Induced responses to herbivory in wild radish: effects on several herbivores and plant fitness.
Plants have evolved defenses against their predators and parasites that do not involve fleeing areas with high risk of attack. Because plant attackers are not always predictable in time and space, and defenses are thought to impose a cost, it is believed that plants use damage as a cue to induce resistance against subsequent herbivores. Such induced responses appear to be common among a very wide array of plants (Karban and Baldwin 1997). Animals, such as mussels (Reimer and Tedengren 1996), bryozoans (Harvell 1990), frogs (McCollum and Van Buskirk 1996), and ants (at the colony level) (Passera et al. 1996), also have well-developed inducible response systems. To understand the evolutionary ecology of inducible response systems, and whether they are examples of adaptive plasticity (sensu Dudley and Schmitt 1996), it is important to measure the fitness consequences (benefits and costs) of induction for the inducing organism.
Karban and Myers (1989) divided induced responses to herbivory into three categories: responses, resistance, and defense. (1) Induced responses are changes in the plant that occur after herbivory. Such changes may be incidental, and range from shifts in water content (Faeth 1992), leaf toughness (Kudo 1996), and nitrogen uptake (Jaramillo and Detling 1988), to changes in secondary chemistry (Baldwin 1994, Karban and Baldwin 1997), trichome density (Baur et al. 1991; A. A. Agrawal, unpublished manuscript), and traits attractive to predaceous arthropods (Smith et al. 1990, Shimoda et al. 1997, Agrawal 1998a, De Moraes et al. 1998, Agrawal and Rutter 1998). (2) Induced resistance is a change in the plant that reduces the preference or performance of subsequently attacking herbivores. Induced resistance can be caused by a variety of nonexclusive biochemical and physical resistance mechanisms. For example, nicotine production is induced in wild tobacco plants and has been causally linked to reduced performance of tobacco herbivores (Baldwin 1991). Similarly, spines of African Acacia trees have been shown to increase in length following giraffe herbivory (Young 1987); plants with larger spines experience reduced herbivory compared to plants with smaller spines (Milewski et al. 1991). (3) Induced defense is a term reserved for cases where induction results in a fitness benefit to induced plants compared to plants that were not induced. Evidence for induced defense is limited, and similarly, most plant traits thought to be defensive are not studied from the plant's perspective (Hartley and Jones 1997). Although induced resistance has been reported from over 100 plant-herbivore systems, and induced responses to herbivory are thought to be adaptive (Green and Ryan 1972, Haukioja and Neuvonen 1985, Baldwin 1988, Wold and Marquis 1997), there is little information available on the effects of induction on plant fitness (Karban and Baldwin 1997).
Negative consequences of induction for herbivores do not necessarily lead to benefits for the plant. For example, slower growing larvae on induced plants may consume more plant tissue than larvae on control plants (sensu Slansky and Feeny 1977). In addition, many plants tolerate herbivory or can compensate for losses to herbivores (Paige 1992, Trumble et al. 1993, Rosenthal and Kotanen 1994, Agrawal et al. 1999b, Strauss and Agrawal 1999). It is necessary to demonstrate fitness benefits to the plant to show that induction is an adaptive plant defense and a strategy that may be favored by natural selection.
I link these three levels (plant responses, resistance, and defense) in the study of induced responses to herbivory in annual wild radish plants. My goal was to examine induced plant responses in the context of the adaptive plasticity hypothesis (Dudley and Schmitt 1996) with special regard to the benefits of induction for plant fitness. Specifically, I asked: (1) Does herbivory induce changes in the density and total number of foliar trichomes, which are putative defenses against herbivores? (2) Does herbivory or application of a natural plant response elicitor, jasmonic acid, induce local and/or systemic resistance affecting herbivores' preference and performance in choice and no-choice experiments? (3) In the field, do induced responses protect plants from subsequent herbivory and increase plant fitness?
Study organisms and general methods
Wild radish plants, Raphanus raphanistrum and R. sativus (Brassicaceae), are cosmopolitan herbaceous annuals common to disturbed sites. My experiments on induced responses to herbivory in wild radish are divided into three categories: responses, resistance, and defense. Pieris rapae (Lepidoptera: Pieridae) larvae were used to induce responses in wild radish plants. P. rapae is a specialist herbivore that feeds only on plants in the Brassicaceae. All P. rapae larvae used in these experiments were obtained from a colony of recently field-collected individuals maintained in the greenhouse. The following general procedure was used to study induction. Plants were randomly divided into control and induced groups. One P. rapae larva was confined to one leaf of the induced plants using a clip cage, and the larva was allowed to feed on the entire leaf. Control plants received a clip cage without the caterpillar. Clip cages were made from the tops of ventilated Petri dishes (5 cm) attached to either side of a hair clip. The cage structure was supported by a wooden stake, so as not to weigh down the leaves.
Induced responses to herbivory
In previous experiments I documented that foliar herbivory to wild radish resulted in systemic increases in concentrations of indole glucosinolates, putative defense compounds (Agrawal et al. 1999b). In the current study I investigated the effects of herbivory on production of setose trichomes, an additional putative defense of wild radish plants [ILLUSTRATION FOR FIGURE 1 OMITTED]. Agren and Schemske (1993) showed that closely related Brassica rapa plants selected for high densities of setose trichomes had a reduced leaf area consumed by P. rapae larvae, compared to plants selected for low densities of trichomes.
