MiR-21 Targets 15-PGDH and Promotes Cholangiocarcinoma Growth

Abstract

MicroRNAs (miRs) are a group of small, non-coding RNAs that modulate the translation of genes by binding to specific target sites in the target mRNA. This study investigated the biological function and molecular mechanism of microRNA-21 (miR-21) in human cholangiocarcinoma. In situ hybridization analysis of human cholangiocarcinoma specimens showed increased miR-21 in cholangiocarcinoma tissue compared to the non-cancerous biliary epithelium. Lentiviral transduction of miR-21 enhanced human cholangiocarcinoma cell growth and clonogenic efficiency in vitro, whereas inhibition of miR-21 decreased these parameters. Over-expression of miR-21 also promoted cholangiocarcinoma growth using an in vivo xenograft model system. The NAD⁺-linked 15-hydroxyprostaglandin dehydrogenase (15-PGDH/HPGD), a key enzyme that converts the protumorigenic prostaglandin E₂ (PGE₂) to its biologically inactive metabolite, was identified as a direct target of miR-21 in cholangiocarcinoma cells. In parallel, cyclooxygenase-2 (COX2) over-expression and PGE₂ treatment increased miR-21 levels and enhanced miR-21 promoter activity in human cholangiocarcinoma cells.

INTRODUCTION

Cyclooxygenase (COX)-2-derived prostaglandin E₂ (PGE₂) is the most abundant prostaglandin in inflammation and various human malignancies(1, 2). The tumorigenic actions of PGE₂ are attributable to its modulation of cell proliferation, survival, migration and invasion. The level of PGE₂ in the inflammatory and tumor microenvironments is controlled by the status of PGE₂ synthesis and degradation. Whereas the cyclooxygenases are rate-limiting key enzyme that controls PGE₂ biosynthesis, the NAD⁺-linked 15-hydroxyprostaglandin dehydrogenase (15-PGDH) is a key enzyme that converts PGE₂ to its biologically inactive metabolite, 13, 14-dihydro-15-keto-PGE₂, thus leading to PGE₂ inactivation(1, 3). Consistent with the anti-inflammatory and anti-tumorigenic effect of 15-PGDH, down-regulation of 15-PGDH expression has been observed in several human cancers, including colon, gastric and breast cancer(4–12). It is conceivable that reduced 15-PGDH may lead to PGE₂ accumulation which sustains carcinogenesis and tumor progression. However, the exact mechanisms for 15-PGDH down-regulation in carcinogenesis have not been fully elucidated.

Cholangiocarcinoma is a highly malignant cancer of the biliary tract. It often arises from background conditions that cause long-standing inflammation, injury and reparative biliary epithelial cell proliferation(13, 14). Cholangiocarcinogenesis involves a series of sequential events including chronic inflammation, cholangiocyte proliferation, dysplasia, and ultimately malignant transformation(15). Studies have documented up-regulation and activation of COX-2/PGE₂ signaling in human cholangiocarcinoma and demonstrated an important role of this cascade in regulating cholangiocarcinoma cell growth and apoptosis(16–20).

MicroRNAs (miRNAs) are a group of small, noncoding RNAs that target specific sites in messenger RNAs (mRNAs), leading to translational inhibition or cleavage of target mRNAs. Recent studies have pointed toward a potentially important role of miRNAs in hepatobiliary carcinogenesis(21), although the functions and mechanisms of miRNAs in hepatobiliary cancers remain to be further defined. In the current study, we have identified 15-PGDH as a novel target of miR-21 in cholangiocarcinoma and our results disclose a novel feed-forward loop between COX-2/PGE₂ and miR-21 signaling pathways which is crucial in cholangiocarcinogenesis and tumor progression.

MATERIALS AND METHODS

Materials

Dulbecco’s modified minimum essential medium (DMEM) and fetal bovine serum (FBS) were purchased from Sigma (St. Louis, MO). Opti-MEM reduced serum medium and Lipofectamine^™ 2000 reagent were purchased from Invitrogen (Carlsbad, CA). MiR-21 expression and scramble control lentivirus particles with enhanced green flurescent protein(eGFP) were purchased from GeneCopoeia (Rockville, MD). Rabbit polyclonal Anti-15-PGDH antibody was purchased from Cayman Chemical (Ann Arbor, MI). Mouse monoclonal anti-β-actin antibody was from Sigma; anti-COX-2 antibody was from Cayman Chemical. Synthetic miR-21 mimic and inhibitor RNAs were purchased from Qiagen (Valencia, CA). 15-PGDH siRNA pool was purchased from Dharmacon (Lafayette, CO); COX-2 siRNAs were from Origene (Rockville, MD). PGE₂ and the COX-2 inhibitor NS398 were purchased from Calbiochem (Billerica, MA).

