Biochemical plasticity and the evolution of diet-breadth in toxic insects

The evolution of host plant feeding is critical for understanding insect evolution and in particular the responses of populations to a changing climate. For example, UK butterfly species that have done well in response to a warming climate have typically also expanded or altered their patterns of host plant use, while those that have suffered typically have a narrow host plant range. Here we will explore how butterflies can alter their biochemical responses to allow the exploitation of different host plants. This is a form of phenotypic plasticity, where a single genotype can produce alternative phenotypes under different environmental conditions. Plasticity is an important adaptation that can allow organisms to survive variable and heterogeneous environments and promote longer term divergence and diversification.

Plant-feeding insects have to deal with toxic plant chemistry, but in some cases such toxins can also provide an important source of defensive chemicals for the insects. Tropical Heliconius butterflies can either obtain cyanogenic toxins from their Passiflora hosts, or can synthesise their own compounds. We have demonstrated that these butterflies can switch between these two strategies dependent on host chemical composition. When the larval diet lacks toxins that can be sequestered, Heliconius respond by increasing biosynthesis of their own defences. This permits use of a wider range of Passiflora species while maintaining their chemical defences. We can readily distinguish toxins derived from the host plant from those that are made by the butterflies, making this an easily quantifiable form of phenotypic plasticity. However, the genetic and biochemical basis of this plasticity remains unknown, as well as it's ecological importance for niche partitioning. We will first address the ecological context, using targeted metabolomics to track the chemical composition of Heliconius erato across seasons at sites with well characterised host plant use in Brazil and Panama. Next, we will explore the fitness trade-offs between different strategies, addressing the reasons for switching between strategies in a plastic species, Heliconius erato. We will compare growth rate and other life history traits of individuals raised on different diets. We will also compare efficiency of sequestration in a derived specialist species, which obtains its toxins only from a specific host plant. Increased efficiency in the derived lineage is predicted by the 'plasticity first' hypothesis. The other major axis of variation for defensive compounds is their influence on predation, and we will measure toxicity and distastefulness of host-plant derived and synthesised toxins. Third, we will explore how plasticity controlled genetically, testing two alternative hypotheses for the molecular control of plasticity. Using transcriptomics we will estimate changes in gene expression in response to larval diet (presence and absence of host-plant derived toxins), and also test whether plasticity is controlled at the level of protein regulation. Finally, we will explore the evolutionary history of cyanogen biosynthesis across the Heliconiines, using molecular evolutionary approaches across a large data of whole genome sequences. We will study the gain and loss of genes involved in cyanogen uptake and synthesis, comparing generalist with those where cyanogen biosynthesis has been lost.

In summary, this integrative study will explore the ecological context, fitness consequences, genetic control and long-term evolutionary trajectory of plasticity in the use of defensive toxins across a diverse group of insects. We will exploit a readily quantifiable and experimentally tractable system in order to understand how butterflies respond metabolically to variation in host plant chemistry. This will have general relevance to understanding how species can respond to a changing climate.