Raphanus sativus plants were grown in a greenhouse in 0.8-L pots using University of California greenhouse soil mix. Thirty-six plants were randomly assigned to control or induced groups. When the plants had developed two fully expanded true leaves, one P. rapae larva was caged on one of the first true leaves of each induced plant. Caterpillars consumed the leaf within 48 h. After 3 wk, when all plants had eight true leaves and were beginning to flower, the plants were harvested for trichome measurements. From each plant, the third, fifth, and seventh leaves were removed, and the area of each leaf was measured using a digital portable area meter (Li-Cor 3000, Lambda Instruments Corporation, Lincoln, Nebraska). Leaves from both treatments were at the same developmental stage and were fully expanded, except for the seventh leaves, which were still expanding. I removed a leaf disc (1.8 [cm.sup.2]) from the apex of each leaf using a number 10 cork borer, and counted the number of trichomes on the top and bottom of each leaf disc.
Two aspects of the number of trichomes on each leaf may be relevant to subsequent insect herbivory. First, an increased density of trichomes may create a less favorable environment for herbivores (Agren and Schemske 1993). However, the density of trichomes may simply reflect changes in leaf size. For example, on two leaves with an equal number of trichomes, there will be a higher density of trichomes on the smaller leaf. Therefore, I estimated both the density and the total numbers of trichomes on each leaf. Densities are presented as trichomes per square centimeter, and total trichomes were calculated by multiplying trichomes per square centimeter by the area of the leaf. A multivariate analysis of variance (MANOVA) was conducted on leaves at each developmental stage (position) to test for the effects of treatment (induction) on leaf size and trichome density. A separate ANOVA was conducted to detect effects of treatment on total trichome number for each leaf. A sequential Bonferroni correction was used to adjust the [[Alpha].sub.005] values for the different analyses of different leaf positions (Rice 1989). Total trichome number was not included in the MANOVA because this variable was calculated by multiplying leaf size by trichome density.
Induced resistance to herbivory
Two experiments were conducted with R. raphanistrum to test for induced resistance to a generalist noctuid herbivore, Spodoptera exigua. S. exigua eggs were obtained from the USDA (Stoneville, Mississippi). In both experiments plants were grown in the greenhouse as described above and randomly assigned to induction and control treatments. In these experiments, the fourth true leaf was used as the treatment leaf, and was exposed to P. rapae herbivory as described above. In the first experiment, an additional treatment was imposed in which I sprayed each plant liberally with 0.5 mmol/L jasmonic acid (control n = 9, caterpillar induced n = 8, jasmonic acid n = 9). Approximately 1 mL of solution was delivered to each plant. Jasmonic acid is naturally found in plants and can be used as an elicitor to stimulate plant responses to herbivory (Bodnaryk and Rymerson 1994, Doughty et al. 1995, Baldwin 1996, Thaler et al. 1996).
Jasmonic acid (JA) is produced as part of the octadecanoid phytochemical pathway and increases in plants following wounding and natural herbivory (Albrecht et al. 1993, McCloud and Baldwin 1997). Its increase in response to herbivory is thought to be one of the initial stages of the octadecanoid cascade of biochemical events relevant to induced resistance, and this response pathway is highly conserved among plant families (Bergey et al. 1996, Karban and Baldwin 1997). Direct application or consumption of jasmonic acid by herbivores does not affect their performance (Avdiushko et al. 1997; J. S. Thaler, personal communication).
Three days after treatment, the newest fully expanded leaf was removed from each plant (in all treatments) with a razor blade and placed in a 90-mm Petri dish with moistened filter paper. A first-instar neonate of S. exigua was added to each Petri dish and the dish was sealed. Caterpillars were weighed wet after 48 h and the effects of induction on mass gain were analyzed using a one-way ANOVA. Initial masses were not taken because all caterpillars were freshly hatched neonate larvae, and it was assumed that they started at equal masses. Masses were natural log(x + 1) transformed to equalize scedasticity. I also performed two preplanned contrasts between (1) control vs. caterpillar-damaged plants, and (2) caterpillar-damaged plants vs. plants sprayed with jasmonic acid.
The second experiment was conducted to test if leaves that expanded after the damage event would still show induced resistance. Plants were grown and subjected to herbivory by P. rapae as described above, but the subsequent challenge was made 1 wk after treatment on a newly formed leaf not present at the time of treatment. Second-instar S. exigua larvae were caged onto the intact assay leaf using a clip cage and were allowed to feed for 3 d. The assay leaves were at the same developmental stage across treatments. The caterpillars were weighed before and after use to calculate mass gain. Mass gain values were natural log(x + 1) transformed to equalize scedasticity and compared between control and induced plants using a t test.
I conducted several additional experiments with R. sativus to characterize induced resistance to herbivory. In the first experiment, plants were grown and assigned to control and induced treatments as above (n = 23 control, and n = 27 induced plants). Three d after treatment the newest fully expanded leaf was removed from each plant with a razor blade and placed in a 90-mm Petri dish with moistened filter paper. A first-instar neonate of S. exigua was added to each Petri dish and it was sealed for 3 d. Caterpillars were weighed and the leaf area consumed was estimated using an acetate grid. In addition, the gross growth efficiency (GGE) was calculated as: mass gained per leaf area consumed (Waldbauer 1968). Because GGE is an estimate of caterpillar growth per unit area of plant consumed, it was used to reflect toxic or antinutritive effects of induction. Effects of induction on these variables were analyzed using MANOVA.
