DNA Repair Proteins

Structure & Function Research Group

The crystal structure of human DNA polymerase β was first solved in 1994. But there is very little structural information that actually addresses the proposed two metal mechanism. Collaborations with the DNA Repair & Nucleic Acid Enzymology Group in the LSB focus on the precise details associated with catalysis in order to come up with a detailed mechanism. To do this, the group is using novel non-hydrolyzible analogs to obtain "all atom" complexes of DNA polymerase β with its substrate and metals.

Human polymerases are also being studied in collaboration with the Kunkel lab in the LSB. These collaborations have resulted in crystal structures of human DNA polymerase γ (an X family polymerase like β) in pre- and post catalytic complexes. These structures allow for direct comparisons with polymerase β to better understand the role of individual residues in catalysis and in determining fidelity. These structures have also provided the greatest insight of the in-line nucleophilic attack associated with nucleotidyltransferase of any polymerase to date. In addition, structures of a "flipped out" nucleotide on the templating strand provide novel insight into the high rate of –1 frame shift errors associated with polymerase γ activity. These structures provide direct evidence for how strand-slippage as proposed by Stresinger results in a –1 deletion. Polymerase γ has been proposed to play a role in the repair of double strand breaks in a process called non-homologous end joining. All of these structures suggest ways by which polymerase γ can accommodate these odd substrates and help repair the breaks.

In addition to the continuing studies of polymerase γ, the group is also focusing attention on other mammalian polymerases and has solved the structure of polymerase µ. Like polymerases β and γ, polymerase µ is an X family member involved in DNA repair. Polymerase µ is believed to play a role in non-homolgous end joining, especially with regards to VDJ recombination events. Polymerase µ is also capable of template independent primer extension.

Figure 1: Active site of DNA polymerase Beta with an incoming dUMPNPP and two magnesium ions bound at the active site. In this orientation the nucleophilic 3’ OH of the DNA primer is in-line with the alpha-phosphate of the dUMPNPP and the bridging oxygen of the pyrophosphate leaving group (represented here as a nitrogen). This model supports the proposed chemistry for the nucleotidyl transferase reaction carried out by DNA polymerases (Batra et al. Structure, 14:1-10 (2006)).

Figure 2: Crystal structure of DNA polymerase lambda bound to DNA containing an unpaired adenine on the templating strand superimposed with the ternary complex of pol lambda with a precatalytic ternary complex (semi-transparent). The unpaired base in the extrahelical position is easily accommodated without distortion to the active site suggesting how pol lambda is able to propagate -1 frameshift errors (Garcia-Diaz et al. Cell, 124:331-342 (2006)).

Figure 3: Ternary complex of DNA polymerase mu. By comparisons to DNA polymerase beta and lambda structures, this structure suggests how the active site of mu may accommodate unusual substrates to aid in double strand break repair of non-homologous ends (Moon et al, Nat Struct Mol Biol, 14:45-53 (2007).

 Back to top