Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

YB‐1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or cisplatin modifications, exhibits endonucleolytic activities and binds several DNA repair proteins

YB‐1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or... Published online January 12, 2004 316±327 Nucleic Acids Research, 2004, Vol. 32, No. 1 DOI: 10.1093/nar/gkh170 YB-1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or cisplatin modi®cations, exhibits endonucleolytic activities and binds several DNA repair proteins Isabelle Gaudreault, David Guay and Michel Lebel* Centre de Recherche en Cance  rologie de l'Universite  Laval, Ho à pital Ho à tel-Dieu de Que  bec, Centre Hospitalier   Universitaire de Quebec, 9 McMahon Street, Quebec, G1R 2J6, Canada Received June 23, 2003; Revised September 24, 2003; Accepted November 15, 2003 ABSTRACT protein tyrosine phosphatase 1B (PTP1B) gene (8), gelatinase A (9) and the multidrug resistance (MDR1) gene (10,11). In YB-1 is a multifunctional protein involved in the addition to the regulation of transcription, YB-1 is a multi- regulation of transcription, translation, mRNA spli- functional protein that also affects the splicing and the cing and probably DNA repair. It contains a con- translation of speci®c mRNAs. Indeed, analysis of the human served cold shock domain and it binds strongly to presplicesome complex has indicated the presence of YB-1 in inverted CCAAT box of different promoters. In this this complex (12). Furthermore, YB-1 interacts with the study, we have found that puri®ed YB-1 oligo- splicing factor SRp30c and confers alternative splice site selection in the E1A minigene model system (13). YB-1 is merizes readily in solutions to form trimers, hexam- also known to bind A/C-rich exon enhancers and to stimulate ers and oligomers of 12 molecules. The presence of splicing of the CD44 alternative exon v4 (14). YB-1 is a ATP changed the conformation of YB-1 in such a component of messenger ribonucleoprotein particles which way that only dimers were detected by gel ®ltration regulate translation in COS cells (15). Consistent with this analyses. Puri®ed YB-1 can separate different DNA observation, YB-1 is capable of speci®cally regulating the duplexes containing blunt ends, 5¢ or 3¢ recessed translation of ferritin mRNA (16). Finally, YB-1 is known to ends, or forked structures. This strand separation increase the stability of IL-2 speci®c mRNA during T-cell activity is increased on cisplatin-modi®ed DNA or activation (17). with duplex molecules containing mismatches. In In recent years, several laboratories have demonstrated that addition to its exonuclease activity, YB-1 exhibits YB-1 is directly involved in the cellular response to genotoxic endonucleolytic activities in vitro. Finally, YB-1 stress. Upon UV irradiation or cisplatin treatments, YB-1 af®nity chromatography experiments have indicated binds to a Y-box element within the MDR1 promoter and increases its expression (11). Moreover, depletion of YB-1 that in addition to prespliceosome factors like expression protein with anti-sense RNA against YB-1 speci®c nucleolin and ALY, YB-1 binds the DNA repair mRNA results in increased sensitivity to cisplatin (11). proteins MSH2, DNA polymerase d, Ku80 and WRN Interestingly, YB-1 is increased in cultured cell lines resistant proteins in vitro. Furthermore, immuno¯uorescence to cisplatin. In fact, several studies have indicated that the studies have shown that YB-1 re-localizes from the level of nuclear expression of YB-1 is predictive of drug cytoplasm to nuclear areas containing either Ku80 resistance and patient outcome in breast tumors, ovarian or MSH2 proteins in human 293 embryonic kidney cancers and synovial sarcomas (18±22). Upon UV irradiation, cells. These results suggest that YB-1 is involved in YB-1 translocates from the cytoplasm to the nucleus (23) and base excision and mismatch repair pathways. is known to bind to modi®ed nucleic acid (24). YB-1 preferentially binds to cisplatin-modi®ed DNA and interacts with PCNA (25), a component of several DNA repair systems INTRODUCTION (26). In addition, YB-1 stimulates an endonuclease involved in YB-1 is a member of a family of DNA-binding proteins base excision repair (27). All these observations suggest that containing a cold shock domain. It was originally identi®ed by YB-1 is important in DNA repair and in conferring drug its ability to bind to the inverted CCAAT box (Y-box) in the resistance on tumor cells. promoter region of MHC class II genes (1). Subsequently, YB- It has been reported that YB-1 creates single-stranded 1 has been shown to regulate the expression of a number of regions in the DRA promoter (28) and it is believed that this genes including proliferating cell nuclear antigen (PCNA) (2), activity is required in part for the regulation of target thymidine kinase (3), DNA polymerase a (4), myosin light- promoters (29). In recent years, YB-1 has been shown to chain 2v gene (5), Fas antigen (6), collagen a1(I) gene (7), bind preferentially to single-stranded nucleic acids and to *To whom correspondence should be addressed. Tel: +1 418 691 5281; Fax: +1 418 691 5439; Email: [email protected] Nucleic Acids Research, Vol. 32 No. 1 ã Oxford University Press 2004; all rights reserved Nucleic Acids Research, 2004, Vol. 32, No. 1 317 exhibit 3¢-5¢ exonuclease activity (30). In this report, we YB-1 puri®cation and gel ®ltration investigated the strand separation activity of human YB-1 BL21 cells expressing GST±YB-1 fusion proteins were lysed against different double-stranded DNA substrates in vitro. in NETN buffer (0.5% NP-40, 20 mM Tris±HCl pH 8.0, Different deletion mutants of YB-1 have indicated that amino 100 mM NaCl and 1 mM EDTA) and incubated overnight acids 39±205 are required for the DNA strand separation with glutathione±Sepharose beads. The next day, beads were activity. We have also found that YB-1 actively promotes washed with NETN buffer and treated with biotinylated strand separation of duplex DNA containing either mis- thrombin (Novagen) for 2 h at room temperature in thrombin matches or cisplatin modi®cations independently of the cleavage buffer (20 mM Tris±HCl pH 8.4, 150 mM NaCl, nucleotide sequence. It also exhibits an endonuclease activity 2.5 mM CaCl ). Beads were spun down and the supernatant on double-stranded DNA. Finally, in vitro YB-1 af®nity was kept for the next step. Thrombin was captured by chromatography and immuno¯uorescence analyses have incubation with streptavidine agarose (Novagen) for 2 h on a shown that several DNA repair proteins can interact with rocking platform at room temperature. Agarose beads were YB-1 reinforcing the notion that this multifunctional protein is spun down and YB-1 protein from the supernatant was involved in the repair of speci®c DNA damage. concentrated onto Centricon-30 ®lters (Amicon). Protein concentration was determined using the Bradford assay. Proteins were then loaded onto a Superdex-200 column for gel ®ltration analysis using an AKTA-FPLC as indicated by MATERIALS AND METHODS the manufacturer (Amersham Pharmacia). Proteins from each fraction of the column were visualized by Coomassie staining. Cell lines and antibodies Human 293 embryonic kidney cells were maintained in Strand separation and endonuclease assays DMEM supplemented with 10% fetal bovine serum. The DNA substrates used for the strand separation and Polyclonal antibodies against the human WRN were pur- nuclease assays are depicted in Table 1. One strand was chased from Novus Biologicals (Littleton, CO). Antibodies labeled with T4 polynucleotide kinase and [g- P]ATP and against PARP-1 and DNA polymerase d were purchased from annealed to its complementary strand as described previously Transduction Laboratories (Lexington, KY). Antibodies (32). DNA with cisplatin adducts were created by incubating against ALY, REF1 and XRCC1 were purchased from Santa oligonucleotides with cisplatin overnight as described previ- Cruz Biotechnology (Santa Cruz, CA). Antibodies against ously (33). DNA substrate was incubated with the recombi- Ku80 were purchased from NeoMarkers (Fremont, CA). nant proteins as indicated in the ®gure legends for 30 min at Antibodies against DNA-PK, MLH1, MSH2 and PMS2 were 37°C in the reaction buffer (40 mM Tris±HCl, pH 7.5, 4 mM purchased from Oncogene Research Products (Boston, MA). MgCl , 5 mM DTT, 2 mM ATP and 0.1 mg/ml bovine serum Antibodies against nucleolin were purchased from Medical albumin). Four microliters of loading buffer was added to the and Biological Laboratories Co. (Watertown, MA). Rabbit reaction (40% glycerol, 50 mM sodium EDTA, 2% SDS and polyclonal antibody against human YB-1 and the correspond- 1% bromophenol blue, pH 8.0) and the DNA was analysed on ing pre-immune serum was kindly provided by Dr P. E. native 12% polyacrylamide gel (in TBE buffer). For the DiCorleto (The Cleveland Clinic Foundation, Cleveland, OH). endonuclease and exonuclease assays, radioactive DNA Finally, all horseradish peroxidase-conjugated secondary substrates were incubated with the indicated puri®ed proteins antibodies were purchased from Amersham Pharmacia. The in the same reaction buffer as indicated for the strand above antibodies were used as indicated by the manufacturers. separation reaction. Cleaved DNA products were separated Western blots were performed as described previously (31). on a denaturing gel (14% polyacrylamide, 8 M urea in TBE) and analyzed by autoradiography. Plasmids YB-1 af®nity puri®cation or `pull down' assay Several GST-fusion proteins were constructed for the `pull down' or YB-1 af®nity puri®cation assay. Human YB-1 Approximately 10 human 293 cells were lysed in 10 mM coding sequence was ampli®ed by PCR with appropriate Tris±HCl (pH 7.5), 1% (v/v) Triton X-100, 150 mM NaCl, 5 oligonucleotides for subsequent cloning into the BamHI/ mM EDTA, 50 mM NaF in the presence of a protease inhibitor EcoRI sites of the pGEX-2TK vector. In addition, YB-1 cocktail (Boehringer Mannheim) at 4°C for 20 min. Cells were cDNA was cut with SmaI and EcoRI (amino acid residues sonicated three times for 5 s and cell debris were spun down. 39±312 of YB-1), SmaI and SalI (residues 39±205), SalI and Af®nity matrices were prepared by immobilizing GST alone or GST±YB-1 fusion proteins on glutathione±Sepharose beads EcoRI (residues 205±312) and these fragments were cloned (Amersham Pharmacia) as described (34). Freshly prepared into the appropriate modi®ed restriction sites in the pGEX- cell lysates were incubated with the af®nity matrices overnight 2TK vector. A pGEX-2TK construct coding for a GST-fusion exo at 4°C. After extensive washing with lysis buffer, bound peptide containing the exonuclease domain of p53 (p53 ) proteins were released by boiling in SDS sample buffer and was kindly provided by the laboratory of Jacques Co à te  (Centre analyzed by SDS±PAGE and western blotting. de Recherche en Cance Ârologie, Que Âbec City, Canada). ProScan analyses on p53 have indicated that its exonuclease Indirect immuno¯uorescence domain is within amino acids 185±290. Plasmids were transfected into BL21 bacteria for fusion protein production. Human 293 embryonic kidney cells were maintained in Proteins were visualized by Coomassie staining when DMEM supplemented with 10% fetal bovine serum. indicated. They were plated on glass coverslips the day before the 318 Nucleic Acids Research, 2004, Vol. 32, No. 1 Table 1. DNA Structures Figure 1. Characterization of puri®ed human YB-1 on gels stained by immuno¯uorescence experiments. Cells were ®xed in Coomassie. (Top, left) This gel contains GST±YB-1 and GST proteins cold methanol for 10 min and permeabilized with 0.15% puri®ed on glutathione±Sepharose beads. (Top, right) GST±YB-1 was Triton X-100 at 4°C for 10 min. After washing with PBS, cells treated with thrombin to remove the GST portion of the fusion protein as described in Materials and Methods. The size of YB-1 is ~42±44 kDa. were blocked with 2% milk at 4°C for 30 min. After blocking, (Bottom) These gels contain YB-1 proteins eluted from a Superdex-200 gel antibodies (1:250 for anti-Ku80, 1:50 for anti-MSH2, 1:2000 ®ltration column with or without 2 mM ATP in the buffer. The numbers for anti-WRN or 1:500 for anti-YB-1) diluted in blocking above each gel represent the position of marker proteins eluted from such a buffer were applied to the coverslips and incubated overnight column. The numbers below the gel represent the fraction number. at 4°C. Coverslips were washed with PBS and incubated for 1 h at room temperature with rhodamine-secondary antibody for WRN or YB-1 detection (Santa Cruz), and FITC-secondary of ~42±44 kDa (Fig. 1, top right). Puri®ed YB-1 was loaded antibody (Santa Cruz) for Ku80 or MSH2 detection. After onto a Superdex-200 column and the eluted proteins were washing, coverslips were mounted on glass slides. Slides were analyzed on a 10% SDS±polyacrylamide gel. Coomassie viewed at 603 magni®cation (1.4NA oil-immersion 603 staining revealed that YB-1 eluted in fractions 12±18. In a objective) and zoomed 23 for image acquisition on a Nikon parallel experiment, molecular weight protein markers were inverted diaphot confocal microscope equipped with Krypton/ also eluted from a Superdex-200 column and analyzed on a Argon lasers (488 and 568 nm). Images were captured with a gel. As indicated in Figure 1, the YB-1 peptide was recovered BioRad MRC1024 confocal microscopy system. Finally, in multiple peaks that co-eluted with molecular weight images were colored and merged in Adobe photoshop. markers of 530, 232, 140 and 67 kDa (bottom). However, YB-1 co-eluted with the molecular weight marker of 67 kDa upon addition of 2 mM ATP in the solution. A dimer of YB-1 would be ~80 kDa which elutes with the 67 kDa molecular RESULTS weight marker. These results indicate that in addition to Puri®ed human YB-1 forms oligomers in solution monomers and dimers, YB-1 can form trimers, hexamers and oligomers of 12 YB-1 molecules in solution. In the presence of Recent studies have indicated that YB-1 may act as multimers ATP, only dimers are formed. and that two C-tail domains of YB-1 are required for homodimerization (28,30). To determine whether YB-1 can Human YB-1 separates double-stranded DNA in the oligomerize in solution, the GST±YB-1 fusion protein was presence or absence of ATP produced in