Genetics of Female Resistance to Male-Induced Harm in Drosophila Melanogaster

When the two sexes maximize their fitness in different ways, a sexual conflict emerges, in which traits favored in one sex are not favored in the other (Parker 1979, Arnqvist and Rowe 2005). Under interlocus sexual conflict, a type of sexual conflict involving alleles at separate loci (Rowe et al. 2018), one sex evolves traits that improve its reproductive success at the expense of the other. In turn, this generates selection on the other sex to resist the effects of these traits. The outcome is thought to be an evolutionary arms race between the two sexes known as sexually antagonistic coevolution (SAC) (Parker 1979). Most studies on SAC rely on phenotypic data alone, thus compelling genomic evidence for how SAC evolves is lacking (Rowe et al. 2018). Moreover, there is a historical disparity in research focus: much more attention has been paid to the genetic mechanisms underlying male trait evolution than female trait evolution (Ah-King et al. 2014; Orr et al. 2020). To establish the importance of female evolution for driving SAC, we implemented a combination of phenotypic and genetic approaches to address the gaps in our knowledge of female evolution within SAC using a model organism that exhibits extensive sexual conflict, Drosophila melanogaster. Male D. melanogaster have evolved proteins in the seminal fluid (SFPs) that increase their own fitness (Ravi Ram and Wolfner 2007b; Hollis et al. 2016) but decrease the lifespan (Chapman et al. 1995) and fecundity (Wigby and Chapman 2005) of their female partners. We first measured the variation in male harm through both lifespan and fecundity among 61 different inbred lines from the Drosophila Genetic Reference Panel (DGRP). This allowed us to identify low- and high-harm lines for later assays of female resistance. We found that males from the different inbred lines did, in fact, vary in their degree of harm to their female partners through both lifespan and fecundity, supporting the results of prior studies (Friberg 2005; Lew and Rice 2005; Filice and Long 2016). Hereafter, we focused on harm through fecundity, as fitness is ultimately determined by reproductive success. We then quantified the variation in female resistance to fecundity-based harm among 139 different inbred lines from the DGRP by mating females from each line to high-harm and low-harm males identified in the male-harm assays. By comparing the fecundity values of females of a given line who were mated to high-harm males to females of the same line who were mated to low-harm males, we obtained a measure of resistance with the expectation that more resistant lines would show little difference in fecundity between the two treatments (Lew et al. 2006). We found the lines did vary in their degree of female resistance, supporting the few prior studies on this topic (Linder and Rice 2005; Lew et al. 2006). The results also supported the validity of the male harm assays, as most lines assessed for female resistance showed higher fecundity values when mated to the low-harm males than when mated to the high-harm males. Using the lifespan and fecundity data from the male harm assays and the fecundity data from the female resistance assays, we performed genome-wide association studies (GWAS) for the three variables using the DGRP's GWAS pipeline. We were not able to identify variants associated with any of the three variables after correcting for multiple comparisons but were able to for all three under less strict significance thresholds (α = 1 x 10-5). There were 12, 6, and 17 annotated variants associated with lifespan-based male harm, fecundity-based male harm, and female resistance, respectively, that had a p-value less than 1 x 10-5. Finally, we used the GAL4/UAS system and RNA interference to knockdown the expression and assess the effects of two of the top variants associated with fecundity-based male harm and the 11 top variants associated with female resistance. We found that knocking down ed reduced male harm. Of the female resistance genes we examined, the knockdowns of Pif1B and CG15373 positively affected female resistance, implying they normally reduce female resistance when expressed. Furthermore, the knockdowns of bun, Sh, and CngA decreased female resistance, indicating they normally function to increase female resistance when expressed. These results support some of our findings from our GWAS and provide candidates for further research. In conclusion, this work provides evidence of sexually antagonistic genetic variation in both male-induced harm and female resistance to male harm, supporting the idea of sexually antagonistic coevolution, and provides insight into the underlying genetic architecture and molecular functions involved in fecundity-based harm and female resistance in our study system. It also addresses a long-standing gap in our knowledge of the female side of sexual conflict and reproductive biology (Orr and Hayssen 2020; Orr et al. 2020; Wigby et al. 2020).