RB·E2F1 Complex Mediates DNA Damage Responses through Transcriptional Regulation of ZBRK1*

RB plays an essential role in DNA damage-induced growth arrest and regulates the expression of several factors essential for DNA repair machinery. However, how RB coordinates DNA damage response through transcriptional regulation of genes involved in growth arrest remains largely unexplored. We examined whether RB can mediate the response to DNA damage through modulation of ZBRK1, a zinc finger-containing transcriptional repressor that can modulate the expression of GADD45A, a DNA damage response gene, to induce cell cycle arrest in response to DNA damage. We found that the ZBRK1 promoter contains an authentic E2F-recognition sequence that specifically binds E2F1, but not E2F4 or E2F6, together with chromatin remodeling proteins CtIP and CtBP to form a repression complex that suppresses ZBRK1 transcription. Furthermore, loss of RB-mediated transcriptional repression led to an increase in ZBRK1 transcript levels, correlating with increased sensitivity to ultraviolet (UV) and methyl methanesulfonate-induced DNA damage. Taken together, these results suggest that the RB·CtIP (CtBP interacting protein)/CtBP (C terminus-binding protein) /E2F1 complex plays a critical role in ZBRK1 transcriptional repression, and loss of this repression may contribute to cellular sensitivity of DNA damage, ultimately leading to carcinogenesis.

The DNA damage response that responds to genotoxic stress induced by radiation, chemicals, and endogenous reactive oxygen species is highly conserved in higher eukaryotes (1,2). The cellular responses to DNA damage include activation of cell-cycle arrest, apoptosis, and DNA damage repair (3). In mammals, multiple partially overlapping DNA repair mechanisms, including base excision, recombination, and mismatch repair, are required for proper DNA damage repair. Defects in these repair pathways frequently lead to irreparable DNA damage, aging, and cancer.
RB, a prototypic tumor suppressor (4), is essential for regulating cell cycle progression through interaction with binding partners such as E2F (5). In quiescent and early G 1 cells, RB associates with the E2F family of transcription factors to repress the expression of E2F-responsive genes involved in cell cycle progression (6). As cells progress toward S-phase, RB is phosphorylated by cyclin-dependent kinases, releasing E2F, which opens the DNA replication origin, and induces transcription of S-phase genes (7). When RB is lost or inactivated, DNA replication origins become readily accessible, resulting in uncontrolled transcription of S-phase genes and ultimately leading to premature S-phase progression (8). RB also plays an essential role in DNA damage-induced growth arrest (9) and transcriptional regulation of several DNA damage-repair factors involved in the ultraviolet (UV) damage repair process, including FEN1, XPC, RPA2-3, RFC4, and proliferating cell nuclear antigen (10). Although previous studies have established that phosphorylation of RB inactivates its binding activity with its interacting partners and promotes cell cycle progression (11); a recent study, however, indicated that the phosphorylation of RB at Ser-612 represents an activating event in response to DNA damage, suggesting a novel role of RB in DNA damage response (12).
Of all the E2F family members, E2F1 appears to be unique in its ability to induce both proliferation and apoptosis (13). In addition, it displays properties of both an oncogene and a tumor suppressor (14 -18). Recent studies have revealed at least two mechanisms of E2F/RB-mediated gene repression: the recruitment of a histone deacetylase (19,20) or the CtBP corepressor complex (21,22). Although the expressions of many cell cycle genes are subject to RB/E2F repression (23), it remains unclear whether any of these genes participate in RB-mediated DNA damage response.
ZBRK1 was identified as a transcriptional repressor that recognizes a consensus DNA element, GGGXXXCAGXXXTTT (24). Interestingly, ZBRK1 exhibits two different modes of repression: its C-terminal domain binds the BRCA1⅐CtIP (CtBP-interacting protein) co-repressor complex (25), whereas the N-terminal KRAB domain binds to the KAP1 co-repressor (26). ZBRK1 thus represents the first KRAP zinc finger protein (KRAB-ZFP) harboring two independent repression domains. ZBRK1 association with BRCA1 and CtIP repress transcription of Angiopoietin 1 (27) and high mobility group AT-hook 2 (28). The same repressor complex can also regulate the DNA damage response gene GADD45A for the maintenance of genomic integrity (29 -31). ZBRK1 is rapidly degraded upon DNA damage through a ubiquitin-proteasome pathway (32), suggesting that the genes repressed by ZBRK1 may be activated during DNA damage. This further substantiates the role of ZBRK1 in DNA damage response.