Two choice tests were conducted to test for effects of induced responses on herbivory by more mobile herbivores. In the first experiment, a natural outbreak of leaf-mining flies, Liriomyza sp., (Diptera: Agromyzidae) was present in the greenhouse, and 39 plants were introduced (20 control, 19 induced) for 1 wk. The plants were arranged randomly with respect to treatment over two greenhouse benches. After allowing another week for the flies to oviposit and the eggs to hatch, I counted the number of successfully initiated mines on each plant. The number of mines on control and induced plants was compared using a t test. In the second choice test, I used naturally occurring R. sativus plants growing in a population near the Orchard Park vicinity of the University of California at Davis campus, and a mobile noctuid caterpillar, Helicoverpa zea. H. zea eggs were obtained from the USDA (Stoneville, Mississippi). I randomly assigned 13 of 26 plants to be damaged by a caged P. rapae larva, while the other 13 remained as controls. Four days after treatments were imposed, I removed a single undamaged leaf from each of the 26 plants and paired a control leaf with a similarly-sized leaf from an induced plant. Paired leaves were placed in a 90-mm petri dish with moistened filter paper and a third-instar H. zea. The percentage of area consumed of each leaf was estimated after 48 h using an acetate grid. A nonparametric sign test was employed to detect effects of induction on the feeding choice of the caterpillars. The leaf with a greater percentage of area consumed was classified as the preferred leaf, and replicates in which an equal amount of leaf area was removed were omitted from the analysis (Zar 1996).
Finally, I conducted a no-choice experiment to test for effects of induced responses to herbivory in R. sativus on growth of a specialist herbivore, P. rapae. Plants were grown and treated with control and induced treatments as above, and one first-instar neonate caterpillar was placed on leaves in petri dishes (n = 12 control, and n = 7 induced plants). Mass gain after 3 d was compared using a t test.
Induced defense against herbivory
I conducted field experiments over 2 yr to detect fitness consequences of induced responses to herbivory for annual wild radish plants. The experiments were conducted at the Blodgett Forest Research Station, near Georgetown, California, in the Sierra Nevada mountain range (1300 m). The research plot was a plowed, disturbed site where several weedy brassicaceous plants had been growing for several years. In both years of the experiment, cotyledonary plants were transplanted to the field from seeds germinated in greenhouse plug trays. In 1996, only 42 experimental R. raphanistrum plants were available after transplant because of mortality from early-season drought. In 1997, 148 R. sativus plants were used.
In both years, plants were equally divided into three treatments: induced plants, leaf damage controls, and overall (unmanipulated) controls. The plants were treated just before the time that grasshoppers (the main herbivore at this site) begin to severely damage the plants. I induced plants by caging a caterpillar larva (P. rapae) on one leaf at the four-leaf stage and allowing it to consume the entire leaf. Leaf damage controls had one leaf clipped off at the petiole with a scissors. Such clipping resulted in an amount of leaf tissue removed equal to that in the induced treatment, but without the associated induced plant response (Haukioja and Neuvonen 1985, Hartley and Lawton 1987, Mattson and Palmer 1988, Baldwin 1990, Bodnaryk 1992, Mattiacci et al. 1995, Alborn et al. 1997, Agrawal et al. 1999b). Induced plant responses are thought to be minimized by clipping damage because of the absence of herbivore saliva, the greatly reduced area of actual leaf tissue that is damaged, and quick nature of the removal. The leaf damage control treatment was used to factor out effects of leaf removal on herbivores and plant fitness. In other words, leaf removal associated with induction causes changes in plant size, which may be perceived by herbivores, as well as an energetic drain on the plant. Because the induction treatment had the same amount of leaf tissue removal as the leaf damage control treatment (without induction), I was able to detect the independent effects of leaf removal and induction on plant fitness. It should be noted that the pattern of leaf removal in wild radish plants has been shown to affect plant fitness (Mauricio et al. 1993).
During each experiment, plants were censused three times for the percentage of leaf area consumed by naturally occurring herbivores. The primary herbivore at this site was the crackling forest grasshopper, Trimerotropis surfusa (A. A. Agrawal, personal observation; M. A. Salser, personal communication). At the end of the season I counted the total number of fruits and seeds produced by each plant and weighed the total fruit mass, to use as indicators of female plant reproductive fitness. Arcsine square-root transformed percentage data (herbivory) were compared using a two-factor repeated-measures ANOVA, with treatment and trial (year) as the main effects. The repeated measures were the censuses of herbivory on the plants three times in each trial. Fitness measures were analyzed using a two-factor (treatment and trial) MANOVA with fruits, fruit mass, and seed number as the response variables. I also performed two preplanned contrasts on the herbivory and fitness data: (1) overall controls vs. induced plants, and (2) overall controls vs. leaf damage controls.
Induced responses to herbivory
Herbivory to the first true leaf affected the leaf size and density of trichomes on subsequently formed leaves (Table 1, [ILLUSTRATION FOR FIGURES 1 AND 2 OMITTED]. The third newly formed leaf was larger on induced plants compared to controls. There was a trend for the third true leaf to have higher densities of trichomes on induced plants compared to controls and a marginal effect indicating the fifth true leaf to be larger on induced plants compared to controls. Leaf size and trichome density on the seventh true leaf were not affected by treatments. Induction caused an increase in the total number of trichomes on the third ([F.sub.1,38] = 7.559, P = 0.009, adjusted [[Alpha].sub.0.05] = 0.017) and marginally on the fifth ([F.sub.1,38] = 4.377, P = 0.043, adjusted [[Alpha].sub.0.05] = 0.025) true leaf, but not the seventh true leaf ([F.sub.1,32] = 0.024, P = 0.878, adjusted [[Alpha]0.05] = 0.05).
TABLE 1. Multivariate analysis of variance (MANOVA) and univariate analyses for effects of induced responses to herbivory on size and trichome density of Raphanus sativus leaves. Variable Wilks' [Lambda] df F P MANOVA (leaf 3) 0.801 2, 37 4.585 0.017(*) Size 1, 38 6.349 0.016 Density 1, 38 2.544 0.119 MANOVA (leaf 5) 0.906 2, 37 1.922 0.161(*) Size 1, 38 2.995 0.092 Density 1, 38 0.014 0.907 MANOVA (leaf 7) 0.997 2, 31 0.048 0.953(*) Size 1, 32 0.098 0.756 Density 1, 32 0.048 0.828 * Sequential Bonferroni corrected [[Alpha].sub.0.05] = 0.017 (leaf 3), 0.025 (leaf 5), and 0.05 (leaf 7).
