Judith A. Kassis, PhD, Head, Section on Gene Expression
J. Lesley Brown, PhD, Staff Scientist
Sarah DeVido, MS, Technician ¹
Deborah Kwon, BS, Postbaccalaureate Fellow
Catherine Stefaniuk, BS, Postbaccalaureate Fellow

During development and differentiation, genes either 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. Recent evidence suggests that modifications of histones in chromatin contribute substantially to determining whether a gene will or will not be expressed. Our group is interested in understanding how chromatin-modifying protein complexes are recruited to DNA. In Drosophila, two groups of genes, the Polycomb group (PcG) and Trithorax group (TrxG), are important for inheritance of the silenced and active chromatin states, respectively. Regulatory elements called Polycomb group response elements (PREs) are cis-acting sequences required for the recruitment of chromatin-modifying PcG protein complexes. Recently, it has become clear that TrxG proteins act through either the same or overlapping cis-acting sequences. Our group is working to understand how PcG and TrxG proteins are recruited to DNA.

DNA sequences that constitute a PRE

Brown, DeVido, Kassis

PREs are DNA elements through which the PcG of transcriptional repressors act. Many PcG proteins are associated in two protein complexes that repress gene expression by modifying chromatin. Both of these protein complexes specifically associate with PREs in vivo; however, it is not known how they are recruited or held at the PRE. PREs are complex elements made up of binding sites for many proteins. Our laboratory has been working to define all the sequences and DNA binding proteins required for the activity of a 181-bp PRE from the Drosophila engrailed gene. At least nine binding sites are present within this 181-bp PRE. Two of the binding sites are for the Polycomb-group proteins Pleiohomeotic (Pho) and Pleiohomeotic-like (Phol). The proteins GAGA factor and Pipsqueak bind to two other sites. The DNA binding proteins Zeste and Dsp1 also are present within the engrailed PRE that we study. Proteins that bind to the other three sites have not been identified. We have recently found that Site 2, one of the sites necessary for PRE activity, can bind to members of the Sp1/KLF family of zinc-finger proteins. This family of proteins comprises transcription factors and has been extensively studied in mammals. In fact, mammals account for 20 Sp1/KLF family members. In Drosophila, of 10 Sp1/KLF family members, nine bind to Site 2. We derived a consensus-binding site for the Sp1/KLF Drosophila family members and showed that the consensus sequence is present in most of the molecularly characterized PREs. The data suggest that one or more Sp1/KLF family members plays a role in PRE function in Drosophila.

During the past year, we have been working to determine which of the Sp1/KLF family members in Drosophila might be involved in PcG function. We are in the process of making antibodies to the three most likely candidates. We will then determine the expression patterns of the proteins in embryos and perform chromatin immunoprecipitation experiments to determine whether the proteins are bound to PREs. We plan to mutate the genes for factors bound to PREs and analyze the phenotypes of the mutants for possible PcG phenotypes.

Recruitment of Polycomb group protein complexes to DNA

Brown, Kassis; in collaboration with Jones, Wang

Pho binding sites have been identified in many PREs, suggesting that Pho is a key component for the recruitment of PcG protein complexes to DNA. If such 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. As sequence of the Drosophila genome became known, a possible explanation for this apparent paradox was found: another YY1 homologue exists in Drosophila. We generated mutants in the pho-like gene and observed that organisms that are double-mutant for pho and pho-like (from mothers and fathers heterozygous for the two mutations) develop into larvae and have much more severe derepression of homeotic genes than pho mutants alone, showing that pho-like enhances the pho phenotype. In fact, the derepression of homeotic genes seen in imaginal disks from pho-like/pho double mutants is comparable to that seen in many other PcG mutants, suggesting that Pho and Pho-like play redundant but highly important roles in PcG repression.

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. Our aim was to determine how the two proteins function. Using chromatin-immunoprecipitation (ChIP) experiments, we showed that Pho and Pho-like are required for the binding of the PcG proteins Enhancer of zeste (E(z)) and Polycomb (Pc) to a PRE from the homeotic gene Ultrabithorax (Ubx). Thus, treatment of Drosophila tissue culture cells with RNAi to Pho (to deplete Pho protein levels) led to a loss of E(z) and Pc from the Ubx PRE. In addition, we observed no E(z) or Pc bound to the Ubx PRE in pho-like/pho double mutants. Consistent with our previous experiments on the redundancy of pho-like and pho, E(z) and Pc were still bound to the Ubx PRE in either single mutant. Further, our data suggest an order of recruitment of Pc complexes and indicate that Pho and Pho-like directly recruit an E(z) protein complex, which helps recruit Pc to DNA. However, the system is more complex. Studies from other laboratories suggest that Pho also interacts directly with Polycomb and Polyhomeotic, indicating that at least two Pho binding sites might be important for recruiting PcG complexes to DNA. In fact, most PREs contain more than one Pho binding site. Our laboratory is currently testing the hypothesis that several Pho binding sites are required for PRE function. Much remains to be learned about how PcG complexes are recruited to DNA.

The role of PREs and flanking sequences at the engrailed gene

Brown, DeVido, Kwon, Stefaniuk, Kassis

The Drosophila engrailed gene encodes a homeodomain protein that plays an important role in the development of several parts of the embryo, including formation of the segments, nervous system, head, and gut. It also plays an important role in the development of the adult, specifying the posterior compartment of each imaginal disk. Accordingly, engrailed is expressed in a highly specific and complex manner in the developing organism. The 181-bp engrailed PRE that we have been studying is located near the engrailed promoter from -576 to -395 upstream of the transcription start site. We are interested in determining the role of this PRE in the control of engrailed expression. Indeed, in our studies, we learned that this PRE is redundant with other flanking PREs in the endogenous engrailed gene. Another strong PRE is located from -1100 to -1500, probably near other weak PREs. In fact, when we examined the location of Ph and Pho proteins on engrailed DNA by chromatin-immunoprecipation (ChIP), we found that the proteins are bound to a 2.5 kb region extending from the engrailed promoter to about -2.5kb upstream. Therefore, it is perhaps not surprising that a 500bp deletion that includes the 181-bp PRE and flanking sequence did not lead to ectopic engrailed expression. The remaining PREs were apparently sufficient to recruit PcG proteins. However, it is surprising that loss of this DNA led to a loss-of-function phenotype, suggesting that this DNA must also play a positive role in the expression of engrailed. Recent experiments suggest that several positive elements either overlap or are coincident with the PREs. We are working to determine whether positive and negative sequences can be separated.

The regulatory sequences for the engrailed gene extend over a 70 kb region. We used reporter constructs to find sequences important for expression in stripes, the nervous system, the head, and so forth and discovered discrete regulatory elements located throughout the 70kb region as well as at least seven additional PREs also located throughout the region. PcG protein complexes have been shown in vitro to bring together DNA fragments, and it is possible that they cause looping in vivo. We are interested in learning whether the additional PREs are involved in mediating interactions between distant enhancers and the engrailed promoter.

¹ Left the laboratory in March 2006.

COLLABORATORS

Richard S. Jones, PhD, Southern Methodist University, Dallas, TX
Liangjun Wang, PhD, Southern Methodist University, Dallas, TX

For further information, contact jkassis@mail.nih.gov.