Despite being a member of the mismatch repair family of proteins, the biological functions of hMSH5 in human cells are presently elusive. Here, we report a novel physical and functional interaction between hMSH5 and c-Abl; the latter is a critical non–receptor tyrosine kinase involved in many critical cellular functions including DNA damage response, in which the kinase activity is normally suppressed in the absence of biological challenges. Our data indicate that hMSH5 associates with c-Abl in vivo, which is mediated by a direct physical interaction between the NH2 terminus (residues 1-109) of hMSH5 and the c-Abl SH3 domain. This physical interaction facilitates the activation of c-Abl tyrosine kinase and the phosphorylation of hMSH5 in response to ionizing radiation. Our data also indicate that the hMSH5 P29S variant overactivates the c-Abl tyrosine kinase activity. Furthermore, it seems that the tyrosine phosphorylation of hMSH5 promotes the dissociation of hMSH4-hMSH5 heterocomplex. Together, the revealed physical and functional interaction of hMSH5 with c-Abl implies that the interplay between hMSH5 and c-Abl could manipulate cellular responses to ionizing radiation–induced DNA damages. (Cancer Res 2006; 66(1): 151-8)

Proteins that are involved in the DNA mismatch repair system are emerging as essential components of the DNA damage response pathways in mammalian cells (1, 2). Through dynamic interplay with the DNA damage response network, mismatch repair proteins play important roles in maintaining faithful genetic transmission in both mitotic and meiotic processes. Increasing evidence suggests that in addition to their apparent roles in the process of mismatch repair, members of the mismatch repair pathway are also involved in mediating cellular responses to a wide spectrum of DNA damages, in signaling cell cycle checkpoint control, and in the process of homologous recombination (2). It is generally known that the process of mismatch repair in mammalian cells requires the concerted action of multiple MutS and MutL homologues. Defects in any component of this pathway will render cells susceptible to genetic instability and predisposition to the development of cancer, in which the most prominent effects are seen in the development of hereditary nonpolyposis colorectal cancer (1). In addition, the mismatch repair pathway is critically linked to cell cycle checkpoint control, and mismatch repair deficiency represents an important factor in mediating antineoplastic drug resistance to many clinically important alkylating drugs used in cancer chemotherapy (35).

In spite of recent advances in our understanding of the mechanistic details involved with the human mismatch repair pathway, relatively little is known about the molecular mechanisms involving with human MutS homologue hMSH5 (6). Recent studies have shown that mammalian MSH5 interacts with MSH4, leading to the formation of a heterocomplex (710). Although experiments done with wild type or mice lacking functional Msh5 or Msh4 genes offered compelling evidence to suggest that Msh5 and Msh4 could act together in both early and late stages of meiotic homologous recombination (1113), the precise roles of these two proteins as well as the functional necessity of their interaction during different stages of meiotic homologous recombination are presently undefined. Consistent with the speculated roles of hMSH5 and hMSH4 in homologous recombination is the recent demonstration that the purified hMSH5-hMSH4 heterodimer could specifically bind to Holliday junctions in vitro (14).

In addition to its anticipated involvement in meiotic recombination, experimental evidence has also suggested that MSH5 could possess multiple cellular functions. For instance, mouse Msh5 played a critical role in chromosome synapsis, and Msh5 deficiency resulted in testicular and ovarian degeneration due to the massive induction of apoptosis (11, 12), and a MSH5 mutant, msh5-22, in Coprinus cinereus could cause defective premeiotic DNA replication (15). Furthermore, the involvement of MSH5 in DNA damage response has also been implicated in Saccharomyces cerevisiae (16, 17). In particular, a gain-of-function yeast MSH5 Y823H allele has been shown to possess a dominant effect in mediating tolerance to DNA-alkylating agents (17). To gain a better understanding of the functional roles of hMSH5 in humans, we have uncovered a novel functional association between hMSH5 and c-Abl. Our experiments show that hMSH5 physically interacts with c-Abl both in vivo and in vitro. In addition, our data indicate that hMSH5 represents a novel c-Abl substrate, and the common hMSH5 P29S variant overstimulates c-Abl tyrosine kinase activity.

Cell lines and mammalian transfection. The 293T/f-hMSH5 and 293T/f45 cell lines were described previously (18, 19). Cells were maintained in DMEM containing 10% fetal bovine serum (Biomeda, Foster City, CA) and 10 μg/mL blasticidin (Invitrogen, Carlsbad, CA). For 293T/f45 cells, 2 μg/mL puromycin (A.G. Scientific, San Diego, CA) was also included in the culture medium. The cDNA open reading frames (ORF) encoding the wild type and the kinase-inactive c-Abl (1a) K290R were cloned into pcDNA3.1-Myc (Invitrogen) to generate corresponding mammalian expression constructs. To express c-Abl in 293T/f-hMSH5 cells, relevant expression constructs were used to perform transient transfection with a standard calcium phosphate procedure.

Western blot analysis and antibodies. Western blotting was done as described previously (18, 19). Antibodies used in this study included α-c-Myc, α-HA, and α-His6 (Roche, Indianapolis, IN), α-hMSH4 (20), α-hMSH5 (18), α-FLAG (Sigma, St. Louis, MO), α-c-Abl monoclonal 8E9 (BD PharMingen, San Diego, CA), α-p-Tyr and α-p-c-Abl Tyr245 (Cell Signaling, Beverly, MA), α-GST (Amersham Pharmacia Biotech, Piscataway, NJ), α-GFP polyclonal (Clontech, Palo Alto, CA), and α-hMRE11 (Novus Biologicals, Inc., Littleton, CO). The Bio-Dot Microfiltration Apparatus (Bio-Rad, Hercules, CA) was used to immobilize purified proteins on nitrocellulose membranes for subsequent far-Western analysis, for which a procedure described previously was adapted (21).

