Human cervical cancer cell lines (ME-180, CaSki, HeLa, SiHa), an immortalized cervical epithelial cell line (NC104), and a human embryonic kidney cell line (HEK293T) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin under a humidified atmosphere of 5% CO₂ at 37°C.

Fresh paired human cervical tissues and adjacent normal tissues were collected from 64 patients who were undergoing surgery for cervical carcinoma between 24th July 2005 and 13th November 2008 in the Women’s Hospital, School of Medicine, Fudan University (Shanghai, China). Informed consent was obtained from each patient. The study was approved and supervised by the Ethics Committee of Fudan University (Shanghai, China) and it was performed in compliance with the Helsinki Declaration. All tissue samples were snap-frozen in liquid nitrogen and stored at −80°C until total RNA was extracted.

Total RNA was obtained from tissue samples and cells using TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. The relative level of miR-26a was determined by RT-qPCR using the miRCURY LNA™ microRNA PCR system (Exiqon, Woburn, MA, USA). Briefly, 10 ng of total RNA was reverse-transcribed with the miRNA corresponding RT Primer (Exiqon, Woburn, MA, USA) and Transcriptor Reverse Transcriptase (Exiqon), and the cDNA was used as a template for the qPCR reaction using the miRNA specific LNA™ PCR primer (Exiqon) and Universal PCR primer (Exiqon). ΔΔCt values were normalized with the endogenous U6 small nuclear RNA.

For analysis of the expression of PRL-1 mRNA, 500 ng of the total RNA was reverse transcribed using a Transcriptor First Strand cDNA Synthesis kit (Takara, Dalian, China) with random primers under standard conditions. The primers used were as follows: Sense, 5′-CACCATCTTCCAGG-AGCGAG-3′ and antisense, 5′-TCACGCCACAGTTTCCCGGA-3′ for GAPDH; and sense, 5′-AGGGACAAGCCTACCCCTC-3′ and antisense, 5′-CTCATCTCCCGTCAGTTGGT-3′ for PRL-1. GAPDH was used as the endogenous control. qPCR was performed in triplicate on the ABI Prism Sequence Detection system 7900HT (Applied Biosystems, Foster City, CA, USA). The gene expression was normalized to the internal controls and the fold-changes were calculated using relative quantification. All of the experiments were performed three times with three technical replicates.

To construct the miR-26a overexpression vector, a DNA fragment encoding the miR-26a pre-miRNA was amplified by PCR using the following oligonucleotide primers: Sense: 5′-GGCGAA TTCCCCACTGCTGACCCATTC-3′ and antisense: 5′-TATGG ATCCCCACAAGACTCCTCGTTGC-3′. The PCR product was TOPO^®-cloned into the pCR^®4-TOPO^® vector (Invitrogen Life Technologies). The construct was sequenced and the pre-hsa-miR-26a fragment was sub-cloned into pcDNA4/myc-HisA to generate pcDNA4/miR-26a.

The human PRL-1 3′UTR luciferase reporter, containing putative binding sites for miR-26a, was generated by cloning the PRL-1 mRNA 3′UTR sequence into the downstream of the luciferase gene of the pGL3-control vector (Promega Corporation, Madison, WI, USA) using the following primers: Sense: 5′-GGC TCTAGAGGGCCTACAGGAGGGGTTA-3′ and antisense: 5′-GGCTCTAGATGTGATTAAAGTAAAATGCA ATTCA-3′. The plasmids was termed PRL-1-UTR–WT and site-directed mutagenesis of the miR-26a target site in the PRL-1 3′UTR was performed using the Quick-change mutagenesis kit (Stratagene, La Jolla, CA, USA) to generate the PRL-1-UTR–MUT plasmids. Correct vector construction was verified by direct sequencing. For the mutated construct, the miR-26a target site UACUUGAC was substituted with a UAGAACAC fragment.

To silence the PRL-1 expression, lentivirus vector shPRL-1 was constructed to establish PRL-1 silencing as described previously (19) using the following primer sequences: Sense: 5′-CCGGTTCTTGCTGTCAGCATATAAACTCGA GTTTATATGCTGACAGCAAGAATTTTTG-3′ and antisense: 5′-AATTCAAAAATTCTTGCTGTCAGCATATAAA CTCGAGTTTATATGCTGACAGCAAGAA-3′.

