. F. Raffa. 1999. Effects of elicitation treatment and genotypic variation on induced resistance in Popidus: impacts on gypsy moth {Lepidoptera: Lymantriidae) developoment and feeding behavior. Oecologia 120: 295-303.

PAGE 133

124 Hedin, P. A. 1986. Developing research trends in the chemistry' of plant resistance to pests, pp 2-14. In M. B. Green and P. A. Hedin [eds.], Natural Resistance of Plants to Pest Role of Allelochemicals, ACS Symposium Series 296. American Chemical Society, Washington, D. C. Heupel, R. C. 1985. Varietal similarities and differences in the polycyclic isopentenoid composition of sorghum. Phytochem. 24: 2929-2937. Horber, E. 1980. Types and classification of resistance, pp 15-21. In F. G. Maxwell and P. R. Jennings [eds.]. Breeding Plants Resistant to Insects. John Wiley & Sons, New York. Hori, M. 1999. Antifeeding, settling inhibitory and toxic activities of labiate essential oils against the green peach aphid, Myzus persicae (Sulzer) (Homoptera: Aphididae). Appl. Entomol. Zool. 34: 113-118. Hori, K., and R. Atalay. 1980. Biochemical changes in the tissue of Chinese cabbage injured by the bug Lygus disposi. App. Entomol. Zool. 15: 234-241. Hsiao, T. H., and G. Fraenkel. 1968. Isolation of phagostimulative substances from the host plant of the Colorado potato beetle. Ann. Entomol. Soc. Am. 61 : 476-484. Hussein, H. I., A. Kamel, M. Abou-Xied, A. H. El-Sebae, and M. A. Saleh. 1994. Uscharin, the most potent molluscicidal compound tested against land snail. J. Chem. Ecol. 20: 135-140. Jones, A., P. D. Dukes, and F. P. Cuthbert, Jr. 1976. Mass selection in sweet potato: breeding for resistance to insects and diseases and for horticultural characteristics. J. Am. Soc. Hort. Sci. 101: 701-704. Jones, G. C, and J. S. Coleman. 1988. Leaf disc size and insect feeding preference: implications for assays and studies on induction of plant defense. Entomol. Exp. Appl. 47: 167-172. Jones, G. C, R. R. Hopper, J. S. Coleman, and V. A. Krischik. 1993. Control of systemically induced herbivore resistance by plant vascular architecture. Oecologia 93: 452-456. Johnson, H. B. 1975. Plant pubescence: an ecological perspective. Bot. Rev. 41: 233-258. Juniper, B. and S. R. Southwood. 1986. Insects and the Plant Surface. Edward Arnold, London. Karban, R., and J. H. Myers. 1989. Induced plant responses to herbivory. Annu. Rev Ecol. Syst. 20: 331-348.