Immunoaffinity purification and immunoprecipitation. Nuclear extracts of 293T/f-hMSH5 and 293T/f45 cells were prepared with NP40 lysis buffer containing 1× complete EDTA-free protease inhibitor cocktail (Roche, Mannheim, Germany), or with the CelLytic-M Mammalian Cell Lysis/Extraction Reagent (Sigma). Purification of FLAG-hMSH5 and its associated proteins was carried out with the anti-FLAG M2-agarose beads (Sigma) by the procedure recommended by the manufacturer. Coimmunoprecipitation analysis of hMSH5 and c-Abl interaction in 293T/f-hMSH5 cells was done with 10 μg of α-c-Abl or 10 μg of α-hMSH2 antibodies at 4°C.

Yeast two-hybrid and three-hybrid analyses. cDNA fragments encoding full-length and relevant regions of hMSH5, hMSH4, and c-Abl 1a proteins were cloned into vectors pBTM116, pGBKT7, pGADT7, pAS2-1, pACT2, and pBridge. Yeast two-hybrid and three-hybrid analyses were carried out with the reporter yeast strains AH109 or L40 as described previously (1820).

Ionizing irradiation. Irradiation of cells was carried out at room temperature with a cobalt-60 source at a dose rate of 11 Gy/min. Irradiated cells were harvested 1 hour after treatment.

Glutathione S-transferase-pulldown assay. Glutathione S-transferase (GST)-pulldown assay was done as described previously (18). Specifically, coding sequences for c-Abl aa1-130 and hMSH5 aa1-109 were cloned in-frame into pGEX-6p (Pharmacia, Piscataway, NJ) and pET-28a (Novagen, Madison, WI) vectors, respectively. The resulting expression constructs were then introduced into BL21(DE3)-RIL cells (Stratagene, La Jolla, CA) to produce GST-c-Abl aa1-130 and HA-hMSH5 aa1-109 recombinant proteins. Glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) were used to capture GST fusion proteins from soluble fractions of cell lysates. Cofractionated proteins were analyzed by Western blotting done with α-hMSH5 and α-GST antibodies.

Expression of recombinant proteins in Sf9 cells. The Bac-to-Bac Baculovirus Expression System (Invitrogen) was used to express recombinant hMSH5, c-Abl, and hMRE11 proteins in Sf9 cells. The hMSH5 cDNA ORF sequence was cloned into pFastBac HT to create pFastBac/hMSH5 construct, whereas pFastBac/c-Abl and pFastBac/hMRE11 constructs were kindly provided by Dr. Yosef Shaul (Weizmann Institute of Science) and Dr. Tanya Paull (University of Texas at Austin), respectively. Recombinant baculoviruses encoding each protein were amplified in Sf9 cells for three times to produce high-titer baculoviral stocks before final infection.

Production and purification of recombinant proteins. hMSH5 and hMSH4 cDNA fragments encoding protein interaction domains were subcloned into pET-28a, and the resulting constructs were transformed into BL21(DE3)-RIL to produce recombinant hMSH5 cp-1 and hMSH4 aa844-936 as His6 fusion proteins. Recombinant proteins were purified under denaturing condition with the TALON Metal Affinity Resins (Clontech). Proteins were renatured gradually by four serial dialyses in 1:1 dilution of initial 6 mol/L guanidine-PBS with the last dialysis in PBS alone.

Amino acid sequence analysis revealed that the hMSH5 NH2-terminal region contains a contiguous (Px)5 dipeptide repeat flanked by two PxxP motifs (Fig. 1). Intriguingly, the (Px)5 dipeptide repeat is disrupted in the hMSH5 P29S (Pro29Ser) variant encoded by a common hMSH5 polymorphic allele (18). It is known that the PxxP motif sequence represents a core scaffold in mediating interaction with SH3 domain-containing proteins; therefore, we have explored the possible link between hMSH5 and c-Abl non–receptor tyrosine kinase. In addition to its role in DNA damage response, c-Abl interacts with numerous DNA repair proteins, such as Rad51, Rad52, BRCA1, and WRN (2225), proteins known to be involved in the process of DNA homologous recombination. Furthermore, it has been shown that the c-Abl and p73 association is mediated through a p73 PxxP motif and the c-Abl SH3 domain (26). These intriguing observations encouraged us to investigate whether hMSH5 could interact with c-Abl in vivo.

Figure 1.

Schematic depiction of a (Px)5 and two PxxP motifs located at the NH2-terminal region of hMSH5. Only the NH2-terminal portion of hMSH5 is shown.

Figure 1.

Schematic depiction of a (Px)5 and two PxxP motifs located at the NH2-terminal region of hMSH5. Only the NH2-terminal portion of hMSH5 is shown.

Close modal

To this end, we first purified hMSH5-associated proteins by the anti-FLAG M2-agarose affinity beads from nuclear extracts of 293T/f-hMSH5 clone 3E (Fig. 2A). Immunoblotting analysis of FLAG-affinity purified protein mixture clearly showed that hMSH5 coexisted with c-Abl in the same protein complex, whereas neither was detected in affinity purified fractions from nuclear extracts of parental 293T cells (Fig. 2B). To validate the observed in vivo interaction between hMSH5 and c-Abl, reciprocal coimmunoprecipitation was done with a monoclonal α-c-Abl antibody (Fig. 2C). It was evident that the α-c-Abl monoclonal antibody could effectively immunoprecipitate both c-Abl and hMSH5, suggesting hMSH5 interacted with c-Abl in vivo (Fig. 2C). The specificity of the observed interaction between hMSH5 and c-Abl was validated as the α-c-Abl antibody could not immunoprecipitate hMSH2, whereas α-hMSH2 antibody could only immunoprecipitate hMSH2 but neither hMSH5 nor c-Abl (Fig. 2C). Thus, our experiments have shown the existence of an in vivo physical interaction between hMSH5 and c-Abl. In consideration of the potential roles of hMSH5 and c-Abl in DNA damage response and homologous recombination, the in vivo association of hMSH5 with c-Abl underscores the possibility of hMSH5 in DNA damage response and repair.

Figure 2.

