Ohlenschläger O et al. (2012), The N-terminus of the human RecQL4 helicase is ...

XB-ART-45563

Nucleic Acids Res 2012 Sep 01;4017:8309-24. doi: 10.1093/nar/gks591.

Show Gene links Show Anatomy links

The N-terminus of the human RecQL4 helicase is a homeodomain-like DNA interaction motif.

Ohlenschläger O , Kuhnert A , Schneider A , Haumann S , Bellstedt P , Keller H , Saluz HP , Hortschansky P , Hänel F , Grosse F , Görlach M , Pospiech H .

???displayArticle.abstract???
The RecQL4 helicase is involved in the maintenance of genome integrity and DNA replication. Mutations in the human RecQL4 gene cause the Rothmund-Thomson, RAPADILINO and Baller-Gerold syndromes. Mouse models and experiments in human and Xenopus have proven the N-terminal part of RecQL4 to be vital for cell growth. We have identified the first 54 amino acids of RecQL4 (RecQL4_N54) as the minimum interaction region with human TopBP1. The solution structure of RecQL4_N54 was determined by heteronuclear liquid-state nuclear magnetic resonance (NMR) spectroscopy (PDB 2KMU; backbone root-mean-square deviation 0.73 Å). Despite low-sequence homology, the well-defined structure carries an overall helical fold similar to homeodomain DNA-binding proteins but lacks their archetypical, minor groove-binding N-terminal extension. Sequence comparison indicates that this N-terminal homeodomain-like fold is a common hallmark of metazoan RecQL4 and yeast Sld2 DNA replication initiation factors. RecQL4_N54 binds DNA without noticeable sequence specificity yet with apparent preference for branched over double-stranded (ds) or single-stranded (ss) DNA. NMR chemical shift perturbation observed upon titration with Y-shaped, ssDNA and dsDNA shows a major contribution of helix α3 to DNA binding, and additional arginine side chain interactions for the ss and Y-shaped DNA.

???displayArticle.pubmedLink??? 22730300
???displayArticle.pmcLink??? PMC3458545
???displayArticle.link??? Nucleic Acids Res

Species referenced: Xenopus
Genes referenced: lgals4.2 recql4 spr topbp1

???attribute.lit??? ???displayArticles.show???

