Insect Antifeedants

INSECT ANTIFEEDANTS

© 2005 by CRC Press LLC

INSECT ANTIFEEDANTS

Opender Koul Insect Biopesticide Research Centre Jalandhar, India

CRC PR E S S Boca Raton London New York Washington, D.C.

© 2005 by CRC Press LLC

Library of Congress Cataloging-in-Publication Data Koul, Opender. Insect antifeedants / by Opender Koul. p. cm. Includes bibliographical references (p. ). ISBN 0-415-33400-4 (alk. paper) 1. Insect antifeedants. 2. Biological insecticides. 3. Insecticidal plants. I. Title SB931.K786 2004 632'.7--dc22 2004051077

This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe.

Visit the CRC Press Web site at www.crcpress.com © 2005 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0-415-33400-4 Library of Congress Card Number 2004051077 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0

© 2005 by CRC Press LLC

ACKNOWLEDGMENTS I thank Prof. Sukh Dev (India), Prof. Murray B. Isman (Canada), Prof. A. J. Mordue Luntz (UK), Prof. W. Kraus (Germany), Prof. L. M. Schoonhoven (Netherlands), and Dr. Michael J. Smirle (Canada) for critical reading and valuable suggestions on various chapters in this book. I also thank my students, particularly Gurmeet Singh for editorial assistance in arranging references.

v © 2005 by CRC Press LLC

PREFACE In the recent past, the virtual dependence on neurotoxic chemicals to control pest insects has provided the impetus for studies into alternative methods of pest control that could avoid the environmental hazards associated with broad-spectrum insecticides. Overuse of synthetic insecticides that share a neurotoxic mode of action for pest management in agriculture, forestry, and managed landscapes has often induced negative impacts on natural enemies, pollinators, and other non-target organisms and often leads to the development of resistance. Fortunately new alternatives for pest control can be found within the large group of natural products or the synthetic derivatives thereof, which have the advantage of providing novel modes of action, therefore reducing the risk of cross-resistance. Naturally occurring mixtures of substances provide a multifactorial selective pressure on pests that also slows down the development of resistance. More important, research in this area has uncovered subtle but effective mechanisms of pest control, such as the behavior-modifying method of feeding deterrence. Therefore, the concept of using insect antifeedants as crop protectants is intuitively attractive. Most plant defensive chemicals discourage insect herbivory, either by deterring feeding and oviposition or by impairing larval growth, rather than by killing insects outright. One application of our understanding of plant defensive chemistry, then, is the identification of putative deterrent substances that could be isolated in sufficient quantities or synthesized for use as crop protectants. In fact, an insect antifeedant is a behavior-modifying substance that deters feeding through a direct action on peripheral sensilla in insects. This definition excludes chemicals that suppress feeding by acting on the central nervous system (following ingestion and absorption), or a substance that has sublethal toxicity to the insect. During the past three decades scores of compounds have been isolated from various natural sources, or semi-synthetic derivatives have been prepared that have the potential to inhibit feeding of a variety of insect species. However, all these studies are scattered through the biological and chemical literature, and it was felt necessary to assemble this data in the form of a comprehensive treatise on this expanding area of study and application that would aid investigators and lead them to more effective and desirable solutions to insect control. The present volume, therefore, is an attempt to compile all the data as a single text that deals specifically, as far as possible, with various aspects of insect antifeedants discussed in seven chapters. Chapter 1 introduces the subject with an emphasis on definitions and the role of antifeedants as a whole. Food selection among insect herbivores is a highly specialized phenomenon. While olfactory and physical aspects of plants or their organs can be important in insect host finding and acceptance, the choice of food is based primarily upon contact chemoreception of various allelochemicals. In particular, dietary experience has been found to influence the ability of insects to taste plant chemicals that may serve as signals of suitability or unsuitability. Certain dietary constituents appear to suppress the development of taste sensitivity to deterrents in an insect. Avoidance of allelochemicals, when looked at from a behavioral point of view, is the outcome of interactions with chemoreceptors characterized by an often-broad sensitivity spectrum of deterrents; therefore, Chapter 2 discusses the concepts and mechanisms involved in the process.

vii © 2005 by CRC Press LLC

In relation to the response of insects to these chemicals, the aspect of evaluation is important and accordingly various bioassay procedures have been developed, which are discussed in Chapter 3 and are mostly species specific. The overall picture, which emerges from various evaluations, shows that small structural variations can produce drastic changes in the activity profile of compounds. A critical examination of functional groups present in the active molecules provides crucial information about the optimal relative stereochemistry required to stimulate an antifeedant response in insects. The main aim of Chapter 4 is directed in this direction and generalizes sufficient structure-activity information within specific skeletal systems to allow rational modification of readily available feeding deterrents to be made into potential insect control agents. Commercialization aspects, practical applications, and conclusions drawn from various studies are discussed in Chapters 5 and 6. The last chapter is the monograph section, which presents relevant information on nearly 900 compounds (in alphabetical order) that is directly accessible. It has been the endeavor to give complete details on the latest structural information and biological data for those compounds that deter feeding of insects. I hope the book will prove useful to all those interested in promoting the cause of new pest control allelochemicals so that sustainability in agriculture systems and environmental protection for future generations is achieved.

viii © 2005 by CRC Press LLC

Dedicated to the memory of my mother Uma Koul

© 2005 by CRC Press LLC

ABOUT THE AUTHOR Opender Koul, Fellow of the National Academy of Agricultural Sciences and the Indian Academy of Entomology, is an insect toxicologist/physiologist/chemical ecologist and currently the Director of the Insect Biopesticide Research Centre, Jalandhar, India. After obtaining his Ph.D. in 1975 he joined the Regional Research Laboratory (CSIR), Jammu, and then became Senior Group Leader of Entomology at Malti-Chem Research Centre, Vadodara, India (1980–1988). He has been a visiting scientist at the University of Kanazawa, Japan (1985–1986), University of British Columbia, Canada (1988–1992), and Institute of Plant Protection, Poznan, Poland (2001). His extensive research experience concerns insect–plant interactions, spanning toxicological, physiological, and agricultural aspects. Honored with an Indian National Science Academy medal (INSA) and the Kothari Scientific Research Institute award, he has authored over 140 research papers and articles, and is the author or editor of the books Insecticides of Natural Origin, Phytochemical Biopesticides, Microbial Biopesticides, Predators and Parasitoids, Neem: Today and in the New Millennium, Integrated Pest Management: Potential, Constraints and Challenges, and Transgenic Crop Protection: Concepts and Strategies. He has also been an informal consultant to BOSTID, NRC of the United States at ICIPE, Nairobi.

x © 2005 by CRC Press LLC

CONTENTS 1 Introduction........................................................................................................................1 References............................................................................................................................6 2 Concepts and Mechanisms ...............................................................................................9 Chemosensory System.......................................................................................................10 Stereoselective Perception.................................................................................................11 Mechanisms .......................................................................................................................14 GABA Antagonistic Mechanism.......................................................................................15 Biogenic Amine Inhibition Mechanism ............................................................................17 Mechanisms Related to Specific Allelochemicals ............................................................17 References..........................................................................................................................19 3 Bioassays...........................................................................................................................25 Leaf Disk Assay ................................................................................................................26 Artificial Diet Feeding.......................................................................................................29 Styropor Assay ..................................................................................................................31 Glass Fiber Disk Test ........................................................................................................31 Paper Towel Disk Test.......................................................................................................31 Wafer Assay.......................................................................................................................32 Electrophysiological Assay ...............................................................................................33 Other Miscellaneous Methods...........................................................................................34 Simulation Bioassay .....................................................................................................34 Dipteran Assays ............................................................................................................35 WireWorm (Melanotus Communis) Assay...................................................................36 Boll Weevil Assay.........................................................................................................36 Scale Insect Assay ........................................................................................................36 Sawfly Assay.................................................................................................................36 Leaf Beetle Assay .........................................................................................................37 Oral Dosing...................................................................................................................37 Field Trials ....................................................................................................................38 References..........................................................................................................................38 4 Structure-Activity Relationships....................................................................................43 Limonoids ..........................................................................................................................43 Quassinoids........................................................................................................................50 Diterpenes ..........................................................................................................................53 Sesquiterpenes ...................................................................................................................56 Monoterpenes ....................................................................................................................58 Coumarins..........................................................................................................................59 xi © 2005 by CRC Press LLC

Isoflavonoids ......................................................................................................................63 Alkaloids............................................................................................................................64 Maytansinoids....................................................................................................................65 Ellagitannins ......................................................................................................................65 Aristolochic Acids .............................................................................................................67 References..........................................................................................................................68 5 Commercialization...........................................................................................................73 References..........................................................................................................................77 6 Practical Applications and Conclusions........................................................................79 Conclusions........................................................................................................................82 References..........................................................................................................................83 7 Bioefficacy Monographs..................................................................................................85

© 2005 by CRC Press LLC

1 INTRODUCTION Research over the last 35 years has provided substantial evidence that plants and animals often produce substances that affect the growth, development, behavior, and distribution of other organisms. Such naturally occurring substances are called “allelochemicals.” The term “allelochemic” was first proposed by Whitaker and Feeny (1971) and has been defined as a chemical that is significant to organisms of a species different from its source for reasons other than food as such (Nordlund, 1981). In fact, the term allelochemical comes from many terms that have been designated for chemicals that convey information between organisms; “semiochemicals” is the umbrella term, which comprises both information-conveying chemicals and toxins. Dicke and Sabelis (1988) have rightly used the term “information conveying chemicals,” as they differ from general toxins and nutrients in that the former are not themselves detrimental or beneficial, but may be through the responses they elicit. Thus using the term “infochemicals” is based on: 1. Whether the interaction is intra- or interspecific (pheromone versus allelochemical) 2. Which costs and benefits fall to each of the two interacting organisms 3. The identity of the producer and the receiver Four major categories of allelochemicals have now been recognized: the allomones, kairomones, synomones, and apneumones. However, Whitman (1988) has added another category, in which neither interactant benefits, and called them antimones. There is also a possibility that one chemical that benefits an organism in one interaction may also have side effects in other interactions (Whitman, 1988). However, all terms are context specific rather than chemical specific. To be precise, “infochemicals” in the present context is appropriate and can be characterized as in Figure 1.1.

© 2005 by CRC Press LLC

2

Opender Koul

INFOCHEMICALS

PHEROMONES

ALLELOCHEMICALS

1. (+, −) Pheromone Benefit to emitter

1. Allomone

2. Kairomone 2. (−, +) Pheromone Benefit to receiver

3. Synomone 4. Apneumone

3. (+, +) Pheromone Benefit to both

5. Antimone

FIGURE 1.1 Infochemical terminology.

The objective of this book is to explore the studies of insect antifeedants and accordingly deal with allelochemicals in general and allomones in particular. “Allomones” have been defined as substance(s) produced or acquired by an organism that, when it contacts an individual of another species in the natural context, evokes in the receiver a behavioral or physiological response that is adaptively favorable to the emitter but not to the receiver (Nordlund, 1981). As such, allomones are differentiated from pheromones because they mediate interspecific, rather than intraspecific interactions. Receiving organisms respond to allomones in a variety of ways. Subtle changes in behavior and physiology of the receiver can result in host-shifts in phytophages or parasites, or extended developmental times due to reductions in nutritional value of foodstuffs. At the other end of the spectrum, violent reactions leading quickly to injury and death are often the result of encounters with highly toxic defensive allomones. This tremendous diversity, coupled with the intensity of allomonemediated interspecific interactions, makes allomonal chemicals potential agents for insect pest control. Although allomones mediate a wide variety of complex interactions, allomonal chemicals fall into one of two basic categories. The first of these includes materials produced by the organisms and released into the environment, mostly volatile compounds that exert their influences at some distance from the emitter. For instance, it is well known that green plants release characteristic volatiles arising from the metabolism of leaf lipids such as linoleic acids, which by oxidative degradation produce a variety of 6-carbon alcohols and aldehydes. Such volatiles include a wide variety of short chain alcohol and aldehydes, ketones, esters, aromatic phenols, mono- and sesquiterpenes, and a host of other secondary metabolites. The second group of allomones includes compounds produced or acquired for defense, which remain in the body of the producer. This group includes toxins sequestered by insects for defense and the vast array of plant allelochemicals or secondary plant compounds. There are several characteristics required for the evolution of the ability of polyphagous insects (in category II) to use plant secondary metabolites as defense compounds. The main prerequisites for the evolution of unpalatability due to such compounds could be the:

© 2005 by CRC Press LLC

Insect Antifeedants • • • •

3

Feeding on a toxic plant thus having potential defense compounds in the insect’s diet Elimination of detoxification mechanism by inactivating detoxification enzymes so that defense compounds are not broken down Ability of compounds to move through the gut and reach the haemolymph Accumulation of defense compounds in the haemolymph against an osmotic gradient and retention of compounds through the stadium (Bowers, 1992)

In other words, the allomones are the phytochemicals produced from secondary metabolic pathways and are the major mechanisms by which plants are protected from excessive herbivory. Behavioral mechanisms provide a system of avoidance of non-host chemicals by which insects select their food, though the molecular basis for action of chemical deterrents on both gustatory and olfactory sensory systems in insects is only poorly understood. Among plant anti-herbivore chemistry, a strong link does not exist between feeding deterrence and internal toxicity in insects, suggesting that behavioral rejection is not an adaptation to ingested effects but more an outcome of deterrent receptors with wide chemical sensitivity (Mullin et al., 1991, 1994). Many of these substances are bitter, and acceptance of host plants by herbivores requires chemoreception of favorable levels of phagostimulants relative to antifeedants (Dethier, 1980). This restricts the application of a very liberal definition for an antifeedant, namely, “any substance that reduces consumption by an insect” to a more precise definition: “A peripherally mediated behavior-modifying substance (i.e., acting directly on the chemosensilla in general and deterrent receptors in particular) resulting in feeding deterrence” (Isman, 1994). This definition, however, excludes chemicals that suppress feeding by acting on the central nervous system (following ingestion or absorption), or a substance that has sublethal toxicity to the insect (Isman, 2002). Feeding deterrents with a wide diversity of structures are not known to directly interfere with insect taste cell responses to phagostimulants such as sugars (Lam and Frazier, 1991; Schoonhoven et al., 1992). Presently the mode of action of feeding modifying chemicals in insect gustatory systems is largely unknown (Frazier, 1992; Schoonhoven et al., 1992), though some molecular targets have been identified (Koul, 1997). Taste receptor proteins are only now beginning to be biochemically purified and cloned. The determination of the molecular basis for action of feeding deterrents in the insect gustatory system is thus a primary goal among basic and applied entomologists interested in insect–plant interactions or in the control of herbivore pests. According to the theory of biochemical coevolution it should be possible to develop an evolutionary pattern of antifeedants on the basis of their distribution in different plant families and their biosynthetic pathways. Accordingly it has been possible to draw an evolutionary scheme as shown in Figure 1.2 (modified from Harborne, 1988). However, the pattern of distribution varies among families. One plant family may concentrate on one type of deterrent molecule, like limonoids in the Rutales (Champagne et al., 1992) and also, within a family, individual members may have developed further barriers to feeding. For instance, it is clear that flavonoids in plants can modulate the feeding behavior of insects, though mechanisms associated with these behavioral responses are not clearly understood (Simmonds, 2001). Other families may diversify their deterrents; for example, non-protein amino acids (e.g., L-canavanine), alkaloids, cyanogens, and isoflavones are found in the Fabaceae. Plants produce all these and many varied compounds in the first instance as protective devices against insect feeding. Thus a majority of plant families rely on secondary plant metabolites for protection from phytophagous insects. One might surmise that within such a family the more advanced members are better protected than others. Berenbaum (1983) has pointed to

© 2005 by CRC Press LLC

4

Opender Koul Sesquiterpene Lactones Alkaloids

Flavones

Monoterpene Lactones HERBACIOUS Sesquiterpenes Cardiac Glycosides

Non-protein Aminoacids

Cucurbitacins

Limonoids

WOODY

Diterpenes Saponins

Quassinoids

Polyacetylenes

Coumarins

ANGIOSPERMS -------------------------------------------------------------------------------------------------------------------Ecdysteroids Tannins Cyanogens Juvenoids

FATTY ACID

GYMNOSPERMS FERNS

TERPENOID

PHENOLICS

NITROGEN

BIOSYNTHETIC PATHWAYS

FIGURE 1.2 Possible evolutionary scheme of insect antifeedants in plants (modified from Harborne, 1988).

good evidence in the Apiaceae where plant defense is based on hydroxycoumarins, linear furanocoumarins, and angular furanocoumarins, which are biosynthetically and toxicologically related. It is also evident from various studies that as a result of coevolutionary pressures, plants have a startling number of plant chemicals including chromenes, polyacetylenes, saponins, quassinoids, cuccurbitacins, cyclopropanoid acids, phenolics, alkaloids, various types of terpenes, and their derivatives, and each insect species may process these allomones in a thoroughly idiosyncratic way, so that the same compound may have very different fates and consequences in different species of insects (Koul, 1993; Blum et al., 1987). These various insect–plant interactions are consistent with the idea of reciprocal evolutionary interactions based on secondary metabolites. This, however, could be related to the evolution of deterrent

© 2005 by CRC Press LLC

Insect Antifeedants

5

receptors in insects, too. There is a clear indication that no two insect species are equipped with an identical sensory system. Each species has a unique sensory window, which can discriminate between host and non-host plants (Schoonhoven, 1982). Even in very closely related species the chemical senses show striking differences (Drongelen, 1979). It can be visualized from such information that the contact chemical senses may in evolutionary terms be easily adapted to changing circumstances, as has been well evidenced in two strains of Mamestra brassicae in response to sinigrin and naphtyl-β-glucoside (Wieczorek, 1976). It can also be visualized that insect feeding deterrents may be perceived either by stimulation of specialized deterrent receptors or by distortion of the normal function of neurons, which perceive phagostimulating compounds. Some sugars are very important components of an insect’s sustained feeding; the inhibition of the receptors is an effective antifeedant action. Some antifeedants influence the feeding activity through a combination of the two principal modes of action mentioned above. Initial discoveries of antifeedant chemicals were simply made by chance when organometallic compounds and a few insecticides were found to reduce insect feeding (Ascher and Rones, 1964; Jermy and Metolcsy, 1967). This clearly emphasizes the point that many synthetic compounds could be potential antifeedants for insect pests (Koul, 1993), of course in addition to the allomones or their derivatives from natural sources. As early as 1932 Metzger and Grant tested about 500 plant extracts against Popillia japonica, though results were not substantially encouraging. Pradhan et al. (1962) evaluated extracts of the Indian neem tree, Azadirachta indica, which prevented feeding by the desert locusts, and today nearly 900 compounds have been identified to possess feeding deterrence against insects (see Chapter 7), though terrestrial plants produce a diverse array of secondary metabolites, likely more than 100,000 unique compounds (Isman, 2002). In addition to various compounds isolated or synthesized as insect antifeedants, a number of studies demonstrate the antifeedant efficacy in metabolite mixtures of plant essential oils or total extracts against a variety of insect species. In recent years studies have revealed the antifeedant potential of plant essential oils against post-harvest pests, aphids, thrips, lepidopterans, termites, and mite pests (Hori, 1999; HouHouaMin et al., 2002a, 2002b; Koschier et al., 2002; Maistrello et al., 2003). Similarly, during the past few years ample emphasis has been in demonstrating the antifeedant efficacy in total plant extracts (Mancebo et al., 2000a, 2000b; Wang et al., 2000; Jannet et al., 2001; Lababidi and Koudseieh, 2001; Schlyter, 2001; Wheeler and Isman, 2001; Mehta et al., 2002; Jayasinghe et al., 2003) as they seem to exhibit the activity as multicomponent systems. However, it is also well known that antifeedants show interspecific variability (Chapman, 1974; Schoonhoven and Jermy, 1977; Isman, 1993). Such interspecific differences, as shown for many insect species, encourage the need to search selectively for specific feeding deterrents. Van Beek and deGroot (1986) have suggested three considerations for the selection of plants that are to be evaluated for antifeedant activity: 1. The species can be selected at random. 2. They could be selected on chemotaxonomic bases. 3. The species can be selected on the basis of ethnobotanical or entomological data. All three methods have been used, and the second method has led to the most frequent successes. It is, therefore, necessary to focus attention on particular chemical groups occurring naturally in various sources, or environmentally safe synthetics, and thus obtain a large number of active antifeedant compounds, which could be successfully introduced in Insect Pest Management (IPM) programs. However, simultaneous to such developments it is necessary to understand the concepts and mechanisms involved in antifeedant interactions. It is

© 2005 by CRC Press LLC

6

Opender Koul

also important to point out here that not much success has been achieved so far in establishing a good commercial antifeedant for crop protection. However, the commercial potential and the drawbacks will be discussed in Chapter 6. Recently, it has also been demonstrated that neuropeptide-based pesticides potentially offer levels of activity, specificity, and environmental compatibility absent in conventional insecticides. However, neuropeptides are generally poor candidates for insecticides because they do not easily penetrate the cuticle and degrade rapidly in the environment and insect gut. Manduca sexta allatostatin (Manse-AS) regulates juvenile hormone biosynthesis in moths and has myoregulatory action on the gut. Moreover, Manse-AS produces marked reductions in feeding and growth when injected into larvae of the tomato moth, Lacanobia oleracea. Snowdrop lectin (GNA) is detectable in the haemolymph of larvae following oral administration. To determine whether GNA could transport neuropeptides across the gut, a recombinant expression system was used to produce a GNA/Manse-AS fusion protein (FP). Following expression in Escherichia coli, purified FP was incorporated in an artificial diet and offered to tomato moth larvae. Intact FP appeared in the haemolymph following oral administration, which resulted in an almost total cessation of feeding and growth by larvae exposed to the FP diet. These results offer the possibility of developing a whole range of novel, orally active, target-specific antifeedants based on insect neuropeptides (Edwards et al., 2002).

