19 Phytochemical Modification of Taste: An Insect Model J. Alan A. Renwick CONTENTS 19.1 Introduction 19.2 Repellents, Antifeedants, and Toxins 19.3 Cabbage Butterfly, Pieris rapae 19.4 Diet-Dependent Sensitivity 19.5 Conclusions References 19.1 Introduction Plant life in its various forms is widely recognized as the most readily available, abundant source of new chemicals or leads for synthetic chemicals to meet the growing needs of the agricultural and pharmaceutical industries. The array of chemicals produced by plants is enormous, but the number of known compounds represents only a fraction of the total. Fur- thermore, the diversity of biological activities exhibited by different groups of compounds continues to offer countless opportunities for practical applications. While we have become more aware of this chemical treasure and we increasingly exploit the unique properties of individual phytochemicals, especially as pharmaceuticals, the adaptive significance of these natural chemicals in the evolution of plants tends to be either ignored or forgotten. Although some controversy exists over the relative importance of pathogens and herbivores in the evolution of secondary metabolite production by plants, it is clear that the selective pres- sure exerted by invading organisms has played a major role. 1 Most chemicals that we now value for their biological activity are, therefore, likely to be products of the evolution of chem- ical defenses against such attacks. The properties that cause a specific phytochemical to deter oviposition by an insect pest, to function as an antibiotic, or to exhibit anticancer activity, may be considered as side benefits of this evolutionary process of self-protection. Compounds that defend a plant against invertebrates or vertebrates may now be used to protect crop plants in an agricultural setting. Other compounds that are highly toxic to vertebrates have proved to have valuable pharmacological properties, and phytochemicals that combat pathogen attack are often active against human pathogens (Figure 19.1). In many cases, common biochemical or physiological pathways may be involved in the activity of a chemical in different organ- isms. Such similarities point to the possible value of using one organism as a model for study- ing the reaction of another organism, especially humans, to phytochemicals. © 1999 by CRC Press LLC 19.2 Repellents, Antifeedants, and Toxins Many studies designed to develop crop plants that are resistant to insects or to find new con- trol agents have utilized basic information about the mechanisms used by plants to protect themselves against herbivores. 2 Chemical defenses of a plant against insects can be roughly categorized as repellents, antifeedants, or toxins. Repellents generally prevent landing by adult insects or movement onto plant surfaces by larvae. Antifeedants inhibit or deter feed- ing by those insects that venture onto the plant, whereas toxins either kill or immobilize those insects that do feed. However, the production of these chemicals by a plant may be dramatically influenced by many environmental factors, both abiotic and biotic. 3 Nutrition, particularly, nitrogen, sulfur, and phosphorus levels, can affect the biosynthesis of com- pounds that are rich in any of these elements. 4,5 Allelochemical production may be induced by herbivory, pathogens, or mechanical damage, 6 and exposure to UV-B radiation or air pol- lutants may have profound effects on some biosynthetic pathways (Figure 19.2). When we examine the way that plant chemistry may influence the selection of hosts by phytophagous insects, we are dealing with basic principles of sensory evaluation and pal- atability of potential foods. As humans, our discriminatory eating habits depend on percep- tion of both olfactory and gustatory chemical signals associated with the food, and personal preferences or individual reactions depend on the way that this chemical information is pro- cessed by the brain. Similarly, insects are often guided by the volatiles from a potential food plant for their initial approach (or avoidance), and then by nonvolatile gustatory stimuli for acceptance or rejection of the plant. The sense of taste, therefore, plays a major role in deter- mining host ranges, or dietary discrimination, of insects as well as higher animals. 19.3 Cabbage Butterfly, Pierus rapae The cabbage butterfly, Pieris rapae, has been used as a model for studying the interplay of gustatory cues that affect the host selection behavior of a specialist insect. The female but- terfly has contact chemoreceptors on its tarsi that are responsible for perceiving both posi- tive and negative gustatory stimuli at the plant surface. 