Down-Regulated Mir-22 as Predictive Biomarkers for Prognosis of Cervical Cancer

Cervical cancer is a highly prevalent cancer in women all over the world, which causes about 300,000 deaths each year, continues to be the leading cause of death among gynecological malignancies [1]. Although the cervical cytology screening benefits a lot in the early diagnosis and treatments, cervical cancer outcomes vary significantly [2]. The exactly molecular mechanisms of cervical cancer initiation and progression have not yet been fully elucidated. More importantly, it tends to present at more and more young cases in recent years [3]. So there is an urgent need for understanding more sensitive molecular prognostic biomarkers to better predict the outcome of cervical cancer.


Cell lines and transfections
Cervical cancer cell lines, Hela and Siha, were obtained from American Type Culture Collection (ATCC) and cultured in T-medium with 5% Fetal Bovine Serum at 37˚C, 5% CO2. Hela and Siha cells were transfected with Lipofectamine 2000 (Invitrogen, CA, USA), according to the manufacturer's protocol. miR-22 mimics and its negative control were purchased form Genepharmacy Company (Shanghai, China).. A scrambled sequence without significant homology to any rat, mouse or human gene was used as a negative control (vector group) (ABI PRISM, Carisbad, CA). The double-stranded oligonucleotide encoding miR-22 mimic and negative control were annealed and inserted into the linearized eukaryotic pFU-GW-009 vector (GeneChem). All these vectors were confirmed by sequencing. SiHa and HeLa cells were transfected with the miR-22-expressing plasmids using Lipofectamine 2000 (Invitrogen, CA, USA), and G418 (400 µg/ml, Sigma) was used to select the clones stably overexpressing miR-22.

Clonogenic assays
Cell viability was examined using the CellTiter 96 Non-Radioactive Cell Proliferation Assay (MTS assay, Promega), according to the manufacturer's protocol. For clonogenic assays, transfected cells were calculated and seeded 48 hrs post-transfection at low density in 6-well plates. Cells were incubated for 10-14 days, then fied and stained with 0.2% methylene blue in 50% methanol. The surviving fraction was calculated by comparison with control cells.

Invasion assays
Briefly, cells (0.5 × 106 cells/well) were seeded with medium containing low serum (1%) in the upper chamber. The lower chamber was filed with medium containing high serum (20%) as a chemoattractant. Cells were incubated for 24 hrs and then membranes were stained using Diff-Quick (Siemens). A light microscope was used to count the number of invading cells. Invasion assays were performed using the BD BioCoat Matrigel Invasion Chambers and Control Inserts (BD Biosciences), respectively.

Cell wound healing assay
The migration ability of the cervical cancer cells was determined using a wound healing assay. At 24 h post-transfection, a monolayer of cells was scratched with the angularity of a small X-ray film. Wound closures were photographed every 4 h.

Statistical analysis
All data were processed using PASW Statistics 16. Data were presented as mean ± SEM. using t tests for 2-group comparisons. While the results did not display normal distribution, we chose to analyze the data with non-parametric methods. (Mann-Whitney U test between two groups and Kruskal-Wallis H test for three or more groups). A P value less than 0.05 is considered statistically significant. The relationship between miR-22 and MMP1/MAPK1 expression was explored by Spearman's correlation analysis. Significant associations between miR-22 changes and clinicopathologic parameters were assessed using a χ2 test. Two-sided P-values < 0.05 were considered statistically significant.

Relative expression of microRNA-22 is associated with advanced clinicopathologic characteristics of patients with cervical cancer
Relative miR-22 expression was detected by qRT-PCR between paired tumor tissues and normal tissues from 62 patients with cervical cancer. Of the 62 patients with cervical cancer, 32 were placed in the low miR-22 expression group and 30 were placed in the high miR-22 expression group. The associations between clinicopathologic features and miR-22 expression were summarized in Table 1. Our results demonstrated that the relative expression of miR-22 were markedly downregulated in cervical cancer samples compared with the normal tissues (2.162 ± 0.2044 vs 5.819 ± 0.3157), (P < 0.0001; Figure 1A). To evaluate the clinical value of miR-22 in cervical cancer patients, we divided the patients into different groups and he association between relative miR-22 expression and clinicopathologic information was then analyzed (Table 1). A significant low miR-22 expression level was correlated with FIGO stage (P = 0.00004), tumor differentiation (P = 0.0002), and lymph-node-metastasis stage (P = 0.0099) ( Figure  1B, Table 1). No significant association was found between miR-22 expression and other clinical characteristics, such as age and pathology (Table 1). Hence, downregulated miR-22 expression was closely related to cervical cancer metastasis.

