The expression of C. elegans vulval cell fates requires an RTK/Ras signaling pathway and is antagonized by the synthetic multivulva (synMuv) genes. synMuv genes can be grouped into two redundant classes: A and B. Animals with mutations in one class are non-Muv, whereas animals with mutations in both classes are Muv. Many class B proteins have homologs that act in chromatin remodeling and transcriptional repression. The class A synMuv genes might also function in transcription, as four of the five class A proteins contain zinc-finger or zinc-finger-like motifs, and the two class A proteins tested are nuclear. In a screen for new class A synMuv alleles we isolated the mutation
n4441.
n4441 is dominantly Muv in combination with class B synMuv mutations but not as a single mutant or in combination with class A synMuv mutations. Therefore,
n4441 causes a dominant class A synMuv phenotype. This observation could be explained if
n4441 were a mutation in a synMuv target gene that abolished class A synMuv-mediated repression. We determined that
n4441 is an allele of
lin-3, which encodes an EGF ligand that controls vulval cell fates. Loss-of-function mutations in
lin-3 cause a vulvaless phenotype, whereas overexpression of
lin-3 causes a Muv phenotype. Cui et al. (Dev. Cell 10, 667-672, 2006) recently showed that the two classes of synMuv genes redundantly repress
lin-3 expression. Using real-time RT-PCR, we found that
lin-3(
n4441) , like other class A synMuv mutations, synthetically causes an increase in
lin-3 mRNA levels when combined with a class B synMuv mutation.
lin-3 has multiple transcription start sites several kb apart. Because there are multiple
lin-3 transcripts and the
n4441 mutation is closest to the start of the
lin-3a transcript, we tested whether the synMuv genes might specifically regulate
lin-3a. Preliminary evidence shows that in a
lin-3(
n4441); class B double mutant, the
lin-3a transcript is upregulated more than is total
lin-3 relative to wild-type levels, suggesting that the
n4441 mutation might cause derepression of the
lin-3a transcript specifically. Because a mutation in
lin-3 can entirely recapitulate the class A synMuv phenotype,
lin-3 is likely to be the major functionally relevant target of the class A synMuv genes in vulval development. Based on these data, we propose that the class A synMuv proteins bind to the
lin-3 promoter at the site of the
n4441 mutation and repress transcription of
lin-3a. Currently, we are determining the exact DNA sequences in the
lin-3 promoter required for class A synMuv-mediated repression and testing whether the class A synMuv proteins directly bind to the
lin-3 promoter.