MOLECULAR AND CELLULAR BIOLOGY, Feb. 2010, p. 1041–1048 0270-7306/10/$12.00 doi:10.1128/MCB.01198-09 Copyright © 2010, American Society for Microbiology. All Rights Reserved.
Vol. 30, No. 4
Pirh2 E3 Ubiquitin Ligase Targets DNA Polymerase Eta for 20S Proteasomal Degradation䌤 Yong-Sam Jung, Gang Liu, and Xinbin Chen* Center for Comparative Oncology, University of California, Davis, California 95616 Received 3 September 2009/Returned for modification 22 October 2009/Accepted 7 December 2009
DNA polymerase eta (PolH), a Y family translesion polymerase, is required for repairing UV-induced DNA damage, and loss of PolH is responsible for early onset of malignant skin cancers in patients with xeroderma pigmentosum variant (XPV), an autosomal recessive disorder. Here, we show that PolH, a target of the p53 tumor suppressor, is a short-half-life protein. We found that PolH is degraded by proteasome, which is enhanced upon UV irradiation. We also found that PolH interacts with Pirh2 E3 ligase, another target of the p53 tumor suppressor, via the polymerase-associated domain in PolH and the RING finger domain in Pirh2. In addition, we show that overexpression of Pirh2 decreases PolH protein stability, whereas knockdown of Pirh2 increases it. Interestingly, we found that PolH is recruited by Pirh2 and degraded by 20S proteasome in a ubiquitin-independent manner. Finally, we observed that Pirh2 knockdown leads to accumulation of PolH and, subsequently, enhances the survival of UV-irradiated cells. We postulate that UV irradiation promotes cancer formation in part by destabilizing PolH via Pirh2-mediated 20S proteasomal degradation. independent manner. We also showed that upon knockdown of Pirh2, PolH is accumulated and, consequently, desensitizes cells to UV-induced cell killing. Based on these observations, we postulate that UV irradiation promotes cancer formation in part by destabilizing PolH via Pirh2-mediated 20S proteasome degradation.
Polymerase eta (PolH) is a member of the Y family translesion DNA polymerases and capable of translesion synthesis over UV-induced cyclobutane pyrimidine dimers (CPDs) and (6-4) photoproducts (7). PolH is also involved in doublestranded break repair via homologous recombination (15, 23). Human PolH is the product of the xeroderma pigmentosum variant (XPV) gene (14, 22). XPV, an autosomal recessive disorder, exhibits clinical phenotypes of extreme sun sensibility, cutaneous and ocular deterioration, and early onset of malignant skin cancers. Thus, it is postulated that loss of PolH is responsible for accumulation of UV-induced lesions, which lead to early onset of multiple skin cancers in XPV patients. The ubiquitin-dependent degradation pathway plays a key role in many cellular processes, including cell proliferation, differentiation, and DNA repair (6, 10, 11). The pathway involves multiple enzymatic reactions catalyzed by a single ubiquitin-activating enzyme (E1), several ubiquitin-conjugating enzymes (E2s), and a large number of ubiquitin ligases (E3s). Protein polyubiquitination serves as a signal for rapid degradation by 26S proteasome, whereas monoubiquitination modulates protein function (3, 30). 26S proteasome is a multisubunit protease consisting of a core 20S proteasome and two 19S regulatory particles (24). 20S proteasome on its own is a broad-spectrum ATP- and ubiquitin-independent protease. 19S regulatory particles recognize and thread polyubiquitinated proteins into 20S proteasome for degradation in an ATP-dependent manner. The RING-H2 type E3 ligase (Pirh2) is regulated by p53 and targets p53 for degradation (19). Recently, studies showed that Pirh2 interacts with and potentially serves as an E3 ligase for TIP60 (21) and p27Kip1 (8). Here, we show that PolH protein stability is reduced by UV irradiation via Pirh2 in a ubiquitin-