7,8 Before accepting a plant for FIGURE 19.1 Evolution and human benefits of secondary metabolite production in plants. © 1999 by CRC Press LLC oviposition, the female often exhibits rapid movement or tapping with the forelegs. This behavior has been called “drumming”, and presumably allows more receptor hairs to come in contact with the chemicals that stimulate oviposition. 9,10 These receptors, or sen- silla, are peg-like structures which are arranged in rows between spines on the foretarsi (Figure 19.3). A single pore at the tip of the sensillum is typical of contact chemoreceptors, and electrophysiological recordings have been used to confirm that receptor cells associ- ated with these sensory hairs respond to active plant constituents. 7,8 FIGURE 19.2 Dynamics of allelochemical production in the defense of plants against insects. (Adapted from Renwick, J.A.A., in Phytochemical Diversity and Redundancy in Ecological Interations, Plenum Press, New York, 1996.) FIGURE 19.3 (Left) SEM of ventral side of the distal segment of a female Pieris rapae foretarsus showing a cluster of contact chemosensory sensilla (arrow). (Right) Enlarged view of a sensillum showing the single pore at the tip, typical of contact chemoreceptors. © 1999 by CRC Press LLC Tasting by larvae of the cabbage butterfly also depends on the use of contact chemorecep- tors that perceive the active chemicals at the surface of a plant. These receptors are located on the galea within the mouth of the caterpillar. Two pairs of sensilla styloconica, medial and lateral, are generally involved in the process of food recognition and discrimination, resulting in either stimulation or inhibition of feeding. 11 As in the case of tarsal receptors, electrophys- iological recordings can be performed to show good correlations between the responses of sensory receptor cells and behavioral responses to specific compounds. 12 Several compounds responsible for acceptance of suitable host plants and for rejection of unsuitable plants by adults and larvae of P. rapae have now been identified. For example, wormseed mustard, Erysimum cheiranthoides, is protected from attack by several specific cardenolides. 13 However, the most effective oviposition deterrents are not the same as the most effective feeding deterrents. The best oviposition deterrents are strophanthidin glyco- sides, whereas the strongest feeding deterrents are glycosides of digitoxigenin. 14,15 When extracts of E. cheiranthoides are subjected to solvent partitioning to remove the deterrents, aqueous extracts actually become stimulatory to ovipositing females of P. rapae. 16 This has served to demonstrate the fact that plants may contain both stimulants and deterrents and that the balance of these positive and negative chemical messages determines whether a plant is accepted or rejected. 17 This balance is likely to be influenced by the physiological state of an individual insect, perception of the compounds, and processing of the informa- tion that reaches the central nervous system (Figure 19.4). The possibility of manipulating the balance of positive and negative sensory cues to cause an insect to reject plants that they would normally accept has been suggested as a means to reduce crop losses to insect pests. Oviposition or feeding deterrents could con- ceivably be applied to protect the plants or they might be introduced through plant breed- ing or genetic engineering to produce resistant plants. Alternatively, we might attempt to interfere with the ability of an insect to taste chemical constituents of plants that they encounter. If an insect loses its sensitivity to stimulants in a plant, it would no longer be capable of recognizing a good host on the basis of secondary plant chemistry. On the other hand, a loss of sensitivity to deterrents might result in feeding on unsuitable food plants that would have a deleterious effect. 19.4 Diet-Dependent Sensitivity Recent studies on larvae of P. rapae have resulted in the discovery of diet-dependent sensi- tivity to an antifeedant in a normally acceptable host plant. The butterflies readily lay eggs on garden nasturtium, Tropaeolum majus, and the hatching larvae feed and develop nor- mally. However, if larvae that have fed and developed on cabbage plants are transferred to nasturtium, they refuse to feed. The effect is so extreme that the larvae will starve to death rather than feed on the nasturtium. 18 Similar results were obtained when larvae were trans- ferred from a range of other host plants, including both crucifers and noncrucifers, to nas- turtium. This rejection behavior was explained by the presence of chlorogenic acid and some additional, unidentified constituents of T. majus. 