Antiherbivore defenses alter natural selection on plant reproductive traits

While many studies demonstrate that herbivores alter selection on plant reproductive traits, little is known about whether antiherbivore defenses affect selection on these traits. We hypothesized that antiherbivore defenses could alter selection on reproductive traits by altering trait expression through allocation trade‐offs, or by altering interactions with mutualists and/or antagonists. To test our hypothesis, we used white clover, Trifolium repens, which has a Mendelian polymorphism for the production of hydrogen cyanide—a potent antiherbivore defense. We conducted a common garden experiment with 185 clonal families of T. repens that included cyanogenic and acyanogenic genotypes. We quantified resistance to herbivores, and selection on six floral traits and phenology via male and female fitness. Cyanogenesis reduced herbivory but did not alter the expression of reproductive traits through allocation trade‐offs. However, the presence of cyanogenic defenses altered natural selection on petal morphology and the number of flowers within inflorescences via female fitness. Herbivory influenced selection on flowers and phenology via female fitness independently of cyanogenesis. Our results demonstrate that both herbivory and antiherbivore defenses alter natural selection on plant reproductive traits. We discuss the significance of these results for understanding how antiherbivore defenses interact with herbivores and pollinators to shape floral evolution.     

Triterpene-based plant defenses

Pentacyclic triterpenes are abundant in the plant kingdom and have a wide array of pharmacological activities. They also have insect antifeedant effects and therefore apparently play a role in plant defense. In this paper, we describe the insecticidal activity of pentacyclic triterpenes of plant origin from different chemical classes on several insect pests (Spodoptera littoralis, Leptinotarsa decemlineata and Myzus persicae), their phytotoxic properties and their selective cytotoxic effects on insect-derived Sf9 and mammalian CHO cells. We also discuss the role they play in plant defense based on these activities.     

Surface lipids and plant defenses

The major function of the plant epidermis is to form the cuticle, a functional permeability barrier of the cell wall which prevents excessive water loss and the entry of harmful substances and pathogens into the host. This type of cell wall modification is mainly composed of a polyester matrix, cutin, and soluble waxes embedded in the matrix and deposited on the external surface. Cuticle-associated proteins may also be important. Recent observations are starting to reveal complex inter-relationships between cuticular lipids and immunity. This suggests that the cuticle is not simply a physical barrier, but a dynamic host defense with signaling circuits and effector molecules. Furthermore, these studies have also demonstrated that cuticular lipids and immunity may intersect in common pathways, although the significance of this is not fully understood. In this review, we examine the functions of the plant cuticle in host–pathogen interactions, and discuss the possibilities of integrating the membrane and cuticular glycerolipid biosynthesis.     

