DNA Deformation Exerted by Regulatory DNA-Binding Motifs in Human Alkyladenine DNA Glycosylase Promotes Base Flipping

Abstract

Human alkyladenine DNA glycosylase (AAG) is a key enzyme that corrects a broad range of alkylated and deaminated nucleobases to maintain genomic integrity. When encountering the lesions, AAG adopts a base-flipping strategy to extrude the target base from the DNA duplex to its active site, thereby cleaving the glycosidic bond. Despite its functional importance, the detailed mechanism of such base extrusion and how AAG distinguishes the lesions from an excess of normal bases both remain elusive. Here, through the Markov state model constructed on extensive all-atom molecular dynamics simulations, we find that the alkylated nucleobase (N3-methyladenine, 3MeA) everts through the DNA major groove. Two key AAG motifs, the intercalation and E131-N146 motifs, play active roles in bending/pressing the DNA backbone and widening the DNA minor groove during 3MeA eversion. In particular, the intercalated residue Y162 is involved in buckling the target site at the early stage of 3MeA eversion. Our traveling-salesman based automated path searching algorithm further revealed that a non-target normal adenine tends to be trapped in an exo site near the active site, which however barely exists for a target base 3MeA. Collectively, these results suggest that the Markov state model combined with traveling-salesman based automated path searching acts as a promising approach for studying complex conformational changes of biomolecules and dissecting the elaborate mechanism of target recognition by this unique enzyme.

Automated Path Searching Reveals the Mechanism of Hydrolysis Enhancement by T4 Lysozyme Mutants

“…According to the convergence analysis of TAPS by MDS [ 66 ] and [ 50 , 51 , 63 ], all three T4L variant systems reached convergence to their final MFEPs (orange) within 40 to 60 iterations ( Figure S2A–C ). Such convergence rates corresponded to the accumulated sampling of ~102.46 ns, ~165.48 ns, and ~72.49 ns, respectively, for T4L-L99A, T4L-L99A,G113A, and T4L-L99A,G113A,R119P ( Figure S2D–F ).…”

“…As described in our previous studies [ 50 , 51 ], the convergence of TAPS optimization can be conveniently checked in two ways:…”

“…the distance between the path in its current iteration and the initial path, as measured by [ 50 , 51 , 63 ]. See the details provided in Equation (2):…”

“…Recently, we developed an efficient path-searching method—the travelling-salesman-based automated path-searching (TAPS) method—which aims to efficiently locate high-dimensional minimum-free-energy paths (MFEPs) between two stable states of a biomolecular system at affordable cost [ 49 ]. The robustness and efficiency of this method have been demonstrated in a variety of complex systems, with sizes ranging from tens to hundreds of residues [ 50 , 51 ]. Though not directly providing kinetic mechanisms, the high-dimensional MFEP found by TAPS is sufficient in offering atomistic insights into the effect of a mutation on the dynamics of the protein.…”

Unveiling the methyl transfer mechanisms in the epigenetic machinery DNMT3A-3 L: A comprehensive study integrating assembly dynamics with catalytic reactions

“…To fully depict the entire catalytic mechanism of DNMT3A in a meaningful biological assembly, we have deployed cutting-edge computational strategies in this study. Building Markov State Models (MSM) is the advanced concurrent technique for characterizing low-frequency essential dynamics upon the massive molecular dynamics simulations (MD) data [52] , [53] , [54] , [55] , [56] . We applied it to our massive (28.8 µs in total) conventional MD data to investigate the meaningful movements related to biological function and search the near attack conformations (NACs).…”