REFERENCES Ascher, K.R.S. and Rones, G. (1964) Fungicide has residual effect on larval feeding. Int. Pest Control., 6, 6–9. Berenbaum, M. (1983) Coumarin and caterpillars, a case for coevolution. Evolution, 37, 163–179. Blum, M.S., Whitman, D.W., Severson, R.F., and Arrendale, R.F. (1987) Herbivores and toxic plants: evolution of a menu of options for processing allelochemicals. Insect Sci. Applic., 8, 459–563. Bowers, M.D. (1992) The evolution of unpalatability and the cost of chemical defense in insects. In B.D. Roitberg and M.B. Isman (eds.), Insect Chemical Ecology: An Evolutionary Approach, Chapman & Hall, New York, pp. 216–244. Champagne, D.E., Koul, O., Isman, M.B., Towers, G.H.N., and Scudder, G.G.E. (1992) Biological activity of limonoids from the rutales. Phytochemistry, 31, 377–394. Chapman, R.F. (1974) The chemical inhibition of feeding by phytophagous insects. A review. Bull. Entomol. Res., 64, 339–363. Dethier, V.G. (1980) Evolution of receptor sensitivity to secondary plant substances with special references to deterrents. Am. Nat., 115, 45–66. Dicke, M. and Sabelis, M.W. (1988) Infochemical terminology: should it be based on costbenefit analysis rather than origin of compounds? Funct. Ecol., 2, 131–139. Edwards, J.P., Fitches, E.C., Audsley, N., and Gatehouse, J.A. (2002) Insect neuropeptide fusion proteins—a new generation of orally active insect control agents. In Pests-anddiseases, Proceedings BCPC Conf., Brighton, UK, pp. 25–31. Frazier, J.L. (1992) How animals perceive secondary plant compounds. In G.A. Rosenthal and M.R. Berenbaum (eds.), Herbivores: Their Interaction with Secondary plant Metabolites, Evolutionary and Ecological Processes, 2nd edition, Vol. 2, Academic Press, San Diego, pp. 89–134. Harborne, J.B. (1988) Introduction to Ecological Biochemistry. Academic Press, New York.

© 2005 by CRC Press LLC

Insect Antifeedants

7

Hori, M. (1999) Antifeeding, settling inhibitory and toxic activities of Labiatae essential oils against the green peach aphid, Myzus persicae (Sulzer) (Homoptera:Aphididae). Appl. Entomol. Zool., 34, 113–118. Hou, HouaMin, Zhang Xing, Hou, H.M., and Zhang, X. (2002a) Effect of essential oil of plants on three lepidopterous insects: antifeeding and growth inhibition. Acta Phytophylacica Sinica, 29, 223–228. Hou, HouaMin, Feng, J.T., Chen AnLiang, Zhang Xing, Hou, H.M., Feng, J.T., Chen, A.L., and Zhang, X. (2002b) Studies of the bioactivities of essential oils against insects. Nat. Prod. Res. Develop., 14, 27–30. Huang, Y., Lam, S.L. and Ho, S.H. (2000) Bioactivities of essential oils from Elletaria cardemomum (L.) Maton to Sitophilus zeamais Motschulsky and Tribolium castaneum (Herbst.). J. Stored Prod. Res., 36, 107–117. Isman, M.B. (1993) Growth inhibition and antifeedant effects of azadirachtin on six noctuids of regional economic importance. Pestic. Sci., 38, 57–63. Isman, M.B. (1994) Botanical insecticides and antifeedants: New sources and perspectives. Pestic. Res. J., 6, 11–19. Isman, M.B. (2002) Insect antifeedants. Pestic. Outlook, 13, 152–157. Jannet, H.B., Skhiri, F., Mighri, Z., Simmonds, M.S.J., and Blaney, W.M. (2001) Antifeedant activity of plant extracts and a new natural diglyceride compounds isolated from Ajuga pseudoiva leaves against Spodoptera littoralis larvae. Industr. Crops Prod., 14, 213–222. Jayasinghe, U.L.B., Kumarihamy, B.M.M., Bandara, A.G.D., Waiblinger, J., and Kraus, W. (2003) Antifeedant activity of some Sri Lankan plants. Nat. Prod. Res., 17, 5–8. Jermy, T. and Metolcsy, G. (1967) Antifeedant effects of some systemic compounds on chewing phytophagous insects. Acta Phytopath. Acad. Sci. Hung., 2, 219–224. Koul, O. (1993) Plant allelochemicals and insect control: An antifeedant approach. In T.N. Ananthakrishanan and A. Raman (eds.), Chemical Ecology of Phytophagous Insects, IBH & Oxford Publishers Pvt. Ltd., New Delhi, pp. 51–80. Koul, O. (1997) Molecular targets for feeding deterrents in phytophagous insects. In A. Raman (ed.), Ecology and Evolution of Plant Feeding Insects in Natural and Man-Made Environments, International Scientific Publications, New Delhi, pp. 123–134. Koschier, E.H., Sedy, K.A., and Novak, J. (2002) Influence of plant volatiles on feeding damage caused by the onion thrips, Thrips tabaci. Crop Protection, 21, 419–425. Lababidi, M.S. and Koudseieh, S. (2001) Laboratory evaluation of the biological activity of several plant extracts against adults of the two-spotted spider mite, Tetranychus urticae Koch (Acari:Tetranychidae). Arab J. Plant Prot., 19, 86–91. Lam, P.Y.-S. and Frazier, J.L. (1991) Rational approach to glucose taste chemoreceptor inhibition as novel insect antifeedants. In D.R. Baker, J.G. Fenyes, and W.K. Moberg (eds.), Synthesis and Chemistry of Agrochemicals II, ACS Symp. Ser. 443, American Chemical Society, Washington, D.C., pp. 400–412. Maistrello, L., Henderson, G., and Laine, R.A. (2003) Comparative effects of vetiver oil, nootkatone, and disodium octaborate tetrahydrate on Coptotremes formosanus and its symbiotic fauna. Pest Managm. Sci., 59, 58–68. Mancebo, F., Hilje, L., Mora, G.A., and Salazar, R. (2000a) Antifeedant activity of plant extracts on Hypsipyla grandella larvae. Rev. Fores. Centroamericana, 31, 11–15. Mancebo, F., Hilje, L., Mora, G.A., and Salazar, R. (2000b) Antifeedant activity of Quassia amara (Simaroubaceae) extracts on Hypsipyla grandella (Lepidoptera:Pyralidae) larvae. Crop Prot., 19, 301–305.

© 2005 by CRC Press LLC

8

Opender Koul

Mehta, P.K., Sood, A.K., Parmar, S., and Kashyap, N.P. (2002) Antifeedant activity of some plants of North-Western Himalayas against cabbage caterpillar, Pieris brassicae (L.). J. Entomol. Res., 26, 51-54. Metzger, F.W. and Grant, D.H. (1932) Repellency of the Japanese beetle of extracts made from plants immune to attack. Tech. Bull. USDA No. 299, 21 pp. Mullin, C.A., Mason, C.H., Chou, J., and Linderman, J.R. (1991) Phytochemical antagonism of γ-aminobutyric acid based resistance in Diabrotica. In C.A. Mullin and J.G. Scott (eds.), Molecular Mechanisms of Insecticide Resistance: Diversity Among Insects, ACS Symp. Ser. 505, American Chemical Society, Washington, D.C., pp. 288–308. Mullin, C.A., Chyb, S., Eichenseer, H., Hollister, B., and Frazier, J.L. (1994) Neuroreceptor mechanism in insect gustation, a pharmacological approach. J. Insect Physiol., 40, 913–931. Nordlund, D.A. (1981) Semiochemicals: A review of the terminology. In D.A. Nordlund, R.L. Jones, and W.J. Lewis (eds.), Semiochemicals, Their Role in Pest Control, Plenum Press, New York, pp. 13–28. Pradhan, S., Jotwani, M.S., and Rai, B.K. (1962) The neem seed deterrent to locusts. Indian Farming, 12, 7–11. Schoonhoven, L.M. (1982) Biological aspects of antifeedants. Entomol. Exp. Appl., 31, 57–69. Schoonhoven, L.M. and Jermy, T. (1977) A behavioural and electrophysiological analysis of insect feeding deterrents. In N.R. Mcfarlane (ed.), Crop Protection-Their Biological Evaluation, Academic Press, London, pp. 133–146. Schoonhoven, L.M., Blaney, W.M., and Simmonds, M.S.J. (1992) Secondary coding of feeding deterrents in phytophagous insects. In E.A. Bernays (ed.), Insect Plant Interactions, Vol. 4, CRC Press, Boca Raton, Florida, pp. 59–79. Schlyter, F. (2001) Antifeedants as plant protection against Hylobius pine weevils. Vaxtskyddsnotiser, 65, 47–53. Simmonds, M.S.J. (2001) Importance of flavonoids in insect–plant interactions: feeding and oviposition. Phytochemistry, 56, 245–252. Van Beek, T.A. and de Groot, A.C. (1986) Terpenoid antifeedants Part I. An overview of terpenoid antifeedants of natural origin. Recuril des Trav. Chimiq. Des Pays-Bas, 105, 513–527. Van Drongelen, W. (1979) Contact chemoreception of host plant specific chemicals in larvae of various Yponomenta species (Lepidoptera). J. Comp. Physiol., 134A, 265–279. Wang, S.F., Liu, A.Y., Ridsdill-Smith, T.J., and Chisalberti, E.L. (2000) Role of alkaloids in resistance of yellow lupin to red legged earth mite Halotydeus destructor. J. Chem. Ecol., 26, 429–441. Wheeler, D.A. and Isman, M.B. (2001) Antifeedant and toxic activity of Trichilia americana extract against the larvae of Spodoptera litura. Entomol. Exp. Appl., 98, 9–16. Whitaker, R.H. and Feeny, P.P. (1971) Allelochemics: chemical interactions between species. Science, 171, 757–770. Whitman, D.W. (1988) Allelochemical interactions among plants, herbivores and their predators. In P. Barbosa and D.K. Letourneau (eds.), Novel Aspects of Insect–Plant Interactions, John Wiley, pp. 11–64. Wieczorek, H. (1976) The glycoside receptor of the larvae of Mamestra brassicae L. (Lepidoptera: Noctuidae). J. Comp. Physiol., 106A, 153–176.

© 2005 by CRC Press LLC

2 CONCEPTS AND MECHANISMS Food selection among insect herbivores is a highly specialized phenomenon. While olfactory and physical aspects of plants or their organs can be important in insect host finding and acceptance (Miller and Strickler, 1984), the choice of food is primarily based upon contact chemoreception of various allelochemicals (Frazier, 1986; Stadler, 1992). In particular, dietary experience has been found to influence the ability of insects to taste plant chemicals that may serve as signals of suitability or unsuitability. Certain dietary constituents appear to suppress the development of taste sensitivity to deterrents in an insect (Renwick, 2001). Avoidance of allelochemicals, when looked at from a behavioral point of view, is the outcome of interactions with chemoreceptors characterized by an often-broad sensitivity spectrum of deterrents (Mullin et al., 1994). According to Schoonhoven et al. (1992) there are four basic reasons why the chemosensory perception of feeding deterrents by phytophagous insects warrants special attention: 1. Feeding deterrents are apparently more important in host-plant recognition than phagostimulants. 2. A huge number of feeding deterrents exist with variable molecular structures adding to their diversity. 3. There are fewer deterrent receptors. 4. Different deterrents may elicit different behavioral reactions, indicating the presence of a differential sensory coding system. On the whole the mode of action of feeding modifying chemicals in insect chemoreceptor systems is largely unknown, and no biochemically purified or cloned taste receptor proteins

© 2005 by CRC Press LLC

10

Opender Koul

have been identified. However, a number of molecular targets for feeding deterrents have been identified (Koul, 1997), and there is evidence to show the existence of several sensory mechanisms involved. Therefore, to understand the concepts and mechanisms of feeding deterrents in an insect gustatory system, a search for candidate neuroreceptors and various behavioral end points is required. To achieve this, at first place one must look into the chemosensory equipment involved in the process.

CHEMOSENSORY SYSTEM The surface of the insect body is richly supplied with sensilla of various shapes and densities. The sensillum is the structural unit from which the majority of insect sensory organs are derived. Ectodermal in origin, a sensillum develops by differentiation from a mother epidermal cell. It consists of cuticular parts, one or more sense cells, and two or more sheath cells. The sense cells vary in number from 1 to 40 or more and have large nuclei located below the epidermis. These bipolar sense cells send their dendrites to the cuticular parts where their form, ultrastructural features, and methods of attachment are characteristic for cells of different modalities. Their axons extend into the sensory nerve parallel with other sensory axons, often extending directly to the central nervous system (CNS) before making synaptic connections to second-order neurons. Thus, they are primary sense cells that contain both a sensory receptor area on their dendrites and an impulse-conducting membrane along their axons. Usually sheath cells vary in number and are of three types: the basal, the outer, and the inner sheath cells. These cells have tight and gap junctions among them and form a sort of insulating barrier between the extracellular space surrounding the dendrites and the haemolymph space below the epidermis (Kuppers and Thurm, 1982). The cuticular projections of insect sensilla are the most visible portions, and their size, shape, and position have been the basis for classifying them. With various microscopic examinations and impulse recording techniques, various features of structure and function have been demonstrated. Insect sensilla on the outside of the body consist of the major types based on shape of the cuticular part, the presence or absence of pores, and the type of attachment to the cuticle (Frazier, 1985). They have been classified as: • • • • • •

Sensillum in a flexible socket with a single sense cell containing a tubular body Sensillum without a flexible socket containing a sense cell with lamellated dendrite Uniporous sensillum in a flexible socket containing one cell with a tubular body and one or more cells with dendrites extending to the terminal pore Uniporous sensillum without a flexible socket containing two or more cells with unbranched dendrites Multiporous sensillum with a single wall and multiple cells with branched dendrites Multiporous sensillum with a double wall and multiple cells with unbranched dendrites

Out of these six major types the sensilla that possess only a single terminal pore (thickwalled) are of gustatory nature and are concentrated on the mouth parts, though taste hairs also occur on tarsae, antennae, and ovipositors. They possess flexible sockets; 2 to 20 sensory cells, 1 dendrite with tubular unbranched body or inflexible sockets; 2 to 9 sensory cells and unbranched dendrites. They are usually uniporus. However, uniporous sensilla with inflexible sockets are fewer in number, but dome-shaped sensilla occur often in the preoral cavity, where they serve to monitor the food being eaten.

© 2005 by CRC Press LLC

Insect Antifeedants

11

Lepidopterous larvae have been observed to carry in each maxilla a palpus and a galea, the latter carrying two sensilla styloconica, non-socketed pegs with an apical papilla. These taste hairs are innervated by four bipolar neurons, the dendrites of which extend through the length of the hollow cuticular peg ending just below the pore at the tip (i.e., within a few milliseconds of diffusion time from the external chemical environment) (Schoonhoven, 1987; Descoins, 2001). The tip of the maxillary palp is covered with eight sensilla basiconica. The palp tip sensilla are innervated by 14 to 19 neurons in total (Schoonhoven and Dethier, 1966). This number, however, varies in different insect species (Devitt and Smith, 1982). As most of these sensilla are gustatory in nature, they are also involved in food recognition (Descoins, 2001). Palpation of the intact leaf surface, prior to biting activity, is related to contact chemoreception during which chemicals on the leaf cuticle are perceived (Devitt and Smith, 1985). An epipharyngeal taste sensillum in Leptinotarsa decemlineata larvae was studied using electron microscopy, which showed that the sensillum is innervated by five neurons. Electrophysiological experiments showed that one of these cells responds to water, a second to sucrose, and a third to two feeding deterrents that were also effective in a behavioral test. The response of the sucrose-sensitive cell was strongly inhibited by one of the two feeding deterrents and only slightly by the other feeding deterrent. It was concluded that probably both the response of the deterrent cell and peripheral interactions exerted by feeding deterrents on the sucrose-sensitive cell determine the potency of feeding deterrents. These results provide a physiological basis for the hypothesis that the presence or absence of feeding deterrents in potential food plants is a decisive cue in food plant selection by L. decemlineata larvae (Messchendorp et al., 1998). However, differential neurosecretory response of this insect species has also been recorded, for instance, against glycoalkaloids (Hollister et al., 2001). Thus one can easily surmise that gustatory chemosensilla must be regulating feeding behavior. It is obvious that many cells furnish information during the feeding sequence. In grasshoppers, for instance, receptor complement is large in number and low in specificity, and in caterpillars the number is low and relatively high in specificity (Frazier, 1986). In both extremes there is, however, redundancy among chemosensory cells, both with respect to specificity as well as overlap of sensitivity ranges of individual receptor cells (Blom, 1978). Obviously it is vital to have extensive and dependable information about plant allelochemicals that reduce or inhibit feeding. This link between single chemosensory cell input and behavioral output must be known before we are able to correlate the effects of allelochemicals on single cells in electrophysiological studies with their effects on the feeding behavior of the whole insect (Frazier, 1986).

STEREOSELECTIVE PERCEPTION Antifeedant properties of a plant compound may be revealed either by direct observation or by using electrophysiological methods that need thorough understanding of an insect’s chemoreceptory system. The latter procedure provides information on sensory mechanisms underlying the perception of antifeedant chemicals. However, no two insects possess fully identical chemoreceptory systems, but rather show different responses to various stimuli. Consequently, plant compounds may evoke different behavioral reactions even in closely related insect species (Schoonhoven, 1987). According to Schoonhoven (1988) life at a macroscopic scale usually presents itself in symmetrical forms. At the molecular level, however, asymmetry prevails. Nature often produces only one type of a stereospecific molecule and not its stereoisomer(s). Since chemoreception is a process of molecular interactions, the phenomenon of stereoisomerism may have consequences for the process of stimulus

© 2005 by CRC Press LLC

12

Opender Koul

recognition. Therefore, the question arises: What is the role of stereospecificity of insect chemoreceptors vis-a-vis antifeedants? As mentioned above, the sense of taste in insects is localized in specialized receptors on the mouthparts, on the preoral cavity, on the tarsi, and on the antennae—often at several of these sites in the same insect. Extensive studies performed mainly on blowfly (Dethier, 1976) and lepidoteran larvae (Schoonhoven, 1987) have shown that receptors are usually not highly specific, and responses could be multineural. A correlation of the electrophysiological response with behavioral discrimination in caterpillars has provided evidence supporting the idea that patterns of multireceptor activity constitutes the basic code for recognition and discrimination. The sensory code may be altered due to the stimulation of specialized receptors or modulation of the activity of receptors tuned to other compounds. In lepidopteran larvae several specialized deterrent receptors have been described that respond to various alkaloids, phenolic compounds, and glycosides and that inhibit food intake. The deterrent receptors in different species often overlap in their sensitivity spectra, but show at the same time characteristic interspecific variations (Schoonhoven, 1982). Feeding deterrents may also change the activity of receptors that signal the presence of feeding stimulants, for instance when suppressing sugar receptors, and thereby act as strong antifeedants (Kennedy and Halpern, 1980). Azadirachtin, a terpenoid isolated from the neem tree, stimulates a deterrent receptor in a number of herbivorous insects (Schoonhoven, 1988), but appears to suppress sugar and inositol receptors in other species (Schoonhoven, 1988). On the whole, several specialized deterrent receptors have been described mainly in lepidopteran larvae. For instance, Bombyx mori possesses a bitter receptor that is located in one of the two sensilla styloconica on the maxilla and responds to various alkaloids acting as feeding inhibitors (Ishikawa, 1966) or responds to limonoid inhibitors, in the case of Helicoverpa armigera and H. assulta (Tang et al., 2000). Pieris brassicae larvae and several other lepidopteran species have one or more deterrent receptors, which overlap in their sensitivity spectra (Schoonhoven, 1982; Chapman, 1982). Colorado beetles also have deterrent receptors in their tarsal sensilla, responding to various solanaceous plant alkaloids (Sturckow, 1959). Specific deterrent receptors are also present in the preoral cavity of lepidopteran larvae (Ma, 1972; de Boer et al., 1977). Electrophysiological studies of Blaney (1980) emphasize the fact that deterrent receptors cannot be of a single and simple category. Therefore, even today the conclusion of Dethier (1980) that in insects with few receptors, multiple receptor sensitivity occurs and that “there is no generalized deterrent receptor,” seems to be highly plausible. As it is clear now that deterrent receptors vary from species to species, it won’t be an exaggeration to conclude that the contact chemical senses may in evolutionary terms be easily adapted to changing circumstances (Schoonhoven, 1982). Receptor sensitivity and specificity, however, is genetically determined, and changes in them apparently occur by a gradual replacement of certain receptor sites in the dendritic membrane by other types of sites, which bind different stimulants. For example, Wieczorek (1976) showed that the deterrent cell in two strains of Mamestra brassicae show quantitative differences in their response to some chemicals (Table 2.1), which may be explained by a different ratio between two types of receptor sites present in the receptor membrane. As shown in Table 2.1, one strain is very sensitive due to the presence of many sinigrin receptor sites, whereas the other strain is more easily stimulated by naphtyl-β-glucoside (Wieczorek, 1976). This is consistent with the model of Bernays and Chapman (1994), which suggests that differences in taste sensitivity to deterrent compounds could account for the difference in host range. It is also possible that diet breadth has a direct link with sensitivity of the deterrent receptor cells. For instance, genetic differences in the sensitivity of the deterrent

© 2005 by CRC Press LLC

Insect Antifeedants

13

TABLE 2.1 Sensitivity (in impulses/sec.) of a deterrent receptor to a standard concentration of sinigrin and 1-naph-β-glucoside in two strains of Mamestra brassicae (calculated from Wieczorek, 1976). Strain

Response

1 Sinigrin

16–21 Imp. Freq.