19 Chlorogenic acid was strongly deterrent to cabbage-reared larvae, but only slightly deterrent to nasturtium-reared individuals. Fur- thermore, larvae that were reared on a wheat germ artificial diet were almost completely insensitive to chlorogenic acid. It appears, therefore, that P. rapae larvae are insensitive to deterrents at the time of hatching and will feed on a wide range of plants or artificial diet. However, as they feed on cabbage or other crucifers, they develop sensitivity that results in their refusal of food that contains deterrents. We have concluded that continuous exposure © 1999 by CRC Press LLC FIGURE 19.4 Factors affecting the balance of positive and negative chemical cues that influence acceptance or rejection of a potential host plant by an insect. (Adapted from Renwick, J.A.A. and Huang, X., in Functional Dynamics of Phytophagous Insects, Oxford & IBH Publishing, New Delhi, 1994.) © 1999 by CRC Press LLC to and/or consumption of food containing deterrents results in a type of habituation, or more precisely, suppression of sensitivity development. 19 Since habituation is usually defined as a “waning of response to a repeatedly presented stimulus over time,” the term is not precisely applicable to the lack of sensitivity develop- ment seen in this insect. However, experiments conducted to test the effects of dietary experience with one or more compounds on responses to unrelated compounds seem to indicate that a type of “cross-habituation” is possible. When larvae were reared on nastur- tium, they were less sensitive than cabbage-reared larvae to a range of feeding deterrents. 20 Furthermore, larvae reared on wheat germ diet were almost completely insensitive to the same compounds. When larvae fed and developed on cabbage leaves that had been treated with selected deterrents, including cardenolides and cucurbitacin glycosides, they remained less sensitive to the nasturtium deterrents. 20 The development of sensitivity to feeding deterrents can be followed at any larval stage. When larvae were transferred from nasturtium to cabbage, they became progressively more sensitive to the nasturtium deterrents. Although all instars show the same general effect, second instars of P. rapae appear to be most plastic in their development of sensitivity as they feed on cabbage. 21 The possible involvement of an “inducer” in cabbage has been discounted, since larvae that were reared on cabbage foliage treated with nasturtium extract remained insensitive to the deterrents from that plant. It is likely, therefore, that sen- sitivity develops naturally as larvae feed on host plants that lack deterrents or other com- pounds that suppress this development. This phenomenon of increased sensitivity within an instar after removal from plants containing suppressors has prompted the proposal to use P. rapae as a model organism for future analysis of the physiological and biochemical processes involved in chemosensory development. 22 The idea of feeding experience changing the characteristics of taste receptors is not com- pletely new. Already in 1969, Schoonhoven 23 found that the sensitivity of a so-called deter- rent receptor in Manduca sexta larvae to salicin was considerably lower for larvae reared on artificial diet that contained this compound than for larvae reared on control diet . Similar electrophysiological results have since been obtained for larvae of Pieris brassicae in response to chlorogenic acid, proline, and cyanin chloride, 24 and for larvae of Spodoptera species in response to azadirachtin, nicotine, or sinigrin. 25 The effect of wheat germ diet on the ability of insects to taste and respond to deterrents is particularly interesting. Artifical diets containing wheat germ have long been used by entomologists for rearing a wide range of insects for various purposes. However, many insects that are normally reared on wheat germ diets will refuse this artificial food if they have previously fed on a host plant. Extracts of an artificial diet containing wheat germ were found to be deterrent to Manduca sexta larvae, and wheat germ itself was thought to be the source of deterrent components, as omission of the wheat germ increased the accept- ability of the diet. 26 Similarly, larvae of the cabbage butterfly refuse to feed on wheat germ diet when transferred from a cabbage plant, and the presence of deterrents has been dem- onstrated to explain this behavior. Both hexane extracts and butanol-soluble material from aqueous extracts of the whole wheat germ diet were highly deterrent to cabbage-reared lar- vae of P. rapae. 