Induced resistance to herbivory
Caterpillar herbivory or exposure to the natural plant hormone jasmonic acid induced resistance in R. raphanistrum plants when I assayed growth of S. exigua, ([ILLUSTRATION FOR FIGURE 3 OMITTED], experiment 1: [F.sub.2,23] = 29.743, P [less than] 0.001; contrast: control vs. caterpillar damaged: [F.sub.1,23] = 20.176, P [less than] 0.001). The jasmonic acid dose used (0.5 mmol/L) induced greater resistance than did herbivory to the fourth true leaf ([ILLUSTRATION FOR FIGURE 3 OMITTED], contrast: caterpillar damaged vs. jasmonic acid: [F.sub.1,23] = 8.650, P = 0.007). Induced resistance persisted in newly formed leaves of damaged plants, reducing the mass of caterpillars feeding on induced plants compared to uninduced controls ([ILLUSTRATION FOR FIGURE 3 OMITTED], experiment 2; t = 5.194, df = 14.8, P [less than] 0.001).
Leaf herbivory to R. sativus increased resistance to generalist S. exigua larvae ([ILLUSTRATION FOR FIGURE 4 OMITTED], Table 2). Induction appeared primarily to reduce the amount of leaf area consumed (and subsequent gain in mass), but did not affect gross growth efficiency (Table 2). I did not detect any effect of induced responses on the growth of specialist P. rapae larvae (mean [+ or -] SE: controls = 0.867 [+ or -] 0.168 g; induced = 1.086 [+ or -] 0.352 g; t = 0.635, df = 17, P = 0.534).
In choice experiments, induction reduced the number of generalist leaf mines successfully initiated by agromizid flies ([ILLUSTRATION FOR FIGURE 5a OMITTED], t = 2.232, df = 37, P = 0.032). When a mobile generalist caterpillar, H. zea, was offered paired leaves from control and induced R. sativus plants, the caterpillars preferred control leaves in 8 out of 9 replicates where there was a difference in the percent leaf area consumed ([ILLUSTRATION FOR FIGURE 5b OMITTED], P = 0.039, sign test). Overall, caterpillars fed nearly twice as much on leaves from control plants (70.6 [+ or -] 6.9%) compared to leaves from induced plants (35.4 [+ or -] 6.2%).
Induced defense against herbivory
In field experiments, induced responses to early-season caterpillar herbivory decreased subsequent herbivory by grasshoppers in both years of study ([ILLUSTRATION FOR FIGURE 6 OMITTED], Table 3, MANOVA contrast: control vs. induced: Wilks' [Lambda] = 0.923, df = 3, 142, F = 3.953, P = 0.010). Experimental clipping damage with a scissors removed an equal amount of leaf area as the induction treatment, but did not affect the amount of subsequent herbivory experienced by the plants ([ILLUSTRATION FOR FIGURE 6 OMITTED], Table 3, MANOVA contrast: overall control vs. leaf damage control: Wilks' [Lambda] = 0.983, df = 3, 142, F = 0.808, P = 0.492). These effects of induced responses on herbivory resulted in lifetime fitness differences between plants. Induced plants outperformed overall controls and leaf damage controls in both years of this study ([ILLUSTRATION FOR FIGURE 7 OMITTED], Table 4). Results for univariate analyses were consistent with the MANOVA for each of the fitness components measured: fruit number, fruit mass, and seed number (P [less than] 0.015 in all cases).
Foliar herbivory in wild radish plants resulted in induced biochemical and physical responses, increased [TABULAR DATA FOR TABLE 2 OMITTED] resistance to several herbivores, and increased seed set compared to uninduced control plants. Investigators rarely examine plant responses to herbivory at these three scales. Measures of the changes in the plant following herbivory provide information about potential mechanisms responsible for resistance to herbivores and are the most common measure of induction. Demonstrations of induced resistance are less common, and are necessary to infer the mode of the subsequent effects on plant performance. Effects of induction on plant fitness are almost completely lacking (Karban 1986, 1993, Brown 1988, Baldwin et al. 1990, Karban and Baldwin 1997, Baldwin 1998). The lack of evidence for fitness benefits of induced responses is particularly surprising because such demonstrations are needed to consider induced responses as defensive. The current study is one of the few demonstrations of such fitness benefits (Agrawal 1998b, Baldwin 1998).
Induced responses that take the form of increases in physical defenses have been reported much less often than biochemical responses (see McNaughton and Tarrants 1983, Young 1987, Baur et al. 1991, Myers and [TABULAR DATA FOR TABLE 3 OMITTED] Bazely 1991, Gowda 1997, Young and Okello 1998; A.A. Agrawal, unpublished manuscript). Herbivory to the first true leaf of wild radish induced an increase in the densities of trichomes on the third true leaf. Such a response may potentially be mediated by a change in leaf size (i.e., smaller leaves may have higher densities of trichomes). For example, Obeso (1997) showed that browsed branches of holly trees produced leaves with increased spinescence. However, this effect was largely due to a decrease in the size of leaves on browsed branches. In the current study, although trichome [TABULAR DATA FOR TABLE 4 OMITTED] densities increased, the mean size of the third true leaf of damaged plants was also larger than that of undamaged plants. Thus, although an increase in leaf size could have reduced the density of trichomes due to a "dilution effect," it did not, because of an overall increase in trichome production. For the fifth true leaf, the densities of trichomes were the same for damaged and control plants. However, total numbers of trichomes were marginally increased on damaged plants, indicating that induced plants produced larger leaves. It is not clear why induced plants produced larger leaves, although this may be a compensatory response to the herbivory and loss of photosynthetic area (Strauss and Agrawal 1999).
