- Top of page
- Materials and Methods
Methyl jasmonate (MeJA) is a volatile signaling compound in plants that, in addition to numerous developmental roles, mediates many important ecological interactions. Accumulation of MeJA after herbivore or pathogen attack is involved in the induction of direct defenses (Farmer et al., 1992) and in the production of volatiles that attract parasitoids and predatory arthropods to plants (Kessler & Baldwin, 2001). In addition, MeJA has been proposed as an airborne signal mediating interplant communication (Farmer et al., 1992; Karban et al., 2000). MeJA is formed when the enzyme S-adenosyl-l-methionine : jasmonic acid carboxyl methyltransferase (JMT) specifically methylates jasmonic acid (JA) (Seo et al., 2001; Schaller et al., 2005). In Arabidopsis thaliana this enzyme is inducible by wounding or exogenous MeJA treatment, which leads to accumulation of MeJA and activation of MeJA-responsive genes. Transgenic A. thaliana plants overexpressing JMT constitutively produce threefold higher MeJA and twofold higher total jasmonate levels in their leaves than wild-type plants, with no differences in JA levels (Seo et al., 2001). In turn, JMT plants exhibit constitutively elevated MeJA-responsive genes in their leaves, including defensins, various PR proteins and oxidative stress-related genes, and are more resistant than wild-type plants to attack by the fungus Botrytis cinerea (Seo et al., 2001) and the bacterium Pseudomonas syringae pv. tomato (Jung et al., 2003). Fitness benefits to the plant of the overproduction of MeJA may be inferred from such results, although they have not been quantified. Fitness costs of the overproduction of MeJA have never been examined in A. thaliana or in any other plant.
Costs of responses induced by MeJA or other signals in plants can be realized in several ways. Direct physiological costs of induced responses are seen independently of the environment, and presumably result from internal allocation trade-offs between growth and defense production, and other pleiotropic effects (Herms & Mattson, 1992; Heil & Baldwin, 2002; Cipollini et al., 2003). Ecological costs of induced responses are seen only under particular ecological conditions (Heil, 2001, 2002; Cipollini et al., 2003), and can include increased susceptibility to alternate attackers (Agrawal et al., 1999; Cipollini et al., 2004); trade-offs between resistance and tolerance to herbivory (Strauss & Agrawal, 1999); and trade-offs between resistance and competitive ability (van Dam & Baldwin, 1998; Dietrich et al., 2005). Although there is growing experimental support for physiological and ecological costs of induced responses (Heil, 2001, 2002; Heil & Baldwin, 2002; Cipollini et al., 2003), conditions under which they are expressed and their influence on the ecology and evolution of plant defenses are still widely debated (Cipollini, 2002; Zangerl, 2003; Koricheva et al., 2004; Dietrich et al., 2005).
Since the identification of important defense hormones such as JA/MeJA, many ecologists have used purified forms of these hormones to manipulate plant defenses in order to address costs of induced responses (Baldwin, 1998; Agrawal et al., 1999; Cipollini, 2002). While these studies have greatly advanced our understanding of the effects of these hormones, pharmacological studies have been criticized for use of sometimes unnatural levels or placement of hormones; use of inappropriate isomers; or producing unwanted side-effects that may be partly responsible for some of the observed effects (Purrington, 2000; Cipollini et al., 2003; Jung et al., 2003). The use of MeJA-overproducing JMT plants provides an opportunity to examine the costs and benefits of the constitutive production of elevated levels of this important hormone in the absence of at least some of these confounding factors. In particular, JMT plants have been engineered to manufacture and accumulate high levels of their own MeJA from their own endogenous JA, circumventing problems associated with uptake of exogenous sprays, and ensuring the presence of appropriate isomers of MeJA. Although JMT plants exhibit high levels of MeJA as a result of genetic manipulation, some highly defended plants occurring in natural plant populations may be the result of constitutive activation of normally inducible defenses. The JMT plant serves as an appropriate model of the fitness costs and benefits incurred by such plants. In this study, I explored the physiological and ecological consequences of the overexpression of MeJA on seed production, tolerance to defoliation and competitive effect and response, by comparing the responses of JMT plants to vector controls and their wild-type Columbia parent ecotype in a series of experiments.
