Cell proliferation was inhibited following forced over-expression of miR-30a in the ovary cancer cell line A2780DX5 and the gastric cancer cell line SGC7901R. Interestingly, miR-30a targets the DNA replication protein RPA1, hinders the replication of DNA and induces DNA fragmentation. Furthermore, ataxia telangiectasia mutated (ATM) and checkpoint kinase 2 (CHK2) were phosphorylated after DNA damage, which induced p53 expression, thus triggering the S-phase checkpoint, arresting cell cycle progression and ultimately initiating cancer cell apoptosis. Therefore, forced miR-30a over-expression in cancer cells can be a potential way to inhibit tumour development.

INTRODUCTION

Cancer is a disease characterized by uncontrolled cell proliferation [1]. The preferred therapy for tumours is to inhibit cancer cell proliferation. In mammals, a cell multiplies itself by way of mitosis, a process that includes: interphase, during which the cell grows, accumulates nutrients and duplicates its DNA; the mitotic (M) phase, in which the cell splits itself into two distinct cells; and the final phase, where the new cell is completely divided [2].

DNA molecules are replicated in the S (synthesis) phase of interphase at replication origins in DNA. A number of proteins are associated with the initiation of replication and continuation of DNA synthesis [3]. Among them, replication protein A (RPA) helps to unwind the DNA duplex and recruit polymerase α to the replication origins after binding to the single-stranded (ss) DNA; it also stimulates the DNA polymerase to read the templates and incorporate the correct nucleoside triphosphate in the opposite position for new DNA synthesis and reduces misincorporation [4,5]. By recruiting Dna2 to cleave the RPA-bound RNA primer and interacting with proliferating cell nuclear antigen (PCNA), RPA also plays a role in stimulating DNA polymerases δ and ε to replicate DNA for elongation [69]. In addition, RPA also maintains the integrity of the genome by co-operating with Werner syndrome ATP-dependent helicase (WRN) during the final S-phase [10].

RPA defects were reported to cause spontaneous DNA damage and a DNA replication deficiency. Because it is implicated in the recruitment and activation of the kinase ATR (ataxia telangiectasia and Rad3-related protein kinase), the RPA defect phosphorylates ATM (ataxia telangiectasia mutated), transmits checkpoint signals via the phosphorylation of downstream effectors, and slows S-phase progression, thus arresting the cell cycle, suppressing cell mitosis and ultimately causing cell death [10,11].

miRNAs are small non-coding RNA molecules (approximately 22 nucleotides) found in plants and animals. They function in the transcriptional and post-transcriptional regulation of gene expression. By base-pairing with some complementary sequences within mRNA molecules, they usually silence genes via translational repression or target degradation [1214]. miR-30a is a type of miRNA. Using the online software ‘microRNA.org-Targets and Expression’, we predicted that miR-30a targets RPA1 (Figure 1).

RPA1 was identified as a direct target of miR-30a

Figure 1
RPA1 was identified as a direct target of miR-30a

(A) Schematic description of the hypothetical duplexes formed by the interactions between the binding sites in the ABCB1 3′-UTR and miR-30a. The mirSVR score of these hybrids was −0.5334, and the PhastCons score was 0.7092, which are well within the range of genuine miRNA-target pairs. The seed recognition site is denoted. (B) Half-life assay of miR-30a mimics in medium. (C) Firefly luciferase reporters containing wild-type (WT) or mutant (MUT) miR-30a-binding sites in the RPA1 3′-UTR were co-transfected into A2780DX5 cells with the pre-miR-control, pre-miR-30a or pre-Scramble and with anti-Scramble or anti-miR-30a. The luciferase reporter of the mutant of the predicted miR-30a-binding site in the RPA1 3′-UTR was unaffected by miR-30a. In contrast, when the resulting plasmid was transfected in A2780DX5 cells along with a transfection control plasmid (β-gal) and anti-miR-30a, there was a 75% increase in luciferase reporter activity compared with the cells treated with the negative control RNA. However, the mutated luciferase reporter was unaffected by the knockdown of miR-30a. (D) Dose-dependent effect of the miR-30a mimics on the RPA1 level. (E) Pearson's correlation scatter plots of the fold change in the levels of miR-30a and the RPA1 protein in the A2780DX5 cells. There is an inverse correlation between the miR-30a levels and RPA1 protein levels. (F) Dose-dependent effect of miR-30a on the level of the RPA1 mRNA. (G) Pearson's correlation scatter plots of the fold change in the miR-30a and RPA1 mRNA levels in the A2780DX5 cells. No significant difference was observed in the RPA1 mRNA levels of the cells treated with different doses of miR-30a. *P<0.05, n=3.

Figure 1
RPA1 was identified as a direct target of miR-30a

(A) Schematic description of the hypothetical duplexes formed by the interactions between the binding sites in the ABCB1 3′-UTR and miR-30a. The mirSVR score of these hybrids was −0.5334, and the PhastCons score was 0.7092, which are well within the range of genuine miRNA-target pairs. The seed recognition site is denoted. (B) Half-life assay of miR-30a mimics in medium. (C) Firefly luciferase reporters containing wild-type (WT) or mutant (MUT) miR-30a-binding sites in the RPA1 3′-UTR were co-transfected into A2780DX5 cells with the pre-miR-control, pre-miR-30a or pre-Scramble and with anti-Scramble or anti-miR-30a. The luciferase reporter of the mutant of the predicted miR-30a-binding site in the RPA1 3′-UTR was unaffected by miR-30a. In contrast, when the resulting plasmid was transfected in A2780DX5 cells along with a transfection control plasmid (β-gal) and anti-miR-30a, there was a 75% increase in luciferase reporter activity compared with the cells treated with the negative control RNA. However, the mutated luciferase reporter was unaffected by the knockdown of miR-30a. (D) Dose-dependent effect of the miR-30a mimics on the RPA1 level. (E) Pearson's correlation scatter plots of the fold change in the levels of miR-30a and the RPA1 protein in the A2780DX5 cells. There is an inverse correlation between the miR-30a levels and RPA1 protein levels. (F) Dose-dependent effect of miR-30a on the level of the RPA1 mRNA. (G) Pearson's correlation scatter plots of the fold change in the miR-30a and RPA1 mRNA levels in the A2780DX5 cells. No significant difference was observed in the RPA1 mRNA levels of the cells treated with different doses of miR-30a. *P<0.05, n=3.

