JBC Anatrace, Inc.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M501083200 on June 1, 2005

J. Biol. Chem., Vol. 280, Issue 31, 28644-28652, August 5, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/31/28644    most recent
M501083200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Choi, Y.-J.
Right arrow Articles by Choi, B.-S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Choi, Y.-J.
Right arrow Articles by Choi, B.-S.

Biophysical Characterization of the Interaction Domains and Mapping of the Contact Residues in the XPF-ERCC1 Complex*

Yun-Jeong Choi{ddagger}§, Kyoung-Seok Ryu{ddagger}, Yun-Mi Ko{ddagger}, Young-Kee Chae¶, Jeffrey G. Pelton||, David E. Wemmer||**, and Byong-Seok Choi{ddagger}{ddagger}{ddagger}

From the {ddagger}Department of Chemistry and National Creative Research Initiative Center, Korea Advanced Institute of Science and Technology, 373-1 Guseong-dong, Yuseong-gu, Daejon 305-701, Korea, the Department of Applied Chemistry, Sejong University, 98 Gunja-Dong, Gwangjin-Gu, Seoul 143-747, Korea, and ||Physical Biosciences Division, Lawrence Berkeley National Laboratory and Department of Chemistry, University of California, Berkeley, California 94720

Received for publication, January 31, 2005 , and in revised form, April 25, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 
XPF and ERCC1 exist as a heterodimer to be stable and active in cells and catalyze DNA cleavage on the 5'-side of a lesion during nucleotide excision repair. To characterize the specific interaction between XPF and ERCC1, we expressed the human ERCC1 binding domain of XPF (XPF-EB) and the XPF binding domain of ERCC1 (ERCC1-FB) in Escherichia coli. Milligram quantities of a heterodimer were characterized with gel filtration chromatography, an Ni2+-NTA binding assay, and analytical ultracentrifugation. Cross-linking experiments at high salt concentrations revealed that XPF interacts with ERCC1 mainly through hydrophobic interactions. XPF-EB was also shown to homodimerize in the absence of ERCC1. NMR cross-saturation methods were applied to map the residues involved in formation of the XPF-EB·XPF-EB homodimer and the XPF-EB·ERCC1-FB heterodimer. Helix H3 and the C-terminal region of XPF-EB were either within or in close proximity to the homodimer interface, whereas the ERCC1-FB binding site of XPF-EB was distributed across helix H1, a small part of H2, H3, and the C-terminal region, most of which exhibited large changes in chemical shift upon ERCC1 binding. The XPF-EB heterodimeric interface is larger than the XPF-EB homodimeric one, which could explain why XPF has a stronger affinity for ERCC1 than for a second molecule of XPF. The XPF binding sites of ERCC1 were located in helices H1 and H3 and in the C-terminal region, similar to the involved surface of XPF. We used cross-saturation data and the crystal structure of related proteins to model the two complexes.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 
DNA lesions, which interfere with DNA replication, transcription, and recombination (1), occur as a result of exogenous and endogenous damaging agents, such as UV irradiation, products of oxidative stress, and DNA-reactive chemicals. To maintain the integrity of the genome, DNA lesions are efficiently removed by DNA repair pathways, such as the nucleotide excision repair pathway, which removes various types of bulky DNA adducts. During nucleotide excision repair, the 5' and 3' sides of a lesion are cut by two endonucleases, the XPF-ERCC1 heterodimeric complex and XPG, respectively (2-5). This dual incision is made asymmetrically around a lesion to allow its release as part of a larger DNA fragment (24-32 nucleotides) (6-8). The remaining gap is filled by DNA synthesis and ligation (9, 10).

The XPF-ERCC1 complex cuts a variety of DNA substrates, including bubbles, stem-loops, splayed arms, and flaps (2, 4, 11, 12). XPF contains the endonuclease activity, while ERCC1 likely modulates the specificity required for DNA incision (13). XPF and ERCC1 form a stable heterodimer in mammalian cells and in vitro, and various lines of evidence suggest that both ERCC1 and XPF are unstable in the absence of the respective partner of each (14-18). The XPF-ERCC1 heterodimer is known to be formed through the interaction of their C-terminal domains, which each include two helix-hairpin-helix (HhH)1 motifs (19-21). However, a detailed characterization of the XPF-ERCC1 heterodimeric interface has not yet been performed.

To more fully characterize the mode of interaction between ERCC1 and XPF, the human ERCC1 binding domain of XPF (XPF-EB) and XPF binding domain of ERCC1 (ERCC1-FB) were expressed in Escherichia coli, and 1-mg quantities of a soluble form of the XPF-EB·ERCC1-FB complex were generated and characterized. To decipher the nature of the XPF-ERCC1 interaction in detail, we used the soluble, recombinant complex in NMR cross-saturation experiments to map the amino acid residues that make contact between the two subunits. This enabled us to study the XPF-ERCC1 interaction at the molecular level and to propose a model for XPF-ERCC1 complex formation. Our results provide insight into the mechanisms by which XPF and ERCC1 interact specifically and tightly and how this binding contributes to the stabilization of the subunits.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
 REFERENCES
 
Cloning, Expression, and Purification of XPF-EB1, XPF-EB2, XPF-EB3, and ERCC1-FB—The DNA coding sequences for XPF-EB1 (amino acids 813-905) and ERCC1-FB (amino acids 224-297) were subcloned into two versions of the bacterial expression vector pET21a (Novagen) using NdeI and XhoI sites, one with a C-terminal His tag and one without. The DNA coding sequence for XPF-EB2 (amino acids 838-901) was subcloned, with an N-terminal His tag, into pET100/D-TOPO® (Invitrogen). The plasmid for expression of the glutathione S-transferase (GST)-XPF-EB3 fusion protein (amino acids 813-847) was generated from pGEX-4T-3 (Amersham Biosciences) using the BamHI and XhoI sites. The resulting proteins are shown in Fig. 1. The nucleotide sequences of the various constructs were confirmed by DNA sequencing.

The expression plasmids harboring the genes for the various versions ERCC1 and XPF were transformed into E. coli strains BL21(DE3) and BL21 for large scale protein preparation. To produce the 15N-labeled and 13C/15N-labeled proteins, the bacteria were grown in M9 minimal medium that contained 1 g/liter of 15NH4Cl as the sole nitrogen source and (for double-labeled protein) 2 g/liter 13C-labeled glucose as the carbon source. For uniformly 2H- and 15N-labeled proteins, the cells were grown in M9 minimal medium containing 98% 2H2O, 15NH4Cl, and [2H]glucose (98 atom %). All chromatographic steps were carried out at 4 °C except for the steps carried out in 6 M urea. The soluble His-tagged ERCC1 binding domain of XPF was purified using both a Ni2+-NTA-agarose column (Qiagen) and a gel filtration column. After thrombin cleavage to release the N-terminal His tag, a benzamidine column (Amersham Biosciences) was used to remove the thrombin. GST-XPF-EB3 was purified using a glutathione-Sepharose column (Amersham Biosciences). ERCC1-FB was overexpressed in inclusion bodies, solubilized in resuspension buffer (100 mM sodium phosphate (pH 8), 200 mM NaCl, 1 mM DTT, 6 M urea), and purified using a Sepharose column (16 mm x 100 cm) charged with Sephacryl S-200 resin (200 ml; Amersham Biosciences) under denaturing conditions.