Although others have shown that densities of setose trichomes can reduce herbivory (Baur et al. 1991, Agren and Schemske 1993, Fernandes 1994), I did not specifically test this hypothesis in my study. However, I did show that leaf herbivory resulted in decreased herbivore growth on undamaged leaves of damaged plants, which was correlated with an increase in trichomes. I detected induced resistance on leaves that were present at the time of initial damage and on leaves that were produced following damage. Because leaves present at the time of damage could not induce trichomes, this result indicates that the observed induced resistance was not entirely due to induction of trichomes, and probably was related to phytochemical induction. Induced resistance affected the leaf area consumed, and hence the growth of larvae, but did not have a direct negative impact on gross growth efficiency, an indicator of growth per unit of leaf area consumed (Waldbauer 1968). In other words, induced resistance did not appear to be due to poisonous or antinutritive effects, but rather due to deterrent effects. In choice experiments, leaf-mining flies and noctuid larvae preferred uninduced control plants over induced plants. Therefore, induced responses in wild radish affect a diverse array of herbivores including noctuids, grasshoppers, and leaf-mining flies, as well as aphids, flea beetles, and earwigs (Agrawal 1998b). Although further studies are needed, induced responses to herbivory in wild radish do not appear to affect specialist herbivores, such as larvae of P. rapae.
Induced responses to herbivory may increase plant fitness via a variety of mechanisms (Agrawal and Karban 1999). First, induction may directly reduce preference or performance of herbivores. Induction of defenses against herbivores is also often accompanied by induction of volatiles that attract predaceous or parasitic arthropods to the plant (Pare and Tumlinson 1996, Takabayashi and Dicke 1996, Shimoda et al. 1997, De Moraes et al. 1998). In addition, a reduction in food quality also may cause omnivores to reduce plant feeding and increase predation on herbivores of the plant (Agrawal et al. 1999a). Two benefits of induction that have been less well explored are (1) increased competitive ability of induced plants against uninduced neighboring plants (e.g., Sadras 1997), and (2) increased tolerance to subsequent herbivory of induced plants compared to controls (e.g., Wittmann and Schoenbeck 1996).
Importance of studying plant defense
The interactions between herbivores and plants have long been viewed as antagonistic, with each participant having the ability to affect selection on the other (Dethier 1954, Painter 1958, Fraenkel 1959, Ehrlich and Raven 1964, Fritz and Simms 1992, but see McNaughton et al. 1997). In particular, if herbivores decrease plant fitness and plants possess genetic variation for traits that reduce herbivory and increase plant fitness, then herbivores may select for more resistant plants. Evolutionary ecologists have recognized this, and there have been many recent studies documenting the fitness consequences of herbivory for plants (reviewed by Marquis 1992, Bigger and Marvier 1998). However, fewer studies have shown that variation in plant traits that affect herbivores also have fitness consequences for the plant. Recent quantitative genetic studies with short-lived plants provide the best evidence that plant resistance traits have important consequences for plant fitness (Berenbaum et al. 1986, Simms and Rausher 1987, 1989, Mauricio and Rausher 1997). For example, Simms and Rausher (1989) found that morning glory plants exhibited genetic variation for resistance to corn earworms, which decreased plant fitness. These herbivores were found to impose natural selection for increased resistance. Thus, although rarely demonstrated empirically, resistance traits can have positive effects on plant fitness by reducing herbivory.
Induced resistance in plants has been largely studied from the herbivore's perspective. Plant resistance mechanisms may benefit plants through a variety of mechanisms, and demonstrating effects on plant fitness is key to documenting that they truly serve a defensive function. For wild radish, induced responses to herbivory are correlated with a net fitness benefit in environments with herbivory. This effect on fitness has two important implications. First, inducible resistance appears to be an example of adaptive plasticity in plants. Adaptive plasticity is defined as the higher relative fitness individuals have when they express one phenotype rather than another in a particular environment (Bradshaw 1965, Thompson 1991, Gotthard and Nylin 1995, Fox et al. 1997). For induced resistance in wild radish, not only is the induced phenotype associated with higher relative fitness in environments with herbivory (Agrawal 1998b), but the induced phenotype is associated with reduced fitness in environments lacking herbivory (i.e., fitness cost of expressing the "wrong" phenotype) (Agrawal et al. 1999b). These results demonstrating benefits and costs of induction are important because phenotypic plasticity is thought to evolve as a mechanism for organisms to express adaptive phenotypes in variable environments. Few studies have documented adaptive plasticity (reviewed in Dudley and Schmitt 1996).
The second implication of this study concerns the evolution of inducible defenses against herbivory, and of phenotypically plastic traits in general. In order for such traits to evolve by natural selection, there must be heritable variation that affects fitness. I did not attempt to detect genetic variation for induction. In a previous study, however, I detected a significant family by treatment interaction for effects of induction in wild radish on resistance to aphids ([F.sub.24,852] = 1.555, P = 0.044, data from Agrawal 1998b), indicating that there may be genetic variation for induction. Similarly, a significant family by induction interaction was detected for seed production of wild radish plants, again indicating that there may be genetic variation for induction or its effects on fitness (Agrawal et al. 1999b). Few other studies have detected genetic variation for induction. Studies by Zangerl and Berenbaum (1990) and van Dam and Vreiling (1994) detected genetic variation for chemical induction, although these studies did not characterize variation in induced resistance to herbivores. No study to date has asked whether genetic variation for induction is associated with differential fitness. If, however, the observed phenotypic correlations between induction and fitness (Agrawal 1998b, Agrawal et al. 1999b; A. A. Agrawal, unpublished manuscript) are reasonable estimates of their genetic counterparts, as is the case in many situations (Cheverud 1988), then induction may have evolved as a defense strategy because of benefits in the presence of herbivores and cost savings in the absence of herbivores. Given that there is likely to be genetic variation for induced resistance, future studies that can detect such variation and determine its fitness consequences will be important for directly measuring natural selection on inducible plant defenses.