2

21–30

1 1-naph-β-glucoside

38–46 Imp. Freq.

2

23–29

receptor cells of Bombyx mori in relation to diet breadth (Asaoka, 1994) imply that the effect could not be peripheral; however, the same interpretation does not hold true for Heliothis species and suggests central nervous system mediated differences (Bernays et al., 2000). Reduced feeding on deterrent diets is, in fact, a consequence either of rejection without any ingestion or of rejection following some ingestion. Rejection without ingestion indicates that deterrent compound is detected by chemoreceptors on the mouthparts. Rejection following some ingestion apparently results from the accumulation of sensory information, since deterrent receptors sometimes adapt relatively slowly (Schoonhoven et al., 1998). There could be post-ingestive feedbacks that allow limited intake (Bernays et al., 2000). An intriguing question concerns the origin of deterrent receptors. It has been suggested that herbivorous insects, rather than evolving receptors for some specific deterrents, have developed from a “common chemical sense,” resulting in a receptor type that is sensitive to a wide variety of compounds, even including chemicals to which a particular species has never been exposed before (Dethier, 1980). It may be concluded from state-of-the-art studies that insect deterrent receptors cannot be considered as a primitive or uniform type of receptor, but rather as compound receptor types with a high degree of plasticity. According to Schoonhoven (1982) this plasticity on the one hand insures that the insect may quickly adapt to changes in its environment, but maintain the capacity to recognize unpalatable plants, and on the other hand, has led to considerable divergence resulting in no two insects being identical. In terms of CNS interpretation of the sensory code, feeding activity obviously requires motor output from the CNS, whereas the presence of feeding deterrents signaled via chemosensory input may inhibit feeding motor output, leading to a refusal to eat (Ma, 1972). Presently it is difficult to study the process as underlying the evaluation of sensory input by the CNS, and resulting in either continuation or cessation of feeding activity. However, the sensory inputs can be analyzed and the principles upon which central neural integration is based can be hypothesized. Some basic considerations put forth are: • • •

The gustatory sense has a leading role in feeding activity. The epipharyngeal organs do not add new information to that of the maxillary hairs. Sensory input from the maxillae is sent to the suboesophageal ganglion, that from the epipharyngeal organ to the tritocerebrum. The message indicating whether a plant is acceptable or not must be hidden in the sensory pattern it evokes (Schoonhoven, 1987).

© 2005 by CRC Press LLC

14

Opender Koul

If the CNS is able to read this message, it is in principle also decipherable to us, and accordingly some messages may permit feeding activity and others may not. Sensory coding of feeding deterrents is based upon neural activity in one or more neurons. Three basic types of sensory coding are known (Schoonhoven et al., 1992): 1. Labeled lines: Each neuron conveys a specific message, which can be understood by the CNS without additional information from other neurons. 2. Across-fiber patterns: The message is contained in a neural activity pattern, transmitted by two or more receptors, possessing different stimulus spectra. 3. Temporal patterns: Stimulus quality affects nerve impulse interval patterns and adaptation rates, which may contain additional information. These coding principles could be cited in several cases and often occur in combination in insects (Dethier and Crnjar, 1982; Schoonhoven and Blom, 1988). A temporary distortion of such sensory codes can result in the inhibition of feeding. When in Leptinotarsa decemlineata the responses were compared between host and non-host potato saps. The response patterns for the non-host stimuli appeared to be considerably less consistent than the patterns evoked by the sap from the host plant (Mitchell et al., 1990; Schoonhoven et al., 1992). This suggests that such variable patterns are interpreted by the CNS as “nonsense,” with the result that no feeding or only limited feeding occurs, a pattern that has also been observed in various lepidopteran larvae (Simmonds and Blaney, 1990). Several chemicals, including some heavy metal ions, may distort the functioning of chemoreceptors in such a way that, even in the presence of an acceptable plant, the neural acceptance profile that the CNS requires for initiating feeding behavior is not evoked (Schoonhoven, 1987; Schoonhoven and Jermy, 1977).

MECHANISMS Secondary plant substances are in principle noxious because they interfere with the normal structure and function of insect cells and thus disturb their integrity. Thus insects, like other animals, have developed various mechanisms to reduce or prevent harmful effects of secondary plant substances when contacting them or after ingesting them (Brattsten and Ahmad, 1986). As we have seen, chemoreceptors in insects are primary sense cells and thus true neurons generally protected from the deleterious effects of secondary plant compounds. This is supported by the fact that insects have sensory neurons that respond to sugars, amino acids, or salts and function normally despite the presence of these host-specific noxious compounds, as was demonstrated in the case of polyhydroxy alkaloids against Spodoptera and Helicoverpa species (Simmonds et al., 1990). If some receptor cells have retained their primordial sensitivity to different kinds of secondary plant compounds, they would be ideally suited to signal the presence of chemicals to be avoided. Thus, the primitive, unmodified taste cell may be considered as the primordial deterrent receptor, which still possesses a sensitivity to odd plant substances originally shown by all primitive neurons. That does not mean that the present-day deterrent receptors are unchanged and wholly identical to their ancestral neural cell type. The modern deterrent receptors, while retaining sensitivity to various secondary plant compounds, have developed a physiological mechanism, which protects them against the harmful effects of their adequate stimuli. Not only has the basic sensitivity to secondary plant substances been preserved in these receptors, it also became connected to the action potential generating system, resulting in a change of impulse frequency upon stimulation (Schoonhoven, 1991). Thus, in contrast

© 2005 by CRC Press LLC

Insect Antifeedants

15

to sugar and salt receptors, deterrent receptors have preserved their general sensitivity, which has been linked to a neural response mechanism. In fact, all lepidopteran larvae possess a pair of maxillary palps that “drum” the surface of foods during feeding. These chemosensory organs contain over 65 percent of a larva's taste receptor cells, but their functional significance remains largely unknown. Their role in rejection of plant allelochemicals was examined, using the tobacco hornworm, Manduca sexta, as a model insect and an extract from a plant species, Grindelia glutinosa, as a model stimulus. This system was selected because hornworms reject foods containing Grindelia extract, and because preliminary studies indicated that their maxillary palps respond to this extract. It was hypothesized that Grindelia extract elicits rejection through stimulating (i) olfactory receptor cells, (ii) taste receptor cells, (iii) oral mechanoreceptors, and (iv) a post-ingestive response mechanism. The results were consistent only with hypothesis (ii); larvae approached Grindelia-treated diets without apparent hesitation, but rejected it within 6 seconds of initiating biting. Grindelia-treated solutions stimulated taste receptor cells in the maxillary palp, but not the other gustatory chemosensilla, and ablating the maxillary palps eliminated rejection of Grindelia-treated diets. The results demonstrate that taste receptor cells in the maxillary palps mediate rejection of Grindelia extract and provide the first direct evidence for a role of maxillary palps in rejection of plant allelochemicals (Glendinning et al., 1998). The possibility exists that insects use some other codes for taste quality, such as assessment of the temporal sequence of firing, which gives a continuous evaluation of the activity of individual neurons. It is also likely that simultaneous evaluation of inputs from different neurons allows contradictory signals, indicating the presence of phagostimulants or antifeedants, and is assessed concurrently (Schoonhoven, 1987). In addition to these neural mechanisms, it should be mentioned that some other targets are also vulnerable to antifeedants, like γ-amino butyric acid (GABA) antagonistic mechanisms, biogenic amine inhibition, and so on.

GABA ANTAGONISTIC MECHANISM GABA and related aminobutyric acids are known to stimulate feeding and evoke taste cell responses among herbivorous insects of various taxa, like Orthoptera, Homoptera, Coleoptera, and Lepidoptera (Mullin et al., 1994). However, it has also been established that allelochemicals antagonize GABA phagostimulants, like the isoquinoline alkaloid papavarine does in the Colorado potato beetle, thereby inducing feeding deterrence (Mitchell, 1987). GABAgated chloride channels respond to many classes of chemicals in insects (Sattelle, 1990; Anthony et al., 1993). The antagonism of GABA binding allows increased depolarization within an excitable cell and functions at both the neuromuscular junction and central synapses within the nervous system of insects. The present view is that inhibitory GABAA (Cl– conducting) receptors belong to a gene superfamily of ligand-gated ion channels that include excitatory nicotinic acetylcholine (Na+, K+) and inhibitory glycine (Cl–) receptors (Anthony et al., 1993). In turn, the α-carboxylated and precursor form of GABA, glutamic acid, gates a more distantly related family of both excitatory K+/Na+ and inhibitory Cl– channels (Darlison, 1992; Sattelle, 1992). On the whole it has been shown that the GABAA and glycine receptor complexes must incorporate two or three different four-transmembrane-domain subunits (Mullin et al., 1994). Association of GABA/glycine receptors with sensory systems has been demonstrated. For instance, bicuculline insensitivity at GABAA sites in insects has been found in CNS interneurons of the cockroach (Walker et al., 1971) and Manduca sexta (Waldrop et al., 1987).

© 2005 by CRC Press LLC

16

Opender Koul TABLE 2.2 Neuroreceptor antagonists on Western corn rootworm showing feeding deterrency. Chemical Strychnine HCl β-Hydrastine Bicuculline PK 11195 Gabazine Bulbocapnine HCl d-Tubocararine Cl Cyproheptadine HCl 7-Chlorokynurenic acid Harmine 2-Hydroxysaclofen Piperine Azadirachtin Parthenolide Agrophyllin A

ED50 nmol/dish 2.2 4.6 6.3 7.1 9.1 17.0 29.0 30.0 35.0 40.0 > 50.0 110.0 0.1 mM 1.5 mM 2.15 mM

Neuroreceptor Type Glycine (also GABAA) GABAA GABAA Periphral benzodiazepine GABAA Dopaminergic GABAA 5HT2 NMDA glutamergic GABAA GABAA Voltage sensitive Na gate GABAA ? GABAA ? GABAA ?

Source: Mullin et al. (1994).

In M. sexta GABA was found to mediate olfactory behavior via inhibitory interneurons in the antennal lobe of the deutocerebrum. However, only β-like subunits of GABA receptors from the CNS of Drosophila spp. (Henderson et al. 1993) and yellow fever mosquito, Aedes aegypti (Thompson et al., 1993), have been cloned from insect species. An interesting study of Mullin et al. (1991a, 1991b) shows the association of an antifeedant with a GABA/glycine receptor. Epoxy sesquiterpene lactone antifeedants from sunflower exhibit picrotoxinin-like GABA-gated chloride channel neurotoxicities in adult Western corn rootworm. In fact, terpenoid epoxides and isoquinoline and related alkaloids, such as azadirachtin, a strong antifeedant from neem (Koul, 1996), bicuculline, and so on, are interesting antifeedants of this category (Table 2.2). Mullin and coworkers (1994) have used three-dimensional structure-function relationships in Diabrotica to demonstrate antifeedant potency of compounds proposed to interact at a common binding site. Compounds were co-fitted through use of Alchemy III molecular modeling software (Tripos Associates). Common binding features for high antifeedant activity among the polycyclic terpenoid epoxides like azadirachtin, agrophylin, picrotoxinin, and caryophyllene oxide include an epoxide and π bonding sites separated by 0.5 to 0.6 nm, one or more electronegative oxygen centers, and a trisubstituted oxirane. Polyoxygenation may maintain sufficient polarity to allow diffusion to and interaction with the taste receptor. The 3D structural similarity between argophyllin (Mullin et al., 1991b) and picrotoxinin and dieldrin (Matsumura et al., 1987) suggest action through a shared picrotoxinin receptor site. The above studies also indicate that optimal polarity for molecular interactions at an exterior chemosensory receptor is different from internal interaction requirements with excitable cells, since membrane penetration and transport by binding proteins are not necessary (Mullin et al., 1994). A hydrophobic nature of compounds makes them non-inhibitory to feeding, as has been determined by using partition coefficient techniques. Many deterrents tested against Diabrotica have been shown to cause firing of a single taste neuron, and this chemosensory response correlates well with their feeding deterrency. In fact, GABA antagonism at the taste cell level may after neural processing result in net inhibition or excitation,

© 2005 by CRC Press LLC

Insect Antifeedants

17

respectively, of the dominant adductor, with a converse effect on the adductor. Clearly higher CNS inputs into mandibular opening and closing are also required. The actual inhibitory and excitatory inputs at each synaptic level, their means of integration, and the responsible neurotransmitters, receptors, and ion movements for insect gustation mostly remain to be clarified (Frazier, 1992).

BIOGENIC AMINE INHIBITION MECHANISM Biogenic amines are widely distributed within the insect CNS and thought to act as neurohormones, neuromodulators, or neurotransmitters (Evans, 1980). To get information about the mechanism of insect feeding, the insect response at biogenic amine levels against the feeding deterrents has been investigated (Ikemoto et al., 1995). For example, chlordimeform and aristolochic acid are well-known insect antifeedants and have been used as a probe of antifeedant activity. Five typical biogenic amines (5-hydroxytryptamine, dopamine, epinephrine, norepinephrine, and octopamine) using HPLC with an electrochemical detector have been investigated in the CNS of last instar Spodoptera litura larvae. It has been demonstrated that chlordimeform causes an increase in N-acetyldopamine levels in cerebral and suboesophageal ganglia and a decrease in 5-hydroxytryptamine (5HT) and n-acetyloctopamine levels in the cerebral, suboesophageal, and thoracic ganglia. On the other hand, aristolochic acid I, an antifeedant from Aristolochia species, did not cause any significant change in any amine levels except for dopamine and 5-hydroxytryptamine in suboesophageal ganglia and tyranine in thoracic ganglia (Ikemoto et al., 1995). Decrease in 5HT has also been reported in the cockroach cerebral ganglia (Omar et al., 1982). Inhibitory activity of chlordimeform against N-acetyltransferase has been shown in several insect species (Wierenga and Hollingworth, 1990). Although these studies suggest that some antifeedants have a mechanism of action through bioamine system in insects, comparing the effect on biogenic amine levels, no similar alterations have been observed. The relationship between such alterations and the antifeedant treatment is not clear as yet, but this could be one of the directions to study the mechanism of antifeedants and to understand the biochemical and physiological meaning of such alterations occurring due to feeding deterrents.

MECHANISMS RELATED TO SPECIFIC ALLELOCHEMICALS Chapter 7 illustrates more than 800 compounds that inhibit feeding of a variety of insect species using various bioefficacy procedures. However, the question remains: How do these chemicals affect the insect chemosensory cells, though a generalistic concept has been discussed above, and do they indicate the multiplicity of actions that can reduce feeding? Substantial data has been obtained in this regard for a diverse group of allelochemicals, the alkaloids. They inhibit impulse generation in sugar-sensitive cells in lepidopterans (Frazier, 1986; Simmonds et al., 1990) and competitively block sucrose responses in flesh flies (Morita et al., 1977). They also reduce the firing of the sugar-sensitive cells. Alkaloids as inhibitors of pyranose and furanose receptor sites have been established for flies (Wieczorek et al., 1988). The steroidal glycoalkaloids elicit irregular firing from several cells in the galeal and tarsal sensilla of adult and the larval α-sensilla of the Colorado potato beetle (Mitchell and Harrison, 1985). On the contrary, deterrent effects of various alkaloids, when tested against black blow flies, Phormia regina, in order to determine tarsal threshold for mixtures of sucrose and alkaloids, using kinetic analysis of electrophysiological data, ruled out competitive, no competitive, and

© 2005 by CRC Press LLC

18

Opender Koul

un-competitive inhibition at receptor sites. Even no correlation of thresholds with available data on lipid solubility or octanol/water partition coefficients was observed. This suggests that there is no uniform limiting mechanism for this multiform array of compounds (Dethier and Bowdan, 1989). Terpenes of various classes also inhibit insect feeding. Azadirachtin, one of the most potent deterrents known, has been shown to induce the firing of one cell in the labial palps and one in the A3 sensillum of the clypeo-labrum of Schistocerca gregaria (Haskell and Schoonhoven, 1969). It also induces the firing of cells in the medial sensilla styloconica of Pieris brassicae and Lymantria dispar larvae (Schoonhoven and Jermy, 1977; Schoonhoven, 1982). Azadirachtin effects in other caterpillar species are characterized by the firing of large spikes in the lateral and medial sensilla styloconica. This cell appears to fire independently of the sugar-sensitive cell (Simmonds and Blaney, 1984). This confirms a general observation that the effects of azadirachtin (and many other compounds as well) are different in different species; that is, affecting more than one chemosensory cell type in more than one way. Luo et al. (1995) describe a significant correlation between behavior and response of the medial deterrent cell for three triterpenoids, azadirachtin, salannin, and toosendanin. They showed a relationship between sensory input and feeding inhibition, supporting the hypothesis that the response of the medial deterrent cell directly causes inhibition of feeding in Pieris brassicae (Messchendorp et al., 1996). However, interference with the lateral glucosinolate- and sugar-sensitive receptor cells measured for toosendanin (Schoonhoven and Luo, 1994) did not contribute to a closer relationship between sensory response and inhibition of feeding on cabbage leaf disks in P. brassicae, mentioned above. Toosendanin has been shown to modulate the sensory code underlying feeding behavior via several different peripheral sensory mechanisms; that is, stimulation of the deterrent receptor cell located in the medial maxillary sensillum styloconicum and inhibition of responses of both the sugar and glycosinolate receptor cells (Schoonhoven and Luo, 1994). Other limonoids have also been shown to deter feeding in a variety of insect species (Champagne et al., 1992), but there is no electrophysiological data available to compare the effects on taste receptor cells. This information gap is mainly due to the fact that limonoids are insoluble in water, and this makes it difficult to apply the tip recording technique in an electrophysiological bioassay of limonoids. Some workers have solved the problem by using mixtures of 50% tetrahydrofuran and 50% aqueous sodium chloride as a solvent system (Waladde et al., 1989). In these studies the compounds, like deoxylimonin, obacunone, and pedonin, were used to investigate responses of Eldana saccharina maxillary styloconic sensilla and exhibited an inhibition of the sugar receptor cells. The sesquiterpene warburganal produces irregular firing of more than one cell and then blocks the responsiveness of the sucrose- and inositol-sensitive styloconic cell of Spodoptera exempta (Ma, 1977). It was suggested that in this case the deterrent acts via interaction with protein sulfhydral groups located at the receptor membrane. Some studies also suggest that warburganal reversibly blocks chemoreceptors, but the observation that feeding behavior of larvae of Spodoptera eridania, Schistocerca gregaria, and Manduca sexta is little affected may indicate that sensory input to the brain in these species does not inhibit food intake (Schoonhoven and Yan, 1989). It is well evident that such dialdehydic sesquiterpenoids (including polygodial, muzigadiol, etc.) affect not only the phagostimulant receptors, but also the deterrent cells located in the medial hair of insects (Schoonhoven and Yan, 1989). This suggests a mechanism of interference common to all taste receptors. Therefore, it remains unexplained why different receptors show different degrees of inhibition and different recovery periods. However, what is certain is that these sesquiterpenoids induce antifeedant effects

© 2005 by CRC Press LLC

Insect Antifeedants

19

in various insect species by (i) stimulation of a deterrent receptor and (ii) decreased sensitivity of most or all other receptors. Clerodin, an antifeedant diterpene, induces greater feeding deterrency when applied to the maxillary palps as compared to the sensilla styloconica (Antonious et al., 1984), which is in contrast to what has been observed in formamidine compounds (discussed above). Ginkgolides from Ginkgo biloba, when tested electrophysiologically for neural responses in the maxillary taste sensilla, show a strong stimulation of the deterrent receptors of two types in Pieris brassicae and P. rapae. However, in P. brassicae the medial sensillum is more strongly stimulated than the lateral sensillum, whereas in P. rapae the reverse is true (Yan et al., 1990). This illustrates the marked difference between the chemoreceptory systems of the two species. Drimanes with a lactone group on the B-ring appear to be the most potent antifeedants at 5 mM level (Messchendorp et al., 1996). The positive correlation between feeding inhibition and response of the deterrent cell suggests that these compounds exert a direct inhibitory effect on the feeding centers in the CNS. At the same time few compounds, though highly deterrent, do not evoke strong responses from the deterrent cells. This suggests that other mechanisms, either sensory or post-ingestive, are also involved in feeding inhibition. One of the drimanes tested in the above studies depressed the neurons sensitive to feeding stimulants. Whether or not this interference contributes to feeding inhibition remains to be elucidated. What could be concluded from this study is that highly effective drimane antifeedants can be selected electrophysiologically on the basis of response intensity of the medial deterrent cells, but further details of the mechanisms underlying feeding inhibition await to be revealed. There is also the evidence that mechanisms for antifeedants may vary within a species. In another study, for instance, 11 analogous synthetic drimane antifeedant compounds were evaluated for their feeding-inhibiting effects on larvae of the large white butterfly Pieris brassicae in no-choice tests on the host plant Brassica oleracea. The results show that the five analogous antifeedants differentially influence feeding behavior and locomotion activity. Some are most likely sensory-mediated antifeedants. Habituation to these compounds occurs soon after the onset of the tests (i.e., within 0.5 to 1.5 h). Others, like confertifolin, probably are not direct sensory-mediated antifeedants and rather induce post-ingestive anorexia. In conclusion, the behavioral observations performed in this research indicate that analogous drimanes inhibit feeding by P. brassicae larvae through multiple mechanisms of action (Messchendorp et al., 2000). The antifeedant activity of chalcones, flavones, and flavanones is due to the predominant stimulation of the deterrent neurons in the medial sensillum stylonicum, and more than one receptor may be involved (Simmonds et al., 1990). These studies suggest that there are at least two different receptor types involved, each having a different structure-function type of response. From the preceding discussion it is clear that the molecular structure of compounds visa-vis the neural responses associated with feeding deterrence mechanisms should throw some light on various molecular parameters such as chirality, functional groups, molecular size, and lipophilicity of the compounds. However, it appears difficult, if not impossible, to ascertain any common molecular conformation to all active molecules and their induction of a specific type of neural/receptor response towards a specific deterrent.