Refolding—Refolding of urea-treated ERCC1-FB, either alone or in the presence of the ERCC1 binding domain of XPF, was performed at 4 °C using two different refolding methods. One method involved stepwise dialysis to remove the denaturing agent (urea) gradually. Proteins were dialyzed sequentially over a period of 8 h against 4, 2, 1, 0.1, and 0 M urea, each in dialysis buffer (100 mM sodium phosphate buffer (pH 8), 1 mM DTT, 100 mM NaCl). A pH of 8.0 was used because disulfide bond formation occurs more rapidly at higher pH than at neutral pH (22); the XPF and ERCC1 binding assays also were performed at a pH near pH 8.0 as in other reports (13, 19). L-Arginine (250 mM) was added to the 1 M urea buffer to facilitate protein refolding during stepwise dialysis (23). The second refolding method used was rapid dilution. The ERCC1-FB protein, either alone or in the presence of the ERCC1 binding domain of XPF (final protein concentration was ~80 µg/ml), was refolded by making 100-fold dilutions in refolding buffer (pH 8, 100 mM sodium phosphate buffer, 1 mM DTT, 100 mM NaCl, 10 mM reduced glutathione, 1 mM oxidized glutathione, and 250 mM L-arginine). Refolding was allowed to proceed for 20 h with stirring. At the end of 20 h, the refolding solution was dialyzed against dialysis buffer. The soluble and insoluble fractions of the final protein solution were separated by centrifugation and subjected to PAGE to identify the fractions enriched for the protein of interest.

Binding Assays Using Ni2+-NTA-agarose—A Ni2+-NTA-agarose column was pre-equilibrated with binding buffer (100 mM sodium phosphate buffer (pH 7.5), 100 mM NaCl, 5 mM imidazole, 0.05% Nonidet P-40, 0.5 mM DTT). The detergent Nonidet P-40 was added to the buffer to prevent nonspecific binding between proteins. After loading the column with protein, it was washed with 10 bed volumes of binding buffer and His tag washing buffer (50 mM imidazole, 100 mM sodium phosphate buffer (pH 7.5), 500 mM NaCl, 0.05% Nonidet P-40, 0.5 mM DTT). The proteins were then eluted with His tag elution buffer (250 mM imidazole, 100 mM sodium phosphate buffer (pH 7.5), 500 mM NaCl, 0.05% Nonidet P-40, 0.5 mM DTT), and the fractions were analyzed by SDS-PAGE and Coomassie Blue staining of the gel.

Binding Assays of GST-XPF-EB3-(813-847) using GST Resin—The GST-XPF-EB3-(813-847) fusion protein and GST were prepared and incubated in the presence of ERCC1-FB during the refolding process described above. A glutathione-Sepharose column was pre-equilibrated with GST-binding buffer (100 mM sodium phosphate buffer (pH 7.5), 100 mM NaCl, 0.05% Nonidet P-40, 1 mM DTT). After loading the column with protein, it was washed with 10 bed volumes of GST-binding buffer. Proteins were eluted with GST elution buffer (10 mM glutathione, 50 mM Tris-HCl, pH 8.0), and fractions were analyzed by SDS-PAGE and Coomassie Blue staining of the gel.

Gel Filtration Chromatography—Gel filtration chromatography (Superdex 75 prep grade (16 mm x 120 cm)) was monitored by measuring absorbance at 280 nm. The column was equilibrated with XPF-ERCC1 binding buffer (50 mM sodium phosphate buffer (pH 6), 500 mM NaCl and 1 mM DTT) before the protein preparation was applied. Molecular size standards were bovine serum albumin (67 kDa), ovalbumin (43.5 kDa), carbonic anhydrase (29 kDa), and cytochrome C (12.4 kDa) (Sigma). Molecular sizes of the recombinant proteins of interest were calculated from a plot of the elution volume versus the log of the molecular size.

Chemical Cross-linking—Refolded heterodimers (20 µM) were allowed to equilibrate at 4 °C in the following buffers: 50 mM sodium phosphate buffer (pH 7), 100 mM NaCl, and 1 mM DTT with various concentrations of NaCl (75, 150, 300, 500, or 1000 mM). Glutaraldehyde is a homobifunctional amine cross-linker that reacts with primary amino groups on proteins and thus has been used as a protein cross-linking reagent. Glutaradehyde was then added to the refolded heterodimer mixture to a final concentration of 0.005%. After incubation for 2 h, the reaction was quenched with a Tris (pH 6.8) solution to a final concentration of 50 mM. Products were analyzed by SDS-PAGE and visualized by Coomassie Brilliant Blue R250 staining.

Analytical Ultracentrifugation—We carried out sedimentation equilibrium ultracentrifugation experiments at 4 °C and 20 °C on a Beckman model XLA analytical ultracentrifuge using an AN-60-Ti rotor. Most experiments were carried out using the standard 12-mm path length, six-channel, charcoal-filled epon cells with quartz windows. Continuous radial scanning at 280 nm was used for both the refolded XPF-EB1·ERCC1-FB complex (3.9, 3.8, and 4.1 mg/ml) and for XPF-EB1 (3 mg/ml). The six channels were scanned at speeds of 18,000 and 20,000 rpm. The cells were scanned every 0.001 cm, and 50 scans were averaged. The standard buffer was XPF-ERCC1 binding buffer. The default value of 1.00017 g/ml was used for the density of the solvent, and a partial specific volume of 0.73 ml/g was calculated from the weight average of the partial specific volumes of individual amino acids.

CD—10 µM protein samples were prepared in XPF-ERCC1 binding buffer, and CD spectra of the proteins were measured on a JASCO J-720 CD spectropolarimeter at room temperature using a cell path length of 1 mm. For urea denaturation experiments, urea was dissolved in XPF-ERCC1 binding buffer, and the samples were incubated in the urea buffer for 12 h prior to analysis by CD.

NMR Spectroscopy—NMR experiments were carried out at 298 K using a 600-MHz Varian Unity Inova spectrometer and a Bruker DMX500 spectrometer with a cryoprobe. A series of three-dimensional triple resonance experiments (HNCA, HN(CO)CA, HNCACB, CBCA-(CO)NH, HNCO, C(CC-TOC-CO)NH, and H(CC-TOC-CO)NH) were recorded for spectral assignments (24). Data were processed and analyzed using the nmrPipe/nmrDraw programs (25). NMR samples contained 1 mM 13C,15N-labeled proteins dissolved in 1H2O/2H2O (9:1, v/v) containing 50 mM phosphate (pH 6), 1 mM DTT, and 500 mM NaCl.