In vivo association of hMSH5 and c-Abl. A, Western blot analysis of 293T stable transfectants expressing FLAG-tagged hMSH5. B, the presence of c-Abl in hMSH5-associated protein complex. Nuclear extracts from clone 3E and parental 293T cells were used to purify hMSH5 and its associated proteins with the anti-FLAG M2-agarose beads. The presence of hMSH5 and c-Abl in the immunoaffinity-purified preparation was analyzed by immunoblotting done with α-hMSH5 and α-c-Abl antibodies. C, coimmunoprecipitation analysis of hMSH5 and c-Abl interaction in 293T cells. Cell extracts of 293T/f-hMSH5 were incubated separately with α-c-Abl and α-hMSH2 antibodies. The resulting immunoprecipitates were analyzed by Western blots done with α-hMSH5, α-c-Abl, and α-hMSH2 antibodies. Lysate, 293T/f-hMSH5 cell extracts. kDa, molecular weight (Mr) in thousands.

Figure 2.

In vivo association of hMSH5 and c-Abl. A, Western blot analysis of 293T stable transfectants expressing FLAG-tagged hMSH5. B, the presence of c-Abl in hMSH5-associated protein complex. Nuclear extracts from clone 3E and parental 293T cells were used to purify hMSH5 and its associated proteins with the anti-FLAG M2-agarose beads. The presence of hMSH5 and c-Abl in the immunoaffinity-purified preparation was analyzed by immunoblotting done with α-hMSH5 and α-c-Abl antibodies. C, coimmunoprecipitation analysis of hMSH5 and c-Abl interaction in 293T cells. Cell extracts of 293T/f-hMSH5 were incubated separately with α-c-Abl and α-hMSH2 antibodies. The resulting immunoprecipitates were analyzed by Western blots done with α-hMSH5, α-c-Abl, and α-hMSH2 antibodies. Lysate, 293T/f-hMSH5 cell extracts. kDa, molecular weight (Mr) in thousands.

Close modal

It is conceivable that the NH2-terminal region of hMSH5 harboring two PxxP motifs and the (Px)5 repeat could be involved in interaction with c-Abl. To investigate whether this region physically associates with c-Abl through its SH3 domain, we next did GST-pulldown and two-hybrid analyses to investigate the potential physical interaction between hMSH5 aa1-109 and c-Abl SH3 domain. Specifically, recombinant GST-c-Abl aa1-130 and HA-hMSH5 aa1-109 fusion proteins were generated from pGEX-6p and pET-28a expression constructs, respectively. Soluble fractions of whole-cell lysates containing both HA-hMSH5 aa1-109 and GST-c-Abl aa1-130 proteins (lysate 1), or HA-hMSH5 aa1-109 alone (lysate 2) were used to perform the GST-pulldown assay (Fig. 3A). Glutathione-Sepharose 4B beads were incubated with lysates 1 and 2, and captured proteins were eluted into five contiguous fractions. As shown in Fig. 3A, immunoblotting analysis showed that hMSH5 aa1-109 was consistently coeluted with GST-c-Abl aa1-130. Moreover, the glutathione-Sepharose 4B beads were unable to capture hMSH5 aa1-109 in the absence of GST-c-Abl aa1-130 (Fig. 3A,, bottom). The direct physical interaction between hMSH5 and c-Abl SH3 was then validated in a two-hybrid analysis done with a series of c-Abl deletion constructs in the two-hybrid vector pGBKT7 (Fig. 3B). The results of this analysis indicated that hMSH5 interacted with the c-Abl SH3 domain, whereas other regions of the c-Abl protein, such as the SH2, kinase, and COOH-terminal domains, were not involved in the interaction (Fig. 3B), thus confirming the existence of a specific interaction between hMSH5 and the c-Abl SH3 domain. Interestingly, no measurable interaction between the full-length hMSH5 and c-Abl was detected by yeast two-hybrid analysis, although such an interaction was readily detected in human cells (Fig. 2). Although the Gal4 fusion portions (BD or AD) could potentially affect the interaction between these two proteins, the most likely scenario is that hMSH5 could be a c-Abl substrate and hMSH5 phosphorylation would negatively regulate the interaction between hMSH5 and c-Abl. It has been well established that in human cells, the c-Abl kinase activity is tightly controlled by intramolecular scaffold (SH3 domain) and multiple cellular components functioning as c-Abl inhibitors (2729), whereas there is no functional c-Abl homologue and its corresponding cellular inhibitors in yeast (30, 31). Therefore, the kinase activity of exogenously expressed c-Abl in S. cerevisiae is solely controlled by autoinhibition through the SH3 domain, which is consistent with the observation that c-Abl expressed in S. cerevisiae is not constitutively active (32). It is possible that the interaction with the c-Abl SH3 domain could activate c-Abl and cause concomitant hMSH5 tyrosine phosphorylation, presumably leading to a conformational change and the subsequent reduction of protein interaction. An analogous scenario has been described in the regulation of the interplay between c-Abl and BRCA1 (24). In addition, studies have also shown that the activation of c-Abl tyrosine kinase can result in the phosphorylation of DNA-PKcs, leading to the release of Ku70/Ku80 heterodimer (33, 34).

Figure 3.

Analysis of direct physical association between hMSH5 aa1-109 and c-Abl SH3 domain. A, recombinant hMSH5 aa1-109 and c-Abl aa1-130 proteins were produced as His6-tag or GST fusions, respectively. Copurification of hMSH5 aa1-109 and c-Abl aa1-130 was carried out with glutathione-Sepharose 4B. Cell lysates containing only His6-hMSH5 aa1-109 were used as a specificity control. Five elution fractions were analyzed with Western blotting to detect the presence of GST-c-Abl aa1-130 and His6-hMSH5 aa1-109. B, the full-length hMSH5 and a series of c-Abl deletion mutants were encoded in yeast two-hybrid vectors as either Gal4-AD or Gal4-BD fusion proteins. Analysis of protein interaction was carried out in reporter strain AH109 by the measurement of histidine prototrophy with three plus signs (+++) representing the positive interaction.

Figure 3.

