Shared Signals and the Potential for Phylogenetic Espionage Between Plants and Animals1

Abstract

Until recently, the study and understanding of plant and animal signalling and response mechanisms have developed independently. Recent biochemical and molecular work is producing a growing list of elements involved in responses to biotic and abiotic stimuli that are very similar across kingdoms. Some of the more interesting examples of these include prostaglandin/octadecanoid-mediated responses to wounding, steroid-based signalling systems, and pathogen-recognition mechanisms. Some of these similarities probably represent evolutionary convergence; others may be ancestral to plants and animals. Ecological and evolutionary implications of such overlaps include the existence of pathogens that can cause disease in plants and animals, the ability of herbivores to manipulate plant responses, usurpation of microbial mechanisms and genes by herbivorous animals and plants, evolution of plant defenses exploiting shared signals in animals, and the medicinal use of plants by humans. Comparative study of the signalling and response mechanisms used by plants, animals, and microbes provides novel and useful insights to the ecology and evolution of interactions across kingdoms.

INTRODUCTION

Most of us usually think of plants more as objects than as organisms. But because adult plants are immobile, they have evolved the ability to detect and respond rapidly to environmental variation by altering phenotypic traits.

While plant response to pathogens has been recognized for decades, we now know that plants can sense and respond dynamically to a much greater range of stimuli, including organic and inorganic chemicals in air, soil, and water (Tuomainen et al., 1996; Miller et al., 1999; Shulaev et al., 1997; Grichko et al., 2000; Reymond et al., 2000), touch, motion and vibration (Cipollini, 1997; Schultz, unpublished), plant and animal pathogens (Holt et al., 2000; Rhame et al., 1997, 2000), insects (Karban and Baldwin, 1997), and light quantity and quality (Martinez-Garcia et al., 2000), among other stimuli. Virtually all metabolic pathways in plants are responsive, sometimes culminating in visible changes in the plant (e.g., pigmentation, growth form) but most are invisible without the use of instrumentation. Plant “behavior” is largely biochemical.

Many plant changes resemble those seen in animals, ranging from immune responses to more overt “behaviors.” While induced disease and parasite resistance, wound healing, touch responsiveness, and other phenomena are commonplace in animal biology, we are just beginning to appreciate them in plants. It now appears that “phenotypic plasticity” (Pigliucci, 1996) may be more important to plant success than are constitutive traits. Phenotypic plasticity not only provides flexibility in dealing with changing circumstances, but offers an opportunity to do so at reduced “cost,” since potentially energy- or material-expensive functions are undertaken only when needed (Karban and Baldwin, 1997).

Some plant-animal similarities probably involve homology (shared ancestral traits) while others represent adaptive convergence on solutions for similar situations or threats. Some organisms have enlisted symbionts to provide traits or functions found in other kingdoms. There may even be some examples of lateral gene transfer that result in genes or traits shared across kingdoms.

Since life on earth depends on plant production, selection has profoundly shaped how microbes and animals exploit plants and how plants resist. This has in turn structured most ecological systems, the evolution of plants, herbivores, and microbes, and the development of much of the earth's biodiversity (Ehrlich and Raven, 1964). While plant responses to pathogens have been known for some time, we have usually thought of other plant-exploiter and plant-environment interactions solely in terms of static traits. Understanding how plants respond dynamically to their environments, enemies, and benefactors provides many new insights into how these interactions function, develop, and evolve.

The main point of this brief review is that interactions between plants and other organisms may be influenced to a large extent by the degree to which they share exploitation and resistance mechanisms. Two organisms that use similar mechanisms or pathways used for a particular function already possess information about each other, since each must organize and regulate those same mechanisms and pathways. I suggest that this represents a kind of “phylogenetic espionage,” in which convergent or ancestral “information” is exploited by one or both participants in their biological or ecological interactions. Below I provide a few examples and discuss the ecological and evolutionary implications of shared mechanisms in plant-animal interactions.

