Frequency-dependent selection at two enzyme loci in Drosophila melanogaster

Abstract

RECENT development in population genetics indicate that it is often misleading to treat selective values as constants^1–5. These studies indicate that selective forces may be a function of genotype frequencies (frequency-dependent selection). For example, if the rarest genotype is most fit then the rare allele will increase in frequency but will not become fixed in the population, because as it gets more common the fitness of its carriers decreases. This mechanism is attractive since it leads to a stable gene frequency equilibrium which may not involve problems of genetic load or inbreeding depression. Experimental evidence has been obtained^6,7 which supports the view that frequency dependence is involved in the maintenance of the esterase-6 (Est-6) and alcohol dehydrogenase (Adh) loci in Drosophila melanogaster. However, these results are subject to two major criticisms^8–11. First, the observed frequency dependence may have resulted from poor experimental design. In the original report⁶ fitness estimates were made by comparisons of adult frequencies with those expected from the gene frequencies in the previous generation. Frequency-dependent selection appears as an artefact of this method¹². Second, since inbred lines were used in these studies problems of linkage disequilibrium may have occurred. If this view is correct then the observed selective differences were due to differences between lines, not to differences between alleles at a locus. As a consequence of these criticisms the role of frequency dependence as a force responsible for maintaining enzyme polymorphism has been disputed⁹. The experiments reported here overcome these criticisms and indicate that frequency-dependent selection should be reconsidered as an important mechanism involved in the maintenance of enzyme polymorphism.