In Situ Hybridization

The MiRCURY LNA microRNA ISH Optimization Kit (Exiqon, Vedbaek, Denmark) was used for in situ hybridization of mature miR-21 in the formalin-fixed and paraffin-embedded tissue specimens surgically resected from cholangiocarcinoma patients according to the approval of the Institutional Review Board. Briefly, the tissue sections were deparaffinized and treated with proteinase-K for 10 min at 37°C. After dehydration, slides were incubated with miR-21 locked nucleic acid probe for 60 min at 50°C, followed by stringent washes with 5×, 1× and 0.2× saline sodium citrate buffers at 50°C; DIG blocking reagent (Roche, Mannheim, Germany) in maleic acid buffer containing 2% sheep serum for 15 min at room temperature; and alkaline phosphatase–conjugated anti-digoxigenin for 60 min at room temperature. The slides were developed by incubating with 4-nitro-blue tetrazolium and 5-brom-4-chloro-3′-Indolylphosphate substrate (Roche) for 2hr at 30°C, followed by nuclear fast red counterstain (Vector Laboratories, Burlingame, CA) for 1min at room temperature. Scrambled probe and U6 small nuclear RNA–specific probe were used as system control. The same procedure was utilized to analyze the cholangiocarcinoma tissue arrays (obtained from BioCat GmbH, Heidelberg, Germany; 44 cases of cholangiocarcinoma and 4 cases of nonneoplastic tissues). The miR-21 staining was evaluated using a 0 to 4+ semi-quantitative scale: 0, completely negative; 1+, weak cytoplasmic staining; 2+, moderate cytoplasmic staining of >50% of the cells; 3+, strong cytoplasmic staining of >50% of the cells; 4+, extremely strong cytoplasmic staining of >50% of the cells.

Cell Culture

Four human cholangiocarcinoma cell lines, including CCLP1, SG231, HuCCT1, TFK1 and one immortalized nontumorigenic human cholangiocyte cell line (H69) were utilized in this study (HuCCT1 and TFK1 cells were obtained from the Japanese Cancer Research Resources Bank; H69 cells were kindly provided by Dr. Gregory J. Gores at the Mayo Clinic College of Medicine, Rochester, MN). The CCLP1, SG231 and HuCCT1 cells were cultured according to our methods as described previously(19, 22, 23); TFK1 cells were cultured in RPMI-1640 medium containing 10% FBS as described previously(24). The H69 cells were cultured in Bronchial Epithelial Cell Basal Medium (Lonza, Walkersville, MD) with supplemental growth factors in BEGM SingleQuot Kit and 10% heat-inactivated FBS. All cells were cultured in a humidified CO2 incubator at 37°C.

Stable Transfection

The CCLP1, SG231 and TFK1 cells were transduced with miR-21 expressing lentiviral particles or scramble control overnight. After 72hr, the medium was replaced with fresh medium containing puromycin for selection and the subsequent cultures were continued in the presence of 1μg/ml puromycin. After the cells reached confluence, total RNA was extracted and quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed to verify the levels of miR-21.

Cell Proliferation Assay

The growth of human cholangiocarcinoma cells was measured by using the WST-1 reagent from Roche (Indianapolis, IN). Briefly, the cells were plated in 96-well plates and cultured for 24 hours to allow attachment. Then the medium was replaced with fresh medium containing 1% FBS or specific reagents as indicated and the cultures were continued for 1 to 5 days. For cell growth measurement, WST-1 reagent was added to each well and the cells were incubated at 37°C for 1 hour; absorbance at 450nm was measured using an ELISA plate reader.