SHARED RESPONSES

Receptors. Like animal behaviors, plant responses require a sensing mechanism and differential expression of genes encoding signalling and metabolic mechanisms for dealing with the perceived stimulus (Cosgrove et al., 2000). The focus of research on plant plasticity has turned to these molecular mechanisms in recent years because a) dynamic responses are now thought to be more important to plant success than are constitutive traits, b) gene expression is often more easily measured than is production of metabolites, and c) response traits must have a heritable, genetic basis if they are to be adaptive and be useful in improving plants. With plant pathology leading the way and the intense focus on sequencing the genomes of a few model plants, we are acquiring more of the details of plant responsiveness daily. The promise of this approach can be seen in early applications of biotechnology to plant improvement (Dixon et al., 1996).

Sensing mechanisms are perhaps the least understood step in plant responses to the environment. A few may resemble those known in animals (Bergey et al., 1996; Groenewald and van der Westhuizen, 1997; Chiu et al., 1999) but this is mostly inference based on the observation that plants can respond to certain stimuli or produce signalling molecules for which there are animal receptors. For example, the production of peptides involved in coordinating plant responses to wounding or herbivores (Bergey et al., 1996) and prostaglandin-like molecules implicated in mediating wound responses (Groenewald and van der Westhuizen, 1997; Imbusch and Mueller, 2000a, b) suggests that some sort of receptor must exist in plants. These have not, however, been identified.

Our rapidly expanding ability to search genomic databases for homologues/analogues may reveal plant receptors for such signals in the near future, causing us to rethink the organization and function of plant metabolism. For example, glutamate receptors, which gate cation-selective ion channels, are present in the surface membranes of excitable cells in insects, where they represent a target for insecticide action and toxins from insect and spider venoms (Elderfrawi et al., 1993; Usherwood, 1994). Plants produce glutamate, as well as many polyamine glutamate receptor antagonists (Usherwood, 1994; Saier, 2000), presumably as defenses against insects or other enemies (Graser and Hartmann, 2000). But recent genomic searches have located sequences encoding plant proteins identified as putative glutamate receptors (Lam et al., 1998). These receptor sequences may comprise an ancestral signalling mechanism possessed by both plants and animals (Chiu et al., 1999).

Catecholamines (CAs) are neurotransmitters in mammals, but are found in 44 plant families, where no essential metabolic function has been established for them. They are precursors of benzoic phenanthridine alkaloids, active ingredients of many medicinal plant extracts. But recent work implicates CAs in plant responses to insect herbivores and injuries, in nitrogen detoxification, plant tissue growth regulation, somatic embryogenesis from in vitro cultures, and flowering (Kuklin and Conger, 1995). CAs inhibit indole-3-acetic acid (auxin) oxidation, and enhance ethylene biosynthesis and various effects of gibberellins (Kuklin and Conger, 1995). Apparently, CAs do have primary functions in plants and plants must have a CA receptor, but none has yet been located.

The classification of entire groups of plant chemicals as “secondary” metabolites on the basis of little or no known function in the plant (Rosenthal and Berenbaum, 1991) may need revisiting as more primary functions, and the necessary receptors, are found for them.

Plants and animals were recently shown to use similar light receptor proteins (Cashmore et al., 1999). Plant phytochromes are differentially responsive to light of several wavelengths, which indicates the presence of nearby competitors (Ballare et al., 1997). Interestingly, the same protein receptors also function as clocks to set cyclic behaviors in both plants and animals, and appear to be ancestral in both (Cashmore et al., 1999).

Plants perceive touch, and are presumed to possess stretch receptors similar to those in animals (Kaiser et al., 1994; Ramahaleo et al., 1996; Trewavas and Knight, 1994). Distorting plant cell walls and membranes triggers characteristic cell and systemic signalling cascades producing altered growth form and rates (Trewavas and Knight, 1994). Some of the biochemical events that occur in response to mechanical stress also occur in plant responses to insects and disease, producing overlapping physiological and ecological outcomes, such as pest resistance enhanced by wind motion (Cipollini, 1997). A specific stretch receptor protein in plants has yet to be identified.