With a view to restrain the malignant proliferation of cancer cells, we tried forcing miR-30a over-expression in the SGC7091R gastric cell line and A2780DX5 ovarian cancer cell line to decrease RPA1 expression and inhibit DNA replication and cell proliferation.

MATERIALS AND METHODS

Materials

The antibodies against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (sc-32233), RPA1 (sc-46504), p-ATM (sc-29761), phospho-checkpoint kinase 2 (p-CHK2) (sc-29271), phospho-cyclin-dependent kinase 2 (p-CDK2) (Thr160) (sc-101656), cyclin A2 (sc-46B11), p-Akt (sc-1619) and IGF1R (sc-113594) were obtained from Santa Cruz Biotechnology. The antibody against PI3K (GW21071) was purchased from Sigma–Aldrich. The A2780DX5 ovarian cancer cell line and SGC7091R gastric cancer cell line were obtained from KeyGEN Corp. Double-stranded siRNA sequences targeting the RPA1 mRNA (E-015749-00-0005) and their non-specific duplexes (D-001810-10-20) were obtained from Dharmacon.

Cell culture

The A2780DX5 and SGC7091R cells were cultured in RPMI 1640 medium supplemented with 10% FBS, 100 units/ml penicillin and 100 mg/ml streptomycin at 37°C and ventilated with 5% CO2. After 12 h, the cells were divided into three groups. Those in the first group were cultured with normal medium and were established as the control group, those in the second group were transfected with 100 ng/ml of the miR-30a mimics and those in the third group were transfected with 100 ng/ml of a mixture of pre-miR-30a and its antagonistic RNA (the ratio of pre-miR-30a to the antagonistic RNA was 1:1). After an additional 6 h of incubation, all of the cells were harvested for use in the following assays: DNA replication, senescence, comet assay, apoptosis assays and Western blotting to quantify RPA1 and p53 expression and ATM–CHK2 phosphorylation.

Dose-dependent effects of the miR-30a mimics and half-life assay

The dose-dependent effect of the miR-30a mimics on RPA1 and their half-lives were analysed with Western blotting and RT-qPCR (quantitative reverse transcription-PCR), respectively. The miR-30a mimic concentration (100 ng/ml) that was sufficient to inhibit RPA1 was used in the subsequent experiments.

Luciferase reporter assay

To confirm the predicted target and test the direct binding of the miRNA, the entire 3′-UTR of human RPA1 was inserted into the p-MIR reporter plasmid. The specificity of binding was determined using an equivalent luciferase reporter, which was into inserted a mutant RPA1 3′-UTR, and the original sequences that interacted with the miRNA seed sequence were mutated from GAUGUUUAC to CUACAAAUG. The A2780DX5 cells were group cultured and transfected with the firefly luciferase reporter plasmid, β-galactosidase (β-gal) expression plasmid, and equal amounts of pre-miR-30a, anti-miR-30a or the scrambled negative control RNA. At 24 h later, the cells were assayed using a luciferase assay kit.

Construction of the plasmid containing the RPA1 cDNA lacking the 3′-UTR and transfection

To verify that RPA1 is indeed regulated by miR-30a via complementary binding to the 3′-UTR, plasmids containing the RPA1 complementary DNA (cDNA) lacking the 3′-UTR (miR-30a-resistant) were constructed and transfected into untreated A2870DX5 cells using Lipofectamine 2000. Specifically, the RPA1 cDNA was cloned into the pGEM3Zf (−) vector using the EcoRI sites. An XbaI fragment containing the 3′-UTR-free RPA1 cDNA was excised and blunt-ended using the DNAPolI Klenow fragment (GE Healthcare), according to the manufacturer's instructions. The fragment was cloned into the HpaI site of the pRUFneo retroviral expression vector21 and checked for correct orientation. The pRUFneo (RPA1 lacking 3′-UTR) construct was expanded in Escherichia coli DH10 cells and the plasmids were purified using a midi-prep kit (Qiagen).

RPA1 RNAi

To verify that DNA replication was inhibited by the reduced RPA1 expression and that the RPA1 deficiency could hinder DNA unwinding during synthesis, siRNA duplexes were designed to target the coding region of the RPA1 mRNA at nucleotides 424–443 with the sequences 5′-GUCAGCUGAAGCAGUUGG dTdT-3′ and 3′-dTdTCAGUCGACUUCGUCAACC-5′. The DNA oligonucleotides were synthesized and inserted into a BamHI/HindIII-linearized pSilencer 2.1-U6 hygro vector according to the manufacturer's instructions. The sequence of the insertion that expressed the hairpin siRNA is 5′-AACACUCUAUCCUCUUUCAUG-3′. A phosphorothioate-modified antisense oligonucleotide (the sequence was 5′-GCTGAGAGTGTAGGATGTTTACA-3′) bracketing the start codon of the RPA1 mRNA was inserted into the corresponding sites of the vector and served as the negative control plasmid.

The A2780DX5 cells (6×105 cells in 200 μl of culture medium) were independently co-transfected with the RPA1 siRNA expression plasmid (100 ng/ml, Genepharm) or the negative control plasmid (Genepharm) and the same quantities of the Renilla luciferase reporter vector, pRL-TK and pGL3-GFPf. At 48 h later, the cells were lysed and analysed for their levels of firefly and Renilla luciferase activity. Parallel cells were submitted to RNA extraction, RPA1 mRNA RT-qPCR and an analysis of the protein levels and phosphorylation of ATM and CHK2. The comet assay and levels of DNA replication, senescence and apoptosis were also determined individually, using the cells described above.

Cell cycle and DNA replication assay

The cells were fixed with 2.5% glutaraldehyde for 1 h and then permeabilized with 1% Triton-X 100. At 1 h later, the cells were concentrated and stained with propidium iodide (PI) for 15 min in the dark at room temperature, and then submitted to the FACScan flow cytometer. The phycoerythrin emission signal detector FL3 (excitation 585 nm, red) was used to analyse the cell cycle. In the histogram of the result, the areas under the curve were quantified as the diploid (2n, left) and tetraploid (4n, right) cells, which represent the cells in interphase and those after DNA replication, respectively.

