p38 MAPK/MK2-mediated induction of miR-34c following DNA damage prevents Myc-dependent DNA replication

Abstract

The DNA damage response activates several pathways that stall the cell cycle and allow DNA repair. These consist of the well-characterized ATR (Ataxia telangiectasia and Rad-3 related)/CHK1 and ATM (Ataxia telangiectasia mutated)/CHK2 pathways in addition to a newly identified ATM/ATR/p38MAPK/MK2 checkpoint. Crucial to maintaining the integrity of the genome is the S-phase checkpoint that functions to prevent DNA replication until damaged DNA is repaired. Inappropriate expression of the proto-oncogene c-Myc is known to cause DNA damage. One mechanism by which c-Myc induces DNA damage is through binding directly to components of the prereplicative complex thereby promoting DNA synthesis, resulting in replication-associated DNA damage and checkpoint activation due to inappropriate origin firing. Here we show that following etoposide-induced DNA damage translation of c-Myc is repressed by miR-34c via a highly conserved target-site within the 3^′ UTR. While miR-34c is induced by p53 following DNA damage, we show that in cells lacking p53 this is achieved by an alternative pathway which involves p38 MAPK signalling to MK2. The data presented here suggest that a major physiological target of miR-34c is c-Myc. Inhibition of miR-34c activity prevents S-phase arrest in response to DNA damage leading to increased DNA synthesis, DNA damage, and checkpoint activation in addition to that induced by etoposide alone, which are all reversed by subsequent c-Myc depletion. These data demonstrate that miR-34c is a critical regulator of the c-Myc expression following DNA damage acting downstream of p38 MAPK/MK2 and suggest that miR-34c serves to remove c-Myc to prevent inappropriate replication which may otherwise lead to genomic instability.

The protein encoded by the proto-oncogene c-Myc (hereafter referred to as Myc) functions as an essential regulator of G₁-S transition by promoting the transcription of mRNAs encoding proteins that drive cell cycle progression and cell growth (1). Myc is tightly regulated at multiple levels and its deregulated expression is associated with a wide range of cancers, particularly those of B-cell origin (2). Further to these well-established functions, Myc also has a role in DNA replication which is distinct from its transcriptional activity. It has been shown that Myc directly interacts with components of the prereplicative complex and localizes to early sites of DNA synthesis. Overexpression of Myc in this context results in checkpoint activation and DNA damage due to replication stress imposed by inappropriate replication origin activity (3). Furthermore, overexpression of Myc following DNA damage prevents cell cycle arrest and promotes apoptosis (4).

The DNA damage response (DDR) has traditionally been divided into two distinct yet parallel pathways, which utilize the upstream kinases ATR and ATM (5). These signal to the checkpoint kinases, CHK1 or CHK2, respectively (6, 7). Recently a third pathway emerged that involves ATM/ATR signalling to p38 MAPK which in turn activates the newly identified checkpoint kinase MAPKAPK2 (MK2) (8, 9). Both ATM/ATR and p38 MAPK activate p53 which transcriptionally regulates gene expression following DNA damage resulting in cell cycle arrest and/or apoptosis (10–12). However, the DDR is not solely dependent upon p53, since p53-deficient cells can still respond to and repair DNA damage. In the absence of p53 the p38 MAPK pathway is essential for survival in response to chemotherapeutic drugs that cause DNA damage (13), due to loss of MK2-mediated S-phase and/or G2/M arrest.

A newly identified component of the cellular response to DNA damage is the induction of microRNAs (miRNAs) which have apoptotic or cell cycle arrest promoting activities. MiRNAs are noncoding RNAs, 21–25 nucleotides in length, that posttranscriptionally regulate gene expression by binding with imperfect complementarity to the 3^′ UTR of target mRNAs and inhibiting their translation (14). In particular p53 transcriptionally activates the miR-34 family of miRNAs and these appear to be vital components of the DDR (15–20). In vertebrates this family comprises miR-34a and miR-34b/c which are expressed on a separate transcript and are subject to coordinated regulation (18). These microRNAs are up-regulated in response to DNA damage and have roles in regulating apoptosis and/or cell cycle arrest.