These few examples suggest that fundamental sensing mechanisms are shared by plants and animals. But while signal-responsive gene sequences are being identified, the search for classical biochemical receptors (e.g., proteins) for common signals is still in its infancy.

Signals. Plant signalling cascades are a focus of intense research at present (Cosgrove et al., 2000). The list of demonstrated and putative signals organizing plant responses to various stimuli has expanded dramatically beyond the original growth hormones (e.g., auxins, cytokinins, gibberellins) to include sugars (Smeekings, 2000), glycoproteins (Zinkl et al., 1998), phenolics (Cooper-Driver and Bhattacharya, 1998), oxidized fatty acids (octadecanoids, Farmer, 1997; Creelman and Mullet, 1997), aldehydes and alcohols (Bate and Rothstein, 1998; Arimura et al., 2000), steroids (Clouse and Sasse, 1998), lectins (Gabius, 2000), peptides (Howe and Ryan, 1999), reactive oxygen species including H₂O₂ and nitric oxide (Grant and Loake, 2000), and electrical impulses (Herde et al., 1999). With such a dramatic expansion of known signalling mechanisms, it is no wonder that examples of signals common to plants and animals are being found. I will focus here on one example of a shared signalling system, the fatty acid oxidation products “oxylipins” (plants) and “eicosanoids” (animals).

Fatty acid oxidation products help coordinate wound responses in most plants so far examined (Karban and Baldwin, 1997). These “oxylipins” or “octadecanoids” (Farmer, 1997; Reymond et al., 2000) are produced via oxidation of linoleic or linolenic acid (Creelman and Mullet, 1997; Mueller, 1997). A major product of this pathway, jasmonic acid (JA; Fig. 1), is produced via a pathway in which linolenic or linoleic acids released from cellular glycerides by lipases are oxidized by one of several lipoxygenases (LOXs) in a stereospecific manner, producing characteristic mixes of hydroperoxides. Several hydroperoxide-metabolizing pathways follow, producing a suite of fatty acid oxidation products that function as signals in plant physiology. Allene oxide synthase (AOS) and AO cyclase produce a group of ketols and fatty acids leading to production of 12-oxo-phytodienoic acid (PDA) and eventually jasmonic acid and relatives. JA and its derivatives can modulate aspects of fruit ripening, production of viable pollen, senescence, root growth and tuberization, stomatal resistance, abscission, germination, tendril coiling, protein synthesis and catabolism, and plant resistance to insects and pathogens (Gardner, 1995; Creelman and Mullet, 1997; Karban and Baldwin, 1997) among other functions. Methylated JA (MeJA) is volatile, and evidence is accumulating that wound-generated MeJA elicits responses in nearby, unwounded plants (Farmer and Ryan, 1990; Arimura et al., 2000; Karban et al., 2000). Another of these pathways, in which hydroperoxide lyase is key, produces aldehydes thought to be antimicrobial or antiherbivore in function, as well as products that may be attractive to both herbivores and their enemies (Arimura et al., 2000; Turlings et al., 2000).

The structures and biosynthesis of JA, its precursors and derivatives, mirror those of animal eicosanoids, derivatives of arachidonic (instead of linolenic/linoleic) acid which are synthesized and released for immediate and local, as well as systemic, action (Fig. 1). Eicosanoids include prostaglandins (PGs), prostacyclins (PCs), and thromboxanes, all produced via cyclooxygenase (COX) enzymes, as well as the lipoxygenase products hydroperoxyeicosatetraenoic acids (HPETEs), hydroxyeicosatetraenoic acids (HETEs), and leukotrienes (LTs; Gardner, 1995) (Figs. 1 and 2).

