How plants balance competitive growth and defence : an analysis of virtual plants in dynamic interactions

Abstract

Plants grow in dynamic environments where their fitness is determined by a multitude of biotic and abiotic factors in the environment. These environments house complex webs of interactions in which plants must compete with neighbours for limited resources such as light, while having to defend against a multitude of heterotrophic organisms such as insect herbivores. Plants respond to cues related to neighbour presence and herbivore damage with a suite of induced shade avoidance and defence responses, respectively. One of the signals that interlinks these shade avoidance and defence responses is the red to far-red ratio (R:FR), which is a neighbour detection cue that induces shade avoidance responses and downregulates direct defence responses. This physiological linkage has been hypothesised as a mechanism that balances a resource allocation trade-off by promoting resource acquisition through competitive growth, at the expense of defence. However, the optimal expression of traits related to competitive growth and defence are not only dependent on biotic environmental factors, but also on interactions with the plant and herbivore communities. The aim of this thesis was to elucidate how ecological interactions shape the optimal balance between the acquisition and protection of resources. To reach this objective, I used a functional-structural plant (FSP) modelling approach to simulate physiological mechanisms that interact with the environment, how these mechanisms drive ecological interactions between plants and herbivores, and the evolutionary dynamics that arise from these interactions. In the general introduction (chapter 1), the linkage between competitive growth and defence is introduced from physiological, ecological and evolutionary perspectives, and FSP modelling is introduced as a tool to study the balance between competitive growth and defence. In this thesis, I describe and utilise two FSP models; an FSP model of Brassica nigra that focusses on detailed physiological responses and is parameterised and validated using field and greenhouse experiments, and an FSP model of a generic dicotyledon that focusses on evolutionary dynamics. In chapter 2, I outline my view on using FSP modelling to study interactions between the plant and herbivore communities and accentuate the importance of considering these interactions in a dynamic environment. In this chapter, I propose four alternative hypotheses to the origin of the physiological linkage between shade avoidance and defence responses that go beyond the perspective of resource allocation trade-offs between growth and defence. The downregulation of direct defence by a low R:FR cue may function to (i) focus defence on valuable tissues, (ii) deter specialist herbivores, (iii) control interactions within the insect herbivore community or (iv) prevent the emission of volatile cues to which neighbouring plants may respond. Chapter 3 presents an FSP model of B. nigra that combines mechanistic simulation of plant growth, shade avoidance responses and herbivore feeding, which are parameterised and validated using field and greenhouse experiments. This model is used to analyse the impact of herbivore feeding at different locations within the plant that relate to herbivore specialisation on plant fitness in a dynamic environment where plants compete for light. This analysis shows that herbivore damage had a larger effect on plant fitness when plants were competing for light and when the damage was directed at young leaves at the top of the canopy. Chapter 4 expands on the FSP model presented in chapter 3 with the addition of plant defence, and is used to analyse how the direct costs (e.g. metabolic costs) and benefits of defence as well as indirect costs and benefits of defence imposed through ecological interactions impact plant fitness. This analysis shows that ecological costs of defence imposed by inter-genotypic competition is the strongest determinant of plant fitness, amplifying size differences resulting from differences in defence investment. Additionally, the model shows that the benefit of plant defence relies on reducing herbivore damage rather than dispersing herbivore damage away from young leaves and towards older leaves that are of less value to the plant. Chapter 5 expands on the FSP model presented in chapter 4 with the addition of R:FR mediation of plant defence, to analyse the function of this physiological regulation from the perspective of simple versus competitive optimisation (e.g. mono-stands vs mixtures). The model results show that plant-level defence investment was a strong determinant of plant fitness, and that leaf-level mediation of plant defence by R:FR may provide an additional fitness benefit in high plant densities. Furthermore, the model shows that the optimal plant-level defence expression does not monotonically decrease with plant density. This indicates that R:FR mediation of defence alone is not sufficient to optimise plant-level defence between densities, as R:FR does monotonically decrease with plant density. Chapter 6 presents a new FSP model that simulates the evolutionary dynamics of a plant population, driven by competition for light and nitrogen in combination with herbivore damage. This evolutionary FSP model is used to analyse how optimal biomass allocation patterns changed with plant density, nitrogen availability and herbivore damage, and how these factors impact the optimal balance between the acquisition and protection of resources. The model results conform with the functional equilibria described by optimal partitioning theory (OPT), which predicts an increase in biomass allocation towards plant parts that partake in the acquisition of a limiting resource. The model further shows that optimal defence levels negatively correlate with nitrogen availability, and positively correlate with plant defence if an increase in plant density decreases the nitrogen availability per plant. This chapter concludes that the adaptive value of plant defence is dependent on the availability of, and competition for both above- and belowground resources. In the general discussion (chapter 7), I discuss the concepts and results presented in chapters 2-6 with respect to the main objective specified in the general introduction (chapter 1). First, I discuss how ecological interactions in a dynamic environment shape growth-defence integration and the challenges and opportunities of using FSP models to simulate these dynamic environments. To illustrate this I present additional simulations that show the effect of a variable plant density on optimal stem allocation, which indicates selection for a high density phenotype. Second, I discuss the role and viability of tolerance as an alternative to plant defence, which highlights the importance of modelling form and function when studying growth-defence integration. Third, I discuss the importance of considering herbivore community dynamics when studying growth-defence integration. To illustrate this importance I present additional simulations that show how specialist herbivores select against taxon-specific forms of defence expression, while generalist herbivores select for overexpression of taxon-specific forms of defence. Fourth, I discuss how considering multiple resources may change the optimal balance between the acquisition and protection of resources. Finally, I present my view on the future perspectives for FSP models of growth and defence, discussing the integration of multiple resources, neighbour detection cues, and dynamics resulting from a complex representation of plant and herbivore communities.