| Edited by Kazuo Yamamoto. Masayuki Seki: Corresponding author. E-mail: seki@mail.pharm.tohoku.ac.jp |
Werner’s syndrome (WS) is a rare autosomal recessive disorder characterized by premature aging and early onset of age-related diseases, including arteriosclerosis, malignant neoplasms, melituria and cataract (Epstein et al., 1966). Somatic cells derived from WS patients show chromosome instability, a shorter life span in in vitro culture, accelerated telomere shortening, and defects in DNA replication (Martin, 1977; Fujiwara et al., 1977; Fukuchi et al., 1989; Schulz et al., 1996). We originally identified Werner helicase-interacting protein 1, WRNIP1, as a mouse protein that interacts with the gene product responsible for WS (Kawabe et al., 2001); we named it WHIP (Werner Helicase Interacting Protein), and it was later renamed as WRNIP1 (WRN Interacting Protein 1). WRNIP1 is a highly conserved protein from E. coli to humans (Kawabe et al., 2001). Mutation in the budding yeast orthlog for WRNIP1 caused hyper recombination and early aging of yeast cells (Kawabe et al., 2001; Branzei et al., 2002a). The gene encoding budding yeast orthlog for WRNIP1 was also found independently as one which encodes a protein belonging to the AAA+ class of ATPase and designated as MGS1 (Maintenance of Genome Stability 1) (Hishida et al., 2001).
Genetic analyses using MGS1 mutants revealed that Mgs1 is required for preventing the genome instability caused by replication arrest and is not involved in the repair of DNA lesions (Hishida et al., 2002). Overproduction of Mgs1 is lethal or very toxic in combination with mutations in genes that encode proteins involved in DNA replication, such as DNA polymerase δ (Polδ), RFC, PCNA, and RPA (Branzei et al., 2002b). Mgs1 physically and functionally interacts in vivo with budding yeast Pol31, the second subunit of Polδ (Vijeh Motlagh et al., 2006). Human WRNIP1 also directly interacts with human Polδ and stimulates Polδ activity by increasing the initiation frequency of DNA synthesis on template primers (Tsurimoto et al., 2005).
The mutation of MGS1 has been found to be synthetic lethal with rad6 and rad18 and to exhibit a synthetic growth defect with rad5 (Hishida et al., 2001). These are members of the RAD6 epistasis group, which consists of RAD6, RAD18, RAD5, MMS2, and UBC13 and plays central roles in damage tolerance. A Rad6-Rad18 complex ubiquitinates PCNA on lysine 164. Mono-ubiquitinated PCNA is targeted to stalled replication forks to initiate translesion DNA synthesis by specific DNA polymerases, such as Polη, which enables DNA synthesis to continue across the damage (Hoege et al., 2002; Stelter and Ulrich, 2003; Kannouche et al., 2004; Haracska et al., 2004). In addition, the Ubc13-Mms2 complex and Rad5 polyubiquitinate PCNA on lysine 164 through lysine 63-linked ubiqutin chains (Hoege et al., 2002). When PCNA is polyubiquitinated in this manner, it promotes damage bypass through template-switching (Zhang and Lawrence, 2005; Torres-Ramos et al., 2002; Blastyak et al., 2007). Interestingly, deletion of RAD18 or MMS2 suppresses the growth defect of Polδ mutants, and a similar phenomenon was observed after deleting MGS1 (Branzei et al., 2002b). Moreover, mgs1 mutation suppresses the growth defect of a Polδ mutant caused by expression of E. coli RuvC, a bacterial Holliday junction resolvase, suggesting that Mgs1 generates a Holliday junction-like structure that may be formed in the process of the template-switch type of damage bypass (Hishida et al., 2002). Furthermore, Mgs1 physically associates with PCNA and acts to prevent the RAD6 damage tolerance pathway (Hishida et al., 2006). These findings suggest that Mgs1 may function in a DNA-damage tolerance pathway that is similar to, but distinct from, the pathway involving Rad6, Rad18, Rad5, and Mms2.