Cell senescence detection

After removing the medium and washing with PBS, the cells were fixed for 1 h with the fixative solution provided by the Senescence Detection Kit (BioVision, K320-250) and then stained overnight. After a PBS wash, the cells were observed under a microscope to analyse the development of the blue colour.

Cell proliferation assay

The cells were cultured in the medium provided with the EDU Kit (C00031, RIBOBIO) for 2 h, washed with PBS and fixed with 4% formaldehyde. At 30 min later, the cells were permeabilized with 0.5% Triton X-100. After a PBS wash, the cells were stained with Apollo solution for 30 min in the dark at room temperature, and then stained with a Hoechst 33342 solution. After a 30 min incubation, the solution was discarded and the cells were washed with PBS. Then, the cells were observed under a confocal microscope.

Preparation of replication intermediate DNA for electron microscopic examination

To know whether the RPA1 inhibition by the over-expression of miR-30a or the RPA1 RNAi could hinder DNA replication, total DNA was extracted from the A2780DX5 cells. In brief, the cells were harvested and cross-linked with trioxsalen (10 mg/ml, Sigma–Aldrich) in TMP buffer (10 mM Tris/HCl, pH 8.0, 0.5 mM EDTA and 25 mM NaCl). Then, the total DNA was isolated and digested with BamHI; meanwhile, the RNAs were eliminated by RNase digestion. Replication intermediates were enriched with benzoylated naphthoylated diethylaminoethyl (BND) cellulose and spread on a monolayer of benzalkonium chloride (BAC) on the surface of bidistilled water for electron microscopic examination.

Electron microscopy

Electron micrographs were taken at ×100000 magnification and an accelerating voltage of 120 kV with an FEI Tecnai 20.

Cell comet assay

To certify that the forced over-expression of miR-30a can cause DNA strand breaks, parallel group of cells were suspended in a molten and ethidium bromide-stained low-melting-point agarose (Trevigen) a ratio of 1:10 (v/v) at 37°C. Then, the mono-suspension was cast on a microscope slide. A glass cover slip was held at an angle and the mono-suspension applied to the point of contact between the coverslip and the slide. As the coverslip was lowered on to the slide, the molten agarose was spread to form a thin layer. The agarose was gelled at 4°C and then the slides plated with cells were placed in a horizontal electrophoresis tank containing electrophoresis solution [300 mM sodium acetate (Merck) and 10 mM Tris/HCl (1st BASE), pH 8.5], and electrophoresis was performed at a constant 25 V for 20 min. Comet images of the cells were captured under a fluorescence microscope (BZ-9000, Keyence). The extent of DNA damage was expressed as a measure of the amount of DNA in the tail.

pVenus-IGF1R plasmid construction and transfection

miR-30a was reported to cause DNA damage through the Akt signalling pathway by targeting IGF1R [22]. To exclude the possibility that the IGF1R deficiency caused DNA damage, we transfected the SGC7901R cells with pVenus-IGF1R plasmids and gradient doses of the miR-30a mimics using Lipofectamine 2000 and then observed the effects on the DNA and the cells. The pVenus-IGF1R plasmids used in the present study were purchased from GenePharma Inc. and were constructed by ligating the cDNA of IGFR to eukaryotic expression vector pVenus using HindIII and KpnI. The primers used were: forward 5′-ATGAAGTCTGGCTCCGGAG-3′ and reverse 5′-TCAGCA-GGTCGAAGACT-3′.

To offset the inhibitory effect of miR-30a on IGF1R, a balance between the miR-30a mimics and pVenus-IGF1R plasmids was established by Western blotting. The doses that depleted RPA1 but maintained IG1FR expression (120 ng/ml for the miR-30a mimic and 200 ng/ml for the pVenus-IGF1R plasmid) were used in the IGF1R deficiency-associated pathway exclusion test.

Immunohistochemistry for observing the γH2AX foci

Upon exposure to damaging agents, the DNA can break and the cell can rapidly activate kinases to phosphorylate histone H2AX and repair the damages. Hence, the γH2AX foci are used to identify DNA strand breaks [27].

To observe the γH2AX foci, cells grown on coverslips were fixed for 20 min in freshly prepared 2% paraformaldehyde and then incubated with mouse monoclonal antibodies against γH2AX (Upstate, 1:500 dilution). The cells were viewed using a Zeiss epifluorescence microscope and the images were analysed to determine the number of foci/nucleus. The experiments were repeated three times and independent results for each sample were plotted.

Western blotting

Cells were lysed in a radio immunoprecipitation assay (RIPA) buffer; the protein concentration was determined using a BCA kit (Thermo Company). Equal amounts (60 μg) of cell lysates were resolved by SDS/PAGE and transferred on to PVDF membranes (GE Healthcare). The PVDF membranes were incubated in TBS for 1 h and then immunoblotted with antibodies. The bound antibodies were detected by enhanced chemiluminescence (ECL; Pierce Biotechnology). Equal protein loading was detected by blotting the same samples with an antibody against GAPDH.

Real-time PCR

Total RNA was extracted from the cells using TRIzol (Invitrogen, 15596-026) according to the manufacturer's protocol. After the concentrations were determined, real-time PCR was performed in triplicate using a TaqMan PCR kit (TaqMan, N8080228) on an Applied Biosystems 7500 Sequence Detection System. The primers used were: forward 5′-CGGGCAGCCGCAAGTAGCTC-3′ and reverse 5′-AGGGTGCCTTTCGAGAAA-3′.

Apoptosis assay

Cell apoptosis was determined using an annexin V–FITC apoptosis detection kit (Becton Dickinson). Briefly, 2.0×105 cells were suspended in 0.5 ml of binding buffer and incubated with annexin V–FITC and PI for 10 min in the dark at room temperature. A FACScan flow cytometer (Becton Dickinson), which was equipped with a FITC signal detector FL1 (excitation 488 nm, green) and a phycoerythrin emission signal detector FL3 (excitation 585 nm, red), was used to analyse cell apoptosis. The results were calculated using the CellQuest™ Pro software (Becton Dickinson) to obtain the percentage of apoptotic cells of the total number of cells.

Statistical analysis

All data were presented as the means±S.E.M. FOR three or more independent experiments, and the differences were considered statistically significant at P<0.05 using a Student's t test.