The current study provides evidence for fitness benefits of induction in environments with herbivory, demonstrating that induced responses can truly serve a defensive function. Inducible defenses are an example of adaptive plasticity because the induced phenotype realizes the higher fitness in an environment with strong herbivory, whereas the uninduced phenotype realizes the higher fitness in an environment with low herbivory.
Rick Karban provided excellent advice and support throughout. I thank Chris Wardlaw and Jennifer Thaler for help with field work. Many thanks also to Bob Heald and the staff of Blodgett Forest Research Station for logistical support. Mark Salser kindly identified the grasshopper species and Mike Stout provided the jasmonic acid. Jay Rosenheim and his laboratory offered technical assistance with the leaf-area meter. The manuscript was improved by thoughtful reviews by Rick Karban, Jennifer Thaier, Jay Rosenheim, Lynn Adler, Sharon Lawler, Carrie Black, Pekka Kaitaniemi, Greg English-Loeb, and Jim Coleman. This study was financially supported by the Center for Population Biology (University of California, Davis), Jastro Shields Awards Program (University of California, Davis), Northern California Chapter of Phi Beta Kappa, and NSF Dissertation Improvement grant DEB-9701109.
Agrawal, A. A. 1998a. Leaf damage and associated cues induced aggressive ant recruitment in a neotropical ant-plant. Ecology 79:2100-2112.
-----. 1998b. Induced responses to herbivory and increased plant performance. Science 279:1201-1202.
Agrawal, A. A., and R. Karban. 1999. Why induced defenses may be favored over constitutive strategies in plants. Pages 45-61 in R. Tollrian and C. D. Harvell, editors. The ecology and evolution of inducible defenses. Princeton University Press, Princeton, New Jersey, USA.
Agrawal, A. A., C. Kobayashi, and J. S. Thaler. 1999a. Influence of prey availability and induced host-plant resistance on omnivory by western flower thrips. Ecology 80: 518-523.
Agrawal, A. A., and M. T. Rutter. 1998. Dynamic anti-herbivore defense in ant-plants: the role of induced responses. Oikos 83:227-236.
Agrawal, A. A., S. Y. Strauss, and M. J. Stout. 1999b. Effects of induced responses and tolerance to herbivory on male and female fitness components of annual wild radish plants. Evolution, in press.
Agren, J., and D. W. Schemske. 1993. The cost of defense against herbivores: an experimental study of trichome production in Brassica rapa. American Naturalist 141:338-350.
Alborn, H. T, T C. J. Turlings, T. H. Jones, G. Stenhagen, J. H. Loughrin, and J. H. Tumlinson. 1997. An elicitor of plant volatiles from beet armyworm oral secretion. Science 276:945-949.
Albrecht, T A. K., K. Stahl, H. D. Knofel, G. Sembdner, and E. W. Weiler. 1993. Quantification of rapid, transient increases in jasmonic acid in wounded plants using a monoclonal antibody. Planta 191:86-94.
Avdiushko, S. A., G. C. Brown, D. L. Dahlman, and D. F. Hildebrand. 1997. Methyl jasmonate exposure induces insect resistance in cabbage and tobacco. Environmental Entomology 26:642-654.
Baldwin, I. T. 1988. Short-term damage-induced increases in tobacco alkaloids protect plants. Oecologia (Berlin) 75: 367-370.
-----. 1990. Herbivory simulations in ecological research. Trends in Ecology and Evolution 5:91-93.
-----. 1991. Damage-induced alkaloids in wild tobacco. Pages 47-69 in D. W. Tallamy and M. J. Raupp, editors. Phytochemical induction by herbivores. Wiley, New York, New York, USA.
-----. 1994. Chemical changes rapidly induced by folivory. Pages 1-23 in E. Bernays, editor. Insect plant interactions. Volume V. Chemical Rubber Company, Boca Raton, Florida, USA.
-----. 1996. Methyl jasmonate-induced nicotine production in Nicotiana attenuata: Inducing defenses in the field without wounding. Entomologia Experimentalis et Applicata 80:213-220.
-----. 1998. Jasmonate-induced responses are costly but benefit plants under attack in native populations. Proceedings of the National Academy of Sciences (USA)95:8113-8118.
Baldwin, I. T, C. L. Sims, and S. E. Kean. 1990. The reproductive consequences associated with inducible alkaloidal responses in wild tobacco. Ecology 71:252-262.
Baur, R., S. Binder, and G. Benz. 1991. Nonglandular leaf trichomes as short-term inducible defense of the gray alder, Alnus incana (L.), against the chrysomelid beetle, Agelastica alni L. Oecologia (Berlin) 87:219-226.
Berenbaum, M. R., A. R. Zangerl, and J. K. Nitao. 1986. Constraints on chemical coevolution: wild parsnips and the parsnip webworm. Evolution 40:1215-1228.
Bergey, D. R., G. A. Howe, and C. A. Ryan. 1996. Polypeptide signaling for plant defensive genes exhibits analogies to defense signaling in animals. Proceedings of the National Academy of Sciences (USA) 93:12053-12058.
Bigger, D. S., and M. A. Marvier. 1998. How different would a world without herbivory be? Integrative Biology 1:60-67.
Bodnaryk, R. P. 1992. Effects of wounding on glucosinolates in the cotyledons of oilseed rape and mustard. Phytochemistry 31:2671-2677.