REFERENCES Anthony, N.M., Harrison, J.B., and Sattelle, D.B. (1993) GABA receptor molecules of insects. In Y. Pichon (ed.), Comparative Molecular Neurobiology, Birkhauser Verlag, Basel, pp. 172–209.

© 2005 by CRC Press LLC

20

Opender Koul

Antonious, A.G., Saito, T., and Nakamura, K. (1984) Electrophysiological response of the tobacco cutworm Spodoptera litura (F) to antifeedant compounds. J. Pestic. Sci., 9, 143–146. Asaoka, K. (1994) Different spectrum in responses of deterrent receptor cells in Sawa-J, a strain of the silkworm, Bombyx mori, with abnormal feeding habit. Zool. Sci (suppl.), 102. Bernays, E.A. and Chapman, R.F. (1994) Host Plant Selection by Phytophagous Insects. Chapman and Hall, New York. Bernays, E.A., Oppenheim, S., Chapman, R.F., Kwon, H., and Gould, F. (2000) Taste sensitivity of insect herbivores to deterrents is greater in specialists than in generalists: A behavioural test of the hypothesis with two closely related caterpillars. J. Chem. Ecol., 26, 547–563. Blaney, W.M. (1980) Chemoreception and food selection of locusts. Olfaction and Taste, 7, 127–130. Blom, F. (1978) Sensory activity and food intake: a study of input-output relationships in two phytophagous insects. Netherl. J. Zool., 28, 277–340. Boer de, G., Dethier, V.G., and Schoonhoven, L.M. (1977) Chemoreceptors in the preoral cavity of the tobacco hornworm, Manduca sexta, and their possible function in feeding behaviour. Entomol. Exp. Appl., 21, 287–298. Brattsten, L.M. and Ahmad, S. (1986) Molecular Aspects of Insect Plant Associations. New York: Plenum Press. Champagne, D.E., Koul, O., Isman, M.B., Towers, G.H.N., and Scudder, G.G.E. (1992) Biological activity of limonoids from the rutales. Phytochemistry, 31, 377–394. Chapman, R.F. (1982) Chemoreception: the significance of receptor numbers. Adv. Insect Physiol., 16, 247–285. Darlison, M.G. (1992) Invertebrate GABA and glutamate receptors: molecular biology reveals predictable structures but some unusual pharmacologies. TINS, 15, 469–473. Descoins, C. Jr. (2001) Sensing of antifeeding agents by phytophagous caterpillars (Lepidoptera). Annee-Biologique, 40, 55–73. Dethier, V.G. (1976) The Hungry Fly. Cambridge: Harvard University Press. Dethier, V.G. (1980) Evolution of receptor sensitivity to secondary plant substances with special reference to deterrents. Amer. Nat., 115, 45–66. Dethier, V.G. and Crnjar, R.M. (1982) Candidate codes in the gustatory system of caterpillars. J. Gen. Physiol., 79, 549–569. Dethier, V.G. and Bowdan, E. (1989) The effect of alkaloids on sugar receptors and the feeding behaviour of the blowfly. Physiol. Entomol., 14, 127–136. Devitt, B.D. and Smith, J.J.B. (1982) Morphology and fine structure of mouth part sensilla in the darksided cutworm, Euxoa messoria (Harris) (Lepidoptera: Noctuidae). Int. J. Insect Morph. Embryol., 11, 225–270. Devitt, B.D. and Smith, J.J.B. (1985) Action of mouth parts during feeding in the darkside cutworm, Euxoa messoria (Lepidoptera: Noctuidae). Can. Entomol., 117, 343–349. Evans, P.D. (1980) Biogenic amines in the insect nervous system. Adv. Insect Physiol., 15, 317–330. Frazier, J.L. (1985) Nervous system: Sensory system. In M.S. Blum (ed.), Fundamentals of Insect Physiology, John Wiley & Sons., New York, pp. 287–356. Frazier, J.L. (1986) The perception of plant allelochemicals that inhibit feeding. In L.B. Brattsten and S. Ahmad (eds.), Molecular Aspects of Insect Plant Associations, Plenum Press, New York, pp. 1–42.

© 2005 by CRC Press LLC

Insect Antifeedants

21

Frazier, J.L. (1992) How animals perceive secondary plant compounds. In G.A. Rosenthal and M.R. Berenbaum (eds.), Herbivores: Their Interaction with Secondary Plant Metabolites, Evolutionary and Ecological Processes, 2nd ed., Vol. 2, Academic Press, San Diego, pp. 89–134. Glendinning, J.I., Valcic, S., and Timmermann, B.N. (1998) Maxillary palps can mediate taste rejection of plant allelochemicals by caterpillars. J. Comp. Physiol. (A), 183, 35–43. Haskell, P.T. and Schoonhoven, L.M. (1969) The function of certain mouth part receptors in relation to feeding in Schistocerca gregaria and Locusta migratoria migratorioides. Entomol. Exp. Appl., 12, 423–440. Henderson, J.E., Soderlund, D.M., and Knipple, D.C. (1993) Characterization of putative γamino butyric acid (GABA) receptor β-subunit gene from Drosophila melanogaster. Biochem. Biophys. Res. Commun., 193, 474–482. Hollister, B., Dickens, J.C., Perez, F., and Deahl, K.L. (2001) Differential neurosensory responses of adult Colorado potato beetle, Leptinotarsa decemlineata, to glycoalkaloids. J. Chem. Ecol., 27, 1105–1118. Ikemoto, Y., Matsuzawa, Y., and Mizutani, J. (1995) The effect of antifeedants against the level of biogenic amines in the central nervous system of the lepidopteran insect (Spodoptera litura). Pestic. Biochem. Physiol., 52, 60–70. Ishikawa, S. (1966) Electrical response and functions of a bitter substance receptor associated with the maxillary sensilla of the larvae of the silkworm Bombyx mori. J. Cell Physiol., 67, 1–12. Kennedy, L.M. and Halpern, B. (1980) Fly chemoreceptors: A novel system for the taste modifier ziziphin. Physiol. Behavior, 24, 1–9. Koul, O. (1996) Mode of azadirachtin action. In N.S. Randhawa and B.S. Parmar (eds.), Neem, New Age International Publishers Ltd., New Delhi, pp. 160–170. Koul, O. (1997) Molecular targets for feeding deterrents in phytophagous insects. In A. Raman (ed.), Ecology and Evolution of Plant Feeding Insects in Nature and Man-Made Environments, International Scientific Publications, New Delhi and Backhuys Publishers, Leiden, The Netherlands, pp. 123–134. Kuppers, J. and Thurm, U. (1982) On the functional significance of ion circulation induced by electrogenic tissue. In A.D.F. Addink and N. Spronk (eds.), Exogenous and Endogenous Influences on Metabolic and Neural Control, Pergamon Press, Oxford, pp. 313–327. Luo, L.-E., van Loon, J.J.A., and Schoonhoven, L.M. (1995) Behavioural and sensory responses to some neem compounds by Pieris brassicae larvae. Physiol. Entomol., 20, 134–140. Ma, W.C. (1972) Dynamics of feeding responses in Pieris brassicae L. as a function of chemosensory input: a behavioural, ultrastructural and electrophysiological study. Meded. Landbouwhogeschool Wageningen 72/11, 162 pp. Ma, W.C. (1977) Alteration of chemoreceptor function in armyworm larvae (Spodoptera exempta) by a plant-derived sesquiterpenoid and sulfhydral reagent. Physiol. Entomol., 2, 199–207. Matsumura, F., Tanaka, K., and Ozoe, Y. (1987) GABA-related systems as targets for insecticides. In R.M. Hollingworth and M.B. Green (eds.), Sites of Action for Neurotoxic Pesticides, Symp. Ser. 356, American Chemical Society, Washington, D.C., pp. 44–70. Messchendorp, L., van Loon, J.J.A., and Gols, G.J.Z. (1996) Behavioural and sensory responses to drimane antifeedants in Pieris brassicae larvae. Entomol. Exp. Appl., 79, 195–202.

© 2005 by CRC Press LLC

22

Opender Koul

Messchendorp, L., Smid, H.M., and van Loon, J.J.A. (1998) The role of an epipharyngeal sensillum in the perception of feeding deterrents by Leptinotarsa decemlineata larvae. J. Comp. Physiol. (A), 183, 255–264. Messchendorp, L., Gols, G.J.Z., and van Loon, J.J.A. (2000) Behavioural observations of Pieris brassicae larvae indicate multiple mechanisms of action of analogous drimane antifeedants. Entomol. Exp. Appl., 95, 217–227. Miller, J.R. and Strickler, K.L. (1984) Finding and accepting host plants. In W.J. Bell and R. Carde (eds.), Chemical Ecology of Insects, Sinauer Associates, Sunderland, MA, pp. 127–157. Mitchell, B.K. (1987) Interaction of alkaloids with galeal chemosensory cells of Colorado potato beetle. J. Chem Ecol., 13, 2009–2022. Mitchell, B.K. and Harrison, G.D. (1985) Effects of Solanum glycoalkaloids on chemosensilla in the Colorado potato beetle; a mechanism of feeding deterrence. J. Chem. Ecol., 11, 73–83. Mitchell, B.K., Rolseth, B.M., and McCashin, B.G. (1990) Differential responses of galeal sensilla of the adult Colorado potato beetle, Leptinotarsa decemlineata (Say) to leaf saps from host and non-host plants. Physiol. Entomol., 15, 61–72. Morita, H., Enomoto, K., Nakashima, M.N., Shimada, I., and Kijima, H. (1977) The receptor site for sugars in chemoreception of the flesh fly and the blow fly. In J. LeMagnen and P. MacLeod (eds.), Proceedings of Sixth International Symp., Olfaction and Taste, Information Retrieval, London, pp. 39–46. Mullin, C.A., Alfatafta, A.A., Harman, J.L., Serino, A.A., and Everett, S.L. (1991a) Corn rootworm feeding on sunflower and other composite: influence of floral terpenoid and phenolic factors. In P.A. Hedin (ed.), Naturally Occurring Pest Bioregulators, Symp. Ser. 449, American Chemical Society, Washington, D.C., pp. 278–292. Mullin, C.A., Alfatafta, A.A., Harman, J.L., Everett, S.L., and Serino, A.A. (1991b) Feeding and toxic effects of floral sesquiterpene lactones, diterpenes and phenolics from sunflower (Helianthus annuus L.) on western corn rootworm. J. Agric. Food Chem., 39, 2293–2299. Mullin, C.A., Chyb, S., Eichenseer, H., Hollister, B., and Frazier, J.L. (1994) Neuroreceptor mechanisms in insect gustation: A pharmacological approach. J. Insect Physiol., 40, 913–931. Omar, D., Murdock, L.L., and Hollingworth, R.M. (1982) Actions of pharmacological agents on 5-hydroxytryptamine and dopamine in the cockroach nervous system (Periplaneta americana L.). Comp. Biochem. Physiol., 73, 423–429. Renwick, J.A.A. (2001) Variable diets and changing taste in plant–insect relationships. J. Chem. Ecol., 27, 1063–1076. Sattelle, D.B. (1990) GABA receptors of insects. Adv. Insect Physiol., 22, 1–113. Sattelle, D.B. (1992) Receptors for L-glutamate and GABA in the nervous system of an insect (Periplaneta americana). Comp. Biochem. Physiol., 103C, 429–438. Schoonhoven, L.M. (1982) Biological aspects of antifeedants. Entomol. Exp. Appl., 31, 57–69. Schoonhoven, L.M. (1987) What makes a caterpillar eat? The sensory code underlying feeding behaviour. In R.F. Chapman, E.A. Bernays, and J.G. Stoffolano (eds.), Perspectives in Chemoreception and Behaviour, Springer-Verlag, New York, pp. 69–97. Schoonhoven, L.M. (1988) Stereoselective perception of antifeedants in insects. In E.J. Ariens, J.J.S. van Rensen, and W. Welling (eds.), Stereoselectivity of Pesticides; Biological and Chemical Problems, Elsevier, Amsterdam, pp. 289–302. Schoonhoven, L.M. (1991) The sense of distaste in plant feeding insects: A reflection on its evolution. Phytoparasitica, 19, 3–7.

© 2005 by CRC Press LLC

Insect Antifeedants

23

Schoonhoven, L.M. and Dethier, V.G. (1966) Sensory aspects of host-plant discrimination by lepidopterous larvae. Arch. Neerl. Zool., 16, 497–530. Schoonhoven, L.M. and Jermy, T. (1977) A behavioural and electrophysiological analysis of insect feeding deterrents. In N.R. Mcfarlane (ed.), Crop Protection Agents, Academic Press, New York, pp. 133–146. Schoonhoven, L.M. and Blom, F. (1988) Chemoreception and feeding behaviour in a caterpillar: towards a model of brain functioning in insects. Entomol Exp. Appl., 49, 123–129. Schoonhoven, L.M. and Yan Fu-Shun (1989) Interference with normal chemoreceptor activity by some sesquiterpenoid antifeedants in an herbivorous insect, Pieris brassicae. J. Insect Physiol., 35, 725–728. Schoonhoven, L.M. and Luo, L.-E. (1994) Multiple mode of action of the feeding deterrent toosendanin, on the sense of taste in Pieris brassicae larvae. J. Comp. Physiol., 175A, 519–524. Schoonhoven, L.M., Blaney, W.M., and Simmonds, M.S.J. (1992) Sensory coding of feeding deterrents in phytophagous insects. In E. Bernays (ed.), Insect–Plant Interactions, Vol. 4, CRC Press, Boca Raton, FL, pp. 59–79. Schoonhoven, L.M., Jermy, T., and van Loon, J.J.A. (1998) Insect–Plant Biology. Chapman and Hall, London. Simmonds, M.S.J. and Blaney, W.M. (1984) Some neurophysiological effects of azadirachtin on lepidopteran larvae and their feeding response. Proc. 2nd Int. Neem Conf., GTZ, Rauischazhausen, pp. 163–179. Simmonds, M.S.J., and Blaney, W.M. (1990) Gustatory codes in lepidopterous larvae. Symp. Biol. Hung., 39, 17–27. Simmonds, M.S.J., Blaney, W.M., and Fellows, L.E. (1990) Behaviour and electrophysiological study of antifeedant mechanisms associated with polyhydroxy alkaloids. J. Chem. Ecol., 16, 3167–3196. Stadler, E. (1992) Behavioural responses of insects to plant secondary compounds. In G.A. Rosenthal and M. R. Berenbaum (eds.), Herbivores: Their Interaction with Secondary Plant Metabolites; Evolutionary and Ecological Processes, Academic Press, San Diego, pp. 44–88. Sturckow, B. (1959) Ueber den Geschmackssinn und den Tastsinn von Leptinotarsa decemlineata Say (Chrysomelidae). Z. Vergl. Physiol., 42, 255–302. Tang, D.L., Wang, C.Z., Luo, L., and Qin, J.D. (2000) Comparative study on the responses of maxillary sensilla styloconica of cotton bollworm Helicoverpa armigera and Oriental tobacco budworm H. assulta larvae to phytochemicals. Sci. China Ser.C, Life Sci., 43, 606–612. Thompson, M., Shotkoski, F., and ffrench-Constant, R. (1993) Cloning and sequencing of the cyclodiene insecticide resistant gene from the yellow fever mosquito, Aedes aegypti. FEBS Letters, 325, 187–190. Waldrop, B., Christensen, T.A., and Hildebrand, J.G. (1987) GABA mediated synaptic inhibition of projection neurons in the antennal lobes of the sphinx moth, Manduca sexta. J. Comp. Physiol., 161A, 23–32. Walker, R.J., Crossman, A.R., Woodruff, G.N., and Kerkut, G.A. (1971) The effect of bicuculline on the γ-amino butyric acid (GABA) receptors of neurons of Periplaneta americana and Helix aspersa. Brain Res., 33, 75–82. Waladde, S.M., Hassanali, A., and Ochieng, S.A. (1989) Taste sensilla responses to limonoids, natural insect antifeedants. Insect Sci. Applic., 10, 301–308.

© 2005 by CRC Press LLC

24

Opender Koul

Wieczorek, H. (1976) The glycoside receptor of the larvae of Mamestra brassicae L. (Lepidoptera: Noctuidae). J. Comp. Physiol., 106A, 153–176. Wieczorek, H., Shimada, I., and Hopperdietzel, C. (1988) Treatment with pronase, uncouples water and sugar reception in the labellar water receptor of blowfly. J. Comp. Physiol., 163A, 413–419. Wierenga, J.M. and Hollingworth, R.M. (1990) Octopamine uptake and metabolism in the insect nervous system. J. Neurochem., 54, 479–489. Yan Fu-shun, Evans, K.A., Stevens, L.H., van Beek, T.A., and Schoonhoven, L.M. (1990) Deterrents extracted from the leaves of Ginkgo biloba: effects on feeding and contact chemoreceptors. Entomol. Exp. Appl., 54, 57–64.

© 2005 by CRC Press LLC

3 BIOASSAYS Bioassays against insects have been used for decades as a means of elucidating the activity of many chemical components. The major goals achieved by employing bioassay techniques are to determine the roles of naturally occurring chemicals, identify the mechanism of resistance in crop plants, and to find various insect control agents. As the aim of this book is to understand chemicals that inhibit feeding, the emphasis will be on antifeedant allelochemical bioassays. However, before going into the details of various types of bioassays for feeding-deterrent activity, it is imperative to know about certain fundamental requirements for such evaluations. The basic design to study deterrents is to present to an insect a substrate with the candidate chemical and to measure the response of the insect. Therefore, substrate choice and presentation are important factors for a successful bioassay. Both natural and artificial substrates are used, depending upon the goal of the experiment. On one hand one may emphasize that artificial substrates offer uniformity, but at the same time studies have shown that thresholds for the same deterrent may vary as much as 1000 times between natural and artificial substrates (Schoonhoven, 1982), perhaps due to differences in porosity or uptake rates by the insect. For sucking insects, the principal artificial substrate used has been a chemically defined liquid presented between natural or artificial membranes (Mittler and Dadd, 1962). However, whatever the substrate may be, it is important that no textural differences should occur between the control and test substrates. Color differences may also influence insects tested (Meisner and Ascher, 1973). Care is needed to ensure the least hindrance with the insect’s chemoreceptor’s encounter, which should be in the usual way (i.e., for edge feeders substrates are placed above the floor level). Natural substrates could be whole plant, leaves, leaf disks (more frequently used), or specialized substrates like twigs, blocks of wood, board, and paper towel disks. Artificial substrates usually include agar-based artificial diets, simple

© 2005 by CRC Press LLC

26

Opender Koul

liquid-based artificial diets, styropors (which are good model substrates for many insects using lamellae), or disks of foamed polystyrene, styrofoam, or polyurethane, and glass fiber disks. After the choice of the substrate, an important step is the chemical application. For antifeedant testing, concentrations need to be selected to achieve the goal of finding the effective concentrations for crop protection. Higgins and Pedigo (1979a) used a foliar phytotoxicity threshold base as the presence of leaf necrosis to determine maximum acceptable concentration. One could use the sensitivity of the insect chemosensilla as a criterion for concentration, but it is difficult as very little is known about these sensitivities. The role of solvents is another important factor. Ascher et al. (1981) found that grades of a common solvent, methanol, could differ in their effects. My studies with antifeedant evaluations have revealed that alcohols interfere with the texture of natural substrates (particularly leaf materials), and therefore should be avoided. The best way is to depend on waterbased emulsified solutions, which hold a small quantity of the solvent. Therefore, natural control substrates with and without solvent should be tested to verify that there is no alteration in palatability due to solvents. In electrophysiological studies appropriate electrolyte use is essential. The next important factors for evaluation are the conditions of the test. Jermy et al. (1968) used a sytem of evaluation where tests were terminated when 50% of either substrate in a choice trial was consumed, so the insects did not lose discrimination due to hunger. This system is adopted even today in many studies. There is also evidence that previous experiences affect insect diet choice by induction, habituation, food aversion learning, and associative learning. Therefore, such problems need to be avoided and apparently could be achieved by using insects only once in a test and for a short duration. Pilot observations are necessary to establish time of maximum feeding, which varies among various insect species. Temperature, humidity, light levels, and population of test insects are other parameters that need to be determined from field observations for better results in the antifeedant assay. In fact, performing short- and long-term tests provides the most information, as data on changes in behavior through lengthy exposure to a chemical may be useful (Lewis and Bernays, 1985). The above-mentioned step is then followed by measurements and observations. These parameters will depend upon the feeding behavior of the insect; that is, methodologies will depend upon whether the bioassay is conducted against a chewing insect, a sucking insect, and so on. Depending on the different modes of feeding in insects, various types of bioassay procedures have been developed in laboratories to evaluate insect antifeedants. I have categorized these bioassays as shown in Figure 3.1. However, some specific assays used against specific insect species are also described.

