miR-192 Regulates Dihydrofolate Reductase and Cellular Proliferation through the p53-microRNA Circuit Free

MicroRNAs (miRNA) are noncoding, single-stranded RNAs of ∼22 nucleotides, processed from larger pre-miRNAs by the RNase III enzyme, Dicer, into miRNA duplex complexes (1). One strand of this duplex can associate with the RNA-induced silencing complex, with the other strand generally degraded by cellular nucleases (1). The miRNA-RNA-induced silencing complex has been shown to bind to specific mRNA targets resulting in translational repression or cleavage of these mRNA sequences. miRNAs modulate protein expression by promoting RNA degradation, inhibiting mRNA translation, and also affecting transcription. Currently, there are >450 human miRNAs that have been identified and the final number will likely be much higher (2). Although miRNA-mediated mRNA degradation occurs in mammals, in most cases, the effect of miRNA on its targets is thought to use a secondary mechanism of gene regulation via imperfect base-pairing to the 3′-untranslated regions (3′-UTR) of the mRNA targets. This results in the repression of target gene expression post-transcriptionally most likely at the translational level of gene expression (3–5). Such translational regulation provides the cell with a more precise, immediate, and energy-efficient way of controlling expression of a given protein (6). Such regulation can induce rapid changes in protein synthesis without the need for transcriptional activation and subsequent steps in mRNA processing. Additionally, the translational control of gene expression has the advantage of being readily reversible, providing the cell with great flexibility in responding to various cytotoxic stresses (7). Clearly, it is essential to know not only the expression levels of individual mRNAs but also what extent mRNAs are being translated into their corresponding proteins.

Post-transcriptional control mediated by miRNAs has become an area of intense research over the last few years. The notion of miRNAs mediating gene expression at the translational level is based on the study of the first two identified miRNAs, lin-4 and let-7, in Caenorhabditis elegans (8). The lin-4 miRNA attenuates the translation, but not the mRNA level, of two target genes, lin-14 and lin-28, by imperfect base-pairing to complementary sequences in the 3′-UTR of the target mRNAs (3, 8). Although the exact function of many newly discovered miRNAs are just emerging, their ability to regulate cell proliferation and cell death has been shown recently (9). Recent reports have shown that the expression of miRNAs is altered in cancer cells and that some miRNAs can function as translational attenuators (5).

The first indication that miRNAs may function as tumor suppressors was derived from a study by Calin et al., who found that miR-15a and miR-16-1 were commonly deleted in >65% of patients with B-cell chronic lymphocytic leukemia (10). Further, Cimmino et al. showed that miR-15a and miR-16-1 negatively regulated Bcl2, an antiapoptotic protein that is often overexpressed in many tumor types (11). Several expression profiling studies have also reported deregulated miRNAs in colon cancer, breast cancer, and other types of solid tumors (2, 12–14).

miRNAs are shown to be critical in oncogenesis, and alterations in miRNA expression are found to be associated with several human cancers (9, 10, 14–16). miRNAs can act as either oncogenes or tumor suppressor genes based on their mRNA targets (14). We have identified previously that the expression of several miRNAs was altered due to the loss of p53 tumor suppressor gene (12). Subsequently, several groups have reported that miR-34 was directly involved in the p53 tumor suppressor network and regulated directly by p53 (17–19). Based on these results, we reasoned that there should be additional miRNAs that are involved in the p53 tumor suppressor network. By performing an in silico analysis coupled with experimental validations to search for these potential miRNAs, we report here the discovery that the miR-192 promoter contains a conserved p53-binding site and that one of the key miR-192 target genes is dihydrofolate reductase (DHFR), an important target for antifolate-based anticancer chemotherapy such as methotrexate (20). We also experimentally confirmed that the downstream pathway of miR-192 in cell cycle control is mediated through increased p53 expression by the induction of p21. Many studies have indicated that expression of DHFR is regulated at least in part at the translational level (21–26). Recently, Mishra et al. reported that a miR-24 binding site polymorphism in DHFR gene led to methotrexate resistance (27). Translational control provides cells with acute response to growth condition changes and it is readily reversible (28). We provide evidence that miR-192 is another player in the p53-miRNA circuit and contributes to the regulation of cell cycle control and cellular proliferation.