Colony-Formation Assay

For soft agar colony-formation assay, the cell culture medium and the 5% agar solution were warmed to 40°C in a water bath. Then the cell culture medium and the agar solution were mixed at the proportion of 9:1 and added to 6-well plates at 1ml of mixture per well. The plates were placed under room temperature until solidification. For preparation of top agar, the cell suspensions and the 3.5% agar solution were warmed to 40°C and then mixed at the proportion of 9:1. 1ml of mixture containing 500 cells was added on the top of base agar in each well of 6-well plates. The cells were cultured for 15 days and the colonies were counted under a microscope. For plate colony-formation assay, the 1 ×10³ cells were cultured in 10cm dishes for 10 days to allow colony formation. Colonies were fixed in 100% methanol and stained with 0.1% crystal violet solution (Amersco, Solon, OH) and counted.

Cell Invasion Assay

The cell invasion assay was performed in Matrigel-coated transwell chambers (BD Biosciences Discovery Labware, Bedford, MA). 500ul cell suspension (5 × 10⁴ cells/ml) was added to each of the upper chambers. Cell culture medium containing 5% FBS was added to each of the lower chambers as chemoattractant. After incubating cells at 37°C for 24hr, the cells on the upper surface of the membrane were removed with a cotton swab. The invading cells on the lower surface of the membrane were fixed in 100% methanol and stained with 0.05% crystal violet solution. Then the invading cells were counted under a microscope and for each chamber eight fields were randomly selected for counting.

Transient Transfection

The cells were transfected with the 15-PGDH expression plasmid (Origene) (pCMV6-AC-GFP as control plasmid) or the COX-2 expression plasmid (pcDNA3 as control plasmid) using Lipofectamine2000 reagent according to manufacturer’s instructions. After transfection at indicated time periods, the transfection mixtures were removed and the cells were used for further experiments. Transfection of miR21 inhibitor, 15-PGDH and COX-2 siRNAs was performed by using Lipofectamine2000 reagent. The levels of 15-PGDH and COX-2 proteins in the transfected cells were verified by Western blotting.

QRT-PCR

Cellular total RNA was isolated using the Trizol reagent (Invitrogen). For quantitation of mature miR-21 levels, reverse transcription was performed by using the miScript Reverse Transcription Kit (Qiagen). The mature form of miR-21 was amplified by using the miScript SYBR Green PCR Kit (Qiagen) on Bio-Rad C1000 Thermal Cycler. U6 small nuclear 2(U6B) was used as the internal control. The U6B (Hs_RNU6B_2) and miR-21 (Hs_miR-21_2) primers were purchased from Qiagen. For quantitation of 15-PGDH mRNA levels, reverse transcription was performed by using the Superscript^™ II Reverse Transcriptase Kit (Invitrogen). Then the 15-PGDH gene was amplified by using the QuantiFast SYBR Green PCR Kit (Qiagen). GAPDH was used as the internal control. The GAPDH (Hs_GAPDH_2_SG) and 15-PGDH (Hs_HPGD_1-SG) primers were from Qiagen. MiR-21 and 15PGDH mRNA expression was normalized to their corresponding internal control genes and relative change was calculated using the 2^−ΔΔCT method.

Western Blotting

At the end of each indicated treatment, the cells were washed once with ice-cold phosphate buffered saline (PBS), scraped off the plates and centrifuged. The cells were then lysed in a lysis buffer containing 50 mM HEPES, 1 mM EDTA (pH 8.0), protease inhibitors and phosphatase inhibitors (Roche). After sonication on ice, the cell lysates were centrifuged at 12,000g for 15 min at 4°C and the supernatants were collected for further analysis. The protein concentration was measured by using the Bio-Rad Protein Assay Kit (Bio-Rad, Hercules, CA). The samples were prepared by boiling for 5min at 95°C in the Laemmli sample buffer (Bio-Rad) containing 10% 2-mercaptoethanol. The protein samples were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto the nitrocellulose membranes (BioRad). Non-specific binding was blocked by incubating the membranes in PBS-T (0.1% Tween 20 in PBS) containing 5% nonfat milk for 1hr at room temperature. The membranes were then incubated overnight at 4°C with individual primary antibodies at the dilutions recommended by the manufacturers in PBS-T containing 5% nonfat milk. Following four washes with PBS-T, the membranes were incubated with the IRDye secondary antibodies (Licor, Lincoln, NE) at 1: 20000 dilutions in PBS-T containing 5% nonfat milk for 1hr at room temperature. After four washes with PBS-T, protein bands were visualized using the ODYSSEY Infrared Imaging System (Licor).

