Genetic differentiation between alpine and lowland populations of a butterfly is related to PGI enzyme genotype


  • Isabell Karl,

  • Thomas Schmitt,

  • Klaus Fischer

I. Karl (, Dept of Animal Ecology I, Bayreuth Univ., DE-95440 Bayreuth, Germany. – T. Schmitt, Dept of Biogeography, Trier Univ., DE-54296 Trier, Germany. – K. Fischer, Dept of Animal Ecology I, Bayreuth Univ., DE-95440 Bayreuth, Germany, and Zoological Inst. and Museum, Greifswald Univ., DE-17489 Greifswald, Germany.


Understanding the ecological process of population differentiation and identifying the molecular changes that contribute to adaptation is a central aspect of evolutionary biology. In this study we analyzed the geographic variation in allozyme allele frequencies (based on 15 enzyme systems representing 18 loci) across 18 populations of the butterfly Lycaena tityrus from different altitudes. Population genetic analyses showed that within population genetic diversity (e.g. mean number of alleles per loci: 1.8; expected heterozygosity: 12%) was within the typical value range for the Lepidoptera. The populations of L. tityrus investigated showed a remarkable genetic differentiation (FST: 0.065), being clearly divided into an alpine (high-altitude) and a non-alpine (low-altitude) cluster. This differentiation was almost entirely caused by variation at a single enzyme locus, PGI. Although the involvement of historical events cannot be ruled out, several lines of evidence suggest that the specific pattern of allozyme (and in this case particularly PGI) variation found is caused by thermal selection. For example, genetic variation was highly locus-specific rather than relatively uniform, as should be expected for effects of natural selection. Further, the PGI 2–2 genotype dominating in alpine (in contrast to low-altitude) populations is known to exhibit increased cold stress resistance and relatively long development times typical of alpine populations. Thus, PGI (or possibly a closely linked gene) is an obvious target for thermal selection in L. tityrus. This study exemplifies how allozyme analyses can be used to detect candidate loci likely to be under selection.