Analyzing plant defenses in nature

A broad range of chemical plant defenses against herbivores has been studied extensively under laboratory conditions. In many of these cases there is still little understanding of their relevance in nature. In natural systems, functional analyses of plant traits are often complicated by an extreme variability, which affects the interaction with higher trophic levels. Successful analyses require consideration of the numerous sources of variation that potentially affect the plant trait of interest. In our recent study on wild lima bean (Phaseolus lunatus L.) in South Mexico, we applied an integrative approach combining analyses for quantitative correlations of cyanogenic potential (HCNp; the maximum amount of cyanide that can be released from a given tissue) and herbivory in the field with subsequent feeding trials under controlled conditions. This approach allowed us to causally explain the consequences of quantitative variation of HCNp on herbivore-plant interactions in nature and highlights the importance of combining data obtained in natural systems with analyses under controlled conditions.Key words: natural systems, plant defensive traits, optimal defense hypothesis (ODH), cyanogenesis, lima bean, Phaseolus lunatus L., plant-herbivore interaction, plant-pathogen interaction, multiple defense syndromesAnalyzing plant defenses against herbivores in nature is often complicated by an extreme variability in multiple factors. Plant populations generally show high genetic variability resulting in substantial intraspecific variation of plant traits.¹ In addition to genotypic variability, phenotypic plasticity of plants is a source of variation.² At the level of individual plants, expression of defensive traits strongly depends on plant organ and ontogeny of plants or plant parts. Within an individual plant, it is quite common for reproductive structures and young leaves to be better chemically defended than older leaf tissues. To explain these within-plant variations of defenses, the optimal defense hypothesis (ODH) was formulated. Concerning the variability of chemical defenses of leaves, the ODH predicts that within the total foliage of a plant, young leaves make a larger contribution to plant fitness than old leaves as they have a higher potential photosynthetic value resulting from a longer expected life-time.^3–5 In addition, younger leaves are often more nutritious and thus more attractive to herbivores⁶ and should consequently be better defended.⁷ In this line, the basic assumption of the ODH is that three main factors—cost of defense, risk of attack and value of the respective plant organ—determine the investment in defensive secondary metabolites.^8,9 Thus, the higher the risk of a given plant tissue to be consumed by herbivores and the higher its value for plant fitness, the more energy should be allocated to its defense.^10,11 Beyond genotypic and ontogenetic variability of a given defense, potential co-variation with other defensive or nutritive traits expressed by the same plant individual can strongly affect its efficiency as defense against herbivores.^12,13 In addition to these endogenous sources of defensive variability, the expression of plant traits strongly depends on multiple external factors such as temperature or availability of plant nutrients, water or light (Fig. 1).¹⁴ At the same time, the outcome of herbivore-plant interactions is crucially determined by biotic interactions. Plant interactions with mutualistic microorganisms such as Rhizobia, mycorrhiza and above-ground fungal endophytes as well as tri-trophic interactions with predators and parasitoids of herbivores can all strongly impact plant fitness.¹⁵Open in a separate windowFigure 1Factors influencing variability of plant defenses. Plant defensive traits are affected by various endogenous and external factors. Endogenous factors comprise plant genotype and ontogeny of plants or plant parts. External factors can be categorized as abiotic or biotic. Important abiotic factors that can influence plant defenses are light exposure, temperature, soil salinity, as well as water and nutrient availability. Biotic factors that can have an effect on plant defense are interspecific interaction with Rhizobia (in the case of legumes), mycorrhizal and endophytic fungi, pathogens as well as the interaction with conspecifics or different plant species.Variability in herbivore-plant interactions can also be associated with herbivore variation. Different attackers of a particular plant species might be affected in different ways by toxins in food plants (Fig. 1). The efficiency of a specific defensive compound can also depend on the feeding mode, i.e., sucking or chewing, as well as on the degree of specialization of the herbivore to the respective plant.¹⁶ Defenses mediated by secondary plant compounds are generally believed not to affect specialist herbivores, because of their capacity to tolerate or to detoxify defensive compounds of their hosts by behavioral or physiological adaptations.^17–20 In this context, the specialist herbivore paradigm predicts that adapted herbivores are less affected by a given chemical defense than generalists,^21,22 although exceptions have been noted.^23–25While it is important to consider these numerous sources of variation affecting the outcome of herbivore-plant interactions when designing functional studies, a significant fraction of the variability in natural systems will always remain unidentified. Consequently, approaches combining field observations with experiments under controlled conditions provide a powerful tool to uncover functional interactions between plants and their multiple antagonists in nature.In a recent study, we analyzed the importance of wild lima bean''s cyanogenesis—i.e., the release of toxic hydrogen cyanide from preformed precursors in response to cell damage—as plant defense at a natural site in South Mexico.²⁵ Although cyanogenesis is generally considered an efficient direct defense against herbivores, in numerous studies plant cyanide production had little or no effect on herbivores.^26–28 One would like to think that most of these inconsistencies in cyanogenesis-based herbivore defense efficiency could be explained by one or more sources of variation mentioned above. Nevertheless, field studies analyzing the action of plant cyanogenesis on a quantitative basis have been scarce. In our study, a two-step approach was used to gain insight into the function of cyanogenesis in nature.²⁵ First, cyanide concentration and herbivore damage were quantified by measuring removed leaf area of individual leaves derived from different individual plants while considering microclimate conditions. Significant negative correlations between cyanogenesis and leaf damage were observed. Second, since existing correlations do not necessarily indicate causal associations, we conducted consecutive feeding experiments under controlled conditions. To consider natural variability of lima beans'' cyanogenesis observed in nature in our analysis, we prepared clones from field-grown plants with different but defined cyanogenic features. These clonal plants showed high constancy of cyanogenic traits compared to their respective mother plants and thus, could be used in comparative analyses. Every effort was made to duplicate natural conditions and so herbivore species selected for feeding trials represented those identified in the field as the most important plant consumers at the respective site (pers. observ.). Feeding trials supported our hypothesis that cyanogenesis has quantitative effects on herbivore behavior in nature and explained the negative correlation of lima bean''s cyanogenesis and herbivory observed in the field.Analytical approaches combining field observations with controlled experiments help to explain natural patterns and may represent a powerful methodological approach for functional analyses of herbivore-plant interactions.     