MATERIALS AND METHODS Antibodies. Antibodies used in this study were as follows: rabbit polyclonal and mouse monoclonal anti-PolH (Santa Cruz Biotechnology), mouse monoclonal anti-ubiquitin (Santa Cruz Biotechnology), anti-20S (PW8155; Affiniti), mouse monoclonal anti-19S (p45-110; Affiniti), rabbit polyclonal anti-Pirh2 antibody (Bethyl Laboratories), monoclonal anti-HA (HA11; Covance), anti-FLAG monoclonal antibody (Sigma), anti-p53 monoclonal antibodies (DO-1, PAb1801, PAb240, and PAb421), antiactin (Sigma), and anti-p21 (C-19) (Santa Cruz Biotechnology). Measurement of protein half-life. RKO cells were incubated with cycloheximide (CHX, 10 g/ml; Sigma) to inhibit de novo protein synthesis for different time points before analysis along with MG132 (5 M; Sigma) or lactacystin (5 M; A.G. Scientific). Protein levels were quantified from three independent assays and plotted as log scale versus time (h), which was then used to calculate the half-life of PolH and p53. Plasmids and mutagenesis. All constructs were verified by DNA sequencing. Pirh2 cDNA was amplified with total RNAs purified from RKO cells with forward primer Pirh2-FF (5⬘-GGAGAATTCCACCATGGCGGCGACGGCCC GG-3⬘) and reverse primer Pirh2-FR (5⬘-GTACTCGAGTCATTGCTGATCCAG TGT-3⬘) and then cloned into a pcDNA4 expression vector (Invitrogen). To generate 2⫻ FLAG-tagged Pirh2, the cDNA fragment was amplified with Pirh2-FF1 (5⬘-GGATGGATCCATGGCGGCGACGGCCCGGGAAG-3⬘) and Pirh2-FR. Various Pirh2 mutants were generated by PCR with forward primer Pirh2-FF1 along with reverse primer Pirh2-137R (5⬘-ACACCTCGAGAATACACTTGTG TCTTCCTTGAAG-3⬘) for Pirh2(1–137), Pirh2-179R (5⬘-GTCTCTCGAGTTC TTTCAACATTTCTTCATAACACG-3⬘) for Pirh2(1–179), and Pirh2-186R (5⬘CAGACTCGAGACATAATGGACATCTGTAGCCTTC-3⬘) for Pirh2(1–186). Pirh2 with mutations at amino acids (aa) 137 to 261 [Pirh2(137–261)] was amplified with forward primer 5⬘-GAGGGGATCCATTGAAAATGTGTCCCGA CAGAATTG-3⬘ and reverse primer Pirh2-FR. Pirh2 which lacks the putative p53 binding domain (aa 120 to 137) [Pirh2(⌬120–137)] was produced by ligation of the fragment bearing aa 1 to 119 (amplified by forward primer Pirh2-FF1 and reverse primer Pirh2-119R [5⬘-TACAGATATCACAATGGAAAAAATCTTC CTTTGGACC-3⬘]) and the fragment bearing aa 138 to 261 (amplified by forward primer Pirh2-138F [5⬘-GGATATCGAAAATGTGTCCCGACAGAATTG TC-3⬘] and reverse primer Pirh2-FR). Pirh2(⌬145–186), which lacks the RING
* Corresponding author. Mailing address: Center for Comparative Oncology, 2128 Tupper Hall, Davis, CA 95616. Phone: (530) 754-8404. Fax: (530) 752-6042. E-mail:
[email protected]. 䌤 Published ahead of print on 14 December 2009. 1041
1042
JUNG ET AL.
finger domain (aa 145 to 186), was produced by ligation of the fragment bearing aa 1 to 144 (amplified by forward primer Pirh2-FF1 and reverse primer Pirh2144R [5⬘-CATATGATATCATTCTGTCGGGACACATTTTCAATACAC-3⬘]) and the fragment bearing aa 187 to 261 (amplified by forward primer Pirh2-187F [5⬘-CCAGATATCATGCACTCTGCTTTAGATATGACCAGG-3⬘] and reverse primer Pirh2-FR). Pirh2(⌬171–179) was produced by ligation of the fragment bearing aa 1 to 170 (amplified by forward primer Pirh2-FF1 and reverse primer Pirh2-170R [5⬘-TCTATGTAAAAGATGTCCACATGGCAAG-3⬘]) and the fragment encoding aa 180 to 261 (amplified by forward primer Pirh2-180F [5⬘-GGCTACAGATGTCCATTATGTATGCAC-3⬘] and reverse primer Pirh2FR). These products were cloned into a pcDNA3-2⫻ FLAG vector via BamHI and XhoI sites. To generate hemagglutinin (HA)-tagged PolH, cDNA fragment was amplified from pcDNA3-PolH (20) with forward primer PolH-FF (5⬘-GAATTCATGTA CCCATACGATGTTCCAGATTACGCTGCTACTGGACAGGATCGAGTG GTTG-3⬘) and reverse primer PolH-FR (5⬘-CTCGAGGGATCCCTAATGTG TTAATGGCTTAAAAAATG-3⬘). Various PolH mutants were amplified with forward primer PolH-FF1 (5⬘-CCGCGGATCCGCTACTGGACAGGATCGA GTGGTTGC-3⬘) along with reverse primer PolH-365R (5⬘-TCACCTCGAGC TGGGTGGCTACCCTGTCATTATTATC-3⬘) for PolH(1-365); with reverse primer PolH-505R (5⬘-GGGCTCGAGAGTGGGAGCAGTAAGAGATGATT G-3⬘) for PolH(1-505); and with PolH-635R (5⬘-CTTCTCGAGGGGCACTTG GTCCTCAGCAGCTAG-3⬘) for PolH(1-635). PolH(351–713) was amplified with forward primer PolH-351F (5⬘-CCCAGGATCCAGACTGACTAAAGAC CGAAATGATAATG-3⬘) and reverse primer PolH-FR. PolH(506–713) was amplified with forward primer PolH-506F (5⬘-GCTGGATCCCAGGCTCCCA TGAGCAATTCACCATC-3⬘) and