Here we show that Myc is translationally repressed in response to DNA damage by the induction of miR-34c. Although miR-34c induction is maximal in p53-proficient cells, miR-34c induction and subsequent Myc repression also occur in cells that are devoid of p53, albeit to a lesser extent. In p53-deficient cells an alternative pathway mediates miR-34c induction which involves p38 MAPK signalling to MK2. Functionally, miR-34c-mediated repression of Myc following DNA damage is required to inhibit DNA synthesis and block cells in S-phase to prevent replication of damaged DNA. In the absence of this miRNA-mediated repression, there is a Myc-dependent increase in checkpoint activation and DNA damage over and above that induced by etoposide alone. These findings suggest that following DNA damage miR-34c functions to repress Myc to prevent inappropriate replication.

Results

miR-34c Represses Myc Following DNA Damage.

We have previously shown that Myc is repressed by miR-34c in HeLa cells via a target-site within its 3^′ UTR (21). Inhibition of miR-34c with a 2^′ O-methyl oligonucleotide (anti-miR) relieves repression of a luciferase reporter harboring the wild-type c-myc 3^′ UTR (pLSVM3^′) relative to a version with a point mutation in the miR-34c target-site (pLSVM3^′mut) (Fig. 1A). Anti-miRs directed against miR-34a or miR-34b lead to an increase in luciferase activity from the wild-type c-myc 3^′ UTR reporter but this failed to reach significance (Fig. S1B). While anti-miRs allow us to determine the role of endogenous microRNAs they may cross-react with closely related microRNA species such as other members of the miR-34 family. To further assess the role of miR-34c in Myc repression in a different way, the 3^′ UTR luciferase reporters were transfected into cells that have low miR-34c expression but high levels of miR-34a (HEK293 cells, Fig. S2 A and B) along with a mimic corresponding to mature miR-34c or a control sequence. Transfection of miR-34c mimic led to over 50% repression of luciferase activity from the construct containing the wild-type c-myc 3^′ UTR (pLSVM3^′) relative to the mutant version (pLSVM3mut) (Fig. 1B). Overexpression of miR-34a or miR-34b led to repression of luciferase activity from the wild-type c-myc 3^′ UTR construct but this failed to reach significance (Fig. S1C). The microRNA mimics were functional since they were equally efficient at reducing luciferase activity from a reporter mRNA with a perfectly matched target-site in its 3^′ UTR (Fig. S1D). Enforced expression of all three miR-34 family members reduced Myc protein abundance; however, miR-34c was the most potent (Fig. S1E). Due to the potential cross-reactivity of anti-miRs we cannot rule out the possibility that miR-34b may function in Myc repression; however, our data would suggest that miR-34a does not have a major role Myc repression (Fig. 2 and Fig. S3B). In both overexpression and inhibition studies miR-34c has the most potent affect on the wild-type c-myc 3^′ UTR reporter construct so we chose to study this miR-34 family member further.

Fig. 1.

miR-34c represses translation of c-Myc. (A) Sequestering miR-34c relieves repression of a reporter construct harboring the wild-type c-myc 3^′ UTR in HeLa cells. HeLa cells were cotransfected with pLSVM3^′ or pLSVM3^′mut and either a control anti-miR or anti-miR-34c. Luciferase acitivity of the wild-type c-myc 3^′ UTR (gray bars) construct are expressed as a percentage of luciferase activity of the seed-mutant (black bars) normalized to a LacZ transfection control. Values are mean ± SD (t-test, n = 3, compared to control anti-miR transfected cells). (B) Transfection of exogenous miR-34c represses translation of a reporter construct harboring the wild-type c-myc 3^′ UTR. HEK293 cells were transfected with pLSVM3^′ or pLSVM3^′mut and either a control siRNA or a miR-34c mimic. Values are expressed and normalized as in (B). Values are mean ± SD (t-test, n = 3, compared to control siRNA cells transfected cells). (C) Myc is repressed at the posttranscriptional level following etoposide treatment. HEK293 cells were treated with etoposide at the indicated concentrations for 24 h. Parallel samples were taken for western and northern analysis as indicated. (D) The c-myc 3^′ UTR reporter construct is repressed in response to etoposide in a miR-34c-dependent manner. HEK293 cells were transfected as indicated and 24 h later treated with 12.5 μM etoposide. Luciferase activity was normalized to a LacZ transfection control and expressed as a percentage of luciferase activity from DMSO treated cells (t-test, n = 3, compared to DMSO control anti-miR transfected cells). (E) Endogenous Myc is repressed in response to etoposide in a miR-34c-dependent manner. HEK293 cells were transfected with control anti-miR or anti-miR-34c and 24 h later treated with 12.5 μM etoposide. Percentages indicate relative translational efficiency compared to control.