Luciferase Reporter Activity Assay

Cells were seeded in 12-well plates and cultured overnight to allow attachment. The cells were then co-transfected with miR-21 mimic and 0.3μg per well of pEZX-MT01-15-PGDH 3′UTR reporter plasmid (Genecopoeia). The cell lysates were obtained 48hr after transfection and luciferase activity was measured in a centro xs3 lb 960 microplate fluoroscence reader (Berthold Technologies, Oak Ridge, TN) by using the Dual-Luciferase reporter assay system (Promega, Madison, WI). The Renilla luciferase activity was used as internal control and the ratio of Firefly to Renilla activity was calculated for normalization. For miR-21 promoter luciferase reporter activity measurement, the cells were transfected with 1.6μg pEZX-PG04-miR-21 promoter reporter plasmid (Genecopoeia). After 24hr, the cells were treated with specific reagents as indicated in the figure legends. The cell culture medium was collected after the treatment and luciferase activity was measured in a FLUOstar Omega microplate reader (BMG Labtech, Cary, NC) by using the Secrete-Pair Dual Luminescence Assay Kit (Genecopoeia). The Secreted Alkaline Phosphatase (SEAP) luminescence was used as internal control and the ratio of Gaussia luciferase (Gluc) activity to SEAP was calculated for normalization.

PGE₂ Enzyme Immunoassay

The cells were incubated for overnight at 37°C in serum-free medium. The cell culture medium was then collected and centrifuged for 10min at 12,000g to remove the floating cells and cell debris. The concentration of PGE₂ in the culture medium was measured by using a specific PGE₂ EIA kit (Cayman Chemical).

Tumor Xenograft Studies

Four-week-old male NOD CB17-prkdc/SCID mice were purchased from Jackson Laboratory (Bar Harbor, ME). For each mouse, 3×10⁶ miR-21 overexpressed or control CCLP1 cells in 100μl PBS were subcutaneously inoculated in the left and right flank areas (N=6). The mice were observed over 16 days for tumor formation. Upon sacrifice, the tumors were recovered and the wet weight of each tumor was measured. Portion of each tumor was used for H&E staining, Western blotting and qRT-PCR analysis.

Statistical Analysis

The in vitro data was presented as mean ± standard deviation (SD) or standard error (SE) from a minimum of three replicates. Differences between groups were evaluated by SPSS 13.0 statistical software with Student’s t-test, one-way analysis of variance (ANOVA), repeated measures ANOVA or Wilcoxon signed ranks test. A p-value less than 0.05 was considered as statistically significant.

RESULTS

The expression of miR-21 is increased in human cholangiocarcinoma tissues

We performed in situ hybridization to determine the expression of miR-21 in human cholangiocarcinoma tissues. The tissue specimens analyzed include archived formalin-fixed paraffin-embedded human cholangiocarcinoma tissues from nine patients who underwent surgical resections and forty-four cases of cholangiocarcinoma tissue arrays. In the archived cholangiocarcinoma tissues, miR-21 was expressed in of 8 of 9 cases (88.9%). In non-neoplastic peribiliary glands, miR-21 was expressed in only 1 of 9 cases (11.1%). The staining intensity of miR-21 was significantly higher in cholangiocarcinoma cells compared to the matched non-neoplastic bile duct epithelial cells (p<0.01) (Figure 1A and 1B). In the tissue array analysis, miR-21 was expressed in 24 of 44 cholangiocarcinoma cases at different levels (47.7% 1+, 4.5% 2+, 2.3% 4), but not in the 4 non-neoplastic cases (Figure 1C). These findings confirm that the expression of miR-21 is up-regulated in human cholangiocarcinoma cells.

Figure 1. In-situ hybridization for mature miR-21 in human cholangiocarcinoma tissues.

A, Representative in-situ hybridization images showing the expression of miR-21 in human cholangiocarcinoma tissues. The dark blue color represented positive signals; nuclei were counterstained as red. I, Positive miR-21 staining in cholangiocarcinoma cells and negative staining in normal bile duct epithelial cells; II and III, high power fields of I from the highlighted areas. B, Graphical presentation of miR-21 staining intensity in archived human cholangiocarcinoma tissues (N=9, p<0.01 Wilcoxon signed ranks test). C, Summary of miR-21 staining intensity in human cholangiocarcinoma tissue arrays.