bacteria and puri®ed with glutathione±Sepharose beads (Fig. 1, top left). As the GST portion of this fusion YB-1 is known to create single strand regions in the DRA protein could modify the behavior of YB-1 in vitro, the GST promoter (28). We thus tested whether puri®ed YB-1 can portion was cleaved off with thrombin (see Materials and separate the strands of different DNA duplexes (Table 1). A Methods). This puri®ed YB-1 protein has a molecular weight forked DNA structure was ®rst tested in the presence or Nucleic Acids Research, 2004, Vol. 32, No. 1 319 absence of ATP. ATP was also tested in these reactions because PROSCAN analysis of YB-1 protein has indicated that residues 28±58 form a motif with 69% homology to an ATP-binding signature region. As shown in Figure 2A, puri®ed YB-1 is capable of separating the strands like a DNA helicase enzyme. The amount of displaced radioactive strand was YB-1 concentration dependent (Fig. 2A). YB-1 proteins from fractions 12, 15, 16 and 18 of the Superdex-200 column (Fig. 1) were tested. All these fractions showed similar strand displacement activities (data not shown). YB-1 was capable of separating the strands of a forked DNA structure without ATP (Fig. 2B). However, this activity was doubled in the presence of 2 mM ATP in the reaction (Fig. 2B and C). The percentage of displacement was calculated from several reactions by excising the corresponding bands from the gels and counting the amount of radioactivity with a b-counter. Similar results were obtained with different DNA structures (data not shown). These observations indicate that puri®ed YB-1 can separate the strands of a forked DNA structure even in the absence of ATP. Finally, ATPase assays were performed and no ATP hydrolysis was detected with our puri®ed YB-1 in the presence or absence of DNA molecules (data not shown). Human YB-1 can separate the strands of either blunted, 5¢ or 3¢ recessed DNA structures The capacity of YB-1 to separate DNA strands was examined with different duplex structures (see Table 1). As shown in Figure 3A±E, YB-1 was equally capable of separating DNA strands from a 22mer duplex containing either blunt ends, a 5¢ recessed or a 3¢ recessed end. In addition, YB-1 displaced GC-rich or AT-rich radioactive strands equally well (see Table 1 for oligonucleotide sequences). Overall, an 8±16% strand displacement was observed with different duplex sequences and this independently of the structures at the ends of the DNA substrates (Fig. 3C). YB-1 is a transcription factor that binds strongly to an inverted CCAAT box (1). A 22mer duplex containing this sequence (Y-box sequence) was also examined. As shown in Figure 3F, YB-1 separated the two strands of the Y-box duplex very ef®ciently. Close to 50% of total duplex was separated by YB-1 (Fig. 3G). All these results indicate that YB-1 can separate a 22mer duplex DNA independently of the nucleotide Figure 2. Strand separation activity of human YB-1 on a forked DNA sequence. However, YB-1 has a very strong preference for a duplex structure. (A) Increasing amounts of puri®ed YB-1 proteins duplex DNA containing an inverted CCAAT box. (indicated in ng) or 40 ng of GST were incubated with a radioactive 22 bp forked duplex under standard conditions for helicase activity (see Materials Human YB-1 does not separate a 36mer DNA duplex and Methods) for 30 min at 37°C. Reactions were stopped in the appropriate with no inverted CCAAT box sequence dye buffer and the DNA products were run on a 12% native polyacrylamide gel. The double-stranded and single-stranded structures are depicted on the The DNA substrates used in the strand separation reactions in left. The asterisk represents the labeled strand at its 5¢ end. The triangle the preceding sections contained <23 bp. We therefore asked represents heat denatured DNA. (B) YB-1 (40 ng) was incubated with a whether the length of the DNA duplex affects YB-1 strand radioactive 22 bp forked duplex as in (A), in the presence or absence of separation activity. As indicated in Figure 4A (top), YB-1 was 2 mM ATP. (C) Histogram representation of the YB-1 strand separation reactions performed in (B). All experiments were done in duplicates with unable to separate the strands of a 36 bp DNA duplex. 40 ng of puri®ed human YB-1. The percentage strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of Human YB-1 separates 36mer DNA duplex treated with displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was cisplatin or containing mismatches calculated for each band of the gel as indicated in Materials and Methods. It has been shown that YB-1 binds preferentially to cisplatin- modi®ed DNA (25). Thus, the 36mer duplex structure was strands of the 36mer DNA substrate containing cisplatin incubated overnight with cisplatin and the resulting modi®ed DNA was used in YB-1-mediated strand separation reaction. modi®cations. Approximately 14% of cisplatin-modi®ed As shown in Figure 4A, YB-1 was capable of separating the strands were separated by YB-1 (Fig. 4D). 320 Nucleic Acids Research, 2004, Vol. 32, No. 1 YB-1-mediated strand separation reaction was also exam- ined on duplex DNA containing one or two mispaired bases. Figure 4B indicates that YB-1 can separate the strands of a 36mer duplex DNA containing one mismatch. Approximately 9% of the total DNA duplex was separated (Fig. 4D). Finally, a 22mer DNA duplex structure (GC rich) with two mismatches was tested. Up to 25% of this DNA substrate was separated into single-stranded DNA (Fig. 4C and D). All these results indicate that YB-1 can separate duplex DNA with either mispaired bases or cisplatin-modi®ed lesions. The DNA strand separation activity of human YB-1 requires amino acids 40±205 To demonstrate that the strand separation activity was intrinsic to YB-1, a series of GST±YB-1 fusion proteins were analyzed (Fig. 5A). Although we did see some protein degradation, all fusion proteins were treated with thrombin and the puri®ed YB-1 fragments were assayed on the Y-box duplex. These experiments were repeated three times. As seen in Figure 5B, full-length YB-1 separated the two strands as shown before. YB-1 proteins lacking the ®rst 40 amino residues (including part of the potential ATP-binding site) showed weak activity on the gel. Approximately 10% of the DNA duplex was separated by this mutant compared with 27% for the full- length YB-1 (Fig. 5C). The strand separation activity of this mutant YB-1 (amino acids 40±313) was similar in the presence or absence of ATP (Fig. 5B, last panel on the right). In contrast, the C-terminus part of YB-1 (amino acids 204±313) did not exhibit strand separation activity. Finally, a YB-1 fragment corresponding to residues 39±205 (Fig. 5B, middle) exhibited a very weak activity (<5% strand displace- ment in Fig. 5C). As a negative control, a GST-fusion peptide exo ) was containing the exonuclease domain of p53 (p53 assayed under the same conditions and no strand separation activity was detected (Fig. 5B and C). These results indicate that the strand separation activity is localized to the region of YB-1 between amino acids 40 and 205. However, both the N- terminus and C-terminus portions of YB-1 are obviously required for optimal activity. Similar results were obtained with a forked DNA substrate (data not shown). Human YB-1 exhibits endonucleolytic activities on double-stranded DNA Exposition of the native gels in Figure 5B for an additional 3 days showed some degradation of the DNA substrates (data Figure 3. Strand separation activity of YB-1 on different duplex structures. not shown). This suggests that nucleolytic activity was also (A±F) The indicated amounts (in ng) of puri®ed human YB-1 proteins were incubated with the indicated radioactive DNA substrates under standard present in the strand separation reactions. Consistent with this conditions for helicase activity (see Materials and Methods) for 30 min at observation, a recent report has indicated that GST±YB-1 37°C. Again, 40 ng of GST proteins were used as negative controls when chimeric proteins exhibit 3¢ to 5¢ exonuclease activity (30). indicated. Reactions were stopped in the appropriate dye buffer and the However, this activity was described only for single-stranded DNA products were run on a 12% native polyacrylamide gel. The double- stranded and single-stranded DNA structures with their sequence character- DNA molecules. Similar analyses were thus performed with istics are depicted on the left of each autoradiogram. The 5¢-labeled strand our puri®ed YB-1 proteins on different duplex substrates. To of the duplex is represented by an asterisk. (A) and (B) represent reactions determine if our puri®ed YB-1 protein exhibited any with blunt-ended DNA. (C) and (D) represent reactions with a structure hav- exonuclease activity, a single-stranded DNA substrate was ing a 3¢ recessed end. (E) represents reactions with a structure having a 5¢ ®rst analyzed. As shown in Figure 6A (left), YB-1 degraded recessed end. In (F), the Y-box sequence is an inverted CCAAT sequence. All DNA substrates contain 22 paired bases. (G) Histogram representation the single-stranded DNA. However, the autoradiogram had to of the YB-1 strand separation reactions performed in (A±F). The strand be exposed for 3 days to detect the exonuclease activity. displacement data were obtained from experiments done in duplicate with A double-stranded DNA substrate was incubated with YB-1 40 ng of the indicated puri®ed proteins. The percentage strand displacement to examine the exonuclease activity. As shown in the second is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was panel of Figure 6A, the pattern of degradation was different calculated for each band of the gel as indicated in Materials and Methods. from the single-stranded substrate. To demonstrate that the Nucleic Acids Research, 2004, Vol. 32, No. 1 321 Figure 4. Strand separation activity of YB-1 on duplex structures containing mismatches or cisplatin cross-links. (A±C) The indicated amounts (in ng) of puri®ed human YB-1 proteins were incubated with the indicated radioactive DNA substrates under standard conditions for helicase activity (see Materials and Methods) for 30 min at 37°C. Forty nanograms of GST were used as a negative control. The panels in (A) represent strand separation reactions performed with a 36 bp duplex. The panel on the right contains a reaction performed with a 36mer duplex treated with cisplatin. In (B), reactions were performed with a 36mer duplex containing one base mismatch. In (C), reactions were performed with a 22mer duplex structure containing two nucleotide mismatches. (D) Histogram representation of the YB-1 strand separation reactions performed in (A). The strand displacement data were obtained from experiments done in duplicate with 40 ng of the indicated puri®ed proteins. The percentage strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was calculated for each band of the gel as indicated in Materials and Methods. exonuclease activity was intrinsic to YB-1, a series of puri®ed was stronger with a YB-1 molecule lacking the ®rst 40 amino YB-1 fragments were analyzed. Full-length YB-1 exhibited acid residues. These results indicate that the ®rst 40 amino nucleolytic activity on duplex DNA (Y-box duplex) (Fig. 6A, acid residues of YB-1 will affect its nicking activity. In right). The absence or presence of ATP did not affect the contrast, a YB-1 fragment comprising residues 39±205 did not nucleolytic activities of full-length YB-1 (data not shown). A exhibit nuclease activity on this DNA substrate. As expected, YB-1 protein lacking the ®rst 40 amino acid residues also the exonuclease domain of p53 exhibited a strong nuclease exhibited similar nucleolytic activity but at a lower ef®ciency activity with a cleavage pattern similar to the N-terminus (Fig. 6B). The C-terminus portion of YB-1 (residues 204±313) truncated YB-1 peptide (Fig. 7A). Similar experiments were did not show nuclease activity. A peptide fragment comprising performed on the same DNA substrate except that the opposite residues 39±205 of YB-1 cleaved the 5¢ radioactive nucleotide strand was labeled at the 5¢ end (Fig. 7B). Although labeling of exo of the duplex. Finally, the exonuclease domain of p53 (p53 ) the short strand did show some heterogeneity of the probe, showed very little nuclease activity on this Y-box DNA only the full-length YB-1 exhibited a weak nucleolytic duplex. These results indicate that YB-1 introduces nicks or activity. Less than 2% of the probe was cleaved with 40 ng breaks into double-stranded DNA molecules. of YB-1. The cleavage pattern indicated nicking near the Additional experiments were performed with YB-1 frag- labeled 5¢ end of the strand. None of the mutated YB-1 peptide exo ments (fragments exhibiting nicking activities in the previous showed nicking on the shorter stand of the duplex. The p53 section) on DNA substrates labeled on either strand to peptide cleaved the labeled 5¢ of this substrate (Fig. 7B, last compare the cleavage patterns. As shown in Figure 7, full- lane on the right panel). These results suggest that YB-1 can length YB-1 also cleaved a DNA duplex containing a 3¢ introduce double-stranded breaks and exhibit endonucleolytic overhang. However, the cleavage pattern was different from activities. However, this activity was weak as the autoradio- the patterns obtained with duplex DNA possessing blunt ends grams had to be exposed for 3 days to detect the cleavage (compare Figs 6A with 7A). Interestingly, the cleavage pattern products. 