In the developing C. elegans embryo, maternally-expressed PIE-1 protein is required for specification of germ cell precursors. In
pie-1 (pharyngeal and intestinal excess) mutants, these cells instead form pharyngeal and endodermal cells by a pathway that is dependent on the
skn-1 gene. Germ cell precursors normally appear to lack both embryonic mRNA transcription, and a particular phosphoepitope on the C-terminal repeat of RNA Polymerase II. PIE-1 is expressed specifically in these cells and is required for their transcriptional silencing, and thus for preventing them from being directed to differentate by SKN-1 or other transcription factors. The sequence of PIE-1 does not suggest an obvious model for how it might function. PIE-1 contains two predicted Cys(3)-His zinc finger motifs and an intervening RS-rich region. Other Cys(3)-His zinc finger proteins have been implicated in RNA processing, post-transcriptional gene regulation, and growth factor responses, but the roles of their zinc fingers remain undefined. An effect of PIE-1 on transcription could be mediated at any of a variety of levels, including signaling, an effect on expression or modification of a general transcription factor, or a direct repressive effect on the transcriptional machinery. We have used a mammalian cell transfection assay to investigate the last of these possibilities. Our experiments indicate that PIE-1 can repress transcription driven by various types of RNA Polymerase II promoters in mammalian cell transfection assays. By attaching portions of PIE-1 to the yeast GAL4 DNA binding domain, we have identified a region that can function as a powerful repressor domain when it is thus tethered to a promoter. Critical residues within this repressor domain are strikingly similar to the Pol II C-terminal repeat sequence, suggesting that PIE-1 might act on proteins that bind to this repeat structure. Experiments are in progress to test this model. Our findings demonstrate that PIE-1 contains sequences that can act directly at promoters to repress transcription. We hypothesize that PIE-1 acts on an evolutionarily conserved component of the transcriptional regulatory machinery, and brings to it residues that are capable of effecting repression.