WRNIP1 foci overlap with replication factories, suggesting that WRNIP1 may function at DNA replication forks and that some WRNIP1 foci overlap with RAD18 foci upon DNA damage (Crosetto et al., 2008). We previously identified a functional interaction between WRNIP1 and Rad18 by disrupting one or both of the genes in chicken DT40 cells (Yoshimura et al., 2006a). However, the functions of WRNIP1 in higher eukaryotic cells, as well as its functional relationship with Rad18, are not well understood. Thus, elucidating the functions of WRNIP1 will provide new insight into the template-switching DNA damage tolerance pathway.
In this study, to understand the functions of WRNIP1 in DNA damage tolerance, we studied biochemical characteristics of WRNIP1, particularly with regard to its interaction with RAD18 in binding to DNA.
RAD18 was purified as a complex with RAD6B as described (Tsuji et al., 2008). Briefly, human RAD18 and RAD6B proteins were expressed in Sf9 cells by using recombinant baculoviruses. The RAD6B protein had a polyhistidine tag at the N-terminal region. Cell lysates containing the RAD18-RAD6B complex were prepared from Sf9 cells and incubated with Ni-NTA (QIAGEN) at 4°C for 1 h. The resin was washed with buffer A (10 mM Na2HPO4, 10 mM NaH2PO4, 0.5 M NaCl, 5 μg/ml leupeptin, 5 μg/ml pepstatin A) containing 10 mM imidazole, and then washed with buffer A containing 60 mM imidazole. The proteins bound to the resin were eluted with buffer A containig 100 mM imidazole. Eluted fractions were applied to a MonoQ column and fractionated by elution with a linear gradient of 0.15 to 0.5 M NaCl in buffer B (30 mM HEPES-NaOH, pH 7.4, 1 mM dithiothreitol (DTT), 0.25 mM EDTA, 0.25% sucrose, 0.01% NP-40, 0.15 M NaCl, 2.5 μg/ml leupeptin, 2.5 μg/ml pepstatin A).
Preparation of virus for expression of human WRNIP1 has been previously described (Tsurimoto et al., 2005). Cell lysates were prepared from High Five cells expressing FLAG-tagged WRNIP1 and incubated with anti-FLAG agarose beads (SIGMA) at 4°C for 1 h. FLAG-WRNIP1 was eluted with buffer 1 (25 mM HEPES-NaOH, pH 7.4, 0.1 M NaCl, 1 mM EDTA, 10% glycerol, 0.01% TritonX-100, 2 μg/ml leupeptin) containing 100 μg/ml FLAG peptide. The eluate was fractionated on a MonoQ column by elution with a linear gradient of 0.1 to 0.6 M NaCl in buffer 2 (25 mM HEPES-NaOH, pH 7.4, 0.1 M NaCl, 1 mM EDTA, 10% glycerol, 0.1% Tween 20, 2 μg/ml leupeptin).
Oligonucleotides 49N2, 120N2, and 120N2R-80ss are previously described (Tsuji et al., 2008). The oligonucleotide 49N2R is complementary to 49N2. The blunt end double-stranded (ds) DNA was formed by annealing of 49N2 and 49N2R. The template/primer DNA with a 40-nucleotide ds region and 80-nucleotide single-stranded (ss) region was prepared by annealing 120N2 with 120N2R-80ss. The forked DNA was prepared as described previously (Komori et al., 2002; Hishida et al., 2004; Tsuji et al., 2008). Biotin labeled DNA was prepared using the Biotin 3’-End DNA Labeling Kit (PIERCE).
