Ecological pharmacodynamics: prey toxin evolution depends on the physiological characteristics of predators

• We present a model of predator decision making, based on changes in physiology.
• The model allows prey to evolve toxins in relation to predator physiology.
• It shows nutritional value of prey is important in determining toxin investment.
• Toxins should be strongly influenced by rate of metabolism of nutrients and toxins.
• Toxins may enable prey to evolve bigger body sizes.

The use of toxic chemical defences to repel and deter predators is widespread across living organisms, yet there are surprisingly few formal models of toxin evolution. Published models tend to focus on a trade-off between individual benefits and costs of toxicity, and treat predators as simple agents of selection, reducing future attacks when they encounter toxic prey. In this paper we argue, however, that the physiological characteristics of predators may be crucial in determining the nature and outcomes of toxin evolution. To examine this idea we devised and explored a model in which prey defence evolves in the context of predator physiology. We represented this as dose–effect relationships in predators for nutrition and toxins along with variable rates of predator metabolism. Incorporating variables of predator physiology can change views of toxin evolution. A key point is that inclusion of predator physiological variables requires that the nutritional value of prey is explicitly represented in the model, and this directly affects predictions for toxin evolution. In our model costly toxins generally evolve to the point that they are ‘minimally unprofitable’: just toxic enough to make prey typically unprofitable given their nutritional value to predators. As the nutritional value of prey increases, so the minimally unprofitable toxin level of prey tends to increase in step; hence another general prediction from this model is that toxin levels within prey should often correlate with the nutritional value of the prey. Predator physiology and cognition also contribute to variation in the social nature of defence. We argue that incorporating representations of predator physiology is important in the comprehension of toxin evolution and make suggestions for directions of future work.