In the genus Caenorhabditis, some species make XX females, whereas others make XX hermaphrodites. These self-fertile hermaphrodites produce sperm as well as oocytes, and evolved independently in C. elegans, C. tropicalis, and C. briggsae. Thus, each convergent transition to hermaphroditism was probably caused by distinct alterations to germline sex determination. To learn how the sex-determination pathway was modified in C. tropicalis, we are using a forward genetic screen to identify mutants that make oocytes instead of sperm. The first mutation we recovered,
v426ts, is recessive and transforms both somatic and germline development from male to female. Sequence analysis revealed a missense mutation in the last codon of
fem-1. Moreover, null alleles of ctr-
fem-1, induced with CRISPER/Cas9, show complete feminization of both XX and XO animals. Thus, our results show that
fem-1 is necessary for spermatogenesis in both sexes of C. tropicalis, just as in C. elegans. By contrast, none of the femgenes are required for spermatogenesis in C. briggsae. We infer that the fem genes played a dual role in the ancestor of all these species, both repressing
tra-1 and promoting spermatogenesis directly, and that the second role was lost during C. briggsae evolution. To identify additional genes that act in or modify the C. tropicalis sex-determination pathway, we are continuing our screen. To help push it to saturation, we made a ctr-
egl-15 (FGF receptor) mutant that prevents egg laying, which allows for the rapid identification of females among potentially mutant progeny. (This mutation alters one of two conserved splicing isoforms). With this modified screen, we identified three new mutations that have a feminization of the germline phenotype (Fog) in both XXand XOanimals, and two others with an incompletely penetrant Fog phenotype. Using SNPs developed by Matt Rockman's group, we have begun positional mapping. Four of the mutations appear to be on Chromosome I. Cloning and characterization of the affected genes, and identification of additional mutations that block self-fertility, should help us reconstruct the genetic regulation of sex determination in C. tropicalis, and reveal how it was altered to produce self-fertile animals.