Arachidonic acid can be released from cellular phospholipids by phospholipases, after which it may enter either the cyclooxygenase or lipoxygenase pathways. In the cyclooxygenase pathway, it eventually undergoes tissue-specific modifications to form particular prostaglandins by specific PG synthases. As in plants, AOS is involved in this animal pathway; end products include physiologically active aldehydes (Gardner, 1995; Howard and Stanley, 1999).

There are two isoforms of cyclooxygenase: COX-1 and COX-2. COX-1 is constitutively expressed in most cells while COX-2 is inducible. Both enzymes and their products play multiple roles in animal systems, including regulating wound responses, inflammation and immune responses, platelet aggregation, cardiovascular function, and tumor growth, among others (van Ryn et al., 2000).

In the lipoxygenase pathway arachidonic acid can be converted to a suite of HETEs by several LOXs. The 5-lipoxygenase leads to the synthesis of the leukotrienes (LTs), which are peptidolipid acids involved in allergic and other immune responses and disorders (Magnon and Vervloet, 1999).

Imbusch and Mueller (2000b) recently discovered a plant pathway synthesizing isoprostanes analogous to animal pathways in which COX is involved. Isoprostanes are markers of oxidative stress in animals, an observation now extended to plants (Imbusch and Mueller, 2000b). Prostaglandins also have been found in plants (Groenewald and van der Westhuizen, 1997, 1998; Imbusch and Mueller, 2000a, b), but it is uncertain whether they are enzymatically produced or are instead nonenzymatic oxidation products (Groenwald and van der Westhuizen, 1997).

It is clear that plants and animals possess parallel, analogous fatty acid signalling systems (Fig. 2). Since most plants do not contain arachidonic acid, this plant pathway differs from the animal pathway in beginning with alpha-linolenic acid. However, recent work indicates that potato can use arachidonic acid from pathogenic fungi as a LOX substrate to initiate defense responses (S. D. Pechous, personal communication). Lipid-based signalling is merely one example of what may be many signalling systems shared by or analogous in plants and animals, either by convergence or homology.

IMPLICATIONS OF SHARED SIGNALLING

Observations like those above, which suggest greater than expected similarity between plants and animals in terms of coordinated responses to their environments, imply a largely unappreciated basis for interactions between the two kingdoms. The fact that mechanisms underlying responsiveness to the biotic and abiotic environment can be shared by plants and animals (and microbes) means that each brings to its interactions with the other(s) a form of information that can be exploited.

If, for example, plants and animals mediate wound responses using similar fatty acid-based signals, each has the means to manipulate the other's signalling system. Evidence is beginning to appear to support this view. Plant responses to pathogens most often involve an independent signalling system in which the polyphenol, salicylate (SA), is key (Fidantsef et al., 1999; Feys and Parker, 2000). SA is a COX and LOX inhibitor, and its accumulation in plants responding to microbes suppresses the production of wound-related signals (Feys and Parker, 2000). Oxidative stress and the accumulation of reactive oxygen species are involved in triggering disease resistance mechanisms in both plants and animals (Grant and Loake, 2000; Sandermann, 2000; Urquiaga and Leighton, 2000). Eichenseer et al. (1999) have shown that some insects produce H₂O₂ in saliva (via a glucose oxidase system) which triggers plant disease responses. Because these responses to SA accumulate, the net result is suppression of JA-mediated anti-herbivore responses, and a less-defended plant. The same SA-JA relationship forms the basis of our use of aspirin (acetylsalicylate), which suppresses eicosanoid inflammatory responses.