322 Nucleic Acids Research, 2004, Vol. 32, No. 1 Figure 5. Strand separation activity of different mutant YB-1 on a DNA duplex structure. (A) Characterization of the different mutant GST±YB-1 fusion proteins (before thrombin treatments) on gels stained by Coomassie. The YB-1 amino acids present in the fusion proteins are indicated above each lane. (B) The different GST-fusion proteins were cleaved with thrombin as described in the Materials and Methods and the cleaved products were used in strand exo displacement reactions. Approximately 40 ng of the puri®ed YB-1, mutant YB-1 proteins or p53 peptide (exonuclease domain) were incubated with a radioactive 22 bp duplex containing a Y-box sequence under standard conditions for helicase activity (see Materials and Methods) for 30 min at 37°C in the absence or presence of 2 mM ATP. Reactions were stopped in the appropriate dye buffer and the DNA products were analyzed on a 12% native polyacrylamide gel. The double-stranded and single-stranded structures are depicted on the left. The asterisk represents the labeled strand at its 5¢ end. The triangle represents heat denatured DNA. (C) Histogram representation of the strand separation reactions performed in (B). All experiments were done in exo duplicate with 40 ng of puri®ed YB-1, mutant YB-1 proteins or p53 (exonuclease domain) in reaction buffer containing 2 mM ATP. The percentage of strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was calculated for each band of the gel as indicated in Materials and Methods. Several DNA repair proteins bind to human YB-1 that the GST±YB-1 fusion construct can be used for af®nity immobilized on a matrix chromatography experiments. Poly(ADP-ribose) polymerase-1 (PARP-1), REF1 endonu- YB-1 binds to cisplatin-modi®ed DNA (25) and can separate clease (also known as APE1), and XRCC1 are proteins the strands of such modi®ed DNA (preceding sections). It also involved in short or long patch excision repair pathways (36± interacts with PCNA (25) and an endonuclease involved in 38). Western blot analyses revealed that none of these proteins base excision repair (27). These observations suggest that YB- bound to the GST±YB-1 matrix (Fig. 8 and data not shown). 1 may be involved in certain aspects of DNA repair. To As YB-1 can separate the strands of a duplex DNA identify other DNA repair proteins that may interact with YB- containing mismatches, antibodies against proteins known to 1, whole cell extracts from the human 293 embryonic kidney be involved in the mismatch repair pathway were also tested. cell line were loaded on af®nity column containing the GST± The presence of MLH1, MSH2 and PMS2 was examined. As YB-1 fusion protein immobilized on glutathione±Sepharose indicated in Figure 8, only MSH2 bound to the GST±YB-1 beads. GST immobilized on glutathione±Sepharose beads was beads but not to the GST control beads. used as a control. Bound proteins were analyzed by western Proteins involved in homologous recombination and non- blots with antibodies speci®c to different DNA repair proteins. homologous end-joining were also examined. These include Western blots were ®rst examined with antibodies against RAD51, RAD51C and RAD52 for the homologous recombi- proteins known to interact with YB-1 or known to be in the nation pathway and DNA-PK, Ku80 and WRN for the non- same cellular complex. Nucleolin and ALY were analyzed as homologous end-joining pathway (39±41). RAD51, RAD51C, both are found in prespliceosomes with YB-1 (35). As shown RAD52 and DNA-PK did not bind the GST±YB-1 beads (data in Figure 8, both nucleolin and ALY bound to the GST±YB-1 not shown). In contrast, Ku80 and WRN proteins bound beads but not to the GST control beads. These results indicate strongly to the GST±YB-1 matrix but not to the GST control. Nucleic Acids Research, 2004, Vol. 32, No. 1 323 Finally, an antibody against DNA polymerase d was also tested on the western blots. As indicated in Figure 8, DNA polymerase d bound to the GST±YB-1 matrix with low af®nity as suggested by the weak signal observed on the blot. Altogether, these results indicate that several proteins involved in DNA repair bind YB-1 in vitro. These ®ndings strongly support the involvement of YB-1 in speci®c DNA repair pathways. Re-localization of YB-1 to nuclear areas containing DNA repair proteins in cisplatin-treated cells Co-localization experiments with antibodies against Ku80 and YB-1 were performed on intact 293 cells that had been treated with or without 7.5 mg/ml of cisplatin for 2 h. FACS analyses revealed no evidence of an increase in apoptotic ®gures in cisplatin-treated cultures compared with untreated cells at the 2 h time point (data not shown). Thus, we were analyzing cells before the onset of apoptosis. As shown in Figure 9A, YB-1 is found mainly in the cytoplasm of untreated 293 cells with some nuclear localization. In contrast, Ku80 is only localized to the nucleus of untreated cells in a diffuse pattern (Fig. 9B). Immuno¯uorescence images for both antibodies were captured in separate channels and overlaid to see the extent of co-localization. As shown in Figure 9C, few nuclear foci contained both proteins. In the presence of cisplatin, some YB-1 proteins translocated to the nucleus (Fig. 9D). The Ku80 distribution pattern was similar in both cisplatin-treated and untreated cells (compare Fig. 9B with E). Interestingly, >90% of YB-1 proteins found in the nucleus of cisplatin-treated cells co-localized with Ku80 (yellow ¯uorescence in Fig. 9F). These results indicate that upon cisplatin treatment, YB-1 re-localizes into multiple nuclear foci containing Ku80 proteins. Pre-immune serum from the YB-1 immunized rabbit did not show such results (data not shown). Western analysis on 293 whole cell lysates indicated that the rabbit immune serum used in the immuno¯uorescence study recognizes only one band speci®c to YB-1 (Fig. 9G). However, the anti-YB-1 antibody immunoprecipitated YB-1 from 293 cell lysate only poorly (data not shown). Hence, co-immunoprecipitation could not be performed with this antibody. Co-localization experiments were also performed with antibodies against MSH2 and YB-1 (Fig. 10). Again, YB-1 is found in the cytoplasm of untreated 293 cells. There are few foci in the nucleus containing YB-1 (Fig. 10A). Upon cisplatin treatment, YB-1 re-localizes into the nucleus of 293 cells (Fig. 10D). As shown in Figure 10B and E, MSH2 is found in the nucleus of untreated and cisplatin-treated 293 cells in a weak diffuse pattern. Merging the images shows that there are few nuclear foci containing both YB-1 and MSH2 proteins (yellow ¯uorescence in Fig. 10C) in untreated 293 cells. However, the number of nuclear foci containing both MSH2 Figure 6. Exonuclease and DNA nicking activity of YB-1. (A) Forty and YB-1 proteins increased substantially in cisplatin-treated exo nanograms of full-length, mutant YB-1 or p53 (exonuclease domain of cells (Fig. 10F). Approximately 30% of MSH2 nuclear foci p53) were incubated for 30 min at 37°C as described in Materials and contained YB-1 proteins upon cisplatin treatment. These Methods with the different DNA substrates indicated above each gel. results indicate that upon cisplatin treatment, a proportion of Reactions were stopped in the appropriate dye buffer and cleaved DNA cellular YB-1 re-localizes into multiple nuclear areas con- products were analyzed on 14% denaturing polyacrylamide gels to analyze the exonuclease activity. (B) Histogram representation of the YB-1 nuclease taining MSH2 proteins. activity performed in (A) with the duplex DNA substrate. The percentage Both anti-WRN and YB-1 antibodies were generated in nuclease activity was calculated by scanning the ®gures and using the rabbits. Consequently, to detect co-localization of YB-1 with formula: intensities of cleaved products/(intensities of cleaved products + WRN proteins in treated 293 cells, a speci®c epitope-tagged intensity of uncleaved probe) 3 100. Results were calculated from duplicate experiments. version of the proteins will be constructed. This is part of 324 Nucleic Acids Research, 2004, Vol. 32, No. 1 exo Figure 7. Endonucleolytic activity of YB-1. (A and B) Forty nanograms of full-length, mutant YB-1 or p53 (exonuclease domain of p53) peptides were incubated with the indicated DNA substrates in reaction buffer for 30 min at 37°C as described in Materials and Methods. DNA products were analyzed on 14% denaturing polyacrylamide gels. The brackets on the right side of each gel indicate the cleavage products of each probe by the proteins. Experiments were repeated twice. (C) Histogram representation of the YB-1 nuclease activity performed in (B). The percentage nuclease activity was calculated by scanning the ®gures and using the formula: intensities of cleaved products/(intensities of cleaved products + intensity of uncleaved probe) 3 100. Results were calculated from duplicate experiments. another study. Finally, the antibody against DNA polymerase were performed with our puri®ed YB-1 and it did not d used in this study does not work in immuno¯uorescence. hydrolyze ATP even in the presence of DNA. Our puri®ed YB-1 can separate the strands of different DNA duplex molecules under buffer conditions used for DISCUSSION helicase assays (44). The strand separation activity of YB-1 is increased 2-fold in the presence of ATP (Fig. 2). Importantly, Recent studies have implicated YB-1 in genome stability as it a mutant YB-1 lacking part of the potential ATP-binding interacts with proteins involved in DNA repair such as PCNA domain did not exhibit increased strand separation activity in (25), human endonuclease III (27) and p53 (42,43). YB-1 also the presence of ATP (Fig. 5B). Thus, our data indicate that exhibits exonuclease activities on single-stranded DNA (30). binding of ATP changes the conformation of YB-1 and affects In this report, we investigated the strand separation and its strand separation activities. YB-1 does not seem to have a nucleolytic properties of YB-1 on different DNA duplex preferred directionality like an helicase. It can separate the substrates in vitro. For such analysis, we produced GST±YB-1 strands from a duplex molecule with either a 3¢ or 5¢ recessed in bacteria and designed a puri®cation scheme involving the end, blunt-ended DNA, or a forked DNA structure equally removal of the GST portion of the fusion construct with well. The strongest strand separation activity was observed thrombin and a gel ®ltration step. Elution from the Superdex- with DNA duplexes containing an inverted CCAAT box. 200 gel ®ltration column has indicated that YB-1 can readily Additional analyses with deletion mutants have indicated that form trimers, hexamers and oligomers of 12 molecules in the DNA strand separation activity is intrinsic to YB-1 protein solution. However, the presence of ATP changed the con- and the active site is situated between amino acid residues 40 formation of the protein as YB-1 mainly formed dimers, a and 205. Importantly, this region of YB-1 is known to contain complex more likely to be present in vivo. This con®rms the DNA-binding domain (30). The length of the duplex earlier reports suggesting that YB-1 is acting as an oligomer in vivo (28,30). PROSCAN analyses of YB-1 amino acid structure affects YB-1 strand separation activities. YB-1 can sequence have indicated that it possesses an ATP-binding separate a 22 bp molecule but not a 36 bp molecule. However, signature at its N-terminus (residues 28±58). ATPase assays it will separate a 36mer containing one or two base Nucleic Acids Research, 2004, Vol. 32, No. 1 325 Figure 8. Immunoblots against DNA repair proteins bound to GST±YB-1 af®nity Sepahrose beads. Human 293 embryonic kidney whole cell extracts (WCE) were incubated with either 50 mg of GST±YB-1 or GST linked glutathione±Sepharose beads overnight. Proteins bound to the af®nity beads were analyzed by SDS±PAGE with antibodies against the indicated proteins on the left of each blot. mismatches or cisplatin modi®cations. This strongly suggests that YB-1 will bind and process DNA molecules with such lesions. YB-1 is not only capable of separating the strands, but it can also cleave double-stranded DNA molecules. However, this YB-1 nicking activity is weaker than the strand separation activities. Unlike the strand separation activity, it takes several days of exposition to see the nucleolytic cleavage pattern on an autoradiogram. This suggests that the DNA strand separ- ation activity of YB-1 is stronger than its nucleolytic activity under the in vitro conditions employed in this study. Our Figure 9. Co-localization of YB-1 and Ku80 by immuno¯uorescence. preliminary observations indicate that YB-1 cleavage activ- Human 293 embryonic kidney cells were untreated (A±C) or incubated with ities seem to be sequence and/or structure dependent. 7.5 mg/ml of cisplatin for 2 h (D±F), then ®xed and subsequently incubated simultaneously with rabbit anti-YB-1 and mouse anti-Ku80 as described in Noticeably, the exonucleolytic activity on a DNA duplex Materials and Methods. Anti-rabbit rhodamine-labeled and anti-mouse substrate containing a 3¢ overhang increased dramatically FITC-conjugated secondary antibodies were used to visualize YB-1 and when YB-1 lacked the ®rst 40 N-terminus amino acid Ku80 by confocal microscopy at 568 and 488 nm, respectively. Images de- residues. This increased nucleolytic activity was not observed pict representative cells from each of the non-treated and cisplatin-treated cultures. In the merged images (C and F) a yellow color appears where on blunt duplex. The reason for the enhanced nucleolytic YB-1 (red) and Ku80 (green) ¯uorescence signals coincide. In (G), western activity with the mutant YB-1 (40±313) peptide is unknown. analysis of 293 total cell lysate with anti-YB-1 and pre-immune sera. Presumably the deletion of the ®rst 40 amino acids of YB-1 has affected its interaction with different DNA structures in vitro. The only repair proteins that bound the YB-1 matrix in our Our data on the enzymatic activities of YB-1 toward duplex study were DNA polymerase d, MSH2, Ku80 and WRN. In with mismatches and cisplatin modi®cations suggest that YB- addition, immuno¯uorescence studies have indicated re- 1 might be involved in base excision repair and mismatch localization of YB-1 to regions of the nucleus containing repair pathways. It was thus logical to ask whether YB-1 can either Ku80 or MSH2 proteins in cisplatin-treated cells (Figs 9 bind to proteins involved in such DNA repair pathways. and 10). Accordingly, af®nity chromatography experiments were The MSH2/YB-1 interaction is an interesting observation as performed on a YB-1 matrix with 293 whole cell lysates. cancer cell lines with a mutation in the MSH2 gene are also 326 Nucleic Acids Research, 2004, Vol. 32, No. 1 It is believed that WRN protein is a dual exonuclease/ helicase enzyme not only involved in non-homologous end- joining, but also in long patch base excision repair (42). Mutations in the WRN gene are responsible for the human progeroid disorder Werner syndrome. Importantly, WRN protein translocates to speci®c nuclear foci upon UV treatment (49). These foci are believed to be sites of DNA repair. The co- localization of WRN and YB-1 and their direct interaction at sites of DNA repair upon UV treatment require experiments with good antibodies against both proteins. In summary, we have found that YB-1 can separate different DNA duplex structures and this activity is increased when duplex molecules contain either mispaired bases or cisplatin modi®cations. YB-1 exhibits endonucleolytic activ- ity in addition to its exonuclease. Finally, YB-1 binds in vitro proteins involved in both base excision repair and mismatch repair pathways suggesting an involvement of YB-1 in these repair pathways. Additional in vitro as well as in vivo analyses are required to con®rm the functional interactions between YB-1 and these DNA repair proteins. ACKNOWLEDGEMENTS We are grateful to Dr P. E. DiCorleto from the Cleveland Clinic Foundation (Cleveland, OH) for the rabbit pre-immune à  and anti-YB-1 antibodies and to Jacques Cote from the Centre   de Recherche en Cancerologie (Quebec City) for the GST-p53 construct. We thank F. Deschenes for technical assistance and Figure 10. Co-localization of YB-1 and MSH2 by immuno¯uorescence. Dr P. O. de Campos-Lima for his helpful discussion. This Human 293 embryonic kidney cells were untreated (A±C) or incubated with 7.5 mg/ml of cisplatin for 2 h (D±F), then ®xed and subsequently incubated work was supported by a grant from the Canadian Institutes of simultaneously with rabbit anti-YB-1 and mouse anti-MSH2 as described in Health Research (CIHR) to M.L. M.L. is a scholar of the Materials and Methods. Anti-rabbit rhodamine-labeled and anti-mouse CIHR. FITC-conjugated secondary antibodies were used to visualize YB-1 and MSH2 by confocal microscopy at 568 and 488 nm, respectively. Images de- pict representative cells from each of the non-treated and cisplatin-treated cultures. In the merged images (C and F) a yellow color appears where YB- REFERENCES 1 (red) and MSH2 (green) ¯uorescence signals coincide. 