Genes have to be replicated before every cycle of cell division. Although DNA polymerase has a proofreading mechanism to minimize the errors during replication, occasionally mismatch due to replication-errors still happens. In all living organisms there are mismatch repair systems to prevent such mutations from occurring. E. coli has a methyl-directed mismatch repair system comprising MutS, MutL and MutH proteins. Homologues of MutS and MutL proteins are also found in human. Mutations in these proteins are identified in 90% of the hereditary nonpolyposis colorectal cancers. During the last year, we have determined crystal structures of MutH, a 229aa sequence-specific endonuclease which is activated by MutS upon its recognition of mismatch. We have obtained two crystal forms of MutH, and have solved and refined MutH structures of both crystal forms. The crystal structure of MutH allow us to identify the active site of this enzyme. We then made point mutations to confirm the catalytic residues. Based on the crystal structures, we postulate a mechanism for how MutH is activated. We also identified the structual similarity between MutH and restriction endonucleases, such as PvuII, EcoRV and Sau3AI and proposed that type II restriction enzymes are evolved from a common ancestor. We have now determined the crystal structure of a 40Kd N-terminal fragment of MutL. MutL and its homologues, although indispensable for DNA mismatch repair from bacterial to human, has no known function prior to our structural characterization. Based on the structural homology of MutL to an ATPase-containing fragment of DNA gyrase and results we obtained from various functional studies, we conclude that MutL is an ATPase. We have since shown that ATP-binding induces conformational changes in MutL and such changes are essential for MutL's function in DNA repair. We have made various mutants which lacks ATPase activity and are currently further characerizing these MutL mutants. To continue the studies of DNA recombination, we have focused our attention on a bacterial transposase, TN10. We have generated ample amounts of pure transposase, IHF (a cofactor for DNA transposition) and various DNA substrates for crystallographic studies of the system. V(D)J gene rearrangement in vertebrates is essential for the maturation of immune systems. It allows the generation of antibodies and T-cell receptors to build up the defense system. Such gene rearrangement has to be tightly controlled during cell development. Erroneous rearrangement often leads to gene truncation or chromosome translocation that becomes causes of various types of lymphomas. V(D)J gene rearrangement is a type of site-specific DNA recombination. Two proteins, RAG-1 and RAG-2 (recombination activation gene products), are necessary and sufficient to turn on the gene rearrangement in vivo. Dr. Martin Gellert's group at NIH is the first to demonstrate purified RAG-1 and RAG-2 proteins can initiate gene rearrangement in vitro. Active RAG proteins from mouse have been over-expressed in insect cells. My group has tested expression of RAG proteins in E. coli. After making dozens of different fusion constructs, we have finally succeeded in making active RAG-1 in E. coli, which paves the road for both mutational studies as well as crystallographic studies of this extremely important protein. We have also cloned human RAG proteins and made constructs for expression in both E. coli and insect cells. Eventually we are going to determine the three-dimensional structures of RAG proteins and their complexes with the DNA recognition sequences using x-ray crystallographic techniques.
Showing the most recent 10 out of 34 publications
