Background

The study of plant-insect interactions is necessarily multidisciplinary. Historically, however, the leaps forward have usually involved specific aspects of the interaction. In the early the 20th century, several biologists recognized the importance of plant secondary metabolites in host choice (e.g. Vershaffelt, Brues, Fraenkel) and in the 1950s and 1960s the mechanisms of host-plant choice by insects was a major focus of study (e.g. Kennedy, Schoonhoven, Dethier, Jermy, Ishikawa). The 1960s and 1970s saw a flowering of theories invoking the importance of chemical defenses in the co-evolution of plants and their associated insects (e.g. Ehrlich, Raven, Feeny, Rhoades and Cates). Other groups, meanwhile, were contributing to our understanding of broader community-level interactions (e.g. Janzen, Root, Strong, Zwolfer, Gilbert, Southwood, Price). Eventually, sophisticated studies of chemicals and their roles in interactions evolved, and the field of chemical ecology was born (e.g. Eisner, Berenbaum, Brattsten, Rosenthal, Rothschild).

More recently, a host of additional approaches have been added. Nutritional ecology (e.g. Scriber, Slansky, Simpson) phylogenetics (e.g. Mitter, Moran, Farrell, Pasteels), biogeography and genetics (e.g. Singer, Gould, Rausher, Via, Jaenike, Futuyma, Feder, Thompson, Mitchell-Olds) have all provided unique insights, and the impacts of other trophic levels in herbivory are now better appreciated (e.g. Gilbert, Lawton, Strong, Dicke, Stamp, Denno). Finally, it has become abundantly clear that plants are quite active participants in their interactions with herbivores (e.g. Coley, Bryant, Baldwin, Tumlinson, Tallamy, Karban).

In addition to these topics and many new researchers in them, important contributions now come from morphologists, neurobiologists, palaentologists and molecular biologists. My presentation will focus on some new and little-tested ways to think about and study the interactions of plants and insects. My attempt to explain the pertinence of these novel interactions will require that I draw together a number of the disciplines involved, though I cannot attempt to review the whole field.

Introduction

My principal focus concerns the insect rather than the plant and, in particular, on how insect neurobiology affects the plant-insect interaction at all levels. Behavior, including host plant-related behavior is an expression of neurobiology that is modified by many physiological factors. Limitations inherent to the nervous system constrain how much information may be processed and can influence choices made, attentiveness to diverse stimuli, and responsiveness to risk (e.g. of predation). Evolutionary adaptations of the sensory system and brain for accommodating these constraints may govern how host affiliations evolve. Changes in the nervous system as a result of experience may also affect fitness in different ways. Both genetically and environmentally based neurological traits may help explain patterns of herbivore-host associations and diet breadth.

Making a choice among plants

Insects searching for an acceptable host plant must first locate and identify the appropriate plant species. We know that the speed of host-finding may be important. There may be time limits for various reasons, while other ecological circumstances commonly impose a need for speed, such as when resources are rare or scattered and predators make searching risky. The accuracy with which host taxa are selected, and individual plant quality assessed, are also important, especially for insects with narrow host ranges and specific nutritional requirements for larval development. If the speed, accuracy and quality of choices are all to be maximized by very small animals in a very complex sensory world, strong selection for efficient neural processing might be expected as might the adoption of high-contrast signals (Bernays & Wcislo, 1994).

The majority of insect species use a very restricted number of hosts that typically share characteristic phytochemicals, some volatile and some nonvolatile. A subset of these compounds seems to be of great importance for identification of the host (see Bernays & Chapman, 1984; Städler, 1992; Schoonhoven et al., 1998), and in some extreme specialists great sensitivity to one or a few host-specific chemicals totally dominates in host selection (e.g. Ferguson et al., 1983; Pereyra & Bowers, 1988; Roessingh et al., 1997). Plant taxa heavily endowed with relatively unusual chemicals or suites of chemicals (non-apparent plant syndrome of Feeny, 1975) are often hosts for relatively large numbers of specialist insect species (e.g.Berenbaum, 1983). In addition, specialists tend to be deterred more than generalists by non-host secondary metabolites. I will make the case that specialists benefit from the strong contrasts between cues from hosts and non-hosts.

