Genome-Wide Association Analysis of Anoxia Tolerance in Drosophila melanogaster

Abstract

As the genetic bases to variation in anoxia tolerance are poorly understood, we used the Drosophila Genetics Reference Panel (DGRP) to conduct a genome-wide association study (GWAS) of anoxia tolerance in adult and larval Drosophila melanogaster Survival ranged from 0-100% in adults exposed to 6 h of anoxia and from 20-98% for larvae exposed to 1 h of anoxia. Anoxia tolerance had a broad-sense heritability of 0.552 in adults and 0.433 in larvae. Larval and adult phenotypes were weakly correlated but the anoxia tolerance of adult males and females were strongly correlated. The GWA identified 180 SNPs in adults and 32 SNPs in larvae associated with anoxia tolerance. Gene ontology enrichment analysis indicated that many of the 119 polymorphic genes associated with adult anoxia-tolerance were associated with ionic transport or immune function. In contrast, the 22 polymorphic genes associated with larval anoxia-tolerance were mostly associated with regulation of transcription and DNA replication. RNAi of mapped genes generally supported the hypothesis that disruption of these genes reduces anoxia tolerance. For two ion transport genes, we tested predicted directional and sex-specific effects of SNP alleles on adult anoxia tolerance and found strong support in one case but not the other. Correlating our phenotype to prior DGRP studies suggests that genes affecting anoxia tolerance also influence stress-resistance, immune function and ionic balance. Overall, our results provide evidence for multiple new potential genetic influences on anoxia tolerance and provide additional support for important roles of ion balance and immune processes in determining variation in anoxia tolerance.

Keywords: D. melanogaster; DGRP; GWAS; hypoxia; stress tolerance.

Figure 1

Adult survival phenotype showing A.) mean survival ± SE for each line after 6 h of anoxic exposure. Survival is sorted by female survival. B) Male and female survival was strongly correlated (r = 0.839, P < 0.0001).

Figure 2

Larval anoxia phenotype showing A) mean survival ± SE for each line after 1 h of anoxic exposure. B) Larval survival was significantly, but not strongly correlated to adult survival pooled by sex (r = 0.18, P = 0.02).

Figure 3

Results of the RNAi-based functional experiments for seven genes identified in the GWA for adults exposed to 6 h of anoxia. Likelihood ratio tests were used to compare each RNAi knockdown to the control for each sex. The dashed lines represent the survival proportions for control groups exposed to anoxia for each sex. All genes significantly reduced anoxia tolerance using RNAi-mediated knockdown. * P < 0.05, **P < 0.01, ***P < 0.001. †only females were assayed.

Figure 4

Results of the RNAi-based functional experiments for ten genes identified by the GWA for larvae exposed to 1 h of anoxia. Values are shown as the proportion of each genotype surviving 1 h of anoxic exposure. The dashed line represents the expected ratio if there was no effect of the knockdown on surviving anoxia. Likelihood ratio tests were used to compare RNAi knockdown to the control. * P < 0.05, **P < 0.01, ***P < 0.001

Figure 5

Results of the SNP-based functional experiments for two SNPs associated with adult anoxia tolerance. Each row represents the identified SNP. Likelihood ratio tests were used to compare the two alleles separately for each sex. * P < 0.05, **P < 0.01, ***P < 0.001