Fig. 2.

miR-34c-mediated repression of Myc in response to DNA damage is maintained in p53-deficient cells. (A) miR-34c is induced following DNA damage in p53-/- MEFs. p53 -/- MEFs were treated with 25 μM etoposide for 24 h, RNA extracted and subjected to qPCR analysis for miR-34c. Data are expressed as fold-change relative to control. Values are mean ± SD (t-test, n = 3, compared to DMSO treated cells). (B) Myc is repressed in response to etoposide in a miR-34c-dependent manner in p53-/- MEFs. p53 -/- MEFs were transfected as indicated and 24 h later treated with 25 μM etoposide. Data shown are representative of three independent experiments.

To determine whether miR-34c represses Myc following DNA damage, HEK293 cells, which express low basal levels of miR-34c (Fig. S2B), were treated with etoposide to introduce both single and double strand breaks into DNA. Treatment with etoposide resulted in increased p53, p21, and miR-34c expression as expected and a decrease in Myc protein levels with no decrease in c-myc mRNA levels, measured by western and northern analysis, respectively (Fig. 1C). Pulse-chase immuno-precipitation experiments showed no difference in Myc protein half-life upon etoposide treatment (Fig. S1F) demonstrating that Myc is repressed at the translational level following DNA damage. To determine whether the c-myc 3^′ UTR was responsible for Myc repression following DNA damage, HEK293 cells were transfected with the luciferase reporter constructs and treated with etoposide. Exposure of cells to etoposide resulted in a 75% decrease in luciferase activity from the reporter containing the wild-type c-myc 3^′ UTR relative to untreated cells whilst the seed-mutant version was only repressed by 30% (Fig. 1D). Inhibition of miR-34c with a specific anti-miR almost completely relieved repression of this reporter construct when cells were exposed to etoposide. Furthermore, DNA damage-induced repression of endogenous Myc was relieved by more than 40% following miR-34c inhibition (Fig. 1E). These data show that miR-34c represses Myc translation following DNA damage via a highly conserved target-site within its 3^′ UTR.

Repression of Myc by miR-34c Following DNA Damage in p53-Deficient Cells.

Initial qPCR analysis for miR-34 expression in different cell lines revealed that although the basal expression levels of miR-34a (Fig. S2A) correlated with p53 status, levels of miR-34c (Fig. S2B) did not. In fact miR-34c expression was highest in two p53 negative cell lines, SaOS2 and p53 -/- mouse embryonic fibroblasts (MEF) (Fig. S2B). This prompted us to investigate whether miR-34c induction could occur in the absence of p53. Three different p53 negative cell lines were treated with etoposide, all of which showed ∼2-fold or greater increase in miR-34c expression following DNA damage (Fig. 2A and Fig. S2C). Enforced expression of p53 in SaOS2 cells demonstrated that miR-34 induction was enhanced by p53, consistent with previous reports (Fig. S3A) (16, 17). Depletion of p53 by siRNA in HEK293 cells reduced miR-34c induction by half (Fig. S3C). However, expression of p53 in H1299 (also p53-negative) cells, by use of a tetracycline inducible form of wild-type p53, had no effect on miR-34c induction (Fig. S3B). These data demonstrate that although p53 is clearly an important regulator, other factors influence miR-34c expression in its absence. We next sought to determine whether, in p53-deficient cells, Myc was repressed in a miR-34c-dependent manner. To this end p53 -/- MEFs were transfected with a control anti-miR or anti-miR-34c and treated with etoposide. We observed miR-34c-dependent Myc repression in response to DNA damage (Fig. 2B), despite no change in miR-34a levels, suggesting that this induction of miR-34c can repress Myc (Fig. 2). These data demonstrate that although p53 is an important factor for miR-34c regulation in response to DNA damage, in its absence, other pathways can induce expression of miR-34c to maintain Myc repression.

MiR-34 Is Induced by p38 MAPK/MK2 Signalling Following DNA Damage.