Analysis of direct physical association between hMSH5 aa1-109 and c-Abl SH3 domain. A, recombinant hMSH5 aa1-109 and c-Abl aa1-130 proteins were produced as His6-tag or GST fusions, respectively. Copurification of hMSH5 aa1-109 and c-Abl aa1-130 was carried out with glutathione-Sepharose 4B. Cell lysates containing only His6-hMSH5 aa1-109 were used as a specificity control. Five elution fractions were analyzed with Western blotting to detect the presence of GST-c-Abl aa1-130 and His6-hMSH5 aa1-109. B, the full-length hMSH5 and a series of c-Abl deletion mutants were encoded in yeast two-hybrid vectors as either Gal4-AD or Gal4-BD fusion proteins. Analysis of protein interaction was carried out in reporter strain AH109 by the measurement of histidine prototrophy with three plus signs (+++) representing the positive interaction.

Close modal

Hence, we next attempted to address whether the interplay between hMSH5 and c-Abl could result in c-Abl activation and concomitant phosphorylation of hMSH5 in human, bacterial, and insect cells. To investigate the potential functional link between c-Abl-hMSH5 interplay and cellular DNA damage response in human cells, the wild-type and the kinase-inactive c-Abl (K290R) were expressed separately in 293T/f-hMSH5 cells. These cells were then irradiated with 20 Gy ionizing radiation and hMSH5 was immunoaffinity purified from both ionizing radiation treated and untreated control cells. As shown in Fig. 4, the results of Western blot analysis of immunoprecipitates clearly indicated that ionizing radiation led to profound phosphorylation of hMSH5 on tyrosine residues (α-p-Tyr blot). The observed ionizing radiation–induced tyrosine phosphorylation of hMSH5 is most likely mediated by c-Abl, as the expression of the kinase-inactive c-Abl (K290R) could not lead to enhanced tyrosine phosphorylation on hMSH5 over the basal level (Fig. 4). In addition, without exposure to ionizing radiation, expression of c-Abl could not increase hMSH5 phosphorylation, suggesting that c-Abl-mediated hMSH5 phosphorylation is triggered by ionizing radiation. In addition, the levels of hMSH5 and c-Abl protein expression were comparable in cells under different treatments (Fig. 4). Taken together, the data shows that ionizing radiation promotes c-Abl mediated tyrosine phosphorylation of hMSH5 in human cells, suggesting that the interplay of hMSH5 with c-Abl could have a critical function in DNA damage response.

Figure 4.

Ionizing radiation triggers c-Abl dependent hMSH5 phosphorylation. 293T/f-hMSH5 cells were transfected separately with c-Abl/pcDNA3.1-Myc and c-Abl (K290R)/pcDNA3.1-Myc, and then cells were irradiated with 20 Gy ionizing radiation at 48 hours after transfection. Irradiated and control cells were harvested 1 hour after treatment, and α-FLAG antibody was used to immunoprecipitate hMSH5 from cell lysate preparations. Ionizing radiation triggered c-Abl-dependent hMSH5 tyrosine phosphorylation was determined by α-p-Tyr immunoblotting. The presence of equivalent amounts of hMSH5 protein in the immunoprecipitates was validated by Western blot analysis. The expression levels of c-Abl and c-Abl (K290R) in transfected cells were examined by α-Myc immunoblotting. kDa, molecular weight (Mr) in thousands.

Figure 4.

Ionizing radiation triggers c-Abl dependent hMSH5 phosphorylation. 293T/f-hMSH5 cells were transfected separately with c-Abl/pcDNA3.1-Myc and c-Abl (K290R)/pcDNA3.1-Myc, and then cells were irradiated with 20 Gy ionizing radiation at 48 hours after transfection. Irradiated and control cells were harvested 1 hour after treatment, and α-FLAG antibody was used to immunoprecipitate hMSH5 from cell lysate preparations. Ionizing radiation triggered c-Abl-dependent hMSH5 tyrosine phosphorylation was determined by α-p-Tyr immunoblotting. The presence of equivalent amounts of hMSH5 protein in the immunoprecipitates was validated by Western blot analysis. The expression levels of c-Abl and c-Abl (K290R) in transfected cells were examined by α-Myc immunoblotting. kDa, molecular weight (Mr) in thousands.

Close modal

To validate if the functional interaction between hMSH5 and c-Abl involves c-Abl activation and hMSH5 tyrosine phosphorylation, an experimental system free of interference from endogenous protein kinases, and potential c-Abl regulators was used. Specifically, we expressed the full-length c-Abl and hMSH5 cp-1 or the control protein GFP in BL21(DE3)-RIL cells, from which soluble fractions of crude lysates were used for immunoblotting analysis done with α-p-Tyr and α-c-Abl antibodies. The cp-1 mutant is composed of amino acid residues 731 to 833 and 1 to 109 of the hMSH5 protein, which contains the c-Abl interacting motif and represents the fully functional hMSH4-interacting domain (18). This internal deleted form rather than the full-length hMSH5 was tested here because only the hMSH5 cp-1 fragment could be well expressed in bacterial cells as a soluble protein. It is known that phosphorylation of c-Abl SH3 domain at Tyr115 (Tyr134 in c-Abl 1b) along with other tyrosine residues occurs during the process of c-Abl activation (35). Evidently, the coexpression of hMSH5 cp-1 led to c-Abl autophosphorylation at tyrosine residues, which was further substantiated by the lack of c-Abl activation when it was expressed together with GFP (Fig. 5A). Remarkably, the activated c-Abl could efficiently phosphorylate hMSH5 cp-1 (Fig. 5A). We then tested whether the hMSH5 P29S variant could display differential effects on the activation of c-Abl by the use of the same bacterial expression system. As shown in Fig. 5B, in contrast to hMSH5 cp-1, coexpression of hMSH5 cp-1 P29S and c-Abl led to enhanced c-Abl autophosphorylation, which was readily detected by a c-Abl-specific phospho-Tyr245 antibody. Furthermore, the phosphorylation of hMSH5 by c-Abl could be recapitulated in Sf9 cells (Fig. 5C). Specifically, the coexpression of hMSH5 with c-Abl in Sf9 cells could lead to hMSH5 phosphorylation, and this modification was specific for hMSH5 as no hMRE11 phosphorylation was observed in the parallel control experiment (Fig. 5C). Together, the data show that hMSH5 is a c-Abl substrate, and the interaction of hMSH5 with c-Abl through its SH3 domain could potentially be involved in the modification of c-Abl kinase activity.