Cross-saturation experiments were carried out at 25 °C according to the procedure of Takahashi et al. (26). The sample contained 1 mM 2H,15N-labeled protein with the nonlabeled interaction partner protein in 1H2O/2H2O (1:9, v/v) that contained 50 mM phosphate (pH 6), 1 mM DTT, and 500 mM NaCl. The WURST-2 decoupling scheme (27) was used for the saturation of aliphatic protons with the irradiation frequency set to 1 ppm. A recycle delay of 3.0 s (1.8-s saturation time) was employed. The results of significantly overlapped residues were excluded when the reduction ratios of the signal intensities were calculated.

Molecular Modeling—The only structure of a dimeric HhH protein reported at the time of this work was that of the Eph receptor sterile {alpha} motif (SAM) domain (28), Protein Data Bank code 1b0x [PDB] . The differences in the residues comprising helix 5 and differences in the regions of contact in the XPF-EB2/ERCC1-FB complex (indicated by cross-saturation experiments) relative to what is seen in the crystal structure indicated that modeling the interface in the XPF-EB2·ERCC1-FB complex based on the SAM domain structure would be problematic. To generate a model for the XPF-EB2·ERCC1-FB complex, a structure of each protein was first generated by carrying out homology modeling based on the Protein Data Bank using MODELLER 6v2 (29). Models for XPF-EB2 and ERCC1-FB were built onto the SAM domain (Protein Data Bank code 1b0x [PDB] ) and C-terminal domain of UvrC (Protein Data Bank code 1kft [PDB] ) as a template with the highest identity and similarity, respectively. Restrained energy minimization was used for portions of the model, which required manual refinement, and the model consistency was assessed with PROCHECK (30). The resulting structures were then used to generate a model for the heterodimer complex using the program AUTODOCK 3.05 (31). The molecule was allowed to move in a cubic box large enough (68-Å edge). The electrostatic potential grid of this cubic box was calculated using a grid spacing of 0.75 Å. Step sizes of 2 Å for translation and 50° for rotation were chosen, and the maximum number of energy evaluations was set to 250,000. For the genetic algorithm, a population size of 50 was used, and the maximum number of generations was set to 27,000. The best complex was chosen according to the lowest energy. After submission of this manuscript, a new structure appeared for an archaeal XPF homodimer (32) including nuclease and (HhH)2 domains. The sequences of human XPF and ERCC1 were threaded to the individual monomers of this model, using Swiss-PDBViewer 4.7, for comparison.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
REFERENCES
 
Aggregated ERCC1-FB Is Refolded Only in the Presence of Its Interaction Partner XPF-EB—XPF-EB1-(813-905) carrying a C-terminal His tag was expressed as a soluble protein in E. coli strain BL21(DE3). In contrast, non-His-tagged ERCC1-FB-(224-297) was expressed exclusively in inclusion bodies in the same E. coli strain (Fig. 1). Although a variety of expression vectors, E. coli strains, and culturing temperatures were tested, we were unable to produce soluble ERCC1-FB. To try to obtain a soluble form of the protein, we disrupted the aggregated ERCC1-FB in resuspending buffer that contained the denaturing reagent urea and then refolded the protein in vitro at 4 °C by stepwise dialysis or rapid dilution of the urea concentration. Although we modulated various factors, including pH, composition of the refolding medium, and redox conditions, the bulk of the ERCC1-FB remained aggregated during the refolding process.



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 1.
Sequence alignment of the HhH domains of ERCC1, UvrC, the SAM domain of the EphA4 receptor tyrosine kinase, and XPF. Residues shown in boldface type are helical residues from published structures of UvrC (Protein Data Bank code 1kft [PDB] ) (gray, poorly defined helix) and SAM (Protein Data Bank code 1b0x [PDB] ) and residues in helix of ERCC1 and XPF (blue, helix) based on NMR chemical shift index. Underlines in the SAM sequence indicate contact residues (<4 Å) between monomers in the dimer structure. {diamond}, residues in the XPF-ERCC1 heterocomplex for which cross-saturation was observed; {circ}, the same for the XPF homerdimer complex.

 



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 2.
A, denaturation profiles of XPF-EB1 as a function of urea concentration. The CD spectra of the proteins at various concentrations of urea were measured. The ellipticities of each spectrum at 222 nm were used to construct the denaturation curves shown in the figure. B, Ni2+-NTA assay for binding of refolded XPF-EB1 and ERCC1-FB. Lane 1, protein molecular size markers (65, 42, 29, and 12.4 kDa); lane 2, refolded XPF-EB1 and ERCC1-FB; lane 3, wash of resin with binding buffer; lane 4, wash of resin with washing buffer; lanes 5 and 6, elution fractions containing XPF-EB1 with a C-terminal His tag (open arrowhead) and ERCC1-FB without tag (solid arrowhead). The asterisks indicate the XPF-ERCC1 dimer formed by intermolecular disulfide linkages due to the lack of DTT in the assay. C, glutaraldehyde cross-linking experiments. The refolded XPF-EB1·ERCC1-FB complex (lanes 1-6) and XPF-EB1 alone (lanes 7-11), each at a total concentration of 20 µM, were incubated with glutaraldehyde for 2 h at room temperature; products were analyzed by SDS-PAGE with Coomassie Blue staining. The sodium chloride concentrations in the cross-linking assay were 0 mM (lane 1), 75 mM (lanes 2 and 7), 150 mM (lanes 3 and 8), 300 mM (lanes 4 and 9), 500 mM (lanes 5 and 10), and 1000 mM (lanes 6 and 11). A reference sample of the unmodified complex (lane 1) served as a control. The solid arrowhead indicates the migration position of the cross-linked XPF-EB1 homodimer. The open arrowheads align with the migration positions of the refolded XPF-EB1·ERCC1-FB complex.

 
We next tried to refold ERCC1-FB in the presence of XPF-EB1 using stepwise dialysis or rapid dilution. The two recombinant proteins were combined in the 6 M urea initial buffer described above, and the mixture was subjected to stepwise dialysis to eliminate the urea. In addition, ERCC1-FB that had been solubilized in the 6 M urea buffer was subjected to rapid dilution with refolding buffer that contained XPF-EB1. After stepwise dialysis or rapid dilution, most of the ERCC1-FB protein, along with the XPF-EB1 protein, was recovered in the soluble fraction, which could be concentrated to yield an ERCC1-FB·XPF-EB1 preparation that was greater than 1 mM. The 1H NMR spectra of the proteins refolded by the two methods were identical (data not shown), indicating that both methods yielded comparably folded proteins in an efficient manner.

To examine the structural stability of XPF-EB1, urea-induced unfolding of XPF-EB1 was monitored by far UV CD using ellipticity changes at 222 nm (Fig. 2A). Loss of the minimum at 222 nm indicates the disruption of {alpha}-helical structure, which is a major structural component of XPF-EB1. The unfolding of XPF-EB1 did not occur until the urea concentration reached 6 M (Fig. 2A); this finding suggests that, during the refolding of ERCC1-FB in the presence of XPF-EB1 described above, most of the XPF-EB1 had a secondary structure similar to the native state. This result could explain the high yields from the refolding process used for the XPF-EB1·ERCC1-FB complex. In other words, during the refolding of this protein complex, XPF-EB1, which was mostly structurally intact, prevented the aggregation of the exposed hydrophobic regions of ERCC1-FB and/or acted as a template to induce the proper folding of ERCC1-FB.

