In C. elegans, cell fusion requires the activity of the
eff-1 gene. Nematodes that are mutant for this gene fail to fuse their hypodermal, vulval, and pharyngeal cells during development. We find that ectopic expression of
eff-1 is lethal to embryos and larvae, suggesting that precise temporal regulation of
eff-1 expression in these specific cell types is important for normal development. We have also found that full function of the
eff-1 gene in rescuing mutant phenotypes requires ~7.5 of 5'-upstream sequence. Thus we are interested in understanding the transcriptional control of
eff-1 expression. A transcriptional fusion of the
eff-1 promoter to GFP (
eff-1p::gfp) reveals a complex spatio-temporal expression pattern. Many cells that are fated to fuse express
eff-1p::gfp in the moments leading up to cell fusion, suggesting that their fusion fate/competence may be limited specifically by expression of EFF-1 protein. Cell fusions occur during various developmental stages and in various tissues of the animal, and the
eff-1 promoter appears to be differentially activated during the morphogenetic processes in which cell fusion is involved. We have created a series of transgenic lines that harbor progressively 5'-truncated versions of
eff-1p::gfp. We have identified 0.5-kb regions in the promoter that are required for tissue-specific transcriptional activity of the transgene in the pharynx, vulva, and amphid sheath, and a separate region for repression of gut expression. Enhancer assay experiments have further been used to confirm the tissue-specific expression patterns conferred by some of these regions in the
eff-1 promoter. As these upstream domains are likely to contain enhancing and silencing cis regulatory elements, we have now begun site-directed mutagenesis of recognizable transcription-factor binding motifs. In addition, we are currently performing yeast one-hybrid experiments to identify possible transcription factors that would regulate
eff-1 promoter activity in a tissue-specific context. Our initial efforts indicate that PHA-4 may act directly upon
eff-1 to drive pharyngeal expression. Interestingly, we observe a stochastic 'all-or-none' effect on pharyngeal expression when two putative PHA-4 binding sites are mutated. The similar expression of
eff-1p::gfp by each binucleate cell in the metacorpus of a given animal indicates the existence of a 'noise-damping' signaling mechanism that tightly coordinates gene expression levels among these cells of quite distinct lineage.