Positive selection on the MHC class II DLA-DQA1 gene in golden jackals (Canis aureus) from their recent expansion range in Europe and its effect on their body mass index

Abstract

Background: In Europe, golden jackals (Canis aureus) have been expanding their range out of the southern and southeastern Balkans towards central Europe continually since the 1960s. Here, we investigated the level of functional diversity at the MHC class II DLA-DQA1 exon 2 in golden jackal populations from Bulgaria, Serbia, and Hungary. Specifically, we tested for positive selection on and geographic variation at that locus due to adaptation to supposedly regionally varying pathogenic landscapes. To test for potential fitness effects of different protein variants on individual body condition, we used linear modeling of individual body mass indexes (bmi) and accounted for possible age, sex, geographical, and climatic effects. The latter approach was performed, however, only on Serbian individuals with appropriate data.

Results: Only three different DLA-DQA1 alleles were detected, all coding for different amino-acid sequences. The neutrality tests revealed no significant but positive values; there was no signal of spatial structuring and no deviation from the Hardy-Weinberg equilibrium across the studied range of expansion. However, we found a signal of trans-species polymorphism and significant test results for positive selection on three codons. Our information-theory based linear modeling results indicated an effect of ambient temperature on the occurrence of individual DLA-DQA1 genotypes in individuals from across the studied expansion range, independent from geographical position. Our linear modeling results of individual bmi values indicated that yearlings homozygous for DLA-DQA1*03001 reached values typical for adults contrary to yearlings carrying other genotypes (protein combinations). This suggested better growth rates and thus a possible fitness advantage of yearlings homozygous for DLA-DQA1*03001.

Conclusions: Our results indicate a demographic (stochastic) signal of reduced DLA-DQA1 exon 2 variation, in line with the documented historical demographic bottleneck. At the same time, however, allelic variation was also affected by positive selection and adaptation to varying ambient temperature, supposedly reflecting geographic variation in the pathogenic landscape. Moreover, an allele effect on body mass index values of yearlings suggested differential fitness associated with growth rates. Overall, a combination of a stochastic effect and positive selection has shaped and is still shaping the variation at the studied MHC locus.