1. Didier,D.K., Schiffenbauer,J., Woulfe,S.L., Zacheis,M. and Schwartz,B.D. (1988) Characterization of the cDNA encoding a protein binding to the major histocompatibility complex class II Y box. Proc. resistant to platinum-containing compounds (45). This pro- Natl Acad. Sci. USA, 85, 7322±7326. vides a link between the activity of YB-1 and the mismatch 2. Travali,S., Ku,D.-H., Rizzo,M.G., Ottavio,L., Baserga,R. and repair protein MSH2 in the processing of cisplatin-modi®ed Calabretta,B. (1989) Structure of the human gene for the proliferating DNA. Future experiments with puri®ed MSH2 will determine cell nuclear antigen. J. Biol. Chem., 264, 7466±7472. 3. Lipson,K.E., Chen,S.-T., Koniecki,J., Ku,D.-H. and Baserga,R. (1989) whether it can bind directly to YB-1 and alter its activities. S-phase speci®c regulation by deletion mutants of the human thymidine DNA polymerase d is another factor (with PCNA and EXO1) kinase promoter. Proc. Natl Acad. Sci. USA, 86, 6848±6852. associated with mismatch repair (46). It is interesting to note 4. Pearson,B.E., Nasheuer,H.P. and Wang,T.S.-F. (1991) Human DNA that DNA polymerase d bound our GST±YB-1 matrix and that polymerase alpha gene: sequences controlling expression in cycling and PCNA interacts with YB-1 in vivo (25). Again, these results serum-stimulated cells. Mol. Cell. Biol., 11, 2081±2095. 5. Zou,Y. and Chien,K.R. (1995) EFIA/YB-1 is a component of cardiac point to a potential role for YB-1 in the mismatch repair HF-1A binding activity and positively regulates transcription of the pathway. myosin light-chain 2v gene. Mol. Cell. Biol., 15, 2972±2982. Ku80 has been reported to have a role in transcriptional 6. Lasham,A., Lindridge,E., Rudert,F., Onrust,R. and Watson,J. (2000) reinitiation by RNA polymerase II in addition to double-strand Regulation of the human fas promoter by YB-1, Puralpha and AP-1 transcription factors. Gene, 252, 1±13. break repair (47). As YB-1 is a transcription factor, it is 7. Norman,J.T., Lindahl,G.E., Shakib,K., En-Nia,A., Yilmaz,E. and possible that the observed interaction with Ku80 is in the Mertens,P.R. (2001) The Y-box binding protein YB-1 suppresses context of transcription in cells and not DNA repair. However, collagen alpha 1(I) gene transcription via an evolutionarily conserved Ku antigen has also been shown to down regulate the activity regulatory element in the proximal promoter. J. Biol. Chem., 276, of the base excision repair protein endonuclease III in 29880±29890. 8. Fukada,T. and Tonks,N.K. (2003) Identi®cation of YB-1 as a regulator of the context of DNA molecules with closely opposed PTP1B expression: implications for regulation of insulin and cytokine lesions (48). Thus, Ku antigen will also affect some aspects signaling. EMBO J., 22, 479±493. of base excision repair in cells. Additional analyses on a 9. Mertens,P.R., Alfonso-Jaume,M.A., Steinmann,K. and Lovett,D.H. Ku/YB-1 complex are required to distinguish between these (1999) YB-1 regulation of the human and rat gelatinase A genes via possibilities. similar enhancer elements. J. Am. Soc. Nephrol., 10, 2480±2487. Nucleic Acids Research, 2004, Vol. 32, No. 1 327 10. Goldsmith,M.E,, Madden,M.J., Morrow,C.S. and Cowan,K.H. (1993) A 28. MacDonald,G.H., Itoh-Lindstrom,Y. and Ting,J.P. (1995) The Y-box consensus sequence is required for basal expression of the human transcriptional regulatory protein, YB-1, promotes single-stranded multidrug resistance (mdr1) gene. J. Biol. Chem., 268, 5856±5860. regions in the DRA promoter. J. Biol. Chem., 270, 3527±3533. 11. Ohga,T, Uchiumi,T., Makino,Y., Koike,K., Wada,M., Kuwano,M. and 29. Swamynathan,S.K., Nambiar,A. and Guntaka,R.V. (1998) Role of single- Kohno,K. (1998) Direct involvement of the Y-box binding protein YB-1 stranded DNA regions and Y-box proteins in transcriptional regulation of in genotoxic stress-induced activation of the human multidrug resistance viral and cellular genes. FASEB J., 12, 515±522. 1 gene. J. Biol. Chem., 273, 5997±6000. 30. Izumi,H., Imamura,T., Nagatani,G., Ise,T., Murakami,T., Uramoto,H., 12. Hartmuth,K., Urlaub,H., Vornlocher,H.P., Will,C.L., Gentzel,M., Torigoe,T., Ishiguchi,H., Yoshida,Y., Nomoto,M., Okamoto,T., Wilm,M. and Luhrmann,R. (2002) Protein composition of human Uchiumi,T., Kuwano,M., Funa,K. and Kohno,K. (2001) Y box-binding prespliceosomes isolated by a tobramycin af®nity-selection method. protein-1 binds preferentially to single-stranded nucleic acids and Proc. Natl Acad. Sci. USA, 99, 16719±16724. exhibits 3¢®5¢ exonuclease activity. Nucleic Acids Res., 29, 1200±1207. 13. Raffetseder,U., Frye,B., Rauen,T., Jurchott,K., Royer,H.D., Jansen,P.L. 31. Lebel,M., Spillare,E.A., Harris,C.C and Leder,P. (1999) The Werner and Mertens,P.R. (2003) Splicing factor SRp30c interaction with Y-box syndrome gene product co-puri®es with the DNA replication complex protein-1 confers nuclear YB-1 shuttling and alternative splice site and interacts with PCNA and topoisomerase I. J. Biol. Chem., 274, selection. J. Biol. Chem., 278, 18241±18248. 37795±37799. 14. Stickeler,E., Fraser,S.D., Honig,A., Chen,A.L., Berget,S.M. and 32. Bennett,R.J., Sharp,J.A. and Wang,J.C. (1998) Puri®cation and Cooper,T.A. (2001) The RNA binding protein YB-1 binds A/C-rich exon characterization of the Sgs1 DNA helicase activity of Saccharomyces enhancers and stimulates splicing of the CD44 alternative exon v4. cerevisiae. J. Biol. Chem., 273, 9644±9650. EMBO J., 20, 3821±3830. 33. Moggs,J.G., Yarema,K.J., Essigmann,J.M. and Wood,R.D. (1996) 15. Davydova,E.K., Evdokimova,V.M., Ovchinnikov,L.P. and Hershey,J.W. Analysis of incision sites produced by human cell extracts and puri®ed (1997) Overexpression in COS cells of p50, the major core protein proteins during nucleotide excision repair of a 1,3-intrastrand d(GpTpG)- associated with mRNA, results in translation inhibition. Nucleic Acids cisplatin adduct. J. Biol. Chem., 271, 7177±7186. Res., 25, 2911±2916. 34. Banin,S., Truong,O., Katz,D.R., Water®eld,M.D., Brickell,P.M. and 16. Ashizuka,M., Fukuda,T., Nakamura,T., Shirasuna,K., Iwai,K., Izumi,H., Gout,I. (1996) Wiskott-Aldrich syndrome protein (WASp) is a binding Kohno,K., Kuwano,M. and Uchiumi,T. (2002) Novel translational partner for c-Src family protein-tyrosine kinases. Curr. Biol., 6, 981±988. control through an iron-responsive element by interaction of 35. Hartmuth,K., Urlaub,H., Vornlocher,H.P., Will,C.L., Gentzel,M., multifunctional protein YB-1 and IRP2. Mol. Cell. Biol., 22, 6375±6383. Wilm,M. and Luhrmann,R. (2002) Protein composition of human 17. Chen,C.Y., Gherzi,R., Andersen,J.S., Gaietta,G., Jurchott,K., prespliceosomes isolated by a tobramycin af®nity-selection method. Royer,H.D., Mann,M. and Karin,M. (2000) Nucleolin and YB-1 are Proc. Natl Acad. Sci. USA, 99, 16719±16724. required for JNK-mediated interleukin-2 mRNA stabilization during 36. Burkle,A. (2001) Physiology and pathophysiology of poly(ADP- T-cell activation. Genes Dev., 14, 1236±1248. ribosyl)ation. Bioessays, 23, 795±806. 18. Bargou,R.C., Jurchott,K., Wagener,C., Bergmann,S., Metzner,S., 37. Marenstein,D.R., Chan,M.K., Altamirano,A., Basu,A.K., Boorstein,R.J., Bommert,K., Mapara,M.Y., Winzer,K.J., Dietel,M., Dorken,B. and Cunningham,R.P. and Teebor,G.W. (2003) Substrate speci®city of Royer,H.D. (1997) Nuclear localization and increased levels of human endonuclease III (hNTH1). Effect of human APE1 on hNTH1 transcription factor YB-1 in primary human breast cancers are associated activity. J. Biol. Chem., 278, 9005±9012. with intrinsic MDR1 gene expression. Nature Med., 3, 447±450. 38. Vidal,A.E., Boiteux,S., Hickson,I.D. and Radicella,J.P. (2001) XRCC1 19. Janz,M., Harbeck,N., Dettmar,P., Berger,U., Schmidt,A., Jurchott,K., coordinates the initial and late stages of DNA abasic site repair through Schmitt,M. and Royer,H.D. (2002) Y-box factor YB-1 predicts drug protein±protein interactions. EMBO J., 20, 6530±6539. resistance and patient outcome in breast cancer independent of clinically 39. Baumann,P., Benson,F.E. and West,S.C. (1996) Human Rad51 protein relevant tumor biologic factors HER2, uPA and PAI-1. Int. J. Cancer, 97, promotes ATP-dependent homologous pairing and strand transfer 278±282. reactions in vitro. Cell, 87, 757±766. 20. Rubinstein,D.B., Stortchevoi,A., Boosalis,M., Ashfaq,R. and 40. Masson,J.Y., Tarsounas,M.C., Stasiak,A.Z., Stasiak,A., Shah,R., Guillaume,T. (2002) Overexpression of DNA-binding protein B gene McIlwraith,M.J., Benson,F.E. and West,S.C. (2001) Identi®cation and product in breast cancer as detected by in vitro-generated combinatorial puri®cation of two distinct complexes containing the ®ve RAD51 human immunoglobulin libraries. Cancer Res., 62, 4985±4991. paralogs. Genes Dev., 15, 3296±3307. 21. Yahata,H., Kobayashi,H., Kamura,T., Amada,S., Hirakawa,T., 41. Opresko,P.L., Cheng,W.H., Von Kobbe,C., Harrigan,J.A. and Bohr,V.A. Kohno,K., Kuwano,M. and Nakano,H. (2002) Increased nuclear (2003) Werner syndrome and the function of the Werner protein; what localization of transcription factor YB-1 in acquired cisplatin-resistant they can teach us about the molecular aging process. Carcinogenesis, 24, ovarian cancer. J. Cancer Res. Clin. Oncol., 128, 621±626. 791±802. 22. Oda,Y., Ohishi,Y., Saito,T., Hinoshita,E., Uchiumi,T., Kinukawa,N., 42. Okamoto,T., Izumi,H., Imamura,T., Takano,H., Ise,T., Uchiumi,T., Iwamoto,Y., Kohno,K., Kuwano,M. and Tsuneyoshi,M. (2003) Nuclear Kuwano M. and Kohno,K. (2000) Direct interaction of p53 with the expression of Y-box-binding protein-1 correlates with P-glycoprotein Y-box binding protein, YB-1: a mechanism for regulation of human gene and topoisomerase II alpha expression and with poor prognosis in expression. Oncogene, 19, 6194±6202. synovial sarcoma. J. Pathol., 199, 251±258. 43. Zhang,Y.F., Homer,C., Edwards,S.J., Hananeia,L., Lasham,A., Royds,J., 23. Koike,K., Uchiumi,T., Ohga,T., Toh,S., Wada,M., Kohno,K. and Sheard,P. and Braithwaite,A.W. (2003) Nuclear localization of Y-box Kuwano,M. (1997) Nuclear translocation of the Y-box binding protein by factor YB1 requires wild-type p53. Oncogene, 22, 2782±2794. ultraviolet irradiation. FEBS Lett., 417, 390±394. 44. Gray,M.D., Shen,J.C., Kamath-Loeb,A.S., Blank,A., Sopher,B.L., 24. Hayakawa,H., Uchiumi,T., Fukuda,T., Ashizuka,M., Kohno,K., Martin,G.M., Oshima,J. and Loeb,L.A. (1997) The Werner syndrome Kuwano,M. and Sekiguchi,M. (2002) Binding capacity of human YB-1 protein is a DNA helicase. Nature Genet., 17, 100±103. protein for RNA containing 8-oxoguanine. Biochemistry, 41, 12739± 45. Fink,D., Aebi,S. and Howell,S.B. (1998) The role of DNA mismatch repair in drug resistance. Clin. Cancer Res., 4, 1±6. 25. Ise,T., Nagatani,G., Imamura,T., Kato,K., Takano,H., Nomoto,M., 46. Marti,T.M., Kunz,C. and Fleck,O. (2002) DNA mismatch repair and Izumi,H., Ohmori,H., Okamoto,T., Ohga,T., Uchiumi,T., Kuwano,M. mutation avoidance pathways. J. Cell. Physiol., 191, 28±41. and Kohno,K. (1999) Transcription factor Y-box binding protein 1 binds 47. Woodard,R.L., Lee,K.J., Huang,J. and Dynan,W.S. (2001) Distinct roles preferentially to cisplatin-modi®ed DNA and interacts with proliferating for Ku protein in transcriptional reinitiation and DNA repair. J. Biol. cell nuclear antigen. Cancer Res., 59, 342±346. Chem., 276, 15423±15433. 26. Jonsson,Z.O. and Hubscher,U. (1997) Proliferating cell nuclear antigen: 48. Hashimoto,M., Donald,C.D., Yannone,S.M., Chen,D.J., Roy,R. and more than a clamp for DNA polymerases. Bioessays, 19, 967±975. 27. Marenstein,D.R., Ocampo,M.T., Chan,M.K., Altamirano,A., Basu,A.K., Kow,Y.W. (2001) A possible role of Ku in mediating sequential repair of Boorstein,R.J., Cunningham,R.P. and Teebor,G.W. (2001) Stimulation of closely opposed lesions. J. Biol. Chem., 276, 12827±12831. human endonuclease III by Y box-binding protein 1 (DNA-binding 49. Blander,G., Zalle,N., Daniely,Y., Taplick,J., Gray,M.D. and Oren,M. protein B). Interaction between a base excision repair enzyme and a (2002) DNA damage-induced translocation of the Werner helicase is transcription factor. J. Biol. Chem., 276, 21242±21249. regulated by acetylation. J. Biol. Chem., 277, 50934±50940. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nucleic Acids Research Oxford University Press

YB‐1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or cisplatin modifications, exhibits endonucleolytic activities and binds several DNA repair proteins

Download PDF
12 pages

Loading next page...