LEAF DISK ASSAY Leaf disks are commonly used in insect bioassays of preference or consumption with chewing insects. This assay can be conducted in two basic ways: i) choice assay and ii) no-choice assay. Insects can choose either control or treated disks (choice) or insects may be exposed to the test substance only (no-choice). The no-choice situation often is more representative of our agricultural system, especially for monophagous species, but at the same time it is very sensitive. The general procedure adopted in this test is that measured leaf disks are punched out from substrates and treated either on one side or both sides with a known quantity of test material in a carrier solvent. It is preferable to use emulsified solutions in water in order to avoid interference with leaf disk texture due to solvents (Koul et al., 1990). A method has been described by which leaf surfaces can be covered with a uniform amount of a test chemical

© 2005 by CRC Press LLC

Insect Antifeedants

27

LEAF DISC TEST LOCUSTS, GRASSHOPPERS MOSTLY LEPIDOPTERANS

ARTIFICIAL DIET TEST

ARGENTINE STEM WEEVIL GREEN BUGS, LEPIDOPTERANS

PAPER TOWEL DISC TEST

TERMITES

WHEAT WAFER ASSAY

STORED GRAIN PESTS

STYROPOR ASSAY

LEPIDOPTERANS SUCKING INSECTS

ANTIFEEDANT BIOASSAYS GLASS FIBER DISC ASSAY

PIPET ASSAY

SIMULATION ASSAY

IMPREGNATION ASSAY

ELECTRO PHYSIOLOGICAL ASSAY

LEPIDOPTERANS, LOCUSTS ACRIDIDS

DIPTERANS

MANY PHYTOPHAGOUS INSECTS

BEAN BEETLES, SCALE INSECTS, WHITE FLIES, CITRUS MITES, WIREWORMS, LEPIDOPTERANS LEPIDOPTERANS LEAFHOPPERS COCKROACHES

FIGURE 3.1 Bioassay methods.

for bioassay with leaf-feeding insects. Chemicals are dissolved in gelatin solutions, which can be sprayed evenly and which will adhere well to many leaf surfaces. However, calibrations of rates of application are required (Wolfson and Murdock, 1987). After application, the leaf disks are dried at room temperature and then fed to candidate insects. Usually the arenas used are petri dishes of variable sizes (9-cm-diameter size is most common for lepidopteran larvae) in which one treated and one control disk is placed (choice), or both the leaf disks are treated (no-choice). In certain experiments five to ten treated and

© 2005 by CRC Press LLC

28

Opender Koul

untreated leaf disks are used and placed alternately in the petri dishes in a choice situation. The number of larvae introduced into each arena is variable depending upon the size and stadium of the larvae used. There is also considerable variation in the duration of experiments (see Chapter 7), both long term and short term. The consumption in each experiment is measured using various digitizing leaf area meters. In certain studies the choice tests have been referred to as dual choice tests. In these experiments individual larvae are confined for a short term (5 to 6 hours) in a petri dish containing two leaf disks. The treatment disks are painted with an aliquot of test solution and the control disks with solvent solution only. The bioassay is conducted for very short duration or until 50% of either disk is consumed. The amount eaten is then assessed and calculated. In certain cases even more than two leaf disks have been used. For instance, while testing drimane-type antifeedants in dual-choice tests against Pieris brassicae, six cabbage leaf disks (3.8 cm2 area) placed in a circle were assayed in a glass petri dish. The upper surface of alternate disks was painted with 10 µl drimane or control solution, after which they were left to dry for 30 minutes. After 3 hours of ad libitum feeding, the remaining disk areas were measured with a leaf area meter (Messchendorp et al., 1996). However, some findings have implications to the design of disk bioassays. Clearly, the ratio of cut edge to overall leaf disk surface area is an important variable that should be taken into account (Jones and Coleman, 1988). The decision to choose one disk size over another has often been a matter of convenience for the experimenter. Thus, it has been suggested that selection of the appropriate disk size will depend upon several variables (Jones and Coleman, 1988): • • •

Whether the insect feeds in the center or edges of the leaves The size of the insect, particularly the ratio of insect size to disk size The type of bioassay being carried out

For example, if a leaf disk from the host plant of a center-feeding insect were painted with a deterrent chemical, one would predict the deterrent effects would be greater in a large disk assay than in a small disk assay. This is because the cut edge would contribute counteracting host-plant attractant or stimulant signals to a greater extent on small disks. In order to measure and calculate the effective concentrations of insect antifeedants, there is no standard size terminology. In certain cases the consumed area of treated leaf disks is expressed as a percentage of the consumed area of control leaf disks. In others, antifeedance has been calculated on the basis of feeding ratios (i.e., test consumption/control consumption followed by grading) (Zalkow et al., 1979). This ratio has also been calculated as: Feeding control = 100 { 1 – % feeding/% feeding by stock} and graded as +++ (90–100%); ++ (60–90%); + (30–60%) protection, and – (0–30%) showing no protection (Lidert et al., 1985). In many cases an antifeedant is considered to be effective when feeding inhibition of 80–100% is achieved (Bernays and Chapman, 1978). EC90 and EC95 values have also been calculated. On the whole some generalized formulas have been devised to calculate feeding deterrence quantitatively: • •

D = (1 – T/C) × 100, where T and C are percent weights of treated and control leaf disks (dry weight basis) and D is the percent deterrence (Hosozawa et al., 1974). Protection (%) = (% PTS – % PIC)/100 – % PIC, where PTS is protection in treated samples and PIC is protection in control samples (Singh and Pant, 1980).

© 2005 by CRC Press LLC

Insect Antifeedants •

•

29

Feeding inhibition (%) = % CTD × 100/% CTD + % CUD; where CTD is consumption of treated disks and CUD is consumption of untreated disks (Reed and Jacobson, 1983). Feeding deterrence (%) = (C – T)/(C + T) × 100, where C and T are the consumption of control and treated disks, respectively (Koul et al., 1990). This formula is preferred for the measurement of deterrence.

Recently, leaf disk choice bioassay has been very successfully used to study aphids (Lowery and Isman, 1993; Koul et al., 1997). In this procedure two leaf disks with the test material and two disks with carrier alone are allowed to dry and then arranged alternately in small petri dishes (9 × 50 mm) with their edges barely touching. Deterrence of the test material is determined by the proportion of aphids on the treated disks relative to the total number of aphids on treated and untreated disks in each dish. An improved antifeedant bioassay has been devised that allows an accurate measurement of consumed disk surfaces, using a video camera interfaced with a computer. The scanned image of the leaf disk is stored and the eaten areas are measured with the help of videoimage analysis software. This method allows for precise quantification of insect antifeedant activity tested on leaf material (Escoubas et al., 1993). The process of storing the image involves selection of the interesting area in the scanning window via cut-and-paste functions. By this method, the image can be edited, enhanced, and used in later measurements.

ARTIFICIAL DIET FEEDING Artificial diet tests are also commonly used against many lepidopterans, green bugs, and weevils. A general procedure for this evaluation is to incorporate the test material into artificial diets, feed to test insects, and calculate mean percent feeding depression. However, artificial diets usually have one of two problems: they are suboptimal or they can be superoptimal (Wolfson, 1988). Compared with the most susceptible host plants, artificial diets often foster more rapid growth (Reese and Field, 1986) and thus can make insects less susceptible to the allelochemicals being tested. According to Berenbaum (1986), tests of chemicals in artificial diets could be compromised by eliminating naturally occurring synergistic interactions between nutrients and allelochemicals. However, there are some specific examples where artificial diets have been successfully used for evaluation. For instance, artificial diet plugs of known weight were placed in an arena (35 × 10 mm petri dishes) along with two sixth instar larvae of spruce budworm, Choristoneura fumiferana. Both choice and no-choice situations were established. Larvae were allowed to feed for 48 hours, at which time remaining portions of diet plugs were weighed. Results were expressed as weight of plugs consumed/insect/48h. Each group was replicated three times for a total of 24 insects/treatment (Alford and Bentley, 1986). Mean percent feeding depression was calculated as = [ 1 – treatment consumption/control consumption × 100 ]. In the case of the boll weevil, bioassay chambers were constructed by pouring a layer (5 mm thick) of beeswax paraffin (2:1 by weight) into the bottom of a petri dish (1.5 × 10 cm). Three to six media wells were cut into the cooling paraffin with a No. 9 cork borer (13 mm diameter), 2 cm from the center of the dish and equidistant apart. The wells were filled with cooling 2.5% nonnutrient agar solution plus 3.0% pharmamedia or freeze-dried cotton square powder. A lens paper disk cut with a No. 10 cork borer (15 mm diameter) was placed on top of the solidified medium surface. Water (20 µl) was applied to the paper to provide

© 2005 by CRC Press LLC

30

Opender Koul

moisture for the insects and to hold the papers to the medium surface. Chemicals were randomly applied to the lens papers as 20 µl aliquots in methanol and allowed to dry for 30 minutes. 10 male or female boll weevils were released into each assay chamber and the tops were applied to prevent escape. The scores for the number of punctures after 3 to 6 hours were noted using a dissecting microscope, and deterrence effects were calculated (Bird and Hedin, 1986). For Eurema hecabe mandarina (yellow butterfly), larvae pre-starved for 4 hours were placed on each of the following diets: 1. The basal diet (1% agar only). 2. The control diet prepared by incorporating dried leaf powder of Lespedeza cuneata (0.39 g) in 1.4 ml of 1% agar. 3. The test diets prepared by incorporating plant powders or fractions or compounds into dried leaf powder as in (2) in 1 ml of 1% agar at a total weight of 1.28 g. The number of frass pellets deposited during the test period (20 h) was then counted. The feeding inhibition activity was calculated on the basis of feeding ratio:

Feeding ratio =

Mean frass count/test diet – Mean frass count for the basal diet Mean frass count for the control diet – Mean fraass count for the basal diet

When the ratio was < 50% it was considered to be a positive inhibition. When the response was between 0 and 20% the test material was a very strong inhibitor (Numata et al., 1984). Agar-based diets have also been used for Argentine stem weevils, Listronotus bonariensis (Rowan and Gaynor, 1986). After cooling, these diets were cut into 35 to 40 disks with a cork borer (1 cm × 3 mm thick) in a choice bioassay. For treatment experiments, cellulose powder was mixed with the candidate chemical at 2 g of extract/4 g of cellulose powder. The suspensions were made, evaporated to dryness, and agar disks were made. There are artificial diets developed for evaluating antifeedant chemicals against aphids. A generalized pattern has been adopted for Schizaphis graminum. In this procedure small (35 ml) polystyrene catsup cups are used as test chambers. For each test 50 to 75 aphids of all ages are transferred from the colony by brushing them carefully off the plant with a hair paint brush. The polystyrene test chambers consist of tight-fitting plastic snap-on caps, which possess circular holes (1.5 cm) punched with a cork borer for the placement of the diet container. The container is usually made from a soft polyethylene vial cap, which fits snuggly into the hole prepared with the cork borer in the lid. A thin sheet of parafilm is stretched across the vial cap to create a sealed diet chamber. Diet containing the test material is added by injection with a syringe through the topside of each polyethylene cap. The diet-filled containers are placed by a snug fit into the hole bored in the lid of the test chamber with the parafilm membrane facing the interior of the test chamber towards the aphids. The tests are maintained for 24 hours at 24˚C. After 24 hours the number of aphids feeding and the number wandering are counted and compared with appropriate controls. One week later an identical set of experiments is run and all the replicates are averaged (Dreyer et al., 1981). In certain cases in synthetic diet feeding, each substance is tested at a series of concentrations so that a dose-dependent curve could be constructed. From this curve a concentration could be obtained at which half of the aphids (ED50 values) would not feed (Rose et al., 1981). Calculations can also be based on the difference in weight of the larvae in each group of treatments, multiplied by 100 and divided by the average weight of larvae in the control group to obtain a larval weight index (Warthen et al., 1982).

© 2005 by CRC Press LLC

Insect Antifeedants

31

STYROPOR ASSAY This method is usually used for lepidopterans and sucking types of insects (Koul, 1993). Thin 0.6 mm lamellae (6 × 4 cm) of styropor (foamed polystyrene) of density 0.016 (P16) are dipped into solutions of the test compounds of different concentrations containing 5% sucrose. The lamellae are left to dry for 24 hours and then weighed individually. They are then offered singly in large 15-cm-diameter petri dishes to one early sixth instar larvae (e.g., Spodoptera littoralis of 170 to 190 mg body weight), together with water absorbed in cotton wads. The number of replicates varies from five to ten for each level of treatment. The weight of styropor after consumption and the weight of fecal pellets voided are recorded for each larva after 48 hours and used as criteria for antifeedant activity (Ascher and Meisner, 1973).

GLASS FIBER DISK TEST This type of antifeedant bioassay, though used for lepidopteran larvae, has also been useful to evaluate compounds against locusts and other acridids. In this method the test compounds are added to glass fiber disks (Whatman GF/A), a pair of which are presented dry to each individual insect. The disk size varies from 2.1 to 4.25 cm, which are usually dried in a cool air stream. The loading of the disks varies from 100 to 400 µl of test material. Locusts are confined individually in clear plastic boxes (27 × 15 × 10 cm), each screened from its neighbor so that insects in adjacent boxes do not disturb each other. Disks are presented in pairs to facilitate feeding for 24 hours until 30 to 50% of the control disk has been consumed; the amount of disk eaten is measured by an area meter. Lepidopteran larvae are confined in petri dishes (9 cm diameter) with a pair of disks for up to 8 hours so that never more than 50% of any disk is eaten. After removal of the larvae disks are redried and the amount eaten is determined by weight (Blaney et al., 1984). For acridids, glass fiber disks of 4.5 cm diameter have been used by adding known quantities of sucrose solutions to give either 5% or 12% dry weight of sucrose. After drying, disks are treated with a known amount of compound in solution. Approximately 0.4 ml solution saturates this size of disk and gives an average concentration of 0.02 to 2.0% on a dry weight basis. After evaporation of solvent the disks are checked by weight and then fed to insects. All choice tests are carried with well-fed insects approximately halfway through the final larval instar. Experiments are run for 3 hours and usually as 10 to 15 replicates. After the test each remaining disk is measured with an area meter or in certain cases weighed and the amount of each disk eaten/insect is calculated (Bernays and DeLuca, 1981).

PAPER TOWEL DISK TEST Commonly used for termites, this method has been very well described by Scheffrahn and Rust (1983). The natural paper towels, cut into 9.6-cm-diameter disks (72 cm2 area and 360 mg average weight) are individually placed in glass petri dishes. Predetermined amounts of wood extract or pure compounds that would yield calculated specific mass per unit area of paper (mg/cm2) are weighed in glass shell vials (e.g., 36 mg of a compound applied to a disk provided a deposit of 0.5 mg/cm2). The test compound dissolved in 2 ml of an appropriate solvent is poured from the vial evenly onto the disk resting inside a petri dish cover. The cover is then placed on a warm hot plate to evaporate the solvent. Trial tests using an ethersoluble dye can be used for practice to get uniform deposits. Each treated disk is cut into four smaller disks (3.9 cm diameter) and held overnight at room temperature before testing. For deterrency bioassays 0.5 mg/cm2 application is presented for 3 days. A choice test is conducted in which four disks are constructed from two half disks; one is treated with test

© 2005 by CRC Press LLC

32

Opender Koul

material and the other half a solvent blank to demonstrate feeding preference. The half disks are joined with a thin strip of cellophane tape on the surface not exposed to termites. Evaluation of weight loss of the disks due to feeding is averaged for each treatment and a comparison of their means determined statistically (Scheffrahn and Rust, 1983). There are many statistical procedures valid to use in antifeedant assays to establish the significance of the data obtained. In some cases feeding bioassays are based on a comparison of feeding on 1.2-cm-round paper penicillin assay disks. Penicillin disks are prepared by placing them on a clean enamelated tray, and then treated with 50 µl of test material in 95% ethanol (controls with carrier alone). Disks are allowed to dry and placed in 35 × 1.0 mm plastic petri dishes (1/dish). Each disk then receives 50 µl of 0.03M L-proline and 0.3M sucrose solutions. For example, sixth instars of spruce budworm (24- to 48-h-old) have been used in this type of bioassay and placed in petri dishes with test or control disks (three replicates or more). At 48 h after initiation of the test, the larvae were removed and frass pellets counted. The yellowish brown frass pellets derived from the artificial food in this insect species were easily differentiated from the white pellets derived from the paper disks. Frass pellets from artificial food were not included in counts (Bentley et al., 1984). The percentage of deterrence was calculated as:

1–

No. of pellets of frass from disks in teest × 100 No. of pellets of frass from disks in coontrol

WAFER ASSAY Wafer assays have been successfully used in assessing antifeedants against stored product insects (Nawrot et al., 1982; Harmatha and Nawrot, 1984). In this method wheat wafer disks (1 cm diameter) are immersed in test solution (in ethanol). After ethanol evaporation, stored grain insects (e.g., Tribolium, Trogoderma, Sitophilus, etc.) are provided with either two wafer disks immersed in ethanol (CC) or two disks saturated with test solution (TT), or they have a choice between ethanol or test material treated disks (CT). On the basis of amount of food consumed (on a weight difference basis) in the experiments where the insects have a possibility of choice, a relative index of deterrence is calculated (Paruch et al., 2000). The weight of food supersaturated with test material and consumed by the control group is used for the calculation of the absolute index. A compound for which both indices reach the value of 100 and the sum = 200 is called an ideal deterrent. The classification of indices is as follows: Below 0 0 – 50 51 – 100 101 – 150 151 – 200

= Attractant = Poor deterrent = Medium deterrent = Good deterrent = Very good deterrent

The coefficients are calculated as: Absolute coefficient of deterrency = [ (CC – TT)/(CC + TT) × 100 ] {A} Relative coefficient of deterrency = [ (C – T)/(C + T) × 100 ] {R} Total coefficient of deterrency = ∑ {A} + {R} A very specific wafer assay to evaluate feeding deterrents against plant weevils has been documented (Thomas and Bradley, 1975). In this assay compounds have been used on pith

© 2005 by CRC Press LLC

Insect Antifeedants

33

wafers of American elder, Sambucus camadensis L., in petri dishes; 2-mm-thick wafers were initially treated with some feeding stimulant, which would produce uniform feeding. The feeding deterrents were then added and the amount of feeding on the wafers treated with deterrent was compared with the feeding on control wafers (i.e., those treated with stimulants only). Wafers first treated with sucrose solution (15% dry weight) were dried for 1 hour at 65˚C. An ethanol extract of mature Loblolly pine phloem (20 mg/ml) was made and 0.1 ml pipetted evenly over the surface of each wafer and dried again at 65˚C for 1 hour. This was followed by the application of the feeding deterrent in appropriate solvent at the rate of 0.1 ml/wafer. The wafers were then dried to a constant weight for 1 hour at 70˚C under a vacuum. The wafers were then weighed to the nearest 0.1 mg and placed in petri dishes having a 1-cm-thick layer of moistened plaster of paris in the bottom. After feeding for 7 hours by the weevils, the wafers were removed, dried to constant weight for 1 hour at 70˚C under vacuum, and reweighed. The mean weight losses resulting from feeding were ranked by analysis of variance as the parameter of antifeedance.

ELECTROPHYSIOLOGICAL ASSAY Electrophysiological assays are modern techniques to evaluate the physiological deterrency of chemicals. The generalized approach is that two microelectrodes are inserted into the maxillary palp and the sensilla. The electrophysiological responses are then recorded with an oscilloscope as impulses/sec. by contacting the tips of sensilla with filter paper impregnated with the test material. The maximum number of impulses that can be evoked under treatment and control conditions are recorded. The functions of all the receptors, though not clear, it is now known that some respond to feeding deterrents (Ma, 1977; Nakanishi and Kubo, 1977), and is discussed separately in Chapter 2. For instance, a procedure adopted for Spodoptera litura larvae was that larvae were fixed ventral side upwards on a piece of plastic foam plate with three loops of steel around the neck, thorax, and the posterior abdomen segments. Recording was accomplished by means of stainless steel electrodes (5 µm diameter). The electrode was inserted into the proximal part of the hypopharynx by means of micromanipulator. This method permitted simultaneous recording of muscle potential and nerve impulses. Impulses were conducted proximally towards the CNS and also distally towards the hypopharynx tip. The effectiveness was based on the number of impulses. The test compounds were applied topically to various mouthparts and antennae by means of a fine glass capillary adapted to a microsyringe attached to a manual applicator. The recording was done on an oscilloscope to which a recording camera was attached (Antonious et al., 1984). Some specific techniques in electrophysiological evaluation have been used that could be generalized if tested against a variety of insect species. In the case of Ostrinia furnacalis, an attempt has been made to study the gustatory chemoreceptor in larval taste systems for bioassay-guided fractionation of antifeedants. This tip recording technique (Shang et al., 1993) has been used to detect the antifeedant activity by placing electrodes in contact with the tip of medial or lateral sensilla styloconica of the isolated maxillae. The nerve action potential yielding high frequencies of spikes are recorded and relative effects calculated. Jones (1979) established an automatic feeding detector (AFD) for use in evaluating insect angles to form a flat plate. This plate fits through a 1.4-cm-diameter hole drilled in the center of a small petri dish base. The dish is maintained on two small pneumatic pistons (disposable 2 ml syringes)—one fixed and the other free—that permit the dish height and the alignment of the trembler to be adjusted. A lid covers the petri dish base with a gauze guard (the guard prevents the larva from moving over the disk). At an approximate central position on the trembler is located a needle fixed to the central brass rod of a clamped linear differential

© 2005 by CRC Press LLC

34

Opender Koul

transductor. The transductor output is connected to a chart recorder and a light source positioned over the petri dish. The whole apparatus is mounted on a leveled, vibration-free bench. Movement of the trembler results in a deflection on the chart recorder. This deflection is calibrated by adding known weights to the end plate and by moving the end plate a vertical distance. Changes in weight are measurable at an accuracy of ±10 mg from 20 to 500 mg and ±30 mg from 500 to 1000 mg. Vertical distance is measurable at an accuracy of ±0.2 mm over the range of 0.2 to 3.6 mm. A pre-weighed leaf disk is fixed to the end plate using a small amount of inert silicon grease, and the recorder offset adjusted to center the pen at mid-scale. A single insect larva is introduced into the petri dish and behavioral observations made over a 30- to 40-minute period. A number of such trials are carried out to establish a correlation between traces and feeding behavior. The behavior of larvae on disks treated with deterrent chemicals results in characteristic traces, which differ from those obtained with natural host-plant controls. Quantitative measurements of consumption, duration of feeding, number of test bites, and number of rejections of the disk are then made and compared. This detector system is cheap to construct, is robust in operation, and gives rapid analyses. This method can be used with any insect capable of causing a deflection of greater than 0.2 mm. Another interesting device has been used to study aphids. Synthetic diets treated with chemicals (like DIMBOA) are prepared just prior to the recording assay. The diet is injected into plastic vial caps (300 µl) over which is stretched a parafilm membrane. A platinum wire, inserted through the plastic cap into the diet, serves as the voltage input electrode. A gold wire fixed to the dorsum of the test aphid with silver conductive paint serves as a lead to the input of the amplifier. The analysis is based on the duration of I-waves (ingestion), S-waves (salivation), and non-probing over 2 h assay period (Argandona et al., 1983). A similar useful electronically recorded feeding behavior for rice hoppers, Nephotettix virescens, is known to evaluate the efficacy of allelochemicals (Saxena and Khan, 1985). In this experimental bioassay a 5 cm fine (18 µm) gold wire is attached by a small quantity of silver paint to the dorsum of an 8- to 10-h-old female reared on a virus-free 45-d-old TN1rice plant. Before attaching the wire, insects are anesthetized with carbon dioxide gas. The insects are starved but water satiated for 2 hours and then placed on an intact leaf blade of a treated or control plant. The gold wire is connected directly to the negative input terminal of a transistorized automatic null balancing DC chart recorder having 250 nm recording width and input resistance of 1MΩ. The voltage source consists of two 1.5v DC batteries connected in series. The positive battery terminal is connected with plant roots through moistened filter paper and an aluminium foil. The negative battery terminal is connected directly to the positive input terminal of the chart recorder. The recorder pen is adjusted to the chart baseline and insect feeding is monitored for 180 minutes. A chart speed of 1.5 cm/min at 500 mV amplifier power is adequate for distinguishing various waveforms and associated voltage reversals. Each treatment including the control is replicated ten times. The evaluation is based on the reduction in phloem feeding compared with controls. Phloem feeding shows erratic response in treated plants, evidenced by repeated voltage reversals in associated waveforms, which indicates the antifeedant behavior (Saxena and Khan, 1985).

