Background
Recently, miRNAs were found to play critical roles in many different cellular processes, especially in tumor progression. There are nearly 1000 miRNAs and more than 40,000 protein-coding genes in the human genome [
1,
2]. Consequently, it is more feasible to explore reliable miRNA biomarkers from genome-wide miRNA expression data than from genome-wide gene expression data. miRNAs are a class of short non-coding RNAs ranging from 19 to 25 nucleotides in length, which are transcribed as precursors and are matured to active forms by a series of enzymes, including Dicer [
3]. Numerous studies have explored the instrumental roles of these small, non-coding RNA species, mostly through base-pairing to the untranslated region (UTR) of target mRNA, thus leading to its degradation and/or reduced translation [
4]. Generally, an individual miRNA can regulate the expression of multiple target genes, and several miRNAs can synergistically act on one target gene, regulating cell differentiation, proliferation, mobility and apoptosis [
5]. The
miR-181 family contains four miRNAs (
miR-181a/b/c/d).
miR-181a and
miR-181b are transcribed from two separated gene loci (
miR-181a-1/miR-181b-1 and
miR-181a-2/miR-181b-2), while
miR-181c and
miR-181d are transcribed from another locus [
6]. It had been reported that
miR-181a,
-181b,
-181c and
-181d function differently in a tumor series. However, the homology among the miR-181 family members and the contribution of
miR-181a,
-181b,
-181c and
-181d in UM have not yet been clarified.
UM is the most frequent malignant intraocular cancer in adults, and up to 50% of UM patients are at risk of metastasis via hematogenous spread, most commonly to the liver [
7]. Recently, epigenetic events mediated by miRNAs have been implicated in UM development. UM proliferation and progression are regulated by dynamic interactions between UM-specific regulators, including miRNAs, whose aberrant expression has been associated with oncogenesis and tumor suppressor activity [
8]. Recent studies have implicated miRNAs in UM development. For example,
miR-20a functions as an oncogenic miRNA involved in promoting cell growth in UM, and
miR-454 promotes proliferation and invasion by regulating
PTEN in UM [
9,
10]. On the other hand,
miR-32 and
miR-124a both function as tumor suppressors by regulating multiple targets involved in UM development [
11,
12]. Moreover, growing evidence indicates that miRNA expression can potentially be used as a biomarker for the diagnosis and prognosis of different tumors. However, the expression and function of the
miR-181 family members in the pathogenesis of UM had not been established.
In the present study, the homology and function of
miR-181 family members,
miR-181a,
-181b,
-181c, and
-181d, were investigated.
miR-181 family members were found to be highly homologous and have the same target,
CTDSPL. The CTDSPL gene contains 8 exons coding for a 4.8 kb mRNA, which has been previously denoted as HYA22 and RBSP3, is a recently identified phosphatase-like tumor suppressor gene that dephosphorylates the Rb1 serine on Ser-807 and Ser-811 [
13]. The sequence analysis shows that CTDSPL belongs to a gene family of small C-terminal domain phosphatases that may control the RNA polymerase II transcription machinery [
14]. Then, the pattern of miRNA expression in melanoma tissues was analyzed using microarray technology. The microarray results indicated
miR-181b1 and
miR-181b2 were highly expressed in melanoma tissues. Furthermore,
miR-181b was found to be extremely overexpressed in most UM cells. These findings raised the possibility that
miR-181b might have an important role in UM development or pathogenesis. However, the molecular basis for this phenotype has not been elucidated, and the status of the downstream targets of
miR-181b in UM has not been researched. Therefore, a better understanding of the mechanisms responsible for UM and an exploration of the novel diagnostic and therapeutic strategies are crucial for achieving improved patient outcomes.
Methods
Cell culture and transfection
UM cells SP6.5, VUP, OCM1 and 92-1 were maintained in Dulbecco’s Modified Essential Medium (DMEM; Gibco, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS; Gibco) OCM1a and MUM2b were maintained in Iscove’s Modified Dulbecco’s Medium (IMDM; Gibco) with 10% FBS. The normal control cells, RPE, were maintained in DMEM with 10% FBS. Cultures were maintained at 37 °C in a 5% CO2 humidified atmosphere. Cells were treated and harvested for qRT-PCR and Western blot analysis. MUM2b (3 × 105) or OCM1a (5 × 105) cells were cultured overnight in 6-well plates and transfected with 200 nM miR-NC, miR-181 family mimics, or as-miR-181 family members (GenePharma Co. Ltd., Shanghai, China) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Two days later, these cells were either harvested for protein and mRNA or fixed using 70% ethanol for FCM.