The DNA damage response involves a cascade of signalling events, mediated by Serine/Threonine protein kinases, which act on their substrates to enforce the biological output of the DDR. The involvement of these pathways in the regulation of microRNAs, specifically the induction of miR-34, have not been investigated. One of the earliest events in the DNA damage response is the recognition of DNA breaks by one or more of the sensor kinases ATM, ATR, or DNA-PK. These proteins are all members of the highly conserved phosphotidyl-inositol-3-kinase-like (PI3K) family, and as such are inhibited by LY294002, a PI3K inhibitor. To determine the effect of ATM/ATR/DNA-PK inhibition on miR-34c induction HEK293 cells were treated with LY294002 for 30 min prior to etoposide treatment and miR-34 levels measured by qPCR. In the absence of LY294002 miR-34c was induced 14-fold and miR-34a was induced 5-fold (Fig. 3A). However, when cells were pretreated with LY294002, miR-34a and miR-34c induction was prevented (Fig. 3A). These data are consistent with the idea that sites of DNA damage are detected by ATM, ATR, or DNA-PK which in turn activate a pathway that leads to miR-34c induction. Analysis of Myc levels by western blot demonstrated that this reduction in miR-34c levels following DNA damage was sufficient to prevent Myc repression (Fig. 3A). Western analysis confirmed that phosphorylation of ATM/ATR substrates p38 MAPK, CHK1, and CHK2 was reduced following treatment with LY294002 (Fig. 3A). To determine the kinetics of miR-34c activation and Myc repression in the context of known DDR pathways, we performed a time course of etoposide treatment in HEK293 cells (Fig. S4). We observed that induction of miR-34c and repression of Myc correlated with maximal p38 MAPK/MK2 activation. However, an increase in p38 MAPK/MK2 signalling is first observed at 1 h after etoposide treatment and miR-34c induction occurs at 12 h suggesting that other factors, most likely a transcription factor, lie between p38 MAPK/MK2 activation and miR-34c transcriptional induction.

Fig. 3.

The p38 MAPK/MK2 pathway controls miR-34c induction in response to DNA damage. (A) Inhibition of sensor kinases ATM, ATR, and DNA-PK prevents miR-34c induction and Myc repression in response to DNA damage. HEK293 cells were pretreated with LY294002 for 30 min prior to treatment with 12.5 μM etoposide for 24 h. Values are mean ± SD (t-test, n = 3, compared to DMSO treated cells). (B) Inhibition of p38 MAPK but not MEK1/2 prevents miR-34c induction and Myc repression in response to DNA damage. HEK293 cells were pretreated with SB20350 or PD098059 for 30 min prior to treatment with 12.5 μM etoposide for 24 h. Values are mean ± SD (t-test, n = 3, compared to DMSO treated cells). (C) MK2 influences miR-34c induction in response to DNA damage. HEK293 cells were transfected as indicated and 24 h later treated with etoposide. Values are mean ± SD (t-test, n = 3, compared to control DMSO treated cells). (D) Constitutively active MK2 can restore miR-34c induction and Myc repression following p38 MAPK inhibition. p53 -/- MEFs were transfected with empty vector or MK2 EE and treated with etoposide or pretreated with SB203580 for 30 min followed by etoposide for 24 h. Values are mean ± SD (t-test, n = 3, compared to DMSO treated cells).