	Figure 1. The Sld2-homologous N-terminal domain of RecQL4 interacts directly with TopBP1. (A) Schematic depiction of human RecQL4 and TopBP1. (B) Specific interaction of the N-terminal domain of RecQL4 fused to the GAL4 DNA-BD with C-terminal part (a.a. 793–1522) of TopBP1 fused to the GAL4-AD in yeast. Interaction was semi-quantitatively assessed by β-galactosidase liquid culture activity assays. β-galactosidase values represent the mean and standard deviations of three independent experiments. (C) GST pull-down experiments performed with purified GST, GST–TopBP1 and GST–TopBP1 (a.a. 793–1522) and [35S]–methionine–labelled RecQL4 (a.a. 1–675) (lower panel). Coomassie brilliant blue R250-stained SDS–PAGE gels of the purified GST constructs are presented (upper panel). (D) GST pull-down experiments performed with purified GST and GST–RecQL4 (a.a. 1–54) and [35S]-cysteine-labelled full-length TopBP1. Coomassie brilliant blue R250-stained SDS–PAGE gels of the purified GST constructs are presented (upper panel). Input IVT: 10 µl of the in vitro translation product. (E) Reciprocal immunoprecipitation of RecQL4 and TopBP1. Cell extracts from human Hek293 cells transiently co-transfected with expression vectors for RecQL4 and ToPBP1 were subjected to immunoprecipitation using antibodies against RecQL4 and TopBP1 as described in ‘Material and Methods’. The precipitates were then analysed by SDS–PAGE and Western blot against the cognate proteins. Controls without antibody or with non-specific rabbit IgG served as negative control and 10% of the input served as an indicator for the (co-)immunoprecipitation efficiency.
	Figure 2. Characterization of the TopBP1 interaction with RecQL4. (A) Identification of the RecQL4_N54 BD of TopBP1 by GST pull-down. Pull-down experiments were performed as in Figure 1C–D with GST-fusion constructs as bait for in vitro translated proteins as indicated. Only TopBP1 (a.a. 1233–1522) shows efficient binding to GST-RecQL4_N54. Purified GST and GST-TopBP1 (a.a. 1233–1264) stained with Coomassie brilliant blue R250 are shown on the right. (B) Real-time in vitro SPR binding analysis of RecQL4_N54 to anti-GST antibody captured GST-TopBP1 (a.a. 1233–1522). Sensorgrams of 200, 100, 50, 25, 12.5, 6.25 and 3.13 µM RecQL4_N54 binding injected in triplicate (black lines) are shown overlaid with the best fit derived from a 1:1 interaction model (red lines) (left). Fit of the equilibrium data for TopBP1 (a.a. 1233–1522) binding (middle). The purified GST-TopBP1 (a.a. 1233–1593) used for the experiment stained with Coomassie brilliant blue R250 is shown on the right.
	Figure 3. NMR solution structure of RecQL4_N54. (A) Stereo view (side-by-side) of the structure closest to the mean. Heavy atom colouring: C (grey), N (blue), O (red) and S (yellow). Helices are indicated by an orange/yellow ribbon. Amino acid type and residue numbers for the start and end residue of the helical elements are annotated. (B) Superimposition of the 20 calculated structures with the lowest target function, backbone in magenta, heavy atom colouring as in A. (C) Aromatic core, the side chains of W16, F20, Y44 and Y47 are emphasized in red, aspect rotated with respect to (B) to display the core.
	Figure 4. Superimposition of the NMR solution structure of RecQL4_N54 (orange ribbon) with homeodomains (blue ribbon). (A) Antennapedia homeodomain (PDB code 1AHD; backbone r.m.s.d. 1.87 Å), (B) Engrailed homeodomain (1ENH; 1.86 Å). (C) MATa1/MATα2 homeodomain (1YRN; 1.76 Å). (D) Homeodomain from the SCR/EXD complex (2R5Z; 1.79 Å). Homeodomains in (A), (C) and (D) are in the DNA-bound form, DNA not shown. (E) Structure-based sequence alignment of RecQL4_N54 against the depicted homeodomains. The RecQL4_N54 construct used for structure determination starts with two non-native residues derived from the GST tag after thrombin digest. Numbering of the a.a. positions refers to the construct and not the native RecQL4 sequence to maintain consistency with the related PDB (2KMU) and BMRB (16544) database entries. Therefore, the numbering of the residues is shifted by two compared to the native protein.
