MiR-630 inhibits proliferation by targeting CDC7 kinase, but maintains the apoptotic balance by targeting multiple modulators in human lung cancer A549 cells

MicroRNAome analyses have shown microRNA-630 (miR-630) to be involved in the regulation of apoptosis. However, its apoptotic role is still debated and its participation in DNA replication is unknown. Here, we demonstrate that miR-630 inhibits cell proliferation by targeting cell-cycle kinase 7 (CDC7) kinase, but maintains the apoptotic balance by targeting multiple activators of apoptosis under genotoxic stress. We identified a novel regulatory mechanism of CDC7 gene expression, in which miR-630 downregulated CDC7 expression by recognizing and binding to four binding sites in CDC7 3'-UTR. We found that miR-630 was highly expressed in A549 and NIH3T3 cells where CDC7 was downregulated, but lower in H1299, MCF7, MDA-MB-231, HeLa and 2BS cells where CDC7 was upregulated. Furthermore, the induction of miR-630 occurred commonly in a variety of human cancer and immortalized cells in response to genotoxic agents. Importantly, downregulation of CDC7 by miR-630 was associated with cisplatin (CIS)-induced inhibitory proliferation in A549 cells. Mechanistically, miR-630 exerted its inhibitory proliferation by blocking CDC7-mediated initiation of DNA synthesis and by inducing G1 arrest, but maintains apoptotic balance under CIS exposure. On the one hand, miR-630 promoted apoptosis by downregulation of CDC7; on the other hand, it reduced apoptosis by downregulating several apoptotic modulators such as PARP3, DDIT4, EP300 and EP300 downstream effector p53, thereby maintaining the apoptotic balance. Our data indicate that miR-630 has a bimodal role in the regulation of apoptosis in response to DNA damage. Our data also support the notion that a certain mRNA can be targeted by several miRNAs, and in particular an miRNA may target a set of mRNAs. These data afford a comprehensive view of microRNA-dependent control of gene expression in the regulation of apoptosis under genotoxic stress.

Cell division cycle 7 (CDC7) is a conserved serine-threonine kinase essential for the initiation of DNA replication. 1,2 Activation of CDC7 kinase requires its association with one of the regulatory proteins DBF4 and DRF1, 1-3 which are cyclically expressed and reach a peak during the S phase. 4-7 CDC7 modulates S-phase checkpoint in DNA damage response (DDR) 8-10 by attenuating checkpoint signaling and triggering DNA replication reinitiation. 11 CDC7 may also phosphorylate claspin and activate ATR-CHK1 checkpoint pathway. 12 CDC7 expression is very low or undetectable in normal tissues and cell lines but high in many human cancers and tumor cell lines. 13,14 Silencing CDC7 in cancer cells impairs progression through the S phase, inducing p53-independent apoptosis, but does not influence normal cells. 15,16 Therefore, CDC7 becomes an attractive target for cancer therapy. 17,18 MicroRNAs (miRNAs) posttranscriptionally regulate gene expression. MiRNAs control~30% protein-coding genes, 19 and have roles in diverse biologic processes including proliferation, differentiation and apoptosis. As miRNAs may function as either tumor suppressor or oncogene, deregulation of miRNAs is closely related to tumorigenesis. [20][21][22][23][24][25] MiRNAs are involved in DDR. For instance, miRNA-34 family members are regulated by p53 in DDR and have roles in cell-cycle checkpoint and apoptosis. [26][27][28][29] Many miRNAs (miR-24, miR-16, miR-421 and miR-138) have roles in DNA damage and repair. [30][31][32][33] MiRNA-regulated DDR may have the potential to improve the efficacy of cancer therapy relying on induction of DNA damage. Further understanding of miRNA actions in regulating cell death and DNA damage under genotoxic stresses will provide insights into cancer surveillance and limiting tumor progression.