27 Wheat germ diet for insect rearing generally consists of seven components in addition to wheat germ itself and water. These include agar, aureomycin, casein, methylparaben, salt mix, sorbic acid, and a vitamin mix. When tested in a feeding deterrent assay using fourth instars of P. rapae, only the sorbic acid was slightly deterrent. However, extracts of the wheat germ itself were highly deterrent. Since strong deterrent activity was found in hex- ane as well as butanol extracts, a separation scheme was developed to examine both active fractions (Figure 19.5). Seven compounds were isolated by HPLC separation of the butanol © 1999 by CRC Press LLC fraction of the aqueous material. The UV spectra suggested that most of these are apigenin- based flavones. Fractionation of the hexane-soluble material by sequential solvent parti- tioning revealed that most of the active compounds were methanol-soluble, and HPLC of the methanol fraction showed that some compounds were common to both the butanol and hexane fractions, but others were present only in one of the fractions. When neonate larvae were allowed to feed on cabbage leaves that were treated with individual fractions collected from the HPLC, they remained insensitive to deterrents. Thus, individual constit- uents of wheat germ could account for the “cross-habituation” that is necessary for larval acceptance of nasturtium. 27 The sensitivity suppressing activity of specific compounds in wheat germ would suggest that these phytochemicals are acting as taste modifiers. This could have considerable prac- tical significance, since modification of taste in humans is an effect that is often desired and is poorly understood. The value of using insects as models for the investigation of higher animal taste mechanisms has already been suggested, and the validity of comparing the two biological systems is supported by the fact that bitter taste to humans has been used as a guide in the search for insect feeding deterrents in plants. 28 Insects are stimulated to feed by sugars that taste sweet to humans, and the interaction of sweet and bitter tastes in humans may be compared with the interactions between stimulants and deterrents in insects (Figure 19.6). Selected compounds that are bitter tasting to humans have been found to deter both oviposition and feeding by the tobacco budworm, Heliothis virescens, and preliminary electrophysiology has suggested that responses of a sucrose-sensitive neu- ron in the gustatory sensilla of the ovipositor are inhibited by these compounds. 29 In addi- tion, both insects and vertebrates exhibit a rather general phagostimulatory response to low molecular weight amino acids such as glycine, β-alanine, α-aminobutyric acid, γ-amino- butyric acid (GABA), L-arginine, and L-proline. 30 19.5 Conclusions Modification of taste is of interest for several reasons. Specific inhibitors of bitter taste are continually sought for pharmaceutical or food science applications, whereas bitter taste might be considered desirable in other products, such as beer. Suppression of bitterness may FIGURE 19.5 Extraction and isolation scheme for feeding deterrents (sensitivity suppressors) in wheat germ. *Denotes deter- rent activity. © 1999 by CRC Press LLC be accomplished simply by the addition of salt, 31 or the perception of bitter compounds may be modified by the presence of citrate or malate that have a sour taste. 32 However, most studies in this area have focused on modification of sweet taste. Many plant-derived chem- icals have been shown to act as either sweetness inhibitors or sweetness inducers or enhanc- ers. As a result, much of the work on taste modification has capitalized on this knowledge to examine mechanisms that govern sweet taste, and considerable progress has been made in explaining how sweetness modifiers might work using an insect model system. 33 Although information about the biochemistry of taste receptors in insects as well as mammals is still rather fragmentary, significant advances in the molecular biology of taste transduction in mammals have recently been made. 34 Some key component G proteins involved in taste transduction in mammals have been cloned. Electrophysiological, bio- chemical, and molecular biological studies have suggested that bitter as well as sweet taste is transduced by second messenger-mediated pathways involving cyclic adenosine mono- phosphate (AMP). 35 However, G proteins have yet to be isolated and identified from insect taste receptors. As soon as taste cell proteins can be cloned from insects, the restriction imposed by the palate of human tasters will be removed to allow for the design and screen- ing of novel taste agents. 