The p38 MAPK/MK2 pathway is an important regulator of the DNA damage response in both p53-proficient and p53-deficient cells but in the absence of p53 it becomes essential (13). Because of the nature of this correlation (Fig. S4) we hypothesised that miR-34c induction may occur through this signalling pathway. To test this HEK293 cells were pretreated with SB203580, a p38 MAPK inhibitor, or PD098059, a MEK1/2 inhibitor as a control, for 30 min prior to etoposide treatment. Inhibition of p38 MAPK but not MEK1/2 prevented etoposide-induced miR-34c expression (Fig. 3B), miR-34b expression (Fig. S5A), and repression of Myc (Fig. 3B). Western blotting for phosphorylated MK2, a direct p38 MAPK substrate, confirmed inhibition of p38 kinase activity (Fig. 3B). These data show that p38 MAPK inhibition prevents miR-34c induction in response to DNA damage. However, this could be due to disruption of signalling to p53 despite continued p21 induction under these conditions (Fig. 3B). To directly assess whether MK2 could influence miR-34c expression, HEK293 cells were transfected with an siRNA targeting MK2 or expression vectors containing wild-type MK2 (MK2 Wt), a mutant version that is catalytically inactive (MK2 K76R), or a constitutively active version (MK2 EE) (22, 23). Depletion of MK2 by siRNA significantly inhibited miR-34c induction (Fig. 3C); however, this inhibition was not sufficient to prevent Myc repression (Fig. S5C) perhaps due to incomplete depletion of MK2, a role for another p38 substrate, or continued signalling to p53 under these conditions. Overexpression of wild-type MK2 or a constitutively active form of MK2 enhanced miR-34c induction following etoposide treatment although a version of MK2 with a mutation in the catalytic domain had no effect (Fig. 3C). The finding that MK2 can influence miR-34c expression prompted us to ascertain whether MK2 could induce miR-34c independently of p53 in a p53-deficient background. To this end p53 -/- MEFs were transfected with empty vector or constitutively active MK2 (MK2 EE) and treated with SB203580 prior to etoposide treatment. Pretreatment of these cells with SB203580 (Fig. 3D) or transfection with an siRNA targeting MK2 (Fig. S5D) prevented miR-34c induction in response to etoposide which could be overcome by addition of MK2EE (Fig. 3D). We saw a similar effect on the primary-miR-34b/c (pri-miR-34b/c) transcript levels (Fig. S5E) suggesting that regulation of miR-34c by p38 MAPK/MK2 is at the transcriptional level. These data demonstrate that MK2 functions downstream of p38 MAPK in this pathway (Fig. S5F) and that the p38 MAPK influences miR-34c induction via MK2 in both p53-proficient and p53-deficient cells. However, MK2 is only completely required for this induction in cells that lack functional p53 (Fig. S5F).

miR-34c Repression of Myc Mediates S-Phase Arrest and Inhibition of DNA Synthesis Following DNA Damage.

MK2 is an important regulator of the cell cycle following DNA damage mediating both S-phase and G2/M arrest. We hypothesised that miR-34c repression of Myc may play a role in regulating the cell cycle following DNA damage. EdU (5-ethynyl-2^′-deoxyuridine) is a thymidine analog that is incorporated into DNA during active DNA synthesis in S-phase (24). Cells that are arrested in S-phase due to exposure to ionizing radiation or radio-mimetic drugs do not incorporate EdU. However, cells with defects in S-phase checkpoints can continue to incorporate EdU following exposure to such agents.

To examine the effect of miR-34c repression of Myc on the cell cycle, HEK293 cells were transfected with combinations of anti-miR-34c and c-myc siRNA. Cells were transfected as indicated and 24 h later were treated with etoposide. One hour prior to harvest cells were pulse labeled with EdU then fixed and stained with propidium iodide as a measure of DNA content. When control cells were treated with etoposide there was arrest in S and G2/M phases of the cell cycle (Fig. 4Ai and Fig. S6Ai) and a decrease in DNA synthesis measured by EdU incorporation (Fig. 4B, first column and Fig. S6Ai). However, inhibition of miR-34c with an anti-miR prevented S-phase arrest and lead to increased DNA synthesis and a subsequent G2/M arrest (Fig. 4 Aii and B, second column, and Fig. S6Aiii). Importantly when miR-34c was inhibited and Myc levels were reduced using siRNA this phenotype was reversed (Fig. 4 Aiii and B, third column and Fig. S6Aiv). The block observed in S-phase in the presence of miR-34c must therefore result from a lack of Myc expression alone. Western blot analysis confirmed a relief of Myc repression and subsequent knockdown in etoposide treated cells (Fig. 4C). These data demonstrate that Myc repression is required for cells to arrest in S-phase following DNA damage, and that a major physiological target of miR-34c is Myc.

Fig. 4.

miR-34c-mediated repression of Myc promotes S-phase arrest and prevents Myc-induced DNA damage. (A) Inhibition of miR-34c following DNA damage prevents S-phase arrest in a Myc-dependent manner. To determine the effect of miR-34c-mediated repression of Myc HEK293 cells were transfected with combinations of siRNA and anti-miRs for 24 h as indicated prior to treatment with 12.5 μM etoposide for 24 h. Data were analyzed as described in Materials and Methods and are shown as mean ± SD (t-test, n = 3, compared to control). (B) Inhibition of miR-34c increases DNA synthesis following DNA damage in a Myc-dependent manner. Quantifiation of EdU incorporation of cells from (A). Values are mean ± SD (t-test, n = 3, compared to DMSO treated, control anti-miR, control siRNA transfected cells). (C) Inhibition of miR-34c leads to Myc-dependent DNA damage and checkpoint activation following DNA damage. Western analysis of lysates generated from cells in Fig. 1A. Data shown are pooled samples from three independent experiments.

