Pre-steady-state binding of damaged DNA by XPC-hHR23B reveals a kinetic mechanism for damage discrimination

Abstract

The XPC-hHR23B complex (XPC-hHR23B) is a heterodimeric protein required for the initial step of DNA damage recognition in the global nucleotide excision repair (NER) pathway. A strong preference of XPC-hHR23B for UV- and cisplatin-damaged DNA has previously been demonstrated using equilibrium binding assays. To better understand the molecular mechanism of damage recognition by XPC-hHR23B, we carried out the pre-steady-state kinetic analysis of the XPC-hHR23B-DNA interactions using a stopped-flow fluorescence assay. XPC-hHR23B displays a faster k(on) for cisplatin- and UV-damaged duplex DNA than for undamaged DNA, with additional, minor effects on the k(off) rates. XPC-hHR23B has a high affinity for undamaged single-stranded DNA compared to duplex DNA, which can be largely attributed to a high rate of association. However, cisplatin damage on single-stranded DNA reduced the overall level of binding by a factor of 7, with nearly equal contributions from changes to the k(on) and k(off) rates. Together, these results support a model for initial damage recognition by XPC-hHR23B that is dependent on structural changes in the DNA, and not adduct chemistry.

Figure 1

Purity of the XPC–hHR23B protein preparation. XPC–hHR23B was expressed in insect cells infected with recombinant baculovirus and purified by column chromatography as described in Materials and Methods. Samples from each step of the purification (2 μg of total protein) were subjected to SDS–PAGE and were detected by silver stain analysis: lane 1, cell extract; lane 2, P-cell fraction; lane 3, ssDNA fraction; and lane 4, H–S fraction.

Figure 2

XPC–hHR23B equilibrium binding to undamaged and cisplatin-damaged DNA. Binding assays were performed with varying concentrations of XPC–hHR23B and 20 fmol of undamaged or cisplatin-treated single-stranded (A and B) or duplex DNA (C and D). (A and C) XPC–hHR23B–DNA complexes were separated from unbound DNA by 4% native PAGE. (B and D) XPC–hHR23B bound to undamaged (○) and cisplatin-damaged (●) single-stranded DNA (B) or undamaged (□) and cisplatin-damaged (■) duplex DNA (D) was quantified by PhosphorImager analysis and plotted as a function of total XPC–hHR23B concentration. Error bars represent the mean and standard deviations for three independent experiments.

Figure 3

Pre-steady-state kinetics of binding of XPC–hHR23B to undamaged and cisplatin-damaged single-stranded DNA substrates. Reaction mixtures were excited at 290 nm, and fluorescence was monitored via emission at 340 nm. Traces shown are the average of at least eight shots and fit to single-exponential decay functions. (A) XPC–hHR23B (10 nM) was mixed with buffer (top trace) or with 25 nM undamaged 75-mer DNA (bottom trace) and fit to a single-exponential decay function. The top trace was offset by 0.07 for clarity. (B) Undamaged (○) and cisplatin-damaged (●) single-stranded DNA was titrated with 10 nM XPC–hHR23B, and the observed rate of quenching (k_obs) was plotted vs the DNA concentration and fit to a straight line. The slope of the line provides the bimolecular rate constant, k_on, and the y-intercept provides the rate of dissociation, k_off. Each point represents the mean and standard deviation from three independent experiments.

Figure 4

Pre-steady-state kinetics of binding of XPC–hHR23B to undamaged or UV- or cisplatin-damaged duplex DNA substrates. Reaction mixtures were excited at 290 nm, and fluorescence was monitored via emission at 340 nm. (A) XPC–hHR23B (10 nM) was mixed with 100 nM dsDNA that was undamaged (—), cisplatin-damaged (- - -), or UV-damaged (···). Traces are the average of at least eight shots and fit to single-exponential decay functions. For clarity, the intrinsic fluorescence for the UV-damaged and cisplatin-damaged data was offset by 0.0069 and 0.0141, respectively. The observed rates of binding were 1.4 ± 0.5, 23.7 ± 2.6, and 8.4 ± 1.2, respectively. (B) Kinetic traces were measured at a constant XPC–hHR23B concentration (10 nM) and varying concentrations of duplex DNA that was undamaged (○), cisplatin-damaged (●), or UV-damaged (■). The observed rate of quenching (k_obs) was plotted vs the DNA concentration and fit to a straight line. Each point represents the mean and standard deviation from at least three independent experiments.

Figure 5

Pre-steady-state kinetics of binding of XPC–hHR23B to duplex damaged DNA containing 1,2 d(GpG) and 1,3 d(GpG) site-specific cisplatin adducts. Kinetic traces were measured at a constant XPC–hHR23B concentration (10 nM) and varying concentrations of duplex DNA with a 1,2 d(GpG) adduct (○) or a 1,3 d(GpXpG) adduct (●). The observed rate of quenching (k_obs) was plotted vs the DNA concentration and fit to a straight line.

Figure 6

Fluorescence anisotropy measurements of binding of XPC–hHR23B to undamaged and cisplatin-damaged duplex DNA. XPC–hHR23B was titrated with 5 nM fluorescently labeled DNA substrate: (A) undamaged LH10 (75-mer), (B) 1,2 d(GpG) cisplatin adduct (60-mer), and (C) 1,3 d(GpXpG) cisplatin adduct (60-mer). Each point represents the mean and standard deviation from three independent experiments.