Refolded ERCC1-FB and XPF-EB1 Bind Tightly to Each Other in a 1:1 Ratio by Hydrophobic Interactions—In order to determine whether refolded XPF-EB1 and ERCC1-FB are functionally active (ERCC1-FB binds tightly and specifically to XPF-EB1), binding assays using Ni2+-NTA-agarose were performed. A mixture containing the refolded complex (XPF-EB1 with a C-terminal His tag and ERCC1-FB) was loaded on a Ni2+-NTA affinity column, and the resin was extensively washed with Ni2+-NTA binding and His tag washing buffer to eliminate nonspecific binding. ERCC1-FB remained adhered to the column along with XPF-EB1, revealing that ERCC1-FB was bound tightly to XPF-EB1 (Fig. 2B), showing that the refolded proteins retain their ability to bind to each other, which is the key function of these domains. Other reports have shown that active XPF-ERCC1 complexes can be prepared using methods that involve co-expression and co-purification, reflecting strong interactions between XPF-EB1 and ERCC1-FB (2, 12).

On the basis of their amino acid sequences, XPF-EB1 and ERCC1-FB have predicted monomeric molecular masses of 11 and 8.2 kDa, respectively. In order to determine the stoichiometry of the refolded XPF-EB1·ERCC1-FB complex, gel filtration chromatography and equilibrium sedimentation were performed. The hydrodynamic molecular size of nonrefolded XPF-EB1 was analyzed on a Superdex 75 gel filtration column. XPF-EB1 eluted as a single peak at 64.51 min, which corresponds to a molecular mass of 38 kDa, consistent with either a trimer of globular shape or a dimer with an extended shape. The refolded XPF-EB1·ERCC1-FB complex eluted as a single peak at 71.19 min, a significantly longer elution time than that of XPF-EB1 alone, and corresponding to a molecular mass of 25.2 kDa (Table I). We also tested refolded XPF-EB1 alone, which eluted at the same position as native XPF-EB1 on a gel filtration column. These results indicate that the refolded XPF-EB1·ERCC1-FB complex is a 1:1 heterodimeric complex with a slightly extended conformation and that XPF-EB1 binds preferentially to ERCC1-FB despite competition from the self-association reaction of XPF-EB1.


View this table:
[in this window]
[in a new window]
  		TABLE I
Estimation of the molecular sizes of XPF-EB1 alone, XPF-EB2 alone, the XPF-EB1·ERCC1-FB complex, and the XPF-EB2·ERCC1-FB complex by gel filtration

Monomer molar mass of XPF-EB1 (11,000 g·mol–1); monomer molar mass of XPF-EB2 (7100 g·mol–1); monomer molar mass of ERCC1-FB (8200 g·mol–1).

 
To confirm the gel filtration data, we performed analytical ultracentrifugation equilibrium sedimentation experiments. The concentration of protein versus radius data were fitted with various models and were best described by a single component model. The equilibrium sedimentation data yielded a molecular weight of 17,863 ± 600 g·mol-1 for the refolded XPF-EB1·ERCC1-FB complex for the different rotor speeds, protein concentrations, and temperatures used (Table II). This supports the conclusion that the XPF-EB1·ERCC1-FB complex has a stoichiometry of one ERCC1-FB molecule bound per monomer of XPF-EB1 and that the 25-kDa size observed with gel filtration chromatography results from an elongated shape for the complex. In addition, equilibrium sedimentation results showed that XPF-B1 alone had a molecular weight of 21,060 ± 300 g·mol-1 at 4 °C (Table II), which implies that XPF-EB1 exists as a homodimer.


View this table:
[in this window]
[in a new window]
  		TABLE II
Sedimentation equilibrium experiments with the refolded XPF-EB1·ERCC1-FB complex and XPF-EB1 alone

Absorption optics on an XL-A ultracentrifuge were used.

 
To determine whether electrostatic interactions are important in the interaction of XPF-EB1 and ERCC1-FB, we assessed the cross-linking of the complex by glutaraldehyde in the presence of varying sodium chloride concentrations, analyzing products by SDS-PAGE. Variation in the sodium chloride concentration from 0.075 to 1 M did not alter the cross-linked patterns, although the intensity of the protein bands (lanes 5 and 6) in the SDS gel was slightly decreased at high salt concentrations (Fig. 2C), suggesting that formation of this complex is driven mainly by hydrophobic interactions. We also optimized the solvent conditions for NMR studies of this complex and observed that the complex showed reduced tendency to aggregate at higher salt concentrations (data not shown). Finally, the same cross-linking experiment with XPF-EB1 alone indicates that the XPF-EB1 homodimer is also driven by hydrophobic interactions (Fig. 2C). Other studies have highlighted the importance of hydrophobic interactions in distinct structure-specific endonuclease complexes that participate in nucleotide excision repair, such as UvrB and UvrC E. coli (33), and in the binding of Rad1 with Rad10 in Saccharomyces cerevisiae (34).

Characteristics and Secondary Structure of the XPF-EB2 Homodimer and the XPF-EB2·ERCC1-FB Complex—We next used NMR to probe the interface of the XPF-ERCC1 complex. Assigned chemical shifts for the individual subunits and the complex are required for solution structure determinations and for probing the interface in the complex. We first collected a two-dimensional 15N-1H HSQC (heteronuclear single quantum coherence) spectrum of soluble 15N-XPF-EB1. Unfortunately, because of flexible N- and C-terminal regions of XPF-EB1, we observed many overlapped peaks with strong intensities in the central region of the HSQC spectrum and severe overlap of resonances in the three-dimensional NMR spectrum, which hampered assignment of the 13C, 15N, and 1H resonances (data not shown).

To eliminate these problems, we subcloned the DNA coding sequences for XPF-EB2-(838-901) (Fig. 1) with the N-terminal His tag in such a way as to remove the terminal regions that were thought to be flexible on the basis of secondary structure predictions. The resulting construct was expressed as a soluble protein (XPF-EB2) in E. coli BL21(DE3). In order to confirm that refolded XPF-EB2 and ERCC1-FB display the same characteristics as XPF-EB1 and ERCC1-FB, XPF-EB2 was subjected to the same binding assays and gel filtration chromatography as XPF-EB1. The Ni2+-NTA binding assay data showed that ERCC1-FB (which did not have a His tag) was retained on the column (Fig. 3A), which indicates binding to ERCC1-FB. In gel filtration chromatography, XPF-EB2 eluted as a single peak at 78.8 min, which corresponds to a molecular size of 16.4 kDa, consistent with a dimer (Table I). The refolded XPF-EB2·ERCC1-FB complex eluted as a single peak at 77.23 min, corresponding to a molecular size of 17.4 kDa (Table I), again consistent with the theoretical molecular size of 15.3 kDa. These results indicate that refolded XPF-EB2 and ERCC1-FB form a 1:1 heterodimeric complex with the same characteristics as the previously characterized XPF-EB1·ERCC1-FB complex.