Similarly, since plants regulate their own oxylipin signalling, they may have the capacity to do so in enemies. While this possibility has not yet received much attention, insect defense responses to bacteria and parasitoids are mediated by eicosanoids biosynthetically and structurally similar to plant octadecanoids (Gillespie et al., 1997; Howard and Stanley, 1999). Since plants can up- or down-regulate their own systems, they have the information they need to do so in insects. I would expect to find that some plants can inhibit insect “immune” responses, perhaps using phenolics (like SA). Certainly medicinal plant use by humans supports this possibility (Teh et al., 1990; Teng and Ko, 1998; Spencer, 1999).

If plants recognize their own lipid-based signals, they may be able to recognize those of insects and use them to initiate defense responses. Evidence for this is accumulating rapidly. Caterpillar regurgitant contains aminated eicosanoids (peptidolipids) that are perceived by plants and upregulate the plant's octadecanoid pathway, culminating in elevated defenses and the emission of volatiles that attract parasitoids and attack the herbivore (Arimura et al., 2000; Turlings et al., 2000). Hence plants use information about a shared signalling system in insects to activate anti-insect defenses. Similarly, we have obtained preliminary evidence that the major insect prostaglandin, E-2 (Howard and Stanley, 1999) can stimulate production of oak leaf tannins, much as herbivore attack does (Schultz, unpublished). Clearly, plants detect and respond to animal as well as plant fatty acid signals.

Why are shared signals found in insect regurgitant? Insect eicosanoids may be found in most cells, including salivary tissues, and may simply occur inadvertently in saliva. Aminated and other fatty acids, including many phospho- and lysophospholipids, occur in insect guts as surfactants necessary to create micelle structure that permits fatty acid absorption (DeVeau and Schultz, 1992). The structure of these molecules broadly resembles the structure of “regurgitant” signals that activate plant defenses (Fig. 3); perhaps plants respond to this suite of fatty acid surfactants or their metabolites as indicators of insect attack (Spiteller et al., 2000). Natural selection would seem likely to favor masking such signals to block plant perception and response (above), yielding “stealthy” herbivores.

Plant growth hormones also have been found in a wide range of insect species (White et al., 1975; Elzer, 1983; Hori, 1992). Cytokinins or related compounds occur in insect secretions or glands associated with oviposition and initiation of galls (plant tumors elicited by insects, Hovanitz, 1959; Elzer, 1983; Hori, 1992; Leitch, 1994). Galling sawflies (Hymenoptera; Tenthridinidae) inject contents of an accessory gland containing plant cytokinins, related molecules, and uric acid (which has cytokinin activity; Elzer, 1983) into the plant tissues with the egg, initiating a tumorlike swelling within which their larvae develop (Leitch, 1994). Hovanitz (1959) was able to complete gall development in willow leaves after removing the larva with repeated injections of accessory gland contents. Many examples and circumstances suggest that insects use or manipulate cytokinins to achieve their ends (reviewed by Elzer, 1983).

It may seem surprising that insects could produce a plant growth hormone. But the major ribosome-bound tRNA in Drosophila, tRNA7-Ser, is N-6-(delta-2-isopentenyl)adenosine (i-6A), a plant cytokinin (White et al., 1975), which apparently performs a critical tRNA-ribosome stabilization function (White et al., 1975), presumably in all insects. This, plus the fact that members of the relatively ancient orthopteran lineage use cytokinins as ligands to regulate activity of their growth hormones (Tsoupras et al., 1983a, b) suggests that insects may share a metabolic pathway with plants that could permit them to enhance regrowth, alter metabolism, or maintain juvenile traits (all cytokinin functions, Hare and van Staden, 1997; Hare et al., 1997).

Animals depend on plants for the essential structures on which cholesterol and steroidal signal synthesis is based. Hence plants are positioned to manipulate steroidal signalling in animals, and even produce phytoecdysteroids and other molecules that manipulate or provide animal hormone activity (Schmelz, 1999). It is now clear that plants also employ steroidal signals (brassinosteroids, BRs) in regulating growth, development, and defense (Clouse and Sasse, 1998). BRs can have agonistic activity with respect to insect ecdytsteroid receptors (Bach et al., 1991; but see Machakova et al., 1995). It seems likely that here is another shared signalling pathway plants and herbivores could use for mutual exploitation or defense. Further work is needed to determine whether insects might produce a molecule with BR activity.