MicroRNA-630 (MiR-630) is induced by cisplatin (CIS) and 3-Cl-AHPC (an adamantyl retinoid-related molecule), and it causes apoptosis in certain types of cancer cells by targeting different molecules such as BCL2, BCL2L2 and IGF-1R. 34,35 Moreover, miR-630 exerts cytoprotective effects in CISadministered A549 cells, but rather behaves as a specific cell death modulator in oxaliplatin-exposed A549 and CISexposed H1650, H1975 and HCC827 cells. 36 These observations indicate that the role of miR-630 in regulating apoptosis is not fully understood. Besides, direct targeting of a modulator involving in DNA replication by miRNA-630 is unknown. Here, we provide evidence that miR-630 downregulates CDC7 expression in A549 cells, thereby inhibiting CDC7-mediated DNA synthesis and contributing to CIS-induced inhibitory proliferation, but maintains the apoptotic balance by targeting multiple modulators.

Results
MiR-630 downregulates CDC7 by targeting CDC7 3'-UTR. Depletion of CDC7 induces apoptosis in cancer cells. 15,16 MiR-630 may target BCL2, BCL2L2 and IGF-1R to induce apoptosis under genotoxic stresses. 34,35 As an miRNA may have multiple targets, 14,37 we speculated that miR-630-induced inhibitory proliferation and, perhaps, apoptosis might be linked to CDC7. To demonstrate this hypothesis, the potential targets of miR-630 were searched by TargetScan software (http://www.targetscan.org), and CDC7 was selected. To validate whether miR-630 could target CDC7, we performed real-time quantitative PCR (RT-qPCR) to check the transfection efficiency (Supplementary Figure S1) and CDC7 expression after transfection of miR-630 mimic and inhibitor into A549 (p53-wt) cells. RT-qPCR and western blotting revealed that compared with transfection of scrambled siRNA, transfection of miR-630 mimic caused marked decreases in CDC7 mRNA and protein (Figures 1a and b), whereas transfection of miR-630 inhibitor led to significant increases of CDC7 mRNA and protein (Figures 1c and d). CDC7 downregulation was also observed in miR-630 mimic-transfected H1299 (p53-null), MCF7 (p53-wt) and MDA-MB-231 (p53-mutant) cells ( Figure 1e). These data indicate that miR-630 may target and inhibit CDC7 in a cell-type-independent manner. Figure 1 MiR-630 downregulates CDC7 expression by targeting CDC7 3'-UTR. A549 cells were transfected with miR-630 mimic or an inhibitor (50 nM) for 48 h. (a) RT-qPCR for CDC7 mRNA downregulation by miR-630. CDC7 mRNA was quantified by the 2 − ΔΔCt method, in which glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control. The scrambled small interfering RNA (siRNA) was used as a negative control, under which condition the level of CDC7 mRNA was normalized to '1'. Data are presented as mean ± S.D. (n = 3). (b) Western blotting for CDC7 protein downregulation by miR-630. α-Tubulin was used as a loading control. (c) RT-qPCR and (d) western blotting for effects of miR-630 inhibitor on CDC7 mRNA and protein expression. Anti-scrambled siRNA was used as a control. Data are presented as mean ± S.D. (n = 3). (e) Western blotting for CDC7 protein in H1299, MCF7 and MDA-MB-231 cells transfected with miR-630 mimic for 48 h. (f) The predicted miR-630-binding sequences or mutated versions of CDC7 3'-UTR fragments 'A', 'B', 'D' and 'E' in (g). WT, wild type; MT, mutant (mutated bases are underlined). (g) Interpretation of luciferase reporter plasmids containing full-length CDC7 3'-UTR, fragments 'A' to 'E' or mutants (upper panel). The full-length 3'-UTR, truncates 'A' to 'E' and mutants in (f) were inserted into the pMIR-Report plasmid to generate pMIR-Report-PmiR-3'-UTR and its variations. To demonstrate further CDC7 as a target for miR-630, we performed luciferase reporter assays and mutation of the seed sequences. For this purpose, we analyzed the miR-630binding sites within the sequences in silico, and found four predicted miR-630 sites in CDC7 3′-UTR. Full-length 3'-UTR, 3'-UTR fragments 'A', 'B', 'D' and 'E' (containing one putative miR-630 target site, respectively) and their mutants (Figure 1f), and fragment 'C' (no miR-630 site) were inserted downstream of the pMiR-Reporter vector to generate a series of reporter constructs (Figure 1g). The resulting constructs were co-transfected with scrambled siRNA or miR-630 mimic into A549. Luciferase reporter assays showed that miR-630 mimic blocked the luciferase activity expressed by the construct containing full-length CDC7 3'-UTR compared with the activity of pMiR-Reporter (Po0.0001) (Figure 1h). Also, the constructs containing 'A', 'B', 'D' or 'E' fragment significantly reduced reporter enzyme activities (P = 0.0098, P = 0.0053, P = 0.0086, P = 0.0007, respectively) when the miR-630 mimic was transfected, whereas the 'C' fragmentcontaining reporter did not lose its luciferase activity compared with control (pMiR-empty) vector. Moreover, the repression of luciferase activity was abolished when the putative miR-630 sites were mutated. These results indicate that miR-630 targets and downregulates CDC7 through direct binding to CDC7 3'-UTR. This conclusion was further evidenced by co-transfection of CDC7 3'-UTR-containing pEGFP-C3 vector with miR-630 mimic in GFP reporter assays (Supplementary Figure S2).