Cell cycle analysis
Treated UM cells along with control cells were harvested. The cells were washed twice with cold phosphate-buffered saline (PBS), fixed in 70% ethanol and stored at 4 °C overnight. The next day, the cells were washed twice with cold PBS and incubated with propidium iodide/ribonuclease staining solution (Becton Dickinson, NJ, USA) for 15 min at room temperature, following the manufacturer’s instructions. Cell cycle distribution was detected and analyzed using the FACScan instrument and CellQuest program (Becton Dickinson, NJ, USA).
Western blot analysis
After the indicated treatments, the cells were washed with PBS and lysed with ice-cold lysis buffer (RIPA; Sigma Chemical Co, MO, USA). Cell lysates were incubated at 4 °C for 50 min. After centrifugation at 12,000 g for 1 min at 4 °C, protein concentration was determined by a BCA protein assay (Bio-Rad, Hercules, CA, USA). Thirty micrograms of protein were separated on 10% SDS–PAGE and transferred to a PVDF membrane. Membranes were probed with primary antibodies against CTDSPL (Abcam, Cambridge, UK) or E2F1 (Abcam, Cambridge, UK) at 4 °C overnight. Next, the membranes were washed three times with TBS containing 0.1% Tween-20 and incubated with secondary antibody for 1 h. The PVDF membrane was washed three times with Tris-Buffered Saline Tween-20 (TBST). After washing with TBST, the bands were detected using the Odyssey Infrared imaging system (Odyssey; LI-COR, Lincoln, NE).
Dual-luciferase reporter assay
To determine the common target region of the miR-181 family in CTDSPL, a segment of wild-type and mutated 3’-UTR of the human CTDSPL cDNA was constructed. Constructs were validated by sequencing. 293 T cells were plated in 24-well flat-bottomed plates and co-transfected with the wild-type or mutated 3’-UTR of the CTDSPL reporter plasmid, pRL-TK, and miR-181 family members or miR-NC using Lipofectamine 2000 reagent. Firefly and Renilla luciferase activities were determined 24 h after transfection using the dual-luciferase reporter assay system (Promega, Madison, WI, USA). The Renilla values were normalized to firefly luciferase.
Microarray and computational analysis
Briefly, RNA from tissue samples (three melanomas and three normal tissues) was used to synthesize double-stranded cDNA, and double-stranded cDNA was labeled and hybridized to the 2.0 microRNA Expression Microarray (Affymetrix GeneChip Human Gene 2.0 ST Array, Rockville, MD, USA). Raw data were extracted as pair files using NimbleScan software (version 2.5; Roche NimbleGen, Inc., Madison, WI, USA). NimbleScan software’s implementation of RMA offers quartile normalization and background correction. Differentially expressed genes were identified through the random variance model. The AP value was calculated using the paired t-test. The threshold set for up- and downregulated genes was a fold change > 2.0 and a P-value < 0.05. Hierarchical clustering was performed based on differentially expressed miRNAs using Cluster_Treeview software from Stanford University (Palo Alto, CA, USA).
Total RNA from UM cells was isolated using the Trizol reagent (Invitrogen, Carlsbad, CA, USA) following the protocol provided by the manufacturer. The cDNA synthesis reaction was performed according to the manufacturer’s protocol (TaKaRa Bio, Otsu, Japan). DPN1 enzyme (Sangon Biotech, Shanghai, China) was used to delete the genomic DNA from the extracted RNA, which was used to amplify the miRNAs. Screening for miRNAs was performed by qRT-PCR with the primer sets described in Table
1. PCR reactions were performed according to the manufacturer’s protocol (TaKaRa Bio) and were repeated at least three times for each sample. The miRNA loop primers were used first, and then the miRNAs PCR primers used. The relative levels of target gene miRNA transcripts to control
U6 were determined by the 2
-△△CT method.