We recorded two-dimensional 15N-1H HSQC spectra for 15N-XPF-EB2 alone, the15N-XPF-EB2·14N-ERCC1-FB complex, and the 14N-XPF-EB2·15N-ERCC1-FB complex. The 15N-1H HSQC spectrum of 15N-XPF-EB2 was of a higher quality than the previous one for 15N-XPF-EB1 and allowed easy assignment of the resonances. The similarity in shifted resonances showed that 15N-XPF-EB2 had a structure very similar to the folded part of XPF-EB1 (data not shown). Both soluble XPF-EB2 alone and the refolded XPF-EB2/ERCC1-FB complex showed well-dispersed 15N and 1H resonances in the HSQC spectrum, with relatively uniform line widths, indicating that the XPF-EB2 homodimer and the XPF-EB2·ERCC1-FB heterodimeric complex are well folded (Fig. 3, B-D). The number of peaks in the 15N-1H HSQC spectrum of 15N-XPF-EB2 (Fig. 3B) is very close to the number of amino acids in the XPF-EB2 monomer, indicating a symmetric homodimer.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 3.
A,Ni2+-NTA binding assay between refolded XPF-EB2 and ERCC1-FB. Lane 1, refolded XPF-EB2 and ERCC1-FB; lane 2, wash of resin with binding buffer; lanes 3 and 4, wash of resin with washing buffer; lanes 5 and 6, elution fractions containing XPF-EB2 with a C-terminal His tag (open arrowhead) and ERCC1-FB without tag (solid arrowhead). B, two-dimensional 15N-1H HSQC spectra of 15N-XPF-EB2 alone. C, the 15N-XPF-EB2·14N-ERCC1-FB complex. D, two-dimensional 15N-1H HSQC spectrum of the 14N-XPF-EB2·15N-ERCC1-FB complex.

 
We next performed three-dimensional triple resonance experiments with 13C,15N-XPF-EB2 alone, for the 13C,15N-XPF-EB2·14N-ERCC1-FB complex, and for the 14N-XPF-EB2·13C,15N-ERCC1-FB complex, in order to probe the secondary structure of XPF-EB2 and ERCC1-FB. From these three-dimensional NMR analyses, we assigned the main chain resonances of the XPF-EB2 homodimer and the XPF-EB2·ERCC1-FB complex. On the basis of chemical shift index analysis using (H{alpha},Ca, C{beta} CO) (35), we determined the secondary structure in both the XPF-EB2 homodimer and XPF-EB2·ERCC1-FB complex (Fig. 4, A, B, and D). In agreement with previous studies that predicted the existence of (HhH)2 motifs in both XPF-EB2 and ERCC1-FB (21), our data show that the XPF-EB2 homodimer consists of five {alpha}-helices (H1, 838-842; H2, 849-857; H3, 862-866; H4, 870-876; and H5, 879-891) per monomer. XPF-EB2 in the heterodimer complex has a similar secondary structure, with the same five {alpha}-helices (H1, 838-842; H2, 849-857; H3, 862-866; H4, 870-876; and H5, 880-892). These results show that refolding of the ERCC1 binding domain of XPF is not required for conversion of the homodimeric complex to the heterodimeric complex. ERCC1-FB, again in the context of the heterodimeric complex, was also shown to have five {alpha}-helices in (HhH)2 motifs (H1, 233-240; H2, 247-257; H3, 260-265; H4, 268-272; and H5, 280-288) and an additional {alpha}-helix in the N-terminal region (226-229).



View larger version (79K):
[in this window]
[in a new window]
  		FIG. 4.
Cross-saturation experiments for identification of the binding sites on the XPF-EB2 homodimer and the XPF-EB2·ERCC1-FB complex. A, intensity changes of the 1H-15N cross-peaks of XPF-EB2 by the cross-saturation from nonlabeled XPF-EB2. The residues with intensity ratios <0.75 are marked on the XPF sequence (green circles). The {alpha}-helices (H1-H5) are indicated, and dotted lines represent the hairpin regions (h1-h2) in the XPF-EB2 homodimer along with the determined secondary structure from chemical shift index analysis. B, intensity changes for 1H-15N cross-peaks of XPF-EB2 by cross-saturation from nonlabeled ERCC1-FB. The residues with intensity ratios <0.75 are marked on the XPF sequence (pink triangles). {alpha}-Helices (H1-H5) are indicated, and dotted lines represent the hairpin regions (h1-h2) of XPF-EB2 in the heterodimer. C, plots of the weighted averaged 1H and 15N chemical shift changes in XPF calculated with the function {delta}{Delta}= (({delta}{Delta}HN2 + {delta}{Delta} 0.115N2)/2). D, intensity changes of the 1H-15N cross-peaks of ERCC1-FB by the cross-saturation from nonlabeled XPF-EB2. The residues with intensity ratios <0.75 are marked on the ERCC1 sequence (violet triangles). {alpha}-Helices (H1-H5) are indicated, and dotted lines represent the hairpin regions (h1-h2) of ERCC1-FB in the heterodimer along with the determined secondary structure from chemical shift index analysis.

 
Interface of the XPF-XPF Homodimer—We mapped the interface in the XPF-EB2 homodimer using cross-saturation methods (26), which can identify protein-protein interfaces reliably on the basis of saturation transfer from a nonlabeled to a labeled protein (36). Cross-saturation decreases the intensity of 1H-15N cross-peaks that result from amino acid residues that reside at the molecular interface. In our experiments, the resonance corresponding to Ala863 exhibited the greatest decrease in intensity (~55%) (Fig. 4A), implying that it is a residue at the interface with the nonlabeled XPF-EB2 molecule. Other residues that appeared to be at the interface include XPF-EB2 residues Ile862, Glu864, Leu865, Ala866, Ala867, Ile890, His891, Thr892, Phe894, since they were most affected by cross-saturation from the nonlabeled XPF-EB2 molecule. These amino acid residues reside in helix H3 and in the C-terminal region of XPF-EB2. About 60% of the binding interface residues identified are found to be hydrophobic in nature and thus could contribute to the hydrophobic interactions stabilizing the homodimer, as suggested by the cross-linking data.

Formation of the XPF-ERCC1 heterodimer is required to create a stable and active endonuclease. A small amount of XPF protein persists in ERCC1-deficient cells even in the absence of complex formation with ERCC1 (4, 21). The self-associated form of XPF has not yet been observed in vivo; however, previous observations and the findings described here show that XPF has a strong tendency to self-associate in the absence of its partner (21). As described above, the ERCC1 binding region of XPF associates in vitro to form a stable homodimer, a process that seems likely to occur under physiological conditions as well. The endonuclease activity of the XPF-ERCC1 complex is associated with XPF, whereas ERCC1 probably modulates the specificity required for DNA incision during the repair process (13). It has been suggested that it might be advantageous for XPF that is not bound to ERCC1 to exist in a self-associated state that does not have nuclease activity, since this would reduce the amount of nonspecific DNA cleavage. Consistent with this hypothesis, no endonuclease activity has been observed for self-associated XPF (21). Our results suggest that the ERCC1 binding region of XPF serves as the self-association domain in the absence of ERCC1.