reverse primer PolH-FR. PolH(594–713) was amplified with forward primer PolH-594F (5⬘-CAACTGGATCCGAGATGGATT TGGCCCACAACAGCCAAAG-3⬘) and reverse primer PolH-FR. PolH(⌬394– 505), which lacks the polymerase-associated domain (aa 394 to 505), was produced by ligation of the fragment bearing aa 1 to 393 (amplified with forward primer PolH-FF1 and reverse primer PolH-393R [5⬘-GGCTGATATCGTGAG CATCATAGCGGGTAAGGGCAC-3⬘]) and the fragment bearing aa 506 to 713 (amplified by forward primer PolH-506F [5⬘-CTGCTGATATCCAGGCTC CCATGAGCAATTCACC-3⬘] and reverse primer PolH-FR). These fragments were then cloned into a pcDNA3-2⫻ FLAG vector. Pirh2(C145S/C148S) and PoH(D652A) were generated by site-directed mutagenesis (QuikChange; Stratagene). Conserved cysteines in the Pirh2 RING domain were altered to serines to generate Pirh2(C145S/C148S) with sense primer 5⬘-CGACAGAATTCTCCAATATCTTTGGAGGACATTC-3⬘ and antisense primer 5⬘-GAATGTCCTCCAAAGATATTGGAGAATTCTGTCG-3⬘. PolH(D652A), which has an amino acid change at codon 652 from aspartate to alanine, was generated with sense primer 5⬘-GATATGCCAGAACACATGGC CTATCATTTTGCATTG-3⬘ and antisense primer 5⬘-CAATGCAAAATGATA GGCCATGTGTTCTGGCATATC-3⬘. To generate 2⫻ FLAG-tagged ubiquitin, cDNA fragment was amplified with forward primer 5⬘-CCGGGATCCATGCAGATTTTCGTGAAAACCCTTAC G-3⬘ and reverse primer 5⬘-GAAAGTCGACACCACCACGAAGTCTCAACA CAAG-3⬘, which was then cloned into a pcDNA3-2⫻ FLAG vector. To generate a construct that expresses short hairpin RNA (shRNA) against Pirh2 under the control of the tetracycline-regulated H1 promoter, one pair of oligonucleotides was synthesized, annealed, and then cloned into pBabe-H1 at HindIII and BglII sites (20). The resulting construct was designated pBabe-H1siPirh2. The sense oligonucleotide is 5⬘-GATCCCCCATGCCCAACAGACTT GTGTTCAAGAGACACAAGTCTGTTGGGCATGTTTTTA-3⬘ and the antisense oligonucleotide is 5⬘-AGCTTAAAAACATGCCCAACAGACTTGTGTC TCTTGAACACAAGTCTGTTGGGCATGGGG-3⬘ (the siRNA targeting regions are underlined). siRNA. Scramble and Pirh2 small interfering RNAs (siRNAs) were purchased from Dharmacon. Pirh2 siRNA sequences are 5⬘-CAUGCCCAACAGACUUG UG-dTdT-3⬘ (sense) and 5⬘-CACAAGUCUGUUGGGCAUG-dTdT-3⬘ (antisense). Cell culture. H1299 cell lines, which inducibly express Pirh2 or in which Pirh2 can be inducibly knocked down, were generated as reported previously (20). 20S proteasome degradation assay. In vitro 35S-labeled PolH or PolH(D652A) was incubated in a buffer (20 mM Tris-Cl [pH 7.2], 50 mM NaCl, 10 mM MgCl2, and 1 mM dithiothreitol [DTT]) supplemented with 1 g 20S proteasome (Affiniti) along with 80 M MG132 at 37°C for the indicated times. Cell proliferation assay. H1299 cells were seeded at 1 ⫻ 104 per well in triplicate in a six-well plate and then counted over the next few days. Colony formation assay. H1299 cells seeded at 500 per well in a six-well plate were cultured for 11 days and then stained with crystal violet.
MOL. CELL. BIOL.
FIG. 1. PolH is a protein with short half-life that is degraded via proteasome. (A) Extracts from RKO cells treated with CHX at the indicated times (0 to 3 h) were used to measure PolH, p53, and actin. (B) Calculated half-lives of p53 and PolH, using the data from panel A. The percentage of intensity in log10 was plotted versus time, and the t1/2 was calculated from the log10 of 50%. (C) PolH, p53, and actin were measured in nontreated RKO cells or RKO cells treated with 5 M MG132 or lactacystin for 0 to 12 h. (D) PolH, p53, and actin were measured in RKO cells 0 to 12 h following exposure to 15 J/m2 UV in the presence or absence of MG132 or lactacystin.
DNA histogram analysis. Both floating cells in the medium and live cells on the plates were collected 16 h following 15 J/m2 UV irradiation from normal and XPV cells, which were transfected with scramble or Pirh2 siRNA for 3 days. DNA histogram analysis was performed as previously described (20). Statistics. The significances were calculated by Student’s t test.
RESULTS Previously, we showed that PolH is transcriptionally regulated by DNA damage in a p53-dependent manner (20).