The ontogeny of plant indirect defenses

Plants frequently attract natural enemies of their herbivores, resulting in a reduction in tissue damage and often in enhanced plant fitness. While such indirect defenses can dramatically change as plants develop, only recently have ecologists begun to explore such changes and evaluate their role in mediating plant–herbivore–natural enemy interactions. Here we review the literature documenting ontogenetic patterns in plant rewards (i.e. extrafloral nectaries (EFNs), food bodies (FBs) and domatia) and volatile organic compounds (VOCs), and identify links between ontogenetic patterns in such traits and the attraction of natural enemies (ants). In the case of reward traits we concentrate in ant–plant studies, which are the most numerous. We report that all indirect defensive traits commonly vary with plant age but ontogenetic trajectories differ among them. Myrmecophytic species, which provide both food and shelter to their defenders, do not produce rewarding traits until a minimum size is reached. Then, a pronounced increase in the abundance of food rewards and domatia often occurs as plants develop, which explains the temporal succession or colony size increase of mutualistic ant species and, in some cases, leads to a reduction in herbivore damage and enhanced fitness as plants age. In contrast, ontogenetic patterns were less consistent in plant species that rely on VOC emissions to attract natural enemies or those that provide only food rewards (EFNs) but not nesting sites to their associated ants, showing an overall decline or lack of trend with plant development, respectively. Future research should focus on uncovering: (i) the costs and mechanisms underlying ontogenetic variation in indirect defenses, (ii) the relative importance of environmental and genetic components shaping these ontogenetic trajectories, and (iii) the consequences of these ontogenetic trajectories on plant fitness. Advances in this area will shed light on the context dependency of bottom-up and top-down controls of herbivore populations and on how natural selection actually shapes the ontogenetic trajectories of these traits.     

Stimulating natural defenses in poplar clones (OP-367) increases plant metabolism of carbon tetrachloride

Groundwater contamination by carbon tetrachloride (CCl4) presents a health risk as a potential carcinogen and pollutant that is capable of depleting the ozone layer. Although use of poplar trees in a phytoremediation capacity has proven to be cost effective for cleaning contaminated sites, minimizing leaf emission of volatile contaminants remains a pressing issue. We hypothesized that recently fixed carbon plays a key role in CCl4 metabolism in planta yielding nonvolatile trichloroacetic acid (TCA) and that the extent of this metabolism can be altered by heightening plant defenses. Labeling intact leaves with (11)CO2 (t 1/2 20.4 m) can test this hypothesis, because the extremely short half-life of the tracer reflects only those processes involving recently fixed carbon. Using radio-HPLC analysis, we observed [(11)C]TCA from leaf extract from poplar clones (OP-367) whose roots were exposed to a saturated solution of CCl4 (520 ppm). Autoradiography of [(11)C]photosynthate showed increased leaf export and partitioning to the apex within 24 h of CCl4 exposure, suggesting that changes in plant metabolism and partitioning of recently fixed carbon occur rapidly. Additionally, leaf CCl4 emissions were highest in the morning, when carbon pools are low, suggesting a link between contaminant metabolism and leaf carbon utilization. Further, treatment with methyljasmonate, a plant hormone implicated in defense signal transduction, reduced leaf CCl4 emissions two-fold due to the increased formation of TCA.     

Regulation of enteric endophytic bacterial colonization by plant defenses

Bacterial endophytes reside within the interior of plants without causing disease or forming symbiotic structures. Some endophytes, such as Klebsiella pneumoniae 342 (Kp342), enhance plant growth and nutrition. Others, such as Salmonella enterica serovar Typhimurium (S. typhimurium), are human pathogens that contaminate raw produce. Several lines of evidence are presented here to support the hypothesis that plant defense response pathways regulate colonization by endophytic bacteria. An ethylene-insensitive mutant of Medicago truncatula is hypercolonized by Kp342 compared to the parent genotype. Addition of ethylene, a signal molecule for induced systemic resistance in plants, decreased endophytic colonization in Medicago spp. This ethylene-mediated inhibition of endophytic colonization was reversed by addition of the ethylene action inhibitor, 1-methylcyclopropene. Colonization of Medicago spp. by S. typhimurium also was affected by exogenous ethylene. Mutants lacking flagella or a component of the type III secretion system of Salmonella pathogenicity island 1 (TTSS-SPI1) colonize the interior of Medicago spp. in higher numbers than the wild type. Arabidopsis defense response-related genotypes indicated that only salicylic acid (SA)-independent defense responses contribute to restricting colonization by Kp342. In contrast, colonization by S. typhimurium is affected by both SA-dependent and -independent responses. S. typhimurium mutants further delineated these responses, suggesting that both flagella and TTSS-SPI1 effectors can be recognized. Flagella act primarily through SA-independent responses (compromising SA accumulation still affected colonization in the absence of flagella). Removal of a TTSS-SPI1 effector resulted in hypercolonization regardless of whether the genotype was affected in either SA-dependent or SA-independent responses. Consistent with these results, S. typhimurium activates the promoter of PR1, a SA-dependent pathogenesis-related gene, while S. typhimurium mutants lacking the TTSS-SPI1 failed to activate this promoter. These observations suggest approaches to reduce contamination of raw produce by human enteric pathogens and to increase the number of growth-promoting bacteria in plants.     