A dual luciferase assay was conducted by co-transfecting HEK293T cells with 20 ng pcDNA4/miR-26a or empty lentiviral vector, along with 100 ng of firefly luciferase reporter comprising wild type or mutant 3′UTR of PRL-1 gene and 10 ng pRL-TK vector, using Lipofectamine™ 2000 (Invitrogen Life Technologies) per well, according to the manufacturer’s instructions in a 24-well plate format. The cells were harvested 48 h following transfection for luciferase assay using a luciferase assay kit (Promega Corporation) and Renilla luciferase activity was used for transfection variation normalization according to the manufacturer’s instructions. Each experiment was repeated three times.

The cells were seeded at a density of 2,000 cells per well in a 96-well plate containing 100 μl DMEM culture media with 10% FBS. A total of 10 μl cell counting kit (CCK)-8 solution was added to each well and the cells were incubated at 37°C for 2 h. The absorbance values were measured at 450 nm every 24 h following the manufacturer’s instructions. Triplicate wells were measured in each treatment group.

A total of 4×10⁴ cells in 100 μl serum free media were seeded into the upper chamber coated with 150 μg of Matrigel (BD Biosciences, Bedford, MD, USA), and NIH 3T3 fibroblast conditioned medium was added to the lower chamber. After the cells were incubated for 36 h, noninvasive cells were removed from the upper surface of the membrane with a cotton swab, and the invaded cells on the lower membrane surface were fixed and stained with hematoxylin for 30 min. Following drying, the invasive cells were captured using a microscope in five random fields using a DP controller (Olympus, Tokyo, Japan).

The cells were placed in 6-well plates at a density of 500 cells per well in normal culture medium as stated above and incubated for two weeks, the medium was replaced every four days. Next the cells were washed twice with PBS, fixed with 4% polyoxymethylene and stained with 1% crystal violet for 30 min. The number of colonies was counted and a single clone contained >50 cells. Each assay was performed in triplicate.

The cells were washed twice with PBS and then lysed using RIPA buffer supplemented with protease inhibitor cocktail (Roche, Basel, Switzerland) and sonicated (Shengyan, Shanghai, China) with one 10 sec burst. Whole cell extracts containing equal quantities of proteins (50 μg) were separated by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE), transferred to 0.45-μm polyvinylidene difluoride membrane sheets (Millipore, Billerica, MA, USA). The membrane was blocked with 5% skimmed milk at room temperature for 1 h and probed with the following antibodies: Anti-PRL-1 antibody (ab3523, Abcam, Cambridge, UK) and anti-GAPDH antibody (sc-25778; Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA). Following an overnight incubation, the membrane was washed and incubated with the appropriate goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (Kangcheng, Shanghai, China) at room temperature for 1 h. The protein bands were subjected to a chemiluminescence detection assay (chemiluminescence kit; Tiangen, Beijing, China).

The present study was conducted in accordance with the Care and Use of Laboratory Animals of the National Institutes of Health. All of the experimental procedures were approved by the Committee on the Care and Use of Laboratory Animals of the Fudan University. Female nude mice (age, 3–5 weeks) were purchased from Shanghai SLAC Laboratory Animal Co. Ltd. (Shanghai, China). The mice were housed under humidity- and temperature-controlled conditions with a 12 h light/dark cycle, and were fed a normal diet. Xenografts were established by subcutaneous injection of 5×10⁶ Hela cells with vector control or overexpression miR-26a per mouse in the right flank area in a volume of 200 μl FBS-free DMEM medium. The animals were sacrificed by joint dislocation following six weeks and the tumor volume was measured using a caliper every seven days, using the following formula: Volume = (width)² × length/2. Four mice were used in each group and the experiment was repeated three times independently.

The experimental data were demonstrated as the mean ± standard deviation for each group. Student’s t-test (two-tailed) was performed to analyze the data using GraphPad Prism software (La Jolla, CA, USA). P<0.05 were considered to indicate a statistically significant difference.

In order to assess the roles of miR-26a in human cervical cancer development, the miR-26a expression levels in cervical cancer tissues (n=64) and paired adjacent normal tissues were compared by RT-qPCR. It was identified that miR-26a expression was markedly reduced in tumor tissues as compared with that in adjacent normal tissues (Fig. 1A). Furthermore, the expression level of miR-26a in several human cervical cancer cell lines and the NC104 human normal immortalized cervical epithelial cell line was detected. As expected, the expression of miR-26a was decreased in all of the cancer cell lines compared with that in the normal cells (Fig. 1B). These results indicated that miR-26a was significantly downregulated in clinical human cervical cancer tissues and cells, and therefore, may be involved in human cervical cancer development.