Cell culture and reagents. The human colon cancer cell lines HCT-116 (wt-p53) and HCT-116 (null-p53) were gifts from Prof. Bert Vogelstein (The Johns Hopkins University) and were maintained in McCoy's 5A medium (Life Technologies). The other two human colon cancer cell lines, RKO (wt-p53) and HT-29 (mut-p53), were obtained from the American Type Culture Collection. The HT-29 (mut-p53) cell line has a missense mutation in codon 273 of p53 resulting in an Arg-to-His substitution. RKO (wt-p53) and HT-29 (mut-p53) cells were maintained in Eagle's MEM and Iscove's Modified Dulbecco's Medium at the American Type Culture Collection, respectively. All media were supplemented with 10% dialyzed fetal bovine serum (Hyclone Laboratories). All cell lines were grown at 37°C in a humidified incubator with 5% CO₂. Methotrexate was purchased from Sigma-Aldrich.

Transfections of miRNA and small interfering RNA specific to DHFR. HCT-116 (wt-p53), HCT-116 (null-p53), RKO (wt-p53), and HT-29 cells (2 × 10⁵) were plated in 6-well plates and transfected with 100 nmol/L of either miR-192 or miR-24 precursors or nonspecific control miR (Ambion) after 24 h with Oligofectamine (Invitrogen) according to the manufacturer's instructions. Small interfering RNA (siRNA) specific to DHFR (On-Target plus SMARTpool L-008799-00-0010, human DHFR, NM_000791) was purchased from Dharmacon and transfected with Oligofectamine (Invitrogen) according to the manufacturer's protocols at a final concentration of 100 nmol/L. siRNA specific to DHFR was used as the positive control. miR-24, a recently reported miRNA that also targets DHFR (27), was also used as a positive control.

RNA isolation. Total RNA, including miRNA, was isolated from cell lines by using TRIzol reagent (Invitrogen) according to the manufacturer's instructions at 24 h after transfection.

Real-time quantitative reverse transcription-PCR analysis of miRNA. cDNA synthesis was carried out with the High-Capacity cDNA Synthesis kit (Applied Biosystems) using 10 ng total RNA as template. The miRNA sequence-specific reverse transcription-PCR (RT-PCR) primers for miR-192, miR-24, and endogenous control RNU6B were purchased from Ambion. Real-time quantitative RT-PCR (qRT-PCR) analysis was carried out using Applied Biosystems 7500 Real-time PCR System. The PCR Master Mix containing TaqMan 2× Universal PCR Master Mix (No Amperase UNG), 10× TaqMan assay, and reverse transcription products in 20 μL volume were processed as follows: 95°C for 10 min and then 95°C for 15 s, 60°C for 60 s for up to 40 cycles (n = 3). Signal was collected at the endpoint of every cycle. The gene expression ΔC_T values of miRNAs from each sample were calculated by normalizing with internal control RNU6B and relative quantitation values were plotted.

Real-time qRT-PCR analysis of mRNA expression. cDNA was synthesized with the High-Capacity cDNA Synthesis kit (Applied Biosystems) using 2 μg total RNA as the template and random primers. Real-time qRT-PCR analysis was done on the experimental mRNAs. The PCR primers and probes for DHFR and the internal control gene GAPDH were purchased from Applied Biosystems. qRT-PCR was done on an ABI 7500HT instrument under the following conditions: 50°C for 2 min of reverse transcription, 95°C for 10 min, 95°C for 15 s, 60°C for 1 min for up to 40 cycles (n = 3).