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 5.
Use of GST resin to assess binding of refolded GST-XPF-EB3-(813-847) and ERCC1-FB (lanes 1-4) and refolded GST and ERCC1-FB (control) (lanes 5-7). Lane 1, load of mixture containing refolded GST-XPF-EB3-(813-847) and ERCC1-FB; lane 2, wash of resin with GST-binding buffer; lanes 3 and 4, elution of refolded GST-XPF-EB3-(813-847) and ERCC1-FB; lane 5, wash of resin with GST-binding buffer; lanes 6 and 7, elution of GST.

 
Interface of the XPF/ERCC1 Complex—XPF binds specifically and tightly to ERCC1 both in vivo and in vitro (4, 14, 15, 19). In order to understand the characteristics of the XPF-ERCC1 heterodimeric interface, mapping of the contact residues involved in the interaction between XPF and ERCC1 is necessary. To identify amino acid residues involved in formation of the XPF-ERCC1 interface, we again performed cross-saturation experiments. Fig. 4B shows the 21 residues of XPF, Gln838, Asp839, Phe840, Leu842, Lys843, Met844, Met856, His857, Asn861, Ile862, Ala863, Glu864, Leu865, Ala866, Ala867, Phe889, Ile890, His891, Thr892, Phe894, Ala895, for which the intensities of the 1H-15N cross-peaks were significantly affected by the cross-saturation. These residues are distributed along helix H1, a small part of helix H2, helix H3, and the C-terminal region. The resonance of Ala863 in H3 exhibited the greatest decrease in intensity (~63%), as was the case for the XPF homodimer. Other residues that showed decreases in intensity nearly as large were Phe840 of H1 and Ile862 of H3. Met856 and His857 in H2 had smaller decreases in intensity than did the contact residues but still showed a clear decrease in intensity relative to noncontact residues. About ~60% of the binding surface residues at the interface are hydrophobic in the heterodimer, consistent with the cross-linking data. The aggregation of free ERCC1 may be caused by an exposed hydrophobic surface that cannot efficiently homodimerize and is shielded by interaction with XPF in the heterodimer. Helix H3 and the C-terminal region of XPF contain residues that participate in both homodimer and heterodimer formation. We believe that the preference XPF displays for ERCC1 is specified by helix H1 in XPF. Helix H1 has more polar contact residues (e.g. Asp839 and Lys843) than do the other helices in the heterodimer interface. The interface mapping results indicate that the surface of the XPF interface in the heterodimer is larger than that in the homodimer, which could be responsible for driving heterodimer formation.

The combined 1H and 15N chemical shift changes between the ERCC1 bound and homodimeric states of XPF are plotted in Fig. 4C. The residues with large chemical shift changes are localized in H1, H2, H3, and the C-terminal region and are generally consistent with the XPF contacts to ERCC1 based on cross-saturation, excepting Asn851, Cys852, and Arg853 in helix H2. These residues in H2 might reflect a conformational change occurring in or around H1 to form the heterodimer interface.

The ERCC1 binding domain of XPF (residues 814-905) was first identified by de Laat et al. (19), and subsequently McCutchen-Maloney et al. (13) localized it to amino acids 656-843. Together, these two studies implicate the region between amino acids 814 and 843 (13). To further define the contribution of XPF-(813-843) to heterodimer formation, we made a recombinant GST-XPF-EB3-(813-847) fusion protein, which contains the residues of the first helix that is part of the ERCC1 binding domain of XPF. GST-XPF-EB3 was refolded by the same method used for ERCC1-FB and showed clear but weak binding to ERCC1-FB (Fig. 5). This result shows that XPF amino acids 813-847 are sufficient for binding, albeit weakly, to ERCC1-FB, and that the H1 region of XPF-EB is important for recognition of the binding interface of ERCC1. However, the rest of the ERCC1 binding domain of XPF is required for strong binding of XPF to ERCC1-FB.

To identify the binding residues of ERCC1 within the heterodimeric complex, we performed further cross-saturation experiments. Residues that appear to be in the interface region include Arg234, Cys238, Thr240, Thr241, Leu260, Glu261, Ile264, Ala265, Val288, Leu289, His290, Glu291, and Leu294, located in H1, H3, and the C-terminal region of ERCC1 (Fig. 4D). The resonance corresponding to Glu261 in H3 exhibited the greatest decrease in intensity (~47%). Cys238 of H1 also exhibited a decrease in its intensity that was nearly as great as that of Glu261. It is interesting that the parts of the sequence forming the interface in ERCC1 are similar to those in XPF. Considering the secondary structure similarity and ~64% primary sequence homology between the interaction domains of XPF and ERCC1, the heterodimer might be nearly symmetrical in structure, as is the XPF homodimer. de Laat et al. (19) had suggested that residues 293-297 in ERCC1 are important for XPF binding although they are not part of the (HhH)2 motif. Consistent with this suggestion, we also observed that Leu294 of ERCC1 as a contact residue. We could not specifically probe Phe293 because of peak overlap, although a decrease in the peak containing Phe293 was seen, and hence it could be a contact residue. Additional {alpha}-helices in the N terminus that are not part of the (HhH)2 motifs of ERCC1-FB do not appear to be necessary for interface formation.