OTHER MISCELLANEOUS METHODS Simulation bioassay Higgins and Pedigo (1979b) developed a simulation bioassay procedure to demonstrate feeding deterrence in phytophagous insects. This method involved rearing of two cohorts of green clover worm larvae, Plathypene scabra (F), under optimal conditions, but separated in

© 2005 by CRC Press LLC

Insect Antifeedants

35

time of oviposition by 3 days. Thereafter the mean development of the cohort to be used in the chemically defined treatments remained 3 days ahead of the average age and development of the physically stressed individuals. Trifoliates of the primary host, soybean (Glycine max), always were available in amounts exceeding requirements during the pre-experimental rearing period. Twenty-four hours before experimental initiation, larvae were removed from the growth chamber and allowed to acclimatize to room temperature. Upon reaching fourth stage plus 1 day, larvae in the chemically treated cohort were individually presented premeasured leaves selectively dip treated with solutions of guazitine triacetate and allowed to feed for 48 hours in 0.55 liter paper cartons. No significant larval weight differences existed at this stage (test initiation). Every 48 hours until pupation or death, leaves were replaced with equivalent treated trifoliates. Damaged leaves were removed and remeasured to determine the mean of 48 h consumption (An) under chemical stress. This cycle was then replaced until pupation or population extinction through toxicity or starvation. Upon reaching 4th stage +1 larvae of the second cohort (physically stressed) were presented untreated leaf material (Bn) equal in area to the mean daily consumption ±SE of the equivalently aged larvae feeding on chemically treated leaves. Preliminary trials revealed that split offerings of the food into 24 h treatments (Bn and Bn') would more closely model the stress imposed by the antifeedant. Area Bn' = An – (actual mean of Bn), ±1SE of An/2 (maximum of 1 cm2). In this way equal amounts of food were available for consumption by each cohort. This delayed cycle of leaf exchange was continued until pupation or population death or until forced absolute starvation was caused by the lack of feeding by the chemically treated cohort. Larvae of the physical cohort were individually placed in 9.53 × 6.99 × 2.24 cm ventilated, clear plastic snap-lid boxes. To obtain the proper untreated area, leaves were measured and then remeasured after the tip of the terminal leaflet was turned under to simulate feeding loss. The area to be presented to the physical cohort (the area turned under) was found by subtraction. The area turned under was then adjusted, by allowing the leaf to refold until it was within the acceptable bounds set by the average daily consumption of the corresponding chemically treated cohort. The leaflet was then slightly creased and the cage was snapped shut on the fold line (turned under area inside larval cage). As area determination continued, an effort was made to maintain equal numbers of larvae above and below the true desired mean. Thus, mean areas presented to the physical cohort were equivalent ±1SE to that consumed by individuals of the chemical cohort. Two trials exhibited feeding restriction of 40 to 75% depending upon the level of treatment (Higgins and Pedigo, 1979b).

Dipteran Assays In a test against Phormia regina, an adjustable pipette rack was fashioned from wood to hold pipettes and vials (each vial containing one fly). Disposable pipettes of 100 µl capacity in 10 µl graduation were used in combination with 25 ml vials fitted with plastic caps. A central hole for entry of the pipette tip and several smaller ventilation holes were drilled in each vial cap. Male and female flies were used, raised on beef liver and maintained on skim milk powder and sugar. Test compounds were mixed with 0.5 molar stock solution (control) of sucrose at a concentration of 0.01 M and stored at 4˚C in darkness for the duration of each trial (6 days). To ensure that the fly could easily ingest the test solution, the pipette racks were adjusted to an angle that allowed the solution to be readily absorbed on Whatman No. 1 filter paper. Newly emerged adults (1, 3, and 5 days old) were anesthetized with carbon dioxide gas prior to transfer to the vials in which they were given one day to acclimate and starve. Eleven

© 2005 by CRC Press LLC

36

Opender Koul

of the twelve pipettes on each rack were connected to vials containing flies and the 12th served as an evaporation control. Three of the four racks held pipettes with alkaloid diets and the fourth held the 0.5 M solution of sucrose control. Readings were taken 6 hours and 24 hours after the beginning of the trial to determine the volume consumed to the nearest microliter (Blades and Mitchell, 1986).

WireWorm (Melanotus communis) Assay For wireworms both choice and no-choice designs have been adopted (Villani and Gould, 1985). In the choice design two experiments were performed. Plexiglas containers of 2.5 cm3 (Experiment 1) and 30 × 15 × 4 cm (Experiment 2) were filled with moist soil. A treated potato section (Experiment 1) or treated corn seed (Exp eriment 2) was placed 2.5 cm below the surface at one end of each container and a control for each at the opposite end. A single corn wireworm was buried head downwards in the center of the container; 25 containers were used in Experiment 1 and 50 containers in Experiment 2. The containers were placed in the dark, and after 36 hours (Experiment 1) and 8 days (Experiment 2) baits were checked for feeding damage. A potato section or corn seed and sprout, which showed any indication of having been fed upon, was considered damaged. The no-choice test design was as above with the exception that only one bait, either treated or untreated (control), was placed in each Plexiglas container.

Boll Weevil Assay A 1% solution of test fraction or compound was used, and an unpunctured debracted bud from a greenhouse-grown cotton plant was dipped momentarily in the solution. Ten 1- or 2-day-old adult boll weevils unfed from time of emergence or starved for 24 hours were placed in a petri dish with one treated and one untreated bud (dipped in solvent alone) and held for 4 hours. Five control and five test dishes were prepared for each test. The number of feeding punctures per bud was counted under a dissecting microscope (Bird et al., 1987).

Scale Insect Assay Test for red scale, Aonidiella aurantii, and yellow scale, A. citrina, have been conducted by spraying (with an atomizer) half of a green lemon fruit with a dilution of 0.01 to 1.0% as an aqueous emulsion and covering the other half with parafilm to protect it. The crawlers were allowed to migrate from heavily infested lemons onto the treated lemons. The insects were counted after 5 and 25 days post-treatment. A similar test has been conducted for citrus red mite, Panonychus citri (Jacobson et al., 1978).

Sawfly Assay A standard bioassay for sawflies used a 7- to 10-cm twig of 1-year-old jack pine foliage. Foliage was stripped from the twig until ten pairs of needles remained at each end. Twigs were rinsed in distilled water and allowed to dry. Needles at one end of each twig were covered with the resin acid/methanol solution by pipetting a few drops at the base of the needle and allowing it to flow to the tip. Needles on the other end were treated with solvent only. A similar twig received the solvent at one end and nothing at the other, to serve as an additional control. The treated needles were allowed to dry at room temperature for a minimum of 30 minutes. The treated twig was suspended horizontally on an insect pin that passed through the center of the twig; the pin was then inserted through 6.2 cm2 moss-green

© 2005 by CRC Press LLC

Insect Antifeedants

37

paper and into a No. 11.5 rubber stopper. Five 4th or 5th instar sawflies, Neodiprion dubiosus, N. lecontei, and N. rugifrons, were placed on the needles at each end of the twig. Replicates were placed in an environmental chamber at 20˚C under continuous light. A test concentration was recognized as biologically active in inhibiting larval feeding if > 70% of the total number of larvae settled on the end treated with solvent only after 4 hours (Schuh and Benjamin, 1984).

Leaf Beetle Assay Matsuda and Senbo (1986) described a method to test feeding deterrents against leaf beetles of various families. The bioassay procedure was based on a comparison of nibbling by adult beetles on 2 × 2 cm filter papers. The assay chamber was composed of a plastic petri dish of 9 cm diameter and 1.5 cm in depth. Three 7-cm-diameter filter papers were immersed in 2 ml of water and were placed on the bottom of the petri dish, and a doughnut-like plastic disk (3 cm ID and 7 cm OD) was put on the papers. Four pieces of filter paper per dish were equidistantly placed on the doughnut-like disk. Filter papers were treated with 0.075 ml of test chemical in methanol; control papers were treated with carrier alone. Papers were allowed to dry for several minutes, then placed in a petri dish, four to a dish. Twenty adult beetles starved for 24 hours were released into a petri dish with test or control material. All assays were completed in darkness at 24 to 25˚C. Five replicates of test and controls were run for each compound tested. Twenty-four hours after release of the beetles, the feeding response was judged from the differences between test and control pieces and expressed as (- - -) = no nibbling, (- -) = less nibbling than controls, (±) = same nibbling as in controls, and (+) = more nibbling than controls. A negative score was considered to indicate feeding deterrence.

Oral Dosing In this procedure a preliminary assessment is made for potential feeding deterrent compounds. In fact, several earlier studies have shown that deterrent chemicals could be consumed after elimination of chemoreceptors. Recently cannulas have been used to place test compounds into the gut lumen via the oesophagous in order to avoid behavior rejection in some grasshoppers (Cottee et al., 1988). Sutherland et al. (1981) have also used cannulas for oral dosing of scarabid grubs. This technique, however, has problems particularly in handling, which induces deleterious and damaging effects to the foregut during ingestion. Another problem is that doses are necessarily sporadic and do not mimic normal intake patterns. Semi-microgelatin capsules have been successfully used with grasshoppers (Cottee et al., 1988), which conceals the taste completely and there is rapid release in the gut. However, this technique is not feasible for smaller insects and is time consuming. Another technique of oral dosing for feeding deterrents was recently demonstrated for various insect species (Usher et al., 1989). In this method deterrent solutions are enclosed inside lipid vesicles and suspended in non-deterrent solutions that can be offered to the insects to drink. Liposomes containing an aqueous solution can be formed in a number of ways, such as reverse phase evaporation, where reverse micelles are formed of phosphatidylcholine around the aqueous phase in an excess of diethyl ether. The ether is evaporated, causing the lipid to form bilayer vesicles, or liposomes, which are sized to approximately 1 µm in diameter by passage through a polycarbonate filter. The liposomes can be separated from any remaining deterrent in the surrounding medium by passage through a gel filtration column. The liposomes are collected in the void volume of the eluting solvent, while deterrent molecules not inside liposomes are retained in the column.

© 2005 by CRC Press LLC

38

Opender Koul

In the assay each insect is weighed immediately before and after a drinking bout and the volume drunk determined by the weight change. Amounts drunk by control and treatment groups of insects are compared. Thus this technique is a very useful approach to dose insects with compounds that normally deter feeding. Some other processes on similar lines use microspheres, microcoating, and molecular encapsulation (Usher et al., 1989; Clancy et al., 1992).

Field Trials Not many studies are available where feeding deterrents have been used in large-scale field trials. The vast majority of feeding bioassays with plant-derived chemicals have been performed under laboratory (or in some cases greenhouse) conditions, although field tests of metal-containing fungicides as antifeedants have been done (Jermy and Matolcsy, 1967; Ascher, 1979). Though field tests have confirmed laboratory results for some naturally occurring chemicals (All and Benjamin, 1976; Metcalf et al., 1980), laboratory results cannot necessarily be extrapolated to field conditions due to: • • • • •

Alteration of the chemicals in the field by environmental factors Habituation of insects Use of artificial substrates in the laboratory experiments Insect movement away from treated plants Composition of the plant community (Lewis and van Emden, 1986)

In fact there is no specific procedural design devised for field evaluation of antifeedants. The procedures are similar to those used for conventional chemical pesticides and have been followed for various field trials in a similar fashion. Among the antifeedant plant chemicals, the most extensive field trials have been carried out with neem products, mostly due to broad international cooperation (Schmutterer and Ascher, 1987). Ladd et al. (1978) carried out the earliest neem seed extract evaluation in the field against Japanese beetles, Popillia japonica, using plots consisting of four to five plants. Similarly, field trials with Colorado potato beetles, European corn borer, diamond back moth, and various aphids have been conducted to show reasonable control by foliar application of neem extracts and azadirachtin (Isman et al., 1991; Lowery et al., 1993). The proceedings of the World Neem Conference held in Bangalore, India, in 1993 (Singh et al., 1996) reports 24 papers on successful field application of various neem-based formulations. These results show the effect of these formulations on a variety of insects like grasshoppers, rice pests, bean flies, various Helicoverpa species, cotton pests, pod borers, maize pests, fruit flies, mango hoppers, and others. Such studies definitely demonstrate the potential of antifeedants on a large scale. However, there is a need to evaluate other identified antifeedant compounds in the field to determine the practical application potential of such compounds.

REFERENCES Alford, A.R. and Bently, M.D. (1986) Citrus limonoid as potential antifeedants for the spruce budworm (Lepidoptera: Tortricidae). J. Econ. Entomol., 79, 35–38. All, J.N. and Benjamin, D.M. (1976) Potential of antifeedants to control larval feeding of selected Neodiprion saw flies (Hymenoptera: Diprionidae). Can. Entomol., 108, 1137–1143. Antonious, A.G., Saito, T., and Nakamura, K. (1984) Electrophysiological response of the tobacco cutworm, Spodoptera litura (F), to antifeedant compounds. J. Pestic. Sci., 9, 143–146.

© 2005 by CRC Press LLC

Insect Antifeedants

39

Ascher, K.R.S. (1979) Fifteen years (1963–1978) of organotin antifeedants: a chronological bibliography. Phytoparasitica, 7, 117–137. Ascher, K.R.S. and Meisner, J. (1973) Evaluation of methods for assay of phagostimulants with Spodoptera littoralis larvae under various conditions. Entomol. Exp. Appl., 16, 101–114. Ascher, K.R.S., Schmutterer, H., Glotter, E., and Kirson, I. (1981) Withanolides and related ergostane-type steroids as antifeedants for larvae of Epilachna varivestis (Coleoptera: Chrysomelidae). Phytoparasitica, 9, 197–205. Argandona, V.H., Corcuera, L.J., Niemeyer, H.M., and Campbell, B.C. (1983) Toxicity and feeding deterrency of hydroxamic acids from graminae in synthetic diets against the green bug, Schizaphis graminum. Entomol. Exp. Appl., 34, 134–138. Bentley, M.D., Leonard, D.E., Stoddard, W.F., and Zalkow, L.H. (1984) Pyrrolizidine alkaloids as larval feeding deterrents for spruce budworm, Choristoneura fumiferana (Lepidoptera: Tortricidae). Ann. Entomol. Soc. Am., 77, 393–397. Berenbaum, M. (1986) Post ingestive effects of phytochemicals on insects: On paracelsus and plant products. In J.R. Miller and T.A. Miller (eds.), Insect Plant Interactions, Springer Verlag, New York, pp. 121–153. Bernays, E.A. and Chapman, R.F. (1978) Plant chemistry and acridoid feeding behaviour. In J.H. Harborne (ed.), Biochemical Aspects of Plant and Animal Coevolution, Academic Press, New York, pp. 99–141. Bernays, E.A. and DeLuca, C. (1981) Insect antifeedants, properties of an iridoid glycoside: ipolamide. Experientia, 37, 1289–1290. Blades, D. and Mitchell, B.K. (1986) Effect of alkaloids on feeding by Phormia regina. Entomol. Exp. Appl., 41, 299–304. Blaney, W.M., Simmonds, M.S.J., Evans, S.V., and Fellows, L.E. (1984) The role of secondary plant compounds 2,5-dihydroxymethyl-3,4-dihydroxy pyrrolidine as a feeding inhibitor for insects. Entomol. Exp. Appl., 36, 209–216. Bird, T.G. and Hedin, P.A. (1986) An improved feeding bioassay for the boll weevil (Coleoptera: Curculionidae). J. Econ. Entomol., 79, 882–886. Bird, T.G., Hedin, P.A., and Burks, M.L. (1987) Feeding deterrent compounds to the boll weevil, Anthonomus grandis Boheman in Rose-of-Sharon, Hibiscus syriacus L. J. Chem. Ecol., 13, 1087–1097. Clancy, K.M., Foust, R.D., Huntsberger, T.G., Whitaker, J.G., and Whitaker, D.M. (1992) Technique for using microencapsulated terpenes in lepidopteran artificial diets. J. Chem. Ecol., 18, 543–560. Cottee, P.K., Bernays, E.A., and Mordue, L. (1988) Comparisons of deterrency and toxicity of selected secondary plant compounds to an oligophagous and a polyphagous acridid. Entomol. Exp. Appl., 46, 241–247. Dreyer, D.L., Reese, J.C., and Jones, K.C. (1981) Aphid feeding deterrents in sorghum: Bioassay, isolation and characterization. J. Chem. Ecol., 7, 273–284. Escoubas, P., Lajide, L., and Mitzutani, J. (1993) An improved leaf-disk antifeedant bioassay and its application for the screening of Hokkaido plants. Entomol. Exp. Appl., 66, 99–107. Harmatha, J. and Nawrot, J. (1984) Comparison of the feeding deterrent activity of some sesquiterpene lactones and a lignin lactone towards selected insect storage pests. Biochem. Syst. Ecol., 12, 95–98. Higgins, R.A. and Pedigo, L.P. (1979a) Evaluation of guazatine triacetate as an antifeedant feeding deterrent for the green cloverworm on soybean. J. Econ. Entomol., 72, 680–686.