Bodnaryk, R. P., and R. T Rymerson. 1994. Effect of wounding and jasmonates on the physico-chemical properties and flea beetle defence responses of canola seedlings, Brassica napus L. Canadian Journal of Plant Science 74:899-907.
Bradshaw, A. D. 1965. The evolutionary significance of phenotypic plasticity in plants. Advances in Genetics 13:115-155.
Brown, D. G. 1988. The cost of plant defense: an experimental analysis with inducible proteinase inhibitors in tomato. Oecologia 76:467-470.
Cheverud, J. M. 1988. A comparison of genetic and phenotypic correlations. Evolution 42:958-968.
De Moraes, C. M., W. J. Lewis, P. W. Pare, H. T Alborn, and J. H. Tumlinson. 1998. Herbivore-infested plants selectively attract parasitoids. Nature 393:570-573.
Dethier, V. G. 1954. Evolution of feeding preferences in phytophagous insects. Evolution 8:33-54.
Doughty, K. J., G. A. Kiddle, B. J. Pye, R. M. Wallsgrove, and J. A. Pickett. 1995. Selective induction of glucosinolates in oilseed rape leaves by methyl jasmonate. Phytochemistry (Oxford) 38:347-350.
Dudley, S. A., and J. Schmitt. 1996. Testing the adaptive plasticity hypothesis: Density-dependent selection on manipulated stem length in Impatiens capensis. American Naturalist 147:445-465.
Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants: a study in coevolution. Evolution 18:586-608.
Faeth, S. H. 1992. Do defoliation and subsequent phytochemical responses reduce future herbivory on oak trees? Journal of Chemical Ecology 18:915-925.
Fernandes, G. W. 1994. Plant mechanical defenses against insect herbivory. Revista Brasileira de Entomologia 38: 421-433.
Fox, C. W., M. S. Thakar, and T. A. Mousseau. 1997. Egg size plasticity in a seed beetle: an adaptive maternal effect. American Naturalist 149:149-163.
Fraenkel, G. S. 1959. The raison d'etre of secondary plant substances. Science 129:1466-1470.
Fritz, R.S., and E. L. Simms. 1992. Plant resistance to herbivores and pathogens. University of Chicago Press, Chicago, Illinois, USA.
Gotthard, K., and S. Nylin. 1995. Adaptive plasticity and plasticity as an adaptation: a selective review of plasticity in animal morphology and life history. Oikos 74:3-17.
Gowda, J. H. 1997. Physical and chemical response of juvenile Acacia tortilis trees to browsing: experimental evidence. Functional Ecology 11:106-111.
Green, T R., and C. A. Ryan. 1972. Wound induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175:776-777.
Hartley, S. E., and C. G. Jones. 1997. Plant chemistry and herbivory, or why the world is green. Pages 284-324 in M. J. Crawley, editor. Plant ecology. Second edition. Blackwell, Oxford, UK.
Hartley, S. E., and J. H. Lawton. 1987. Effects of different types of damage on the chemistry of birch foliage and the responses of birchfeeding insects. Oecologia (Berlin) 74: 432-437.
Harvell, C. D. 1990. The ecology and evolution of inducible defenses. Quarterly Review of Biology 65:323-340.
Haukioja, E., and S. Neuvonen. 1985. Induced long-term resistance of birch foliage against defoliators: defensive or incidental? Ecology 66:1303-1308.
Jaramillo, V. J., and J. K. Derling. 1988. Grazing history, defoliation, and competition: effects on shortgrass production and nitrogen accumulation. Ecology 69:1599-1608.
Karban, R. 1986. Induced resistance against spider mites in cotton: field verification. Entomologia Experimentalis et Applicata 42:239-242.
-----. 1993. Costs and benefits of induced resistance and plant density for a native shrub, Gossypium thurberi. Ecology 74:9-19.
Karban, R., and I. T Baldwin. 1997. Induced responses to herbivory. University of Chicago Press, Chicago, Illinois, USA.
Karban, R., and J. H. Myers. 1989. Induced plant responses to herbivory. Annual Review of Ecology and Systematics 20:331-348.
Kudo, G. 1996. Herbivory pattern and induced responses to simulated herbivory in Quercus mongolica var. grosseserrata. Ecological Research 11:282-289.
Marquis, R. J. 1992. The selective impact of herbivores. Pages 301-325 in R. S. Fritz and E. L. Simms. Plant resistance to herbivores and pathogens: ecology, evolution, and genetics. University of Chicago Press, Chicago, Illinois, USA.
Mattiacci, L., M. Dicke, and M. A. Posthumus. 1995. Beta-Glucosidase: an elicitor of herbivore-induced plant odor that attracts host-searching parasitic wasps. Proceedings of the National Academy of Sciences (USA) 92:2036-2040.
Mattson, W. J., and S. R. Palmer. 1988. Changes in levels of foliar minerals and phenolics in trembling aspen Populus tremuloides in response to artificial defoliation. Pages 157170 in W. J. Mattson, J. Levieux, and C. Bernard-Dagan, editors. Mechanisms of woody plant defenses against insects: search for pattern. Springer-Verlag, Berlin, Germany.
Mauricio, R., M.D. Bowers, and F. A. Bazzaz. 1993. Pattern of leaf damage affects fitness of the annual plant Raphanus sativus (Brassicaceae). Ecology 74:2066-2071.
Mauricio, R., and M. D. Rausher. 1997. Experimental manipulation of putative selective agents provides evidence for the role of natural enemies in the evolution of plant defense. Evolution 51:1435-1444.
McCloud, E. S., and I. T Baldwin. 1997. Herbivory and caterpillar regurgitants amplify the wound-induced increases in jasmonic acid but not nicotine in Nicotiana sylvestris. Planta 203:430-435.