PAGE 134

125 Karban, R., and I. T. Baldwin. 1997. Induced Responses to Herbivory. University of Chicago Press, Chicago. Keathley, C. P., D. A. Potter, and R. L. Houtz. 1999. Freezing-altered palatability of Bradford pear to Japanese beetles: evidence for decompartmentalization and enzymatic degradation of feeding deterrents. Entomol. Exp. et Appl. 90: 49-59. Kogan, M., and J. Paxton. 1983. Natural inducers of plant resistance to insects, pp. 153171. //? P. A. Hedin [ed.], Plant Resistance to Insects, American Chemical Society Symposium Series 208. American Chemical Society, Washington, D. C. Krysan, J. L. 1986. Introduction: biology, distribution, and identification of pest Diahrotica, pp. 1-23. //; J. L. Krysan and T. A. Miller [eds.]. Methods for the Study of Pest Diahrotica. Springer-Verlag, New York. Kumar, H. 1992. Inhibition of ovipositional responses of Chilo partellus (Lepidoptera: Pyralidae) by the trichomes on the lower leaf surface of a maize cultivar. J. Econ. Entomol. 85: 1736-1739. Layton, M. B., D. J. Boethel, and C. M. Smith. 1987. Resistance to adult bean leaf beetle and banded cucumber beetle (Coleoptera: Chrysomelidae) in soybean. J. Econ. Entomol. 80: 151-155. Lewis, A. C, and H. F. van Emden. 1986. Assays for insect feeding, pp. 95-11 9. In J. R. Miller and T. A. Miller [eds.], Insect-Plant Interactions. Springer-Verlag New York. Lin, S., G. R., Binder, and E. R. Hart. 1998. Insect feeding stimulants from the leaf surface of Popiilus. J. Chem. Ecol. 24: 1781-1790. Lin, S. Y. H., J. T. Trumble, and J. Kumamoto. 1986. Activity of volatile compounds in glandular trichomes of Lycopersicon species against two insect herbivores J Chem Ecol. 13: 837-850. Louda, S., and S. Mole. 1991. Glucosinolates: chemistry and ecology, pp. 124-164. In G. A. Rosenthal and M. R. Berenbaum, [eds.], Herbivores: Their Interactions with Secondary Plant Metabolites,Vol. I. 2nd edition. Academic Press, New York. Louise, E. M. 1999. Evolutionary aspects of plant-carnivore interactions, pp. 3-20. In Insect-Plant Interactions and Induced Plant Defence. Novartis Foundation Symposium 223. John Wiley & Sons, Ltd Chichester, U.K. Ma, W. C. 1972. Alteration of chemoreceptor function in armyworm larvae {Spodoptera exempta) by a plant-derived sesquiterpenoid and by sulfhydryl reagents Physiol Entomol. 2: 199-207.

PAGE 135

126 Maksymowych, R., and M. C. Ledbetter. 1987. Fine structure of epithelial canal cells in petioles oi Xanthium pennsylvanicum. Am. J. Bot. 74: 65-73. Matheis, G., and J. R. Whitaker. 1984. Modification of proteins by polyphenol oxidase and peroxidases and their products. J. Food Biochem. 8: 137-162. McAuslane, H. J., H. T. Albom, and J. P. Toth. 1997. Systemic induction of terpenoid aldehydes in cotton pigment glands by feeding of larval Spodoptera exigua. J. Chem. Ecol. 23: 2861-2879. McCloud, E. S., D. W. Tallamy, and F. T. Halaweish. 1995. Squash beetle trenching behavior: Avoidance of cucurbitacin induction or mucilaginous plant sap? Ecol. Entomol. 20: 51-59. Metcalf, R. L. 1986. Forword. In J. L. Krysan and T. A. Miller [eds.]. Methods for the Study of Pest Diabrotica. SpringerVerlag, New York. Metcalf, R. L., A. M. Rhodes, R. A. Metcalf, J. Ferguson, E. R. Metcalf, and P. Lu. 1982. Cucurbitacin contents and Diabroticite (Coleoptera: Chrysomelidae) feeding upon Cucurbita spp. Environ. Entomol. 11: 931-937. Miller, J. R., and T. A. Miller. 1986. Insect-Plant hiteractions. SpringerVerlag, New York. Mollema, C, and R. A. Cole. 1996. Low aromatic amino acid concentrations in leaf proteins determine resistance to Frankliniella occidentalis in four vegetable crops. Entomol. Exp. Appl. 78: 325-333. Mori, M. 1982. n-Hexacosanol and n-octacosanol: feeding stimulants for larvae of the silkworm, Bombyx mori. J. Insect Physiol. 1: 969-973. Mutikainen, P., M. Walls, and J. Ovaska. 1996. Herbivore-induced resistance in Betula pendida: the role of plant vascular architecture. Oecologia 108: 723-727. Nagata, R. T., L. M. Wilkinson, and G. S. Nuessly. 1998. Longevity, fecundity, and leaf stippling of Liriomyza trifolii (Diptera: Agromyzidae) as affected by lettuce cultivar and supplemental feeding. J. Econ. Entomol. 91: 999-1004. Nuessly, G. S., and R. T. Nagata. 1993. Evaluation of damage by serpentine leafminer and banded cucumber beetles to cos lettuce. Everglades Res. and Ed. Center Res Rpt., EV-1 993-2: 76-77. Nuessly, G. S., and R. T. Nagata. 1994. Differential probing response of serpentine leafminer, Liriomyza trifolii (Burgess), on cos lettuce. J. Entomol. Sci. 29: 330-338. Nugent, P. E., F. P. Cuthbert, Jr., and J. C. Hoffman. 1984. Two genes for cucumber beetle resistance in muskmelon. J. Am. Soc. Hort. Sci. 109: 756-759.