FIG. 2. PolH interacts with Pirh2. (A) Extracts from RKO and MCF7 cells were immunoprecipitated with anti-PolH or a control IgG, which was then used to detect PolH and Pirh2 along with whole-cell lysates as input control. (B) The experiment was performed as described for panel A except that anti-HA was used to immunoprecipitate HA-tagged PolH in RKO and MCF7 cells inducibly expressing HA-PolH. (C) The experiment was performed as in (A) except that anti-HA was used to immunoprecipitate HA-tagged PolH in RKO cells transiently transfected with HA-PolH and FLAG-Pirh2. (D) The experiment was performed as described for panel A except that anti-FLAG was used to immunoprecipitate FLAG-tagged Pirh2 in RKO cells transiently transfected with HA-PolH and FLAG-Pirh2. (E and G) Schematic presentation of PolH (E) and Pirh2 (G) domains and deletion constructs. The highly conserved DNA polymerase motifs (I to V) in PolH, the polymerase-associated domain (PAD) in PolH, the ubiquitin binding/zinc finger motif (UBZ) in PolH, the zinc finger in Pirh2, p53 binding domain (p53) in Pirh2, and the RING finger domain (RING) in Pirh2 are indicated. (F) RKO cells were transfected with FLAG-tagged wild-type or mutant PolH shown in panel E. Thirty-six hours after transfection, cells lysates were immunoprecipitated (IP) with anti-FLAG and then immunoblotted with the indicated antibodies. (H) RKO cells were transfected with HA-PolH along with wild-type or mutant FLAG-Pirh2 shown in panel G. Thirty-six hours after transfection, cell lysates were immunoprecipitated with anti-FLAG and immunoblotted with the indicated antibodies. ␣, anti; WCLs, whole-cell lysates. 1043
1044
JUNG ET AL.
Since PolH is required for DNA repair, it is likely that PolH is tightly controlled via multiple pathways. To test this, we measured the half-life of PolH protein and found that PolH protein was unstable, with a calculated half-life of ⬃28 min (Fig. 1A, lanes 1 to 6, and Fig. 1B). As a control, p53 had a calculated half-life of ⬃25 min, consistent with previous reports (25). We also found that PolH degradation was blocked by the MG132 proteasome inhibitor (Fig. 1A, lanes 7 to 12). To further test this, we measured the steady-state level of PolH and found that the level of PolH protein was increased by MG132 and proteasome inhibitor lactacystin in a time-dependent manner (Fig. 1C, compare lanes 1 to 5 with lanes 6 to 10 and 11 to 15, respectively). As a control, the levels of p53 were increased by these proteasome inhibitors. Since PolH is necessary for repairing UV-induced DNA damage (7), we measured the effect of UV irradiation on PolH stability and showed that the level of PolH protein was decreased by UV in a time-dependent manner, which was abrogated by proteasome inhibitors (Fig. 1D, compare lanes 1 to 5 with lanes 6 to 10 and 11 to 15, respectively). This is consistent with a recent study that showed that Caenorhabditis elegans PolH stability is decreased by UV (16). Since proteasomal degradation of PolH is enhanced by UV, we examined the Pirh2 E3 ligase, which is known to be induced by DNA damage in a p53-dependent manner (19). We showed that endogenous Pirh2 was coimmunoprecipitated with endogenous PolH, and exogenous PolH was coimmunoprecipitated with endogenous Pirh2 in RKO and MCF7 cells (Fig. 2A and B). Similarly, exogenous Pirh2 was coimmunoprecipitated with exogenous PolH, and exogenous PolH was also coimmunoprecipitated with exogenous Pirh2 in RKO cells (Fig. 2C and D). Next, we examined the region in PolH responsible for interaction with Pirh2 (Fig. 2E). We showed that the polymeraseassociated domain (aa 394 to 505) in PolH was required, since Pirh2 interacted with PolH(1–505), PolH(1–635), and PolH(351– 713), but not PolH(1–365), PolH(506–713), PolH(594–713), and PolH(⌬394–505) (Fig. 2F). Similarly, we examined the region in Pirh2 responsible for interaction with PolH (Fig. 2G). We showed that the RING finger domain (aa 145 to 186) in Pirh2 was required, since PolH interacted with Pirh2(1–179), Pirh2(1–186), and Pirh2(137–261), but not Pirh2(1–137), Pirh2(⌬145–186), and Pirh2(⌬171–179) (Fig. 2H). Interestingly, we showed that a small quantity of Pirh2(⌬120–137) was found to interact with PolH (Fig. 2H, lane 9). This may reflect the effect of the small deletion on the folding of Pirh2, as suggested by Sheng et al. (28). To investigate whether PolH stability is regulated by Pirh2, we generated H1299 cell lines in which endogenous Pirh2 can be inducibly knocked down. We showed that the level of PolH was increased in a manner inversely correlated with the level of Pirh2 (Fig. 3A). In addition, knockdown of Pirh2 abrogated UV-induced destabilization of PolH protein (Fig. 3B, compare lanes 1 to 5 with lanes 6 to 10, respectively). Conversely, we showed that overexpression of Pirh2 had an opposite effect on PolH stability in RKO cells transiently expressing PolH and Pirh2, as the levels of exogenous and total PolH were decreased in a manner directly correlated with the expression levels of Pirh2 (Fig. 3C). Moreover, overexpression of Pirh2 led to p53 destabilization (Fig. 3C), consistent with previous
MOL. CELL. BIOL.