The reaction mixture for DNA binding contained 25 mM HEPES-NaOH, pH 7.6, 1 mM DTT, 50 mM KCl, 5 mM ATP, 5 mM MgCl2, 0.1% Tween 20, 2 nM 3’-biotin labeled DNA substrate, the indicated amount of WRNIP1 and/or RAD18-RAD6B. After the reaction mixtures were incubated at 37°C or kept on ice, products were analyzed by 6% or 4.5% native polyacrylamide gel electrophoresis (PAGE) in TBE buffer (45 mM Tris-borate, 1 mM EDTA) and detected using LightShift Chemiluminescent EMSA Kit (PIERCE).
Human 293EBNA cells were grown in DMEM (Dulbecco’s modified Eagle’s medium) supplemented with 10% calf serum. 293ENBNA cells were transfected with constructs encoding FLAG-tagged human WRNIP1 and HA-tagged human RAD18 using Lipofectamine 2000 Reagent (Invitrogen), in accordance with the manufecturer’s instructions. Cell lysates were prepared with lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% NP-40, and protease inhibitor cocktail (Roche)). The lysate was collected by centrifugation and incubated with IgG-Sepharose (GE Healthcare) at 4°C for 30 min. After centrifugation, the supernatant was incubated with anti-HA-agarose beads (SIGMA) at 4°C for 1 h. The beads were washed 5 times with the lysis buffer, resuspended in Laemmli sample buffer (50 mM Tris-HCl, pH 6.8, 2% SDS, 0.1% bromophenol blue, 10% glycerol) containing 1 mM DTT, and boiled for 3 min. Proteins were separated by SDS-PAGE and detected by western blotting using an anti-HA antibody (SIGMA) or an anti-FLAG antibody (SIGMA).
Purified WRNIP1 and RAD18-RAD6B were mixed in phosphate-bufferd saline (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4) containing 0.02% NP-40, 5 mM MgCl2, and protease inhibitor cocktail (Roche), and incubated for 10 min at 37°C. After adding anti-FLAG agarose beads (SIGMA), the mixture was rotated at 4°C for 1 h. The beads were washed 5 times with wash buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM MgCl2, 0.05% NP-40), resuspended in wash buffer containing 100 μg/ml FLAG peptide, and rotated at 4°C for 30 min. After centrifugation, eluted protein complexes were separated with SDS-PAGE and detected by western blotting.
Although WRNIP1 binds to Polδ and stimulates its DNA synthesizing activity in vitro (Tsurimoto et al., 2005), the function of WRNIP1 does not appear to be critical for normal DNA replication, because WRNIP1 is not essential for cell growth in chicken cells (Kawabe et al., 2006; Yoshimura et al., 2006a). Considering this in conjunction with the results obtained from genetic analysis using yeast mgs1 mutants, it seems likely that WRNIP1 function is required only under particular conditions, such as stalled replication forks upon encountering DNA lesions. To test this, we employed electrophoretic mobility shift assay (EMSA) using purified FLAG-WRNIP1 and RAD18/His-RAD6B complex (Fig. 1A) to examine the binding of WRNIP1 toward variously shaped DNAs, including DNA resembling stalled replication forks (Fig. 1B). We observed no binging of WRNIP1 and very little, if any, binding of RAD18 to a 49-mer ssDNA. However, we clearly detected binding of RAD18 to a 120-mer ssDNA, as reported previously (Tsuji et al., 2008), and also detected a small amount of slowly migrating 120-mer ssDNA in the presence of WRNIP1. A very low amount of slowly migrating dsDNA was detected in the presence of WRNIP1, whereas RAD18 did not bind to dsDNA, as reported (Tsuji et al., 2008). Both WRNIP1 and RAD18 bound to a 40-mer dsDNA with an 80-mer ssDNA tail, which mimics the template/primer structure, but the binding of WRNIP1 was much less efficient than that of RAD18. WRNIP1 as well as RAD18 bound efficiently to forked DNA, which resembles a stalled replication fork (Fig. 1B). When an anti-FLAG antibody was added to the reaction mixture for the binding assay containing WRNIP1 and forked DNA, the slowly migrating band was supershifted, and this supershift was reversed by the further addition of FLAG peptide, suggesting that the slowly migrating band is a FLAG-WRNIP1/forked DNA complex (Fig. 1C).