PAGE 136

127 Oghiakhe, S. 1995. Effect of pubescence in cowpea resistance to the legume pod borer Maruca testulalis (Lepidoptera: Pyralidae). Crop Protection 14: 379-387. Orians, C. M. 1995. Preserving leaves for tannin and phenolic glycoside analyses: a comparison of methods using three willow taxa. J. Chem. Ecol. 21: 1235-1243. Ortman, E. E., and D. C. Peters. 1980. Introduction, pp. 3-13. In F. G. Maxwell and P. R Jennings [eds.], Breeding Plants Resistant to Insects. Wiley, New York. Overman, J. L., and L. E. MacCarter. 1972. Evaluating seedlings of cantaloupe for varietal nonpreference-type resistance to Diabrotica spp. J. Econ. Entomol. 65: 1140-1144. Painter, R. H. 1951. Insect Resistance in Crop Plants. University of Kansas Press, Lawrence, Kansas. Panda, N., and G. S. Khush. 1995. Host Plant Resistance to Insects. CAB International, Wallingford, Oxon, U.K. Parrott, W. L. 1990. Plant resistance to insects in cotton. Florida Entomol. 73: 392-396. Pelletier, Y., and A. Smilowitz. 1990. Effect of trichome B exudates of Solanum berthaultii Hawkes on consumption by the Colorado potato beetle, Leptinotarsa decemlineata (Say). J. Chem. Ecol. 16: 1547-1555. Peterson, J. K., and J. M. Schalk. 1985. Semiquantitative bioassay for levels of cucurbitacins using the banded cucumber beetle (Coleoptera: Chrysomelidae). J. Econ. Entomol. 78: 738-741. Pilipenko, L. N., A. A. Kolesnik, and A. A. Todorova. 1994. Surface lipids of leafy vegetables. Technological Institute of Food Industry 29: 591-596. Pitre, H. N., and E. J. Kantack. 1962. Biology of the banded cucumber beetle, Diabrotica balteata, in Louisiana. J. Econ. Entomol. 55:904-906. Pollard, A. J. 1992. The importance of deterrence: responses of grazing animals to plant variation, pp. 216-239. In R. S. Fritz and E. L. Simms [eds.]. Plant Resistance to Herbivores and Pathogens: Ecology, Evolution and Genetics. University of Chicago Press, Chicago, Illinois. Quiring, D. T., P. R. Timmins, and S. J. Park. 1992. Effect of variations in hooked trichome densities of Phaseolus vulgaris on longevity of Liriomyza trifolii (Diptera: Agromyzidae) adults. Environ. Entomol. 21: 1375-1361. Rahbe, Y., B. Delobel, J. Guillaud and C. Nardon. 1990. Storage proteins in Coleoptera: a new class of tyrosine-rich proteins from the pupae of two weevils, Sitophilus oryzae