FIG. 3. Knockdown of Pirh2 increases, whereas overexpression of Pirh2 decreases, PolH expression in a ubiquitin-independent manner. (A) Pirh2, PolH, p21, and actin were measured in H1299 cells induced to knock down Pirh2 for 0 to 72 h with their respective antibodies. (B) Pirh2, PolH, and actin were measured 0 to 12 h following exposure to 15 J/m2 UV in uninduced H1299 cells (control) or H1299 cells induced to knock down Pirh2 (Pirh2-KD) for 3 days. (C) Levels of total and exogenous PolH along with p53 and Pirh2 were measured in RKO cells transfected with HA-PolH along with various levels of FLAGPirh2 for 24 h. (D) ts20 cells were transiently transfected with FLAGPolH along with FLAG-Pirh2 at 35°C for 24 h and then incubated at 35°C or 39°C for an additional 24 h in the presence or absence of MG132 (5 M). The levels of FLAG-Pirh2 and FLAG-PolH were measured by anti-FLAG, and endogenous p53 and actin were measured by their respective antibodies. (E and F) Cells were treated as described for panel D. Cell extracts were collected 6 h following treatment with MG132 (5 M), immunoprecipitated with anti-FLAG or anti-p53 antibodies, and then immunoblotted with the indicated antibodies.
reports (19, 28). We would like to mention that p21 was accumulated by Pirh2-KD in p53 null H1299 cells (Fig. 3A). To determine whether ubiquitination is required for PolH degradation, Pirh2 and PolH were coexpressed in temperature-sensitive murine ts20 cells in which E1 ubiquitin-activating enzyme is not active at the restrictive temperature of 39°C (4). We showed that the level of endogenous wild-type
VOL. 30, 2010
DEGRADATION OF DNA POLYMERASE ETA BY Pirh2
1045
FIG. 4. Pirh2 targets PolH for 20S proteasomal degradation. (A) Extracts from RKO cells transfected with HA-PolH for 36 h were immunoprecipitated with anti-HA or a control IgG, which was then used to detect PolH and 19S/20S proteasomes along with whole-cell lysates as input control. (B) The experiment was performed as described for panel A, except that RKO cells were cotransfected with FLAG-Pirh2 and HA-PolH. (C) In vitro 35S-labeled PolH or PolH(D652A) was incubated with 1 g 20S proteasome in the presence or absence of 80 M MG132 for 30 to 90 min, and levels of 35S-labeled protein run in an SDS-PAGE analysis were detected by autoradiography. (D) Levels of Pirh2 and PolH were measured in RKO cells transfected with FLAG-Pirh2 along with FLAG-PolH or FLAG-PolH(D652A). (E and F) Extracts from RKO cells, which were transiently transfected with FLAG-PolH or FLAG-PolH(D652A), were immunoprecipitated with anti-FLAG or a control IgG, which was then used to detect Pirh2, PolH, and19S/20S proteasomes along with whole-cell lysates as input control. (G) Levels of PolH and Pirh2-DN were measured in RKO cells cotransfected with HA-PolH and various levels of FLAG-Pirh2-DN for 36 h.
p53 was increased at the restrictive temperature (39°C) (Fig. 3D, compare lane 1 with lane 3), consistent with the report that p53 is accumulated in ts20 cells (4). We also showed that Pirh2 was capable of inhibiting p53 expression at the permissive temperature (35°C), but much less efficient at 39°C (Fig. 3D, compare lanes 1 and 3 with lanes 2 and 4, respectively). However, PolH expression was inhibited by Pirh2 at both 35°C and 39°C (Fig. 3D, compare lanes 1 and 3 with lanes 2 and 4, respectively). In addition, the stability of PolH and p53 was restored upon treatment with MG132 (Fig. 3D, lanes 5 to 8). To further verify the effect of Pirh2 on the degradation of PolH, cells were treated as shown in Fig. 3D and subjected to immunoprecipitation with antiFLAG or anti-p53 antibodies. We found that p53 but not PolH was polyubiquitinated at 35°C (Fig. 3E and F). These results indicated that Pirh2 promotes PolH proteasomal turnover independently of ubiquitination. Based on the results above, PolH is degraded by proteasome in a ubiquitin-independent manner, suggesting that 20S proteasome plays a role. Indeed, we found that HA-tagged PolH physically associated with 20S proteasome but not 19S regula-
tory particle (Fig. 4A). We also found that the extent of PolH interaction with 20S proteasome was progressively increased as the level of Pirh2 was gradually increased (Fig. 4B). Nevertheless, a minute amount of 19S regulatory particle was found to associate with PolH when Pirh2 was highly expressed (Fig. 4B, lanes 5 to 6). Next, an in vitro protein degradation assay was performed and showed that PolH was degraded in a timedependent manner by 20S proteasome, which was inhibited by MG132, an inhibitor also known to repress 20S proteasome (Fig. 4C). Since PolH can be monoubiquitinated, it is possible that monoubiquitination facilitates PolH degradation. Thus, mutant PolH(D652A) was generated, which cannot be monoubiquitinated (2) or polyubiquitinated (data not shown). We found that like wild-type PolH, PolH(D652A) was degraded by 20S proteasome, which was inhibited by MG132 (Fig. 4C). We also showed that PolH(D652A) was decreased in a manner correlated with the expression levels of Pirh2 (Fig. 4D). Moreover, like wild-type PolH (Fig. 4E), PolH(D652A) was able to interact with 20S proteasome but not 19S particle (Fig. 4F). To further demonstrate that ubiquitination is not required, Pirh2DN, an E3 ligase-defective mutant (C145S/C148S) (28), was
1046
JUNG ET AL.