The HhH motif has been predicted to exist in several DNA repair proteins and has been shown to mediate mainly non-sequence-specific interactions with DNA (37-39). For instance, the (HhH)2 motifs of the C-terminal domain of the E. coli XPF homolog UvrC are required for DNA binding to a 5' incision point in the prokaryotic nucleotide excision repair process (40). Although the ERCC1 binding domain of XPF and the XPF binding domain of ERCC1 have (HhH)2 motifs, our data show that these structures are used for protein-protein interactions instead of binding to the DNA. The homodimeric structure formed by the SAM domain in the EphA4 receptor tyrosine kinase was determined by x-ray crystallography (28). The SAM domain can homo- and hetero-oligomerize with other SAM domains and is composed of (HhH)2 motifs that mediate the various protein-protein interactions (28). This domain is similar to the (HhH)2 motifs of both XPF and ERCC1 and shares 55% sequence homology with these domains. The interface of the SAM homodimer consists of the N-terminal strand, C-terminal helix H5, helix H1, and H3. The interface in the XPF-ERCC1 complex appears to be related to that of the Eph receptor SAM domain homodimer but with significant differences in details. The structure of the SAM dimer is shown in Fig. 6A, looking down the 2-fold axis. The dimer interface is composed of an N-terminal segment that is extended and the C-terminal portion of helix 5 plus 2 residues beyond the helix. Comparing the sequences and secondary structure (for ERCC1 and XPF based on the NMR data) (Fig. 1), it is immediately apparent that there are differences in both structure and contacts for the equivalent regions of ERCC1 and XPF. First, in the SAM dimer, the contacts to the N-terminal region involve residues before the first helix, whereas contacts in XPF-ERCC1 involve residues in the first helix. Second, in the SAM dimer, there is just one contact in the region of helix 3, whereas this is a major contact region for XPF-ERCC1. Third, in the SAM dimer the intermolecular contacts in the C-terminal region involve residues at the end of its elongated helix 5. In the XPF-ERCC1 complex, contacts are shifted toward the N terminus by at least 6 residues and involve residues that are not helical. Fig. 6 shows the residues involved in contacts in the XPF-ERCC1 complex (Fig. 6A) mapped onto the SAM dimer structure. It is clear that the contact residues are far from the interface. To get a better model for the XPF-ERCC1 dimer, the ERCC1 and XPF structures were each built by homology modeling onto previously determined HhH structures and then brought together using the program AUTODOCK, with the result shown in Fig. 6B. Contact residues defined by cross-saturation are shown as colored highlights. This model brings the appropriate regions into proximity. There are a few unmodeled residues, particularly at the C terminus, that are not helical and may wrap up onto the dimer partner to make additional contacts. Since the individual domain models are only homology models, and the juxtaposition is by docking with very loose restraints, one must view this as a rough model. While this manuscript was in review, the structure of an archaeal XPF homolog dimer, containing nuclease and (HhH)2 domains was reported (32). Threading the human sequences onto the structure of the (HhH)2 domain (Protein Data Bank code 2bhn [PDB] ) yielded the results shown in Fig. 6C, qualitatively similar to the AUTODOCK model (Fig. 6B), the biggest difference being a rotation of the proteins relative to one another. The NMR data reported here indicate small differences in secondary structure in the human XPF-ERCC1 proteins relative to the archaeal, but the similarity is sufficient to support a model for the human XPF homodimer and the XPF-ERCC1 complex that has similar general features. Determination of the full three-dimensional structures of these complexes is currently under way in our laboratories.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6.
The structure of the Eph receptor SAM domain dimer (Protein Data Bank code 1b0x [PDB] ) is shown in A with contact residues from ERCC1 and XPF indicated by red and orange. In B, the AUTODOCK model of the homology-modeled ERCC1 and XPF domains in complex is shown (ERCC1 in green, XPF in blue) with contact residues to the partner again mapped in red and orange. In C, the model based on the archaeal XPF dimer (Protein Data Bank code 2bhn [PDB] ) is shown with the same coloring including contact residues.

 
Biological Implications of XPF-ERCC1 Interface—XPF (xeroderma pigmentosum group F) patients have skin that is highly sensitive to UV irradiation from the sun. These patients carry a truncated version of the XPF protein caused by a frame shift mutation that inserts a premature stop codon in front of the ERCC1 binding domain or a point mutation, R788W, which is known to affect heterodimeric complex formation (41). Reduced concentrations of both the XPF and the ERCC1 proteins are observed in these patients, although the genes encoding XPF and ERCC1 are localized to different chromosomes (16). It has been shown that a normal amount of ERCC1 mRNA is transcribed in XPF-defective cells (42), and no evidence for a lowered level of translation of ERCC1 in XPF-deficient cells exists. The ERCC1 protein concentration can be increased in XPF-defective cells by transfection with an expression vector encoding wild-type XPF (43). These data imply that the stability and function of ERCC1 in the cell are dependent on heterodimer formation with XPF. However, until now, the molecular characteristics of the XPF-ERCC1 interaction had not yet been elucidated. Our results help to clarify the basis for the specific and tight binding that characterize the XPF-ERCC1 interaction. In addition, we suggested that homodimerization of the ERCC1 binding domain of XPF is likely to occur under physiological conditions. Consequently, the ERCC1 binding region of XPF may possess at least two functions, ERCC1 binding and self-association, and these functions probably are important for the regulation of XPF endonuclease activity and, therefore, the DNA repair process.


   FOOTNOTES
 
* This work was supported in part by the Creative Research Initiative Program from the Ministry of Science and Technology, Korea. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

§ Supported in part by the BK21 project. Back

** To whom correspondence may be addressed. Tel.: 510-486-4318; Fax: 510-486-6059; E-mail: dewemmer{at}lbl.gov.

{ddagger}{ddagger} To whom correspondence may be addressed. Tel.: 82-42-869-2868; Fax: 82-42-869-2810; E-mail: byongseok.choi{at}kaist.ac.kr.

1 The abbreviations used are: HhH, helix-hairpin-helix; XPF-EB, ERCC1 binding domain of XPF; ERCC1-FB, XPF binding domain of ERCC1; GST, glutathione S-transferase; DTT, dithiothreitol; SAM, sterile {alpha} motif; NTA, nitrilotriacetic acid. Back


   ACKNOWLEDGMENTS
 
We thank Dr. Sung-Hun Bae for useful discussions and John-Marc Chandonia for help with modeling.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS AND DISCUSSION
 REFERENCES
 

