A molecular explanation of frequency-dependent selection in Drosophila.

Frequency-dependent selection provides a means for maintaining genetic variability within populations, without incurring a large genetic load. There is a wealth of experimental evidence for the existence of frequency-dependent changes in genotypic fitness among a wide variety of organisms. Examples of traits which have been shown to be subject to frequency-dependent selection include the self-incompatibility alleles of plants, chromosomal rearrangements in Drosophila, visible mutations, enzyme variants and rare-male mating advantage in Drosophila. These experiments have been interpreted in a number of different ways. Principally, frequency dependence of genotype fitness may result from intergenotype facilitation due to the production of biotic residues, or from the differential use of resources by the competing genotypes. However, it has proved extremely difficult to isolate and identify any biotic residue of importance or, alternatively, to understand the manner in which genotypes partition the environment. Thus, the difficulty in the interpretation of experiments which show frequency-dependent selective effects stems largely from our lack of understanding of the exact physiological mechanisms which produce these frequency-dependent effects. The principal aim of this study was to investigate the mechanisms associated with frequency-dependent selection at the amylase locus in Drosophila melanogaster. The excretion of catalytically active amylase enzyme and its effect on food medium composition were correlated with the outcome of intraspecific competition between amylase-deficient and amylase-producing genotypes. Amylase-producing genotypes were shown to excrete enzymatically active amylase protein into the food medium. The excreted amylase causes the external digestion of dietary starch; this accounts for the frequency-dependent increase in the viability of the amylase-deficient mutants in mixed cultures, maintained on a starch-rich diet.