The expression of miR-22 is associated with survival of cervical cancer patients
To further analyze the significance of miR-22 in terms of clinical prognosis, Kaplan Meier analysis with the log-rank test was performed, using patient survival ( Figure 2). Our results indicated that low miR-22 expression had a significant impact on overall survival (P = 0.0016;  shown in Figure 2A) and progression-free survival (P = 0.0005; shown in Figure 2B). Multivariate analysis revealed that miR-22 expression level was an independent prognostic factor for overall survival (HR = 2.754, 95% CI: 1.201-5.591), as well as progression-free survival (HR = 3.460, 95% CI: 1.805-7.536) of cervical cancer patients. This supports the notion that low levels of miR-22 in tumor tissues could lead to worse survival rates in cervical cancer patients.

miR-22 promotes colony formation, migration and invasion in vitro
To determine whether miR-22 had any effects on cervical cancer cell lines( SiHa and Hela cells), which are both cervical squamous cell carcinoma lines, we analyzed colony formation, migration and invasion. Both cell lines were transfected with premiR negative control (miR-NC) or pre-miR-22 (miR-22). miR-22 significantly inhibited colony formation compared with the NC control in both cell lines(both P < 0.05; Figure 3A).
Because miR-22 down-regulation was observed to be associated with lymph node metastasis and recurrence in our patients, we performed migration and invasion assays. Consistent with the clinical data, miR-22 overexpression reduced migration and invasion capacities of both Hela and Siha cells (P<0.05; Figures 3B and 3C).
Taken together, these results suggest an oncogenetic role for miR--22 in regulating cervical cancer cell survival, invasion and motility in both Hela and Siha cells.

miR-22 Directly Targets and Inhibits MMP1 and CXCR4
To understand how miR-22 facilitates cervical cancer growth and metastasis, we used four algorithms (Targetscan, miRanda, miRDB and miRWalk) to help identify miR-22 targets in human cervical cancers. Of these target genes that were predicted by all algorithms (Figure 4A), matrix metallopeptidase 1(MMP1) and mitogen-activated protein kinase 1 (MAPK1) attracted our attention immediately as they have been implicated in tumorigenesis and metastasis. We cloned the MMP1 and MAPK1 3'-UTR into a luciferase reporter vector. Luciferase assay revealed that miR-22 directly bound to MMP1 and MAPK1 3'-UTR, and by which it remarkably reduced luciferase activities ( Figure  4B). However, mutation of the putative miR-22 sites in the 3'-UTR of MMP1 and MAPK1 abrogated luciferase responsiveness to miR-22. To directly assess the effect of miR-22 on MMP1 and MAPK1 expression, we performed western blot analysis. As seen in (Figure 4C

Discussion
Evidence has shown miRNAs may play an important role in cancer initiation, development and metastasis in many types of cancers, including cervical cancer [10][11][12]. Thus, studies have focused on cancerspecific miRNAs and its target genes to elucidate biological mechanism [13,14]. Previously, researchers have found that dysregulation of several miRNAs might be associated with the prognosis in patients with cancer.
In this study, we analyzed the expression of miR-22 in cervical cancer tissues, explored the effects of miR-22 on cervical cancer cell lines, and determined the target genes regulated by miR-22. We have discovered that expression levels of miR-22 were downregulated with an increase in the grade of FIGO stage. First, we found that patients with cervical cancer have a much lower level of miR-22 compared with normal cervical tissue (Figure 1). Although considerable evidence has revealed the importance of miR-22 in cancer, its role as an ontogenetic factor is still controversial: it was downregulated in breast cancer, inhibits cell proliferation, cell metastasis, impairs migration in ovarian cancer cells [15,16]. In contrast, miR-22 expression was suggested to be upregulated in prostate cancer [6]. However, the clinical significance and prognostic value of miR-22 in cervical cancer haven't been investigated. Hence, our present study focused on the expression and clinical significance of miR-22 in cervical cancer.
Furthermore, survival analyses showed that post-operative patients with lower levels of miR-22 had a poorer prognosis in both overall survival and progression free survival compared with those with higher miR-22 level (Figure 2). Multivariate Cox analysis also revealed that the post-operative level of miR-22 is an independent prognostic factor for cervical cancer. Taken together, our results support a notion that miR-22 may have potential as a promising biomarker for the prognosis of cervical cancer. However, it still needs to be explored further, as our sample size and the duration of follow-up are limited.
Then the expression of the miR-22 with various clinical features of cervical cancer were analyzed (Table 1). We found that low miR-22 expression was correlated with FIGO stage and lymph node metastasis, suggesting that miR-22 might be involved in the carcinogenesis and metastasis of cervical cancer. In our present study, we confirmed the role of miR-22 in cervical cancer via its ability to inhibit cell colony formation, migration and invasion in vitro. We identified two tumor suppressor genes targeted by miR-22: matrix metallopeptidase 1(MMP1) and mitogen-activated protein kinase 1 (MAPK1). Out of these genes, MMP1 could have a specific suppressive impact as it suppresses cell proliferation and metastasis, correlates with cell apoptosis and proliferation, and is associated with lymph node metastasis in cervical cancer [17][18][19][20]. Thus we conclude that abnormally low expression of miR-22 in cervical cancer that specifically suppresses the function of MMP1 could be a common carcinogenetic mechanism in various cervical cancer. This may also help to explain why miR-22 is depressed in most cervical cancer tissue samples. However, the mechanisms of miR-22 regulation in cervical cancer are complex. Comprehensive studies on the carcinogenic effect of miR-22 are needed to be established as a biomarker for use in both clinical diagnosis and prognosis of patients with cervical cancer.