No abstract

We investigated the heat tolerance of adults of three replicated lines of Drosophila melanogaster that have been evolving independently by laboratory natural selection for 15 yr at "nonextreme" temperatures (18°C, 25°C, or 28°C). These lines are known to have diverged in body size and in the thermal dependence of several life-history traits. Here we show that they differ also in tolerance of extreme high temperature as well as in induced thermotolerance ("heat hardening"). For example, the 28°C flies had the highest probability of surviving a heat shock, whereas the 18°C flies generally had the lowest probability. A short heat pretreatment increased the heat tolerance of the 18°C and 25°C lines, and the threshold temperature necessary to induce thermotolerance was lower for the 18°C line than for the 25°C line. However, neither heat pretreatment nor acclimation to different temperatures influenced heat tolerance of the 28°C line, suggesting the loss of capacity for induced thermotolerance and for acclimation. Thus, patterns of tolerance of extreme heat, of acclimation, and of induced thermotolerance have evolved as correlated responses to natural selection at nonextreme temperatures. A genetic analysis of heat tolerance of a representative replicate population each from the 18°C and 28°C lines indicates that chromosomes 1, 2, and 3 have significant effects on heat tolerance. However, the cytoplasm has little influence, contrary to findings in an earlier study of other stocks that had been evolving for 7 yr at 14°C versus 25°C. Because genes for heat stress proteins (hsps) are concentrated on chromosome 3, the potential role of hsps in the heat tolerance and of induced thermotolerance in these naturally selected lines is currently unclear. In any case, species of Drosophila possess considerable genetic variation in thermal sensitivity and thus have the potential to evolve rapidly in response to climate change; but predicting that response may be difficult.

Genetic variation for resistance to high temperature stress was studied in populations of D. melanogaster and D. buzzatii from different geographic regions. Drosophila melanogaster individuals were presented with either a direct short exposure to a high temperature or exposure to high temperature after receiving a pretreatment, which increased resistance. Heat-stress resistance varied among populations, with one much more resistant than all others under both treatments. Another possessed low stress resistance when exposed without the heat pretreatment; but with pretreatment, resistance increased relative to the other populations. Evidence from reciprocal crosses suggests that the X chromosome of the more resistant population carries alleles that greatly increase resistance, and that one or more factors on the autosomes also affect resistance. Non-additive interaction effects among the three less resistant population, were suggestive that all differ for various elements that contribute to stress resistance, and that some clearly change inducible resistance more than basal levels. In D. buzzatii, the two least resistant populations were genetically very similar. Crosses to the more resistant population gave results suggesting that the low resistance to heat is dominant. A small X-chromosome effect that increased resistance, and a dominant enhancer of male resistance also may have contributed to variation in resistance.