 View Details | Fig. 1 Binding of WRNIP1 and RAD18 to various DNA substrates. (A) Purification of humanWRNIP1 and human RAD18-RAD6B complex. HumanWRNIP1 and a complex of human RAD18-RAD6B were purified as described in “Experimental procedures” and subjected to 12% SDS-PAGE followed by staining with Coomasie Brilliant Blue. Arrow: ubiquitinated RAD18 (Miyase et al., 2005). (B) Substrate specificity. The DNA substrates indicated on the top of the figure (40 fmol of each) were incubated with 4 pmol WRNIP1 or RAD18-RAD6B at 37°C for 40 min. Protein-DNA complexes were analyzed with EMSA after 6% native-PAGE. Open and closed arrowheads indicate bottom of the wells and free DNA, respectively. Asterisks indicate biotinylation of DNA. (C) Confirmation of binding of WRNIP1 to DNA using an anti-FLAG antibody. The reaction mixtures containing 40 fmol forked DNA were incubated at 37°C for 40 min in the presence of WRNIP1 (4 pmol), anti-FLAG antibody, and FLAG peptide as indicated at the top of the figure. The protein-DNA complexes were analyzed with EMSA after 4.5% native PAGE. |
During the course of this study, we noticed that WRNIP1 binds to forked DNA more efficiently at 37°C than on ice (Fig. 2A). As shown in Fig. 2B, a considerable amount of WRNIP1-forked DNA complex was detected after 10 min incubation at 37°C.
 View Details | Fig. 2 Requirements for binding of WRNIP1 to DNA. (A) Dose response. Various amounts (0, 1, 2, 4, 8 pmol) of WRNIP1 were added to 40 fmol forked DNA, and the mixture was incubated at 37°C or on ice for 20 min. Protein-DNA complexes were analyzed with EMSA after 6% native-PAGE. (B) Time course. WRNIP1 (2 pmol) and forked DNA (40 fmol) were incubated at 37°C or on ice for the indicated length of time. Protein-DNA complexes were analyzed with EMSA after 6% native-PAGE. (C) ATP and Mg2+ requirement. WRNIP1 (4 pmol) and forked DNA (40 fmol) were incubated at 37°C for 40 min in the presence or absence of 5 mM ATP and 5 mM Mg2+, as indicated at the top of the figure. Protein-DNA complexes were analyzed with EMSA after 4.5% native-PAGE. Closed arrowheads indicate free DNA. |
WRNIP1 is the AAA+ - family proteins possessing the Walker A and B motifs, which are involved in trapping ATP and promoting ATP hydrolysis. In our previous study, we found that the interaction between WRNIP1 and WRN requires ATP (Kawabe et al., 2006). In addition, Mgs1 reportedly binds to a flap DNA in an ATP-dependent manner (Kim et al., 2005). Therefore, we examined the effect of ATP on the DNA-binding of WRNIP1. Omission of either ATP or Mg2+ abolished the formation of the WRNIP1-forked DNA complex (Fig. 2C). Next, we examined the effect of various nucleotides on the binding of WRNIP1 to the forked DNA. Binding of WRNIP1 to the forked DNA was equally efficient in the presence of ATP-γ-S as in the presence of ATP. In contrast, ADP, CTP, GTP, and UTP hardly supported the formation of the WRNIP1-DNA complex (data not shown).