 
/lp/oxford-university-press/yb-1-promotes-strand-separation-in-vitro-of-duplex-dna-containing-QYKx02GeuA

References (48)

Publisher
Oxford University Press
Copyright
Oxford University Press
ISSN
0305-1048
eISSN
1362-4962
DOI
10.1093/nar/gkh170
pmid
14718551
Publisher site
See Article on Publisher Site

Abstract

Published online January 12, 2004 316±327 Nucleic Acids Research, 2004, Vol. 32, No. 1 DOI: 10.1093/nar/gkh170 YB-1 promotes strand separation in vitro of duplex DNA containing either mispaired bases or cisplatin modi®cations, exhibits endonucleolytic activities and binds several DNA repair proteins Isabelle Gaudreault, David Guay and Michel Lebel* Centre de Recherche en Cance  rologie de l'Universite  Laval, Ho à pital Ho à tel-Dieu de Que  bec, Centre Hospitalier   Universitaire de Quebec, 9 McMahon Street, Quebec, G1R 2J6, Canada Received June 23, 2003; Revised September 24, 2003; Accepted November 15, 2003 ABSTRACT protein tyrosine phosphatase 1B (PTP1B) gene (8), gelatinase A (9) and the multidrug resistance (MDR1) gene (10,11). In YB-1 is a multifunctional protein involved in the addition to the regulation of transcription, YB-1 is a multi- regulation of transcription, translation, mRNA spli- functional protein that also affects the splicing and the cing and probably DNA repair. It contains a con- translation of speci®c mRNAs. Indeed, analysis of the human served cold shock domain and it binds strongly to presplicesome complex has indicated the presence of YB-1 in inverted CCAAT box of different promoters. In this this complex (12). Furthermore, YB-1 interacts with the study, we have found that puri®ed YB-1 oligo- splicing factor SRp30c and confers alternative splice site selection in the E1A minigene model system (13). YB-1 is merizes readily in solutions to form trimers, hexam- also known to bind A/C-rich exon enhancers and to stimulate ers and oligomers of 12 molecules. The presence of splicing of the CD44 alternative exon v4 (14). YB-1 is a ATP changed the conformation of YB-1 in such a component of messenger ribonucleoprotein particles which way that only dimers were detected by gel ®ltration regulate translation in COS cells (15). Consistent with this analyses. Puri®ed YB-1 can separate different DNA observation, YB-1 is capable of speci®cally regulating the duplexes containing blunt ends, 5¢ or 3¢ recessed translation of ferritin mRNA (16). Finally, YB-1 is known to ends, or forked structures. This strand separation increase the stability of IL-2 speci®c mRNA during T-cell activity is increased on cisplatin-modi®ed DNA or activation (17). with duplex molecules containing mismatches. In In recent years, several laboratories have demonstrated that addition to its exonuclease activity, YB-1 exhibits YB-1 is directly involved in the cellular response to genotoxic endonucleolytic activities in vitro. Finally, YB-1 stress. Upon UV irradiation or cisplatin treatments, YB-1 af®nity chromatography experiments have indicated binds to a Y-box element within the MDR1 promoter and increases its expression (11). Moreover, depletion of YB-1 that in addition to prespliceosome factors like expression protein with anti-sense RNA against YB-1 speci®c nucleolin and ALY, YB-1 binds the DNA repair mRNA results in increased sensitivity to cisplatin (11). proteins MSH2, DNA polymerase d, Ku80 and WRN Interestingly, YB-1 is increased in cultured cell lines resistant proteins in vitro. Furthermore, immuno¯uorescence to cisplatin. In fact, several studies have indicated that the studies have shown that YB-1 re-localizes from the level of nuclear expression of YB-1 is predictive of drug cytoplasm to nuclear areas containing either Ku80 resistance and patient outcome in breast tumors, ovarian or MSH2 proteins in human 293 embryonic kidney cancers and synovial sarcomas (18±22). Upon UV irradiation, cells. These results suggest that YB-1 is involved in YB-1 translocates from the cytoplasm to the nucleus (23) and base excision and mismatch repair pathways. is known to bind to modi®ed nucleic acid (24). YB-1 preferentially binds to cisplatin-modi®ed DNA and interacts with PCNA (25), a component of several DNA repair systems INTRODUCTION (26). In addition, YB-1 stimulates an endonuclease involved in YB-1 is a member of a family of DNA-binding proteins base excision repair (27). All these observations suggest that containing a cold shock domain. It was originally identi®ed by YB-1 is important in DNA repair and in conferring drug its ability to bind to the inverted CCAAT box (Y-box) in the resistance on tumor cells. promoter region of MHC class II genes (1). Subsequently, YB- It has been reported that YB-1 creates single-stranded 1 has been shown to regulate the expression of a number of regions in the DRA promoter (28) and it is believed that this genes including proliferating cell nuclear antigen (PCNA) (2), activity is required in part for the regulation of target thymidine kinase (3), DNA polymerase a (4), myosin light- promoters (29). In recent years, YB-1 has been shown to chain 2v gene (5), Fas antigen (6), collagen a1(I) gene (7), bind preferentially to single-stranded nucleic acids and to *To whom correspondence should be addressed. Tel: +1 418 691 5281; Fax: +1 418 691 5439; Email: [email protected] Nucleic Acids Research, Vol. 32 No. 1 ã Oxford University Press 2004; all rights reserved Nucleic Acids Research, 2004, Vol. 32, No. 1 317 exhibit 3¢-5¢ exonuclease activity (30). In this report, we YB-1 puri®cation and gel ®ltration investigated the strand separation activity of human YB-1 BL21 cells expressing GST±YB-1 fusion proteins were lysed against different double-stranded DNA substrates in vitro. in NETN buffer (0.5% NP-40, 20 mM Tris±HCl pH 8.0, Different deletion mutants of YB-1 have indicated that amino 100 mM NaCl and 1 mM EDTA) and incubated overnight acids 39±205 are required for the DNA strand separation with glutathione±Sepharose beads. The next day, beads were activity. We have also found that YB-1 actively promotes washed with NETN buffer and treated with biotinylated strand separation of duplex DNA containing either mis- thrombin (Novagen) for 2 h at room temperature in thrombin matches or cisplatin modi®cations independently of the cleavage buffer (20 mM Tris±HCl pH 8.4, 150 mM NaCl, nucleotide sequence. It also exhibits an endonuclease activity 2.5 mM CaCl ). Beads were spun down and the supernatant on double-stranded DNA. Finally, in vitro YB-1 af®nity was kept for the next step. Thrombin was captured by chromatography and immuno¯uorescence analyses have incubation with streptavidine agarose (Novagen) for 2 h on a shown that several DNA repair proteins can interact with rocking platform at room temperature. Agarose beads were YB-1 reinforcing the notion that this multifunctional protein is spun down and YB-1 protein from the supernatant was involved in the repair of speci®c DNA damage. concentrated onto Centricon-30 ®lters (Amicon). Protein concentration was determined using the Bradford assay. Proteins were then loaded onto a Superdex-200 column for gel ®ltration analysis using an AKTA-FPLC as indicated by MATERIALS AND METHODS the manufacturer (Amersham Pharmacia). Proteins from each fraction of the column were visualized by Coomassie staining. Cell lines and antibodies Human 293 embryonic kidney cells were maintained in Strand separation and endonuclease assays DMEM supplemented with 10% fetal bovine serum. The DNA substrates used for the strand separation and Polyclonal antibodies against the human WRN were pur- nuclease assays are depicted in Table 1. One strand was chased from Novus Biologicals (Littleton, CO). Antibodies labeled with T4 polynucleotide kinase and [g- P]ATP and against PARP-1 and DNA polymerase d were purchased from annealed to its complementary strand as described previously Transduction Laboratories (Lexington, KY). Antibodies (32). DNA with cisplatin adducts were created by incubating against ALY, REF1 and XRCC1 were purchased from Santa oligonucleotides with cisplatin overnight as described previ- Cruz Biotechnology (Santa Cruz, CA). Antibodies against ously (33). DNA substrate was incubated with the recombi- Ku80 were purchased from NeoMarkers (Fremont, CA). nant proteins as indicated in the ®gure legends for 30 min at Antibodies against DNA-PK, MLH1, MSH2 and PMS2 were 37°C in the reaction buffer (40 mM Tris±HCl, pH 7.5, 4 mM purchased from Oncogene Research Products (Boston, MA). MgCl , 5 mM DTT, 2 mM ATP and 0.1 mg/ml bovine serum Antibodies against nucleolin were purchased from Medical albumin). Four microliters of loading buffer was added to the and Biological Laboratories Co. (Watertown, MA). Rabbit reaction (40% glycerol, 50 mM sodium EDTA, 2% SDS and polyclonal antibody against human YB-1 and the correspond- 1% bromophenol blue, pH 8.0) and the DNA was analysed on ing pre-immune serum was kindly provided by Dr P. E. native 12% polyacrylamide gel (in TBE buffer). For the DiCorleto (The Cleveland Clinic Foundation, Cleveland, OH). endonuclease and exonuclease assays, radioactive DNA Finally, all horseradish peroxidase-conjugated secondary substrates were incubated with the indicated puri®ed proteins antibodies were purchased from Amersham Pharmacia. The in the same reaction buffer as indicated for the strand above antibodies were used as indicated by the manufacturers. separation reaction. Cleaved DNA products were separated Western blots were performed as described previously (31). on a denaturing gel (14% polyacrylamide, 8 M urea in TBE) and analyzed by autoradiography. Plasmids YB-1 af®nity puri®cation or `pull down' assay Several GST-fusion proteins were constructed for the `pull down' or YB-1 af®nity puri®cation assay. Human YB-1 Approximately 10 human 293 cells were lysed in 10 mM coding sequence was ampli®ed by PCR with appropriate Tris±HCl (pH 7.5), 1% (v/v) Triton X-100, 150 mM NaCl, 5 oligonucleotides for subsequent cloning into the BamHI/ mM EDTA, 50 mM NaF in the presence of a protease inhibitor EcoRI sites of the pGEX-2TK vector. In addition, YB-1 cocktail (Boehringer Mannheim) at 4°C for 20 min. Cells were cDNA was cut with SmaI and EcoRI (amino acid residues sonicated three times for 5 s and cell debris were spun down. 39±312 of YB-1), SmaI and SalI (residues 39±205), SalI and Af®nity matrices were prepared by immobilizing GST alone or GST±YB-1 fusion proteins on glutathione±Sepharose beads EcoRI (residues 205±312) and these fragments were cloned (Amersham Pharmacia) as described (34). Freshly prepared into the appropriate modi®ed restriction sites in the pGEX- cell lysates were incubated with the af®nity matrices overnight 2TK vector. A pGEX-2TK construct coding for a GST-fusion exo at 4°C. After extensive washing with lysis buffer, bound peptide containing the exonuclease domain of p53 (p53 ) proteins were released by boiling in SDS sample buffer and was kindly provided by the laboratory of Jacques Co à te  (Centre analyzed by SDS±PAGE and western blotting. de Recherche en Cance Ârologie, Que Âbec City, Canada). ProScan analyses on p53 have indicated that its exonuclease Indirect immuno¯uorescence domain is within amino acids 185±290. Plasmids were transfected into BL21 bacteria for fusion protein production. Human 293 embryonic kidney cells were maintained in Proteins were visualized by Coomassie staining when DMEM supplemented with 10% fetal bovine serum. indicated. They were plated on glass coverslips the day before the 318 Nucleic Acids Research, 2004, Vol. 32, No. 1 Table 1. DNA Structures Figure 1. Characterization of puri®ed human YB-1 on gels stained by immuno¯uorescence experiments. Cells were ®xed in Coomassie. (Top, left) This gel contains GST±YB-1 and GST proteins cold methanol for 10 min and permeabilized with 0.15% puri®ed on glutathione±Sepharose beads. (Top, right) GST±YB-1 was Triton X-100 at 4°C for 10 min. After washing with PBS, cells treated with thrombin to remove the GST portion of the fusion protein as described in Materials and Methods. The size of YB-1 is ~42±44 kDa. were blocked with 2% milk at 4°C for 30 min. After blocking, (Bottom) These gels contain YB-1 proteins eluted from a Superdex-200 gel antibodies (1:250 for anti-Ku80, 1:50 for anti-MSH2, 1:2000 ®ltration column with or without 2 mM ATP in the buffer. The numbers for anti-WRN or 1:500 for anti-YB-1) diluted in blocking above each gel represent the position of marker proteins eluted from such a buffer were applied to the coverslips and incubated overnight column. The numbers below the gel represent the fraction number. at 4°C. Coverslips were washed with PBS and incubated for 1 h at room temperature with rhodamine-secondary antibody for WRN or YB-1 detection (Santa Cruz), and FITC-secondary of ~42±44 kDa (Fig. 1, top right). Puri®ed YB-1 was loaded antibody (Santa Cruz) for Ku80 or MSH2 detection. After onto a Superdex-200 column and the eluted proteins were washing, coverslips were mounted on glass slides. Slides were analyzed on a 10% SDS±polyacrylamide gel. Coomassie viewed at 603 magni®cation (1.4NA oil-immersion 603 staining revealed that YB-1 eluted in fractions 12±18. In a objective) and zoomed 23 for image acquisition on a Nikon parallel experiment, molecular weight protein markers were inverted diaphot confocal microscope equipped with Krypton/ also eluted from a Superdex-200 column and analyzed on a Argon lasers (488 and 568 nm). Images were captured with a gel. As indicated in Figure 1, the YB-1 peptide was recovered BioRad MRC1024 confocal microscopy system. Finally, in multiple peaks that co-eluted with molecular weight images were colored and merged in Adobe photoshop. markers of 530, 232, 140 and 67 kDa (bottom). However, YB-1 co-eluted with the molecular weight marker of 67 kDa upon addition of 2 mM ATP in the solution. A dimer of YB-1 would be ~80 kDa which elutes with the 67 kDa molecular RESULTS weight marker. These results indicate that in addition to Puri®ed human YB-1 forms oligomers in solution monomers and dimers, YB-1 can form trimers, hexamers and oligomers of 12 YB-1 molecules in solution. In the presence of Recent studies have indicated that YB-1 may act as multimers ATP, only dimers are formed. and that two C-tail domains of YB-1 are required for homodimerization (28,30). To determine whether YB-1 can Human YB-1 separates double-stranded DNA in the oligomerize in solution, the