Title:	Molecular Dynamics Studies of DNA repair enzymes
SNIC Project:	SNIC 2014/1-216
Project Type:	SNAC Medium
Principal Investigator:	Kwangho Nam <kwangho.nam@chem.umu.se>
Affiliation:	Umeå university
Duration:	2014-06-27 – 2014-10-01
Classification:	10407 10603 10402
Homepage:	http://www.chemistry.umu.se/english/
Keywords:

Abstract

The focus of this proposal is to increase our understanding of the function of DNA repair enzymes, with focuses on how these enzymes discriminate the normal DNA base versus damaged DNA base. During the proposed allocation period, we are particularly focus on the numerical simulation studies of (1) human 8-oxoguanine (oxoG) DNA glycosylase (hOGG1) and (2) DNA polymerase β (polβ). hOOG1 is a key paradigm enzyme in base-excision DNA repair. Specifically, hOGG1 (and its various homologues) corrects DNA mutations caused by oxidative damage to guanines (G). For the proposed work, we are collaborating with Prof. Gregory Verdine (Harvard University, USA) to understand the atom-level recognition mechanism of damaged oxoG by hOGG1. Very recently, Prof. Verdine's lab solved the structure of the lesion recognition complex with hOGG1 in the intrahelical state (private communication). This structure is an important step toward understanding how the hOGG1 repair protein discriminates between the oxidized base (oxoG) and the normal base (G). On the basis of the new structure, we will use all-atom MD simulations to evaluate the thermodynamic stability of the G vs. oxoG in the intra- and extra-helical positions relative to the DNA strand. Specifically, we will use μs-long MD simulations and the string method (under continued development in our laboratory in collaboration with Dr. Victor Ovchinnikov, Harvard University, USA) to construct reversible transition paths between the intra- and extra-helical intermediates. In addition to thermodynamic data, we expect the simulations to show a larger energy barrier to the extrusion of a normal (vs. oxidized) base, in support of a kinetic discrimination mechanism. The enzyme polβ is an error-prone base-excision repair DNA polymerase that preferentially induces transition mutations over transverse mutations (78 % vs. 11 %). Polβ is overexpressed in many cancer cells and overexpression of the enzyme in mammalian cells have been shown to significantly increase spontaneous DNA mutations. In collaboration with Prof. Seongmin Lee (University of Texas at Austin, USA), we aim to understand how polβ recognizes the damaged DNA base before its insertion into DNA. In particular, Prof. Lee’s lab recently solved the structures of polβ in complex with normal and damaged nucleotides in the open and the closed conformational states (private communication). Based on these structures, we propose to carry out multiple MD simulations to elucidate the role of protein dynamics on the discrimination of matched versus mismatched base pairs. Further, we will carry out a series of the targeted molecular dynamics simulations to elucidate the conformational transition pathways between the two conformations. In the long run, we will aim to apply the string method simulation to determine the free energy profiles of the conformational transition paths. The proposed simulations are to be performed in explicit water and thus require continued support from SNIC. In particular, the proposed sting method simulations are demanding because they require an ensemble of MD simulations running concurrently, and because of the long running time needed to obtain converged thermodynamic averages (up to microseconds for highly flexible systems such as the DNA/protein complexes).