MiR-630 is differentially expressed but commonly induced in cancer and immortalized cells upon DNA damage. To study the correlation of CDC7 and miR-630 in cells, CDC7 and miR-630 expression in several cell lines was examined. CDC7 was expressed relatively lower or undetectable in A549 and NIH3T3, but high in H1299, MCF7, MDA-MB-231, HeLa and 2BS cells (Figure 2a  Northern blotting for miR-630 expression in CIS-exposed A549. Cells were exposed to 100 μM CIS for 36 h and Northern blotting was performed. U1, 5SRNA and tRNA were used as loading controls. Unexposed cells were used as control (Con). (e) RT-qPCR for miR-630 expression in CIS-exposed A549. Data are presented as mean ± S.D. (n = 3); *P = 0.0347. (f) RT-qPCR and (g) western blotting for CDC7 mRNA and protein in CIS-exposed A549. Data are presented as mean ± S.D. (n = 3); *P = 0.0171. (h) The effect of anti-miR-630 on CDC7 expression. A549 cells were transfected with miR-630 inhibitor (50 nM) for 48 h, and exposed to CIS for 36 h, followed by western blotting with anti-CDC7 antibody. The blots were screened/quantified and normalized against β-actin level. The value obtained from scrambled small interfering RNA (siRNA)/CIS-unexposed cells was designated as '1' (lane 1, bottom). (i and j) The effects of p53 transfection on miR-630 and CDC7 expression. A549 (p53-wild type) and H1299 (p53-null) cells were transfected with pcDNA3.1-p53 (pcDNA3.1 as control) for 48 h, followed by RT-qPCR for miR-630 (i) and western blotting for CDC7 and p53 (j). Data are presented as mean ± S.D. (n = 3) in (i), and α-tubulin was used as a loading control in (j). (k) P53 status in A549 and H1299. Cells were grown and exposed or unexposed to CIS for 36 h, followed by western blotting for p53 and p21 expression, and β-actin was used as a loading control These data also suggest that miR-630 as well as CDC7 is differentially expressed in a variety of cancer and immortalized cells.
To investigate miR-630 induction by genotoxic stress, we exposed A549 cells to CIS for 36 h. Northern blotting showed that miR-630 was upregulated under CIS exposure ( Figure 2d). RT-qPCR revealed that CIS exposure increased the level of miR-630 by~5.2-folds ( Figure 2e). Whereas the levels of CDC7 mRNA and protein were markedly decreased in CIS-exposed A549 compared with unexposed (control) cells (Figures 2f and g). To demonstrate further CDC7 downregulation by miR-630 under CIS exposure, anti-miR-630 was transfected into A549, and CDC7 was tested 36 h after CIS exposure. CDC7 downregulation was markedly abrogated by anti-miR-630 compared with scrambled siRNA (Figure 2h). The opposite expression of miR-630 and CDC7 was also detectable in camptothecin (CPT)-and CdCl 2exposed A549 (Supplementary Figure S3). Furthermore, miR-630 induction was observed in H1299, MCF7, MDA-MB-231 and 2BS cells exposed to CIS (data not shown). To clarify whether miR-630 induction was associated with p53, pcDNA3.1-p53 expression vector was transfected into A549 and H1299, followed by examining miR-630 and CDC7. Neither miR-630 nor CDC7 was affected in both cells (Figures 2i and j), regardless of the status of p53 and its target p21 before and after exposure ( Figure 2k). These data indicate that miR-630 induction by CIS may not be regulated by p53, and is a common event in human cancer and immortalized cells under genotoxic stresses.