In this communication, we show that the ZBRK1 promoter contains an authentic E2F1 binding site that serves as the platform for the RB⅐E2F1⅐CtIP (CtBP interacting protein)⅐CtBP (C terminus-binding protein) complex-mediated transcriptional repression of ZBRK1. This regulation is intimately correlated with cellular sensitivity toward UV-and methyl methanesulfonate (MMS) 3 -induced DNA damage. These results provide a potential mechanism to elucidate how RB deficiency contributes to an elevated sensitivity to DNA damage.

EXPERIMENTAL PROCEDURES
Materials-Lipofectamine 2000, DMEM, Opti-MEM medium, FBS, TRIzol RNA extraction reagent, and SuperScript TM III were purchased from Invitrogen. Super-Therm DNA polymerase was purchased from Bertec Enterprise CO., Ltd. (Taipei, Taiwan). Oligonucleotides for RT-PCR were synthesized by MDBio Inc. (Taipei, Taiwan). Oligonucleotides for DNA Affinity Precipitation Assay (DAPA) and siRNA assay were synthesized by Invitrogen. TaqDNA polymerases and the luciferase assay system were from Promega (Madison, WI). The yTA cloning vector kit was purchased from Yeastern Biotech Co. (Taipei, Taiwan). The expression plasmid pcDNA3/HA was a gift from Dr. Hsin-Fang Yang-Yen (IMB, Academia Sinica, Taiwan). The E2F1 expression plasmid was a gift from Dr. Pan-Chyr Yang (College of Medicine, National Taiwan University, Taiwan). The E2F6 cDNA was purchased from ATCC (number MGC-15924). The CCND1 reporter (CCND1-p) containing the 3.3-kb cyclin D1 promoter was a gift from Richard Pestell (Georgetown University) (33). Antibodies against E2F1 and CtBP were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against Rb (Rb245) and CtIP (19E8) were purchased from GeneTex (San Antonio, TX). The detection of endogenous ZBRK1 was performed as described in Zheng et al. (25). Antibody against ␤-actin was purchased from Sigma.
Cell Culture and Treatment with DNA Damage Agents-MCF7 and MDA-MB-468 (MB-468) breast cancer cells were cultured in high glucose Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 mg/ml of streptomycin, and 100 units/ml of penicillin. MCF10A cells were cultured as described in Debnath et al. (34). DNA-damage induction was performed using the UV-Crosslinker (Spectroline) at 40 J/m2 or 0.05% MMS (Sigma). To measure the half-life of exogenous GFP-ZBRK1 and endogenous ZBRK1 protein, cells were treated with cycloheximide (50 g/ml) alone or combined with either UV or MMS. Protein lysates were harvested at various time points for Western blot analysis.
Western Blot and DAPA-Cells were lysed in modified RIPA buffer (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1% Nonidet P-40, 0.25% sodium deoxycholate, 1 mM DTT, 10 mM NaF, 1 mM PMSF, 1 g/ml of aprotinin, 1 g/ml of leupeptin, and 1 mM Na 3 VO 4 ). Nuclear extracts were prepared from MCF7 cells for DAPA. DAPA was performed with a slight modification from the procedures detailed in Zhu et al. (35). Briefly, 200 g of nuclear extract and 2 g of biotinylated E2F motif of the ZBRK1 promoter were incubated at 4°C for 1 h, before pulldown by streptavidin-agarose beads. The precipitated products were washed three times with cold PBS and eluted for separation by SDS-PAGE. The sequence of the biotinylated E2F motif of ZBRK1 promoter was 5Ј-GGCCCTCTTTTGGCCCCTCATATG-3Ј.
