mRNA 3' End Processing
We use X-ray crystallography in conjunction with other biochemical and biophysical techniques to study the structure and function of the proteins responsible for 3' end processing of messenger RNA. This processing occurs in all eukaryotic cells, and defects are associated with a variety of human diseases. This project is part of a longstanding collaboration with Claire Moore in the Department of Molecular Biology and Microbiology.
Figure 1. Poly(A) polymerase contains three domains that surround the active site. In the structure shown, electron density for two nucleotides was observed. The one colored red sits at the site of the incoming ATP. The white one occupies the site of the 3' end of the mRNA. Upon RNA binding the blue and purple domains rotate such that they interact with one another and completely encircle the RNA substrate.
Our past work in this area has concentrated on poly(A) polymerase, the template-independent RNA polymerase responsible for mRNA elongation after cleavage at the 3' end. We have determined crystal structures of the enzyme with and without substrates, and shown that it undergoes significant molecular motions during catalysis. We've also published a series of kinetic studies to characterize the mechanism of the polyadenylation reaction and we've studied the complex that poly(A) polymerase makes with Fip1, the protein that tethers it to the remainder of the mRNA processing assembly. Our current work on Poly(A) polymerase involves the design and characterization of inhibitors and further characterization of the linkage between the polymerase and the remainder of the complex.
Cleavage Factor I
We are also very interested in understanding the structure of Cleavage Factor I (CFI), a five-protein complex in the yeast system. This complex interacts with RNA polymerase II and pre-mRNA. It functions to recruit the remaining proteins of the 3' end processing complex (including the nuclease and polymerase) to the nascent mRNA strand. All five of CFI proteins have been the subject of structural studies. These studies have yielded high resolution structures for a number of domains and fragments, but the overall picture is both disjointed and incomplete. We are working to develop a three-dimensional model of this complex by combining the existing high resolution structures and interaction data with EM reconstructions of the complex.
Figure 2. Structures of the five Cleavage Factor I subunits (or their homology models) are shown. Regions that are missing in these structures are represented by spheres that occupy the same volume as structurally-uncharacterized amino acids. The relative orientation of the individual subunits and of the spheres is arbitrary.
Large T Antigen Function in Viral Replication
A second major project in the lab aims to understand how large T-antigen molecules from DNA tumor viruses interact with DNA during the early steps of viral replication. Much of our work in this area is being done in collaboration with Peter Bullock in the Tufts Biochmistry Department. T-antigen interactions involve site-specific binding to repeating elements in the DNA sequence at the viral origin of replication. Interaction with T-antigen leads to melting and unwinding of the DNA. The viral origins we study contain well-defined DNA elements, and only a single protein is required for origin recognition and DNA melting. Origin sequences in higher eukaryotes are not easily identified, and many more proteins are involved in replication initiation. Nonetheless, there is strong evidence that the eukaryotic and viral systems are structurally similar. Thus, understanding the mechanism of viral replication initiation will likely yield a clearer picture of the events that normally occur in dividing eukaryotic cells. Our current work involves the T-antigen molecules from both SV40 and Merkel Cell Carcinoma Virus (MCV).
Figure 3. The DNA binding domain of SV40 T-antigen interacts with the major groove of DNA containing the pentameric sequence GAGGC. Within this domain, segments labled A1 and B2 are the primary determinants of sequence-specific binding. The oligonucleotide duplex is slightly distorted relative to canonical B-form DNA (shown in orange).