Mir-21 enhances cholangiocarcinoma cell growth and invasion in vitro

To further determine the biological role and molecular mechanism of miR-21 in cholangiocarcinoma, we generated cholangiocarcinoma cell lines (CCLP1, SG231 and TFK1) with stable overexpression of miR-21. These cells were produced by infecting the parental cell lines with lentivirus particles carrying the miR-21 and eGFP genes under the control of the same promoter; the stably transduced cells were selected in the presence of 1μg/ml puromycin. Satisfactory infection efficiency was confirmed by high eGFP expression under fluorescence microscopy and by qRT-PCR for miR-21. As shown in Figure 2A, the selected miR-21 lentivirus infected cell lines showed increased miR-21 expression and increased cell growth compared to the corresponding lentivirus control cells. Additionally, miR-21 overexpression increased colonogenic efficiency in soft agar (Supplementary Figure S1) and enhanced tumor cell invasion (Supplementary Figure S2). We next employed a parallel approach in which cholangiocarcinoma cells were transfected with the miR-21 specific inhibitor or the scramble control. HuCCT1 cell line was utilized for this purpose, as this cell line expresses relatively high basal level of miR-21 among the four cholangiocarcinoma cell lines utilized in this study (Figure 2B). We observed that inhibition of miR-21 significantly decreased cell growth (p<0.01) and reduced colonogenic efficiency (p<0.01) (Figure 2C and 2D). These results demonstrate that miR-21 promotes anchorage-dependent cell growth and anchorage-independent colony formation and enhances cell invasion, indicating protumorigenic effect of miR-21 in cholangiocarcinoma.

Figure 2. MiR-21 regulates cholangiocarcinoma cell growth in vitro.

A, (upper panel) Cell growth curves of human cholangiocarcinoma cells (CCLP1, SG231 and TFK1) stably transduced with miR-21 lentivirus (indicated as L/miR-21) or scramble control (indicated as L/control) (N=4; p<0.001, repeated measures ANOVA). (lower panel) MiR-21 levels were measured by qRT-PCR. The data is presented as mean ± SD (N=3; ***p<0.001, **p<0.01, *p<0.05; Student’s t-test). B, The levels of miR-21 in different human cholangiocarcinoma cell lines (CCLP1, SG231, TFK1 and HuCCT1). C, The effect of miR-21 inhibition on cell growth in HuCCT1 cells. The cells were transfected with miR-21 inhibitor (Anti/miR-21) or scramble control (Anti/Control) and WST-1 assay was performed to determine cell growth (the data is presented as mean ± SE; N=6; p<0.01, repeated measures ANOVA). D, The effect of miR-21 inhibition on colony formation in HuCCT1 cells transfected with miR-21 inhibitor (Anti/miR-21) or scramble control (Anti/Control). The numbers of colonies were counted after 10 days. (Left panel) Representative images showing colonies formed in cell culture dishes.(Right panel) Average colony forming efficiency (the data is presented as mean ± SD, N=3, **p<0.01).

MiR-21 targets the tumor suppressor, 15-PGDH

Next, we sought to identify the direct target of miR-21 in human cholangiocarcinoma cells. By using the microRNA.org resource, we found that the 15-PGDH mRNA harbors the miR-21 binding site in the 3′UTR (illustrated in Figure 3A). 15-PGDH is a key enzyme that degrades the pro-inflammatory and protumorigenic PGE₂ and has been shown to inhibit the growth of several human cancers(5, 6), although the role of 15-PGDH in human cholangiocarcinogenesis has not been previously reported. In this study, we observed decreased 15-PGDH protein expression in human cholangiocarcinoma cells (CCLP1, SG231, HuCCT1 and TFK1) compared to the immortalized non-tumorigenic human cholangiocyte cell line, H69 (Figure 3B). QRT-PCR and Western blotting results showed that both 15-PGDH mRNA and protein levels were decreased in cells with stable overexpression of miR-21 (Figure 3C and 3D). Transient transfection of CCLP1 and SG231 cells with miR-21 mimic also decreased 15-PGDH protein level (data not shown). On the other hand, transfection of SG231 cells with anti-miR-21 increased 15-PGDH mRNA and protein levels (Supplementary Figure S3). To document the role of the putative miR-21 binding site for regulation of 15-PGDH in cholangiocarcinoma cells, we co-transfected miR-21 mimic with 15-PGDH wild type or mutant 3′UTR reporter plasmid. As shown in Figure 3E, miR-21 mimic decreased 15-PGDH 3′UTR luciferase reporter activity and this effect was abolished when the one nucleotide in the miR-21 seed binding site was deleted. Given that 15-PGDH is a key enzyme involved in PGE₂ inactivation, we further measured the level of PGE₂ and observed that miR-21 overexpression increased the accumulation of PGE₂ in CCLP1 and SG231 cells (Figure 3F). As cyclooxygenase-2 (COX-2) is a rate-limiting key enzyme that mediates PGE₂ synthesis, we also examined the level of COX-2 protein and found that miR-21 did not alter the expression of COX-2 (data not shown). Our results suggest that miR-21-induced PGE₂ increase in cholangiocarcinoma cells is mediated predominantly through inhibition of 15-PGDH and blockade of PGE₂ degradation. The role of 15-PGDH for inhibition of cholangiocarcinoma growth is supported by the observation that 15-PGDH overexpression inhibited CCLP1 cell growth (Figure 4A) whereas 15-PGDH siRNAs accelerated tumor cell growth (Figure 4B).