Chromatin Immunoprecipitation (ChIP) and Re-ChIP Assay-ChIP assay was performed essentially as described by Wang et al. (36). Briefly, the sheared chromatin fragments were immunoprecipitated with antibodies specific to E2F1, RB, CtIP, CtBP, or control mouse IgG at 4°C overnight. The DNA was dissociated from the immunoprecipitated chromatin prior to PCR amplification. For the re-ChIP assay, the first immune complex was washed twice with wash buffer (50 mM Tris-HCl, pH 8.0, 0.1% SDS, 0.5% Nonidet P-40, 150 mM NaCl, and 2.5 mM EDTA), twice with low-salt buffer (10 mM Tris-HCl, pH 8.0, and 0.1 mM EDTA), resolved in 10 mM DTT at 37°C, diluted in ChIP dilution buffer, and processed according to the same ChIP assay protocol using the indicated antibodies. The primers used for PCR amplification were as follows: ZBRK1p/Ϫ419, 5Ј-CATT-CTCTGGGATACCTACACCTG-3Ј; ZBRK1p/Ϫ107, 5Ј-CTG-TAAAATGGCCGCGCCCATAGC-3Ј; and GL2 primer, 5Ј-CTTTATGTTTTTGGCGTCTTCCA-3Ј. The amplified DNA products were resolved by agarose gel electrophoresis and confirmed by sequencing.
Comet Assay-Cells were embedded in agarose gel and spread on a frosted microscope slide before lysing in lysis buffer (2.5 M NaCl, 10 mM Tris-HCl, pH 7.4, 100 mM EDTA, and 1% Triton X-100) for 1 h at 4°C. After lysis, cells were preincubated for 20 min at 4°C in electrophoresis buffer (0.3 M NaOH and 1 mM EDTA, pH Ͼ 13) before alkaline gel electrophoresis at 24 V/300 mA for 15 min. Following electrophoresis, the slides were rinsed with neutralization buffer and allowed to dry before staining with ethidium bromide for visualization under fluorescence microscopy. The tail moment was quantified by a computerized image analysis program. The tail moment was calculated by multiplying the total intensity of the Comet tail by the migration distance from the center of the Comet head.
Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling (TUNEL) Assay-For the TUNEL assay, the Apoptosis Detection System Fluorescein (Promega) was used. Cells grown on coverslips were treated with UV or MMS and allowed to recover for 2 h at 37°C. Cells were then washed with PBS and fixed in 3.7% formaldehyde for 10 min before storage at Ϫ20°C in 70% ethanol for at least 30 min. The coverslips were then washed in PBS for 5 min at room temperature, treated with 0.2% Triton X-100 in PBS for 5 min, and rinsed twice with PBS for 5 min at room temperature. The percentage of red fluorescing cells (TUNEL positive) was determined using images captured by fluorescence microscope.

Identification of the E2F-recognition Sequences in the ZBRK1
Promoter as a Negative Transcriptional Response Element-RB plays a critical role in the response to DNA damage. However, the precise mechanism by which RB mediates the DNA damage response remains unclear. Because ZBRK1 serves as a repressor regulating DNA damage response genes (25), we hypothesized that RB may act as an up-stream regulator of ZBRK1 for this purpose. To address this possibility, we examined the correlation between the expression of RB protein and ZBRK1 transcript in RB-depleted cells. As shown in Fig. 1, when RB was depleted by siRNA, the ZBRK1 transcript level was increased, suggesting that RB acts as a negative regulator of ZBRK1 transcription. Because RB is known to serve as co-repressor with E2F1 for the regulation of several genes, the ZBRK1 promoter sequence was analyzed using the TFSEARCH website, revealing a putative E2F binding motif (supplemental Fig. S1). To validate whether this putative E2F motif is biologically functional, we examined the promoter activity of ZBRK1 in cells transfected with reporter constructs containing the 5Ј-flanking regions of the human ZBRK1 gene, Ϫ1113/ϩ47 (ZA), Ϫ624/ϩ47 (ZB), or their corresponding E2F motif mutants (ZAmE and ZBmE) ( Fig. 2A). As shown in Fig. 2B, ZAmE and ZBmE induced a 6-and 14-fold induction, respectively, in luciferase activity when compared with wildtype constructs ZA and ZB. It was noted that a higher luciferase activity was observed in the ZBmE mutant construct with a shorter promoter region when compared with the longer ZAmE mutant promoter (Fig. 2B), implicating the presence of a potential negative regulatory element within the Ϫ1113 to Ϫ624-bp region flanking the ZBRK1 promoter. Taken together, these results suggest that the E2F recognition site identified on the ZBRK1 promoter is biologically active.