	Figure 5. RecQL4_N54 binds dsDNA without apparent sequence preference. (A) Electrophoretic mobility gel shift assay were performed with RecQL4_N54 (2 µg/30 µM) and DNA fragments (200 ng) covering the sites of DNA replication initiation of three well-characterized human origins as well as three control fragments from proximal sites. Details on the DNA fragments are provided in Supplementary Table S1. The DNA binding buffer was supplemented with 100 mM KCl where indicated. (B) Binding of DNA fragments derived from the LB2 origin of replication (42) by RecQL4 was evaluated by a DNA bead binding assay. Biotinylated dsDNA was pre-incubated with RecQL4 following binding to magnetic SA beads immobilized to magnetic columns. The flow through (FT), the first wash (W) and the elution (E) fractions were analysed by SDS–PAGE. RecQL4_N54 is detected in the elution fractions of all samples containing DNA indicative for DNA binding irrespective of fragment length. The protein band corresponding to RecQL4_N54 is indicated by an arrow. The additional band in the elution (denoted by an asterisk) is SA released from the magnetic bead during sample preparation for SDS–PAGE. Input represents purified RecQL4_N54. The positions of the molecular weight markers (in kDa) are indicated on the right.
	Figure 6. Interaction of RecQL4_N54 with DNA. (A) [1H,15N]–HSQC spectrum (15N chemical shifts in F1 and 1H chemical shifts in F2) of RecQL4_N54 (170 µM) without (black contours) and with 3-fold excess (red contours) of dsDNA. The inset (upper left) depicts an overlay for selected cross-peaks (boxed in spectra) of spectra recorded with protein:DNA ratios of 1:0 (green contours), 1:0.5 (red), 1:1 (magenta), 1:2 (blue) and 1:3 (coral). Residue assignments for the inset are indicated, a fully assigned [1H,15N]–HSQC is given in the Supporting Information. (B) Chemical shift changes for each backbone amide group of RecQL4_N54 at 3-fold molar excess of the 24 mer dsDNA. The chemical shift change was determined for the 15N/1H cross-peak of the backbone amide group of each residue in the spectrum of free RecQL4_N54 to the nearest 15N/1H cross-peak in the spectrum of the RecQL4_N54:DNA complex using the relationship \|Δδ(1H)\| + \|Δδ(15N)\|/7 (45). Values above the average chemical shift perturbation of ∼0.05 ppm are depicted in red. The X for Ala15 indicates that the corresponding cross-peak was broadened beyond detection at 3-fold molar excess of dsDNA. The cyan bars on the top indicate the positions of the α-helices in RecQL4_N54. Note that the first native RecQL4 residue is M3. The numbering of the a.a. positions in RecQL4_N54 starts with two non-native residues derived from the expression construct to maintain consistency with the related PDB (2KMU) and BMRB (16544) database entries. Therefore, the numbering of the residues is shifted by two compared to the native protein. (C) Left panel: Cyan balls at the position of the nitrogen atom highlight the residues in RecQL4_N54 for which chemical shift perturbation above average was observed. Centre panel: For comparison, the Antennapedia homeodomain–DNA complex [PDB code 1AHD; (35)] is presented. Right panel: Sequence comparison between RecQL4_N54 and Antennapedia. Orange: Residues interacting with the phosphate/sugar backbone of the DNA (p) or contributing in base recognition interactions (b) in the Antennapedia homeodomain–DNA complex. Cyan: Residues experiencing significant chemical shift perturbation in RecQL4_N54.
	Figure 7. RecQL4_N54 binding to Y-shaped DNA. (A) RecQL4_N54 prefers binding to Y-shaped DNA over ssDNA and dsDNA. Left panel: 30 µM RecQL4_N54 was preincubated in the presence of 2 µM of the indicated DNA and the DNA–protein complexes resolved by electrophoretic mobility gel shift assay. Middle panel: ssDNA binding—indicated amounts of RecQL4_N54 were preincubated with 1.3 µM oligomer LB2-F carrying a 5′-fluorescein, and the DNA–protein complexes resolved by electrophoretic mobility gel shift assay. Right panel: Binding to Y-shaped DNA—indicated amounts of RecQL4_N54 were preincubated with 2 µM LB2-derived Y-shaped DNA, and the DNA–protein complexes resolved by electrophoretic mobility gel shift assay. The DNA binding buffer was supplemented with KCl as indicated. Details for the DNA fragments are provided in Supplementary Table S1. (B) [1H,15N]–HSQC spectra (15N chemical shifts in F1 and 1H chemical shifts in F2) of RecQL4_N54 (200 µM) without (black contours) and with 3.5-fold excess (red contours) of Y-shaped DNA. (C) [1H,15N]-HSQC spectra of RecQL4_N54 (200 µM) without (black contours) and with 2-fold excess (red contours) of ssDNA. (D) Real-time binding analysis of RecQL4_N54 to Y-shaped, dsDNA and ssDNA. Fits of the biolayer interferometry steady-state binding of RecQL4_N54 using a HEPES buffer containing 50 mM NaCl.

References [+] :

Altschul, Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. 1997, Pubmed