Figure 3. 15-PGDH is a direct target of miR-21 in cholangiocarcinoma cells.

A, Putative miR-21 binding site in the 3′UTR of 15-PGDH mRNA. A mutant 15-PGDH 3′UTR with one nucleotide deletion was constructed as indicated. B, Western blotting for 15-PGDH in an immortalized human cholangiocyte cell line (H69) and four human cholangiocarcinoma cell lines (CCLP1, SG231, HuCCT1, TFK1). C, qRT-PCR analysis for 15-PGDH mRNA in miR-21 overexpressed and control cells. The data is presented as mean ± SD (N=3; **p<0.01, ***p<0.001; Student’s t-test). D, 15-PGDH protein levels in miR-21 overexpressed and control cells as determined by Western blotting. E, CCLP1 and SG231 cells were transfected with miR-21 mimic or scramble control together with a wild type 15-PGDH 3′UTR reporter plasmid, or a mutant 15-PGDH 3′UTR reporter plasmid. The luciferase activity was analyzed 48hrs after transfection. The data is presented as mean ± SE (***p<0.001, Student’s t-test). F, Equal number of miR-21 overexpressed and control cells were plated in 6-well plates and cultured in serum free medium overnight. The cell culture medium was collected to measure PGE₂ concentration. The data is presented as mean ± SD (*p<0.05, ***p<0.001; Student’s t-test).

Figure 4. 15-PGDH regulates cholangiocarcinoma cell growth in vitro.

A, 15-PGDH overexpression inhibits cholangiocarcinoma cell growth. (Left panel) The growth of CCLP1 cells transfected with 15-PGDH expression plasmid or control vector. The data is presented as mean ± SE (N=6; p<0.05, repeated measures ANOVA). (Right panel) Increased 15-PGDH in 15-PGDH overexpressed CCLP1 cells was confirmed by Western blotting. B, Depletion of 15-PGDH by siRNA enhances cholangiocarcinoma cell growth. (Left panel) The growth curves of CCLP1 cells transfected with 15-PGDH or control siRNAs. The data is presented as mean ± SD (N=6; p<0.001, repeated measures ANOVA). (Right panel) Decreased 15-PGDH protein in CCLP1 cells transfected with 15-PGDH siRNAs was confirmed by Western blotting.

MiR-21 promotes cholangiocarcinoma growth in tumor xenograft model

To investigate whether miR-21 promotes cholangiocarcinoma growth in vivo, miR-21 overexpressed and control CCLP1 cells were subcutaneously inoculated into the flank areas of SCID mice. As shown in Figure 5A and 5B, miR-21 overexpressed tumors were larger in size and had higher tumor weight compared to control (0.72±0.10g vs 0.37±0.18g, p<0.05). QRT-PCR result confirmed elevated miR-21 levels in miR-21 overexpressed tumors (Figure 5C). Consistent with the in vitro results, the levels of 15-PGDH mRNA and protein were also decreased in miR-21 overexpressed tumors (Figure 5D and 5E). Thus, miR-21 reduces 15-PGDH level and enhances cholangiocarcinoma growth, both in vitro and in vivo.

Figure 5. MiR-21 promotes cholangiocarcinoma growth in vivo.