Additional mechanisms for acceptance/rejection used by different insect groups may heighten perceived contrast in various ways. An insect’s response to a chemical mixture may not be predictable based on its responses to each chemical separately. Specifically, interactions among chemical stimulants at the level of the chemoreceptors can result in major changes in concentration-response functions of particular stimulants, with deterrents reducing input from positive inputs and vice versa (e.g. Shields & Mitchell, 1995; Schoonhoven et al., 1998)). I will demonstrate how such interactions could potentially alter the total input from a mixture of conflicting inputs to either a clear positive or a clear negative signal. Such process may be important in producing the particular and synchronous firing of a suite of taste cells, that appears to occur in some beetles only when the requisite mixture of plant chemicals is present (Sperling & Mitchell, 1991). In addition, highly synergistic effects of multiple host compounds are seen in some cases (e.g. Städler & Buser, 1984; Spencer et al., 1999), Thus a variety of mechanisms can provide the clear signal needed for rapid decision-making in a highly complex chemical world. Data so far suggest that the predominant mechanisms vary among insect taxa. In any case, the mechanisms could influence the evolutionary lability of host associations, and the trajectory for a clade of insect herbivores evolving with respect to host affiliation.

Some herbivorous insects alter their preferences as a result of experience. In some cases this results from increased or decreased sensitivity of their chemoreceptors to certain metabolites (Renwick & Lopez, 1999). So far, such changes have been recorded in species that feed on plants in at least several genera and lead to increased acceptability of the experienced food, sometimes with a concomitantly decreased acceptability of other potential foods - once again, an increase in contrast between alternatives. Diet quality also alters relative acceptability of alternatives depending on nutrient need - a flexibility dependent on variation in the strength of inputs from different nutrient chemoreceptors (Simpson & Raubenheimer, 1993).

A minority of insect herbivore species are extreme individual generalists, apparently adapted to situations where food plant quality or abundance is variable or unpredictable, or to situations where the food plants may all be very rich in potentially noxious secondary metabolites. Such herbivores engage in food mixing, eating a variety of plants and frequently making choices about what to eat and what to ignore. Such food mixers often appear to be stimulated by novel chemicals, potentially reducing the inefficiency and complexity of decision making (e.g. Bernays et al., 1997).

Evidence for limited efficiency among generalists

Data will be presented from experiments with butterflies (Janz & Nylin, 1997), caterpillars (Bernays & Minkenberg, 1997), whiteflies (Bernays, 1999), aphids (Bernays & Funk, 1999) and grasshoppers (Bernays, 1998), indicating that having a choice of suitable foods reduces efficiency of foraging, and that the specialists have significant advantages. These benefits include the amount of time taken to reach the host plant, the times taken to make decisions to accept or reject potential food, the time taken to begin ingestion and the time spent in pauses during a meal. In addition, the degree of fidelity to the most suitable host in the presence of less suitable host species and genotypes and the ability to choose superior hosts in the presence of a choice of mixed-quality hosts are shown to be greater in specialists than relative generalists.

Fitness benefits of behavioral efficiency in host choice

Data will be presented indicating that efficient decision-making has positive fitness effects. This appears to be true for ovipositing insects not just with respect to limitation on time overall, but also for evasion of predation through rapid oviposition. Evidence will also be presented for costs associated with the poor quality decisions made by the relatively generalized Lepidoptera and Hemiptera due to reduced growth rate, reduced survivorship and reduced fecundity (Janz & Nylin, 199 7; Bernays & Minkenberg, 1997). Locomotor activity is known to be risky with respect to predator and parasitoid attack, and field studies demonstrate that predation may be 100x more likely during feeding than during resting (Bernays, 1997), thus illustrating fitness costs of with reduced feeding rates. In addition I argue that intermittent, hesitant, or picky feeding behavior and any kind of dithering is dangerous not only because it is conspicuous, but because an animal attentive to food-related activities is unlikely to be attentive to simultaneous environmental risks (Dukas, 1998).