Western immunoblot analysis. At 48 h after transfection with miR-192 or miR-24 precursors or nonspecific control miRNA, the cells were scraped and lysed in radioimmunoprecipitation assay buffer (Sigma). Equal amount of proteins were resolved by SDS-PAGE on 12% gels by the method of Laemmli (29) and transferred to polyvinylidene difluoride membranes (Bio-Rad Laboratories). The membranes were then blocked by 5% nonfat milk in TBS-0.5% Tween 20 at room temperature for 1 h. The primary antibodies used for the analysis included mouse anti-DHFR monoclonal antibody (mAb; 1:250; BD Biosciences), mouse anti-p53 mAb (1:1,000; DO-1), mouse anti-p21mAb (1:1,000; F-5), and mouse anti-α-tubulin mAb (1:1,000; TU-02) purchased from Santa Cruz Biotechnology. Horseradish peroxidase-conjugated antibodies against mouse or rabbit (1:1,000; Santa Cruz Biotechnology) were used as the secondary antibodies. Protein bands were visualized with a chemiluminescence detection system using the Super Signal substrate.

Methotrexate cytotoxicity. HCT-116 (wt-p53) cells were plated in 96-well plates at 1 × 10³ per well in triplicate. They were transfected with miR-192 precursor, nonspecific control miRNA, or siRNA against DHFR in 100 μL medium. Twenty-four hours later, methotrexate in 100 μL medium (final concentration of 25 nmol/L) was added and incubated for 72 h. WST-1 (10 μL; Roche Applied Science) was added to each well. After 2 h incubation, absorbance was measured at 450 and 630 nm, respectively (n = 3). Nonspecific control miRNA alone was used as a negative control, and siRNA incubation with methotrexate was used as a positive control.

Cell proliferation analysis. HCT-116 (wt-p53), HCT-116 (null-p53), RKO (wt-p53), and HT-29 cells were plated in 96-well plates in triplicate at 1 × 10³ per well after transfection with miR-192 precursor or nonspecific control miRNA. Cells were cultured for 24, 48, 72, and 96 h. The absorbance at 450 and 630 nm was measured after incubation with 10 μL WST-1 for 2 h.

Cell cycle analysis. HCT-116 (wt-p53) and HCT-116 (null-p53) cells were transfected with miR-192 precursor and the nonspecific control miRNA described as above. At 36 h after transfection, cells were harvested and resuspended at 0.5 × 10⁵ to 1 × 10⁵/mL in modified Krishan buffer (17, 18) containing 0.1% sodium citrate and 0.3% NP-40 and kept at 4°C. Before being analyzed by flow cytometry, cells were treated with 0.02 mg/mL RNase H and stained with 0.05 mg/mL propidium iodide (Sigma).

Methotrexate treatment. HCT-116 (wt-p53), HCT-116 (null-p53), RKO (wt-p53), and HT-29 cells were seeded in 6-well plates at a density of 2 × 10⁵ per well and then incubated with or without methotrexate (25 nmol/L). The cells were collected at 24 h after incubation, and total RNA and protein were extracted, respectively. The subsequent real-time qRT-PCR of miRNA and Western immunoblot analysis for p53 and α-tubulin were done as described above.

Chromatin immunoprecipitation and qRT-PCR analysis. To show that p53 protein directly interacts with the miR-192 promoter region, we performed chromatin immunoprecipitation-qRT-PCR analysis using p21, a known cell cycle regulator transcriptionally regulated by p53, as a positive control. Mouse mAb (DO-1) against p53 (Santa Cruz Biotechnology) was used for immunoprecipitation of the p53-binding complex. Nonrelated antibody α-tubulin (TU-02; Santa Cruz Biotechnology) was used as a negative control. Immunoprecipitation was done based on the manufacturer's protocols of Active Motif. The primer sequences for the miR-192 promoter and the p21 promoter are listed as follows:

• miR-192 promoter (forward primer) 5′-AGCACCTCCCATGTCACC-3′ and (reverse primer) 5′-CAAGGCAGAGCCAGAGC-3′ and

• p21 promoter (forward primer) 5′-GCTGGTGGCTATTTTGTCCTTGGGC-3′ and (reverse primer) 5′-CAGAATCTGACTCCCAGCACACACTC-3′.

