Quantitative-Genetic Analysis of Temperature Regulation in Mus domesticus. IV. Pleiotropy and Genotype-by-Environment Interactions

Publisher Website
Google Scholar
Abstract
The heritabilities of, and genetic correlations among, several traits related to thermoregulation in house mice were examined in a randomly bred population. The traits studied comprised aspects of morphology, behavior, and physiology. Effects of the environment, both internal and external, on the phenotypic expression and genetic architecture of the traits were considered. Genotype-by-environment interactions of two kinds were found to be important. The external environment had an effect on the expression of genetic influences on behavioral thermoregulation, with shifts in the ambient temperature causing a significant change in heritability. The internal environment affected genetic variation in physiological thermoregulation, with significant differences being found in the heritability of body temperature between males and females. Phenotypic dimorphisms were poor predictors of sex differences at the genetic level. Morphological characters were phenotypically the most dimorphic, but genotypically the least dimorphic; body temperature showed the reverse relationship. Significant additive genetic correlations were found between several of the traits. These were highest for pairs of similar characters, for example, body weight at different ages and nesting behavior at different temperatures. The correlation between morphological and behavioral aspects of temperature regulation was also significant. Physiological aspects of thermoregulation could evolve essentially as independent characters, but behavior and morphological traits would be expected to evolve in concert. The relevance of these results to general predictive theories is discussed. In particular, the assumption of no genotype-by-environment interactions could be misleading, and the occurrence of such interactions is likely to be troublesome when selection pressures fluctuate or differ qualitatively in males and females. Furthermore, genetic correlations between traits can pose a major difficulty, even when the traits vary as much as aspects of behavior and morphology. We conclude that a greater amount of empirical information is needed for a wider range of traits on both genotype-by-environment interactions and additive genetic correlations, an that such information must be incorporated into predictions of specific evolutionary adaptations.