Genotype-by-Environment Interactions and Adaptation to Local Temperature Affect Immunity and Fecundity in Drosophila melanogaster

Natural populations of most organisms harbor substantial genetic variation for resistance to infection. The continued existence of such variation is unexpected under simple evolutionary models that either posit direct and continuous natural selection on the immune system or an evolved life history “balance” between immunity and other fitness traits in a constant environment. However, both local adaptation to heterogeneous environments and genotype-by-environment interactions can maintain genetic variation in a species. In this study, we test Drosophila melanogaster genotypes sampled from tropical Africa, temperate northeastern North America, and semi-tropical southeastern North America for resistance to bacterial infection and fecundity at three different environmental temperatures. Environmental temperature had absolute effects on all traits, but there were also marked genotype-by-environment interactions that may limit the global efficiency of natural selection on both traits. African flies performed more poorly than North American flies in both immunity and fecundity at the lowest temperature, but not at the higher temperatures, suggesting that the African population is maladapted to low temperature. In contrast, there was no evidence for clinal variation driven by thermal adaptation within North America for either trait. Resistance to infection and reproductive success were generally uncorrelated across genotypes, so this study finds no evidence for a fitness tradeoff between immunity and fecundity under the conditions tested. Both local adaptation to geographically heterogeneous environments and genotype-by-environment interactions may explain the persistence of genetic variation for resistance to infection in natural populations.
Author Summary Genetic variation for resistance to infection is ubiquitous in natural animal and plant populations. This observation runs counter to intuition that resistance should be an important determinant of fitness, and that alleles conferring low resistance should be eliminated by natural selection. We use the model insect Drosophila melanogaster to test the hypotheses that species-wide genetic variation for resistance may be maintained by a) adaptation of subpopulations to their local environmental conditions (temperature), b) genotype-by-environment interactions (GxE) determining resistance, and c) correlated fitness costs of resistance, or life history tradeoffs. We measure resistance to bacterial infection and fecundity at three experimental temperatures in D. melanogaster collected from three environmentally distinct subpopulations. Indeed, we find that flies from a tropical African subpopulation are, on average, less resistant to infection and less fecund at low temperature than are flies from temperate and semi-temperate populations in North America. We observe considerable GxE for both traits in all populations. Although individual flies are less fecund when they have more severe infections, the genetic correlations between resistance and fecundity are either positive or nonsignificant under our experimental conditions, so we do not find evidence of a life history tradeoff. We conclude that adaptation to local abiotic environment and GxE may maintain species-wide genetic variation for resistance to infection (and fecundity) in D. melanogaster, and by logical extension, other species.