CDC7 downregulation contributes to CIS-induced inhibitory proliferation and apoptosis in A549 cells. To examine the roles of downregulation of CDC7 by genotoxic agent in cell proliferation and apoptosis, A549 cells were exposed to CIS for 36 h, and apoptosis was determined by Annexin V/PI double staining and flow cytometry ( Figure 3a, upper panel). The early apoptosis induced by CIS was increased by~23% (P = 0.0033) ( Figure 3b). CIS-induced apoptosis was also confirmed by the activation of procaspase-3 that was cleaved to generate a 17 kDa activated product, which in turn led to cleavage of its substrate PARP ( Figure 3c). Furthermore, CIS exposure reduced cell survival by 36.7% in non-radioactive cell proliferation assay (MTS) (P = 0.0049) (Figure 3d). To demonstrate whether CDC7 was associated with CISinduced apoptosis, CDC7 was silenced in A549 by CDC7 siRNA (siCDC7-1). Silencing CDC7 increased early apoptosis from 7.7% (scrambled siRNA) to 12.9% (P = 0.0086) (Figures 3a and e, lower panel), followed by procaspase-3 activation (Figure 3f), whereas cell survival decreased by 19.8% compared with control ( Figure 3g). The effects of silencing CDC7 on apoptosis and cell survival were further demonstrated by a second siRNA (siCDC7-2) (Supplementary Figure S4). Taken together, CDC7 downregulation contributes to CIS-induced inhibitory proliferation and apoptosis.
To further test the role of CDC7 in CIS-induced inhibitory proliferation and apoptosis, CDC7 was overexpressed in A549, followed by CIS exposure. Compared with pcDNA3.1empty vector transfection, overexpressing CDC7 reduced CIS-induced apoptotic cells from 25.8 to 17.7% (P = 0.0194) (Figure 3h and Supplementary Figure S5). Consistently, the activated product of procaspase-3 was reduced compared with CIS-exposed A549 without overexpressing CDC7 ( Figure 3i). Unsurprisingly, overexpressing CDC7 increased cell survival even in the presence of CIS compared with control ( Figure 3j). These data indicate that CDC7 expression increases cell resistance to CIS-induced apoptosis.
MiR-630 inhibits proliferation by blocking CDC7mediated DNA synthesis. MCM2 phosphorylation by CDC7 promotes binding of CDC45 to the pre-RC during replisome assembly. 3,7 To study the effects of CDC7 downregulation by miR-630 on DNA replication initiation and progression, miR-630 mimic or inhibitor was transfected into A549. After 48 and 72 h transfection, CDC7 and phosphorylated MCM2 were downregulated by miR-630 mimic compared with miR-630 inhibitor and scrambled siRNA (  miR-630 mimic or inhibitor was transfected into A549, followed by the detection of apoptosis. Unexpectedly, in Annexin V/PI assays we could not detect any apoptosis in miR-630 mimic-transfected A549 compared with scrambled siRNA and inhibitor transfection (Figure 5a and Supplementary Figure S6), although miR-630 mimic as well as CDC7 siRNA markedly repressed CDC7 expression (Figure 5b). Consistently, procaspase-3 activation was undetectable (data not shown) in miR-630 mimictransfected A549, indicating that miR-630 overexpression MiR-630 targeting CDC7 under genotoxic stress J-X Cao et al failed to induce apoptosis. As apoptosis induced by CDC7 depletion in cancer cells is associated with ATR-activated p38 kinase, 16 we examined phosphorylation/activation of p38 kinase in CDC7-silenced A549 cells, where phosphor-p38 was upregulated (Figure 5b, left panel). Surprisingly, the activated kinase was also detectable in miR-630 mimic-transfected A549 (right panel). These results suggest that the p38 kinase-related apoptotic pathway is still operative in miR-630-expressed or CDC7-silenced A549. Thus, the functional significance of miR-630 induction in CIS-induced apoptosis remains to be established.
