Mark Glover

A. BRCA1 BRCT domain. The C-terminal, BRCT domain of BRCA1 is essential to its tumour suppressor function. BRCT domains are found in a large number of proteins that regulate DNA repair, where these domains act as multi-purpose protein-protein interaction modules. We were the first to determine the structure of the BRCA1 BRCT domain (Williams et al., 2001), and have carried out a detailed analysis of the structural and functional defects associated with a large panel of BRCA1 mutations derived from breast cancer screening programs (Williams et al., 2003; Williams and Glover, 2003). We determined the structure of the BRCA1 BRCT bound to a phospho-peptide target, revealing the structural basis for the biochemical function of this domain (Williams et al., 2004). This work suggested that other members of the BRCT protein family might act as phospho-peptide binding modules (reviewed in Glover et al., 2004). For example, we and others have shown that the BRCT protein MDC1 specifically recognizes the phosphorylated tail of the histone variant, H2AX (Lee et al., 2005). Phospho-H2AX is a critical chromatin signal associated with DNA double strand breaks and the recognition of this mark by MDC1 initiates the assembly of other DNA damage signaling and repair proteins at the DNA lesion. Future work will involve determining the structure and function of other BRCT proteins involved in the DNA damage response. We will use this information as a basis for the design of BRCT inhibitors that could provide leads in the development of new anti-cancer drugs.

 

B. Mammalian polynucleotide kinase (PNK). PNK is a key enzyme in the repair of DNA strand breaks, by both the non-homologous end joining, and base excision/single strand break repair pathways. We determined the crystal structure of the intact enzyme, revealing distinct kinase and phosphatase catalytic domains, and a distinct regulatory FHA domain (Bernstein et al., 2005). Our structure of the FHA domain bound to the scaffold protein XRCC4 reveals how PNK is recruited to sites of repair. We have also characterized the distinct DNA substrate preferences for the phosphatase and kinase domains, which indicates that the two domains act independently (Bernstein et al., 2009). Together with our collaborator, Michael Weinfeld (Cross Cancer Institute), we plan to use this information to develop PNK inhibitors that could be lead compounds for new anti-cancer drug development.

2. Regulation of plasmid DNA architecture and interactions with the conjugative pore complex. The conjugative transfer of F-like plasmids involves the nicking and unwinding of the plasmid DNA and the subsequent active transport of the single stranded plasmid through a multi-protein channel, the conjugative pore, which connects the cytoplasms of donor and recipient cells. The DNA around the site of nicking is ordered within a multiprotein complex which involves several DNA binding proteins which facilitate the activity of TraI, a protein which nicks the DNA and unwinds the single strand for transfer. A key protein in this complex is TraM, which binds multiple sites around the nicking site and facilitates TraI binding and activity. TraM also recruits the plasmid to the conjugative pore for transport to the recipient cell. We hypothesized that TraM undergoes conformational changes which could serve as a molecular signal between the conjugative pore and the plasmid to regulate conjugation in response to mating contact with the donor cell. We showed that the tetramerization domain of TraM is critical in the process, and undergoes a pH-dependent conformational change which modulates conjugative efficiency (Lu et al., 2006). Most recently, we have shown that the tetramerization domain serves as the initial point of contact for the cytoplasmic face of the conjugative pore, TraD (Lu et al., 2008) and have also shown that proteins with similar DNA binding protein architectures function in other plasmid transfer systems (Lu et al., 2009). Future work is directed towards understanding large scale complexes between TraM and its plasmid DNA target sequences, as well as with its major protein partners, TraI and TraD, to understand how plasmid nicking, unwinding and transfer are coordinated during conjugation.

3. Structural studies of Ndt80, a sporulation-specific transcription factor in yeast. Ndt80 drives the transcription of a large set of genes in response to the successful completion of DNA recombination during meiosis in S. cerevisiae. Our work is directed towards understanding the way in which Ndt80 binds its specific DNA target, the MSE, to regulate the expression of genes that are required for progression through meiosis. We used biochemical methods to define the DNA binding domain of Ndt80 and have determined the X-ray crystal structures of complexes of the Ndt80 DNA-binding domain bound to MSE-containing DNAs, as well as the structure of the uncomplexed protein. Surprisingly, our results reveal that Ndt80 is a member of the Ig-fold family of transcription factors, which includes the p53, NF- k B, NFAT, Runt, and STAT sub-families. Our crystals of Ndt80-DNA complexes diffract X-rays to better than 1.40 Å, the highest resolution yet achieved for any protein-DNA complex. Not only is our work revealing the structural elements that define sequence-specific DNA binding in exceptional detail, but, by comparing the structures of the free and DNA-bound forms of Ndt80, we are studying conformational changes that occur in Ndt80 upon DNA binding. Our work has revealed a previously unrecognized, mode of sequence-specific DNA recognition that is conserved in many families of transcription factors and can help to explain the DNA binding preferences of these proteins (Lamoureux et al., 2004, Lamoureux and Glover, 2006). Long-range goals include determining the structures of other domains of Ndt80, particularly its transactivation domain, as well as the structures of another transcription factor, Sum1, which binds MSEs and represses the transcription of sporulation-specific genes during vegetative growth.