E2F1, but Not Other E2F Family Members, Specifically Down-regulate ZBRK1 Transcription-RB interacts with several E2F family members (37). To test which E2F family member is involved in this E2F motif-mediated repression of the ZBRK1 gene, we transfected cells with various E2F expression vectors, including E2F1, E2F4, and E2F6, and examined the expression of ZBRK1 by RT-PCR and Western blot. E2F1, but not E2F4 or E2F6, showed specific inhibition of ZBRK1 expression at the RNA and protein levels (Fig. 3A) although the expressions of E2F1, E2F4, and E2F6 were comparable (supplemental Fig. S2). To further test whether this repression is promoter-mediated, a reporter construct containing the ZBRK1 promoter was co-transfected into cells with E2F1, E2F4, or E2F6 expression vectors. We found that only E2F1 specifically repressed the ZBRK1 promoter activity (Fig. 3B). To confirm the authenticity of E2F1mediated ZBRK1 repression, we examined ZBRK1 promoter activity in comparison to MCL-1, CCND1, and AURKA promoters. MCL-1 was previously reported to be negatively regulated by E2F1 (38), whereas CCND1 and AURKA were reported to be positively regulated by E2F1 (39,40). Consistent with previous reports, we also found MCL-1 to be repressed, whereas CCDN1 and AURKA were activated upon E2F1 expression. Using these controls, ZBRK1 promoter activity was demonstrated to be repressed upon E2F1 expression (Fig. 3C). Consistently, depletion of E2F1 by siRNA resulted in an increase in ZBRK1 expression at both the RNA and protein levels (Fig.  3D). Taken together, these results support the notion that E2F1 specifically represses ZBRK1 transcription.

RB⅐E2F1 Mediates DNA Damage through ZBRK1
OCTOBER 22, 2010 • VOLUME 285 • NUMBER 43 RB Acts as a Negative Regulator of the ZBRK1 Promoter through the E2F Motif-RB plays an important role in E2F1mediated transcriptional repression of several genes, such as cdc25A (41) and cyclin E (42). To investigate whether RB coordinates with E2F1 in repressing ZBRK1 transcription, we performed a ZBRK1 reporter assay in cells with RB depleted by siRNA. As shown in Fig. 4A, depletion of RB enhanced ZBRK1 promoter activity (compare lanes 2 and 3 with lane 1). This was also the case in cells transfected with E2F1 (Fig. 4A, compare  lanes 5 and 6 with lane 4), suggesting that depletion of RB reverses the E2F1-mediated repression of ZBRK1. On the contrary, promoter activity of the ZBmE reporter containing the mutant E2F motif was unaffected by either RB depletion or E2F1 overexpression (Fig. 4A, compare lane 7 with 12). To confirm this finding, we transfected the ZB and ZBmE reporters into RB-negative Saos-2 and RB-reconstituted SR5 cells (43) and found that the presence of RB in SR5 cells negatively affected ZB reporter activity, but not the activity of ZBmE (Fig.  4B). These results suggest that RB coordinates with E2F1 to regulate the ZBRK1 promoter through the E2F motif.

RB⅐E2F1 Recruits the CtIP⅐CtBP Co-repressor Complex to the E2F1 Site of ZBRK1 Promoter-Because
RB is known to bind to the CtIP⅐CtBP co-repressor complex to repress transcription (21), it is likely that this repressor complex may participate in RB⅐E2F1-mediated ZBRK1 gene repression. To investigate this possibility, we used two CtIP mutants: one with CtBP interaction disrupted by the deletion of PLDLS sequence (44) and one with RB interaction disrupted by an N-terminal mutation from LXCXE to LXCXK (Fig. 5A) (21). Ectopic expression of wildtype (WT) CtIP, but not the mutants, repressed ZBRK1 reporter activity (Fig. 5B). To test whether E2F1, RB, CtIP, and CtBP bind to the E2F motif of the ZBRK1 promoter, we performed a DNA precipitation assay using a biotinylated E2F DNA motif and found that E2F1, RB, CtIP, and CtBP all bind to the E2F motif (Fig. 5C). Furthermore, we performed a ChIP assay to test in vivo binding of the E2F1⅐RB⅐CtIP⅐CtBP complex to the ZBRK1 promoter and found that increased binding can be detected upon E2F1 increase (Fig. 5D). To further validate that RB, CtIP, CtBP, and E2F1 exist as one multiprotein repressor complex on the ZBRK1 promoter, a Re-ChIP assay was performed. Immunoprecipitation was first performed with E2F1 antibodies followed by a second immunoprecipitation with RB, CtIP, or CtBP antibodies. The result showed that RB, CtIP, CtBP, and E2F1 formed a complex on the ZBRK1 promoter and binding was enhanced in cells with E2F1 overexpression, suggesting that E2F1 may participate in the recruitment of the RB⅐CtIP⅐CtBP complex to the ZBRK1 promoter (Fig. 5E). To demonstrate the importance of the E2F motif as the platform for the recruitment of RB⅐CtIP⅐CtBP complex, we performed a ChIP assay to examine the presence of the complex on exogenously expressed wild-type and ZBmE promoters. Only the promoter containing the wild-type E2F1 site bound the RB⅐CtIP⅐CtBP complex (Fig. 5F). Similar results were obtained in U2OS cells (supplemental Fig. S3). Taken together, the E2F motif of the ZBRK1 promoter is a platform for the E2F1⅐RB⅐CtIP⅐CtBP complex-mediated transcriptional repression of ZBRK1.