© 2005 by CRC Press LLC

40

Opender Koul

Higgins, R.A. and Pedigo, L.P. (1979b) A laboratory antifeedant simulation bioassay for phytophagous insects. J. Econ. Entomol., 72, 238–244. Hosozawa, S., Kato, N., Munakata, K., and Chen, Y.L. (1974) Antifeedant substances for insects in plants. Agric. Biol. Chem., 38, 1045–1048. Isman, M.B., Koul, O., Arnason, J.T., Stewart, J., and Salloum, G.S. (1991) Developing a neem based insecticide for Canada. Mem. Ent. Soc. Can., 159, 39–47. Jacobson, M., Reed, D.K., Crystal, M.M., Moreno, D.S., and Soderstrom, E.L. (1978) Chemistry and biological activity of insect feeding deterrents from certain weed and crop plants. Entomol. Exp. Appl., 24, 448–457. Jermy, T. and Matolcsy, G. (1967) Antifeedant effects of some systemic compounds on chewing phytophagous insects. Acta. Phytopath. Acad. Sci. Hung., 2, 219–224. Jermy, T., Hanson, F., and Dethier, V. (1968) Induction of specific food preference in lepidopterous larvae. Entomol. Exp. Appl., 11, 211–230. Jones, C.G. (1979) An automatic feeding detector (AFD) for use in insect behaviour studies. Entomol. Exp. Appl., 25, 112–115. Jones, C.G. and Coleman, J.S. (1988) Leaf disk size and insect feeding preference, implications for assays and studies on induction of plant defense. Entomol. Exp. Appl., 47, 167–172. Koul, O. (1993) Plant allelochemicals and insect control: An antifeedant approach. In T.N. Ananthakrishnan and A. Raman (eds.), Chemical Ecology of Phytophagous Insects, Oxford & IBH Publishing Co. Pvt. Ltd., pp. 51–80. Koul, O., Smirle, M.J., and Isman, M.B. (1990) Asarones from Acorus calamus L. Oil: Their effect on feeding behaviour and dietary utilization in Peridroma saucia. J. Chem. Ecol., 16, 1911–1920. Koul, O., Shankar, J.S., and Mehta, N. (1997) Antifeedant activity of neem seed extracts and azadirachtin to cabbage aphid Brevicoryne brassicae (L). Ind. J. Expt. Biol., 35, 994–997. Ladd, T.L., Jacobson, M., and Buriff, C.R. (1978) Japanese beetles: extracts from neem tree seeds as feeding deterrents. J. Econ. Entomol., 71, 810–813. Lewis, A.C. and Bernays, E.A. (1985) Feeding behaviour selection of both wet and dry food for optimal growth by Schistocerca gregaria nymphs. Entomol. Exp. Appl., 37, 105–112. Lewis, A.C. and van Emden, H.F. (1986) Assays for insect feeding. In J.R. Miller and T.A. Miller (eds.), Insect Plant Interactions, Springer Verlag, New York, pp. 95–119. Lidert, Z., Taylor, D.A.H., and Thirungnanam, M. (1985) Insect antifeedant activity of four prieurianin type limonoids. J. Nat. Prod., 48, 843–845. Lowery, D.T. and Isman, M.B. (1993) Antifeedant activity of extracts from neem, Azadirachta indica to strawberry aphid Chaetosiphon fragaefolii. J. Chem. Ecol., 19, 1761–1773. Lowery, D.T., Isman, M.B., and Brard, N.L. (1993) Laboratory and field evaluation of neem for the control of aphids (Homoptera: Aphididae). J. Econ. Entomol., 86, 864–870. Ma, W.-C. (1977) Alteration of chemoreceptor function in armyworm larvae (Spodoptera exempta) by a plant-derived sesquiterpenoid and by sulfhydral reagent. Physiol. Entomol., 2, 199–207. Matsuda, K. and Senbo, S. (1986) Chlorogenic acid as a feeding deterrent for the Salicaceae feeding leaf beetle, Lochmaeae capreae cribrata (Coleoptera: Chrysomelidae) and other species of leaf beetles. Appl. Ent. Zool., 21, 411–416. Meisner, J. and Ascher, K.R.S. (1973) Attraction of Spodoptera littoralis larvae to colours. Nature London, 242, 332–334. Messchendorp, L., van Loon, J.J.A., and Gols, G.J.Z. (1996) Behaviour and sensory responses to drimane antifeedants in Pieris brassicae larvae. Entomol. Exp. Appl., 79, 195–202.

© 2005 by CRC Press LLC

Insect Antifeedants

41

Metcalf, R.L., Metcalf, R.A., and Rhodes, A.M. (1980) Cucurbitacins as kairomones for diabroticite beetles. Proc. Natl. Acad. Sci. USA, 77, 3769–3772. Mittler, T.E. and Dadd, R.H. (1962) Artificial feeding and rearing of the aphid, Myzus persicae (sulzer) on a completely defined synthetic diet. Naturre London, 195, 404. Nakanishi, K. and Kubo, I. (1977) Studies on warburgnal, muzigadial and related compounds. Israel J. Chem., 16, 28–31. Nawrot, J., Bloszyk, E., Grabarczyk, H., and Drozdz, B. (1982) Deterrent properties of sesquiterpene lactones for the selected storage pests. Proc. Nauk. Inst. Ochr. Roslin, 24, 27–36. Numata, A., Katsuno, T., Yamamoto, K., Tomoko, N., Tsuruko, T., and Seto, K. (1984) Plant constituents biologically active to insects IV. Antifeedants for the larvae of the yellow butterfly, Eurema hecabe mandarina in Arachniodes standishii. Chem. Pharm. Bull., 32, 325–331. Paruch, E., Ciunik, Z., Nawrot, J., and Wawrzencyzk, C. (2000) Lactones. 9. Synthesis of terpenoid lactones: active insect antifeedants. J. Agric. Food Chem., 48, 4973–4977. Reed, D.K. and Jacobson, M. (1983) Evaluation of aromatic tetrahydropyranyl ethers as feeding deterrents for the striped cucumber beetle, Acalymma vittatum. Experientia, 39, 378–380. Reese, J.C. and Field, M.D. (1986) Defence against insect attack in susceptible plants. Black cutworm (Lepidoptera: Noctuidae) growth on corn seedling and artificial diets. Ann. Entomol. Soc. Am., 79, 372–376. Rose, A.F., Jones, K.C., Hadden, W.F., and Dreyer, D.L. (1981) Grindelane diterpenoid acids from Grindelia humilis, feeding deterrency of diterpene acids towards aphids. Phytochemistry, 20, 2249–2255. Rowan, D.D. and Gaynor, D.L. (1986) Isolation of feeding deterrents against argentine stem weevil from ryegrass infected with the endophyte. J. Chem. Ecol., 12, 647–658. Saxena, R.C. and Khan, Z.R. (1985) Electronically recorded disturbances in feeding behaviour of Nephotettix virescens (Homoptera: Cicadellidae) on neem oil treated rice plants. J. Econ. Entomol., 78, 222–226. Scheffrahn, R.H. and Rust, M.K. (1983) Dry wood termite feeding deterrents in sugar pine and antitermite activity of related compounds. J. Chem. Ecol., 9, 39–55. Schmutterer, H. and Ascher, K.R.S. (1987) Natural pesticides from neem tree (Azadirachta indica A. Juss) and other tropical plants. Proc. 3rd Int. Neem Conf., Rauischholzhausen, GTZ, Eschborn, Germany. Schoonhoven, L.M. (1982) Biological aspects of antifeedants. Entomol. Exp. Appl., 31, 57–69. Schuh, B.A. and Benjamin, D.M. (1984) The chemical feeding ecology of Neodiprion dubiosus Schedl., N. rugifrons Midd., and N. lecontei (Ritch.) on Jack Pine (Pinus banksiana Lamb.). J. Chem. Ecol., 10, 1071–1079. Shang, Z., Zhao, W., Zhu, Y., and Li, Q. (1993) The use of electrophysiological technique to explore antifeedants in plants. Prog. Natural. Sci., 3, 530–534. Singh, R.P. and Pant, N.C. (1980) Lycorine: A resistant factor in the plants of subfamily Amaryllidodieae (Amaryllidaceae) against desert locust, Schistocerca gregaria. Experientia, 36, 552–553. Singh, R.P., Chari, M.S., Raheja, A.K., and Kraus, W. (1996) Neem and Environment, Vol. 1, Oxford & IBH Publishing Co. Pvt. Ltd., New Delhi. Sutherland, O.R.W., Hutchnis, R.F.N., Russel, G.B., Lane, G.A., and Biggs, D.R. (1981) Biochemical plant resistance mechanisms: an evaluation of basic research. In K.E. Lee (ed.), Proc. Third Aust. Conf. Grassl. Invert. Ecol., S.A. Govt. Printer, Adelaide, pp. 245–253.

© 2005 by CRC Press LLC

42

Opender Koul

Thomas, H.A. and Bradley, E.L. (1975) Feeding deterrents for the pales weevil in a laboratory bioassay. J. Econ. Entomol., 68, 147–149. Usher, B.F., Bernays, E.A., Barbehenn, R.V., and Wrubel, R.P. (1989) Oral dosing of insects with feeding deterrent compounds. Entomol. Exp. Appl., 52, 119–133. Villani, M.G. and Gould, F. (1985) Screening of crude plant extracts as feeding deterrents of the corn wire worm, Melanotus communis. Entomol. Exp. Appl., 37, 69–75. Warthen, J.D. Jr., Redfern, R.E., Uebel, E.C., and Mills, G.D. Jr. (1982) Antifeedant screening of 39 local plants with fall armyworm larvae. J. Environ. Sci. Health, 17A, 885–895. Wolfson, J.L. (1988) Bioassay techniques: An ecological perspective. J. Chem. Ecol., 14, 1951–1963. Wolfson, J.L. and Murdock, L.L. (1987) Method for applying chemicals to leaf surfaces for bioassay with herbivorous insects. J. Econ. Entomol., 80, 1334–1336. Zalkow, L.H., Gordon, M.M., and Lanir, N. (1979) Antifeedants from rayless goldenrod and oil of pennyroyal: Toxic effects for the fall armyworm. J. Econ. Entomol., 72, 812–815.

© 2005 by CRC Press LLC

4 STRUCTURE-ACTIVITY RELATIONSHIPS Biologically active compounds are found among all major classes of secondary plant substances and especially among the higher oxidized metabolites. The diversity of structures is usually large, and many of the biological effects are interpreted in terms of plant defense against predation and pathogenecity. Insect antifeedants are one of such biologically active substances that induce the cessation of feeding either temporarily or permanently. Various aspects of behavioral physiology in response to such chemicals have been described in the previous chapter, but quantitative structural-activity relationships have posed several problems. The overall picture that emerges from various evaluations shows that small structural variations can produce drastic changes in the activity profile of compounds. A critical examination of functional groups present in the active molecules provides crucial information about the optimal relative stereochemistry required to stimulate an antifeedant response in insects, but an analysis based on functionality and skeletal types appears to be difficult to produce any generalization, albeit one can certainly discuss activity variations within a skeletal type. The main aim of this chapter is directed in this direction and generalizes sufficient structure-activity information within specific skeletal systems to allow rational modification of readily available feeding deterrents to be made into potential insect control agents. Accordingly some specific class of compounds will be discussed in terms of structural-activity relationships to bring forth some generalizations.

LIMONOIDS This group of compounds, today, are considered to be most potential antifeedant allelochemicals that could be introduced in an Insect Pest Management (IPM) system. A specific example like azadirachtin (1), a tetranortriterpenoid from the Indian neem tree Azadirachta indica, is

© 2005 by CRC Press LLC

44

Opender Koul

known for its potent antifeedant characteristics (Koul, 1992). It is widely reported that azadirachtin affects the feeding behavior of many insect species (Koul, 1996), and a variation in the structure influences the activity of this compound (Blaney et al., 1990; Rembold, 1989; Ley et al., 1993). For instance, hydrogenation of the dihydrofuran ring as in dihydroazadirachtin (4) does not affect the activity of the molecule, and esters on the A ring do not affect the activity of the compound (Yamasaki and Klocke, 1987), although they could be important in transporting the compounds to the receptor sites. The difference in the level of antifeedance, for instance, among compounds 1 to 7 evaluated against four noctuid larvae (Blaney et al., 1990), has been attributed to the ability of the respective esters at C-1 or C-3 to transport the molecule to the target site. COOCH3 R1O

COOCH3

R4

O

OH

R1O

11

22

O

1

1

3

O OR3

R2O O

O

23

3

O

O

OR3

R 2O O

H3COOC

O (1) R1 =

OH

11

O

H3COOC

R4

O

O , R2 = Ac, R3 = H, R4 = OH

(4) R1 =

, R2 = Ac, R3 = H, R4 = OH

O (2) R1 = H, R2 =

, R 3 = R4 = H

(5) R1 = 2 methylbutenate, R2 = Ac, R3 = H, R4 = OH (6) R1 = pyruvate, R2 = Ac, R3 = H, R4 = OH

(3) R1 = R2 = R3 = R4 = H (7) R1 = R3 = H, R2 = Ac, R4 = OH

However, change in C-1 or C-3 ester in combination with a structural variation at C-11 resulted in decrease of feeding deterrent activity, especially in H. armigera. This suggests that the type of ester present at C-11 is important. These results also show that hydrogenation of C-22,23 double bond in azadirachtin does not significantly influence antifeedant activity, thus confirming the observations of Yamasaki and Klocke (1987). Hein et al. (1999) also report the hydroxy group at C-11 in azadirachtin A is important for high mortality rates, and a single bond between C-22 and C-23 increases the degree of efficiency. An exchange of the large ester group ligands at C-1 and C-3 with hydroxy groups in combination with a single bond between C-22 and C-23 and a hydroxy group at C-11 leads to high feeding activity and a degree of efficiency of about 100 percent. Ley and his co-workers have synthesized a large number of compounds to establish structure-activity relationships. Ley et al. (1993) report screening of 31 compounds related to azadirachtin against Spodoptera littoralis and point to the importance of hydroxyfuranacetal moeity in the high level of potency of this compound. Stereochemistry at C-7 is crucial, and the bridging oxygen substituent at C-6 may play some role. The precise spatial and electrostatic requirements of all the various oxygen substituents, according to Ley, need more detailed studies. These studies also reveal reduction in activity by increasing bulk at C-23. However, similar things do not hold true for other evaluated species like S. frugiperda or H. armigera. In fact the bulky isopropoxy substitute results in a compound with very potent antifeedant activity against S. frugiperda (Blaney et al., 1990) and less bulky ethoxy substitution quite active against H. armigera.

© 2005 by CRC Press LLC

Insect Antifeedants φ8

O O O

φ7 OH O 11 φ11 O D

1 O

φ3

φ4

CH3C

3

φ6

B

A

OH φ12 F

φ9 O φ10

C C φ5

8

13 E 14

7

4

O CH3O

OCH3

C

φ2 φ1

45

20

G O

OH

O

O

FIGURE 4.1 Azadirachtin key structural features. The designation of torsional angles and notation of the different rings are shown.

These studies, therefore, imply that large numbers of functional groups present in azadirachtin and the sensitivity of azadirachtin in different bioassays exhibit variable results. Therefore, the question still remains that specificity of structural features responsible for such an activity remains unpredicted. A recent theoretical study on the conformations of azadirachtin has provided some cruical information by critically examining the functional groups present in this compound and its methyl ether derivatives, which provide the data about the optimal relative conformations required to stimulate antifeedant response (Baldoni et al., 1996). In this study, uniform scanning using molecular mechanics calculations were carried out, and accordingly the conformations arising from the combinations of torsional analysis φ – φ (Figure 4.1) were considered (Dewar et al., 1985). In order to obtain a structure-activity relationship, the results obtained for azadirachtin were compared with those for its 7-methyl ether derivatives (Figure 4.2). In case of azadirachtin the substituents at C-1, C-3, C-4, and C-11 showed a moderate but significant conformational flexibility; in contrast, the tricyclic dihydrofuran ring showed a restricted rotation about the single bond with a highly preferred conformation at φ 9 ≈ 70°. Similar conformations were obtained for methyl ether derivatives. However, different results were obtained for the energy profiles, which reflected the influence of the dihydrofuran ring orientation. After comparing various results it was obvious that OH groups at C-11 and C-20 could collectively play a significant role in conferring the appropriate structural conformation and thus significant decrease in activity. Similarly the monomethyl substituent on C-7, C-11, or C-20 may not be critical to confer the structural conformation leading to biological activity for these compounds. In contrast, the presence of two methyl groups at C-11 and C-20 respectively or trimethylation introduces important changes in the conformational behavior of these compounds, which may be responsible for the lack of the activity (Baldoni et al., 1996). Similar studies with 3-tigloyl azadirachtol and other derivatives of azadirachtin have established that lack of antifeedant activity of 7-keto derivative and other inactive compounds can be explained on the basis of their different conformational behavior (Baldoni et al., 1997). For instance, on the three 2D conformational energy maps a total of 18 conformations were selected on energetic grounds for 3-tigloyl azadirachtol from molecular mechanism calculations. These data together with other experimental findings on the antifeedant activity of closely related compounds suggest that specific ester groups are not required at positions C-1 and C-3 in the azadirachtin nucleus in order to maintain a high level of activity. However, the presence of certain ester groups (e.g., tigloyl) at these positions may provide a favorable hydrophilic/lipophilic balance, necessary for optimum transport across various membranes

© 2005 by CRC Press LLC

46

Opender Koul COOCH3

O O

O

COOCH3

O

OH

OH

O

11

OCH3

O

O

O

13 14

1

20 O

O OCH3

CH3COO

3

H

COOCH3

O OH O

11

OH

O

O

O

13 14

20

7

O

O OH

3

H

OCH3

AZADIRACHTIN-7,20-METHYLETHER

COOCH3

COOCH3

O

OH

OCH3

O

O

13 14

20

1

8

H3COOC

O

O OH

CH3COO

13 14

20

7

7 4

OCH3

11 O

3

OCH3

O

11 1

H

O

H3COOC

O

O

O

CH3COO

AZADIRACHTIN-11-METHYLETHER

O

7

4

O

O

20

8

CH3COO H3COOC

13 14

1

8 4

OCH3

11

1 3

H

AZADIRACHTIN-7,11-DIMETHYLETHER

OCH3

O

OCH3 O

H3COOC

COOCH3 O

O

O

CH3COO

AZADIRACHTIN-7-METHYLETHER

O

7

4

O

H3COOC

20

8

7

4

13 14

1

8 3

OH

11

3

8

4 H

OH

CH3COO

O

AZADIRACHTIN-20-METHYLETHER

H3COOC

O

O H

O

AZADIRACHTIN-11,20-METHYLETHER

FIGURE 4.2 7-methylether derivatives of azadirachtin

and physiological partitions, as these molecules make their way to their target sites or receptors — an observation made by earlier workers as well. It is essential, therefore, to point out here that the highest level of biological activity was obtained when C-1 and C-3 positions were occupied by only OH functional groups (Hansen et al., 1994). Another interesting example of a limonoid from neem showing potential antifeedant activity is salannin (Yamasaki and Klocke, 1989; Koul et al. 1996). Fourteen derivatives of salannin (8) when bioassayed against Colorado potato beetle, Leptinotarsa decemlineata, larvae have revealed four target points which after modification change the activity pattern of salannin. These targets are (i) hydrogenation of the furan ring, (ii) replacement of the acetoxy group, (iii) modification of the tigloyl group, and (iv) saponification of the methyl ester. The hydrogenation of the furan ring to the tetrahydrofuran ring increases the antifeedant activity. The replacement of the acetoxy group at position 3 by a methoxy group increases

© 2005 by CRC Press LLC

Insect Antifeedants

47

the activity, and a similar increase occurs when the acetoxy group at position 3 is replaced by hydrogen. The modification of a tigloyl function, such as hydrogenation, increases the activity at least twofold. On the contrary, deesterfication of the tigloyl or the α-methyl butyrene groups result in a reduction of activity. Saponification of the methyl ester at C-11 increases the activity, for instance, salannic acid (9) is at least eightfold more active than 1,3-diol derivative (10). O

O

OH

O

O

O

OH

O

3

O

3

O

O

CH3COO

HO O

O

(8)

(9)

O O O O

OH

O

O

O

O

3 O

O

7

HO

OH O

(10)

(11)

However, other derivatives in which the methyl ester is chemically modified need to be prepared in order to ascertain the activity of the carbomethoxyl group (Yamasaki and Klocke, 1989). Another group that has shown potential as antifeedant limonoids is that of citrus limonoids. It appears that furan and epoxide groups have to play a major role in the activity of these compounds. A possible role of C-7 is implied by the modest activity of the 7-hydroxylated de-epoxy system (Bentley et al., 1988). For instance, highly reduced activity of deoxyepilimonol (11) against epilimonol (12) and Limonin (13) demonstrate the above conclusion. In certain cases the cyclohexenone A ring and the α-hydroxy enone group in the B ring appear to be important for antifeedant activity. Also, the absence of 14-15 epoxide may not drastically reduce antifeedant activity (Govindachari et al., 1995). Recently, 23 semisynthetic derivatives of citrus limonoids, with a focus on the changes in C-7 carbonyl and the furan ring, have been evaluated against Spodoptera frugiperda larvae. In particular, reduction at C-7 afforded the related alcohols, and from these their acetates, oximes, and methoximes were prepared. Hydrogenation of the furan ring was also performed on limonin and obacunone to establish the significance of the furan ring in the antifeedant activity against insects (Ruberto et al., 2002).

© 2005 by CRC Press LLC

48

Opender Koul

O

O O

O

O 14 O

O

O

C

D 14

15 O

O

7

O

A

A'

O

B

7

15 O

O O

OH

(12)

(13)

If we look at the overall system of limonoid compounds there are several impediments to developing a quantitative understanding of structure-activity relationships. Different investigators seldom utilize the same bioassay species, interspecific differences in the response of the test insect can easily mask any meaningful observations, and differences in larval stadium make comparisons invalid (Champagne et al., 1992). Although some specific examples from neem and citrus have been discussed above and despite the difficulties inherent in comparing data from such diverse array of studies, some qualitative trends could be generalized. Aside from the C-seco limonoids mentioned above, the most active compounds appear to be intact apo-euphol limonoids (14) with a 14,15-epoxide and either a 19/28 lactol bridge or a cyclohexenone (3-oxo-1-ene) A ring. Absence of the 14,15-epoxide results in reduced activity as with azadiradione (15) in comparison to cedrelone (16) or anthothecol (17). OH O O OH O

OAc

O

HO

(14)

(15)

O

O

R

14 15 O O

O

O

(16) R
H (17) R
OH

O

O O

OAc

OH

(18)

© 2005 by CRC Press LLC

Insect Antifeedants

49

Oxidation to a D-seco limonoid (18) appears to correlate with reduced activity. For example, azadirone (19) is almost inactive as a feeding deterrent, but the 16-keto intermediate azadiradione (20) is quite active. Further oxidation to 14-epoxy-azadiradione (21) results in considerable loss of activity, which is scarcely improved by formation of the lactone-D ring in gedunin (18).

