Screening for association between SNPs in candidate genes and heart rate phenotypes in Drosophila melanogaster. N. Nikoh , A. Duty , G. Gibson. Department of Genetics, North Carolina State Univ, Raleigh, NC.

   We are dissecting the molecular basis of genetic variation affecting cardiac rhythm in D. melanogaster. Genetic and physiological analyses implicate monoamine receptor and potassium channel loci as candidate genes for reduction or increase of pupal heart rate. We have sequenced 12 kb of three serotonin receptors (5-HT1A, 5-HT1B, and 5-HT2) and 6 kb of two potassium channels (ether a go-go and seizure) in two datasets of 130 and 85 nearly isogenic lines from NC and CA, respectively. These lines were highly inbred for more than 15 generations, yet the frequency of heterozygotes at each locus within the two populations was between 6 and 35 percent. In the 10 kb of sequence from the 5-HT1A and 5-HT1B loci, a total of 467 substitution and 50 indel polymorphisms were found, of which 184 substitutions and 21 indels are common, in the sense that the rare allele frequency is larger than 5 percent. Haplotype structure analysis shows that linkage disequilibrium (LD) usually decays within 1kb. Interestingly however, haplotype block-like structure was observed near the 2nd intron of the 5-HT1A, where LD extended over 4 kb. Two cardiac rhythm phenotypes, heart rate (HR) and heart rate variability (HRV) have been determined for these lines. Statistical associations between common polymorphic sites and the phenotypes were inferred with two types of tests. Considering each site independently, 15 of the 205 sites were associated with variation for each of the traits HR and HRV at significance level P<0.05. However, no sites were significant after Bonferroni correction for multiple tests. A permutation test allowed pooling the statistics of sites in particular regions of the genes, and provided evidence that the region surrounding the 3rd exon of 5-HT1A are more likely to be associated with HRV than other regions of the gene. Since the mean values of the phenotypes are slightly different between the two populations, adjustments are also being made for population stratification.