WRNIP1 foci partly overlap with RAD18 foci, and both foci increases after treatment of cells with UVC irradiation (Crosetto et al., 2008). In addition, we found a functional interaction between WRNIP1 and RAD18 genetically using cells with the corresponding genes knocked out (Yoshimura et al., 2006a). We therefore examined whether WRNIP1 physically interacts with RAD18 in the cell. We co-expressed FLAG-tagged human WRNIP1 and HA-tagged human RAD18 in human 293EBNA cells and immunoprecipitated RAD18 with an anti-HA antibody. A portion of the WRNIP1 in the cell coimmunoprecipitated with RAD18 (Fig. 3A). To examine whether WRNIP1 interacts directly with RAD18, we mixed purified human WRNIP1 and the human RAD18-RAD6 complex, and immunoprecipitated WRNIP1 with an anti-FLAG antibody. The result clearly indicates the direct interaction between WRNIP1 and RAD18-RAD6 complex (Fig. 3B). We assume that this interaction occurs via a direct binding of WRNIP1 to the RAD18 subunit. It is noteworthy that, unlike the interaction between WRNIP1 and WRN (Kawabe et al., 2006), ATP was not required for the interaction between WRNIP1 and RAD18.
 View Details | Fig. 3 Physical interaction of WRNIP1 with RAD18. (A) Interaction in the cells. Anti-HA agarose beads were used to immunoprecipitate HA-RAD18 from cell lysates prepared from 293EBNA cells expressing both FLAG-WRNIP1 and HA-RAD18. Whole cell extracts (WCE) and the immunoprecipitates (IP) using anti-HA antibody were subjected to SDS-PAGE. WRNIP1 and RAD18 were detected by western blotting (IB) using an anti-FLAG and anti-HA antibodies, respectively. (B) Interaction of purified proteins. One pmol purified FLAG-WRNIP1 and 4 pmol RAD18-RAD6B were mixed and incubated as described in “Experimental procedures”. FLAG-WRNIP1 was immunoprecipitated using an anti-FLAG agarose beads. WRNIP1 and RAD18 were detected by western blotting using an anti-FLAG and anti-HA antibodies, respectively. Unbind; proteins unbound to anti-FLAG agarose. |
Genetic analysis revealed that mgs1 deletion is synthetically lethal in combination with rad6 or rad18 mutations (Hishida et al., 2002). In addition, Mgs1 may suppress the RAD6 damage tolerance pathway in the absence of DNA damaging treatments (Hishida et al., 2006), and we showed a direct physical interaction between WRNIP1 and RAD18 in this study. Considering these results together, it is likely that there is a cross-talk between WRNIP1 and RAD18. Because WRNIP1 and RAD18 bind to forked DNA with similar efficiency (Fig. 1), we examined the possible cross-talk between WRNIP1 and RAD18 in binding to forked DNA.
Neither WRNIP1-DNA nor RAD18-DNA complex was observed when WRNIP1 and RAD18 was incubated with forked-DNA in the absence of ATP, suggesting that WRNIP1 interferes with the binding of RAD18 to forked DNA without the formation of the ternary complex based on ATP-independent association between WRNIP1 and RAD18 (Fig. 4A). On the other hand, the incubation of WRNIP1 and RAD18 with forked DNA in the presence of ATP resulted in the appearance of shifted band, which migrated slower than RAD18-DNA complex and was highly similar to that of WRNIP1-DNA complex, suggesting that WRNIP1-DNA complex without RAD18 is formed in this condition.