Taken together, miR-630 not only induces apoptosis by targeting CDC7 but also protect the cell from apoptosis by direct and indirect downregulation of DDIT4, PARP3, EP300 and p53. Alternatively, miR-630 may exert a bimodal role in the regulation of apoptosis by targeting multiple modulators of apoptosis, thereby maintaining the apoptotic balance.
MiR-630 induction is associated with CIS-induced G1 arrest. To investigate the effects of CIS-induced miR-630 on cell cycle, A549 cells were synchronized at G0/G1-phase by serum starvation and released into the cell cycle by adding 20% fetal calf serum and exposed to CIS for given hours, followed by cell-cycle analysis. G1 arrest was observed immediately after CIS exposure and sustained until at least 36 h (Figures 6a and b). The majority of G1 population with a sub-G1 peak appeared 24 h after exposure (Figure 6a). Meanwhile, miR-630 was upregulated within 3-36 h after exposure (Figure 6c). Contrarily, CDC7 was markedly downregulated after 6 h exposure (Figure 6d). Supporting G1 arrest, CIS exposure increased p-ATM, p53 and p21 expression, whereas cyclin D1 was downregulated after 6 h exposure. Phospho-ATR was consistently detectable, although its levels were not increased. Moreover, phospho-p38 was increased (Figure 6d). These data indicate that CIS exposure may induce G1 arrest and apoptosis in A549. It seems likely that CIS-induced G1 arrest is attributed to the activation of ATM/ATR-p53-p21 signaling pathway, and CISinduced apoptosis is linked to the activation of p53 and p38 kinase.
To demonstrate whether miR-630 induction was associated with CIS-induced G1 arrest, miR-630 mimic or scrambled siRNA was transfected into A549 cells, and the cell cycle was analyzed 48 h after transfection. MiR-630 overexpression significantly increased G1 sub-population (67%) compared with scrambled siRNA-transfected cells (49% G1 sub-population) (Figure 6e), indicating that miR-630 induction is associated with CIS-induced G1 arrest but not apoptosis.

Discussion
In this study, we demonstrate that miR-630 as a DDR component is induced upon CIS exposure, leading to subsequent modulation of several targets involved in DNA replication and apoptosis. First, we identified a novel regulatory mechanism of CDC7 gene, whose expression is suppressed by miR-630 by targeting CDC7 3'-UTR. Second, miR-630-suppressed CDC7 inhibits CDC7-mediated initiation of DNA synthesis and induces G1 arrest, contributing to CISinhibited cell proliferation. Third, miR-630 may exert a bimodal role in the regulation of apoptosis by targeting multiple modulators of apoptosis, therebymaintaining the apoptotic balance.
CDC7 is frequently downregulated in response to genotoxic stresses, resulting in the activation of S-phase checkpoint signal. 8,9,11,15,16 Overexpression of CR/periphilin may also downregulate CDC7 and induce S-phase arrest. 43 Recently, CDC7 was shown to be downregulated by miR-29a under BPDE-induced DNA damage. 44 We found that miR-630 downregulated CDC7 by targeting its 3'-UTR (Figure 1). MiR-630 induction and CDC7 suppression were detectable in A549 exposed to CIS (Figures 2d-h), CPT and CdCl 2 (Supplementary Figure S3). We also observed miR-630 induction in H1299, MCF7, MDA-MB-231, HeLa and 2BS cells under CIS exposure (data not shown). The inverse correlation of miR-630 and CDC7 expression occurred in several cancer and immortalized cells even in the absence of DNA damage agents (Figures 2a-c). We conclude that miR-630 acts as a modulator of CDC7. Our data also suggest that the regulatory mechanism of CDC7 by miR-630 is operative in cancer and immortalized cells, in particular when exposed to DNA-damaging agents. MicroRNAome studies show that several miRNA species are deregulated in cancer cells during CIS-induced DNA damage, in which miR-630 is upregulated. 34,36 Taking the previous findings 34, 36 and our data, miR-630 induction is a DDR component. MiR-29a is not included in the deregulated miRNA species under CIS exposure. 34,36 It seems that targeting CDC7 by miR-630 or miR-29a depends upon cell contexts and genotoxic agents.