RESULTS

RPA1 can be identified as a direct target of miR-30a

Using the computational algorithms PicTar, TargetScan and miRanda, more than 30 miRNAs were predicted to target the RPA1 mRNA, including miR-1, miR-96, miR-495, miR-103, miR-186 and miR-429. Among them, miR-30a was found to target a site in the 3′-UTR of the RPA1 mRNA sequence. The mirSVR score of the hybrids is −0.5334 and the PhastCons score is 0.7092, which are well within the range of genuine miRNA-target pairs, suggesting that miR-30a can potentially down-regulate RPA1 expression [15]. Perfect base-pairings occurred between the seed region (the core sequence that encompasses the first two to ten bases of the mature miRNA) and the cognate targets. Moreover, mutation of the predicted miR-30a-binding site in the RPA1 3′-UTR resulted in no change in the expression of a luciferase reporter following over-expression of miR-30a (Figure 1C). In contrast, when the resulting plasmid was transfected into A2780DX5 cells along with a transfection control plasmid (β-gal) and anti-miR-30a, the knockdown of miR-30a resulted in a 75% increase in luciferase reporter activity compared with the cells treated with the negative control RNA (Figure 1C). miRNAs are generally thought to have expression patterns that are opposite to those of their targets [1618]. Using Western blotting, RT-qPCR, and Pearson's correlation scatter plots, an inverse correlation between the miR-30a levels and RPA1 protein levels (r=−0.9443 and −0.2101) (Figures 1D and 1E), but not mRNA levels, was observed in the A2780DX5 multidrug-resistant cell line (Figures 1D–1G) within the half-life time. These findings suggest that the binding sites strongly contribute to the miRNA–mRNA interaction and mediate the post-transcriptional repression of RPA1 translation.

miR-30a can inhibit cancer cell proliferation and induce apoptosis

As shown in Figures 2(A)–2(C), after the transfection of miR-30a into the A2780DX5 and SGC7091R cells, the rate of cell proliferation (shown as red spots in Figure 2B) decreased to 26%, resulting in 50% fewer proliferating cells than in the controls. The cell renewal defect allowed the miR-30a-transfected cells to age (stained green); 42% of the cells were positively stained by EDU, which was 6-fold higher than the control (Figures 2D and 2E). Furthermore, with longer culture, the aged cells died more frequently; approximately 22% of the A2780DX5 and SGC7091R cells perished in 10 days, which is 3-fold more cells than the cells that were not transfected with miR-30a (Figures 2F and 2G). In contrast with the antisense RNA, the miR-30a-transfected cells resumed proliferating, and their aging and apoptosis were also reversed.

miR-30a inhibited the cancer cell proliferation and induced cell apoptosis

Figure 2
miR-30a inhibited the cancer cell proliferation and induced cell apoptosis

(A) In the cells transfected with the miR-30a mimic, the PCR analysis showed that the miR-30a content in the A2780DX5 and SGC7091R cells was significantly increased, nearly 7-fold, compared with the control. (B and C) The EDU assay indicates that the cell proliferation (red) rates decreased to 26 and 24% for the A2780DX5 and SGC7901R cell lines, respectively, which was a nearly 3-fold decrease compared with the miR-30a-free control. Antagonizing miR-30a with the antisense RNA allowed cell proliferation to resume. (D and E) miR-30a over-expression inhibited DNA replication and the remaining cells aged (stained green); approximately 42% of the cells were stained with Senescence Detection Kit, 6-fold more than the number of control or miR-30a-antagonized cells. (F and G) Over time, flow cytometry indicated that the number of apoptotic cells increased. After 10 days of culture, 22–24% of the miR-30a-transfected cells died, a 3-fold increase compared with the miR-30a-free medium cultured cells. P<0.05, n=3.

Figure 2
miR-30a inhibited the cancer cell proliferation and induced cell apoptosis

(A) In the cells transfected with the miR-30a mimic, the PCR analysis showed that the miR-30a content in the A2780DX5 and SGC7091R cells was significantly increased, nearly 7-fold, compared with the control. (B and C) The EDU assay indicates that the cell proliferation (red) rates decreased to 26 and 24% for the A2780DX5 and SGC7901R cell lines, respectively, which was a nearly 3-fold decrease compared with the miR-30a-free control. Antagonizing miR-30a with the antisense RNA allowed cell proliferation to resume. (D and E) miR-30a over-expression inhibited DNA replication and the remaining cells aged (stained green); approximately 42% of the cells were stained with Senescence Detection Kit, 6-fold more than the number of control or miR-30a-antagonized cells. (F and G) Over time, flow cytometry indicated that the number of apoptotic cells increased. After 10 days of culture, 22–24% of the miR-30a-transfected cells died, a 3-fold increase compared with the miR-30a-free medium cultured cells. P<0.05, n=3.

miR-30a can inhibit DNA replication by decreasing RPA1 expression, arresting the cell cycle at the G1/S-phase and inducing apoptosis

miR-30a was reported to cause DNA damage through the Akt signalling pathway by targeting insulin-like growth factor 1 receptor (IGF1R) [19]. To exclude the possibility that the IGF1R deficiency caused DNA impairment, we supplied IGF1R to the miR-30a-transfected SGC7091R cells using a vector containing the IGF1R gene, the pVenus-IGF1R plasmid. The result indicates that a 90 ng/ml dose of the miR-30a can cause DNA damage (Figures 3A–3E and 5C-c), cell cycle arrest (Figure 3F) and inhibit cell proliferation without triggering the IGF1R-mediated phosphoinositide 3 kinase (PI3K)/Akt pathway (Akt was phosphorylated normally, as shown in Figures 3C, 3D and 3G). Moreover, RPA1 expression was down-regulated in both the SGC7091R and A2780DX5 cell lines (Figure 4B); the number of 4n-chromosome-possessing cells decreased to 3.6% for A2780DX5 and 4.1% for SGC7091R, which was 3-fold fewer cells than the control (Figures 4C and 4D). Most cells were arrested at G1/S-phase, and no DNA replication fork was detected (Figure 5C-c), whereas it is common in normal or miR-30a offset cells (Figures 5C-a and 5C-b), implying that the DNA duplex did not unwind for the originally scheduled replication or transcription. Moreover, in the miR-30a-over-expressing cells, the DNA molecules were fragmented. As shown in the immunofluorescence analysis in Figure 4(E), γH2AX foci (↑), a DNA break and repair marker, appeared in the miR-30a-over-expressing cells, but it was not observed in the normal untreated cells or the miR-30a counter-balanced cells. A comet assay (Figure 4F) further demonstrates that the cells that were transfected with the miR-30a mimics had trailing comets (the character of a cell with DNA damage), whereas the untreated or miR-30a offset cells were round. Moreover, the phosphorylation of the ATM–CHK2 protein complex, which senses DNA damage, was increased (Figure 4B), and p53 expression was also increased (Figure 4B), implying an S-phase checkpoint trigger and the initiation of cell apoptosis (Figures 4G and 4H).