PAGE 137

and Rhynchophorus palmarum (Coleoptera: Curculionidae). Insect Biochem. 20: 331-341. Raina, A. K., P. S. Benepal, and A. Q. Sheikh. . 1980. Effects of excised and intact leaf methods, leaf size, and plant age on Mexican bean beetle feeding. Ent. Exp. Appl. 27: 303-306. Rees, S. B., and J. B. Harbome. 1985. The role of sesquiterpene lactones and phenolics in the chemical defense of the chicory plant. Phytochem. 24: 2225-2231. Regnier, F. E., and J. H. Law. 1986. Insect pheromones. J. Lipid Res. 9: 541-551. Reinink, K., F. L. Dieleman, J. Jansen, and A. M. Montenarie. 1989. Interactions between plant and aphid genotypes in resistance of lettuce to Myzus persicae and Macrosiplmm euphorbiae. Euphytica 43: 215-222. Risch, S. J. 1985. Effects of induced chemical changes on interpretation of feeding preference tests. Entomol. Exp. Appl. 39: 81-84. Roberts, M. F. 1987. Papaver latex and alkaloid storage vacuoles, pp. 513-528. In B. Marin [ed.], Plant Vacuoles: Their Importance in Solute Compartmentation in Cells and Their Applications in Plant Biotechnology. Plenum, New York. Roessingh, P., E. Stadler, J. Hurter, and T. Ramp. 1992. Oviposition stimulant for the cabbage root fly: important new cabbage leaf surface compound and specific tarsal receptors, pp. 141-142. /« B. J. Menken, J. H. Visser and P. Harrenwijn [eds.], Insect-Plant Relationships. Proceedings of the 8th International Symposium, Kluwer Acad. Publ., Dordrecht. Ryder, E. J. 1998. Lettuce, Endive and Chicory. CABI Publishing, Cambridge, UK. Saba, F. 1970. Host plant spectrum and temperature limitations of Diabrotica balteata. Can. Entomol. 102: 684-691. Sams, D. W., F. I. Lauer, and E. B. Radcliffe. 1975. Excised leaflet test for evaluating resistance to green peach aphid in tuber-bearing Solarium germplasm. J. Econ. Entomol. 68: 607-609. SAS histitute. 1999. Guide for Personal Computers, Version 6, SAS Institute, Cary, NC. Schalk, J. M. 1986. Rearing and handling of Diabrotica balteata, pp. 49-56. In J. L. Krysan and T. A. Miller [eds.]. Methods for the Study of Pest Diabrotica, Springer, New York. Schalk, J. M., and A. Jones. 1982. Methods to evaluate sweet potatoes for resistance to the banded cucumber beetle in the field. J. Econ. Entomol. 75: 76-79.

PAGE 138

129 Schalk, J. M., A. Jones, and P. D. Dukes. 1986. Factors associated with resistance in recently developed sweet potato cultivars and gemiplasm to the banded cucumber beetle, Diabrotica balteata LeConte. J. Agric. Entomol. 3: 329-334. Schalk, J. M., J. R. McLaughlin and J. H. Tumlinson. 1990. Field response of feral male banded cucumber beetles to the sex pheromone 6, 1 2-dimethylpentadecan-2-one. Florida Entomol. 73: 292-297. Schoonhoven, L. M. 1972. Secondary plant substances and insects. Rec. Adv. Phytochem. 5: 197-224. Schoonhoven, L. M. 1982. Biological aspects of antifeedants. Entomol. Exp. Appl. 31: 57-69. Schoonhoven, L. M., T. Jermy, and J. J. A. van Loon. 1998. Chapter 3: Plant chemistry: endless variety. Insect-Plant Biology. Chapman & Hall, London, UK. Sen, D. N., and D. D. Chawan. 1972. Leafless Euphorbia on Rajasthan (India) rocks. Vegetatio 24: 193-214. Sengonca, C. and S. Gerlach. 1984. The influence of leaf surface features on the effectiveness of the predatory thrips, Scolothrips longicornis (Thysanoptera: Thripidae). Entomophaga 29: 55-61. Sharpies, G. C. 1964. Polyphenol content of head lettuce. J. Am. Soc. Hort. Sci. 84: 356363. Siddiqui, M. A., and M. Alam. 1990. Potential of plant latex as a botanical nematicide, IPM Practitioner 9: 6-8. Simpson, S. J., and C. L. Simpson. 1990. The mechanisms of nutritional compensation by phytophagous insect, pp. 1 1 1-160. /« E. A. Bemays [ed.], Insect-Plant Interactions, vol. 2, CRC Press, Boca Raton, FL. Smith, C. M. 1989. Plant Resistance to Insects: A Fundamental Approach. John Wiley & Sons, New York. Slansky, F. 1992. Allelochemical-nutrient interactions in herbivore nutritional ecology, pp. 135-1 73. In G. A. Rosenthal and M. R. Berenbaum [eds.], Herbivores: Their hiteractions with Secondary Plant Metabolites, vol. 2: Evolutionary and ecological processes. Academic Press, Inc. San Diego. Southwood, S. R. 1986. Plant surfaces and insects an overview, pp. 1-22. In B. Juniper and S. R. Southwood [eds.], hisects and the Plant Surface. Edward Arnold, London. Spilatro, S. R., and P. G. Mahlberg. 1986. Latex and laticifer starch content of developing leaves oi Euphorbia pulcherrima. Am. J. Bot. 73: 1312-1318.