It is known that overexpression of Myc can result in DNA damage due to inappropriate replication origin firing resulting in checkpoint activation (3). We therefore sought to determine the consequence of continued Myc expression following DNA damage by inhibiting miR-34c. When miR-34c was inhibited following DNA damage, by use of an anti-miR, there was a Myc-dependent increase in CHK2 (Fig. 4C) phosphorylation, indicative of activation by ATM, and a Myc-dependent increase in H2AX phosphorylation, a marker of DNA damage, over and above that induced by etoposide alone (Fig. 4C). These data show that continued expression of Myc following DNA damage causes inappropriate DNA replication (Fig. 4B), thereby perpetuating the DDR in a feed-forward loop that miR-34c functions to prevent.

Discussion

The DNA damage response requires the concerted action of a number of distinct pathways including those that detect the damage, halt cell cycle progression, and mediate DNA repair. The transcriptional pathways that regulate these processes are fairly well described (25); however, more recently it has become apparent that proteins involved in this response are also posttranscriptionally regulated. For example, it has been shown that following exposure to ionizing radiation or radio-mimetic drugs, miR-34 levels are up-regulated, and expression of one of the members of this family, miR-34a, is required for p53-mediated apoptosis under certain conditions (15–20). Although some of the downstream targets of the miR-34 family have been determined, the biological effects mediated by specific targets are not fully understood. We and others have shown that Myc levels are controlled by miR-34 (21, 26). Although it has been shown previously that levels of Myc are reduced following DNA damage by a number of different mechanisms, including changes in the rate of transcription and protein turnover (27–29), the possible involvement of miRNAs has not been investigated. It was therefore timely to investigate the role of miR-34c in the control of Myc repression following DNA damage and its importance in this process.

Our data show that Myc is repressed by miR-34c following DNA damage via a highly conserved target-site within its 3^′ UTR (Fig. 1). Other members of the miR-34 family may repress Myc when overexpressed (Fig. S1E); however, their ability to exert repression via the 3^′ UTR in reporter assays is limited (Fig. S1 B and C). More specifically miR-34b may function in Myc repression since it is expressed on the same primary transcript as miR-34c; these two microRNAs are coordinately regulated (17, 30) and our anti-miR directed against miR-34c could potentially cross-react with miR-34b; however, this has little impact on our findings since miR-34b is also controlled by p38 MAPK (Fig. S5A). Conversely it is unlikely that miR-34a represses Myc under the conditions we have tested since in p53 -/- MEFs Myc is repressed in a miR-34c-depedendent manner despite no change in miR-34a levels (Fig. 2), and in H1299 cells following p53 induction miR-34a is induced 10-fold (Fig. S3Biii) yet the wild-type c-myc 3^′ UTR reporter construct is not repressed (Fig. S3Bii). Taken together our data suggest that miR-34c represses Myc following DNA damage but that miR-34b may play a role.

Inhibition of miR-34c relieves repression of endogenous Myc following DNA damage in HEK293 cells (Fig. 1E); however, this is not complete. In p53 -/- MEFs inhibition of miR-34c appears to completely prevent Myc repression (Fig. 2B). These data suggest that in HEK293 cells other factors contribute to repression of Myc translation such as CPEB (cytoplasmic polyadenylation element binding protein), which regulates Myc translation during senescence in a p53-depdendent manner (31).