miR-30a can inhibit DNA replication through a pathway that is distinct from the IGF1R-mediated PI3K/Akt pathway

Figure 3
miR-30a can inhibit DNA replication through a pathway that is distinct from the IGF1R-mediated PI3K/Akt pathway

(A and B) Dose-dependent effects of the miR-30a mimics and pVenus-IGF1R plasmids on the expression of RPA1 and IGF1R. Both RPA1 and IGF1R decrease with the concentration of miR-30a mimics. When the miR-30a mimics reached 90 ng/ml, IGF1R is still expressed, but RPA1 is not. (C and D) Using 200 ng/ml pVenus-IGF1R plasmids and 90 ng/ml miR-30a mimics, RPA1 was inhibited, but IGF1R expression was maintained, which prevented miR-30a from triggering the PI3K/Akt pathway. (E) Immunofluorescence indicates that γH2AX foci (↑) can be observed in the miR-30a-rich cells, suggesting that DNA damage occurred. (F) miR-30a over-expression triggered the S-phase checkpoint and hindered DNA replication, even in the presence of IGF1R; the number of 4n-chromosome-possessing cells decreased. (G) The EDU assay shows that cell proliferation was inhibited after the cells were transfected with the miR-30a mimics, but the pVenus-IGF1R plasmids-alone transfected cells were spared.

Figure 3
miR-30a can inhibit DNA replication through a pathway that is distinct from the IGF1R-mediated PI3K/Akt pathway

(A and B) Dose-dependent effects of the miR-30a mimics and pVenus-IGF1R plasmids on the expression of RPA1 and IGF1R. Both RPA1 and IGF1R decrease with the concentration of miR-30a mimics. When the miR-30a mimics reached 90 ng/ml, IGF1R is still expressed, but RPA1 is not. (C and D) Using 200 ng/ml pVenus-IGF1R plasmids and 90 ng/ml miR-30a mimics, RPA1 was inhibited, but IGF1R expression was maintained, which prevented miR-30a from triggering the PI3K/Akt pathway. (E) Immunofluorescence indicates that γH2AX foci (↑) can be observed in the miR-30a-rich cells, suggesting that DNA damage occurred. (F) miR-30a over-expression triggered the S-phase checkpoint and hindered DNA replication, even in the presence of IGF1R; the number of 4n-chromosome-possessing cells decreased. (G) The EDU assay shows that cell proliferation was inhibited after the cells were transfected with the miR-30a mimics, but the pVenus-IGF1R plasmids-alone transfected cells were spared.

miR-30a decreased RPA1 expression and thus inhibited DNA replication, triggered the S-phase checkpoint and induced cell apoptosis

Figure 4
miR-30a decreased RPA1 expression and thus inhibited DNA replication, triggered the S-phase checkpoint and induced cell apoptosis

(A) RT-qPCR quantified the miR-30a levels in the differently treated cells. (B) The Western blot assay shows that miR-30a over-expression decreased the RPA1 levels and increased the phosphorylation of the S-phase implicated proteins ATM and CHK2 in both the A2780DX5 and SGC7901R cell lines and enhanced the expression of the apoptosis-related protein p53 in the miR-30a-transfected cells. (C and D) Flow cytometry demonstrates that the forced over-expression of miR-30a in the A2780DX5 and SGC7901R cell lines decreased the number of 4n-chromosome-possessing cells to 4%, which is a 3-fold lower than the control; meanwhile, miR-30a over-expression inhibited DNA replication and arrested cells at G1/S-phase. Upon deleting the 3′-UTR from the RPA1 cDNA and transfecting it with the IGF1R plasmids, the number of cells arrested at the G1/S checkpoint was reduced, but the number of dead cells increased when IGF1R plasmids were not transfected. However, if no IGF1R was supplied (fourth panel), the replication of DNAs would still be hard under abundant miR-30a mimics, even the 3′-UTR of RPA1 was depleted, which proved that miR-30a damaged DNA via targeting IGF1R. (E) Immunocytochemistry demonstrates that γH2AX foci (↑) formed after miR-30a was over-expressed. However, when the antisense RNA was transfected, the number of foci decreased, indicating that the antagonist offset the DNA damage effect; miR-30a had no effect on the ability of the cells transfected with plasmids lacking the 3′-UTR of RPA1 to form foci when co-transfected with the pVenus-IGF1R plasmids; but the foci still appeared in the IGF1R-lacking cells under the influence of miR-30a. (F) The comet assay indicates that forced miR-30a over-expression caused DNA fragmentation and trailing at the rear of the cell; nevertheless, the untreated IGF1R-over-expressing and the 3′-UTR-lacking RPA1 plasmid-transfected cells were round. However, the cells lacking the IGF1R plasmid were still trailing, regardless of the 3′-UTR-lacking RPA1 plasmids being transfected in. (G and H) Flow cytometry indicates that over-expression of miR-30a caused apoptosis to levels of the control or the 3′-UTR-lacking RPA1 plasmids transfected. However, if the cells were pre-transfected with the pVenus-IGF1R plasmids, apoptosis of the cells lacking the 3′-UTR RPA1 plasmids decreased to 6%, even in the presence of miR-30a mimics, which was nearly the same as the untreated cells. But apoptosis of the cells carrying 3′-UTR-lacking RPA1 mRNA still reach 17%, higher than that of the untreated control if extra IGF1R was not transfected in. P<0.05, n=3.