- Top of page
- Materials and Methods
Costs of the overproduction of MeJA and associated plant responses in A. thaliana have been examined here for the first time. Production of threefold higher levels of MeJA in transgenic A. thaliana plants overexpressing JMT, and constitutive upregulation of MeJA-responsive genes (apparently including those encoding trypsin inhibitors), was associated with a significant decline in total seed production and reduced seed germination compared with vector controls or wild-type plants when grown alone in pots. Some portion of the costs observed in JMT plants may be caused by the random insertion of JMT construct(s) in the A. thaliana genome or expression of the kanamycin-resistance gene (Strauss et al., 2002; Jackson et al., 2004), but at least some of these effects were controlled by comparing fitness in replicated JMT lines with several replicated vector control lines. Indeed, although variation in total seed production was evident among replicate lines, JMT lines universally produced lower total seed mass than vector control lines containing the kanamycin-resistance gene. In two other carefully controlled studies of costs of resistance using genetically engineered plants, fitness of several vector-only control lines (the same ones as used here) did not differ significantly from untransformed wild-type plants in a study of herbicide resistance in A. thaliana (Purrington & Bergelson, 1997), and in a study of trypsin inhibitor expression in Nicotiana attenuata (see supplementary material in Zavala et al. (2004)). This suggests that side-effects of the transformation process or expression of the kanamycin-resistance gene probably contribute in only minor ways to the variation in fitness among JMT and wild-type plants seen in this study. Reduced total seed mass in JMT plants could also have extended from a direct reduction in pollen fertility or embryo development caused by elevated MeJA levels. Both a lack of JA production in the fad3-1 fad7-2 fad8 mutant (McConn & Browse, 1996) and the lack of JA sensitivity in the coi1 mutant (Xie et al., 1998) lead to male sterility. However, JMT plants contain normal levels of JA and a functional COI1 gene, and there is as yet no evidence that elevated MeJA or altered MeJA : JA ratios can directly affect male fertility.
The substantial reduction in seed production in JMT plants seen here is better explained by resource allocation trade-offs between the production of high levels of MeJA and alterations in associated defense responses and growth processes (Herms & Mattson, 1992), or other pleiotropic effects of elevated MeJA levels. In A. thaliana, MeJA treatment upregulates numerous defense genes and some growth and maintenance-related genes, but downregulates important photosynthetic genes such as RuBisCO (Schenk et al., 2000). A similar pattern of response to MeJA occurs in N. attenuata (Hermsmeier et al., 2001), which is particularly costly when plants are growing with neighbors (van Dam & Baldwin, 1998; Baldwin & Hamilton, 2000). JMT plants constitutively express a variety of defense-related genes and exhibit downregulated expression of growth-related genes, such as the chlorophyll a/b-binding protein (Jung et al., 2003), which may combine to constrain fitness of these plants. Physiological costs of responses induced by MeJA in A. thaliana have not been reported previously, but Cipollini (2002) showed that JA treatment of A. thaliana three times during the growing season led to an 18% decline in total seed production relative to uninduced plants. Dietrich et al. (2005) showed that A. thaliana has the capacity to compensate for growth depressions caused by a single elicitation of defenses with the salicylate mimic, BION, given the time and soil resources. As expected, the constitutive expression of MeJA and associated responses studied here was much more costly than the upregulation and subsequent relaxation of responses associated with repeated JA application. This finding supports the view that inducibility of defenses is a cost-saving mechanism over constitutive expression.
Although overproduction of MeJA was directly costly to seed production and seed germination, additional ecological costs to tolerance of the removal of a particularly sensitive leaf class were not observed in this study. A lack of a negative genetic correlation between tolerance and resistance to both defoliation (Mauricio et al., 1997) and apical meristem removal (Weinig et al., 2003) has been reported for A. thaliana, but I am aware of no studies that have examined whether induction of jasmonate-mediated defenses is costly to plant tolerance. Although studies have shown negative relationships between the expression of single classes of compound or biological resistance and tolerance (Stowe, 1998), potential benefits of the MeJA-induced upregulation of genes putatively associated with plant tolerance (Strauss & Agrawal, 1999; Schenk et al., 2000) may have counterbalanced any additional costs predicted to accrue in defoliated JMT plants. Identification of genes specifically associated with tolerance would help guide investigations of this mechanism.