DNA damage may regulate miRNA expression at the transcriptional level, in which transcription factors c-Myc, 45 p53 46 and E2F1 [47][48][49] have roles. DNA damage may also regulate miRNA expression by modulating miRNA processing and maturation. 50 We found that p53 overexpression could not induce miR-630 and suppress CDC7 in A549 and H1299 cells (Figures 2i and j). Also, we observed miR-630 induction in p53-deficient H1299, MDA-MB-231 and HeLa cells under CIS exposure (data not shown). Our data indicate that miR-630 induction may not be associated with p53. A recent study showed that regulation of the DICER1 promoter by ATMphosphorylated delta-DNp63a was implicated in CIS-induced alteration in microRNAome (including miR-630). 34 We reported recently that CIS-induced miR-630 was attributed to the promotion of pri-miR-630 processing by E2F1-regulated DROSHA. 51 However, activation of the miR-630 gene by DNA damage is unknown.
CDC7 coordinates with CDK2 kinase to activate the MCM complex bound at the pre-RC, resulting in the recruitment of DNA polymerases and initiation of DNA replication. 52 Downregulation of CDC7 as well as CDK2 and MCM gene family is associated with the inhibitory effects of genistein and TSA on DNA replication. 53 We found that miR-630-inhibited A549 cell proliferation (Figure 4d) was associated with the downregulation of CDC7 and phospho-MCM2 (Figure 4a), which led to the blockage of CDC7-mediated initiation of DNA synthesis (Figures 4b and c). This can at least partly explain CISinduced inhibitory proliferation. Depletion of CDC7 in cancer cells impairs progression through the S phase. 1,11,15,16 Transfection of pre-miR-630 arrests A549 at G0/G1 by p27Kip1 induction upon CIS exposure. 36 Similarly, we found that miR-630 transfection arrested A549 at the G1 phase (Figure 6e), and CIS exposure induced G1 arrest (Figures 6a  and b), followed by miR-630 induction ( Figure 6c). As miR-630 overexpression could partly diminish CIS-induced p53 (Figure 5i) due to targeting EP300 that acetylates p53, 39,40 we suppose that the signal pathway in CIS-induced G1 arrest should differ from miR-630 transfection-resulted G1 arrest. The former depends upon ATM/ATR-p53-p21 signal pathway in A549 (Figure 6d), whereas the latter is attributed to miR-630-mediated CDC7 inactivation, which leads to defective G1-S transition.
We found that silencing CDC7 caused apoptosis in A549 (Figures 3e and a, lower panel), as CIS did so (Figures 3a-c), whereas CDC7 overexpression diminished CIS-induced apoptosis (Figures 3h and i and Supplementary Figure S5), indicating that CDC7 downregulation is responsible at least partly for CIS-induced apoptosis. Surprisingly, we could not detect apoptosis in miR-630 transfected cells (Figure 5a and Supplementary Figure S6), indicating that overexpression of miR-630 and depletion of CDC7 by RNA interference (RNAi) (or downregulation of CDC7 by CIS) do not have equal effect on apoptosis, probably because of the multiple target effects of miR-630. We showed that DDIT4, PARP3 and EP300, which have been shown to be apoptotic activators, 34,[38][39][40] were specific targets for miR-630 (Figure 5c). MiR-630 transfection significantly reduced expression of DDIT4, PARP3 and EP300 mRNAs and proteins in A549 (Figures 5d and e). Importantly, individual silencing, in particular, combined silencing of DDIT4, PARP3 and EP300, significantly inhibited apoptosis (Figures 5f and g and Supplementary Figure S8) and procaspase-3 activation (Figure 5h). Furthermore, miR-630 overexpression markedly decreased acetylated and phosphorylated p53 and total p53 in CIS-exposed A549 (Figure 5i). EP300 may act as a regulator of p53 via its acetylase. 39,40 Therefore, decreased acetylation of p53 should be attributed to miR-630-targeted EP300. Our data suggest that miR-630 may directly or indirectly downregulate both apoptotic activator (s) and inhibitor(s) that have a bimodal role in the regulation of apoptosis. On the one hand, miR-630 promoted apoptosis by downregulating CDC7; on the other hand, it reduced apoptosis by downregulating apoptotic activators DDIT4, PARP3, EP300 and p53, thereby maintaining apoptotic balance in DDR (Figure 7). This can explain why we could not detect apoptosis in miR-630-transfected cells (Figure 5a). Although miR-630 transfection or induction may partly offset p53 acetylation and phosphorylation, the levels of chemomodified p53 in CISexposed cells were still higher than control (Figure 5i). We found recently that p53 was accumulated in HCT116p53 +/+ and A549 upon CIS-induced DNA damage. 54 Our recent observation 54 and the data fromFigure 5i suggest that apoptotic mechanism still works in the presence of CIS, although miR-630 downregulates p53 by targeting EP300. Notably, miR-630 overexpression is an extreme case that differs from miR-630 induction by CIS. Therefore, the bimodal role of miR-630 does not discriminate against the contributions of CDC7 inactivation and other apoptotic regulators to CISinduced apoptosis.