GST±YB-1 fusion protein was presence or absence of ATP produced in bacteria and puri®ed with glutathione±Sepharose beads (Fig. 1, top left). As the GST portion of this fusion YB-1 is known to create single strand regions in the DRA protein could modify the behavior of YB-1 in vitro, the GST promoter (28). We thus tested whether puri®ed YB-1 can portion was cleaved off with thrombin (see Materials and separate the strands of different DNA duplexes (Table 1). A Methods). This puri®ed YB-1 protein has a molecular weight forked DNA structure was ®rst tested in the presence or Nucleic Acids Research, 2004, Vol. 32, No. 1 319 absence of ATP. ATP was also tested in these reactions because PROSCAN analysis of YB-1 protein has indicated that residues 28±58 form a motif with 69% homology to an ATP-binding signature region. As shown in Figure 2A, puri®ed YB-1 is capable of separating the strands like a DNA helicase enzyme. The amount of displaced radioactive strand was YB-1 concentration dependent (Fig. 2A). YB-1 proteins from fractions 12, 15, 16 and 18 of the Superdex-200 column (Fig. 1) were tested. All these fractions showed similar strand displacement activities (data not shown). YB-1 was capable of separating the strands of a forked DNA structure without ATP (Fig. 2B). However, this activity was doubled in the presence of 2 mM ATP in the reaction (Fig. 2B and C). The percentage of displacement was calculated from several reactions by excising the corresponding bands from the gels and counting the amount of radioactivity with a b-counter. Similar results were obtained with different DNA structures (data not shown). These observations indicate that puri®ed YB-1 can separate the strands of a forked DNA structure even in the absence of ATP. Finally, ATPase assays were performed and no ATP hydrolysis was detected with our puri®ed YB-1 in the presence or absence of DNA molecules (data not shown). Human YB-1 can separate the strands of either blunted, 5¢ or 3¢ recessed DNA structures The capacity of YB-1 to separate DNA strands was examined with different duplex structures (see Table 1). As shown in Figure 3A±E, YB-1 was equally capable of separating DNA strands from a 22mer duplex containing either blunt ends, a 5¢ recessed or a 3¢ recessed end. In addition, YB-1 displaced GC-rich or AT-rich radioactive strands equally well (see Table 1 for oligonucleotide sequences). Overall, an 8±16% strand displacement was observed with different duplex sequences and this independently of the structures at the ends of the DNA substrates (Fig. 3C). YB-1 is a transcription factor that binds strongly to an inverted CCAAT box (1). A 22mer duplex containing this sequence (Y-box sequence) was also examined. As shown in Figure 3F, YB-1 separated the two strands of the Y-box duplex very ef®ciently. Close to 50% of total duplex was separated by YB-1 (Fig. 3G). All these results indicate that YB-1 can separate a 22mer duplex DNA independently of the nucleotide Figure 2. Strand separation activity of human YB-1 on a forked DNA sequence. However, YB-1 has a very strong preference for a duplex structure. (A) Increasing amounts of puri®ed YB-1 proteins duplex DNA containing an inverted CCAAT box. (indicated in ng) or 40 ng of GST were incubated with a radioactive 22 bp forked duplex under standard conditions for helicase activity (see Materials Human YB-1 does not separate a 36mer DNA duplex and Methods) for 30 min at 37°C. Reactions were stopped in the appropriate with no inverted CCAAT box sequence dye buffer and the DNA products were run on a 12% native polyacrylamide gel. The double-stranded and single-stranded structures are depicted on the The DNA substrates used in the strand separation reactions in left. The asterisk represents the labeled strand at its 5¢ end. The triangle the preceding sections contained <23 bp. We therefore asked represents heat denatured DNA. (B) YB-1 (40 ng) was incubated with a whether the length of the DNA duplex affects YB-1 strand radioactive 22 bp forked duplex as in (A), in the presence or absence of separation activity. As indicated in Figure 4A (top), YB-1 was 2 mM ATP. (C) Histogram representation of the YB-1 strand separation reactions performed in (B). All experiments were done in duplicates with unable to separate the strands of a 36 bp DNA duplex. 40 ng of puri®ed human YB-1. The percentage strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of Human YB-1 separates 36mer DNA duplex treated with displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was cisplatin or containing mismatches calculated for each band of the gel as indicated in Materials and Methods. It has been shown that YB-1 binds preferentially to cisplatin- modi®ed DNA (25). Thus, the 36mer duplex structure was strands of the 36mer DNA substrate containing cisplatin incubated overnight with cisplatin and the resulting modi®ed DNA was used in YB-1-mediated strand separation reaction. modi®cations. Approximately 14% of cisplatin-modi®ed As shown in Figure 4A, YB-1 was capable of separating the strands were separated by YB-1 (Fig. 4D). 320 Nucleic Acids Research, 2004, Vol. 32, No. 1 YB-1-mediated strand separation reaction was also exam- ined on duplex DNA containing one or two mispaired bases. Figure 4B indicates that YB-1 can separate the strands of a 36mer duplex DNA containing one mismatch. Approximately 9% of the total DNA duplex was separated (Fig. 4D). Finally, a 22mer DNA duplex structure (GC rich) with two mismatches was tested. Up to 25% of this DNA substrate was separated into single-stranded DNA (Fig. 4C and D). All these results indicate that YB-1 can separate duplex DNA with either mispaired bases or cisplatin-modi®ed lesions. The DNA strand separation activity of human YB-1 requires amino acids 40±205 To demonstrate that the strand separation activity was intrinsic to YB-1, a series of GST±YB-1 fusion proteins were analyzed (Fig. 5A). Although we did see some protein degradation, all fusion proteins were treated with thrombin and the puri®ed YB-1 fragments were assayed on the Y-box duplex. These experiments were repeated three times. As seen in Figure 5B, full-length YB-1 separated the two strands as shown before. YB-1 proteins lacking the ®rst 40 amino residues (including part of the potential ATP-binding site) showed weak activity on the gel. Approximately 10% of the DNA duplex was separated by this mutant compared with 27% for the full- length YB-1 (Fig. 5C). The strand separation activity of this mutant YB-1 (amino acids 40±313) was similar in the presence or absence of ATP (Fig. 5B, last panel on the right). In contrast, the C-terminus part of YB-1 (amino acids 204±313) did not exhibit strand separation activity. Finally, a YB-1 fragment corresponding to residues 39±205 (Fig. 5B, middle) exhibited a very weak activity (<5% strand displace- ment in Fig. 5C). As a negative control, a GST-fusion peptide exo ) was containing the exonuclease domain of p53 (p53 assayed under the same conditions and no strand separation activity was detected (Fig. 5B and C). These results indicate that the strand separation activity is localized to the region of YB-1 between amino acids 40 and 205. However, both the N- terminus and C-terminus portions of YB-1 are obviously required for optimal activity. Similar results were obtained with a forked DNA substrate (data not shown). Human YB-1 exhibits endonucleolytic activities on double-stranded DNA Exposition of the native gels in Figure 5B for an additional 3 days showed some degradation of the DNA substrates (data Figure 3. Strand separation activity of YB-1 on different duplex structures. not shown). This suggests that nucleolytic activity was also (A±F) The indicated amounts (in ng) of puri®ed human YB-1 proteins were incubated with the indicated radioactive DNA substrates under standard present in the strand separation reactions. Consistent with this conditions for helicase activity (see Materials and Methods) for 30 min at observation, a recent report has indicated that GST±YB-1 37°C. Again, 40 ng of GST proteins were used as negative controls when chimeric proteins exhibit 3¢ to 5¢ exonuclease activity (30). indicated. Reactions were stopped in the appropriate dye buffer and the However, this activity was described only for single-stranded DNA products were run on a 12% native polyacrylamide gel. The double- stranded and single-stranded DNA structures with their sequence character- DNA molecules. Similar analyses were thus performed with istics are depicted on the left of each autoradiogram. The 5¢-labeled strand our puri®ed YB-1 proteins on different duplex substrates. To of the duplex is represented by an asterisk. (A) and (B) represent reactions determine if our puri®ed YB-1 protein exhibited any with blunt-ended DNA. (C) and (D) represent reactions with a structure hav- exonuclease activity, a single-stranded DNA substrate was ing a 3¢ recessed end. (E) represents reactions with a structure having a 5¢ ®rst analyzed. As shown in Figure 6A (left), YB-1 degraded recessed end. In (F), the Y-box sequence is an inverted CCAAT sequence. All DNA substrates contain 22 paired bases. (G) Histogram representation the single-stranded DNA. However, the autoradiogram had to of the YB-1 strand separation reactions performed in (A±F). The strand be exposed for 3 days to detect the exonuclease activity. displacement data were obtained from experiments done in duplicate with A double-stranded DNA substrate was incubated with YB-1 40 ng of the indicated puri®ed proteins. The percentage strand displacement to examine the exonuclease activity. As shown in the second is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was panel of Figure 6A, the pattern of degradation was different calculated for each band of the gel as indicated in Materials and Methods. from the single-stranded substrate. To demonstrate that the Nucleic Acids Research, 2004, Vol. 32, No. 1 321 Figure 4. Strand separation activity of YB-1 on duplex structures containing mismatches or cisplatin cross-links. (A±C) The indicated amounts (in ng) of puri®ed human YB-1 proteins were incubated with the indicated radioactive DNA substrates under standard conditions for helicase activity (see Materials and Methods) for 30 min at 37°C. Forty nanograms of GST were used as a negative control. The panels in (A) represent strand separation reactions performed with a 36 bp duplex. The panel on the right contains a reaction performed with a 36mer duplex treated with cisplatin. In (B), reactions were performed with a 36mer duplex containing one base mismatch. In (C), reactions were performed with a 22mer duplex structure containing two nucleotide mismatches. (D) Histogram representation of the YB-1 strand separation reactions performed in (A). The strand displacement data were obtained from experiments done in duplicate with 40 ng of the indicated puri®ed proteins. The percentage strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was calculated for each band of the gel as indicated in Materials and Methods. exonuclease activity was intrinsic to YB-1, a series of puri®ed was stronger with a YB-1 molecule lacking the ®rst 40 amino YB-1 fragments were analyzed. Full-length YB-1 exhibited acid residues. These results indicate that the ®rst 40 amino nucleolytic activity on duplex DNA (Y-box duplex) (Fig. 6A, acid residues of YB-1 will affect its nicking activity. In right). The absence or presence of ATP did not affect the contrast, a YB-1 fragment comprising residues 39±205 did not nucleolytic activities of full-length YB-1 (data not shown). A exhibit nuclease activity on this DNA substrate. As expected, YB-1 protein lacking the ®rst 40 amino acid residues also the exonuclease domain of p53 exhibited a strong nuclease exhibited similar nucleolytic activity but at a lower ef®ciency activity with a cleavage pattern similar to the N-terminus (Fig. 6B). The C-terminus portion of YB-1 (residues 204±313) truncated YB-1 peptide (Fig. 7A). Similar experiments were did not show nuclease activity. A peptide fragment comprising performed on the same DNA substrate except that the opposite residues 39±205 of YB-1 cleaved the 5¢ radioactive nucleotide strand was labeled at the 5¢ end (Fig. 7B). Although labeling of exo of the duplex. Finally, the exonuclease domain of p53 (p53 ) the short strand did show some heterogeneity of the probe, showed very little nuclease activity on this Y-box DNA only the full-length YB-1 exhibited a weak nucleolytic duplex. These results indicate that YB-1 introduces nicks or activity. Less than 2% of the probe was cleaved with 40 ng breaks into double-stranded DNA molecules. of YB-1. The cleavage pattern indicated nicking near the Additional experiments were performed with YB-1 frag- labeled 5¢ end of the strand. None of the mutated YB-1 peptide exo ments (fragments exhibiting nicking activities in the previous showed nicking on the shorter stand of the duplex. The p53 section) on DNA substrates labeled on either strand to peptide cleaved the labeled 5¢ of this substrate (Fig. 7B, last compare the cleavage patterns. As shown in Figure 7, full- lane on the right panel). These results suggest that YB-1 can length YB-1 also cleaved a DNA duplex containing a 3¢ introduce double-stranded breaks and exhibit endonucleolytic overhang. However, the cleavage pattern was different from activities. However, this activity was weak as the autoradio- the patterns obtained with duplex DNA possessing blunt ends grams had to be exposed for 3 days to detect the cleavage (compare Figs 6A with 7A). Interestingly, the cleavage pattern products. 