Several aspects of plant response to, and effect on, competitors were altered by overproduction of MeJA. JMT plants were generally less negatively affected by a competitor in the pot than wild-type plants, regardless of the identity of the neighbor and its induction status. This relates, to some extent, to the greater fitness that wild-type plants attained in the absence of competition, making reductions in the presence of competition more apparent in wild-type than in lower-yielding JMT plants. The size of the jasmonate-overproducing cev1 mutant of A. thaliana is reduced less by MeJA treatment than wild-type plants, partly because cev1 plants have a much lower growth potential than wild-type plants (Ellis & Turner, 2001). Likewise, costs of responses induced by MeJA in N. attenuata were more apparent in well fertilized, high-yielding plants than in nutrient-deprived, low-yielding plants (van Dam & Baldwin, 1998). MeJA has also been suggested to be allelopathic to growth or germination of competing species (Preston et al., 2002), which may contribute to the ‘resistance’ of JMT plants to a competitor in the pot. If JMT plants emitted greater amounts of MeJA into the shared airspace or soil solution than wild-type plants, then JMT plants may have had a greater allelopathic effect on neighboring wild-type plants than wild-type plants had on JMT, which may have complemented effects mediated by resource competition. Reciprocal allelopathic effects, combined with resource competition, could also explain why fitness of JMT plants did decline in pots where both the target and its wild-type neighbor were induced with JA.
Opportunity costs and benefits of induction were also different in JMT than in wild-type plants. Wild-type plants incurred an opportunity cost of induction with JA when grown with an uninduced neighbor, a pattern also shown by Dietrich et al. (2005) in their study of A. thaliana plants treated with BION. In contrast, JMT plants did not exhibit an opportunity cost of induction. This could relate to the differential degree of response that JA induces in JMT plants (which already express jasmonate-mediated defenses at high levels) relative to ‘naive’ wild-type plants that are more highly inducible. The jasmonate-lacking fad3-2 fad7-2 fad8 mutant of A. thaliana is much more inducible by JA than wild-type A. thaliana (Cipollini et al., 2004), suggesting that the strength of the response to exogenous jasmonate treatment depends on endogenous jasmonate levels. Here I have shown that JMT plants expressed higher constitutive activity of trypsin inhibitor than wild-type plants, as expected for a jasmonate-mediated defense (Cipollini et al., 2004). But trypsin inhibitor activities were increased by JA treatment much more in wild-type than in JMT plants. In addition to dampened opportunity costs of induction, JMT plants did not exhibit an opportunity benefit in response to induction of their neighbors as the wild-type did, suggesting that JMT plants were less able than wild-type plants to take advantage of a resource opportunity created by induction of their neighbor. Costs of defense production to growth mechanisms responsible for rapid plastic responses to resource opportunities probably explained this pattern. The patterns observed in wild-type A. thaliana here almost exactly match the patterns observed in wild-type N. attenuata (van Dam & Baldwin, 1998), suggesting that opportunity costs and benefits of induction in the presence of neighboring plants are a general phenomenon in plants, a pattern from which JMT plants deviate.
Together, these results suggest that overproduction of MeJA is directly costly to total seed production and seed germination rates, and constrains plasticity in response to alterations in the competitive status of neighboring plants. However, tolerance of defoliation was unaffected by overproduction of MeJA, and in some scenarios JMT plants were more resistant than wild-type plants to competitors. As suggested by Cipollini (2004), JMT plants appear to represent a phenotype that can defend against both herbivores and competitors, but at a cost to phenotypic plasticity. In natural plant populations, overexpression of MeJA-mediated responses should be beneficial to resistance to herbivores, pathogens and competitors, but is directly costly to fitness and probably constrains plasticity in response to changing environmental conditions, including resource opportunities created by manipulation of neighboring plants. Such ecologically dependent and independent costs and benefits of expression of MeJA and associated resistance mechanisms may help maintain polymorphisms in resistance in natural plant populations.