PAGE 139

130 Stadler, E. and H. R. Buser. 1983. Defense chemicals in leaf surface wax synergistically stimulate oviposition by a phytophagous insect. Experientia 40: 11 57-1 159. Stipanovic, R. D. 1983. Function and chemistry of plant trichomes and glands in insect resistance, pp. 69-100. In P. A. Hedin [ed.], Plant Resistance to Insects. ACS Symposium Series 208, Washington, D.C. Stoner, K. A. 1996. Plant resistance to insects: a resource available for sustainable agriculture. Biol. Agric. Hort.l3: 7-38. Stork, N. E. 1980. Role of waxblooms in preventing attachment to brassicas by the mustard beetle, Phaedon cochleariae. Entomol. Exp. et Appl. 28: 100-107. Stout, M. J., and S. S. Duffey. 1996. Characterization of induced resistance in tomato plants. Entomol. Exp. Appl. 79: 273-283. Stout, M. J., K. V. Workman, J. S. Workman, and S. S. Duffey. 1998. Temporal and ontogenetic aspects of protein induction in foliage of the tomato, Lycopersicon esculentum. Biochem. Syst. Ecol. 24: 611-625. Tallamy, D. W., and M. J. Raupp. 1991. Phytochemical Induction by Herbivores. John Wiley & Sons, Inc., New York. Tarksley, S. D., and T. J. Orton. 1983. Isozymes in Plant Genetics and Breeding, Part. A. Elsevier, Amsterdam. Teng, H. J. 1983. Host preference of the banded cucumber beetle Diabrotica balteata LeConte, with emphasis on selecting a trap to be used in potato production. M.S. thesis. University of Florida, Florida. Thaler, S. J., and R. Karban. 1997. A phylogenetic reconstruction of constitutive and induced resistance in Gossypium. Am. Nat. 149: 1 139-1 146. Tuomi, J., P. Neimela, M. Rousi, and S. T. Vuorisalo. 1988. Induced accumulation of foliage phenols in mountain birch: branch response to defohation Am Nat 132602-608. Valle, M. G., G. Appendino, G. M. Nano, and V. Picci. 1987. Prenylated coumarins and sesquiterpenoids from Ferula communis. Phytochemistry 26: 253-256. van Emden, H. F., and M. A. Bashford. 1976. The effect of leaf excision on the performance of Myzus persicae and Brevicoryne brassicae in relation to the nutrient treatment of the plants. Physiol. Entomol. 1: 67-71. van Helden, M., and W. F. Tjallingii. 1993. Tissue localization of lettuce resistance to the aphid Nasonovia hbisnigri using electrical penetration graphs. Entomol Exd AddI 68: 269-278. '