Plant auxins would seem to have no functional or biosynthetic analogs in animals. Yet plant auxins are also found in insects, especially in phloem-feeding species (e.g., aphids), which inject materials into the plant phloem while feeding, manipulating plant metabolism and nutritional content (Hori, 1992). While no primary functions of auxins are known in insects, some insects may synthesize IAA, and auxin and cytokinin concentrations are frequently elevated in the tissues on which they feed (Hori, 1992). IAA oxidase inhibitors have been found in some galling insect species (this would increase auxin concentrations in the plant), as has unidentified auxin “synergist” activity (Hori, 1992). There are no data to suggest how auxin synthesis might be accomplished in insects. It is seductive to note that forms of indole acetic acid (IAA) are byproducts of serotonin metabolism in insects (Vieira and Aldegunde, 1993).

The ability to manipulate responses across kingdoms could be acquired from third parties; plants or insects could exploit microbes or microbial genes to acquire the means to manipulate signalling systems they themselves do not possess. For example, several microbes have the ability to synthesize plant auxins and/or cytokinins (Elzer, 1983). Herbivorous tephritid flies host an endosymbiotic bacterium until recently classified in the genus Erwinia, a galling plant pathogen capable of synthesizing auxins and cytokinins (Lloyd et al., 1986; Drew and Lloyd, 1991). Most or all aphids host symbiotic bacteria (Buchnera) which, among other things, synthesize large amounts of tryptophan, an auxin precursor (Moran and Wernegreen, 2000). Spiteller et al. (2000) provide evidence that the aminated fatty acids (e.g., volicitin) in caterpillar guts that cue plant defense responses are of bacterial, not insect, origin. And recent evidence that plant-feeding nematodes have acquired bacterial genes encoding cell wall-degrading enzymes (Popeijus et al., 2000) while others possess chorismate mutase, a shikimate pathway enzyme otherwise restricted to plants (Lambert et al., 1999) suggests that genetic “tools” could be acquired laterally (Bird and Koltai, 2000). Perhaps this is how root knot nematodes stimulate flavonoid synthesis, blocking auxin transport in roots to produce galls (Hutangara et al., 1999; Goverse et al., 2000).

THE INTEGRATIVE, COMPARATIVE APPROACH

Our perception of plants as a kingdom apart from our own has colored our entire approach to studying plant function and adaptation, especially in plant-herbivore and plant-microbe interactions. Plant traits are usually interpreted as either serving the plant's own (“primary”) needs or managing interactions with other organisms or the environment (“secondary” needs). The suggestion that a plant trait (or molecule) might meet both needs (Berenbaum, 1995) has usually been rejected. I suggest that an integrative and comparative view that encompasses plants and animals (and microbes) offers a much better understanding of each, and is especially useful in grasping how plants and others interact. Moreover, this approach must embrace multiple levels, from molecular to population, to be successful.

It is increasingly evident that plants and animals differ less than we thought in how they respond to their biotic and abiotic environments. They share elements of fatty acid, protein, steroidal, neurotransmitter, reactive oxygen species, nitric oxide, and even plant growth hormone signalling systems. When they do not share a particular system, they may acquire aspects of that system from other organisms, for example symbiotic bacteria. As menages a deux, trois, or plus, plants, animals, and microbes often have information about the regulation of each other's responses, since they use the same coordinated responses themselves. This observation carries several implications.

1. We can expect many stealthy herbivores, pathogens, and parasites that manipulate plant hosts, just as many manipulate animal hosts. Evidence of stealthy herbivores is now appearing, and gall-forming species (which manipulate plant development and defenses) are the ultimate example. Understanding plant signalling systems is needed to explain successful herbivory, disease, and galling.

