Analysis of heat and cold tolerance of a freeze-tolerant soil invertebrate distributed from temperate to Arctic regions: evidence of selection for extreme cold tolerance.

Abstract

Tolerance to thermal extremes is critical for the geographic distributions of ectotherm species, many of which are probably going to be modified by future climatic changes. To predict species distributions it is important to understand the potential of species to adapt to changing thermal conditions. Here, we tested whether the thermal tolerance traits of a common freeze-tolerant potworm were correlated with climatic conditions and if adaptation to extreme cold constrains the evolutionary potential for high temperature tolerance. Further, we tested if evolution of thermal tolerance traits is associated with costs in other fitness traits (body size and reproduction). Lastly, we tested if slopes of temperature-survival curves (i.e., the sensitivity distribution) are related to tolerance itself. Using 24 populations of the potworm, Enchytraeus albidus Henle (Enchytraeidae), collected from a wide range of climatic conditions, we established a common garden experiment in which we determined high and low temperature tolerance (using survival as endpoint), average reproductive output and adult body size. Heat tolerance was not related to environmental temperatures whereas lower lethal temperature was about 10°C lower in Arctic populations than in populations from temperate regions. Reproduction was not related to environmental temperature, but was negatively correlated with cold tolerance. One explanation for the trade-off between cold tolerance and reproduction could be that the more cold-hardy populations need to channel energy to large glycogen reserves at the expense of less energy expenditure for reproduction. Adult body size was negatively related to environmental temperature. Finally, the slopes of temperature-survival curves were significantly correlated with critical temperature limits for heat and cold tolerance; i.e., slopes increased with thermal tolerance. Our results suggest that relatively heat-sensitive populations possess genetic variation, leaving room for improved heat tolerance through evolutionary processes, which may alleviate the effects of a warmer future climate in the Arctic. On the other hand, we observed relatively narrow sensitivity distributions (i.e., less variation) in the most heat tolerant populations. Taken together, our results suggest that both cold and heat tolerance can only be selected for (and improved) until a certain limit has been reached.