Figure 5.

Analysis of hMSH5 phosphorylation by c-Abl. A, BL21(DE3)-RIL cells were used to express full-length c-Abl (1a), HA-hMSH5 cp-1, and GFP recombinant proteins from pET-22- and pET-28-based constructs in different combinations as indicated. The expression of proteins was determined by Western blot analysis. Tyrosine phosphorylation of hMSH5 cp-1 and c-Abl was analyzed with immunoblotting done with α-p-Tyr. B, full-length c-Abl together with hMSH5 cp-1 or hMSH5 cp-1 (P29S) were expressed in BL21(DE3)-RIL cells. Immunoblotting with α-His-tag and α-c-Abl was used to examine the levels of protein expression, and a c-Abl specific α-p-c-Abl (Tyr245) antibody was used to assess c-Abl activation. C, examination of hMSH5 phosphorylation by c-Abl in insect cells. Sf9 cells were infected with different combinations of recombinant baculoviruses encoding c-Abl, hMSH5, and hMRE11. Levels of protein expression were determined at 48 hours after infection by Western blot analysis done with corresponding antibodies. The status of hMSH5 tyrosine phosphorylation was examined by α-p-Tyr immunoblotting. kDa, molecular weight (Mr) in thousands.

Figure 5.

Analysis of hMSH5 phosphorylation by c-Abl. A, BL21(DE3)-RIL cells were used to express full-length c-Abl (1a), HA-hMSH5 cp-1, and GFP recombinant proteins from pET-22- and pET-28-based constructs in different combinations as indicated. The expression of proteins was determined by Western blot analysis. Tyrosine phosphorylation of hMSH5 cp-1 and c-Abl was analyzed with immunoblotting done with α-p-Tyr. B, full-length c-Abl together with hMSH5 cp-1 or hMSH5 cp-1 (P29S) were expressed in BL21(DE3)-RIL cells. Immunoblotting with α-His-tag and α-c-Abl was used to examine the levels of protein expression, and a c-Abl specific α-p-c-Abl (Tyr245) antibody was used to assess c-Abl activation. C, examination of hMSH5 phosphorylation by c-Abl in insect cells. Sf9 cells were infected with different combinations of recombinant baculoviruses encoding c-Abl, hMSH5, and hMRE11. Levels of protein expression were determined at 48 hours after infection by Western blot analysis done with corresponding antibodies. The status of hMSH5 tyrosine phosphorylation was examined by α-p-Tyr immunoblotting. kDa, molecular weight (Mr) in thousands.

Close modal

To examine the potential downstream effects of the interaction between hMSH5 and c-Abl, we analyzed whether c-Abl could regulate the formation of hMSH5-hMSH4 heterocomplex. To this end, a “three-hybrid” approach was employed, in which the full-length c-Abl cDNA was inserted into pBridge-based constructs harboring either hMSH5 or hMSH4 ORF (20). The effects of c-Abl on the interaction between full-length proteins, as well as between hMSH4 and hMSH5 cp-1, were investigated. The results indicated that the expression of c-Abl in AH109 transformants completely abolished the interaction between hMSH5 cp-1 and hMSH4 and reduced the interaction between hMSH5 and hMSH4, whereas all AH109 double transformants grew efficiently on SD/-Leu-Trp medium (Fig. 6A). The observed effect of c-Abl on hMSH5-hMSH4 complex was highly specific, as expression of c-Abl did not affect the interaction between p53 and T-antigen (Fig. 6A). Because no detectable interaction was observed between the full-length c-Abl and hMSH5 in yeast (Fig. 3B), the reduction of hMSH5-hMSH4 interaction by c-Abl was most likely caused by protein posttranslational modification rather than the competition between c-Abl and hMSH4 for the binding of hMSH5. Consistent with the results presented in Fig. 5B, immunoblot analysis showed that the expression of hMSH5 could promote c-Abl autophosphorylation in yeast and this effect was more prominent for the mutant hMSH5 P29S protein (data not shown). However, our semiquantitative yeast two-hybrid analysis suggested that hMSH5 P29S could only slightly enhance protein interaction between hMSH5 and c-Abl SH3 domain, but this effect was not statistically significant (data not shown). It is conceivable that the conformational change caused by the P29S alteration would not significantly affect protein interaction with c-Abl.

Figure 6.

Regulation of hMSH4 and hMSH5 interaction by c-Abl. A, all of the constructs used in the analysis of complex protein interactions were created with pBridge and pACT2 or pGADT7 vectors. The effects of c-Abl on the formation of hMSH5-hMSH4 heterocomplex were evaluated by the growth potentials of AH109 double transformants on selection medium SD/-Leu-Trp-Ade-His. The effects of c-Abl on the interaction between p53 and T-antigen were examined as a specificity control. B, far-Western analysis of the effects of tyrosine phosphorylation on hMSH5-hMSH4 heterocomplex formation. His-tagged hMSH5 cp-1 and hMSH4 aa844-936, as well as phosphotyrosine containing hMSH5 cp-1 from bacterial cells co-expressing a kinase-active GST-c-Abl (top left) were affinity purified. Immunoblotting with α-p-Tyr was used to discriminate proteins containing phospho-tyrosine (top right). Bottom, approximately equal amounts of purified His6-tagged hMSH5 cp-1, pTyr-hMSH5 cp-1 proteins, or BSA were immobilized directly on the same nitrocellulose membrane, which was then probed with purified His6-tagged hMSH4 aa844-936 followed by a conventional Western blotting with α-hMSH4 antibody to detect captured hMSH4. C, analysis of hMSH4-hMSH5 interaction in ionizing radiation–treated 293T/f45 and untreated control cells. Irradiation of cells was carried out with 20 Gy ionizing radiation, and cells were harvested 1 hour after treatment. Approximately 10 μg of α-hMSH4 antibody were used to perform immunoprecipitation analysis. CONTROL, loading and antibody specificity control. kDa, molecular weight (Mr) in thousands.

