Solution structure and DNA-binding properties of the C-terminal domain of UvrC from E.coli

Abstract

The C-terminal domain of the UvrC protein (UvrC CTD) is essential for 5' incision in the prokaryotic nucleotide excision repair process. We have determined the three-dimensional structure of the UvrC CTD using heteronuclear NMR techniques. The structure shows two helix-hairpin-helix (HhH) motifs connected by a small connector helix. The UvrC CTD is shown to mediate structure-specific DNA binding. The domain binds to a single-stranded-double-stranded junction DNA, with a strong specificity towards looped duplex DNA that contains at least six unpaired bases per loop ("bubble DNA"). Using chemical shift perturbation experiments, the DNA-binding surface is mapped to the first hairpin region encompassing the conserved glycine-valine-glycine residues followed by lysine-arginine-arginine, a positively charged surface patch and the second hairpin region consisting of glycine-isoleucine-serine. A model for the protein-DNA complex is proposed that accounts for this specificity.

Fig. 1. (A) Domain organization of the UvrC protein. (B) Sequence alignment of HhH domains of UvrC, ERCC1, RuvA, DNA ligase, DNA polymerase β and HL5. H1, H2, H3 and H4 indicate the presence of an α-helix, Hc represents the presence of a connector helix, and h1 and h2 are the hairpin loops. Conserved regions are indicated in boxes.

Fig. 2. EMSA assay of UvrC CTD binding to different DNA substrates. (A) DNA binding of 1 µM UvrC CTD with dsDNA (10, 20 and 28 bp), bubble DNA, hairpin DNA, ssDNA, a fork with unpaired bases at the 3′ and 5′ end, and a fork with unpaired bases at the 3′ end. The type of DNA and the size of the unpaired bases (T stretch) are indicated schematically at the top of the gel. Free and bound represent the unbound DNA and the DNA in complex with UvrC CTD, respectively. (B) Quantification of three representative binding experiments using the bubble with an 8 bp spacer as a probe. The average fraction bound and the SD was determined for the indicated UvrC CTD concentrations (circles with error bars). Using these data, we calculated the theoretical binding curves assuming respectively one (n = 1, dashed line), two (n = 2, small dashed line) or four (n = 4, dotted line) UvrC CTD molecules per DNA substrate, as described in Materials and methods. The inset shows one of these experiments with 0.08, 0.16, 0.31, 0.625, 1.25 and 5.0 µM UvrC CTD. (C) Off-rate determination. Binding competition by the addition of a 200-fold molar excess of unlabelled homologous DNA at the indicated time points (in min) of a pre-formed complex of UvrC (3 µM) with a bubble with eight unpaired bases. –, no protein added to the binding mixture; 0, no competitor added to the protein–DNA complex. (D) Graphical representation from a gel filtration experiment showing the time after elution from the column (in min) against the molecular weight, calibrated using BSA (66 kDa), ovalbumin (46 kDa), carbonic anhydrase (30 kDa), cytochrome c (12 kDa) and aprotinin (6 kDa) (these proteins are indicated by black dots). The average molecular weight and SD for UvrC (open circle), bubble DNA with 10 unpaired bases (open triangle) and protein–DNA complex (open square) was determined by measuring the elution time of two injections for both the references and the indicated samples.

Fig. 3. The NMR solution structure of the UvrC CTD. (A) Backbone stereo view (residues 28–78) of the NMR ensemble (22 structures); the hairpins are coloured in blue. (B) Ribbon view of a representative UvrC CTD structure (closest to average) for residues 23–78. h1 and h2 are the hairpins of HhH motifs. The structures were displayed using the molecular graphics program MOLMOL (Koradi et al., 1996).

Fig. 4. (A) NMR restraints statistics. (B) Backbone r.m.s.d. (C) Backbone {¹H}–¹⁵N NOEs as a function of residue number. The NOE restraints are intra-residue NOEs, inter-residue short-range sequential NOEs (|i – j| = 1), inter-residue medium-range NOEs (1 < |i – j| < 5) and long-range NOEs (|i – j| > 4) from bottom to top. The boxes at the top indicate the inclusion of TALOS-derived φ and χ angle restraints for that residue. The residue r.m.s.ds were calculated over the ensemble of 22 structures after superposition of residues 36–74.

Fig. 5. Comparison of the overall topologies of the HhH domain of (A) UvrC in blue, (B) RuvA in red, (C) DNA ligase in green and (D) an overlay of all three. The hairpin loops are coloured purple and the helix H1 was removed for the sake of clarity. The backbone r.m.s.d. of UvrC CTD with RuvA is 1.63 Å and with DNA ligase is 1.47 Å. The structures were displayed using the software Molscript and Raster 3D (Esnouf, 1997).

Fig. 6. (A) Plot of chemical shift value changes (the r.m.s.d. of chemical shift changes in backbone amide nitrogen and amide proton) upon titration of UvrC CTD and the bubble DNA versus residue number. The absence of a bar in the plot indicates the presence of a proline residue or an unmeasured shift due to overlap. (B) Part of the HSQC spectrum displaying changes in the chemical shift values of Gly31, Lys35, Met39 and Glu30 residues, respectively. The peak position corresponds to 0, 0.032, 0.068, 0.13, 0.5 and 1.0 equivalent of DNA to UvrC CTD (increasing from light grey to black).

Fig. 7. Model representing the interaction of the UvrC CTD with the ds–ss junction. Glycines are coloured red, lysine in blue and threonine in green. This model was obtained by superimposing the two hairpins of UvrC CTD on to the corresponding loops of RNA polymerase II (domain containing the active site of the Rpb1 subunit, PDB accession No. 16IH, see Materials and methods). The structure was generated using the software VMD (Humphrey et al., 1996).