Constraints on plant resistance and tolerance trade-offs

Plant tissue loss due to herbivory is an important selective force for shaping plant phenotypes (Marquis 1992). Therefore, plants have evolved a variety of mechanisms to mitigate the negative effects of herbivory. Much of what we have learned, to date, on plant responses to their enemies (herbivores and pathogens) have focused on the evolution of chemical and structural traits that reduce or prevent tissue damage by herbivores (resistance) (Simms and Fritz 1990). However, herbivores may also select for traits that allow plants to compensate for tissue loss with little or no detriment in fitness (tolerance) (Painter 1951). Studies have even shown that herbivory leads to, under certain environmental circumstances (Hawkes and Sullivan 2001), an increase in plant reproductive success (overcompensation, i.e., increased fruit and seed production), rather than a decrease. Plant chemical defense and tolerance to damage is often studied in the context of ecological trade-offs and costs of tolerance (Strauss et al. 2002). It was initially theorized that resistance and tolerance represented two alternative and redundant defense strategies given limited nutrients and energy available in the struggle against hervbivory (van der Meijden et al. 1988). According to this hypothesis, both defensive strategies offer the same fitness benefits (Mauricio et al. 1997). Regardless of the suggestions and assumptions that resistance and tolerance may play redundant roles in plant defense, a recent meta-analysis showed most natural populations appear to be comprised of a mixture of both strategies due to selection for the maintenance of both traits (Leimu and Koricheva 2006). What has been missing from previous studies was the genetic and molecular underpinnings of plant tolerance and its relationship to the well-characterized molecular pathway involved in chemical defense, i.e., the shikimate pathway. Previous studies in the lab have shown that the process of endoreduplication is the primary mechanism by which plants compensate for lost tissue following mammalian herbivory (Scholes et al. 2011). Moreover, the gene G6PD1 was also found to play a significant role in the compensatory process, which is the key regulator for the oxidative pentose phosphate pathway (opp) (Siddappaji et al. 2013). This pathway supplies intermediate carbon skeletons for both nucleotides (consistent with the upregulation of DNA with endoreduplication) and for resistance compounds via the shikimate pathway, thus tying both tolerance and resistance within the same pathway. I utilized Arabidopsis thaliana genotypes displaying a range of compensatory responses from under to overcompensation and measured glucosinolate content for each which uncovered a positive association between the two defenses. Recombinant inbred lines from a cross between the overcompensating ecotype Col-4 and the undercompensating Ler-0 of the annual plant Arabidopsis thaliana were utilized to assess the relationship between tolerance and resistance. Total glucosinolate content for each ecotype following simulated mammalian herbivory was measured. These data show overcompensating ecotypes also have the highest resistance chemistry. Similarly, the direct association between tolerance and resistance was demonstrated by genetically manipulating the endoreduplication pathway. By overexpressing ILP1, a positive regulator of endoreduplication, and thus compensation, glucosinolate production and tolerance was increased in the Col-0 ecotype. This approach allowed us to assess plant resistance-tolerance tradeoffs from a molecular genetic point of view. These results indicate that plant tolerance and resistance pathways are tightly integrated within the oxidative pentose phosphate pathway and may represent a general phenomenon among herbaceous plants given that approximately 90% of herbaceous angiosperms endoreduplicate (Nagl 1976, Sugimoto-Shirasu and Roberts 2003). Although a causal link has been established between endoreduplication, fitness compensation, and chemical defense, no one has addressed whether insect leaf-feeding can elicit the same compensatory response as removing the apical meristem which lowers the level of auxin and triggers entry into the endocycle. In Arabidopsis, wounding has been shown to down-regulate a number of genes that are positively associated with auxin production, suggesting a suppressive effect of insect wound-induced signals, like jasmonic acid, salicylic acid and ethylene, on the auxin signal transduction pathway (Onkokesung et al. 2010). Thus, insect leaf-feeding could trigger endoreduplication by the upregulation of wound-induced signals ostensibly lowering auxin production. Results here support this contention; insect leaf-feeding by Trichoplusia ni elicited a compensatory response similar to that elicited by mammalian herbivores - an ecotype-specific response dependent upon the level of endoreduplication. In addition, the interactive effects of mammalian and insect herbivory on each of these inbred lines was assessed to determine whether interactions were additive (pairwise) or nonadditive (diffuse) on fitness compensation (tolerance) and secondary plant metabolite production (resistance). Results show that some ecotypes had non-additive effects of herbivory such as Ler-0 where following clipping plants suffered increased fitness impacts from T. ni compared to unclipped plants. Other ecotypes such as CS1906 displayed additive effects of insect and mammalian herbivory. Despite a failure to detect tradeoffs between the two defenses both defensive strategies utilize carbon skeletons from a shared resource pool in the oxidative pentose phosphate pathway. Therefore, the costs for maintaining both strategies were assessed experimentally. Specifically the cost of resistance in A. thaliana was assessed by utilizing a double knockout mutant for two cytochrome P450s CYP79B2 and CYP79B3, key enzymes in the biosynthetic process of indole glucosinolates, which converts tryptophan to indole-3-acetaldoxime (IAOx). IAOx is then further utilized in the production of indole glucosinolates (Bender and Celenza 2009). Results show that knocking out indole glucosinolate production and thus resistance leads to an increase in the compensatory response compared to wildtype Columbia-0. This shows that despite a positive association there are still physiological costs for maintaining both strategies. My studies have shown that endoreduplication is an important driver of both compensatory and induced resistance responses following both mammalian and insect herbivory. Moreover, these results showing ecotypic variation in endoreduplication, raises some concerns about drawing conclusions on the impacts of herbivory given that we tend to ignore genotypic variation by collapsing it into an overall mean population response. Through multiple chapters I have shown that when we averaged total seeds and glucosinolates across all plant genotypes, ignoring the genetic variation in endoreduplication among ecotypes of A. thaliana, we no longer observe a positive association between the two defenses. This dissertation indicates it will be important to assess the degree of endoreduplication across a population of herbaceous plants prior to measuring fitness and plant resistance responses to herbivory. Thus, additional plant species that endoreduplicate should be investigated to see if this phenomenon is generalizable