322 Nucleic Acids Research, 2004, Vol. 32, No. 1 Figure 5. Strand separation activity of different mutant YB-1 on a DNA duplex structure. (A) Characterization of the different mutant GST±YB-1 fusion proteins (before thrombin treatments) on gels stained by Coomassie. The YB-1 amino acids present in the fusion proteins are indicated above each lane. (B) The different GST-fusion proteins were cleaved with thrombin as described in the Materials and Methods and the cleaved products were used in strand exo displacement reactions. Approximately 40 ng of the puri®ed YB-1, mutant YB-1 proteins or p53 peptide (exonuclease domain) were incubated with a radioactive 22 bp duplex containing a Y-box sequence under standard conditions for helicase activity (see Materials and Methods) for 30 min at 37°C in the absence or presence of 2 mM ATP. Reactions were stopped in the appropriate dye buffer and the DNA products were analyzed on a 12% native polyacrylamide gel. The double-stranded and single-stranded structures are depicted on the left. The asterisk represents the labeled strand at its 5¢ end. The triangle represents heat denatured DNA. (C) Histogram representation of the strand separation reactions performed in (B). All experiments were done in exo duplicate with 40 ng of puri®ed YB-1, mutant YB-1 proteins or p53 (exonuclease domain) in reaction buffer containing 2 mM ATP. The percentage of strand displacement is calculated by using the formula: c.p.m. of displaced strand/(c.p.m. of displaced strand + c.p.m. of double-stranded DNA) 3 100. The c.p.m. was calculated for each band of the gel as indicated in Materials and Methods. Several DNA repair proteins bind to human YB-1 that the GST±YB-1 fusion construct can be used for af®nity immobilized on a matrix chromatography experiments. Poly(ADP-ribose) polymerase-1 (PARP-1), REF1 endonu- YB-1 binds to cisplatin-modi®ed DNA (25) and can separate clease (also known as APE1), and XRCC1 are proteins the strands of such modi®ed DNA (preceding sections). It also involved in short or long patch excision repair pathways (36± interacts with PCNA (25) and an endonuclease involved in 38). Western blot analyses revealed that none of these proteins base excision repair (27). These observations suggest that YB- bound to the GST±YB-1 matrix (Fig. 8 and data not shown). 1 may be involved in certain aspects of DNA repair. To As YB-1 can separate the strands of a duplex DNA identify other DNA repair proteins that may interact with YB- containing mismatches, antibodies against proteins known to 1, whole cell extracts from the human 293 embryonic kidney be involved in the mismatch repair pathway were also tested. cell line were loaded on af®nity column containing the GST± The presence of MLH1, MSH2 and PMS2 was examined. As YB-1 fusion protein immobilized on glutathione±Sepharose indicated in Figure 8, only MSH2 bound to the GST±YB-1 beads. GST immobilized on glutathione±Sepharose beads was beads but not to the GST control beads. used as a control. Bound proteins were analyzed by western Proteins involved in homologous recombination and non- blots with antibodies speci®c to different DNA repair proteins. homologous end-joining were also examined. These include Western blots were ®rst examined with antibodies against RAD51, RAD51C and RAD52 for the homologous recombi- proteins known to interact with YB-1 or known to be in the nation pathway and DNA-PK, Ku80 and WRN for the non- same cellular complex. Nucleolin and ALY were analyzed as homologous end-joining pathway (39±41). RAD51, RAD51C, both are found in prespliceosomes with YB-1 (35). As shown RAD52 and DNA-PK did not bind the GST±YB-1 beads (data in Figure 8, both nucleolin and ALY bound to the GST±YB-1 not shown). In contrast, Ku80 and WRN proteins bound beads but not to the GST control beads. These results indicate strongly to the GST±YB-1 matrix but not to the GST control. Nucleic Acids Research, 2004, Vol. 32, No. 1 323 Finally, an antibody against DNA polymerase d was also tested on the western blots. As indicated in Figure 8, DNA polymerase d bound to the GST±YB-1 matrix with low af®nity as suggested by the weak signal observed on the blot. Altogether, these results indicate that several proteins involved in DNA repair bind YB-1 in vitro. These ®ndings strongly support the involvement of YB-1 in speci®c DNA repair pathways. Re-localization of YB-1 to nuclear areas containing DNA repair proteins in cisplatin-treated cells Co-localization experiments with antibodies against Ku80 and YB-1 were performed on intact 293 cells that had been treated with or without 7.5 mg/ml of cisplatin for 2 h. FACS analyses revealed no evidence of an increase in apoptotic ®gures in cisplatin-treated cultures compared with untreated cells at the 2 h time point (data not shown). Thus, we were analyzing cells before the onset of apoptosis. As shown in Figure 9A, YB-1 is found mainly in the cytoplasm of untreated 293 cells with some nuclear localization. In contrast, Ku80 is only localized to the nucleus of untreated cells in a diffuse pattern (Fig. 9B). Immuno¯uorescence images for both antibodies were captured in separate channels and overlaid to see the extent of co-localization. As shown in Figure 9C, few nuclear foci contained both proteins. In the presence of cisplatin, some YB-1 proteins translocated to the nucleus (Fig. 9D). The Ku80 distribution pattern was similar in both cisplatin-treated and untreated cells (compare Fig. 9B with E). Interestingly, >90% of YB-1 proteins found in the nucleus of cisplatin-treated cells co-localized with Ku80 (yellow ¯uorescence in Fig. 9F). These results indicate that upon cisplatin treatment, YB-1 re-localizes into multiple nuclear foci containing Ku80 proteins. Pre-immune serum from the YB-1 immunized rabbit did not show such results (data not shown). Western analysis on 293 whole cell lysates indicated that the rabbit immune serum used in the immuno¯uorescence study recognizes only one band speci®c to YB-1 (Fig. 9G). However, the anti-YB-1 antibody immunoprecipitated YB-1 from 293 cell lysate only poorly (data not shown). Hence, co-immunoprecipitation could not be performed with this antibody. Co-localization experiments were also performed with antibodies against MSH2 and YB-1 (Fig. 10). Again, YB-1 is found in the cytoplasm of untreated 293 cells. There are few foci in the nucleus containing YB-1 (Fig. 10A). Upon cisplatin treatment, YB-1 re-localizes into the nucleus of 293 cells (Fig. 10D). As shown in Figure 10B and E, MSH2 is found in the nucleus of untreated and cisplatin-treated 293 cells in a weak diffuse pattern. Merging the images shows that there are few nuclear foci containing both YB-1 and MSH2 proteins (yellow ¯uorescence in Fig. 10C) in untreated 293 cells. However, the number of nuclear foci containing both MSH2 Figure 6. Exonuclease and DNA nicking activity of YB-1. (A) Forty and YB-1 proteins increased substantially in cisplatin-treated exo nanograms of full-length, mutant YB-1 or p53 (exonuclease domain of cells (Fig. 10F). Approximately 30% of MSH2 nuclear foci p53) were incubated for 30 min at 37°C as described in Materials and contained YB-1 proteins upon cisplatin treatment. These Methods with the different DNA substrates indicated above each gel. results indicate that upon cisplatin treatment, a proportion of Reactions were stopped in the appropriate dye buffer and cleaved DNA cellular YB-1 re-localizes into multiple nuclear areas con- products were analyzed on 14% denaturing polyacrylamide gels to analyze the exonuclease activity. (B) Histogram representation of the YB-1 nuclease taining MSH2 proteins. activity performed in (A) with the duplex DNA substrate. The percentage Both anti-WRN and YB-1 antibodies were generated in nuclease activity was calculated by scanning the ®gures and using the rabbits. Consequently, to detect co-localization of YB-1 with formula: intensities of cleaved products/(intensities of cleaved products + WRN proteins in treated 293 cells, a speci®c epitope-tagged intensity of uncleaved probe) 3 100. Results were calculated from duplicate experiments. version of the proteins will be constructed. This is part of 324 Nucleic Acids Research, 2004, Vol. 32, No. 1 exo Figure 7. Endonucleolytic activity of YB-1. (A and B) Forty nanograms of full-length, mutant YB-1 or p53 (exonuclease domain of p53) peptides were incubated with the indicated DNA substrates in reaction buffer for 30 min at 37°C as described in Materials and Methods. DNA products were analyzed on 14% denaturing polyacrylamide gels. The brackets on the right side of each gel indicate the cleavage products of each probe by the proteins. Experiments were repeated twice. (C) Histogram representation of the YB-1 nuclease activity performed in (B). The percentage nuclease activity was calculated by scanning the ®gures and using the formula: intensities of cleaved products/(intensities of cleaved products + intensity of uncleaved probe) 3 100. Results were calculated from duplicate experiments. another study. Finally, the antibody against DNA polymerase were performed with our puri®ed YB-1 and it did not d used in this study does not work in immuno¯uorescence. hydrolyze ATP even in the presence of DNA. Our puri®ed YB-1 can separate the strands of different DNA duplex molecules under buffer conditions used for DISCUSSION helicase assays (44). The strand separation activity of YB-1 is increased 2-fold in the presence of ATP (Fig. 2). Importantly, Recent studies have implicated YB-1 in genome stability as it a mutant YB-1 lacking part of the potential ATP-binding interacts with proteins involved in DNA repair such as PCNA domain did not exhibit increased strand separation activity in (25), human endonuclease III (27) and p53 (42,43). YB-1 also the presence of ATP (Fig. 5B). Thus, our data indicate that exhibits exonuclease activities on single-stranded DNA (30). binding of ATP changes the conformation of YB-1 and affects In this report, we investigated the strand separation and its strand separation activities. YB-1 does not seem to have a nucleolytic properties of YB-1 on different DNA duplex preferred directionality like an helicase. It can separate the substrates in vitro. For such analysis, we produced GST±YB-1 strands from a duplex molecule with either a 3¢ or 5¢ recessed in bacteria and designed a puri®cation scheme involving the end, blunt-ended DNA, or a forked DNA structure equally removal of the GST portion of the fusion construct with well. The strongest strand separation activity was observed thrombin and a gel ®ltration step. Elution from the Superdex- with DNA duplexes containing an inverted CCAAT box. 200 gel ®ltration column has indicated that YB-1 can readily Additional analyses with deletion mutants have indicated that form trimers, hexamers and oligomers of 12 molecules in the DNA strand separation activity is intrinsic to YB-1 protein solution. However, the presence of ATP changed the con- and the active site is situated between amino acid residues 40 formation of the protein as YB-1 mainly formed dimers, a and 205. Importantly, this region of YB-1 is known to contain complex more likely to be present in vivo. This con®rms the DNA-binding domain (30). The length of the duplex earlier reports suggesting that YB-1 is acting as an oligomer in vivo (28,30). PROSCAN analyses of YB-1 amino acid structure affects YB-1 strand separation activities. YB-1 can sequence have indicated that it possesses an ATP-binding separate a 22 bp molecule but not a 36 bp molecule. However, signature at its N-terminus (residues 28±58). ATPase assays it will separate a 36mer containing one or two base Nucleic Acids Research, 2004, Vol. 32, No. 1 325 Figure 8. Immunoblots against DNA repair proteins bound to GST±YB-1 af®nity Sepahrose beads. Human 293 embryonic kidney whole cell extracts (WCE) were incubated with either 50 mg of GST±YB-1 or GST linked glutathione±Sepharose beads overnight. Proteins bound to the af®nity beads were analyzed by SDS±PAGE with antibodies against the indicated proteins on the left of each blot. mismatches or cisplatin modi®cations. This strongly suggests that YB-1 will bind and process DNA molecules with such lesions. YB-1 is not only capable of separating the strands, but it can also cleave double-stranded DNA molecules. However, this YB-1 nicking activity is weaker than the strand separation activities. Unlike the strand separation activity, it takes several days of exposition to see the nucleolytic cleavage pattern on an autoradiogram. This suggests that the DNA strand separ- ation activity of YB-1 is stronger than its nucleolytic activity under the in vitro conditions employed in this study. Our Figure 9. Co-localization of YB-1 and Ku80 by immuno¯uorescence. preliminary observations indicate that YB-1 cleavage activ- Human 293 embryonic kidney cells were untreated (A±C) or incubated with ities seem to be sequence and/or structure dependent. 7.5 mg/ml of cisplatin for 2 h (D±F), then ®xed and subsequently incubated simultaneously with rabbit anti-YB-1 and mouse anti-Ku80 as described in Noticeably, the exonucleolytic activity on a DNA duplex Materials and Methods. Anti-rabbit rhodamine-labeled and anti-mouse substrate containing a 3¢ overhang increased dramatically FITC-conjugated secondary antibodies were used to visualize YB-1 and when YB-1 lacked the ®rst 40 N-terminus amino acid Ku80 by confocal microscopy at 568 and 488 nm, respectively. Images de- residues. This increased nucleolytic activity was not observed pict representative cells from each of the non-treated and cisplatin-treated cultures. In the merged images (C and F) a yellow color appears where on blunt duplex. The reason for the enhanced nucleolytic YB-1 (red) and Ku80 (green) ¯uorescence signals coincide. In (G), western activity with the mutant YB-1 (40±313) peptide is unknown. analysis of 293 total cell lysate with anti-YB-1 and pre-immune sera. Presumably the deletion of the ®rst 40 amino acids of YB-1 has affected its interaction with different DNA structures in vitro. The only repair proteins that bound the YB-1 matrix in our Our data on the enzymatic activities of YB-1 toward duplex study were DNA polymerase d, MSH2, Ku80 and WRN. In with mismatches and cisplatin modi®cations suggest that YB- addition, immuno¯uorescence studies have indicated re- 1 might be involved in base excision repair and mismatch localization of YB-1 to regions of the nucleus containing repair pathways. It was thus logical to ask whether YB-1 can either Ku80 or MSH2 proteins in cisplatin-treated cells (Figs 9 bind to proteins involved in such DNA repair pathways. and 10). Accordingly, af®nity chromatography experiments were The MSH2/YB-1 interaction is an interesting observation as performed on a YB-1 matrix with 293 whole cell lysates. cancer cell lines with a mutation in the MSH2 gene are also 326 Nucleic Acids Research, 2004, Vol. 32, No. 1 It is believed that WRN protein is a dual exonuclease/ helicase enzyme not only involved in non-homologous end- joining, but also in long patch base excision repair (42). Mutations in the WRN gene are responsible for the human progeroid disorder Werner syndrome. Importantly, WRN protein translocates to speci®c nuclear foci upon UV treatment (49). These foci are believed to be sites of DNA repair. The co- localization of WRN and YB-1 and their direct interaction at sites of DNA repair upon UV treatment require experiments with good antibodies against both proteins. In summary, we have found that YB-1 can separate different DNA duplex structures and this activity is increased when duplex molecules contain either mispaired bases or cisplatin modi®cations. YB-1 exhibits endonucleolytic activ- ity in addition to its exonuclease. Finally, YB-1 binds in vitro proteins involved in both base excision repair and mismatch repair pathways suggesting an involvement of YB-1 in these repair pathways. Additional in vitro as well as in vivo analyses are required to con®rm the functional interactions between YB-1 and these DNA repair proteins. ACKNOWLEDGEMENTS We are grateful to Dr P. E. DiCorleto from the Cleveland Clinic Foundation (Cleveland, OH) for the rabbit pre-immune à  and anti-YB-1 antibodies and to Jacques Cote from the Centre   de Recherche en Cancerologie (Quebec City) for the GST-p53 construct. We thank F. Deschenes for technical assistance and Figure 10. Co-localization of YB-1 and MSH2 by immuno¯uorescence. Dr P. O. de Campos-Lima for his helpful discussion. This Human 293 embryonic kidney cells were untreated (A±C) or incubated with 7.5 mg/ml of cisplatin for 2 h (D±F), then ®xed and subsequently incubated work was supported by a grant from the Canadian Institutes of simultaneously with rabbit anti-YB-1 and mouse anti-MSH2 as described in Health Research (CIHR) to M.L. M.L. is a scholar of the Materials and Methods. Anti-rabbit rhodamine-labeled and anti-mouse CIHR. FITC-conjugated secondary antibodies were used to visualize YB-1 and MSH2 by confocal microscopy at 568 and 488 nm, respectively. Images de- pict representative cells from each of the non-treated and cisplatin-treated cultures. In the merged images (C and F) a yellow color appears where YB- REFERENCES 1 (red) and MSH2 (green) ¯uorescence signals coincide. 