MOL. CELL. BIOL.
FIG. 5. Pirh2-KD cells are highly resistant to UV-induced cell death. (A) (Left panel) Control H1299 cells were grown in the absence (⫺tet) or presence (⫹tet) of tetracycline for 72 h, exposed to 15 J/m2 UV, and then cultured for 11 days. (Right panel) Effect of tetracycline along with UV exposure on colony formation, using the data from the left panel. The number of colonies for H1299 cells under the control condition was set at 100%. (B) The experiment was performed as described for panel A, except that tetracycline-inducible Pirh2-KD H1299 cells were not induced (control) or induced (Pirh2-KD) to knock down Pirh2. *, P ⬍ 0.04 versus control; **, P ⬍ 0.02 versus control. (C) The growth rates of control and tetracycline-inducible Pirh2-KD H1299 cells, which were grown in the absence or presence of tetracycline, were measured at day 5. *, P ⬍ 0.04 versus control; **, P ⬍ 0.02 versus control. (D) Levels of PolH, p53, and Pirh2 were measured in normal and 2XPV fibroblasts transfected with scramble or Pirh2 siRNA for 3 days. (E and F) DNA content was measured 16 h post-UV irradiation (15 J/m ) in normal and XPV fibroblasts transfected with scramble or Pirh2 siRNA for 3 days by the CellQuest program.
made and found to be capable of promoting PolH degradation in a dose-dependent manner when coexpressed with PolH (Fig. 4G). Cells deficient in PolH are hypersensitive to UV-induced cell death, due to accumulation of UV-induced DNA damage (7). Thus, a colony formation assay was performed to examine whether Pirh2 has an effect on cell survival via PolH in H1299 cells. We showed that tetracycline alone had no effect on cell survival regardless of UV irradiation (Fig. 5A). However, in the absence of UV irradiation, knockdown of Pirh2 decreased the colony-forming ability of the H1299 cells (Fig. 5B), which is probably due to increased levels of p21 (Fig. 3A). Interestingly, upon exposure to UV, cell survival was markedly increased when Pirh2 was knocked down (Fig. 5B). Next, a short-term cell proliferation assay was performed. We showed that tetracycline alone had no effect (Fig. 5C, left panel). However, knockdown of Pirh2 desensitized cells to UV irradiation, whereas in the absence of UV irradiation, knockdown of Pirh2 decreased cell proliferation (Fig. 5C, right panel). Therefore, Pirh2-KD had similar effects on cell proliferation (Fig. 5C) and survival (Fig. 5A and B). To demonstrate the role of PolH,
Pirh2 was transiently knocked down in normal and XPV fibroblasts (Fig. 5D). Consistent with the observation in H1299 cells (Fig. 3A), Pirh2 knockdown led to a marked increase in PolH in normal, but not in XPV, fibroblasts (Fig. 5D, compare lanes 1 and 3 with lanes 2 and 4, respectively). We also showed that upon knockdown of Pirh2, p53 is accumulated in both normal and XPV cells (Fig. 5D). Next, DNA histogram analysis was performed and showed that upon knockdown of Pirh2, UVinduced apoptosis was significantly decreased in normal fibroblasts (13.27% versus 8.18%) (Fig. 5E). This is likely due to the possibility that an increased level of PolH would quickly eliminate UV-induced lesions, which would then decrease UVinduced cell killing by reducing the duration and extent of accumulation of proapoptotic proteins, such as p53. However, in XPV cells, UV-induced apoptosis was not decreased, but instead increased, by knockdown of Pirh2 (16.45% versus 22.23%) (Fig. 5F). This is likely due to the possibility that XPV cells (without PolH) are deficient in repairing UV-induced lesions, which would prolong accumulation of proapoptotic proteins negatively regulated by Pirh2, such as p53, and subsequently increase UV-induced cell killing.