A and B, The gross image and weight of xenograft tumors recovered from the six SCID mice. p<0.05 (paired t-test). C and D, miR-21 and 15-PGDH mRNA levels in xenograft tumor tissues were determined by qRT-PCR. The data is presented as mean ± SD (**p<0.01, ***p<0.001; paired t-test). E, 15-PGDH protein levels in xenograft tumor tissues were determined by Western blotting.

PGE₂ up-regulates miR-21 level and increases miR-21 transcription

We performed further experiments to determine whether miR-21 might be regulated by the COX-2/PGE₂ signaling cascade. As shown in Figure 6A, CCLP1 and SG231 cells transfected with the COX-2 expression plasmid showed increased PGE₂ production and increased miR-21 level. COX-2 overexpression also increased miR-21 level in H69 cells (Supplemental Figure S4). Accordingly, treatment of CCLP1 and SG231 cells with PGE₂ also increased miR-21 expression (Figure 6B). Furthermore, PGE₂ treatment also increased miR-21 promoter luciferase reporter activity (Supplemental Figure S5). Given that the HuCCT1 cell line expresses highest basal level of COX-2 protein among the four cholangiocarcinoma cell lines utilized in this study, the HuCCT1 cells were selected for COX-2 inhibition studies. As shown in Figure 6C and 6D, COX-2 siRNAs significantly decreased miR-21 expression as well as miR-21 promoter luciferase reporter activity in HuCCT1 cells. Moreover, inhibition of COX-2 by its pharmacological inhibitor, NS-398, also significantly decreased miR-21 level in HuCCT1 cells (Supplemental Figure S6). These data suggest that the COX-2-derived PGE₂ up-regulates miR-21 expression in cholangiocarcinoma cells and the effect is mediated at least in part through induction of miR-21 gene transcription.

Figure 6. COX-2 and PGE₂ signaling enhances miR-21 expression in cholangiocarcinoma cells.

A, The effect of COX-2 overexpression on miR-21 level. (Upper panel) CCLP1 and SG231 cells were transfected with COX-2 expression plasmid or control PCDNA3 vector. 48hrs after transfection, total cellular RNA was isolated and miR-21 level was determined by qRT-PCR. The expression of COX-2 was confirmed by Western blotting (Mid panel). The level of PGE₂ released from the transfected cells is shown in the Lower panel. The results were obtained from three experiments (*p<0.05, **p<0.01, ***p<0.001; Student’s t-test). B, The effect of PGE₂ treatment on miR-21 level. CCLP1 and SG231 cells were treated with PGE₂ at indicated concentrations for 24 hrs. Total cellular RNA was isolated and miR-21 level was determined by qRT-PCR. The results were obtained from three individual experiments (left panel, p<0.001; right panel, p<0.05; one-way ANOVA). C, COX-2 siRNAs decreased miR-21 expression in HuCCT1 cells. The cells were transfected with two individual COX-2 siRNAs. 48hrs after transfection, total cellular RNA was isolated and miR-21 level was determined by qRT-PCR (N=3; p<0.001, one-way ANOVA). Decreased COX-2 protein levels in the cells transfected with COX-2 siRNAs were confirmed by Western blotting. D, COX-2 siRNAs decreased miR-21 promoter luciferase reporter activity in HuCCT1 cells. The cells were transfected with miR-21 promoter reporter plasmid. 24hrs after transfection, the cells were transfected with two individual COX-2 siRNAs. 72hrs after the siRNA transfection, the cell culture medium was obtained for analysis of the luciferase reporter activity (N=3; p<0.01, one-way ANOVA).

DISCUSSION

The current study identifies 15-PGDH as a novel target of miR-21 in cholangiocarcinoma. Our results disclose a novel feed-forward loop between COX-2/PGE₂ and miR-21 signaling pathways via 15-PGDH which is crucial in cholangiocarcinogenesis and tumor progression (illustrated in Figure 7). These findings are noteworthy as 15-PGDH is an important tumor suppressor that antagonizes the pro-inflammatory and tumorigenic effect of PGE₂. Moreover, by utilizing in situ hybridization technique, we observed that miR-21 is exclusively expressed in cholangiocarcinoma cells but not in other cell types in the tumor microenvironment (such as myofibroblasts, endothelial cells and inflammatory cells); this observation is consistent with the reported increase of miR-21 in cholangiocarcinoma cells by qRT-PCR analysis(22, 23). In the present study, the tumorigenic effect of miR-21 in cholangiocarcinoma is documented by complementary in vitro and in vivo studies.