O

O

16

O

OAc

O

O

OAc

(19)

(20)

O

O AcO

O

O O

O

O

O O

O

O

OAc

(21)

(22)

O

OAc O OH

O

O

O

O

O

O

O O

O

AcO

OH

AcO H

(23)

(24)

The A,D-secolimonoids like liminin (13), nomilin (22), and obacunone (23) are usually less active than D-seco and many apo-euphol compounds. Model compounds based on the C and D rings, the associated furan ring are slightly more active than limonin, suggesting that this region of the molecule is most critical for bioactivity (Bently et al., 1990). However, few A, B, or B-seco limonoids seem to be less active than above A,D-seco limonoids, but it

© 2005 by CRC Press LLC

50

Opender Koul

is difficult to ascertain functional comparisons due to substantial variability in bioassay procedures. At this point of time it is not possible to draw any specific conclusions due to the paucity of structure-activity data based on monophyletic biosynthetic pathways of limonoids and the process of ring modification represented by the O and S values, which are likely independent of each other. However, there is some relationship in structure activity to some cytotoxicity studies against mammals. Pettit et al. (1983) bioassayed 38 limonoids from Meliaceae and Rutaceae for inhibition of a murine P-388 lymphocytic leukemia cell line. The most active compounds were 14,15-epoxy D ring and a 19-28 lactol bridge compounds like sendanin (24). Compounds with the epoxide and a 3-oxo-1-ene A ring (anthothecol) were somewhat less active and the reduction of the olefin eliminated the activity. In a very recent study, structure-related insect antifeedant relationship of 56 limonoids of both natural ones from various plants and modified forms belonging to the order Rutales was attempted, considering substitution patterns, oxidation states, and hydrophobicity, as well as distant geometry derived through conformational analysis on molecular modeling. It was demonstrated that orientation of the furan and hydroxylation at specific carbon sites influence the antifeedant activity against Spodoptera litura (Suresh et al., 2002). These studies suggest that molecular modeling could be a significant helping tool for designing of compounds. For instance, relaxed bond distance between oxygen atom at C-3 and C-20 of the azadirachtin molecule was found comparable to that of 20-β-hydroxyecdysone, which has a significance in binding to ecdysone receptors; therefore, such a change of the distance between the active sites could have an impact on antifeedant activity as well. These studies also suggest that the most active limonoid could be either an intact apoeuphol compound or a C-seco compound with a hydroxylated furan (an –OH that may overlap with a C-20 –OH in azadirachtins) and a dihydroxy A-ring.

QUASSINOIDS Quassinoids, which are, like limonoids, degraded triterpenes, exhibit some structural relationship vis-a-vis the antifeedant activity. Discovery of bruceantin (25), a quassinoid from Brucea antidysenterica, as a potent antineoplastic compound has generated tremendous interest in quassinoid type of natural and synthetic biologically active compounds (Lidert et al., 1987). Apart from anticancer, antiviral, antiamoebic, antimalarial, and anti-inflamatory properties of such compounds, quassinoids possess anti-insect properties as well, particularly the feeding deterrent effects (Leskinen et al., 1984). The structure activity correlation pattern OH COOCH3

HO O O

OCO A

HO

O

O

(25)

for the feeding inhibition, as demonstrated against tobacco budworm, Heliothis virescens, can be summarized as follows:

© 2005 by CRC Press LLC

Insect Antifeedants •

51

The A-ring enerone function is essential for activity. Reduction of the electrophilic capacity of this michael acceptor results in lowering of activity. Thus the A-ring diosphenols are on the whole less active, as can be seen in compounds 26 versus 27 (Lidert et al., 1987). However, higher electrophilicity of diosphenol achieved by placement of the electron withdrawing trifluoromethyl sulfonyl substituent onto the 3-hydroxy group (28) did not result in increased activity. OH

OH COOCH3

HO

HO

O

O

HO

O

O

O

O

COOCH3

HO

OCO

O

O

(26)

(27) OH COOCH3

HO O O

OCO A

SO2F3CO

O

O

(28)

•

•

The C-ring oxomethylene bridge is very important (compound 29 is of poor activity). C-8 and C-13 linkages seem to be somewhat more advantageous than C-8 to C-11 (compound 30 versus 31 and 32). Ester side chains have in many cases great influence on activity (compound 33 more active than 34). On the whole hydrophilic side-chains seem to be deterimental (35 versus 36) while hydrophobic, unsaturated side chains improve activity (37, 38). OH

HO O

HO C

O

O

(29)

Compounds lacking side chains altogether can be fairly active (30). Their pattern of structure-activity relationships is similar to the ones reported by other workers, too (Klocke et al., 1985; Odjo et al., 1981).

© 2005 by CRC Press LLC

52

Opender Koul OH

OH

HO

HO

HO

HO

O OH

O

O OH

O

OH O

O

O

(30)

O

(31) OH HO HO

O OCO

O

O

O

(32) OH

OH

HO

HO

OAc

OAc

O

O OCO

O

O

HO

OAc

O

O

HO

O

(33)

(34) OH

OH HO

HO

OAc

OAc

O

OCO

O

O

O

OCO

O

HO

(35) OH OAc

HO

HO

O

O

OCO

O

(37)

© 2005 by CRC Press LLC

O

(36)

OH

HO

OH

O

O

HO

O

O

OAc

HO O

O

OCO

O

O

(38)

Insect Antifeedants

53

DITERPENES Various diterpene acids and clerodane types of diterpenes have been identified from various plant sources and shown to deter feeding in various insect species (Hosozawa et al., 1974; Rose et al., 1981; Miyase et al., 1981; Koul, 1982; Wagner et al., 1983; Schuh and Benjamin, 1984; Belles et al., 1985; Enriz et al., 1994; Giordano et al., 2000). In fact, clerodanes occur in different isomeric forms, and the general problem of a structure-activity relationship exists here too, due to variability in bioassay evaluations. Nevertheless, it is possible to draw some conclusions. For instance taking clerodin (39) as the parent compound, changes in different carbons vis-a-vis the activity can be discussed. There are few derivatives that bear a substituent at C-1 as in ajugareptansin (40) and ajugareptansone A (41), but these are weak antifeedants. According to Belles et al. (1985) this may be accounted for by a Skew boat confirmation of the A ring, caused by sterical hinderence between substituents at C-1 and C-9. In clerodendrin A (42) aceylation at C-2 position results in a complete loss of activity; however, in Ivaine I (43), aceylation has no effect on the activity. O

O 15

O

O

14

16

O

O

O

COO

1

3

5 4

AcO

6

AcO

(39)

O

7 CO O

O

OAc

8

10

HO

O

9

1

2

OAc

AcO

(40)

(41)

O

O

O

O

HO

OAc

HO 2

2 COO

COO OAc

O AcO

(42)

OAc

O AcO

OAc

(43)

In any case, stereochemistry at C-2 is important, which is evident from higher activity in 14,15-dihydroajugapitin (44) than Ivaine IV (45). The substitution or stereochemistry at C-3 does not seem to affect the activity in any case. The epoxy ring also does not appear to

© 2005 by CRC Press LLC

54

Opender Koul

hinder the activity due to the similarity in activity among ajugarin II and III (46, 47) and tafricanins (48, 49). On the other hand, stereochemistry at C-4 seems to play an important role in the activity of these compounds. Several decalins with an epoxide of the opposite stereochemistry are less active (Ley et al., 1982; Geuskens et al., 1983) than otherwise identical decalins of the normal stereochemistry. O

O O

O

O

O

HO

HO 2

2

COO

COO O AcO

O

OAc

AcO

(44)

(45)

OAc

O

OH AcO

(46)

O

O

O O

(48) R1 = O, R2 = H, R3 = Cl, R4+R5 = O (49) R1 = O, R2 = R4 = H, R3 = Cl, R5 = OAc

R1 R2O

HO HO

AcO

OAc

R4 R3

R5

AcO

(47)

According to vanBeek and deGroot (1986), an important feature of the clerodane diterpenes might well be the α-CH2OAc side chain at C-5. All derivatives, including those with an intact epoxy group, lacking the acetyl substitution at C-19, are only very weakly active or inactive, like ajugarin V and clerodendrin A tetraol (50, 51). A loss of acetyl group at C6 does not affect the activity, as ajugarin I (52) and ajugarin II (46) do not differ in their antifeedant action. The necessity of some oxygen-containing substituent at C-9 is immediately obvious from the low level of activity of several synthesized decalins relative to clerodin and other natural clerodanes. Many different furofuran side chains in the ajugarin series and a furan side chain in the tefricanins all give rise to deterrent activity. Geuskens et al. (1983) suggest that activity of clerodanes is neither located in the side chain nor the decalin moiety alone but rather two groups appear to exert a synergistic effect on each other with regard to antifeedant activity.

© 2005 by CRC Press LLC

Insect Antifeedants O

55 O

O O

O O

HO

HO

O HO

O

OAc

HO

(50)

O

OH

AcO

(51)

OAc

(52)

Considering the feeding deterrent activity exhibited towards Tenebrio molitor of several diterpenes with clerodane skeleton using electronic and conformational behaviors (Enriz et al., 1994), they seem to mediate at least through two binding sites. The presence of an α,βunsaturated system, or one spiroepoxide substituent at C-4 in the clerodane structure, together with the β-furyl moiety at C-9 is important to evoke antifeedant activity. In addition the free rotation of the β-furyl group could play a significant role in the biological activity. These results will be apparently helpful in the structural identification and understanding of the minimal structural requirements for these molecules and can provide a guide in the design of compounds with antifeedant activity (Enriz et al., 1994). Enriz and his co-workers (2000) have further emphasized that steroelectronic factors are more important than the hydrophobic aspects. A conformational study indicates that the optimum interatomic distance between furan ring in the side chain and a spiro-epoxide range between 9.5 and 10.5 Å. They also found similar steroelectronic response among withanolides and azadirachtin, which for the first time gives an indication of a relative chemical mechanism for these compounds (Enriz et al., 2000). Certain structural transformations in this respect by synthesizing such derivatives have proved the importance of furan moiety and cleavage of the oxirane ring, which causes almost total disappearance of activity (Gallardo et al., 1996). To be precise the presence of β-axial spiro epoxy at C-4 together with β-ethylfuran ring, an ethyl butenolide or a hexahydrofurofuran substituent at C-9 is necessary to elicit the antifeedant activity of neoclerodanes against insects (Camps and Coll, 1993; Rodriguez et al., 1993; Malakov et al., 1994; Urones et al., 1995). Recently a new class of insect antifeedants, the ryanodine diterpenes (53), have been isolated from Persea indica (a Lauraceae plant). The structure-activity relationship of these compounds show that C-14 and C-1 substituents play an important role. Aceylation of these centers results in loss of activity, whereas pyrrolecarboxylate at C-14 (54) confers high OH

1 HO

OH O

14

R=

RO

C N

OH HO

O

(53)

© 2005 by CRC Press LLC

(54)

56

Opender Koul

potency (Gonzalez-Coloma et al., 1996). The comparisons of the mammalian toxicity and insect feeding deterrency of these compounds suggest a mechanism of action of these diterpenes in insects different from the Ca2+ release channel (Gonzalez-Coloma et al., 1996).

SESQUITERPENES Caryophyllene oxide, spathalenol, guaianol, helenalin, eupatoriopicrin, bakkenolide A, bisabolangelone, and various sesquiterpene lactones are active antifeedants against a variety of insect species (see Chapter 7). In fact antifeedant activity of a number of sesquiterpene lactones has been comprehensively reviewed (Picman, 1986), but as usual due to efficacy variabilities it has been difficult to generalize structural features responsible for this activity. Two conclusions that could be emphasized here are the importance of α-methylene group evidenced by diethylamine and methanol adducts of eupatolide being more deterrent than eupatolide (55). Michael type addition with the α-methylene on γ-lactone as well as with α,β-unsaturated ketone or with other exomethylenes could explain the activity of sesquiterpene lactones, which do not contain the α-methylene-γ-lactone moeity and, therefore, are worth consideration. Similarly antifeedant activity of 53 sesquiterpenes of Lactarius origin is known against stored grain pests (Daniewski et al., 1995). The sesquiterpenes with lactarane (56) and marasmane (57) skeletons are much more active than those with an isolactarane skeleton (58). The activity of furans is generally higher than their lactonic counterparts. The activity of furans depends upon the presence of hydroxyl groups in their molecules. The greater the number of OH groups the lower the activity. However, no simple correlation is possible between the antifeedant activity of lactones and the number of –OH groups in their molecules. A change in the position of a carbonyl group from C-5 to C-13 in the lactone ring does not improve the activity. Alteration in the natural characteristic configuration at C-8 causes a decrease in antifeeedant activity in both lactones and furans of lactarane skeleton (Daniewski et al., 1995). R1

OH

O

CH2

O

O

O

CH2R4 R2

O

(55)

O

(56)

(58)

© 2005 by CRC Press LLC

OH

O

OHC

R

(57)

OH

OH

O

R1

R3

HOH2C

OH

(59)

(60)

R2

Insect Antifeedants

57 CHO

CHO

OH

CHO

CHO

OH CHO

CHO

CH2

(61)

(62)

(63)

It could be further demonstrated that a mixture of hemiacetal and γ-hydroxyaldehyde form of isolactarane sesquiterpenes (59) are very active compounds compared with those that exist only in hemiacetal form (60) (Daniewski et al., 1997). Sesquiterpene drimane antifeedants like warburganal (61), polygodial (62), and muzigadial (63) are also known active compounds (Lam and Frazier, 1987) with a reactive enedial functionality that interacts with a chemoreceptor site via pyrrole formation. A series of natural drimanes and related synthetic compounds, when tested against aphids, show least activity due to polygodial or those compounds that do not possess the trans-decalin ring of the natural drimane aldehydes. Despite the high activity of (-) warburganal, of the other 9 α-hydroxy compounds only compound (64), the 9 α-hydroxy analogue of cinnamolide (65), is active. In fact, these compounds, which have been reported to be active against several species of Spodoptera and Heliothis (Blaney et al., 1987), are inactive against aphids. O

9 8

CHO OH

O O

CHO

(64)

(65)

(66)

At the molecular level most active dialdehydes have a double bond in common between C-7 and C-8 and an 11-12 β dialdehyde (Gols et al., 1996). The importance of such a configuration is supported by the lack of activity of compound (66) and (±) isotadeonal (67). The deterrence of lactones is higher when the lactone group is present at the C-8/C-9 position. Forty-one sesquiterpenes with a dihydro-β-agarofuran skeleton (68) and 14 related synthetic compounds have been evaluated against Spodoptera littoralis larvae (Gonzalez et al., 1997). These studies show activity in 38 compounds, the most active being those with isoalatol (69) and 4 β-hydroxyalatol (70) skeletons. Comparing the activities of the compounds with the same skeleton, the activity seems to increase with the number of acetate esters and decrease with the number of benzoate esters. In general products with aromatic esters at C-1 and C-9, whatever the stereochemistry, are moderately active. The introduction of ester groups at C-2 does not seem to change the activity of the compounds. CHO 11

9

1 8

12 CHO

2 3

8 10

A 4

7

5 6 O

14

(67)

© 2005 by CRC Press LLC

(69)

4β - OH, 9α - OBz, 1α,6β,15 - OAc, 8α OCH3

(70)

4β - OH, 9α - OBz, 2α,6β,8α,15 − OAc, 1α − OAng

B 7 C 12

11 13

(68)

58

Opender Koul

Silphenene sesquiterpenes are established chrysomelid antifeedants and have been evaluated against S. littoralis, L. decemlineata, Myzus persicae, Rhopalosiphum padi, Metopolophium dirhodum, Diuraphis noxia, and Sitobion avenae (Gonzalez-Coloma et al., 2002). Small structural changes (see Chapter 7) resulted in drastic differences in antifeedant activity, suggesting a high molecular selectivity for silphinene derivatives on chemoreceptors. The changes from angelate to acetate or tiglate or isobutyrate as C-5 substitution induces a great impact on the antifeedant potency. Esterification at C-5 with different substituents has strong effect on the activity (e.g., 5α-acetoxysilfinen-3-one; see Chapter 7) (Reina et al., 2002). Significant to moderate increase depending on the type of C-5 substituent (258-fold for ang., 187-fold for tig., 4-fold for isobut., and 3-fold for Ac) has been demonstrated (Reina et al., 2002). Importance of C-11 acetate has also been demonstrated and apparently the tricyclic silphinene sesquiterpenes are good antifeedant candidates for future study.

MONOTERPENES Many monoterpenes have been evaluated against insects to show feeding deterrence against them (Koul, 1982). However, capillin (71), capillarin (72), methyl eugenol (73), and arcurcumene (74) isolated from Artemisia capillaris have a promise as antifeedant compounds against cabbage butterfly larvae, Pieris rapae crucivora. The relative strong antifeedant activity of capillin and capillarin suggest that C=O carbonyl group instead of CH2 methylene group, a C ≡ C in a side chain, and a lactone ring are some of the many factors that contribute to the biological activity (Yano, 1987). Various derivatives of these base compounds like methyl eugenol reveal that the 3,4-dimethyl group and 1-substituent of 3,4-dimethoxy-1substituted benzenes (75) contribute to the antifeedant activity (Yano and Kamimura, 1993). Similarly capillin structure has an aromatic carbonyl group and two C ≡ C bonds. In order to demonstrate the importance of these two functions for the candidate activity, various derivatives evaluated against P. rapae crucivora reveal that arylmethyl ketone with a CH3 group (76), instead of an H atom combined with C = O group of aromatic aldehyde, is more active than that of aromatic aldehyde (Yano and Tanaka, 1995). Also a relationship between antifeedant activity using phenyl alkynes suggests that C ≡ C triple bond in the side chain is associated with antifeedant activity. It has also been observed that terminal groups (R) of side chain of C6H5-C≡C-R influences activity considerably, and the intensity of activity of various compounds shows a reasonable trend (77). This suggests that charge separation of C ≡ C triple bond by electron donative effect of alkyl group combination with C ≡ C bond may be correlated with an increase in antifeedant activity, and that a carbon chain enlargement of the alkyl group results in a decrease of antifeedant activity, probably because of the stereochemical hindrance (Yano, 1986). OCH3 OCH3

O O

O

(71)

© 2005 by CRC Press LLC

CH2

(72)

(73)

Insect Antifeedants

59

OCH3 OCH3

C=O R

R

(74)

R = CH3

(75)

(76)

>

>

>

> (77)

It has also been established that the 3,4-dimethoxy group and the 1-substituent of 3,4-dimethoxy-1-substituted benzenes related to methyl eugenol contribute to the activity. As for aromatic carbonyl compounds related to capillin, arylmethylketones [Ar – C(CH3)=O] became more active than aromatic aldehydes [Ar–C(H)=O], when an H atom of aldehyde group of aromatic aldehydes was replaced with a CH3 group (Yano, 1998).

COUMARINS Inhibitory activity of feeding by coumarins isolated from Atlanta recemosa and other related species against Spodoptera litura larvae have made it possible to draw some structural patterns for the said activity in this class of allelochemicals. Xanthotoxin (78) is known from decades to induce inhibitory effects in insects and accordingly has been shown to deter feeding as well (and so has its derivatives). Amongst these, xanthotoxol ethyl ether (79) has shown the highest feeding inhibition. Demethylated products like xanthotoxol (80) and its acetate (81) totally lack activity. Methyl and ethyl ethers of rutaretin (82), which are 2-(α-hydroxyisopropyl) dihydrofurano analogues of xanthotoxin are also inactive. Similarly 2-isopropylexanthotoxin (83) and its ethyl analogues (84) are also inactive (Luthria et al., 1989). OCH3 O

OC2H5 O

(78)

© 2005 by CRC Press LLC

O

O

OH O

(79)

O

O

O

(80)

O

60

Opender Koul OAc O

OR

OH O

O

O

R = CH3

O

O

or R = C2H5

(81)

(82)

OCH3 ipr

O

OC2H5 O

O ipr

O

O

O

(83)

O

O

O

(84)

(85)

OCH3 O O

O

O

OCH3

O

O

O

O

O

OCH3

(86)

(87)

(88)

OCH3 O

O

O

R1

O

O

OCH3

(89)

O

RO

O

R2

(90)

(91)

Psoralen (85), which has a linearly fused furan ring like xanthotoxin but lacks a methoxyl group, is moderately active. However, its isomer angelicin (86), with an angularly fused furan ring, shows several-fold reduction in activity. Isopimpinellin (87), which has an additional methoxyl group at C-4, is sixfold more active than the former compounds. An interesting observation is the least activity in luvangetin (88), racemosin (89) and xanthyletin (90), which are corresponding pyrano-analogues of xanthotoxin, isopimpinellin, and psoralen respectively. Substituted coumarins like umbelliferones (91) without furano and pyrano moieties are also inactive (Luthria et al., 1989). The conclusions that can be drawn from the comparisons are: • •

A linearly fused furan ring along with alkoxy groups at positions 4 and 9 play an important role in determining antifeedant activity. A substituent in the furan ring causes a loss of antifeedant activity.

© 2005 by CRC Press LLC

Insect Antifeedants

61

A series of 3-acyl-4-hydroxy coumarins structurally related to dicoumarol are also known to induce feeding inhibition in insects (Dreyer et al., 1987). It has been suggested that increase in 3-acyl group size in such compounds decreases the activity. These compounds are known antibacterial agents against gram positive bacteria but inactive against gram negative bacteria (Toda et al., 1958). Similar studies on furochromones have revealed some distinctive features responsible for antifeedant activity against insects (Luthria et al., 1993). The assumptions made on the basis of activity suggest that substitution at C-2 lowers the actviity (92 versus 93). Saturation of the 2,3-double bond as in 2,3-dihydronomellin (94) also diminishes the activity compared to that of khellin (95). Absence of cleavage of the furan ring results in drastic reduction in activity, which is evident from the reduced activity in chromones (96) and chromonones (97). Changes at the substituents at the pyrone ring (C-7 position) reduces activity. Thus compounds 98 and 99, which contain a CH2OH group at C-7 and are analogues of 92 and 95, are significantly active. The furanoflavones (100), the aryl analogues of 95, also do not show any significant antifeedant activity. O

Ac

O

O

2

O

> CH3O

CH2

(92)

O

(93) OCH3

OCH3

O

O

O

2

O