Figure 6.

Regulation of hMSH4 and hMSH5 interaction by c-Abl. A, all of the constructs used in the analysis of complex protein interactions were created with pBridge and pACT2 or pGADT7 vectors. The effects of c-Abl on the formation of hMSH5-hMSH4 heterocomplex were evaluated by the growth potentials of AH109 double transformants on selection medium SD/-Leu-Trp-Ade-His. The effects of c-Abl on the interaction between p53 and T-antigen were examined as a specificity control. B, far-Western analysis of the effects of tyrosine phosphorylation on hMSH5-hMSH4 heterocomplex formation. His-tagged hMSH5 cp-1 and hMSH4 aa844-936, as well as phosphotyrosine containing hMSH5 cp-1 from bacterial cells co-expressing a kinase-active GST-c-Abl (top left) were affinity purified. Immunoblotting with α-p-Tyr was used to discriminate proteins containing phospho-tyrosine (top right). Bottom, approximately equal amounts of purified His6-tagged hMSH5 cp-1, pTyr-hMSH5 cp-1 proteins, or BSA were immobilized directly on the same nitrocellulose membrane, which was then probed with purified His6-tagged hMSH4 aa844-936 followed by a conventional Western blotting with α-hMSH4 antibody to detect captured hMSH4. C, analysis of hMSH4-hMSH5 interaction in ionizing radiation–treated 293T/f45 and untreated control cells. Irradiation of cells was carried out with 20 Gy ionizing radiation, and cells were harvested 1 hour after treatment. Approximately 10 μg of α-hMSH4 antibody were used to perform immunoprecipitation analysis. CONTROL, loading and antibody specificity control. kDa, molecular weight (Mr) in thousands.

Close modal

To validate that tyrosine phosphorylation by c-Abl plays a role in the regulation of hMSH5-hMSH4 heterocomplex formation, we purified recombinant hMSH5 cp-1 and hMSH4 aa844-936 (containing hMSH5-interacting domain; refs. 18, 19), as well as hMSH5 cp-1 coexpressed with GST-c-Abl from BL21(DE3)-RIL cells (Fig. 6B,, top). Immunoblotting analysis confirmed that coexpression with c-Abl led to phosphotyrosine containing hMSH5 cp-1 protein (designated as pTyr-hMSH5 cp-1; Fig. 6B). To test whether tyrosine phosphorylation leads to a reduction in protein interaction between hMSH5 and hMSH4, far-Western blotting analysis was done with the purified proteins. As presented in Fig. 6B (bottom), immobilized pTyr-hMSH5 cp-1 and hMSH5 cp-1 proteins but not the control protein bovine serum albumin (BSA) specifically interacted with hMSH4 aa844-936 fragment. Furthermore, with hMSH5 cp-1 as a reference, immobilized pTyr-hMSH5 cp-1 displayed weakened interaction with hMSH4 aa844-936 (Fig. 6B). In addition, the specificity of the far-Western analysis was validated with the inclusion of BSA, as well as the addition of antibody specificity and protein loading controls (Fig. 6B, Control). Together with the results of yeast three-hybrid analysis, the far-Western data show that phosphorylation by c-Abl kinase leads to the weakening of hMSH4-hMSH5 interaction, implying the hMSH4-hMSH5 heterocomplex could be subjected to regulation in the cells. Consistent with this view, experiments done with 293T/f45 cells, stable transfectants expressing both hMSH4 and hMSH5 proteins (19), showed that ionizing radiation (20 Gy) significantly compromised the integrity of the hMSH4-hMSH5 heterocomplex in these cells (Fig. 6C). Specifically, the α-hMSH4 antibody could only coimmunoprecipitate a relatively small amount of hMSH5 from ionizing radiation–treated 293T/f45 cells compared with that of the untreated control cells, despite approximately equal amounts of hMSH4 were present in both immunoprecipitates (Fig. 6C). Furthermore, coimmunoprecipitation experiments done with α-FLAG showed that c-Abl coexisted with hMSH4-hMSH5 complex in untreated cells (data not shown). Together, this series of experiments supports the view that activation of c-Abl tyrosine kinase will lead to hMSH5 phosphorylation, of which one functional consequence is the attenuation of the interaction between hMSH4 and hMSH5.

In the present study, we show that hMSH5 is both physically and functionally associated with c-Abl non–receptor tyrosine kinase; the latter represents a critical mediator for DNA damage–induced cellular response, in which c-Abl activation is primarily caused by ATM-mediated phosphorylation (36, 37). Constitutive activation of c-Abl kinase activity, such as in the case of oncogenic Bcr-Abl, has been linked to the development of chronic myeloid leukemia and a subset of acute lymphocytic leukemia (38). Being one of the critical regulators in the DNA damage response circuit, the kinase activity of the nuclear c-Abl in human cells is controlled by the outcome of interactions among intramolecular scaffold and multiple cellular components functioning as c-Abl inhibitors (27, 28).

Our study suggests that hMSH5 is a potential substrate and an activator for c-Abl. Because the SH3 domain is known to play an essential role in c-Abl autoinhibition through intrinsic intramolecular interactions (28), presumably the activation of c-Abl could be achieved by liberating intramolecular constrains through the interaction between hMSH5 proline-rich NH2 terminus and c-Abl SH3 domain. Apparently, the in vivo c-Abl activation by hMSH5 has to circumvent the inhibitory effects of many other c-Abl regulators. In this context, the observed in vivo association between hMSH5 and c-Abl supports a view that hMSH5-mediated c-Abl activation must be dynamically regulated by a yet-to-be-defined mechanism. Obviously, the molecular definition of hMSH5-associated protein complex, as well as the understanding of its dynamic composition in response to cellular stimuli, is essential to fully appreciate the functions associated with hMSH5.