McCollum, S. A., and J. Van Buskirk. 1996. Costs and benefits of a predator-induced polyphenism in the gray treefrog Hyla chrysoscelis. Evolution 50:583-593.
McNaughton, S. J., F. F. Banyikwa, and M. M. McNaughton. 1997. Promotion of the cycling of diet-enhancing nutrients by African grazers. Science 278:1798-1800.
McNaughton, S. J., and J. L. Tarrants. 1983. Grass leaf silification: natural selection for an inducible defense against herbivores. Proceedings of the National Academy of Sciences (USA) 80:790-791.
Milewski, A. V., T. P. Young, and D. Madden. 1991. Thorns as induced defenses: experimental evidence. Oecologia (Berlin) 86:70-75.
Myers, J. H., and D. Bazely. 1991. Thorns, spines, prickles, and hairs: Are they stimulated by herbivory and do they deter herbivores. Pages 325-344 in D. W. Tallamy and M. J. Raupp, editors. Phytochemical induction by herbivores. Wiley, New York, New York, USA.
Obeso, J. R. 1997. The induction of spinescence in European holly leaves by browsing ungulates. Plant Ecology 129: 149-156.
Paige, K. N. 1992. Overcompensation in response to mammalian herbivory: from mutualistic to antagonistic interactions. Ecology 73:2076-2085.
Painter, R. H. 1958. Resistance of plants to insects. Annual Review of Entomology 3:267-290.
Pare, P. W., and J. H. Tumlinson. 1996. Plant volatile signals in response to herbivore feeding. Florida Entomologist 79: 93-103.
Passera, L., E. Roncin, B. Kaufmann, and L. Keller. 1996. Increased soldier production in ant colonies exposed to intraspecific competition. Nature 379:630-631.
Reimer, O., and M. Tedengren. 1996. Phenotypical improvement of morphological defences in the mussel Mytilus edulis induced by exposure to the predator Asterias rubens. Oikos 75:383-390.
Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225.
Rosenthal, J. P., and P. M. Kotanen. 1994. Terrestrial plant tolerance to herbivory. Trends in Ecology and Evolution 9:145-148.
Sadras, V. O. 1997. Interference among cotton neighbours after differential reproductive damage. Oecologia (Berlin) 109:427-432.
Shimoda, T., J. Takabayashi, W. Ashihara, and A. Takafuji. 1997. Response of predatory insect Scolothrips takahashii toward herbivore-induced plant volatiles under laboratory and field conditions. Journal of Chemical Ecology 23: 2033-2048.
Simms, E. L., and M.D. Rausher. 1987. Costs and benefits of plant resistance to herbivory. American Naturalist 130: 570-581.
Simms, E. L., and M. D. Rausher. 1989. The evolution of resistance to herbivory in Ipomoea purpurea. II. Natural selection by insects and costs of resistance. Evolution 43: 573-585.
Slansky, F., and P. P. Feeny. 1977. Stabilization of the rate of nitrogen accumulation of larvae of the cabbage butterfly on wild and cultivated food plants. Ecological Monographs 47:209-228.
Smith, L. L., J. Lanza, and G. C. Smith. 1990. Amino acid concentrations in extraforal nectar of Impatiens sultani increase after simulated herbivory. Ecology 71:107-115.
Strauss, S. Y., and A. A. Agrawal. 1999. Ecology and evolution of plant tolerance herbivory. Trends in Ecology and Evolution 14, in press.
Takabayashi, J., and M. Dicke. 1996. Plant-carnivore mutualism through herbivore-induced carnivore attractants. Trends in Plant Science 1:109-113.
Thaler, J. S., M. J. Stout, R. Karban, and S. S. Duffey. 1996. Exogenous jasmonates simulate insect wounding in tomato plants (Lycopersicon esculentum) in the laboratory and field. Journal of Chemical Ecology 22:1767-1781.
Thompson, J. D. 1991. Phenotypic plasticity as a component of evolutionary change. Trends in Ecology and Evolution 6:246-249.
Trumble, J. T., D. M. Kolodny-Hirsch, and I. P. Ting. 1993. Plant compensation for arthropod herbivory. Annual Review of Entomology 38:93-119.
Van Dam, N. M., and K. Vrieling. 1994. Genetic variation in constitutive and inducible pyrrolizidine alkaloid levels in Cynoglossum officinale 1. Oecologia (Berlin) 99:374-378.
Waldbauer, G. P. 1968. The consumption and utilization of food by insects. Advances in Insect Physiology 5:229-289.
Wittmann, J., and F. Schoenbeck. 1996. Studies of tolerance induction in wheat infested with powdery mildew or aphids. Zeitschrift fur Pflanzenkrankheiten und Pflanzenschutz (Journal of Plant Disease and Protection) 103:300-309.
Wold, E. N., and R. J. Marquis. 1997. Induced defense in white oak: effects on herbivores and consequences for the plant. Ecology 78:1356-1369.
Young, T P. 1987. Increased thorn length in Acacia depranolobium - an induced response to browsing. Oecologia 71: 436-438.
Young, T. P., and B. D. Okello. 1998. Relaxation of an induced defense after exclusion of herbivores. Oecologia 115:508-513.
Zangerl, A. R., and M. R. Berenbaum. 1990. Furanocoumarin induction in wild parsnip: genetics and populational variation. Ecology 71:1933-1940.
Zar, J. H. 1996. Biostatistical analysis. Third edition. Prentice Hall, Upper Saddle River, New Jersey, USA.
|Printer friendly Cite/link Email Feedback|
|Author:||Agrawal, Anurag A.|
|Date:||Jul 1, 1999|
|Previous Article:||Nectar-robbing bumble bees reduce the fitness of Ipomopsis Aggregata (Polemoniaceae).|
|Next Article:||Why does early-season herbivory affect subsequent budburst?|