Since protein is often at low concentration in leaves (especially older leaves) and the nitrogen requirements of insects tend to be relatively high, herbivores often compensate by eating large amounts. Not only is high quality food better for growth, but the risk of mortality via predation is reduced on nutritious hosts since less time must be spent feeding and vulnerable to predators. Indeed, perhaps the fitness advantage associated with predator avoidance exceeds that enjoyed due to increased growth rate. Safety and growth are important together of course at a larger time scale - feeding on high quality foliage may also reduce development time, reducing the lifetime risk.

Leaves present very diverse physical challenges, and highly diverse solutions have been found by insects through adaptations of mouthpart morphology. The frequency with which certain mandible types have evolved in separate insect lineages with similar types of food indicates the adaptive value of these structures (Bernays & Janzen, 1988; Bernays, 1991). Furthermore, evolution of mouthparts can be very rapid (Carroll & Boyd, 1992). In view of the ever-present risk of predation, structures that determine handling time may be under great selection pressure. Indeed, the preponderance herbivores that feed on young easily-handled leaves, is probably a matter of safety as much as nutrition.

Secondary chemistry of plants

Among the hundreds of thousands of phenols, alkaloids, terpenoids, iridoids, flavonoids, steroids, and other chemical compounds of plants have been the subject of considerable study (Rosenthal & Berenbaum, 1991). Many appear to have no effects at all on insect herbivores, while others stimulate feeding and/or growth. Some are sequestered, and of these many are clearly toxic in general and serve as plant defenses against many herbivores Such chemicals have often been considered toxic when close study demonstrated they were actually deterrent only, the effects on test insects due to starvation. Some herbivores, however, suffer deleterious postingestive effects and perform poorly in some way, or learn to avoid the plant as a result of feeding on it and move elsewhere, both of which can be bad outcomes. In many cases, it is not at all clear whether an ability to deal with toxins has been lost, as suggested in the case of grass-feeding grasshoppers (Bernays, 1990), or whether the plants have evolved specific defenses against particular insect herbivores (Berenbaum, 1983). It seems likely that both scenarios occur.

Sequestering these secondary metabolites for defense against predators is also common among herbivores (Bowers, 1990). Besides clear and well-documented fitness benefits of sequestration in highly aposematic species, there are many more subtle cases, in which insects deposit chemicals in the cuticle yet are not warningly colored (Bernays et al., 1991), and others where they gain protection from predators as a result of the gut contents alone (Sword, 1999). Although such processes are not specifically relevant to the theme of neuroecology, they do highlight once more the crucial importance of natural enemies in the lives of insects on plants.

In this presentation, I will also emphasize the very important roles of plant secondary metabolites in signaling. First of all the remarkable diversity chemicals found among plant taxa and within individual plants allows potentially clear signals for every specialist herbivores at taxonomic levels from plant species to plant family (Bernays, 1996). That butterflies such as checkerspots home in on plants from multiple families, simply because of the presence of iridoid glycosides in host plants that are otherwise extraordinarily diverse physically and chemically (Bowers, 1983), is startling evidence of the influence of simple chemical signals on insect behavior. Evidence is beginning to suggest that such cases are not unusual. In addition, deterrence of non-host compounds, being greater for specialists, increases sensory contrast between host and nonhost.

Diversity and the association of insect and plant clades

It has become clear from both fossil studies (Labandiera, 1998) and molecular phylogenies (Farrell & Mitter, 1993) that among all the herbivorous insect groups studied, great diversification is historically associated with the expansion and increased diversity of angiosperms. Although this could arise through coevolutionary processes associated with arms races, I argue that it could be at least partly the result of herbivore tracking diverse genotypes in a plant population and subsequent speciation of herbivores on established plant host races or species.

The selective pressures on the sensory system and its central nervous projections, would favor those insects that match the fine tuning of their detection of distinctive signals with particular plant chemotypes, and would thus be acting through the agency of ecological risk. So, as plants changed and diversified chemically, insect herbivores, being so dependent on specific cues, also changed and diversified so that discrimination of signals from hosts could be maintained at maximum levels of contrast. In this way, vigilance for predators could be maintained at maximum levels. Evidence for tracking chemicals in this way has been demonstrated in one study of a group of beetles and their host plants (Becerra, 1997), and a model is presented to further illustrate this scenario.