Figure 7. Diagram illustrating the cross-talk between PGE₂ and miR-21 signaling pathways in human cholangiocarcinoma.

Increased PGE₂ signaling in cholangiocarcinoma enhances miR-21 expression. MiR-21 targets 15-PGDH, leading to blockage of PGE₂ degradation. Accumulation of PGE₂ further enhances miR-21 expression. These signaling cascades form a feed-forward loop that drives cholangiocarcinogenesis and tumor progression.

In normal tissues 15-PGDH acts as a physiologic antagonist of the COX2/PGE₂ signaling. Thus loss of 15-PGDH expression in tumor tissues may be one of the key mechanisms for enhanced PGE₂ signaling in human cancers. However, to date, the exact mechanism for loss of 15-PGDH expression in cancer has not been completely defined. The current study shows that the expression of 15-PGDH is reduced in human cholangiocarcinoma cell lines (compared to the non-tumorous biliary epithelial cell line, H69). Our results in this study provide the first evidence that 15-PGDH is post-transcriptionally down-regulated by miR-21. This conclusion is based on the following findings: (1) miR-21 binding site was identified in the 3′UTR of 15-PGDH mRNA by sequence alignment analysis; (2) miR-21 overexpression decreased 15-PGDH mRNA and protein levels; and (3) transfection of miR-21 mimic decreased 15-PGDH 3′UTR luciferase reporter activity and the effect was abolished by miR-21 seed binding site mutation. The observations that 15-PGDH overexpression inhibited tumor cell growth and that 15-PGDH inhibition accelerated tumor cell growth support a tumor suppressive role of 15-PGDH in human cholangiocarcinoma. Our findings described in this study, along with the documented role of 15-PGDH in tumorigenesis, support a key role of miR-21-mediated inhibition of 15-PGDH in cholangiocarcinogenesis and tumor progression.

Although much attention has been paid to miRNA target studies to understand the role of miRNAs in cancer biology, there are fewer studies on how miRNAs are regulated in cancer cells. Whereas some of miRNAs that are contained in introns might be generated as a by-product of pre-mRNA splicing, most miRNAs likely come from intergenic regions and are transcribed from their own promoters(24). A previous study described a miR-21 regulatory region mapping −3,565 to −2,415 upstream of the primary miR-21 (pri-miR-21) which is inducible by interleukin-6 (IL-6)/signal transducer and activator of transcription 3 (STAT3)(25). In addition to transcriptional regulation, miR-21 can be also post-transcriptionally regulated by transforming growth factor beta (TGF-β) and bone morphogenetic protein (BMP) signaling(26). In the current study, we show that miR-21 is induced by PGE₂ at least in part through transcriptional regulation. The latter assertion is based on the findings that COX-2 overexpression and PGE₂ treatment increased miR-21 level and miR-21 promoter luciferase reporter activity and that inhibition of COX-2 decreased miR-21 level and miR-21 promoter luciferase reporter activity.

In summary, this study shows that miR-21 enhances cholangiocarcinoma growth by inhibiting 15-PGDH which leads to PGE₂ accumulation and that enhanced PGE₂ signaling further stimulates miR-21 transcription and increases the level of cellular miR-21. Such a feed-forward regulatory loop between PGE₂ and miR-21 likely plays an important role in cholangiocarcinogenesis and tumor progression. Thus, blocking COX2/PGE₂ signaling in combination with targeting miR-21 may represent a promising therapeutic strategy for the treatment of human cholangiocarcinoma.

Implications.

Cholangiocarcinogenesis and tumor progression is regulated by a novel interplay between COX-2/PGE₂ and miR-21 signaling which converges at 15-PGDH.

ABBREVIATIONS

3′UTR

3′-untranslated region

miR-21

microRNA-21

15-PGDH

15-hydroxyprostaglandin dehydrogenase

PGE₂

prostaglandin E₂

COX-2

cyclooxygenase-2

miRNA

microRNA

mRNA

messenger RNA

DMEM

Dulbecco’s modified minimum essential medium

FBS

fetal bovine serum

eGFP

enhanced green flurescent protein

siRNA

small interfering RNA

qRT-PCR

quantitative reverse transcription polymerase chain reaction

PBS

phosphate buffered saline

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

SCID

severe combined immunodeficiency