2. Signal “jamming” should be a major basis of plant-enemy interactions. The evolution of adaptation and defense should revolve around avoiding detection and manipulating signalling, not just developing and adapting to new classes of defenses.

3. Both plants and animals should manipulate responses to pathogens in the other. Upregulating plant disease responses can benefit insects by downregulating wound responses. Plants appear to have the tools necessary to alter insect disease resistance, which could benefit the plant. Both plants and animals could use “biowarfare” by interfering with immune responses.

4. Plants and animals have apparently converged upon or retained ancestral forms of signalling in defense and exploitation responses. It would be interesting to examine phylogenetic patterns in signalling development to determine which of these is more common.

5. To the extent that plants and animals use similar signalling and defense responses, they may be susceptible to pathogens or parasites that can subvert those responses across kingdoms. Bacterial pathogens of both plants and animals employ similar infectious strategies (e.g., Type III secretion systems), selecting for convergent plant and animal defenses (Rossier et al., 1999; Rahme et al., 2000), and permitting cross-kingdom infection (e.g.,Pseudomonas aeruginosa; Rahme et al., 1997). We can learn a lot about animal diseases from studying plant responses, and vice versa.

6. We should not be surprised to find that plants and animals employ microbes to manipulate each other's signalling systems. It appears that both symbionts and lateral gene transfer offer such possibilities, and that the origin of some (many?) plant-herbivore interaction traits is neither insect nor plant.

7. The key to developing pest-resistant plants and understanding plant-pest interactions lies in understanding signalling and perception rather than constitutive traits. Besides the potential “savings” of not expressing resistance when it is not necessary, dynamic responses offer flexibility and diversity in the face of evolving, diverse pests (Karban and Baldwin, 1997). One ought to be able to manipulate plants (through breeding or biotechnology) to respond specifically to particular stimuli, and perhaps ignore others. Response mechanisms can be examined for enhancement (e.g., degree, rapidity, type of response) and novel responses might even be invented. Understanding how animal signalling systems are organized reveals many additional targets for plant resistance traits, and examining both reveals pesticide targets. Since human, plant and insect signalling systems appear to share many features, understanding these offers many more opportunities to discover and develop pharmaceuticals and other products. While examining the chemistry of medicinal plants (or insects) provides some clues as to potentially useful medicines, foreknowledge of signalling system targets suggests many potentially useful mechanisms and molecules. Moreover, many potentially useful chemicals are not produced until elicited by some environmental stimulus or enemy; understanding dynamic responses can expand the utility of plants and other organisms dramatically.

It is time to stop dividing plants from animals in thinking about how organisms respond to their environments. Comparative aproaches are widely known to be powerful, but cross-kingdom comparisons seem to elude many of us. Some response elements are so ancestral and fundamental that we learn little from comparison (e.g., cell signalling), and others are no doubt restricted to plants or animals. But between these two extremes there is a wealth of understanding to be gained from surprising convergences and commonalities. These can teach us much about adaptation as a long-term process and especially about coevolution between plants and their exploiters. But to reap these benefits, we need to focus not on botany or zoology, but on general phenomena of biology.

Fig. 1. Structures of some fatty-acid-based signals from plants and animals. PGF_2a = prostaglandin (from animals). PGE₂ is common in insects, including herbivores. JA = jasmonic acid; PDA = 12-oxo-phytodienoic acid

Fig. 2. Schematic of fatty acid signalling synthesis in plants and animals. Modified from Gardner (1995).

Fig. 3. A. A typical surfactant phospholipid from larval Lepidoptera midgut. B. Two putative plant defense elicitors from larval Lepidoptera midguts

1

From the symposium Plant/Animal Physiology presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 3–7 January 2001, at Chicago, Illinois.

2

E-mail: [e-mail]

Attendance at this symposium and preparation of this paper were supported by NSF grant DEB-9974067.