We find that although the induction of miR-34c is clearly influenced by p53 (Fig. S3), in cells which lack p53 a parallel pathway exists to induce miR-34c and repress Myc following DNA damage (Fig. 2). This pathway involves signalling through p38 MAPK to MK2. Inhibition of p38 MAPK prevents miR-34c induction and Myc repression in response to DNA damage in HEK293 cells (Fig. 3B) whereas MK2 only appears to be fully required for miR-34c induction in p53-/- MEFs (Fig. 3 C and D). Furthermore, induction of miR-34c can be restored following p38 MAPK inhibition by introduction of a constitutively active form of MK2 (Fig. 3D). To eliminate the possibility that miR-34c directly impinges on p38 MAPK signalling, we measured MK2 phosphorylation following miR-34c inhibition and see no effect (Fig. S5B) demonstrating that miR-34c per se does not affect p38 MAPK activity. Interestingly p38 MAPK-mediated induction of miR-34 may be a conserved mechanism since in Caenorhabditis elegans miR-34 induction following ionizing radiation is the same in the presence or absence of p53/cep-1 (32) and C. elegans possess proteins with homology to both p38 MAPK and MK2.

Here we show that a major biological function of miR-34c following DNA damage is to repress Myc. This repression mediates S-phase arrest thus preventing DNA replication (Fig. 4 A and B), and it is probable that this is one mechanism by which MK2 performs its checkpoint function. When miR-34c repression of Myc is inhibited cells undergo enhanced DNA damage, in addition to that induced by etoposide alone (Fig. 4C), likely due to inappropriate Myc-dependent DNA replication. Taken together these data suggest that miR-34c serves to remove Myc following DNA damage downstream of p38 MAPK/MK2 to prevent replication stress which may otherwise lead to genomic instability. Since miR-34c-mediated repression of Myc is downstream of p38 MAPK (Fig. 3), a general stress response kinase, this mechanism may prevent Myc-dependent replication stress in response to a diverse range of cellular stresses. It was recently proposed that facilitating replication under stress conditions may be an oncogenic function of Myc (33). Therefore, epigenetic silencing of miR-34b/c, frequently observed in cancers (26), could lead to aberrant Myc-induced replication and transformation.

Materials and Methods

Cell Culture and Drug Treatments.

HEK293, SaOS2, p53 -/- MEFs, and HeLa cells were grown under standard conditions. H1229 p53 inducible cells were grown as described previously (34, 35). Cells were treated with etoposide, adriamycin (doxorubicin), or pretreated for 30 min with SB203580 (25 μM), PD098059 (50 μM), or LY294002 (25 μM) (Sigma-Aldrich) as indicated.

Transient Transfections and Luciferase Assays.

DNA transfections were performed using the calcium phosphate method. RNA and DNA cotransfections were performed using Lipofectamine 2000 (Invitrogen). MicroRNA mimic and 2^′ O-methyl-oligonucleotide transfections were performed using Lipofectamine RNAiMax (Invitrogen). Luciferase assays were preformed as described previously (21).

Plasmids, siRNAs, MicroRNA Mimics and 2^′O-methyl Oligonucleotides.

pRLSVM3^′, pRLSVM3^′mut have been described previously (21). pcDNA-MK2Wt-Myc, pcDNA-MK2K76R, and pcDNA-MK2EE were a kind gift from Mathias Gaestel. siRNAs to p53 (20 nM final concentration) and c-Myc (20 nM) were purchased as smart-pool ON-TARGET plus siRNAs from Dharmacon. For oligonucloetide sequences, see SI Materials and Methods.

MicroRNA Northern Blots.

Total RNA was extracted from cells using Tri-reagent (Sigma-Aldrich) according to the manufacturer’s instructions, except that an equal volume of isopropanol was used for the precipitation step. For full details of microRNA northern, see SI Materials and Methods.

Northern Blots.

Norther blots were performed as described previously (21).

Quantitative PCR for MicroRNA.

One hundred ng (nano gram) of total RNA was reverse transcribed and amplified using the Taqman microRNA assay kit (Applied Biosystems) according to the manufacturer’s instructions in a Stratagene Mx3005P cycler. Ct (threshold cycle) values were calculated and data expressed as 2-^ΔCt′ (36). qPCR for pri-miR-34b/c was performed as described previously (17).

Cell Cycle Analysis.

Cells were treated as indicated, 60 min prior to harvest were labeled with 20 μM EdU, harvested and processed using the Click-iT EdU Alexa Fluor 488 Flow Cytometry assay kit (Invitrogen), and stained for DNA content with 25 μg/mL propidium iodide (Sigma-Aldrich). Samples were subject to cell cycle analysis on a BD FACSAria special order system. Data were analyzed using the WIN-MDI and Cylchred software.