ATP hydrolyzing salivary enzymes of caterpillars suppress plant defenses

The oral secretions of herbivores are important recognition cues that can be used by plants to mediate induced defenses. In this study, a degradation of adenosine-5'-triphosphate (ATP) in tomato leaves was detected after treatment with Helicoverpa zea saliva. Correspondingly, a high level of ATPase activity in saliva was detected and three ATP hydrolyzing enzymes: apyrase, ATP synthase and ATPase 13A1 were identified in salivary glands. To determine the functions of these proteins in mediating defenses, they were cloned from H. zea and expressed in Escherichia coli. By applying the purified expressed apyrase, ATP synthase or ATPase 13A1 to wounded tomato leaves, it was determined that these ATP hydrolyzing enzymes suppressed the defensive genes regulated by the jasmonic acid and ethylene pathways in tomato plant. Suppression of glandular trichome production was also observed after treatment. Blood-feeding arthropods employ 5'-nucleotidase family of apyrases to circumvent host responses and the H. zea apyrase, is also a member of this family. The comparatively high degree of sequence similarity of the H. zea salivary apyrase with mosquito apyrases suggests a broader evolutionary role for salivary apyrases than previously envisioned.     

Relaxation of putative plant defenses in a tropical agroecosystem

Evidence of the effects of agriculture on natural systems is widespread, but potential evolutionary responses in nontarget species are largely uncharacterized. To explore whether exposure to agrochemicals may influence selective pressures and phenotypic expression in nonagricultural plant populations, we characterized the expression of putative antiherbivore defense phenotypes in three nonagricultural species found upstream and downstream of irrigated rice fields in Guanacaste Province, Costa Rica. We found that plants downstream of chemically intensive agriculture showed shifts toward reduced expression of putative antiherbivore defenses relative to upstream counterparts. In two of three tested species, leaf extracts from downstream plants were more palatable to a generalist consumer, suggesting a possible reduction of chemical defenses. In one species with multiple modes of putative defenses, we observed parallel reductions of three metrics of putative biotic and physical defenses. These reductions were concurrent with reduced herbivore damage on downstream plants. Together, these results suggest that agriculture has the potential to alter intraspecific phenotypic expression, ecological interactions, and natural selection in nontarget plant populations.     

Early signaling events induced by elicitors of plant defenses

Plant pathogen attacks are perceived through pathogen-issued compounds or plant-derived molecules that elicit defense reactions. Despite the large variety of elicitors, general schemes for cellular elicitor signaling leading to plant resistance can be drawn. In this article, we review early signaling events that happen after elicitor perception, including reversible protein phosphorylations, changes in the activities of plasma membrane proteins, variations in free calcium concentrations in cytosol and nucleus, and production of nitric oxide and active oxygen species. These events occur within the first minutes to a few hours after elicitor perception. One specific elicitor transduction pathway can use a combination or a partial combination of such events which can differ in kinetics and intensity depending on the stimulus. The links between the signaling events allow amplification of the signal transduction and ensure specificity to get appropriate plant defense reactions. This review first describes the early events induced by cryptogein, an elicitor of tobacco defense reactions, in order to give a general scheme for signal transduction that will be use as a thread to review signaling events monitored in different elicitor or plant models.     

Linking aboveground and belowground interactions via induced plant defenses

Plants have a variety of chemical defenses that often increase in concentration following attack by herbivores. Such induced plant responses can occur aboveground, in the leaves, and also belowground in the roots. We show here that belowground organisms can also induce defense responses aboveground and vice versa. Indirect defenses are particularly sensitive to interference by induced feeding activities in the other compartment, and this can disrupt multitrophic interactions. Unravelling the involvement of induced plant responses in the interactions between aboveground and belowground communities associated with plants is likely to benefit from comprehensive metabolomic analyses. Such analyses are likely to contribute to a better understanding of the costs and benefits involved in the selection for induced responses in plants.     