 View Details | Fig. 4 Interference of WRNIP1 with RAD18 in DNA binding. (A) Inhibition of RAD18 binding to forked DNA by WRNIP1. Reaction mixtures containing 40 fmol forked DNA and 4 pmol RAD18-RAD6B, 16 pmol WRNIP1, and ATP, as indicated at the top of the figure, were incubated at 37°C for 40 min, and protein-DNA complexes were analyzed with EMSA after 4.5% native PAGE. (B) Inhibition of RAD18 binding to template/primer DNA by WRNIP1. Reaction mixtures containing 40 fmol template/primer DNA and 4 pmol RAD18-RAD6, 16 pmol WRNIP1, and ATP, as indicated at the top of the figure were incubated at 37°C for 40 min, and protein-DNA complexes were analyzed with EMSA after 4.5% native PAGE. (C) Enhancement of the binding of WRNIP1 to template/primer DNA by RAD18. Reaction mixtures containing 40 fmol template/primer DNA, ATP, and either 16 pmol WRNIP1 or 4 pmol RAD18-RAD6B, were incubated at 37°C for 20 min, and then various amounts of RAD18-RAD6 (1, 2, 4 pmol) or WRNIP1 (4, 8, 16 pmol) were added to the reaction mixtures containing WRNIP1 or RAD18-RAD6, respectively. Incubation was continued for a further 20 min, and protein-DNA complexes were analyzed with EMSA after 4.5% native PAGE. Open and closed arrowheads indicate bottom of the wells and free DNA, respectively. |
We next examined whether WRNIP1 interferes with the binding of RAD18 to template/primer DNA, to which WRNIP1 binds less efficiently than RAD18. Again, WRNIP1 abolished the binding of RAD18 to template/primer DNA, regardless of the presence of ATP (Fig. 4B). Interestingly, in the presence of ATP, the band corresponding to the WRNIP1-template/primer DNA complex was greatly increased in the presence of RAD18, suggesting that RAD18 increases the efficiency of WRNIP1 binding to template/primer DNA by recruiting WRNIP1 to the DNA. To clarify this issue, we first incubated a fixed amount of WRNIP1 or RAD18 with template/primer DNA and then added increasing amounts of RAD18 or WRNIP1 to the reaction mixture. When increasing amounts of RAD18 were added to the reaction after fixed amount of WRNIP1 was incubated with template/primer DNA, the band corresponding to DNA-bound WRNIP1 increased in proportion to the amount of RAD18 (Fig. 4C). In contrast, when the DNA was incubated with a fixed amount of RAD18 and then increasing amounts of WRNIP, the RAD18- template/primer DNA band was diminished by the addition of WRNIP1with a concomitant increase of the band corresponding to WRNIP1-template/primer DNA.
WRNIP1 was identified as a protein that interacts with WRN, the gene product responsible for WS, and is conserved from E. coli to humans (Hishida et al., 2001; Kawabe et al., 2001). Studies using Saccharomyces cerevisiae suggested that Mgs1, the yeast orthlog for WRNIP1, may participate in a DNA damage tolerance pathway that is an alternative to the Rad6-Rad18 pathway. In this study, we showed that WRNIP1 binds in an ATP dependent manner to forked DNA, which resembles stalled replication forks, and less efficiently to template/primer DNA. Because ATP-γ-S, a non-hydrolyzable analog of ATP, also supports the binding of WRNIP1 to the DNAs, it seems likely that ATP acts as a molecular switch for changing the structure of WRNIP1. The higher efficiency of binding of WRNIP1 to DNA at 37°C than on ice seems to be related to a conformational change of WRNIP1.
We have shown for the first time physical interaction between WRNIP1 and RAD18, although we cannot completely exclude the possibility that WRNIP1 indirectly interacts with RAD18 via interaction with RAD6B. Unlike the interaction between WRNIP1 and WRN (Kawabe et al., 2006), this interaction does not require ATP. WRNIP1 also interacts directly with Polδ and stimulates its activity on template/primer in the absence of ATP (Tsurimoto et al., 2005). Thus, it seems possible that WRNIP1 binds to template/primer weakly in the absence of ATP, and we were unable to detect this binding by gel-shift assay because of its weakness. In addition, WRNIP1 may interact with RAD18 or Polδ in a different mode from that of the interaction between WRNIP1 and WRN.