1. Didier,D.K., Schiffenbauer,J., Woulfe,S.L., Zacheis,M. and Schwartz,B.D. (1988) Characterization of the cDNA encoding a protein binding to the major histocompatibility complex class II Y box. Proc. resistant to platinum-containing compounds (45). This pro- Natl Acad. Sci. USA, 85, 7322±7326. vides a link between the activity of YB-1 and the mismatch 2. Travali,S., Ku,D.-H., Rizzo,M.G., Ottavio,L., Baserga,R. and repair protein MSH2 in the processing of cisplatin-modi®ed Calabretta,B. (1989) Structure of the human gene for the proliferating DNA. Future experiments with puri®ed MSH2 will determine cell nuclear antigen. J. Biol. Chem., 264, 7466±7472. 3. Lipson,K.E., Chen,S.-T., Koniecki,J., Ku,D.-H. and Baserga,R. (1989) whether it can bind directly to YB-1 and alter its activities. S-phase speci®c regulation by deletion mutants of the human thymidine DNA polymerase d is another factor (with PCNA and EXO1) kinase promoter. Proc. Natl Acad. Sci. USA, 86, 6848±6852. associated with mismatch repair (46). It is interesting to note 4. Pearson,B.E., Nasheuer,H.P. and Wang,T.S.-F. (1991) Human DNA that DNA polymerase d bound our GST±YB-1 matrix and that polymerase alpha gene: sequences controlling expression in cycling and PCNA interacts with YB-1 in vivo (25). Again, these results serum-stimulated cells. Mol. Cell. Biol., 11, 2081±2095. 5. Zou,Y. and Chien,K.R. (1995) EFIA/YB-1 is a component of cardiac point to a potential role for YB-1 in the mismatch repair HF-1A binding activity and positively regulates transcription of the pathway. myosin light-chain 2v gene. Mol. Cell. Biol., 15, 2972±2982. Ku80 has been reported to have a role in transcriptional 6. Lasham,A., Lindridge,E., Rudert,F., Onrust,R. and Watson,J. (2000) reinitiation by RNA polymerase II in addition to double-strand Regulation of the human fas promoter by YB-1, Puralpha and AP-1 transcription factors. Gene, 252, 1±13. break repair (47). As YB-1 is a transcription factor, it is 7. Norman,J.T., Lindahl,G.E., Shakib,K., En-Nia,A., Yilmaz,E. and possible that the observed interaction with Ku80 is in the Mertens,P.R. (2001) The Y-box binding protein YB-1 suppresses context of transcription in cells and not DNA repair. However, collagen alpha 1(I) gene transcription via an evolutionarily conserved Ku antigen has also been shown to down regulate the activity regulatory element in the proximal promoter. J. Biol. Chem., 276, of the base excision repair protein endonuclease III in 29880±29890. 8. Fukada,T. and Tonks,N.K. (2003) Identi®cation of YB-1 as a regulator of the context of DNA molecules with closely opposed PTP1B expression: implications for regulation of insulin and cytokine lesions (48). Thus, Ku antigen will also affect some aspects signaling. EMBO J., 22, 479±493. of base excision repair in cells. Additional analyses on a 9. Mertens,P.R., Alfonso-Jaume,M.A., Steinmann,K. and Lovett,D.H. Ku/YB-1 complex are required to distinguish between these (1999) YB-1 regulation of the human and rat gelatinase A genes via possibilities. similar enhancer elements. J. Am. Soc. Nephrol., 10, 2480±2487. Nucleic Acids Research, 2004, Vol. 32, No. 1 327 10. Goldsmith,M.E,, Madden,M.J., Morrow,C.S. and Cowan,K.H. (1993) A 28. MacDonald,G.H., Itoh-Lindstrom,Y. and Ting,J.P. (1995) The Y-box consensus sequence is required for basal expression of the human transcriptional regulatory protein, YB-1, promotes single-stranded multidrug resistance (mdr1) gene. J. Biol. Chem., 268, 5856±5860. regions in the DRA promoter. J. Biol. Chem., 270, 3527±3533. 11. Ohga,T, Uchiumi,T., Makino,Y., Koike,K., Wada,M., Kuwano,M. and 29. Swamynathan,S.K., Nambiar,A. and Guntaka,R.V. (1998) Role of single- Kohno,K. (1998) Direct involvement of the Y-box binding protein YB-1 stranded DNA regions and Y-box proteins in transcriptional regulation of in genotoxic stress-induced activation of the human multidrug resistance viral and cellular genes. FASEB J., 12, 515±522. 1 gene. J. Biol. Chem., 273, 5997±6000. 30. Izumi,H., Imamura,T., Nagatani,G., Ise,T., Murakami,T., Uramoto,H., 12. Hartmuth,K., Urlaub,H., Vornlocher,H.P., Will,C.L., Gentzel,M., Torigoe,T., Ishiguchi,H., Yoshida,Y., Nomoto,M., Okamoto,T., Wilm,M. and Luhrmann,R. (2002) Protein composition of human Uchiumi,T., Kuwano,M., Funa,K. and Kohno,K. (2001) Y box-binding prespliceosomes isolated by a tobramycin af®nity-selection method. protein-1 binds preferentially to single-stranded nucleic acids and Proc. Natl Acad. Sci. USA, 99, 16719±16724. exhibits 3¢®5¢ exonuclease activity. Nucleic Acids Res., 29, 1200±1207. 13. Raffetseder,U., Frye,B., Rauen,T., Jurchott,K., Royer,H.D., Jansen,P.L. 31. Lebel,M., Spillare,E.A., Harris,C.C and Leder,P. (1999) The Werner and Mertens,P.R. (2003) Splicing factor SRp30c interaction with Y-box syndrome gene product co-puri®es with the DNA replication complex protein-1 confers nuclear YB-1 shuttling and alternative splice site and interacts with PCNA and topoisomerase I. J. Biol. Chem., 274, selection. J. Biol. Chem., 278, 18241±18248. 37795±37799. 14. Stickeler,E., Fraser,S.D., Honig,A., Chen,A.L., Berget,S.M. and 32. Bennett,R.J., Sharp,J.A. and Wang,J.C. (1998) Puri®cation and Cooper,T.A. (2001) The RNA binding protein YB-1 binds A/C-rich exon characterization of the Sgs1 DNA helicase activity of Saccharomyces enhancers and stimulates splicing of the CD44 alternative exon v4. cerevisiae. J. Biol. Chem., 273, 9644±9650. EMBO J., 20, 3821±3830. 33. Moggs,J.G., Yarema,K.J., Essigmann,J.M. and Wood,R.D. (1996) 15. Davydova,E.K., Evdokimova,V.M., Ovchinnikov,L.P. and Hershey,J.W. Analysis of incision sites produced by human cell extracts and puri®ed (1997) Overexpression in COS cells of p50, the major core protein proteins during nucleotide excision repair of a 1,3-intrastrand d(GpTpG)- associated with mRNA, results in translation inhibition. Nucleic Acids cisplatin adduct. J. Biol. Chem., 271, 7177±7186. Res., 25, 2911±2916. 34. Banin,S., Truong,O., Katz,D.R., Water®eld,M.D., Brickell,P.M. and 16. Ashizuka,M., Fukuda,T., Nakamura,T., Shirasuna,K., Iwai,K., Izumi,H., Gout,I. (1996) Wiskott-Aldrich syndrome protein (WASp) is a binding Kohno,K., Kuwano,M. and Uchiumi,T. (2002) Novel translational partner for c-Src family protein-tyrosine kinases. Curr. Biol., 6, 981±988. control through an iron-responsive element by interaction of 35. Hartmuth,K., Urlaub,H., Vornlocher,H.P., Will,C.L., Gentzel,M., multifunctional protein YB-1 and IRP2. Mol. Cell. Biol., 22, 6375±6383. Wilm,M. and Luhrmann,R. (2002) Protein composition of human 17. Chen,C.Y., Gherzi,R., Andersen,J.S., Gaietta,G., Jurchott,K., prespliceosomes isolated by a tobramycin af®nity-selection method. Royer,H.D., Mann,M. and Karin,M. (2000) Nucleolin and YB-1 are Proc. Natl Acad. Sci. USA, 99, 16719±16724. required for JNK-mediated interleukin-2 mRNA stabilization during 36. Burkle,A. (2001) Physiology and pathophysiology of poly(ADP- T-cell activation. Genes Dev., 14, 1236±1248. ribosyl)ation. Bioessays, 23, 795±806. 18. Bargou,R.C., Jurchott,K., Wagener,C., Bergmann,S., Metzner,S., 37. Marenstein,D.R., Chan,M.K., Altamirano,A., Basu,A.K., Boorstein,R.J., Bommert,K., Mapara,M.Y., Winzer,K.J., Dietel,M., Dorken,B. and Cunningham,R.P. and Teebor,G.W. (2003) Substrate speci®city of Royer,H.D. (1997) Nuclear localization and increased levels of human endonuclease III (hNTH1). Effect of human APE1 on hNTH1 transcription factor YB-1 in primary human breast cancers are associated activity. J. Biol. Chem., 278, 9005±9012. with intrinsic MDR1 gene expression. Nature Med., 3, 447±450. 38. Vidal,A.E., Boiteux,S., Hickson,I.D. and Radicella,J.P. (2001) XRCC1 19. Janz,M., Harbeck,N., Dettmar,P., Berger,U., Schmidt,A., Jurchott,K., coordinates the initial and late stages of DNA abasic site repair through Schmitt,M. and Royer,H.D. (2002) Y-box factor YB-1 predicts drug protein±protein interactions. EMBO J., 20, 6530±6539. resistance and patient outcome in breast cancer independent of clinically 39. Baumann,P., Benson,F.E. and West,S.C. (1996) Human Rad51 protein relevant tumor biologic factors HER2, uPA and PAI-1. Int. J. Cancer, 97, promotes ATP-dependent homologous pairing and strand transfer 278±282. reactions in vitro. Cell, 87, 757±766. 20. Rubinstein,D.B., Stortchevoi,A., Boosalis,M., Ashfaq,R. and 40. Masson,J.Y., Tarsounas,M.C., Stasiak,A.Z., Stasiak,A., Shah,R., Guillaume,T. (2002) Overexpression of DNA-binding protein B gene McIlwraith,M.J., Benson,F.E. and West,S.C. (2001) Identi®cation and product in breast cancer as detected by in vitro-generated combinatorial puri®cation of two distinct complexes containing the ®ve RAD51 human immunoglobulin libraries. Cancer Res., 62, 4985±4991. paralogs. Genes Dev., 15, 3296±3307. 21. Yahata,H., Kobayashi,H., Kamura,T., Amada,S., Hirakawa,T., 41. Opresko,P.L., Cheng,W.H., Von Kobbe,C., Harrigan,J.A. and Bohr,V.A. Kohno,K., Kuwano,M. and Nakano,H. (2002) Increased nuclear (2003) Werner syndrome and the function of the Werner protein; what localization of transcription factor YB-1 in acquired cisplatin-resistant they can teach us about the molecular aging process. Carcinogenesis, 24, ovarian cancer. J. Cancer Res. Clin. Oncol., 128, 621±626. 791±802. 22. Oda,Y., Ohishi,Y., Saito,T., Hinoshita,E., Uchiumi,T., Kinukawa,N., 42. Okamoto,T., Izumi,H., Imamura,T., Takano,H., Ise,T., Uchiumi,T., Iwamoto,Y., Kohno,K., Kuwano,M. and Tsuneyoshi,M. (2003) Nuclear Kuwano M. and Kohno,K. (2000) Direct interaction of p53 with the expression of Y-box-binding protein-1 correlates with P-glycoprotein Y-box binding protein, YB-1: a mechanism for regulation of human gene and topoisomerase II alpha expression and with poor prognosis in expression. Oncogene, 19, 6194±6202. synovial sarcoma. J. Pathol., 199, 251±258. 43. Zhang,Y.F., Homer,C., Edwards,S.J., Hananeia,L., Lasham,A., Royds,J., 23. Koike,K., Uchiumi,T., Ohga,T., Toh,S., Wada,M., Kohno,K. and Sheard,P. and Braithwaite,A.W. (2003) Nuclear localization of Y-box Kuwano,M. (1997) Nuclear translocation of the Y-box binding protein by factor YB1 requires wild-type p53. Oncogene, 22, 2782±2794. ultraviolet irradiation. FEBS Lett., 417, 390±394. 44. Gray,M.D., Shen,J.C., Kamath-Loeb,A.S., Blank,A., Sopher,B.L., 24. Hayakawa,H., Uchiumi,T., Fukuda,T., Ashizuka,M., Kohno,K., Martin,G.M., Oshima,J. and Loeb,L.A. (1997) The Werner syndrome Kuwano,M. and Sekiguchi,M. (2002) Binding capacity of human YB-1 protein is a DNA helicase. Nature Genet., 17, 100±103. protein for RNA containing 8-oxoguanine. Biochemistry, 41, 12739± 45. Fink,D., Aebi,S. and Howell,S.B. (1998) The role of DNA mismatch repair in drug resistance. Clin. Cancer Res., 4, 1±6. 25. Ise,T., Nagatani,G., Imamura,T., Kato,K., Takano,H., Nomoto,M., 46. Marti,T.M., Kunz,C. and Fleck,O. (2002) DNA mismatch repair and Izumi,H., Ohmori,H., Okamoto,T., Ohga,T., Uchiumi,T., Kuwano,M. mutation avoidance pathways. J. Cell. Physiol., 191, 28±41. and Kohno,K. (1999) Transcription factor Y-box binding protein 1 binds 47. Woodard,R.L., Lee,K.J., Huang,J. and Dynan,W.S. (2001) Distinct roles preferentially to cisplatin-modi®ed DNA and interacts with proliferating for Ku protein in transcriptional reinitiation and DNA repair. J. Biol. cell nuclear antigen. Cancer Res., 59, 342±346. Chem., 276, 15423±15433. 26. Jonsson,Z.O. and Hubscher,U. (1997) Proliferating cell nuclear antigen: 48. Hashimoto,M., Donald,C.D., Yannone,S.M., Chen,D.J., Roy,R. and more than a clamp for DNA polymerases. Bioessays, 19, 967±975. 27. Marenstein,D.R., Ocampo,M.T., Chan,M.K., Altamirano,A., Basu,A.K., Kow,Y.W. (2001) A possible role of Ku in mediating sequential repair of Boorstein,R.J., Cunningham,R.P. and Teebor,G.W. (2001) Stimulation of closely opposed lesions. J. Biol. Chem., 276, 12827±12831. human endonuclease III by Y box-binding protein 1 (DNA-binding 49. Blander,G., Zalle,N., Daniely,Y., Taplick,J., Gray,M.D. and Oren,M. protein B). Interaction between a base excision repair enzyme and a (2002) DNA damage-induced translocation of the Werner helicase is transcription factor. J. Biol. Chem., 276, 21242±21249. regulated by acetylation. J. Biol. Chem., 277, 50934±50940.

Journal

Nucleic Acids Research – Oxford University Press

Published: Jan 16, 2004

There are no references for this article.