Understanding the impacts of plant secondary metabolism on the health, behaviour and fitness of bee pollinators

Flowering plants rely on their floral secondary metabolism to recruit pollinating insects and pollinator recruitment is critical for plant fitness. However, plants not only interact with mutualistic pollinators but also come in frequent contacts with herbivores and additional biotic and abiotic stresses. Understanding how plants maintain effective pollinator-mediated reproduction while also adaptively responding to environmental variation has fundamental implications for the ecology and evolution of plants and insects. Herbivory stress causes upregulation of leaf defence metabolites, but also flower metabolites, and the costs of these responses for pollination are well-documented but poorly understood, and we lack a general hypothesis for floral trait plasticity. In this thesis I develop a framework for understanding how flowering plants respond to biotic stress based on variation in the mating system, which dictates a plant’s reliance on insect pollinators. I then empirically evaluate the ecology and evolution of stress-induced floral traits using a study system, wild tomato, Solanum habrochaites, which exhibits population level variation in mating system ranging from self-incompatibility (SI) to self-fertilizing (self-compatibility, SC). The first chapter of my thesis presents background of past and current research on plant-insect interactions and effects of herbivory-induced secondary metabolism on floral rewards, with a particular focus on pollen. In chapters two and three, I focus on identifying adaptive variation in floral trait plasticity with a special focus on floral scent and floral reward chemistry (i.e., pollen secondary metabolism) in outcrossing. The third chapter assesses how these induced responses affect plant fitness under semi-natural conditions, and tests for compensatory plant responses within each mating system. My work provides evidence for adaptive phenotypic plasticity of floral odour cues and reward chemistry when experiencing stress from herbivory, suggesting that phytohormone-mediated attraction of pollinators may be a previously unappreciated strategy by which plants cope with herbivore attack. My results indicate that pollen secondary metabolomic profiles and floral scent profiles of herbivory-induced plants are divergent, and mating system specific, with indications of convergence within each mating system. This study is one of the first to demonstrate the drivers for the evolution of phenotypically plastic pollinator attraction, including the mechanisms involved in more or less attractive floral scent profiles. I show that, contrary to most models of scent-based pollinator recruitment, pollinator choice in our system may be based on preference for low-emission flowers, consistent with a hypothesised role of floral VOCs as signals of reduced reward quality. For the first time, I provide evidence to show that each mating system shows divergent adaptations, i.e., selfing plants under relaxed selection from pollinators show striking upregulation of pollen defences while outcrossers have evolved to minimise these responses due to strong selection driven by higher pollinator-dependence. My use of replicate wild populations indicates that a driver of this convergent evolution of plasticity is likely to have been reliance on pollinators. In chapter three, I provide evidence that selfing and outcrossing plants are equipped with compensatory mechanisms to maintain plant fitness under herbivory stress. This study confirms the results of Chapters 2 and 3, by showing induced attraction to pollinators in outcrossing plants and methyl jasmonate mediated plasticity in self-fertilization in SC plants. Overall, my research findings have yielded novel insight into the evolution of phenotypic plasticity of floral scent and reward chemistry, while focusing on the impacts on pollinator behaviour and fitness, as well as plant fitness. I believe my findings lay the groundwork for future studies exploring adaptive plasticity of floral traits under abiotic or biotic stress, and the implications of this plasticity in natural and managed ecosystems.