Cold adaptations along a range limit in an obligate symbiosis

Abstract

Abstract Symbionts can have profound effects on host fitness, adaptations and range distributions. Stress‐induced evolution is difficult to show in obligate symbioses; however, adaptive evolution within an obligate symbiosis can be investigated experimentally or by correlating trait variation with stress along an ecological cline (i.e. temperature‐stress gradient). We investigated the cold tolerance of the fungus‐growing ant Trachymyrmex septentrionalis by performing cold tolerance assays comparing two populations collected from either the southernmost range of their distribution (Bastrop, Texas) or from a site that is approximately 600 km further north (Norman, Oklahoma). We first compared isolated fungal symbionts grown on artificial media to determine cold tolerance of fungus alone. Subsequently, we conducted cross‐fostering experiments between northern and southern host and symbionts to test for synergisms between the partners in generating adaptations of cold tolerance. Ants of the northern fungal populations were more cold adapted than southern fungal populations. Northern nests were deeper and northern colonies initially rejected fungi from the southern population. The cross‐fostering experiments demonstrated that only one partner must be cold tolerant to confer maximum cold tolerance to the ant–fungus symbiosis, because northern ants growing southern fungus under cold stress performed just as well as northern ants growing northern fungi. Our results suggest that cold stress has been an important selective factor during the migration of this ant–fungus symbiosis into northern latitudes during the last 10,000 years, and that cold tolerance likely is an energetically demanding trait that may be traded off with other aspects of the symbiosis' life history. The symbiosis also appears to have evolved several additional adaptations that increase survival in cold environments, such as building deeper nests that insulate the fungi from cold surface. Read the free Plain Language Summary for this article on the Journal blog.