PAGE 140

131 van Helden, M., P. N. F. Hanneke, V. Heest, A. V. B. Teris, and W. F. Tjallingii. 1994a. Development of a bioassay to test phloem sap samples from lettuce for resistance to Nasonovia ribisnigri (Homoptera, Aphididae). J. Chem. Ecol. 21: 761-775. van Helden, M., W. F. Tjallingii, and T. A. van Beek. 1994b. Phloem sap collection from lettuce {Lactuca sativa L.): chemical comparison among collection method. J. Chem. Ecol. 20: 3191-3206. van Sumere, C. F., H. Geiger, D. Bral, G. Fockenier, K. Vande Casteele, M. Martens, R. Hanselaer, and L. Gevaert. 1983. Freeze-drying and analyses of plant and other biological material. Anal. Biochem. 13: 530-532. Wolfenbarger, D. O. 1963. The banded cucumber beetle and its control. J. Econ. Entomol. 56: 770-773. Woodhead, S. 1982. r-Hydroxybenzaldehyde in the surface wax of sorghum: its importance in seedling resistance to acridids. Ent. Exp. Appl. 31 : 296-302. Woodhead, S. 1983. Surface chemistry of Sorghum bicolor and its importance in feeding by Lociista migratoria. Physiol. Entomol. 8: 345-352. Woodhead, S., and E. Bemays. 1977. Changes in release rates of cyanide in relation to palatability of Sorghum to insects. Nature 270: 235-236. Woodhead, S. and R. F. Chapman. 1986. Insect behaviour and the chemistry of plant surface waxes, pp. 123-135. In B. Juniper, and S. R. Southwood [eds.], hisects and the Plant Surface. Edward Arnold, London. Woodhead, S., and D. E. Padgham. 1988. The effect of plant surface characteristics on resistance of rice to the brown planthopper, Nilaparvata lugens. Entomol. Exp Appl. 47: 15-22. Yencho, G. C, J. A. A. Renwick, J. C. Steffens, and W. M. Tingey. 1994. Leaf surface extracts ofSolanum berthaiiltii Hawkes deter Colorado potato beetle feeding. J. Chem. Ecol. 20: 991-1007. Zalucki, M. P., and L. P. Brower. 1992. Survival of first instar larvae ofDanaus plexippus (Lepidoptera: Danainae) in relation to cardiac glycoside and latex content of Asclepios humistrata (Asclepiadaceae). Chemoecol. 3: 81-93. Zalucki, M. P., and S. B. Malcolm. 1999. Plant latex and first-instar monarch larval growth and survival on three North American milkweed species J Chem Ecol 251827-184.

PAGE 141

BIOGRAPHICAL SKETCH Juan Huang was bom on June 8, 1967, in Shanghai, P. R. China. She received her bachelor's degree in agronomy from Shanghai Agricultural College in 1990. At the same year, she was admitted to Shanghai Institute of Entomology, Academia Sinica and received a master's degree in entomology in 1993. She worked at the same institution for three years right after her graduation. In 1 996, she was accepted at the University of Florida for her Ph.D. program and offered a graduate research assistantship to work on host plant resistance to the banded cucumber beetle under the supervision of Dr. Gregg Nuessly and Heather McAuslane in the Entomology and Hematology Department. 132

PAGE 142

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Gregg S. Nuessly, Chair Associate Professor of Entomology and Nematology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Heather J. McAuslane, Cochair Associate Professor of Entomology and Nematology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosopl Frani J. SlanskV T Professor of Entomols^ and Nematology I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Russell T. Nagata Associate Professor of Horticultural Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Anson H. Moye Professor of Food Science and Human Nutrition Department

PAGE 143

This dissertation was submitted to the Graduate Faculty of the College of Agricultural and Life Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor o f Philosophy. December 2000 y cuy^_ 1 J-^-rT ^ Dean, College of Agricultural ana^ife Sciences Dean, Graduate School