Figure 4
miR-30a decreased RPA1 expression and thus inhibited DNA replication, triggered the S-phase checkpoint and induced cell apoptosis

(A) RT-qPCR quantified the miR-30a levels in the differently treated cells. (B) The Western blot assay shows that miR-30a over-expression decreased the RPA1 levels and increased the phosphorylation of the S-phase implicated proteins ATM and CHK2 in both the A2780DX5 and SGC7901R cell lines and enhanced the expression of the apoptosis-related protein p53 in the miR-30a-transfected cells. (C and D) Flow cytometry demonstrates that the forced over-expression of miR-30a in the A2780DX5 and SGC7901R cell lines decreased the number of 4n-chromosome-possessing cells to 4%, which is a 3-fold lower than the control; meanwhile, miR-30a over-expression inhibited DNA replication and arrested cells at G1/S-phase. Upon deleting the 3′-UTR from the RPA1 cDNA and transfecting it with the IGF1R plasmids, the number of cells arrested at the G1/S checkpoint was reduced, but the number of dead cells increased when IGF1R plasmids were not transfected. However, if no IGF1R was supplied (fourth panel), the replication of DNAs would still be hard under abundant miR-30a mimics, even the 3′-UTR of RPA1 was depleted, which proved that miR-30a damaged DNA via targeting IGF1R. (E) Immunocytochemistry demonstrates that γH2AX foci (↑) formed after miR-30a was over-expressed. However, when the antisense RNA was transfected, the number of foci decreased, indicating that the antagonist offset the DNA damage effect; miR-30a had no effect on the ability of the cells transfected with plasmids lacking the 3′-UTR of RPA1 to form foci when co-transfected with the pVenus-IGF1R plasmids; but the foci still appeared in the IGF1R-lacking cells under the influence of miR-30a. (F) The comet assay indicates that forced miR-30a over-expression caused DNA fragmentation and trailing at the rear of the cell; nevertheless, the untreated IGF1R-over-expressing and the 3′-UTR-lacking RPA1 plasmid-transfected cells were round. However, the cells lacking the IGF1R plasmid were still trailing, regardless of the 3′-UTR-lacking RPA1 plasmids being transfected in. (G and H) Flow cytometry indicates that over-expression of miR-30a caused apoptosis to levels of the control or the 3′-UTR-lacking RPA1 plasmids transfected. However, if the cells were pre-transfected with the pVenus-IGF1R plasmids, apoptosis of the cells lacking the 3′-UTR RPA1 plasmids decreased to 6%, even in the presence of miR-30a mimics, which was nearly the same as the untreated cells. But apoptosis of the cells carrying 3′-UTR-lacking RPA1 mRNA still reach 17%, higher than that of the untreated control if extra IGF1R was not transfected in. P<0.05, n=3.

RPA1 RNAi inhibited DNA replication and induced DNA damage and cell apoptosis

Figure 5
RPA1 RNAi inhibited DNA replication and induced DNA damage and cell apoptosis

(A) In the cells transfected with the siRNA targeting RPA1, RPA1 expression was decreased, and thus the phosphorylation of the proteins in the S-phase checkpoint-implicated complex, ATM and CHK2, was increased. Meanwhile, the expression of the apoptosis-related protein p53 was increased. (B) Depleting RPA1 expression with an siRNA decreased the number of 4n-chromosome-possessing cells and induced the S-phase checkpoint. (C) RPA1 inhibition hampers DNA unwinding and the formation of the replication fork. The duplex DNA of normal or nonsense RNA-transfected cells was unwound at the replication fork sites (↑). Forced miR-30a over-expression caused DNA fragmentation (△). An RNA anti-miR-30a offset the miR-30a-induced DNA damage. Depleting RPA1 with a siRNA also hampered the unwinding of DNA and fragmented the DNA (△). (D) Upon depleting RPA1 with an siRNA, γH2AX foci (↑) appeared. (E) Cell electrophoresis indicates that depleting RPA1 with an siRNA produced a cell comet (↑), suggesting that a DNA strand break occurred. (F and G) The DNA replication deficiency (F) decreased cell proliferation and (G) promoted aging in the remaining cells (green stained). (H and I) The flow cytometry assay shows that RPA1 depletion caused DNA replication deficiency-induced cell apoptosis. Compared with the control, the death rate of the RPA1-depleted cells reached 23%, which was significantly increased compared with the control (4%). P<0.05, n=3.

Figure 5
RPA1 RNAi inhibited DNA replication and induced DNA damage and cell apoptosis

(A) In the cells transfected with the siRNA targeting RPA1, RPA1 expression was decreased, and thus the phosphorylation of the proteins in the S-phase checkpoint-implicated complex, ATM and CHK2, was increased. Meanwhile, the expression of the apoptosis-related protein p53 was increased. (B) Depleting RPA1 expression with an siRNA decreased the number of 4n-chromosome-possessing cells and induced the S-phase checkpoint. (C) RPA1 inhibition hampers DNA unwinding and the formation of the replication fork. The duplex DNA of normal or nonsense RNA-transfected cells was unwound at the replication fork sites (↑). Forced miR-30a over-expression caused DNA fragmentation (△). An RNA anti-miR-30a offset the miR-30a-induced DNA damage. Depleting RPA1 with a siRNA also hampered the unwinding of DNA and fragmented the DNA (△). (D) Upon depleting RPA1 with an siRNA, γH2AX foci (↑) appeared. (E) Cell electrophoresis indicates that depleting RPA1 with an siRNA produced a cell comet (↑), suggesting that a DNA strand break occurred. (F and G) The DNA replication deficiency (F) decreased cell proliferation and (G) promoted aging in the remaining cells (green stained). (H and I) The flow cytometry assay shows that RPA1 depletion caused DNA replication deficiency-induced cell apoptosis. Compared with the control, the death rate of the RPA1-depleted cells reached 23%, which was significantly increased compared with the control (4%). P<0.05, n=3.

Nevertheless, if the IGF1R-rich cells were pre-transfected with a 3′-UTR-lacking RPA1 plasmid, DNA damage or cycle arrest rarely occurred, even when the miR-30a mimics were transfected. As shown in the fourth panel of Figure 4(E), the number of γH2AX foci in these cells was reduced, and the number of 4n-possessing cells was nearly the same as the untreated cells, suggesting that they were growing as well as the control. However, if IGF1R was not extra-supplied, the 3′-UTR-lacking RPA1-containing cells were still ailing even though the miR-30a mimic was abundant in the cells (the fourth panel in Figures 4C–4G), hinting that miR-30a injured DNAs via the IGF1R-mediated pathway, which was different from RPA1.

