| In all species with two sexes, every individual is faced with a crucial decision: which sex to become. Different species make this decision in a staggering variety of ways, using a range of both environmental and genetic cues. Such sex determination mechanisms (SDMs) have been the focus of active research in many biological sub-disciplines, as they make an excellent model system for genetic and developmental biologists and can have significant ecological and evolutionary consequences. Unfortunately, relatively little is known about within-species variation in SDMs, and how this variation leads to between-species divergence. In this dissertation, I examine the microevolution of SDMs using strains of the nematode "worm" Caenorhabditis elegans with temperature-sensitive mutations in the sex-determining genes tra-2 and her-1. This species has two sexes: obligately outcrossing males, and self-fertilizing hermaphrodites which can facultatively outcross with males.;I first characterize these mutant strains at the phenotypic and DNA sequence levels, showing that they display thermal sex ratio reaction norms similar to many reptiles with temperature-dependent sex determination. They also differ from each other at the molecular level, suggesting that transitions between SDMs can occur in a variety of ways, and that even SDMs appearing outwardly similar may differ in their genetic basis. However, these mutations have deleterious pleiotropic effects on overall fitness, implying that SDM evolution is constrained, or that compensatory mutations are necessary to ameliorate these effects if new sex determination mutations become fixed.;Next, I show that fog-2, a gene thought to be key for the evolution of self-fertility in this species, is not necessarily so critical after all, as self-fertility in fog-2 null mutants can be restored in mutants with altered tra-2 activity. Therefore, self-fertility could have evolved prior to fog-2, which may have been a later innovation to fine-tune tra-2 regulation and perhaps avoid some negative pleiotropic effects. Molecular dating of the duplication event that generated fog-2, however, shows that it probably evolved almost simultaneously with the origin of self-fertility in this species. Thus, although fog-2 was not the first step in the evolution of self-fertility in C. elegans, it was surely an important step in the process strongly favored by selection.;Finally, I introgress the mutations carried by one of these strains into several additional wild genetic isolates and discover that the mutations' effects are significantly dependent upon the genetic background in which they occur, demonstrating the microevolutionary potential of this SDM. I then use quantitative trait locus (QTL) mapping to identify background loci responsible for this variation. I find that most of it can be explained by only a handful of loci with moderate to large effects. Some of these loci lack known candidate sex-determining genes, highlighting the potential for these previously undiscovered genes with minor or redundant functions to play a role in SDM evolution.;Together, these studies show that SDMs can change in a variety of ways. The presence of cryptic genetic variation suggests that there may be abundant standing variation to compensate for the deleterious effects of new mutations, and that neutral developmental drift may play a role in SDM divergence. They also showcase how studies of natural and lab-generated variation can complement one another to yield novel insights, providing a useful starting point for more detailed exploration of sex determination and its evolution in this species. |