Conserved p53 binding miR-192 promoter reporter activity assay. Luciferase reporter assay was used to determine the transcriptional activation of conserved p53-binding promoter of miR-192. pGL3-Basic promoterless luciferase reporter plasmid (Promega) was used in this study. Double-stranded DNA oligonucleotides of conserved p53-binding sequence of miR-192 was synthesized and annealed and cloned upstream of firefly luciferase in the pGL3-Basic plasmid (miR-192-pGL3). The p53-binding site oligonucleotide (bold) contains MluI at the 5′-end and BglII sequence (italic) at the 3′-end (5′-ACGCGTCCATGTCACCACCAGGGGTCGCCATGCCTCCTGGCCTTGCCCAGCAGATCT-3′). Control vector and miR-192-pGL3 vector were transfected into both HCT-116 (wt-p53) and HCT-116 (null-p53) cells. To further induce p53 expression, transfected HCT-116 (wt-p53) and HCT-116 (null-p53) cells were also treated with 5 μmol/L 5-fluorouracil for 24 h. The promoter activity of each construct was quantified by dual luciferase assay (Promega) 24 h after transfection. Firefly luciferase activity was normalized with Renilla luciferase internal control under each condition.

Statistical analysis. All experiments were repeated at least twice. Statistical significance was evaluated by Student's t test (two-tailed) comparison between two groups of data. Asterisks indicate significant differences of experimental groups compared with the corresponding control condition. Statistical analysis was done using GraphPad Prism software (GraphPad). Differences were considered statistically significant at P < 0.05.

miRNA regulates the mRNA translation rate by perfect or imperfect base-pairing with the 3′-UTR regions of their targets (1). It has been predicted that one miRNA can potentially regulate translation of up to hundreds of mRNAs (1). This has created a challenge for experimentally validating miRNA-specific targets. Previous studies from our laboratory have discovered that nearly half of the miRNAs promoter regions contain putative p53-binding site(s) (12). To further validate the significance of some of these candidate miRNAs, we took a systematic approach by first analyzing which miRNA may target critical anticancer target genes. We also select the miRNA candidates with high predicted ranking scores of p53-binding promoter. This allows us to focus on miR-192 with DHFR as one of its predictive target, an important anticancer target. DHFR is the key enzyme responsible for intracellular folate metabolism and a target of methotrexate (30). The regulation of DHFR is complex and includes post-transcriptional control by an auto feedback mechanism (31).

Translational regulation of DHFR expression by miR-192. We investigated the roles of miRNAs in regulating the expression of DHFR. Based on the structural analysis of 3′-UTR of the DHFR gene and miRNA target analysis (TargetScan, PicTar, and miRnaDa), we identified several miRNAs that potentially interact with the 3′-UTR region of DHFR mRNA. Bioinformatic analysis of the secondary structure of the 3′-UTR of the DHFR mRNA and miRNA binding sites led us to reduce the candidate miRNAs to a small number. This allows us to efficiently identify miRNAs that involved in regulating key targets like DHFR.

Figure 1A shows the target sequence of 3′-UTR of the DHFR mRNA that interacts with miR-192. To experimentally confirm that the expression of DHFR was regulated by miR-192, a miR-192 precursor was transfected into HCT-116 (wt-p53) cells. A nonspecific miR was used as a negative control. It has been reported that the expression of DHFR is regulated by miR-24 (27). We thus used both a DHFR siRNA and miR-24 as positive controls. Overexpression of miR-192 and miR-24 was confirmed by real-time qRT-PCR analysis using U6 RNA to normalize the expression (Supplementary Data 1). The expression of DHFR protein was analyzed using Western immunoblot analysis and the results are shown in Fig. 1B. Overexpression of miR-192 clearly decreased the expression of DHFR protein (Fig. 1B, lane 4). The potency of miR-192 for decreasing DHFR expression was comparable with miR-24 (Fig. 1B, lane 5). We also analyzed the expression level of DHFR mRNA using real-time qRT-PCR analysis and the results (Fig. 1C) indicated that there was no reduction in DHFR mRNA expression by miR-192 (lane 4) and miR-24 (lane 5). Thus, the suppression of DHFR expression was regulated at the translational level without the degradation of DHFR mRNA. By contrast, the decreased expression of DHFR by siRNA was clearly caused by mRNA degradation (Fig. 1C, lane 3).

