Solution structure of a multifunctional DNA- and protein-binding motif of human Werner syndrome protein

Abstract

Werner syndrome (WS) is an autosomal recessive disease that results in premature aging. Mutations in the WS gene (WRN) result in a loss of expression of the WRN protein and predispose WS patients to accelerated aging. As a helicase and a nuclease, WRN is unique among the five human RecQ helicase family members and is capable of multiple functions involved in DNA replication, repair, recombination, and telomere maintenance. A 144-residue fragment of WRN was previously determined to be a multifunctional DNA- and protein-binding domain (DPBD) that interacts with structure-specific DNA and a variety of DNA-processing proteins. In addition, DPBD functions as a nucleolar targeting sequence of WRN. The solution structure of the DPBD, the first of a WRN fragment, has been solved by NMR. DPBD consists of a winged helix-like motif and an unstructured C-terminal region of approximately 20 aa. The putative DNA-binding surface of DPBD has been identified by using known structural and biochemical data. Based on the structural data and on the biochemical data, we suggest a surface on the DPBD for interacting with other proteins. In this structural model, a single winged helix domain binds to both DNA and other proteins. Furthermore, we propose that DPBD functions as a regulatory domain to regulate the enzymatic activity of WRN and to direct cellular localization of WRN through protein-protein interaction.

Fig. 1.

Secondary structure of the DPBD of WRN and sequence alignments of WT DPBD with five WS-associated mutants. WS1, IVS25–1G→ C mutation in the last base of intron 25 (4); WS2, IVS26+1G→ C mutation in the first base of intron 26 (10). WS1 and WS2 give an identical protein, and only WS1 is shown. WS3, 3265-3266delGA mutation in exon 25 (10); WS4, 3259–3262delCAAA mutation in exon 25 (10); WS5, 3004delG mutation in exon 23 (11). The modified sequences in the mutants are highlighted in pink.

Fig. 2.

Stereoview of the ensemble of the 10 simulated annealing structures of DPBD. The backbone coordinates of residues 955-1068 are superimposed. The ensemble within the box displays all 144 residues and illustrates the disordered N and C termini.

Fig. 3.

Ribbon diagrams of the DPBD and structurally homologous proteins. (a and b) Ribbon diagrams showing two orthogonal views of DPBD. The putative DNA-recognition helix (H4) is shown in red, and other helices are shown in purple, green, and dark blue. The β-strands are shown in blue. (c–e) Comparison of the WH-like fold topology of the DPBD with the crystal structures of the WH domains of E. coli RecQ (c), of MarR (d), and of BlaI (e). The known DNA-recognition helix of BlaI and the putative DNA-recognition helices of other WH domains are shown in red and are marked with H; other helices are shown in purple and green. The β-strands are shown in blue.

Fig. 4.

Structure-based sequence alignments. Shown are structure-based sequence alignments of the DPBD with the most structurally homologous WH-containing proteins (RecQ, MarR, and BlaI) and the WH-containing proteins with known protein–DNA cocrystal structures (BlaI, RTP, DtxR, D2P, and CAP). The sequence number for the first residue on each line is indicated on the left. Sequence regions that do not align with the DPBD are not shown. Residues that are known to contact DNA in the protein–DNA complexes are indicated in red. Bold red type in the DPBD sequence of WRN indicates a residue previously identified (17) to interact with DNA. The putative DNA-binding residues of DPBD are highlighted in green.

Fig. 5.

Potential and identified DNA-binding residues in the DPBD of WRN. The residues with positive charged potential on the putative DNA-binding face are shown in blue, and residues with negative charged potential, and likely involved in hydrogen bonding on the same face, are shown in red. The residues of E1012 and S1013 are not shown for clarity. The DNA-binding residue K1016 previously identified (17) is shown in bold blue with an asterisk.

Fig. 6.

Model for the regulation of the WRN enzyme by DPBD. Only three functional domains, helicase core (HC), putative Zn-binding domain (ZBD), and DPBD, of WRN are shown. The active site of HC is shown. The DPBD has the strongest DNA-binding pocket of WRN, which collaborates with ZBD to form an even stronger DNA-binding site to hold on to the upstream duplex region of a DNA substrate, whereas HC unwinds at the fork junction of the DNA substrate. The DPBD–DNA interaction dictates a specific DNA processing pathway. The protein–protein interaction between DPBD and the regulatory proteins as discussed in the text directs the cellular localization of WRN or regulates the enzymatic activity of HC.