Genetic variation in plant chemistry : consequences for plant-insect interactions

Plants form the basis of many food webs and are consumed by a wide variety of organisms, including herbivorous insects. Over the course of evolution, plants have evolved mechanisms to defend themselves against herbivory, whereas herbivorous insects have evolved counter-mechanisms to overcome these defences (a.k.a. co-evolutionary arms races). Plant-insect interactions are not restricted to plants and their herbivores (bi-trophic interactions), but also involve natural enemies of the herbivores such as parasitoids and predators (tri-trophic interactions). Plant quality can affect the quality of the host or prey for parasitoids and predators, respectively. In addition, other plant traits are important in providing shelter, alternative food sources, or chemical cues that can be used for host/prey location. Moreover, as plants reside in both soil and air, they mediate interactions between organisms above- and belowground through changes in plant quality. Plant quality is determined by secondary metabolites and morphological traits that may negatively affect the performance of insects, as well as by primary metabolites that plants produce in order to grow, develop and reproduce, which also provide essential nutrients for insects. Natural plant populations often exhibit genetic variation in various plant traits that include, amongst others, primary and secondary chemistry. Genetic variation in plant defence traits, such as the production of secondary metabolites, can be under selection pressure from a suite of biotic and abiotic factors that vary in space and time. Herbivorous insects may encounter a wide range of plant metabolites because the total concentrations of primary and secondary metabolites and the concentrations of individual compounds vary between genetically different plants. Also as a consequence of genetic variation, plants can respond differently to herbivory in terms of induced defence chemistry and re-allocation of metabolites. The main aim of this thesis was to study how genetic variation in plant chemistry affects (multi)trophic interactions between wild cabbage plants and associated insects, both above- and belowground. As a model system I used five naturally occurring populations of wild cabbage (Brassica oleracea) located in the Dorset area in the UK. These populations have been shown to genetically differ in their defence chemistry profiles even though they are located in relatively close proximity to each other. Wild cabbages belong to the Brassicaceae, a plant family that is characterized by the production of glucosinolates, a group of secondary metabolites. Together with the enzyme myrosinase they form the chemical defence system of Brassicaceous plants including wild cabbage. Glucosinolates and myrosinases are stored separately in plant tissues but upon tissue damage they come into contact with each other upon which the glucosinolates are hydrolysed into potentially toxic break down products. The wild cabbage populations used in this thesis differ in their total glucosinolate concentrations as well as in the expression of individual glucosinolates. In chapter 1 I describe plant-insect interactions in a multi-trophic framework, including both the above- and belowground compartments. Genetic variation in plant traits is introduced as the main topic of this thesis, and I present the main aim and outline of my work. In chapter 2 I discuss how aboveground-belowground interactions influence the evolution and maintenance of genetic variation in plant defence chemistry. I review literature on AG-BG interactions as selection pressures for genetic variation, discuss hypotheses about plant mediation of AG-BG interactions, identify gaps in our knowledge such as the influence of spatial-temporal variation in AG-BG interactions, and in the end present new data on genetic variation in secondary chemistry of wild cabbage and related species. The co-evolutionary arms race between plants and insects has resulted in adaptations in herbivores to cope with plant defence traits. Some insect herbivore species concentrate or sequester secondary metabolites from their food plant and use them in defence against their own enemies. In chapter 3 I studied whether sequestration of glucosinolates by a specialist herbivore is an effective defence mechanism against a generalist predatory bug. I used the sequestering herbivore Athalia rosae as one prey species, and the non-sequestering herbivore Pieris rapae as the control prey species. I compared the performance of the predatory stink bug Podisus maculiventris on these two prey species. As an extra factor, the two prey species were each reared on three different wild cabbage populations to test if plant population would have an effect on the predator through the sequestering herbivore. I found no consistent effect of plant population on the performance of the predator, and prey species only marginally affected its performance. Based on the results I suggest that in some trophic interactions sequestration is not an effective defence mechanism but merely an alternative way of harmlessly dealing with plant secondary metabolites. In addition to aboveground plant-insect interactions, belowground interactions were considered as well. To test whether the performance of the belowground specialist herbivore Delia radicum, of which the larvae feed on root tissues, was influenced by population-related variation in defence chemistry, I reared this species on the five wild cabbage populations (chapter 4). Chemical analyses of root tissues revealed that there were differences amongst the populations in plant primary (amino acids and sugars) and secondary (glucosinolates) chemistry, but this did not affect the performance of the root herbivore, suggesting that D. radicum is well adapted to a wide range of total concentrations and concentrations of individual metabolites. Whereas in chapters 3 and 4 I only focused on one compartment (aboveground and belowground respectively), in chapter 5 I included both compartments in one experiment. I studied the effect of belowground herbivory by larvae of the root fly D. radicum on the performance of an aboveground multi-trophic food chain, and whether this effect differed among three wild cabbage populations. I found that belowground herbivory differentially affected the performances of a specialist aboveground herbivore, the diamondback moth Plutella xylostella, and its parasitoid, Cotesia vestalis, with the parasitoid being more affected than the herbivore. Their performance also differed between the wild cabbage populations, often in interaction with the presence/absence of the belowground herbivore. For both the above- and belowground herbivore I found correlations between performance and plant chemistry, which differed between the insect species and also between males and females. In chapter 6 I discuss the results of my experiments in relation to other studies. I finish with a general conclusion about my work and provide some ideas for future studies that could contribute to our knowledge in the field of (multi)trophic above-belowground interactions with regard to genetic variation in plant chemistry. In my thesis I show that genetic variation in plant chemistry can affect the outcome of above-belowground plant-insect interactions. Herbivores and higher trophic levels were differently affected by the wild cabbage populations, and this difference was also influenced by the location of herbivory (i.e. aboveground or belowground). In both chapter 4 and chapter 5 I found no strong, unidirectional links between plant chemistry and insect performance, suggesting that other metabolites may have played a role in the observed differential effects of the wild cabbage populations. I also show that sequestration of plant allelochemicals in some herbivores is an alternative way of harmlessly dealing with plant secondary metabolites instead of an effective defence mechanism against predators (chapter 3).