miR-192 sensitizes HCT-116 (wt-p53) cells to methotrexate. Because DHFR is a target for methotrexate, the increased expression of miR-192 may also contribute to the sensitivity to methotrexate treatment. With equal molar concentration of methotrexate at 25 nmol/L (IC-10), cell proliferation was reduced by 10% in nonspecific control miR-treated cells (Fig. 2, lane 2). Cell proliferation was reduced by ∼70% in cells transfected with miR-192, showing a synergistic effect in combination with methotrexate (Fig. 2, lane 4). By contrast, cells treated with siRNA against DHFR were inhibited by 55% (Fig. 2, lane 3). The more potent effect of miR-192 plus methotrexate compared with siRNA targeting specifically to DHFR suggests that miR-192 may also target additional mRNA targets through imperfect base-pairing. Future studies are clearly needed to systematically identify additional miR-192 regulated mRNA transcripts.

Effect of overexpression of miR-192 on cellular proliferation. To assess the functional significance of miR-192, we evaluated the effect of miR-192 on cellular proliferation using HCT-116 (wt-p53), HCT-116 (null-p53), RKO (wt-p53), and HT-29 (mut-p53) colon cancer cell lines. A nonspecific miR was used as a negative control. Our results show that the overexpression of miR-192 can suppress cellular proliferation in HCT-116 (wt-p53) cells by >55% (n = 3; Fig. 3A) and RKO (wt-p53) cells by 48% (n = 3; Fig. 3B), with less effect on HCT-116 (null-p53; 15%; n = 3; Fig. 3C) and HT-29 (24%; n = 3; Fig. 3D) cell lines. By contrast, the nonspecific control miR has no effect on cellular proliferation, indicating that this effect caused by miR-192 is highly specific. These results clearly show that the effect of miR-192 on the inhibition of cellular proliferation is more potent in colon cancer cell lines containing wild-type p53 than in the p53-null or mutant p53 cell lines, further indicating that the function of miR-192 depends on the status of p53.

Effect of cell cycle control by miR-192. To determine whether the effect of miR-192 on cellular proliferation was related to cell cycle control, we analyzed the effect of miR-192 on cell cycle control by flow cytometry using HCT-116 (wt-p53) and HCT-116 (null-p53) cells transfected with nonspecific control miR or miR-192. Our results show that miR-192 increases both G₁-S (>2-fold) and G₂-S (>3-fold) ratio in HCT-116 (wt-p53) cells (Fig. 4A). By contrast, this effect has not been observed in HCT-116 (null-p53) cells (Fig. 4B). Thus, the cell cycle analysis data are highly consistent with the cell proliferation results that the function of miR-192 is dependent on the presence of wild-type p53 for cell cycle control.

Effect of miR-192 on cell cycle control genes. To further analyze the cell cycle control genes involved in miR-192 overexpression, we analyzed several cell cycle control genes (p53, p21, and Bax). Figure 5 shows the results of Western immunoblot analysis in HCT-116 (wt-p53) cells (Fig. 5A) and RKO (wt-p53) cells (Fig. 5B). Ectopic expression of miR-192 increased the expression of the p53 protein (Fig. 5A, lane 4) by >10-fold and p21 by 10-fold. By contrast, siRNA against DHFR (Fig. 5A, lane 3) did not cause an increase in expression of p53 and p21. The expression of Bax was not altered by miR-192, indicating that miR-192 may not trigger apoptosis. Similar results were obtained for RKO (wt-p53) cells (Fig. 5B, lane 3, miR-192; lane 1, nonspecific miR; and lane 2, siRNA of DHFR). It has been well characterized that the induction of the p53-dependent cell cycle checkpoint control gene p21 is the key to trigger cell cycle arrest at both G₁ and G₂ phases (32, 33). Thus, our results further confirm the notion that the function of miR-192 is clearly dependent on the status of wild-type p53.