34 The development of taste sensitivity in mammals or invertebrates is a phenomenon that still remains a mystery. Observations with human newborns have indicated that responses to a bitter compound increase within a period of 14 to 180 days. 36 These results appear to parallel the gradual development of sensitivity to deterrents found in newly hatched larvae of P. rapae. These caterpillars, therefore, represent an ideal model for detecting taste recep- tor proteins as the sensory system develops. Such critical information about the processes involved in taste modification may then provide a starting point for the needed isolation and cloning of taste receptor proteins. The identities and exact role of phytochemicals in the modification of taste will be key components of this puzzle that now seems possible to solve in the foreseeable future. FIGURE 19.6 The role of taste in food selection and discrimination by insects and humans. © 1999 by CRC Press LLC References 1. Spencer, K.C., Chemical Mediation of Coevolution, Academic Press, San Diego, 609, 1988. 2. Hedin, P.A, Plant Resistance to Insects, American Chemical Society Symp. Series No. 208, 375, 1983. 3. Kogan, M. and Paxton, J., in Plant Resistance to Insects, Hedin, P.A., Ed., American Chemical Society Symp. Series No. 208, 153, 1983. 4. Hugentobler, U. and Renwick, J.A.A., Oecologia, 102, 95, 1995. 5. Waring, G.L. and Cobb, N.S., in Insect-Plant Interactions, vol. IV, Bernays, E.A., Ed., CRC Press, Boca Raton, FL, 167, 1992. 6. Baldwin, I., in Insect Plant Interactions, vol. V, Bernays, E.A., Ed., CRC Press, Boca Raton, FL, 1, 1994. 7. Du, Y.J., Loon, J.J.V., and Renwick, J.A.A., Physiol. Entomol., 20, 164, 1995. 8. Städler, E., Renwick, J.A.A., Radke, C.D., and Sachdev-Gupta, K., Physiol. Entomol., 20, 175, 1995. 9. Ilse, D., J. Bombay Nat. Hist. Soc., 53, 486, 1956. 10. Fox, R.M., J. Res. Lepid., 5, 1, 1966. 11. Schoonhoven, L. M., in Perspectives in Chemoreception and Behavior, Chapman, R.F., Bernays, E.A., and Stoffolano, J.G., Jr., Eds., Springer-Verlag, New York, 69, 1986. 12. van Loon, J.J.A., Entomol. Exp. Appl., 80, 7, 1996. 13. Sachdev-Gupta, K., Renwick, J.A.A., and Radke, C.D., J. Chem. Ecol., 16, 1059, 1990. 14. Sachdev-Gupta, K., Radke, C.D., Renwick, J.A.A., and Dimock, M.B., J. Chem. Ecol., 19, 1355, 1993. 15. Renwick, J.A.A., in Phytochemical Diversity and Redundancy in Ecological Interations, Romeo, J.T., Sunders, J.A., and Barbosa, P., Eds., Plenum Press, New York, 57, 1996. 16. Renwick, J.A.A., Radke, C.D., and Sachdev-Gupta, K., J. Chem. Ecol., 15, 2161, 1989. 17. Renwick, J.A.A. and Huang, X., in Functional Dynamics of Phytophagous Insects, Ananthakrish- nan, T.N., Ed., Oxford & IBH Publishing, New Delhi, 79, 1994. 18. Renwick, J.A.A. and Huang, X.P., J. Chem. Ecol., 21, 465, 1995. 19. Huang, X.P. and Renwick, J.A.A., J. Chem. Ecol., 21,1601, 1995. 20. Huang, X. and Renwick, J.A.A., Entomol. Exp. Appl., 76, 295, 1995. 21. Renwick, J.A.A. and Huang, X.P., Entomol. Exp. Appl., 80, 90, 1996. 22. Renwick, J.A.A. and Huang, X.P., in Phytochemicals and Health, Gustine, D.L. and Flores, H.E., Eds., American Society of Plant Physiologists, vol. 15, 271, 1995. 23. Schoonhoven, L.M., Koninkl. Nederl. Akademie van Wetenschappen, 72, 491, 1969. 24. van Loon, J.J.A., J. Comp. Physiol. A, 166, 889, 1990. 25. Simmonds, M.S.J., Simpson, S.J., and Blaney, W.M., J. Exp. Biol., 162, 73, 1992. 26. Städler, E. and Hanson, F.E., Physiol. Entomol., 3, 121, 1978. 27. Huang, X.P. and Renwick, J.A.A., J. Chem. Ecol., 23, 51, 1997. 28. Kubo, I., in Recent Advances in Phytochemistry, Downum, K.R., Romeo, J.T., and Stafford, H.A., Eds., Plenum Press, New York, vol. 27, 133, 1993. 29. Ramaswamy, S.B., Cohen, N.E., and Hanson, F.E., Entomol. Exp. Appl., 65, 81, 1992. 30. Mullin, C.A., Chyb, S., Eichenseer, H., Hollister, B., and Frazier, J.L., J. Insect Physiol., 40, 913, 1994. 31. Breslin, P.A.S. and Beauchamp, G.K., Chem. Senses, 20, 609, 1995. 32. King, N.L.R. and Bradbury, J.H., J. Sci. Food Agric., 68, 223, 1995. 33. Kennedy, L.M., Bourassa, D.M., and Rogers, M.E., in Sweet-Taste Chemoreception, Mathlouth, M., Kanters, J., and Birch, G., Eds., Elsevier Applied Science, London, 317, 1993. 34. Margolskee, R.F., Bioessays, 15, 6455, 1993. 35. Kolesnikov, S.S. and Margolskee, R.F., Nature, 376, 85, 1995. 36. Kajura, H., Cowart, B.J., and Beauchamp, G.K., Developmental Psychobiology, 25, 375, 1992. © 1999 by CRC Press LLC . to low molecular weight amino acids such as glycine, β-alanine, α-aminobutyric acid, γ-amino- butyric acid (GABA), L-arginine, and L-proline. 30 19. 5 Conclusions Modification of taste is of interest. 19 Phytochemical Modification of Taste: An Insect Model J. Alan A. Renwick CONTENTS 19. 1 Introduction 19. 2 Repellents, Antifeedants, and Toxins 19. 3 Cabbage Butterfly, Pieris rapae 19. 4 Diet-Dependent. these are apigenin- based flavones. Fractionation of the hexane-soluble material by sequential solvent parti- tioning revealed that most of the active compounds were methanol-soluble, and HPLC