VOL. 30, 2010
DEGRADATION OF DNA POLYMERASE ETA BY Pirh2
DISCUSSION Here, we present evidence that upon UV irradiation, the stability of PolH protein is compromised via Pirh2 in a ubiquitin-independent manner. We also showed that upon knockdown of Pirh2, PolH is accumulated and, consequently, desensitizes cells to UV-induced cell killing. Based on these observations, we postulate that UV irradiation promotes cancer formation in part by destabilizing PolH via Pirh2-mediated 20S proteasome degradation. Pirh2 targets p53 for ubiquitin-dependent degradation (19). Here, we showed that Pirh2 recruits 20S proteasome to degrade PolH. Thus, Pirh2 has properties similar to those of Mdm2. It is well defined that Mdm2 targets p53 for degradation via a ubiquitin-dependent pathway (9, 12, 17, 31). In addition, Mdm2 interacts with, and degrades, p21 and RB through 20S proteasome in a ubiquitin-independent manner (13, 27, 32). Furthermore, Mdmx, which does not have ubiquitin E3 ligase activity, acts as an adaptor of Mdm2 to degrade p21 in a ubiquitin-independent manner (13). Therefore, future studies are warranted to examine the possibility that PolH may be regulated by Mdm2 and/or Mdmx. Although the level of PolH was decreased upon UV irradiation, the level of Pirh2 was not substantially altered in H1299 cells (Fig. 3B). Thus, the question is whether and how Pirh2 is regulated upon UV irradiation. In cells with endogenous wildtype p53, it is likely that the activation of p53 upon UV irradiation would lead to increased Pirh2 expression, since Pirh2 is a well-defined p53 target gene (19). However, in cells with no functional p53, such as the H1299 cells used in Fig. 3B, other mechanisms are likely to regulate Pirh2 activity. Previous studies showed that Pirh2 can be phosphorylated in a calmodulindependent manner and that phosphorylated Pirh2 is deficient in polyubiquitinating and degrading p53 (5). Since UV irradiation leads to activation of various DNA damage-induced kinases, it is possible that the activity of Pirh2 to target PolH may be increased by phosphorylation. Additionally, Pirh2 exists as a homodimer or a heterodimer with PLAGL2 (pleomorphic adenoma gene-like 2) (33). Thus, it is possible that UV irradiation may alter the partner of Pirh2, which then enhances Pirh2 to target PolH. Although Pirh2 promotes PolH degradation independent of ubiquitination, it remains possible that Pirh2 may still ubiquitinate PolH, albeit not polyubiquitinate it. This is significant, especially considering that PolH is known to be monoubiquitinated and that monoubiquitination modulates PolH functions in repairing UV-induced DNA damage and homologous recombination repair (2, 26). For example, PolH(D652A), which has a mutation in the UBZ (ubiquitin binding zinc finger) domain, cannot be monoubiquitinated and is incapable of interacting with monoubiquitin (2). Thus, future studies are necessary to determine whether Pirh2 and other DNA damage-inducible E3 ligases catalyze PolH monoubiquitination. We showed that upon knockdown of Pirh2, p21 expression was increased in H1299 cells (Fig. 3), which might be responsible for decreased cell proliferation and survival (Fig. 5A to C). Like p53 and PolH, p21 is a short-half-life protein and subject to proteasome-dependent degradation. Several ubiquitin E3 ligases are known to target p21 for degradation, including CRL4Cdt2 and CRL4Ddb2, which polyubiquitinate p21
1047
for 26S proteasomal degradation (1, 29), and SCFSkp2 and Mdm2, which target p21 for 20S proteasomal degradation (13, 18, 32). Thus, whether p21 is directly regulated by Pirh2 is worth further investigation. ACKNOWLEDGMENTS This work is supported in part by NIH grants (CA076069 and CA081237). We thank Jim Xiao and H. Ozer for providing ts20 cells. REFERENCES 1. Abbas, T., U. Sivaprasad, K. Terai, V. Amador, M. Pagano, and A. Dutta. 2008. PCNA-dependent regulation of p21 ubiquitylation and degradation via the CRL4Cdt2 ubiquitin ligase complex. Genes Dev. 22:2496–2506. 2. Bienko, M., C. M. Green, N. Crosetto, F. Rudolf, G. Zapart, B. Coull, P. Kannouche, G. Wider, M. Peter, A. R. Lehmann, K. Hofmann, and I. Dikic. 2005. Ubiquitin-binding domains in Y-family polymerases regulate translesion synthesis. Science 310:1821–1824. 3. Chau, V., J. W. Tobias, A. Bachmair, D. Marriott, D. J. Ecker, D. K. Gonda, and A. Varshavsky. 1989. A multiubiquitin chain is confined to specific lysine in a targeted short-lived protein. Science 243:1576–1583. 4. Chowdary, D. R., J. J. Dermody, K. K. Jha, and H. L. Ozer. 1994. Accumulation of p53 in a mutant cell line defective in the ubiquitin pathway. Mol. Cell. Biol. 14:1997–2003. 5. Duan, S., Z. Yao, D. Hou, Z. Wu, W. G. Zhu, and M. Wu. 2007. Phosphorylation of Pirh2 by calmodulin-dependent kinase II impairs its ability to ubiquitinate p53. EMBO J. 26:3062–3074. 