In C. elegans, XX animals are self-fertile hermaphrodites, and XO animals are males. The only other male/hermaphrodite species in this genus is C. briggsae. However, our phylogeny shows that these species are not sisters. Furthermore, work by Eric Haag and Dave Pilgrim indicates that the fem genes do not regulate germ cell fates in C. briggsae, although they are essential in C. elegans. Thus, it seems likely that hermaphroditism evolved separately in these species. How did this happen? And why did it happen twice within this genus?
To answer these questions, we are taking two approaches. Here, we describe a genetic analysis of C. briggsae, to identify the genes that control hermaphrodite development. In a separate abstract, C. Baldi and R. Ellis describe what changes would be needed in the male/female species C. remanei to create hermaphrodites.
By screening for mutations that transform XX animals into females, we identified these genes:
glf-1 IV. GLF-1 is required for XX animals to develop as hermaphrodites, but plays no role in males. Thus, it acts like
fog-2 does in C. elegans. However, Tim Schedl could not identify a homolog of
fog-2 in C. briggsae, so GLF-1 probably acts by a novel mechanism. The original allele,
v35, is temperature-sensitive, and we have isolated several suppressors. We also found two more alleles in our screen for XX females, and four alleles in a non-complementation screen. All are ts, and all affect only XX animals.
Three alleles on LGI that cause both sexes to produce oocytes. We suspect that two are alleles of
fog-1, and that one is an allele of
fog-3, and are testing this hypothesis by sequencing DNA from each strain.
At least five alleles on LGII that cause both sexes to produce oocytes. These mutations fail to complement each other, and a new gene,
glf-2.
At least one allele of
tra-1 on LGIII that causes oogenesis in both sexes, perhaps like rare
tra-1(gf) alleles do in C. elegans.
Taken together, these results suggest that some parts of the sex-determination pathway are similar in both C. elegans and C. briggsae, like the roles of
fog-1,
fog-3 and
tra-1. Other parts, in particular the functions of
glf-1 and
glf-2, are likely to be novel.
Since
glf-1 appears to be the key to hermaphroditic development in C. briggsae, we have been using SNP mapping to clone it. We have narrowed its location down to one of two finger-printed supercontigs on LGIV, and are using fine-scale mapping to identify the locus.