  1. Benhamou, S., and Sarasin, A. (2000) Mutat. Res. 462, 149-158[CrossRef][Medline] [Order article via Infotrieve]
  2. de Laat, W. L., Appeldoorn, E., Jaspers, N. G. J., and Hoeijmakers, J. H. J. (1998) J. Biol. Chem. 273, 7835-7842[Abstract/Free Full Text]
  3. Matsunaga, T., Mu, D., Park, C.-H., Reardon, J. T., and Sancar, A. (1995) J. Biol. Chem. 270, 20862-20869[Abstract/Free Full Text]
  4. Sijbers, A. M., de Laat, W. L., Ariza, R. R., Biggerstaff, M., Wei, Y.-F., Moggs, J. G., Carter, K. C., Shell, B. K., Evans, E., de Jong, M. C., Rademakers, S., de Rooij, J., Jaspers, N. G. J., Hoeijmakers, J. H. J., and Wood, R. D. (1996) Cell 86, 811-822[CrossRef][Medline] [Order article via Infotrieve]
  5. O'Donovan, A., Davies, A. A., Moggs, J. G., West, S. C., and Wood, R. D. (1994) Nature 371, 432-435[CrossRef][Medline] [Order article via Infotrieve]
  6. Huang, J. C., Svoboda, D. L., Reardon, J. T., and Sancar, A. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 3664-3668[Abstract/Free Full Text]
  7. Moggs, J. G., Yarema, K. J., Essigmann, J. M., and Wood, R. D. (1996) J. Biol. Chem. 271, 7177-7186[Abstract/Free Full Text]
  8. Mu, D., Park, C.-H., Matsunaga, T., Hsu, D. S., Reardon, J. T., and Sancar, A. (1995) J. Biol. Chem. 270, 2415-2418[Abstract/Free Full Text]
  9. Aboussekhra, A., Biggerstaff, M., Shivji, M. K. K., Vilpo, J. A., Moncollin, V., Podust, V. N., Protic, M., Hubscher, U., Egly, J.-M., and Wood, R. D. (1995) Cell 80, 859-868[CrossRef][Medline] [Order article via Infotrieve]
  10. Friedberg, E. C., Walker, G. C., and Siede, W. (1995) DNA Repair and Mutagenesis, American Society for Microbiology Press, Washington, D. C.
  11. de Laat, W. L., Jaspers N. G., and Hoeijmakers, J. H. (1999) Genes Dev. 13, 768-785[Free Full Text]
  12. Bessho, T., Sancar, A., Thompson, L. H., and Thelen, M. P. (1997) J. Biol. Chem. 272, 3833-3837[Abstract/Free Full Text]
  13. McCutchen-Maloney, S. L., Giannecchini, C. A., Hwang, M. H., and Thelen, M. P. (1999) Biochemistry 38, 9417-9425[CrossRef][Medline] [Order article via Infotrieve]
  14. Biggerstaff, M., Szymkowski, D. E., and Wood, R. D. (1993) EMBO J. 12, 3685-3692[Medline] [Order article via Infotrieve]
  15. van Vuuren, A. J., Aplleldoorn., E Odijk, H. Yasui, A., Jaspers, N. G. J. Bootsma, D., and Hoeijmaker, H. J. (1993) EMBO J. 12, 3693-3701[Medline] [Order article via Infotrieve]
  16. Yagi, T., Wood, R. D., and Takebe, H. (1997) Mutagenesis, 12, 41-44[Abstract/Free Full Text]
  17. Kaliraman, V., Mullen, J. R., Fricke, W. M., Bastin-Shanower, S. A. and Brill, S. J. (2001) Genes Dev. 15, 2730-2740[Abstract/Free Full Text]
  18. Bardwell, L., Cooper, A. J., and Friedberg, E. C. (1992) Mol. Cell. Biol. 12, 3041-3049[Abstract/Free Full Text]
  19. de Laat, W. L., Sijbers, A. M., Odijk, H., Jaspers, N. G., and Hoeijmakers, J. H. (1998) Nucleic Acids Res. 26, 4146-4152[Abstract/Free Full Text]
  20. Sijbers, A. M., van der Spek, P. J., Odijk, H., van den Berg, J., van Duin, M., Westerveld, A., Jaspers, N. G., Bootsma, D., and Hoeijmakers, J. H. (1996) Nucleic Acids Res. 24, 3370-3380[Abstract/Free Full Text]
  21. Gaillard, P. H., and Wood, R. D. (2001) Nucleic Acids Res. 29, 872-879[Abstract/Free Full Text]
  22. Creighton, T. E. (1997) Biol. Chem. 378, 731-744[Medline] [Order article via Infotrieve]
  23. Suenaga, M., Ohmae, H., Tsuji, S., Itoh, T., and Nishimura, O. (1998) Biotechnol. Appl. Biochem. 28, 119-124
  24. Clore, G. M., and Gronenborn, A. M. (1994) Methods Enzymol. 239, 349-363[Medline] [Order article via Infotrieve]
  25. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A. (1995) J. Biomol. NMR 6, 277-293[Medline] [Order article via Infotrieve]
  26. Takahashi, H., Nakanishi, T., Kami, K., Arata, Y., and Shimada, I. (2000) Nat. Struct. Biol. 7, 220-223[CrossRef][Medline] [Order article via Infotrieve]
  27. Kupce, E., and Freeman, R. (1995) J. Magn. Reson. A 117, 246-256[CrossRef]
  28. Stapleton, D., Balan, I., Pawson, T., and Sicheri, F. (1999) Nat. Struct. Biol. 6, 44-49[CrossRef][Medline] [Order article via Infotrieve]
  29. Sali, A., and Blundell, T. L. (1993) J. Mol. Biol. 234, 779-815[CrossRef][Medline] [Order article via Infotrieve]
  30. Laskowski, R. A., MacArthur, M. W., Moss, D. S., and Thornton, J. M. (1993) J. Appl. Crystallogr. 26, 283-291[CrossRef]
  31. Morris, G. M., Goodsell, D. S., Halliday, R. S., Huey, R., Hart, W. E., Belew, R. K., and Olson, A. J. (1998) J. Comput. Chem. 19, 1639-1662[CrossRef]
  32. Newman, M., Murray-Rust, J., Lally, J., Rudolf, J., Fadden, A., Knowles, P. P., White, M. F., and McDonald, N. Q. (2005) EMBO J. 24, 895-905[CrossRef][Medline] [Order article via Infotrieve]
  33. Moolenaar, G. F., Franken, K. L. M. C., Dijkstra, D. M., Thomas-Oates, J. E., Visse, R., van de Putte, P., and Goosen, N. (1995) J. Biol. Chem. 270, 30508-30515[Abstract/Free Full Text]
  34. Bardwell, A. J., Bardwell, L., Johnson, D. K., and Friedberg, E. C. (1993) Mol. Microbiol. 8, 1177-1188[Medline] [Order article via Infotrieve]
  35. Wishart, D. S., Watson, M. S., Boyko, R. F., and Sykes, B. D. (1997) J. Biomol. NMR 10, 329-336[CrossRef][Medline] [Order article via Infotrieve]
  36. Wüthrich, K. (2000) Nat. Struct. Biol. 7, 188-189[CrossRef][Medline] [Order article via Infotrieve]
  37. Doherty, A. J., Serpell, L. C., and Ponting, C. P. (1996) Nucleic Acids Res. 24, 2488-2497[Abstract/Free Full Text]
  38. Aravind, L., Walker, D. R., and Koonin, E. V. (1999) Nucleic Acids Res. 27, 1223-1242[Abstract/Free Full Text]
  39. Shao, X., and Grishin, N. V. (2000) Nucleic Acids Res. 28, 2643-2650[Abstract/Free Full Text]
  40. Singh, S., Folkers, G. E., Bonvin, A. M., Boelens, R., Wechselberger, R., Niztayev, A., and Kaptein, R. (2002) EMBO J. 21, 6257-6266[CrossRef][Medline] [Order article via Infotrieve]
  41. Sijbers, A. M., van Voorst Vader, P. C., Snoek, J. W., Raams, A., Jaspers, N. G., and Kleijer, W. J. (1998) J. Invest. Dermatol. 110, 832-836[CrossRef][Medline] [Order article via Infotrieve]
  42. van Duin, M., Vredeveldt, G., Mayne, L. V., Odijk, H., Vermeulen, W., Klein, B., Weeda, G., Hoeijmakers, J. H., Bootsma, D., and Westerveld, A. (1989) Mutat. Res. 217, 83-92[Medline] [Order article via Infotrieve]
  43. Yagi, T., Matsumura, Y., Sato, M., Nishigori, C., Mori, T., Sijbers, A. M., and Takebe, H. (1998) Carcinogenesis 19, 55-60[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
D. Yu, X. Zhang, J. Liu, P. Yuan, W. Tan, Y. Guo, T. Sun, D. Zhao, M. Yang, J. Liu, et al.
Characterization of Functional Excision Repair Cross-Complementation Group 1 Variants and Their Association with Lung Cancer Risk and Prognosis
Clin. Cancer Res., May 1, 2008; 14(9): 2878 - 2886.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/31/28644    most recent
M501083200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Choi, Y.-J.
Right arrow Articles by Choi, B.-S.
Right arrow Search for Related Content
PubMed