6. Goldberg, A. L. 2003. Protein degradation and protection against misfolded or damaged proteins. Nature 426:895–899. 7. Guo, C., J. N. Kosarek-Stancel, T. S. Tang, and E. C. Friedberg. 2009. Y-family DNA polymerases in mammalian cells. Cell. Mol. Life Sci. 66:2363– 2381. 8. Hattori, T., T. Isobe, K. Abe, H. Kikuchi, K. Kitagawa, T. Oda, C. Uchida, and M. Kitagawa. 2007. Pirh2 promotes ubiquitin-dependent degradation of the cyclin-dependent kinase inhibitor p27Kip1. Cancer Res. 67:10789–10795. 9. Haupt, Y., R. Maya, A. Kazaz, and M. Oren. 1997. Mdm2 promotes the rapid degradation of p53. Nature 387:296–299. 10. Hershko, A. 1996. Lessons from the discovery of the ubiquitin system. Trends Biochem. Sci. 21:445–449. 11. Hershko, A., and A. Ciechanover. 1998. The ubiquitin system. Annu. Rev. Biochem. 67:425–479. 12. Honda, R., H. Tanaka, and H. Yasuda. 1997. Oncoprotein MDM2 is a ubiquitin ligase E3 for tumor suppressor p53. FEBS Lett. 420:25–27. 13. Jin, Y., H. Lee, S. X. Zeng, M. S. Dai, and H. Lu. 2003. MDM2 promotes p21waf1/cip1 proteasomal turnover independently of ubiquitylation. EMBO J. 22:6365–6377. 14. Kannouche, P., and A. Stary. 2003. Xeroderma pigmentosum variant and error-prone DNA polymerases. Biochimie 85:1123–1132. 15. Kawamoto, T., K. Araki, E. Sonoda, Y. M. Yamashita, K. Harada, K. Kikuchi, C. Masutani, F. Hanaoka, K. Nozaki, N. Hashimoto, and S. Takeda. 2005. Dual roles for DNA polymerase eta in homologous DNA recombination and translesion DNA synthesis. Mol. Cell 20:793–799. 16. Kim, S. H., and W. M. Michael. 2008. Regulated proteolysis of DNA polymerase eta during the DNA-damage response in C. elegans. Mol. Cell 32:757–766. 17. Kruse, J. P., and W. Gu. 2009. Modes of p53 regulation. Cell 137:609–622. 18. Lee, H., S. X. Zeng, and H. Lu. 2006. UV induces p21 rapid turnover independently of ubiquitin and Skp2. J. Biol. Chem. 281:26876–26883. 19. Leng, R. P., Y. Lin, W. Ma, H. Wu, B. Lemmers, S. Chung, J. M. Parant, G. Lozano, R. Hakem, and S. Benchimol. 2003. Pirh2, a p53-induced ubiquitinprotein ligase, promotes p53 degradation. Cell 112:779–791. 20. Liu, G., and X. Chen. 2006. DNA polymerase , the product of the xeroderma pigmentosum variant gene and a target of p53, modulates the DNA damage checkpoint and p53 activation. Mol. Cell. Biol. 26:1398–1413. 21. Logan, I. R., V. Sapountzi, L. Gaughan, D. E. Neal, and C. N. Robson. 2004. Control of human PIRH2 protein stability: involvement of TIP60 and the proteosome. J. Biol. Chem. 279:11696–11704. 22. Masutani, C., R. Kusumoto, A. Yamada, N. Dohmae, M. Yokoi, M. Yuasa, M. Araki, S. Iwai, K. Takio, and F. Hanaoka. 1999. The XPV (xeroderma pigmentosum variant) gene encodes human DNA polymerase eta. Nature 399:700–704. 23. McIlwraith, M. J., A. Vaisman, Y. Liu, E. Fanning, R. Woodgate, and S. C. West. 2005. Human DNA polymerase eta promotes DNA synthesis from strand invasion intermediates of homologous recombination. Mol. Cell 20: 783–792. 24. Murata, S., H. Yashiroda, and K. Tanaka. 2009. Molecular mechanisms of proteasome assembly. Nat. Rev. Mol. Cell Biol. 10:104–115. 25. Oren, M., W. Maltzman, and A. J. Levine. 1981. Post-translational regulation
1048
26.
27.
28.
29.
JUNG ET AL.
of the 54K cellular tumor antigen in normal and transformed cells. Mol. Cell. Biol. 1:101–110. Plosky, B. S., A. E. Vidal, A. R. Fernandez de Henestrosa, M. P. McLenigan, J. P. McDonald, S. Mead, and R. Woodgate. 2006. Controlling the subcellular localization of DNA polymerases iota and eta via interactions with ubiquitin. EMBO J. 25:2847–2855. Sdek, P., H. Ying, D. L. Chang, W. Qiu, H. Zheng, R. Touitou, M. J. Allday, and Z. X. Xiao. 2005. MDM2 promotes proteasome-dependent ubiquitinindependent degradation of retinoblastoma protein. Mol. Cell 20:699–708. Sheng, Y., R. C. Laister, A. Lemak, B. Wu, E. Tai, S. Duan, J. Lukin, M. Sunnerhagen, S. Srisailam, M. Karra, S. Benchimol, and C. H. Arrowsmith. 2008. Molecular basis of Pirh2-mediated p53 ubiquitylation. Nat. Struct. Mol. Biol. 15:1334–1342. Stoyanova, T., T. Yoon, D. Kopanja, M. B. Mokyr, and P. Raychaudhuri.
MOL. CELL. BIOL.
30.
31. 32.
33.
2008. The xeroderma pigmentosum group E gene product DDB2 activates nucleotide excision repair by regulating the level of p21Waf1/Cip1. Mol. Cell. Biol. 28:177–187. Voges, D., P. Zwickl, and W. Baumeister. 1999. The 26S proteasome: a molecular machine designed for controlled proteolysis. Annu. Rev. Biochem. 68:1015–1068. Vousden, K. H., and C. Prives. 2009. Blinded by the light: the growing complexity of p53. Cell 137:413–431. Zhang, Z., H. Wang, M. Li, S. Agrawal, X. Chen, and R. Zhang. 2004. MDM2 is a negative regulator of p21WAF1/CIP1, independent of p53. J. Biol. Chem. 279:16000–16006. Zheng, G., J. Ning, and Y. C. Yang. 2007. PLAGL2 controls the stability of Pirh2, an E3 ubiquitin ligase for p53. Biochem. Biophys. Res. Commun. 364:344–350.
