Research

    We are interested in the role that chromatin structure plays in gene regulation, considering both effects from packaging large domains and local effects of the nucleosome array. We work with Drosophila, combining biochemical, genetic and cytological approaches. We have used a transposable P-element containing a copy of the white gene, a visible marker for gene silencing, and a copy of hsp26, a well-characterized inducible gene, to examine the effect of insertion into different chromosomal domains. While these genes are fully active in euchromatic domains, silencing similar to Position Effect Variegation can be observed when the P-element is inserted into pericentric heterochromatin, telomeres, or the small fourth chromosome.  Investigations to examine the mechanism(s) of gene silencing have shown that changes in the local nucleosome array, as well as spatial organization in the nucleus, are both critical.

    We are exploiting this system to ask what features of genome organization target the formation of heterochromatin to particular domains.  Mapping using this P-element as the reporter indicates that most of the fourth chromosome, in particular the distal 1.2 kb arm containing ~80 genes, is packaged into heterochromatin.  However, we have identified several P-element insertion sites on the fourth chromosome that allow full gene expression, suggesting that this region has some domains permissive for gene expression. More detailed examination has shown that proximity to repetitious elements, in particular  the 1360 transposable element, can be critical for heterochromatin formation on the fourth chromosome.  In a further test, we have found that the 1360 element can enhance silencing at many sites in the euchromatic regions proximal to the pericentric heterochromatin.  Further mapping studies are underway to determine what conditions and sequence elements favor heterochromatin assembly (see Heterochromatin Abstracts).

    Earlier work in the lab identified Heterochromatin Protein 1 (HP1) as a protein preferentially associated with the pericentric heterochromatin, and in a banded pattern along the small fourth chromosome (see figure below). Subsequent analysis showed that HP1 is encoded by Su(var)2-5; both mutations that would be expected to reduce the level of HP1, and a point mutation (shown by others to disrupt the interaction of HP1 with H3-mK9), result in suppression of Position Effect Variegation, demonstrating that HP1 plays a key role in establishing heterochromatic structure. Using a yeast two-hybrid screen, we  identified HP2, an interacting protein that shows a similar distribution pattern on polytene chromosomes. We have recovered 17 alleles of HP2; some are strong suppressors of PEV, while others are not (see selected HP1 and HP2 Abstracts).  We are now screening for proteins that interact with HP2.

    As RNAi appears to play a role in heterochromatin formation in Drosophila, we wonder whether the DNA transposons are targeted for silencing by this system.  In a collaborative study with the Birchler laboratory (U Missouri), we observed that mutations in piwi, aubergine, and homeless can result in a loss of silencing at some of our reporter loci described above, as well as at other repeat loci.  A collaborative effort with Haifan Lin and colleagues (Yale University) has provided evidence for a specific interaction between HP1 and PIWI in Drosophila melanogaster (see selected RNAi Abstracts).  We are currently investigating the interaction of PIWI and HP1 in more detail, using both genetic and biochemical appoaches.

    The above studies required a detailed knowledge of the test gene, hsp26. Previous work from our lab and others has shown that correct assembly of the hsp26 regulatory region in an activatable form requires two (CT)n sites, which bind GAGA factor (see selected hsp26 Abstracts). In a collaboration with D. Gilmour (Penn State), we have found that the presence of an immediately adjacent TFIID binding site is critical. The well-characterized hsp26 gene has also proven useful for studies of silencing by the Polycomb system, carried out in collaboration with V. Pirrotta, Rutgers.