Defensive symbiosis and the evolution of virulence

Abstract

Although environments rife with enemies should cause selection for defensive traits, such enemy-rich environments should also select for greater virulence in co-occurring symbionts, yet many defensive symbionts cause little to no damage while protecting their hosts from enemies. Thus, co-infection of a defensive symbiont and a parasite is predicted to select both for increased virulence in co-infecting symbionts and for increased defense in the protective symbiont. Why then do we observe defensive mutualists that protect hosts while causing little damage? To address this question, we build a symbiont-centered model that incorporates the evolution of two independent traits: defense and virulence. Virulence is modeled as a continuous trait spanning mutualism (negative virulence) and parasitism (virulence) and thus accounts for the entire range of direct effects that symbionts have on host mortality. Defense is modeled as a continuous trait that ameliorates the costs to the host associated with co-infection with a deleterious parasite. We obtain the counterintuitive result that the evolution of increased defense in one symbiont largely leads to the evolution of lower virulence in both symbionts and may even facilitate pathogens evolving to mutualism. However, results are context-dependent and when defensive traits are costly, the evolution of greater defense may also lead to higher virulence.