ZBRK1 and GADD45 Expression Profiles in Cells upon UV and MMS Treatment-Because both RB and ZBRK1 participate in DNA damage response, we next examined the relationship between ZBRK1 and RB expression profiles upon treatment with DNA damage inducers. We compared the mRNA and protein levels of ZBRK1 and its downstream target GADD45A in RB-proficient MCF7 and RB-deficient MDA-MB-468 (MB-468) cells. ZBRK1 expression was found to be higher in MB-468 cells than MCF7 cells, and GADD45A expression was inversely correlated with ZBRK1 expression patterns. When MCF7 and MB-468 cells were treated with DNA damage inducers (UV and MMS), expression of ␥-H2AX and GADD45A correlated with a reduction in ZBRK1 expression (Fig. 6, A and B). These results support the notion that upon UV and MMS treatment, RB forms a complex with E2F1 to repress ZBRK1 expression, in turn to derepress the expression of GADD45A. In the absence of RB, the repression of ZBRK1 was lost and the expression of GADD45A was compromised upon DNA damage, likely leading to improper resolution of DNA damage. To confirm that DNA damage inducers lead to E2F1/RB-mediated repression of ZBRK1 via binding to the ZBRK1 promoter, a luciferase reporter assay was performed. As shown in Fig. 6C, a decrease in ZBRK1 promoter activity was observed upon UV and MMS treatments in an E2F1-and RBdependent manner. In addition, a ChIP assay was performed to detect the binding of the E2F1⅐RB⅐CtIP⅐CtBP complex at the ZBRK1 promoter upon DNA damage. As shown in Fig.  6D, UV and MMS treatments induced binding of the E2F1⅐RB⅐CtIP⅐CtBP complex at the ZBRK1 promoter. Together, these results suggest that the E2F1⅐RB⅐CtIP⅐ CtBP complex negatively regu lates ZBRK1 expression in re sponse to DNA damage.

Depletion of ZBRK1 Protects Cells from UV-and MMS-induced DNA
Damage-Based on the observations that the loss of RB results in an accumulation of DNA damage (45) and the induction of ZBRK1 expression (this study), we proceeded to examine whether accumulation of DNA damage responses is a consequence of ZBRK1 expression. The Comet assay was performed to detect the level of DNA damage as a function of ZBRK1 expression in MB-468 cells. Before treatment with DNA damage inducers, cells were pretreated with lentiviral shZBRK1 for 24 h. Inhibition of ZBRK1 expression attenuated the level of DNA damage after treatment with UV or MMS (Fig. 7A). In addition, to verify whether inactivation of ZBRK1 reduces UV-and MMS-induced apoptosis, a TUNEL assay was performed using the described cell line. As shown in Fig.  7B, inactivation of ZBRK1 protects cells from UV-and MMS-induced apoptosis. These results suggest that ZBRK1 depletion can protect cells against UV-and MMS-induced DNA damage and cell death.