The observation that hMSH5 and c-Abl exist in a common protein complex in human cells (Fig. 2) underscores the potential involvement of hMSH5 in DNA damage response and repair. Increasing evidence indicates that both hMSH5 and c-Abl are involved in the process of homologous recombination. This view is particularly pertinent to the observation that the strand exchange activity of Rad51 is regulated by c-Abl, and that the phosphorylation of Rad51 by c-Abl modifies its function in homologous recombination repair of DNA double-strand breaks (22, 39). Consistent with the view that homologous recombination is a prerequisite for functional chromosome pairing in mammals (11, 40), mouse lacking functional Msh5 displayed defective homologous chromosome pairing (11, 12). Besides interaction with many proteins that are involved in DNA homologous recombination, c-Abl is also known to negatively regulate DNA-PK complex leading to the release of DNA-PKcs and Ku70/Ku80 heterodimer (33, 34). Thus, the interplay between hMSH5 and c-Abl could impinge on the balance between homologous recombination and nonhomologous end joining–mediated repair of DNA double-strand breaks.

Consonant with our observation that overexpression of hMSH5 in human cells confers increased cellular sensitivity to ionizing radiation,1

1

W. Yi and C. Her, unpublished data.

the interaction between hMSH5 and c-Abl could be expected to play a role in DNA damage sensing and signaling. The current available evidence supports a role of hMSH5 in cellular response to DNA damage, and this effect of hMSH5 could be resulted from the dynamic interplay with c-Abl. Obviously, a better understanding of the functional consequences associated with hMSH5 phosphorylation by c-Abl awaits the future identification of the involved tyrosine residues and the clarification of the precise functional roles of hMSH5 phosphorylation in DNA damage response. The potential involvement of hMSH5 in DNA damage response has gained experimental support as a Y823H mutant of the S. cerevisiae hMSH5 homologue was found to mediate cellular tolerance to DNA-alkylating agents (17). To reconcile the apparent absence of a functional c-Abl homologue in yeast, we speculate that yeast and human MSH5 homologues play overlapping but distinct functional roles in DNA damage response. This view is well reflected by the observed functional diversity associated with yeast and mammalian MSH5 in meiotic homologous recombination, of which yeast MSH5 promotes chromosomal crossover events, whereas the mouse counterpart is required for homologous chromosome pairing or possibly homology searching (11, 41). Alternatively, it is possible that other kinases, unrelated to c-Abl, could act together with MSH5 in yeast to mediate DNA damage responses.

Although the precise functional consequence of c-Abl-mediated hMSH5 tyrosine phosphorylation is largely unknown, one functional implication of their interplay is reflected by the observation that c-Abl-mediated phosphorylation could regulate the availability of the hMSH5-hMSH4 heterocomplex in cells. Because the formation of hMSH5-hMSH4 complex is essential for subsequent recruitment of GPS2 (19), the regulation of this heterocomplex by c-Abl should be expected to have a significant influence on the dynamic molecular composition of the functional complex involved in DNA damage response. In this regard, it is of great interest to know whether c-Abl-mediated tyrosine phosphorylation will augment the action of hMSH5 in homologous recombination or simply modify the dynamics of the hMSH5-associated complex to end its participation in that process. Clearly, the detailed molecular mechanisms associated with hMSH5 and its phosphorylated form, as well as the cellular signals that promote hMSH5-mediated c-Abl activation remain to be clarified.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Y. Chen and Dr. N. Zhao for their technical assistance, Dr. Y. Shaul for pFastBac/c-Abl, and Dr. T. Paull for pFastBac/hMRE11 construct.