Although I will make the case for tracking of chemotypes, it is not impossible for diversification of insect herbivores, driven by such neural processes, to result from coevolutionary processes. However, rather than invoking toxins, the currency would be in terms of signal information.

Plants as active players

The widespread occurrence of herbivory-induced chemical changes in plants (Karban & Baldwin 1997; Agrawal et al., 1999) will be discussed not only as a direct defense but also as an indirect one, that plays on the vulnerability of these small herbivores to multiple risks from predators and parasitoids. Plants may encourage mortality of herbivores by causing them to decrease vigilance. This could involve increased searching and foraging activity, intermittent feeding, and restlessness that is induced by unusual or increasing levels of ingested secondary metabolites.

Conclusions

The synthesis presented here depends on knowing the insect - understanding it as an organism. The details of behavior and physiology, especially neurophysiology, have suggested a theoretical approach to the study of insect-plant interactions, namely the constraints on neural processing and the diverse effects of these constraints in ecology and evolution. I believe that this approach will allow us to understand more about all aspects of the insect-plant interaction in a way that has been difficult in recent decades. The importance of avoiding anthropomorphism and subjectivity in the study of animals may have mitigated against the study of them as individuals with behavioral and neural limitations that impact every aspect of their lives.

References Agrawal, A.A., Tuzun, S. & Bent,E. 1999. Induced Plant Defenses Against Pathogens And Herbivores. Biochemistry, Ecology, And Agriculture. American Phytopathological Society Press, St. Paul, MN.

Becerra, J.X. 1997. Insects on plants: macroevolutionary trends in host use. Science 276, 253-256.

Berenbaum, M. 1983. Coumarins and caterpillars: a case for coevolution. Evolution 37, 163-179.

Bernays, E.A. 1990. Plant secondary compounds deterrent but not toxic to the grass specialist Locusta migratoria: implications for the evolution of graminivory. Ent.Exp. Appl. 54, 53-56.

Bernays, E.A. 1991. Evolution of insect morphology in relation to plants. Phil. Trans.R.Soc.Lond.B 333, 257-264.

Bernays, E.A. 1996. Selective attention and host-plant specialization. Ent. Exp. Appl. 80, 125-131.

Bernays, E.A. 1997. Feeding by caterpillars is dangerous. Ecol. Ent. 22, 121-123.

Bernays, E.A. 1998. The value of being a resource specialist: behavioral support for a neural hypothesis. Amer. Nat. 151,451-464.

Bernays, E.A. 1999. When choice is a problem for a generalist herbivore: experiments with the whitefly Bemisia tabaci. Ecol. Ent. 24, 260-267.

Bernays, E.A. & Chapman, R.F. 1984. Host-Plant Selection by Phytophagous Insects. Chapman & Hall, N.Y.

Bernays, E.A. & Funk, D. 1999. Specialists make faster decisions than generalists: experiments with aphids. Proc. R. Soc. Lond. B 266, 1-6.

Bernays, E.A. & Janzen, D.1988. Saturniid and sphingid caterpillars: two ways to eat leaves. Ecology 69,1153-1160.

Bernays, E.A. & Minkenberg, O. 1997. Insect herbivores: different reasons for being a generalist. Ecology 78, 1157-1169.

Bernays, E.A. & Wcislo, W. 1994. Sensory capabilities, information processing and resource specialization. Q.Rev.Biol. 69, 187-204.

Bernays, E.A., Howard, J.J., Champagne, D. and Estesen, B.J. 1991. Rutin: a phagostimulant for the grasshopper Schistocerca americana. Ent. Exp. Appl. 60: 19-28.

Bernays, E.A., Augner, M. & Abbot, D.K. 1997. A behavioral mechanism for incorporating an unpalatable food in the diet of a generalist herbivore. J.Insect Behav. 10, 841-858.