DISCUSSION
In this communication, we addressed how RB coordinates the DNA damage response through transcriptional regulation of genes involved in growth arrest. We found that the ZBRK1 promoter contains an authentic E2F-recognition sequence, which specifically binds E2F1, but not E2F4 or E2F6 (Fig. 3). RB, together with E2F1 and chromatin modifiers CtIP and CtBP, was recruited to this binding site to form a repression complex for ZBRK1 transcription. Furthermore, we found that loss of RB-mediated transcriptional repression led to an increase in ZBRK1 transcript levels, which correlated with increased sensitivity to UV-and MMS-induced DNA damage. These results revealed a potential mechanism whereby the RB⅐E2F1 complex represses ZBRK1 transcription, and loss of this repression contributes to cellular sensitivity to DNA damage, ultimately leading to accumulation of the genomic mutation and carcinogenesis.  (CtIPEK). B, luciferase reporter assay of ZBRK1-B or ZBmE after cotransfection with CtIPwt (WT), CtIP⌬CtBP, or CtIPEK expression vectors in MCF10A cells. ***, p Ͻ 0.001 versus control. C, ZBRK1 promoter pulldown assay of the E2F1⅐Rb⅐CtIP⅐CtBP complex. DAPA was performed by incubating the biotinylated E2F motif of the ZBRK1 promoter with the MCF7 nuclear extract. The pulldown lysates were further analyzed by Western blot with E2F1, RB, CtIP, and CtBP antibodies as indicated. D, ChIP assay was performed with E2F1 antibody followed by a second immunoprecipitation with the indicated antibodies. The sheared formaldehyde cross-linked chromatin samples extracted from MCF7 cells with or without E2F1 transfection were subject to immunoprecipitation and PCR amplification with specific primers for the ZBRK1 promoter as described under "Experimental Procedures." E, the in vivo binding of E2F1⅐Rb⅐CtIP⅐CtBP complex on the ZBRK1 promoter. The re-ChIP assay was performed using the E2F1-immunoprecipated lysate and indicated antibodies. F, ZBRK1 ChIP assay by E2F antibody using lysates from MCF7 cells transfected with ZB or ZBmE reporter constructs. PCR amplification was performed with primers that recognize exogenously expressed ZBRK1 promoters. OCTOBER 22, 2010 • VOLUME 285 • NUMBER 43

RB⅐E2F1 Mediates DNA Damage through ZBRK1
In this study, we used the Comet and TUNEL assays to examine the response of ZBRK1 upon DNA damage in cell lines with varying RB expression. MCF7 and MB468 cell lines were chosen to be representatives of RB-proficient and RB-deficient cells, respectively. It is important to note that MCF7 and MB468 cells are also well known to be estrogen receptor positive and negative, respectively (46). Estrogens induce RB phosphorylation and thereby disrupt the formation of the RB⅐E2F1 complex, leading to cell cycle progression (47). Interestingly, our preliminary result demonstrated that the ZBRK1 transcript level decreased following estrogen-induced E2F1 expression in MCF7 but not in MB468. 4 Therefore, how the RB⅐E2F1 com-plex-mediated ZBRK1 repression responses to estrogen is an interesting question that warrants further investigation.
Although E2F1 was characterized as a transcriptional activator (48), several reports demonstrated that E2F1 can also act as a transcriptional repressor by inhibiting promoter activity (38, 49 -51). RB can form a complex with E2F1 to shut off an otherwise active promoter (52), Binding of RB to E2F1 was observed to increase after UV treatment (53). This is consistent with the finding that phosphorylation of RB at Ser-612 is induced by UV or ionizing radiation through the ATM-Chk1/2 pathway, leading to elevated formation of the complex between RB and E2F1 (12). The increased binding of RB and E2F1 was not only triggered by phosphorylation of RB, but also enhanced by stabilization of the E2F1 in response to DNA damage (12). These findings suggest that phosphorylation of RB plays an important role in the DNA damage response through enhancement of E2F1⅐RB complex formation. The identification of ZBRK1 as the in vivo downstream target controlled by the RB⅐E2F1 complex in response to DNA damage further supports this notion. This mode of action of RB is opposite to the role of RB in the cell cycle control, where RB dissociates from E2F1 binding upon phosphorylation. The issue of how different kinases specifically phosphorylate different residues of RB to modulate E2F1 binding is quite interesting. Detailed analysis of this interaction is warranted. Furthermore, a recent study revealed that BRCA1 is autoregulated by a dynamic coregulatory complex containing BRCA1, E2F1, and RB at the BRCA1 promoter. This complex can be disrupted by DNA-damaging agents leading to up-regulation of its own transcription (53). Utilizing a reporter assay, our result showed that BRCA1 overexpression has no effect on ZBRK1 reporter activity, indicating that BRCA1 is not involved in RB⅐E2F1-mediated repression of ZBRK1. 