In addition to physical binding of WRNIP1 to RAD18, WRNIP1 interferes with the binding of RAD18. The results shown in Fig. 4 suggest that WRNIP1 preferentially recognizes the RAD18-DNA complex via interaction with RAD18 and binds to the DNA, probably resulting in displacement of RAD18 from the DNA. RAD18 is involved in at least two post-replication repair (PRR) pathways, translesion synthesis (TLS) and template-switching (Hoege et al., 2002; Stelter and Ulrich, 2003; Kannouche et al., 2004; Haracska et al., 2004; Zhang and Lawrence, 2005; Torres-Ramos et al., 2002; Blastyak et al., 2007). The RAD18-RAD6 complex recruited onto stalled replication forks mono-ubiquitinates PCNA and then allows the recruitment of damage-tolerant DNA polymerases, such as Polη for TLS, to the site containing DNA lesions (Stelter and Ulrich, 2003; Kannouche et al., 2004; Haracska et al., 2004). After the exchange of replicative DNA polymerase for a TLS polymerase and translesion synthesis, a second polymerase exchange is required to restore replication forks. In this context, the functional interaction between RAD18 and WRNIP1 found in this study may be necessary for the second polymerase change during TLS, because RAD18 recruits WRNIP1 and WRNIP1 recruits the replicative DNA polymerase, Polδ.
On the other hand, genetic evidences obtained from analyses of budding yeast mutants indicate that Mgs1 may function in a DNA damage tolerance pathway that is similar to, but distinct from, the pathway involving Rad6, Rad18, Rad5, Mms2, and Ubc13, and has been postulated to be a template-switch type of damage avoidance pathway (Hishida et al., 2002; Branzei et al., 2002b, 2004). Very recently, biochemical evidences supporting a Rad18- and Rad5-driven template-switch at stalled forks was reported (Branzei et al., 2008). Thus, another possible function of WRINP1 is to facilitate damage avoidance via a template-switching. In agreement with this idea, Mgs1 reportedly generates a Holliday junction-like structure in Polδ mutants, which may be formed in the process of the template-switch damage bypass (Hishida et al., 2002). In addition, Mgs1 has activity for annealing single-stranded DNA that is stimulated by ADP (Hishida et al., 2001).
Why would cells have two similar, but distinct, template-switch damage avoidance pathways? Because E. coli has a WRNIP1 homolog but not a Rad18 system, the pathway involving WRNIP1/Mgs1 seems to be an evolutionally conserved template-switch system for damage bypass, and the more sophisticated Rad18 system may have developed during evolution. However, the function of Rad18 is not restricted to template-switching, and Rad18 is also involved in translesion synthesis that is occasionally mutagenic (Stelter and Ulrich, 2003; Kannouche et al., 2004; Haracska et al., 2004), and recombination (Yamashita et al., 2002; Szuts et al., 2006; Yoshimura et al., 2006b). When lesions requiring template-switching are present, cells may primarily use the pathway involving WRNIP1/Mgs1 to avoid attendant risk of mutation, accomplishing this by replacing Rad18 with WRNIP1/Mgs1. In addition, when WRNIP1/Mgs1 pathway is defected, this pathway may be backed-up by the Rad18 template-switching pathway because mgs1 mutants show very mild phenotype. The fact that overexpression of Mgs1 suppresses damage-induced mutagenesis and causes sensitivity to damage-inducing agents in wild-type cells (Branzei et al., 2002b) can be explained by the suppression of all pathways involving Rad18 by WRNIP1/Mgs1. Thus, the above scenario is not in conflict with the data so far obtained.
WRNIP1 possesses the ubiquitin-binding zinc finger motif that is very similar to that of RAD18 (Kawabe et al., 2001) and binds to polyubiquitin chains and ubiquitinated proteins (Bish and Myers, 2007). In addition, WRNIP1 binds to WRN and Polδ (Kawabe et al., 2001; Tsurimoto et al., 2005), and Mgs1 binds to PCNA and Polδ (Hishida et al., 2006; Vijeh Motlagh et al., 2006). In this study, we only analyzed physical and functional interaction between WRNIP1 and RAD18. Studies using WRNIP1, RAD6-RAD18 pathway proteins, PCNA, ubiquitinated PCNA, Polδ and others will shed light on the mechanisms of template-switch damage avoidance and the relationship of these two template-switch pathways.
|
Index