Noncognate DNA damage prevents the formation of the active conformation of the Y-family DNA polymerases DinB and DNA polymerase κ

Abstract

Y-family DNA polymerases are specialized to copy damaged DNA, and are associated with increased mutagenesis, owing to their low fidelity. It is believed that the mechanism of nucleotide selection by Y-family DNA polymerases involves conformational changes preceding nucleotidyl transfer, but there is limited experimental evidence for such structural changes. In particular, nucleotide-induced conformational changes in bacterial or eukaryotic Y-family DNA polymerases have, to date, not been extensively characterized. Using hydrogen-deuterium exchange mass spectrometry, we demonstrate here that the Escherichia coli Y-family DNA polymerase DinB and its human ortholog DNA polymerase κ undergo a conserved nucleotide-induced conformational change in the presence of undamaged DNA and the correct incoming nucleotide. Notably, this holds true for damaged DNA containing N(2) -furfuryl-deoxyguanosine, which is efficiently copied by these two polymerases, but not for damaged DNA containing the major groove modification O(6) -methyl-deoxyguanosine, which is a poor substrate. Our observations suggest that DinB and DNA polymerase κ utilize a common mechanism for nucleotide selection involving a conserved prechemical conformational transition promoted by the correct nucleotide and only preferred DNA substrates.

Keywords: DNA replication; conformational change; hydrogen exchange mass spectrometry; nucleotide selection; substrate specificity.

Fig. 1

Local deuterium uptake for DinB and pol κ in the absence of substrates. Peptide-level HDX data for DinB (top) and pol κ (bottom) are mapped together with atomic-level structural information to visualize the intrinsic local HDX of the free polymerases in solution. For each time point, the percent deuterium uptake for each peptide is displayed as a color gradient on the respective ternary crystal structures (DinB, PDB: 4IRC; pol κ, PDB: 2OH2). No correction for deuterium back-exchange was performed (see Materials and Methods). The results are the average of three replicate experiments. DNA and nucleotides are not shown. The part of the pol κ palm loop insertion that is unresolved in the crystal structure (residues 225–281) is shown as a dashed line and colored according to its relative percent deuterium uptake. Portions of this figure were previously published in Nevin, et al. [43] and are reprinted here with permission of Elsevier.

Fig. 2

Substrate-dependent HDX in pol κ N-clasp domain. (a) Deuterium uptake for pol κ N-clasp peptides as a function of time for free pol κ (black diamonds), pol κ-DNA (gray triangles, dashed line), pol κ-DNA-dGTP_incorrect (blue squares), and pol κ-DNA-dCTP_correct (green circles). Deuterium levels are not corrected for back-exchange and the theoretical maximum number of deuterated amides in each peptide equals its number of residues after subtracting one for the N-terminus and one for each proline residue and is set as the maximum on the Y axis. Error bars represent variation of triplicate experiments. (b) Peptides highlighted on the crystal structure of pol κ (PDB: 2OH2) according to the colors indicated in (a).

Fig. 3

DinB and pol κ are protected from exchange in the presence of undamaged DNA and the correct dNTP. (a) Minimal model for nucleotide incorporation. The free pol is denoted E, the binary pol-DNA complex ED, and the ternary pol-DNA-dNTP complex EDN. The states in the box were probed by HDX-MS. (b) Differences in deuterium uptake between the indicated DinB states. The templating base was G and the incorrect and correct dNTPs were dGTP and dCTP, respectively. For each peptide and time point, the relative deuterium level of the second state was subtracted from the first and displayed as a color gradient. The data are the average at least two replicate experiments. Differences >0.6 Da are significant at 98% confidence. (c and d) Regions of DinB with deuterium uptake differences between the indicated states of at least 1 Da at any time point displayed as a color gradient on ternary complex crystal structure (PDB: 4IRC). In these images, brown indicates regions for which no data were obtained. (e) Differences in deuterium uptake between the indicated pol κ states calculated and displayed as in (b). (f and g) Regions of pol κ with deuterium uptake differences between the indicated states of at least 1 Da at any time point displayed as a color gradient on ternary complex crystal structure (PDB: 2OH2).

Fig. 4

Thermodynamic stability of DinB and pol κ in different states. Melting transition midpoints (T_m) were determined from thermal denaturation transitions of DinB and pol κ in complex with the indicated substrates. The templating base was G. The data are the average of three experiments and error bars represent ±SD.

Fig. 5

Only preferred template lesions and correct dNTPs promote “active” conformation. (a) Primer extension opposite G, N²fG, and O⁶meG by DinB (top) and pol κ (bottom). Reactions were quenched after 0, 5, 30, and 60 min and analyzed by 16% denaturing polyacrylamide gel electrophoresis and subsequent phosphorimaging (b) Average differential deuterium uptake between indicated states of DinB (top) and pol κ (bottom) at each indicated time point (0.17, 1, 10, 60, 240 min) for peptides (same as in Figure 3) displayed as a color gradient. Differences >0.6 Da are significant at 98% confidence. For each peptide and time point, the relative deuterium level of the second state was subtracted from the first. The templating base and the incoming dNTP are indicated.

Fig. 6

Proposed reaction pathway for DinB and pol κ. The free pol binds DNA which results in a relatively closed and stable binary complex. In the case of correct dNTP incorporation opposite preferred DNA bases, there is a conserved conformational change towards a relatively closed and “active” ternary complex resulting in fast nucleotide incorporation. In the case of incorrect dNTP incorporation or non-preferred DNA, a non-productive ternary complex is favored, resulting in a slower reaction.