miR-192 expression was dependent on p53. With ectopic overexpression of miR-192 by transfection, both HCT-116 (wt-p53) and RKO (wt-p53) cells undergo cell cycle arrest at the G₂ phase leading to decreased cellular proliferation. Bioinformatic analysis also reveals that there is a putative p53-binding site in the miR-192 promoter region. To confirm the direct regulatory relationship with p53, we performed the following experiments. First, the expression of the p53 protein was induced by treatment with methotrexate. The induction of p53 protein expression by methotrexate treatment in HCT-116 (wt-p53) and RKO (wt-p53) cells (Fig. 6A) caused a significant increase of miR-192 expression (Fig. 6B). By contrast, methotrexate treatment in HCT-116 (null-p53) and HT-29 (mut-p53) cells did not cause any change in the expression of miR-192 (Fig. 6B). Furthermore, the expression level of miR-192 in HCT-116 (wt-p53) cells was ∼4-fold higher than HCT-116 (null-p53) cells (Supplementary Data 4). These results suggest that the endogenous expression of miR-192 depends on the wild-type p53 after genotoxic stress by methotrexate treatment. These findings combined with the data in Fig. 5, which shows that p53 is induced by ectopic expression of miR-192, indicate that there might be a positive feedback loop between p53 and miR-192 to ensure proper cell cycle control. miR-192 may down-regulate some key cell cycle-related genes to trigger p53 induction. Further studies are currently under way to define the detailed mechanism of this positive feedback loop between miR-192 and p53.

We have predicted in our previous study that the promoter site of miR-192 contains a well-conserved p53-binding sequence (12). The binding sequence is 5′-CGCCATGCCT…GGCCTTGCCC-3′ with a 3-bp gap with a ranking site score of 90 based on TFBS (34). Our results are subsequently confirmed by a recent report using a similar type of analysis, which shows that the promoter of miR-192 contains a p53-binding site (35). To experimentally confirm a direct interaction between the p53 protein and the miR-192 promoter, we used chromatin immunoprecipitation-qRT-PCR analysis to isolate p53-bound chromosome DNA. The isolated p53-specific binding DNA was PCR amplified using primers that span the predicted p53-binding sites of the miR-192 promoter or the positive control p21 promoter transcriptionally regulated by p53 protein. We show that the p53 protein directly interacts with the miR-192 promoter based on chromatin immunoprecipitation-qRT-PCR analysis with a 4-fold enriched signal with p53-specific mAb compared with the nonspecific antibody control DNA (Supplementary Data 2A and B). These results validate our bioinformatic prediction of the existence of a conserved p53-binding site at the promoter region of miR-192 (12). We further showed directly that the conserved p53-binding site at the promoter region of miR-192 can activate luciferase expression only in HCT-116 (wt-p53) cells. The activation was further enhanced by induced p53 expression in HCT (wt-p53) cells treated with 5-fluorouracil. By contrast, the induction of luciferase activity was completely absent from the HCT-116 (null-p53) cells (Supplementary Data 3). This suggests that miR-192, like miR-34, is another miRNA that is involved in the p53 tumor suppressor network. It is well established that p53 is one of the most frequently altered tumor suppressor genes in colorectal cancer (36). We hypothesize that the potential function of multiple miRNAs involved in p53 tumor suppressor network is to provide the p53 with greater flexibility in rapidly responding to different growth condition changes perhaps by having unique miRNAs (e.g., miR-34 and miR-192) mediate the regulation of the key mRNA targets. We also discovered recently that the expression of miR-192 was decreased in a panel of colorectal tumor specimens compared with the adjacent normal tissues (data not shown). This is consistent with a recent report by Schetter et al., which showed that miR-192 was one of the miRNAs with reduced expression in a large cohort of colon cancer patient samples, further supporting the potential effect and clinical relevance of miR-192 in colon cancer (37). We speculate that the decrease or loss of the suppressive function of miR-192 in colon cancer may be an important factor related to cell cycle control and chemosensitivity to antifolate-based therapy.

In conclusion, our study provides evidence that miR-192 is another candidate miRNA that is directly involved in the regulation of a key anticancer target DHFR. The expression and function of the miR-192 is largely dependent on the presence of functional wild-type p53. This raises the potential of using miR-192 as a novel therapeutic option for treating cancer via an effective delivery system either alone or in combination with antifolate compounds. Further studies are needed to identify additional miR-192-mediated targets and its relationship with other miRNAs involved in the same pathway.

No potential conflicts of interest were disclosed.