1
Harfe BD, Jinks-Robertson S. DNA mismatch repair and genetic instability.
Annu Rev Genet
2000
;
34
:
359
–99.
2
Stojic L, Brun R, Jiricny J. Mismatch repair and DNA damage signalling.
DNA Repair (Amst)
2004
;
3
:
1091
–101.
3
Hawn MT, Umar A, Carethers JM, et al. Evidence for a connection between the mismatch repair system and the G2 cell cycle checkpoint.
Cancer Res
1995
;
55
:
3721
–5.
4
Kong S, Amos CI, Luthra R, Lynch PM, Levin B, Frazier ML. Effects of cyclin D1 polymorphism on age of onset of hereditary nonpolyposis colorectal cancer.
Cancer Res
2000
;
60
:
249
–52.
5
Lage H, Dietel M. Involvement of the DNA mismatch repair system in antineoplastic drug resistance.
J Cancer Res Clin Oncol
1999
;
125
:
156
–65.
6
Her C, Doggett NA. Cloning, structural characterization, and chromosomal localization of the human orthologue of Saccharomyces cerevisiae MSH5 gene.
Genomics
1998
;
52
:
50
–61.
7
Her C, Wu X, Wan W, Doggett NA. Identification and characterization of the mouse MutS homolog 5: Msh5.
Mamm Genome
1999
;
10
:
1054
–61.
8
Her C, Wu X, Bailey SM, Doggett NA. Mouse MutS homolog 4 is predominantly expressed in testis and interacts with MutS homolog 5.
Mamm Genome
2001
;
12
:
73
–6.
9
Winand NJ, Panzer JA, Kolodner RD. Cloning and characterization of the human and Caenorhabditis elegans homologs of the Saccharomyces cerevisiae MSH5 gene.
Genomics
1998
;
53
:
69
–80.
10
Bocker T, Barusevicius A, Snowden T, et al. hMSH5: a human MutS homologue that forms a novel heterodimer with hMSH4 and is expressed during spermatogenesis.
Cancer Res
1999
;
59
:
816
–22.
11
Edelmann W, Cohen PE, Kneitz B, et al. Mammalian MutS homologue 5 is required for chromosome pairing in meiosis.
Nat Genet
1999
;
21
:
123
–7.
12
de Vries SS, Baart EB, Dekker M, et al. Mouse MutS-like protein Msh5 is required for proper chromosome synapsis in male and female meiosis.
Genes Dev
1999
;
13
:
523
–31.
13
Kneitz B, Cohen PE, Avdievich E, et al. MutS homolog 4 localization to meiotic chromosomes is required for chromosome pairing during meiosis in male and female mice.
Genes Dev
2000
;
14
:
1085
–97.
14
Snowden T, Acharya S, Butz C, Berardini M, Fishel R. hMSH4-hMSH5 recognizes Holliday Junctions and forms a meiosis-specific sliding clamp that embraces homologous chromosomes.
Mol Cell
2004
;
15
:
437
–51.
15
Cummings WJ, Merino ST, Young KG, et al. The Coprinus cinereus adherin Rad9 functions in Mre11-dependent DNA repair, meiotic sister-chromatid cohesion, and meiotic homolog pairing.
Proc Natl Acad Sci U S A
2002
;
99
:
14958
–63.
16
Bawa S, Xiao W. A mutation in the MSH5 gene results in alkylation tolerance.
Cancer Res
1997
;
57
:
2715
–20.
17
Bawa S, Xiao W. A single amino acid substitution in MSH5 results in DNA alkylation tolerance.
Gene
2003
;
315
:
177
–82.
18
Yi W, Wu X, Lee T-H, Doggett NA, Her C. Two variants of MutS homolog hMSH5: prevalence in humans and effects on protein interaction.
Biochem Biophys Res Commun
2005
;
332
:
524
–32.
19
Lee T-H, Yi W, Griswold MD, Zhu F, Her C. Formation of hMSH4-hMSH5 heterocomplex is a prerequisite for subsequent GPS2 recruitment. DNA Repair (Amst) 2005. Epub 2005 Aug 22.
20
Her C, Wu X, Griswold MD, Zhou F. Human MutS homologue MSH4 physically interacts with von Hippel-Lindau tumor suppressor-binding protein 1.
Cancer Res
2003
;
63
:
865
–72.
21
Vo AT, Zhu F, Wu X, et al. hMRE11 deficiency leads to microsatellite instability and defective DNA mismatch repair.
EMBO Rep
2005
;
6
:
438
–44.
22
Yuan ZM, Huang Y, Ishiko T, et al. Regulation of Rad51 function by c-Abl in response to DNA damage.
J Biol Chem
1998
;
273
:
3799
–802.
23
Kitao H, Yuan ZM. Regulation of ionizing radiation-induced Rad52 nuclear foci formation by c-Abl-mediated phosphorylation.
J Biol Chem
2002
;
277
:
48944
–8.
24
Foray N, Marot D, Randrianarison V, et al. Constitutive association of BRCA1 and c-Abl and its ATM-dependent disruption after irradiation.
Mol Cell Biol
2002
;
22
:
4020
–32.
25
Cheng WH, von Kobbe C, Opresko PL, et al. Werner syndrome protein phosphorylation by Abl tyrosine kinase regulates its activity and distribution.
Mol Cell Biol
2003
;
23
:
6385
–95.
26
Agami R, Blandino G, Oren M, Shaul Y. Interaction of c-Abl and p73α and their collaboration to induce apoptosis.
Nature
1999
;
399
:
809
–13.
27
Pluk H, Dorey K, Superti-Furga G. Auto-inhibition of c-Abl.
Cell
2002
;
108
:
247
–59.
28
Wang JY. Controlling Abl: auto-inhibition and co-inhibition?
Nat Cell Biol
2004
;
6
:
3
–7.
29
Hantschel O, Superti-Furga G. Regulation of the c-Abl and Bcr-Abl tyrosine kinases.
Nat Rev Mol Cell Biol
2004
;
5
:
33
–44.
30
Walkenhorst J, Goga A, Witte ON, Superti-Furga G. Analysis of human c-Abl tyrosine kinase activity and regulation in S. pombe.
Oncogene
1996
;
12
:
1513
–20.
31
Manning G, Plowman GD, Hunter T, Sudarsanam S. Evolution of protein kinase signaling from yeast to man.
Trends Biochem Sci
2002
;
27
:
514
–20.
32
Brasher BB, Roumiantsev S, Van Etten RA. Mutational analysis of the regulatory function of the c-Abl Src homology 3 domain.
Oncogene
2001
;
20
:
7744
–52.
33
Jin S, Kharbanda S, Mayer B, Kufe D, Weaver DT. Binding of Ku and c-Abl at the kinase homology region of DNA-dependent protein kinase catalytic subunit.
J Biol Chem
1997
;
272
:
24763
–6.
34
Kharbanda S, Pandey P, Jin S, et al. Functional interaction between DNA-PK and c-Abl in response to DNA damage.
Nature
1997
;
386
:
732
–5.
35
Musacchio A, Wilmanns M, Saraste M. Structure and function of the SH3 domain.
Prog Biophys Mol Biol
1994
;
61
:
283
–97.
36
Baskaran R, Wood LD, Whitaker LL, et al. Ataxia telangiectasia mutant protein activates c-Abl tyrosine kinase in response to ionizing radiation.
Nature
1997
;
387
:
516
–9.
37
Shafman T, Khanna KK, Kedar P, et al. Interaction between ATM protein and c-Abl in response to DNA damage.
Nature
1997
;
387
:
520
–3.
38
Advani AS, Pendergast AM. Bcr-Abl variants: biological and clinical aspects.
Leuk Res
2002
;
26
:
713
–20.
39
Slupianek A, Schmutte C, Tombline G, et al. BCR/ABL regulates mammalian RecA homologs, resulting in drug resistance.
Mol Cell
2001
;
8
:
795
–806.
40
Mahadevaiah SK, Turner JM, Baudat F, et al. Recombinational DNA double-strand breaks in mice precede synapsis.
Nat Genet
2001
;
27
:
271
–6.
41
Hollingsworth NM, Ponte L, Halsey C. MSH5, a novel MutS homolog, facilitates meiotic reciprocal recombination between homologs in Saccharomyces cerevisiae but not mismatch repair.
Genes Dev
1995
;
9
:
1728
–39.