In summary, we have demonstrated that miR-630 targets CDC7, thereby inhibiting CDC7-mediated initiation of DNA synthesis and inducing G1 arrest, but has a bimodal role in the regulation of apoptosis because of its multiple target effects, maintaining apoptotic balance under genotoxic stress. These data afford a comprehensive view of miRNA-dependent regulation of gene expression in regulating apoptosis in DDR.

Materials and Methods
Cell culture and drug treatment. Human lung cancer A549 and H1299 cells and human cervical adenocarcinoma HeLa cells were from American Type Culture collection (ATCC, Manassas, VA, USA). Human breast cancer MCF7 and MDA-MB-231 cells were a gift from Dr Yong-Feng Shang (Peking University Health Science Center, Beijing, China). Cells were cultured in DMEM medium supplemented with 10% fetal bovine serum (GIBCO-BRL, Carlsbad, CA, USA), 100 IU/ml penicillin and 100 mg/ml streptomycin, and grown at 37°C with 5% CO 2 . CIS (100 μM) (Sigma, St Louis, MO, USA), CPT (1 μM) (Sigma) and Cdcl2 (50 μM) (Sigma) in DMSO were added to cultures for given hours, respectively.
Transfection was performed with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions. Cells were plated in 6-well plates at 50% confluence and transfected with the miRNA mimics, miRNA inhibitors and siRNAs at final concentrations of 50 nM, and collected 48 h after transfection for further analysis.
Northern blotting analysis. Total RNA was extracted using Trizol reagent (Invitrogen), resolved on 12.5% denaturing polyacrylamide gel and electrotransferred to Hybond N + membranes (Amersham Biosciences, Little Chalfont, UK), followed by UV crosslinking the membrane two times at 120 mJ for 30 s each. After prehybridization for 2 h at 42°C in a highly efficient hybridized solution (Mylab, Beijing, China), the membrane was hybridized overnight with the Dig-labeled probes corresponding to miR-630 (synthesized by Sangon, Shanghai, China) in a highly efficient hybridized solution. Following hybridization, immunologic detection was performed using DIG Detection Kit (Mylab). The membrane was washed two times with 2 × SSC and 0.05% SDS at room temperature (RT) for 5 min each, two times with 0.1 × SSC and 0.1% SDS at 50°C for 15 min each and one time with MABT (0.1 M maleic acid, 0.15 M NaCl, 0.3% Tween-20, pH 7.5) at RT for 5 min. The membrane was blocked with blocking buffer at RT for 30 min, and then hybridized overnight with Dig-AP in 1 × blocking buffer at 4°C. After incubation with Dig-AP, blots were visualized by Enhanced Chemiluminescence Regents CDP-star (Mylab).
Western blotting. Whole-cell extracts were prepared in lysis buffer and the concentration of proteins was determined using the BCA Protein Assay Reagent Kit (Thermo scientific). Proteins were separated in 10% SDS-PAGE and transferred to nitrocellulose membrane. The membrane was blocked using TBST buffer containing 5% (w/v) non-fat milk 1 h at RT and probed with specific antibodies