To elucidate the potential mechanisms underlying the effects of miR-26a in cervical cancer cells, miR-26a was restored in the Hela and SiHa cell lines by a lentiviral vector. miR-26a transduction significantly increased the miR-26a expression in Hela and SiHa cells compared with the vector cells, as demonstrated by RT-qPCR analysis (Fig. 2A), reflecting efficient overexpression of miR-26a in these cervical cancer cells. Next, several functional analyses were performed. A CCK-8 assay demonstrated that forced expression of miR-26a significantly reduced cell proliferation in the two cervical cancer cell lines (Fig. 2B and C). To further evaluate the proliferation ability, the effects of the restoration of miR-26a on colony formation were examined. As demonstrated in Fig. 2D, there were notably fewer and smaller colonies of the cells overexpressing miR-26a compared with the vector transductions.

Invasion is the key element of tumor metastasis, thus it was investigated whether reintroducing miR-26a affects cell invasion by utilizing transwell migration assays. As exhibited in Fig. 2E, miR-26a significantly reduced the number of Hela and SiHa cells invading through the Matrigel basement membrane. Together, these observations suggested that the loss of miR-26a in cervical cancer may, at least partially, contribute to tumor growth and invasive capacity.

Considering the aforementioned results that demonstrated that forced expression of miR-26a inhibited cell growth in cervical cancer cells in vitro, an in vivo model was used to examine the effect of miR-26a restoration on tumor growth. It was identified that miR-26a restoration markedly inhibited tumor formation and reduced the tumor size and weight after five weeks compared with the vector control (Fig. 3A and B). These data indicate that overexpression of miR-26a reduces tumor growth in Hela cells in vivo.

To examine the mechanisms through which miR-26a regulated cervical cancer cell growth, two publicly available algorithms were used to facilitate predicting the miR-143 targets. A bioinformatic prediction (Targetscan and Pictar) for putative targets of miR-26a was used and it was identified that PRL-1 was a potential target of miR-34a. Increased expression of PRL-1 has been reported and implicated in tumor progression and angiogenesis in human lung cancer and colon cancer (20). The present study focused on the possible regulation of PRL-1 by miR-26a. Sequence analyses revealed that the 3′-UTR of PRL-1 mRNA was partially complementary to miR-26a (Fig. 4A). To experimentally validate the direct miR-26a-PRL-1 interaction, PRL-1-UTR-WT and PRL-1-UTR-MUT plasmids were constructed. Upregulation of miR-26a led to a significant decrease in luciferase activity when the reporter contained a wild-type sequence, but no change in the luciferase activity of the mutant sequence (Fig. 4B). Furthermore, the endogenous expression of PRL-1 in Hela and SiHa cells upon miR-26a forced expression was examined, and it was identified that the endogenous PRL-1 protein level was markedly reduced by miR-26a (Fig. 4C). These results demonstrated a direct interaction between miR-26a and PRL-1 in cervical cancer cell lines.

To evaluate whether PRL-1 is potentially involved in miR-26a-regulated cell proliferation and invasiveness, shPRL-1 efficiently knocked down endogenous PRL-1 expression (Fig. 5). Compared with the negative control, silencing PRL-1 markedly suppressed cell proliferation (Fig. 5A and B), colony formation (Fig. 5C) and invasive ability (Fig. 5D) of Hela and SiHa cells. As PRL-1, a direct target of miR-26a, has an important role in extracellular signal regulated kinases (ERK)1/2 activation (21), the expression of ERK1/2 and the downstream effectors following PRL-1 silencing were detected. As expected, PRL-1 knockdown reduced ERK1/2 and RhoA activation in the cervical cancer cells (Fig. 5E). To further confirm that PRL-1 is a target gene protein of miR-26a, it was investigated whether miR-26a affected the expression of ERK1/2 and RhoA through suppression of PRL-1 expression. Consistent with the results of PRL-1 silencing, miR-26a overexpression significantly decreased the expression of ERK1/2 and RhoA (Fig. 5F). These results provide evidence suggesting that PRL-1 is involved in miR-26a-regulated cervical cancer cell growth and invasiveness.

To further determine the correlation between miR-26a and PRL-1, the mRNA level of PRL-1 in primary human cervical cancer tissues was detected and compared with that in the normal adjacent tissue in the same patient cohort that was used for measuring the miR-26a levels. The expression of PRL-1 was markedly higher in the tumor tissue compared with the adjacent normal tissue (Fig. 6A). Notably, there was an inverse correlation between the expression level of miR-26a and PRL-1 in the cervical cancer tissues, as determined by Pearson’s correlation coefficient (Fig. 6B). Collectively, these data supported the evidence for a reciprocal correlation between the levels of miR-26a and PRL-1 in human cervical cancer.