Strategies used by bacterial pathogens to suppress plant defenses

Plant immune systems effectively prevent infections caused by the majority of microbial pathogens that are encountered by plants. However, successful pathogens have evolved specialized strategies to suppress plant defense responses and induce disease susceptibility in otherwise resistant hosts. Recent advances reveal that phytopathogenic bacteria use type III effector proteins, toxins, and other factors to inhibit host defenses. Host processes that are targeted by bacteria include programmed cell death, cell wall-based defense, hormone signaling, the expression of defense genes, and other basal defenses. The discovery of plant defenses that are vulnerable to pathogen attack has provided new insights into mechanisms that are essential for both bacterial pathogenesis and plant disease resistance.     

Avoidance and suppression of plant defenses by herbivores and pathogens

Plants are nutritious and hence herbivores and phytopathogens have specialized to attack and consume them. In turn, plants have evolved adaptations to detect and withstand these attacks. Such adaptations we call ‘defenses’ and they can operate either directly between the plant and the plant consumer or indirectly i.e. when taking effect via other organisms such as predators and parasitoids of herbivores. Plant defenses put selection pressure on plant-consumers and, as a result, herbivores and pathogens have evolved counter-adaptations to avoid, resist, or manipulate plant defenses. Here we review how plant consumers have adapted to cope with plant defenses and we will put special emphasis on the phenomenon of suppression of plant defenses.     

Community-level relaxation of plant defenses against herbivores at high elevation

Ecological gradients shape the evolution of plant functional traits through variation in climate, abiotic factors, and biological processes. While recent observations highlight elevation clines in functional trait variation, several traits vary independently of others and vary in species-specific manner, thereby hindering community-level generalization. In this study, we examined whether whole plant communities at different elevations independently converge to elevation-specific trait values. Using a combination of field-derived trait quantification and a literature survey, we analyzed eight functional traits relating to plant palatability [specific leaf area (SLA), leaf dry matter content, carbon-to-nitrogen ratio], and defenses (spines, trichomes, silica, and the relative importance of phenolic- or non-phenolic-based chemical defenses in plants) of plants growing along several mountain transects in Switzerland, and computed community-level means for each trait. We observed a general decline of most defense traits in communities at high elevations. The individual traits’ patterns underlined a concerted decline of the physico-chemical defense syndrome at high elevation, and low SLA values at high elevation also indicate investment in the production of leaves with slow turnover rates. In addition, we found that high-elevation plant trait values tend to converge, while at low elevation, to diverge within communities. Our data suggest that community-level defense syndromes along elevation gradients are influenced by ecological filtering. Lower herbivore pressures select for reduced plant defenses, while harsh environmental conditions select for investment in tolerance-related traits. Secondly, since species are highly variable in their functional trait space within communities, niche-partitioning theory might explain which variabilities of the defense functional space are selected within communities, particularly at low elevation.     

Gall insects can avoid and alter indirect plant defenses

Parasitic species can dramatically alter host traits. Some of these parasite-induced changes can be considered adaptive manipulations that benefit the parasites. Gall-inducing insects are parasites well known for their ability to alter host-plant morphology and physiology, including the distribution of plant defensive compounds. Here it was investigated whether gall-inducing species alter indirect plant defenses, involving the release of volatile compounds that are attractive to foraging natural enemies. Using field and factorial laboratory experiments, volatile production by goldenrod (Solidago altissima) plants was examined in response to attack by two gall-inducing species, the tephritid fly Eurosta solidaginis and the gelechiid moth Gnorimoschema gallaesolidaginis, as well as the meadow spittlebug, Philaenus spumarius, and the generalist caterpillar Heliothis virescens. Heliothis virescens elicited strong indirect defensive responses from S. altissima, but the gall-inducing species and spittlebugs did not. More significantly, infestation by E. solidaginis appeared to suppress volatile responses to subsequent attack by the generalist caterpillar. The extensive control that E. solidaginis apparently exerts over host-plant defense responses may reduce the predation risk for the gall inducer and the subsequent herbivore, and could influence community-level dynamics, including the distribution of herbivorous insect species associated with S. altissima parasitized by E. solidaginis.     

Ecological consequences of evolution in plant defenses in a metacommunity

Dispersal can affect the assembly of local communities in a metacommunity as well as evolution of local populations in a metapopulation. These two processes may also affect each other in ways that have not yet been well studied and that may have novel effects on community structure. Here, we illustrate the interaction of these two processes on community structure with a model of adaptive evolutionary dynamics of plant defenses in a metacommunity food web involving multiple patches along a productivity gradient. We find an enhanced suite of adaptive plant types in our metacommunity model than is predicted in the absence of dispersal. We also find that this, and the movement of nutrients among patches via dispersal, alters patterns of food web architecture, trophic structure and diversity along the productivity gradient. Overall, our model illustrates that evolutionary and metacommunity dynamics may influence communities in complex interactive ways that may not be predicted by models that ignore either of these types of processes.