Bowers, M.D. 1990. Recycling plant natural products for insect defense. In: D.L.Evans & J.O.Schmidt (eds) Insect Defenses, State University of N.Y. Press, pp 353-383.

Bowers, M.D. 1983. Iridoid glycosides and larval hostplant specificity in checkerspot butterflies. J.Chem.Ecol. 9, 475-493

Carroll, S.P. and Boyd, C. 1992. Host race radiation in the soapberry bug: natural history with the history. Evolution 46, 1052-1069.

Dukas, R. 1998 (ed) Cognitive Ecology. Chicago University Press. Farrell, B.D. & Mitter, C. 1993. Phylogenetic determinants of insect/plant community diversity. In: Species diversity in ecological communities: historical and geographical perspectives. R.E. Ricklefs and D Schluter (eds), Chicago University Press, pp 253-266.

Feeny, P. 1975. Biochemical coevolution between plants and their insect herbivores. In: Gilbert, L.E. and Raven, P.H. (eds) Coevolution of Animals and Plants, University of Texas Press, Austin.

Ferguson, J.E., Metcalf, E.R., Metcalf, R.L. and Rhodes, A.M. 1983. Influence of cucurbitacin content in cotyledons of Cucurbitaceae cultivars upon feeding behavior of Diabroticine beetles (Coleoptera: Chrysomelidae). J. Econ. Ent. 76, 47-51.

Janz, N. & Nylin,S. 1997. The role of female search behaviour in determining host plant range in plant feeding insects: a test of the information processing hypothesis. Proc.R.Soc.Lond. B 264, 701-707.

Karban, R. & Baldwin I.T. 1997. Induced Responses to Herbivory, Chicago Univ. Press. Labandiera, C.C. 1997. Insect mouthparts: ascertaining the paleobiology of insect feeding strategies. Ann.Rev.Ecol. Syst. 28, 153-193.

Pereyra, P.C. & Bowers, M.D.1988. Iridoid glycosides as oviposition stimulants for the buckeye butterfly, Junonia coenia. J. Chem. Ecol., 14, 917-928.

Renwick, J.A.A. and Lopez, K. 1999. Experienced-based food consumption by larvae of Pieris rapae: addiction to glucosinolates? Ent. Exp. Appl. 91, 51-58.

Roessingh, P., Städler, E., Baur, R., Hurter, J. & Ramp, T. 1997. Tarsal chemoreceptors and oviposition behaviour of the cabbage root fly sensitive to fractions and new compounds of host-leaf surface extracts. Physiol Ent. 22, 140-148.

Rosenthal, G.A. & Berenbaum, M.R. 1991 (eds). Herbivores: Their Interaction with Secondary Plant Metabolites, Academic Press, N.Y.

Schoonhoven, L.M., Jermy, T. & Van Loon, J.A.A. 1998. Insect-Plant Biology. Chapman & Hall, London.

Shields, V.D.C. & Mitchell, B.K. 1995. The effect of phagostimulant mixtures on deterrent receptor(s) in two crucifer-feeding lepidopterous species. Phil. Trans.R. Soc.Lond.B 347, 459-464.

Simpson, S.J. & Raubenheimer, D. 1993. The central role of the haemolymph in the regulation of feeding. Physiol.Ent. 18, 395-403.

Spencer, J., Pillai, S. & Bernays, E.A. 1999. synergism in the ovipositional behavior of Plutella xylostella: sinigrin and wax compounds. J. Insect Behav. 12, 483-500.

Sperling,J.H.L. & Mitchell, B.K.1991. A comparative study of host recognition and the sense of taste in Leptinotarsa. J. Exp. Biol. 157, 439-459.

Städler, E. 1992. Behavioral responses of insects to plant secondary compounds. In: Rosenthal, G.A. & Berenbaum, M.R. 1991 (eds). Herbivores: Their Interaction with Secondary Plant Metabolites, Academic Press, N.Y pp 45-88.

Städler, E. & Buser, H.-R. 1984. Defense chemicals in leaf surface wax synergistically stimulate oviposition by a phytophagous insect. Experientia 40, 1157-1159.

Sword, G.A. 1999. Density dependent warning coloration. Nature 397, 217.

---