4 The cellular response to DNA damage is a complex process that involves a network of interacting signal transduction pathways (2,54). Such genomewide transcriptional response is very complex and tightly regulated, thus allowing a coordinated global response to DNA damage. This process is initiated by as yet unidentified proteins that detect or sense DNA damage and  and MMS (0.05%). C, depletion of E2F1 and RB reversed UV-and MMS-induced repression of ZBRK1 reporter activity. MCF7 cells were co-transfected with the ZBRK1-B reporter and various siRNAs as indicated, and treated with UV (40 J/m 2 ) or MMS (0.05%) for 2 h. Lysates from transfectants were harvested for luciferase assay. *, p Ͻ 0.05 versus scrambled control (siC); **, p Ͻ 0.01 versus siC. D, increases in E2F1, RB, CtIP, and CtBP binding to the ZBRK1 promoter after treatment with DNA damage inducers. Extracts with chromatin-protein complexes harvested from MCF7 cells treated with or without DNA damage inducers were immunoprecipitated with indicated antibodies and PCR amplified with specific primers on the ZBRK1 promoter.
subsequently transmit a signal by activating a cascade of phosphorylation events. This results in a variety of cellular responses, including activation of cell cycle checkpoints, commencement of transcriptional programs, execution of DNA repair, and in some cases, induction of cell death. We demonstrate in this study that DNA damage caused by UV or MMS leads to the repression of ZBRK1 transcription through binding of the CtBP⅐CtIP⅐RB⅐E2F1 complex to the ZBRK1 promoter. This triggers an increase in expression of a cluster of DNA damage response genes, including GADD45A (Fig. 8). Importantly, previous studies have shown that ZBRK1 is degraded through the ubiquitin-proteosome pathway upon UV treatment (32). These two events are coordinately regulated to ensure that DNA damage response genes repressed by ZBRK1 can be activated in response to DNA damage. ZBRK1 serves as a transcriptional repressor either by interacting with KAP1 through its N-terminal KRAB domain or by interaction with BRCA1 through its C-terminal domain (25). A group of genes have been identified to be tentatively regulated by ZBRK1 (25). Among these, GADD45A, Angiopoietin 1, and high mobility group AT-hook 2 have been shown to be regulated by ZBRK1 through interaction with the BRCA1⅐CtIP complex (27,28,32). In addition, our preliminary results showed that constitutive exogenous expression of ZBRK1 had no effect on the activation of cell cycle arrest and apoptosis in HeLa and U2OS cells, suggesting that transcriptional activation of genes essential for growth arrest or cell death are dependent on ZBRK1 repression. Moreover, our recent study demonstrates that ZBRK1 can suppress metastasis in cervical cancer (55). It is likely that ZBRK1, together with the KAP1 co-repressor, may repress other genes critical for the DNA repair process as well as other physiological functions, although solid evidence and examples remain to be demonstrated. It will be of interest to learn how ZBRK1 represses different genes through distinct co-repressor partners and how these complexes respond to different signals. This study is the first report to provide direct in vivo binding evidence demonstrating the E2F1⅐RB⅐CtIP⅐CtBP complex-mediated transcriptional repression of ZBRK1 and its role in the DNA damage response. RBmediated gene repression was shown to have both HDAC-dependent and -independent pathways (21). CtIP was reported to participate in the HDAC-independent path (56). Interestingly, our preliminary DAPA data showed that HDAC1 can be recruited to the wild-type E2F motif of the ZBRK1 promoter, 4 suggesting that formation of a CtIP⅐HDAC complex is in a The results are shown in the right panel. **, p Ͻ 0.01 versus siC. C, depletion of ZBRK1 resulted in an increase in GADD45A expression. The protocol was the same as that of A. Cells were harvested after 2 h of treatment with either UV or MMS. Western blotting was performed using antibodies as indicated. OCTOBER 22, 2010 • VOLUME 285 • NUMBER 43 gene-specific manner. However, the detailed mechanism of HDAC1-regulated ZBRK1 transcription needs to be further investigated.