RPA1 RNAi confirmed that inhibiting RPA1 can block DNA replication and induce apoptosis

To verify the inhibitory effect of miR-30a on DNA replication, we silenced RPA1 in the SGC7091R cells with siRNA (Figure 5A). The result shown in Figure 5(B) indicates that the number of 4n-chromosome-containing cells decreased after the treatment, implying that the DNA replication was inhibited and the cells were arrested at G1/S-phase. The depletion of RPA1 made it difficult for the cells to unwind the double-stranded DNA. As shown in the TEM images in Figure 5(C-e), the replication fork is not observed in the RPA1 RNAi cells; however, it exists in the untreated (Figure 5C-a) or null-RNA plasmid-transfected cells (Figure 5C-b). As a result, DNA replication is inhibited, accompanied by DNA damage (shown by Figure 5C-e and γH2AX foci in Figure 5D) and cell trailing (Figure 5E). The DNA lesions increased ATM and CHK2 phosphorylation, thus triggering the S-phase checkpoint (Figure 5A). Therefore, fewer cells could generate renascent cells (Figure 5F), and the remaining cells aged (Figure 5G). As expected, the DNA replication deficiency caused by the RPA1 siRNA increased p53 expression (Figure 5A), and the percentage of dead cells increased to 23 from 4% (Figures 5H and 5I).

DISCUSSION

Unlimited cancer cell proliferation is the cause of tumours [1]. Efforts have been made to inhibit the multiplication of malignant cells to identify a cure for cancer.

Mitosis is the main way for eukaryotic cells to proliferate. In this process, the DNA and organelles are replicated and then separated into daughter cells; the hereditary material is thereby passed on to the next generation, and the regenerative cells begin growing [20]. Any deficiency in DNA replication or DNA damage will trigger a checkpoint, delaying the progression of the cell cycle to facilitate DNA repair. If the error is irretrievable, the cells will launch a death programme to eliminate the hazard [21].

RPA plays a role in the replication of DNA. It assists in unwinding the duplexed DNA, recruiting polymerases to incorporate nucleoside triphosphates into the newly synthesized DNA strand and maintaining the genomic integrity [10,22]. An RPA1 defect will cause spontaneous DNA damage and induce a DNA replication checkpoint, thus suppressing mitosis or launching cell death [10,11]. Therefore, decreasing RPA1 expression can be a way to inhibit the malignant proliferation of cancer cells.

In our experiment, we predicted that miR-30a can target RPA1. By forcing its over-expression in cancer cells, we showed that it not only damages the DNA and inhibits cell proliferation by blocking the IGF1R-mediated PI3K/Akt pathway (Figure 3), but also can decrease the expression of the DNA replication protein RPA1 (Figure 4B). When IGF1R was over-expressed in cells, which excluded the effect of IGF1R deficiency on the PI3K/Akt pathway, we observed that RPA1 depletion made it difficult for the cells to unwind the duplex DNA (Figures 5C-c and 5C-e) and replicate the DNA (Figures 4C and 4D), and caused DNA fragmentation (shown in Figures 5C-c and 5C-e). The decreased RPA1 expression increased the phosphorylation of ATM and CHK2, which subsequently triggered the S-phase checkpoint and arrested the cells at G1/S-phase (Figures 4B–4D) [23], ultimately decreasing cell proliferation (Figures 2B and 2C). Without cell renewal, the remaining cells were subjected to aging (Figures 2D and 2E) and the intracellular organelles were damaged. Therefore, hazardous factors, such as oxy-radicals, leaked out from the broken mitochondria or peroxisomes, destroying macromolecules, such as proteins, and ultimately caused cell death [24,25]. DNA fragmentation can also activate forkhead box O3 (FOXO3) to increase the phosphorylation of the ATM–CHK2–p53 complex, launching cell apoptosis and thus directly reducing the number of cancer cells [26]. Therefore, the forced over-expression of miR-30a in cancer cells may be a way to inhibit tumour growth.

AUTHOR CONTRIBUTION

Zhenyou Zou, Mengjie Ni and Jing Zhang designed the study. Yongfeng Chen, Hongyu Ma, Shihan Qian and Longhua Tang performed the TEM. Jiamei Tang, Hailun Yao, Chengbin Zhao and Xiongwen Lu designed all of the plasmids used in the experiments. Hongyang Sun, Jue Qian, Dan Zong, Xiaoting Mao, Xulin Lu and Qun Liu performed the Western blots and flow cytometry. Juping Zen, Hanbing Wu, Zhaosheng Bao and Shudan Lin performed the PCR analysis. Hongyu Sheng, Zhiqiang Chen, Yunlong Li and Yong Liang statically analysed the data.

FUNDING

This work was supported by the Public Welfare Technology Research Grant for Zhejiang Social Development [grant number 2015C33248]; the Taizhou University Research Fund [grant number 0104010004]; the Taizhou University Talent Fostering Fund [grant number 2015PY028]; and the Tumor Institute of Taizhou University [grant number 2012R428024].

