During development and differentiation, genes become competent to be expressed or are stably silenced in an epigenetically heritable manner. This selective activation/repression of genes leads to the differentiation of tissue types. Our group is interested in the molecular mechanisms that lead to the heritable transmission of the silenced state. To address this problem, we are studying the mechanism of gene silencing by the Polycomb group genes (PcG) in Drosophila. The PcG genes encode a diverse group of proteins known to be important for silencing of homeotic and segmentation genes during development. Many PcG genes encode chromatin-associated proteins and it has been proposed that they silence transcription by forming protein complexes that inactivate chromatin. PcG proteins act through poorly defined cis-acting DNA sequences called Polycomb group Response Elements (PREs). PREs are thought to recruit PcG protein complexes to the DNA. We have completed a detailed study of one PRE from the segmentation gene engrailed. Within a 139-bp minimal PRE, there are binding sites for eight DNA binding proteins. Our functional data have shown that at least 5 of these sites are required for PRE activity. One of the proteins that binds to this PRE is the product of the pleiohomeotic (pho) gene, a known PcG gene. pleiohomeotic encodes a Drosophila homolog of the mammalian zinc-finger transcription factor Yin Yang 1 (YY1). Another is the sequence GAGAG, known to bind two Drosophila proteins, GAGA factor and pipsqueak. There are two Pho binding sites and two GAGAG sequences within the engrailed PRE. The proteins that bind to the other 4 sites have not been identified. Isolation of these proteins is a major focus of our laboratory. PHO binding sites have been identified in many different PREs suggesting that PHO is a key component for the recruitment of PcG protein complexes to the DNA. If this were the case, then the phenotype of pho mutants should be severe derepression of homeotic genes. Curiously, in pho mutants homeotic genes are only mildly derepressed. With the sequence of the Drosophila genome, a possible explanation to this apparent paradox was found. Another YY1 homolog exists in Drosophila. We have completed an analysis of this pho-related gene that we call pho-like. We find that flies mutant for pho-like are viable, but female sterile. Homozygous pho-like mothers lay eggs that are fertilized but undergo no nuclear divisions. Thus, Pho-like must be deposited in the egg by the mother for development to occur. Pho-like binds to Pho binding sites in vitro and pho-like, pho double mutants show more severe misexpression of homeotic genes than the single mutants. These results suggest that Pho and Pho-like act redundantly to repress homeotic genes. PcG chromosomal proteins are associated with about 100 sites on polytene chromosomes. If Pho and Pho-like are required to anchor PcG protein complexes to the DNA, then the chromosomal association of PcG proteins might be lost in pho, pho-like double mutants. In general, this was not the case. We examined the distribution of five PcG proteins on polytene chromosomes from pho-like, pho double mutants. Pc, Psc, Scm, E(z), and Ph remain bound to polytene chromosomes at most sites in the absence of Pho and Pho-like. At a few chromosomal locations, however, some of the PcG proteins are no longer present in the absence of Pho and Pho-like, suggesting the presence of both Pho-like/Pho-dependent and Pho-like/Pho-independent PREs. We are currently trying to characterize these two types of PREs.