Abbreviations

     
  • ATM

    ataxia telangiectasia mutated

  •  
  • CHK2

    checkpoint kinase 2

  •  
  • β-gal

    β-galactosidase

  •  
  • GAPDH

    glyceraldehyde-3-phosphate dehydrogenase

  •  
  • IGF1R

    insulin-like growth factor 1 receptor

  •  
  • PCNA

    proliferating cell nuclear antigen

  •  
  • PI

    propidium iodide

  •  
  • PI3K

    phosphoinositide 3 kinase

  •  
  • RPA

    replication protein A

  •  
  • RT-qPCR

    quantitative reverse transcription-PCR

  •  
  • WRN

    Werner syndrome ATP-dependent helicase

References

References
1
Balmain
A.
Cancer genetics: from Boveri and Mendel to microarrays
Nat. Rev. Cancer
2001
, vol. 
1
 (pg. 
77
-
82
)
[PubMed]
2
Cooper
G.M.
Chapter 14: The Eukaryotic Cell Cycle. The Cell: A Molecular Approach
2000
2nd edn
Washington, DC
ASM Press
3
Berg
J.M.
Tymoczko
J.L.
Stryer
L.
Clarke
N.D.
DNA replication of both strands proceeds rapidly from specific start sites
Biochemistry
2002
New York
W.H. Freeman and Company
4
Tanaka
T.
Nasmyth
K.
Association of RPA with chromosomal replication origins requires an Mcm protein, and is regulated by Rad53, and cyclin- and Dbf4-dependent kinases
EMBO J.
1998
, vol. 
17
 (pg. 
5182
-
5191
)
[PubMed]
5
Zou
L.
Stillman
B.
Assembly of a complex containing Cdc45p, replication protein A, and Mcm2p at replication origins controlled by S-phase cyclin-dependent kinases and Cdc7p-Dbf4p kinase
Mol. Cell. Biol.
2000
, vol. 
20
 (pg. 
3086
-
3096
)
[PubMed]
6
Waga
S.
Bauer
G.
Stillman
B.
Reconstitution of complete SV40 DNA replication with purified replication factors
J. Biol. Chem.
1994
, vol. 
269
 (pg. 
10923
-
10934
)
[PubMed]
7
Waga
S.
Stillman
B.
Anatomy of a DNA replication fork revealed by reconstitution of SV40 DNA replication in vitro
Nature
1994
, vol. 
369
 (pg. 
207
-
212
)
[PubMed]
8
Loor
G.
Zhang
S.J.
Zhang
P.
Toomey
N.L.
Lee
M.Y.
Identification of DNA replication and cell cycle proteins that interact with PCNA
Nucleic Acids Res.
1997
, vol. 
25
 (pg. 
5041
-
5046
)
[PubMed]
9
Dianov
G.L.
Jensen
B.R.
Kenny
M.K.
Bohr
V.A.
Replication protein A stimulates proliferating cell nuclear antigen-dependent repair of abasic sites in DNA by human cell extracts
Biochemistry
1999
, vol. 
38
 (pg. 
11021
-
11025
)
[PubMed]
10
Dodson
G.E.
Shi
Y.
Tibbetts
R.S.
DNA replication defects, spontaneous DNA damage, and ATM-dependent checkpoint activation in replication protein A-deficient cells
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
34010
-
34014
)
[PubMed]
11
You
Z.
Kong
L.
Newport
J.
The role of single-stranded DNA and polymerase α in establishing the ATR, Hus1 DNA replication checkpoint
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
27088
-
27093
)
[PubMed]
12
Chen
K.
Rajewsky
N.
The evolution of gene regulation by transcription factors and microRNAs
Nat. Rev. Genet.
2007
, vol. 
8
 (pg. 
93
-
103
)
[PubMed]
13
Bartel
D.P.
MicroRNAs: target recognition and regulatory functions
Cell
2009
, vol. 
136
 (pg. 
215
-
233
)
[PubMed]
14
Kusenda
B.
Mraz
M.
Mayer
J.
Pospisilova
S.
MicroRNA biogenesis, functionality and cancer relevance
Biomed. Pap. Med. Fac. Univ. Palacky Olomouc Czech Repub.
2006
, vol. 
150
 (pg. 
205
-
215
)
[PubMed]
15
Betel
D.
Koppal
A.
Agius
P.
Sander
C.
Leslie
C.
Comprehensive modeling of microRNA targets predicts functional non-conserved and non-canonical sites
Genome Biol.
2010
, vol. 
11
 pg. 
R90
 
[PubMed]
16
Ambros
V.
The functions of animal microRNAs
Nature
2004
, vol. 
431
 (pg. 
350
-
355
)
[PubMed]
17
Bartel
D.P.
MicroRNAs: genomics, biogenesis, mechanism, and function
Cell
2004
, vol. 
116
 (pg. 
281
-
297
)
[PubMed]
18
He
L.
Hannon
G.J.
MicroRNAs: small RNAs with a big role in gene regulation
Nat. Rev. Genet.
2004
, vol. 
5
 (pg. 
522
-
531
)
[PubMed]
19
Wen
X.P.
Ma
H.L.
Zhao
L.Y.
Zhang
W.
Dang
C.X.
MiR-30a suppresses non-small cell lung cancer progression through AKT signaling pathway by targeting IGF1R
Cell. Mol. Biol. (Noisy-le-Grand)
2015
, vol. 
61
 (pg. 
78
-
85
)
[PubMed]
20
De Souza
C.P.
Osmani
S.A.
Mitosis, not just open or closed
Eukaryot. Cell
2007
, vol. 
6
 (pg. 
1521
-
1527
)
[PubMed]
21
Jiri
B.
Jiri
L.
Chk1 and Chk2 kinases in checkpoint control and cancer
Cancer Cell
2003
, vol. 
3
 (pg. 
421
-
429
)
[PubMed]
22
Dodson
G.E.
Shi
Y.
Tibbetts
R.S.
Expression of mutant RPA in human cancer cells causes telemere shorting
Biosci. Biotechnol. Biochem.
2010
, vol. 
74
 (pg. 
382
-
385
)
[PubMed]
23
Rhind
N.
Russell
P.
The Schizosaccharomyces pombe S-phase checkpoint differentiates between different types of DNA damage
Genetics
1998
, vol. 
149
 (pg. 
1729
-
1737
)
[PubMed]
24
Baker
B.M.
Haynes
C.M.
Mitochondrial protein quality control during biogenesis and aging
Trends Biochem. Sci.
2011
, vol. 
36
 (pg. 
254
-
261
)
[PubMed]
25
Löw
P.
The role of ubiquitin-proteasome system in ageing
Gen. Comp. Endocrinol.
2011
, vol. 
172
 (pg. 
39
-
43
)
[PubMed]
26
Chung
Y.M.
Park
S.H.
Tsai
W.B.
Wang
S.Y.
Ikeda
M.A.
Berek
J.S.
Chen
D.J.
Hu
M.C.
FOXO3 signaling links ATM to the p53 apoptotic pathway following DNA damage
Nat. Commun.
2012
, vol. 
3
 pg. 
1000
 
[PubMed]
27
Kinner
A.
Wu
W.
Staudt
C.
Iliakis
G.
Gamma-H2AX in recognition and signaling of DNA double-strand breaks in the context of chromatin
Nucleic Acids Res.
2008
, vol. 
36
 (pg. 
5678
-
5694
)
[PubMed]

Author notes

1

These authors contributed equally to this study.