Table 1
Primers used in this study
hsa-miR-181a-5p loop
| GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACTCACCG |
hsa-miR-181a-5p F
| TGCGCAACATTCAACGCTGTCG |
hsa-miR-181a-5p R
| CTCAAGTGTCGTGGAGTCGGCAA |
hsa-miR-181b-5p loop
| GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACCCACCG |
hsa-miR-181b-5p F
| TGCGCAACATTCATTGCTGTCG |
hsa-miR-181b-5p R
| CTCAAGTGTCGTGGAGTCGGCAA |
hsa-miR-181c-5p loop
| GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACTCACCG |
hsa-miR-181c-5p F
| TGCGCAACATTCAACCTGTCG |
hsa-miR-181c-5p R
| CTCAAGTGTCGTGGAGTCGGCAA |
hsa-miR-181d-5p loop
| GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACCCACCG |
hsa-miR-181d-5p F
| TGCGCAACATTCATTGTTGTCG |
hsa-miR-181d-5p R
| CTCAAGTGTCGTGGAGTCGGCAA |
U6 F
| CGCTTCGGCAGCACATATAC |
U6 R
| AAATATGGAACGCTTCACGA |
CTDSPL-F
| GTGGCTGACCTCCTAGACC |
CTDSPL-R
| TTCACGTAGTTCCCACGATGA |
GAPDH-F
| GGCTGTTGTCATACTTCTCATGG |
GAPDH-R
| GGCTGTTGTCATACTTCTCATGG |
CTDSPL-si1
| GCAGCAUCCUUAGCUCCUUTT |
CTDSPL-si2
| UCCACCAGCUAAGUACCUUTT |
Overexpressing miR-181b plasmid construction, lentivirus package, cloning and stable transfection in UM cells
The miR-181b sequence was amplified and sequenced without mutations. Then, 293 T cells were used to package the lentivirus. 293 T cells were cultured in DMEM supplemented with 10% FBS and maintained at 37 °C at a concentration of 6 × 106 cells/ml and transfected using Lipofectamine 2000 reagent with 3 μg PL-shRNA-HSA-MIR-181b-5p, 3 μg pMD2.D, and 6.0 μg PsPax. After incubation overnight with the 293 T cells, the media was replaced with 5 mL of fresh medium. The viral supernatants were concentrated and used to obtain stably transfected miR-181b-overexpressing UM cells. Stable MUM2b and OCM1a cell lines were established by lentiviral infection and blasticidin selection. The colonies with GFP expression were selected for subsequent culture after incubation with 4 g/mL blasticidin for 3 weeks. Transduction efficiency was determined by EGFP expression and measured by qRT-PCR.
Statistical analysis
All experiments were carried out in triplicate. All statistical analyses were performed using SPSS19.0 software. The statistical analysis was performed with a double-sided Student’s t-test for comparison of two groups. All data are expressed as the mean ± standard error of the mean (SEM). Differences at P < 0.05 were considered statistically significant.
Discussion
Recently, miRNAs have emerged as important cellular regulators that mediate cellular proliferation and progression. The
miR-
181 family consists of
miR-
181a,
miR-
181b,
miR-
181c, and
miR-
181d, and
miR-
181b, which divided into
miR-
181b1 and
miR-
181b2 and is transcribed from two separate gene loci [
6]. However, there are few studies on the correlations among
miR-
181 family members. Here, we explored the correlations among
miR-
181 family members. First, we found that these miRs are highly evolutionarily conserved in different species, including
Homo sapiens,
Mus musculus, Rattus norvegicus, Bos taurus and
Pan troglodytes (Fig.
1a), which suggests functional conservation. Cell cycle experiments demonstrated that pretreatment with
miR-
181 family mimics promoted cell cycle progression, while inhibitors resulted in cell cycle arrest (Fig.
1b). Similar effects usually shared the same mechanism and we know that small, non-coding RNA species mostly function through base-pairing to the untranslated region (UTR) of its target mRNA, leading to its degradation and/or reduced translation [
18]. To explore the mechanism of the
miR-
181 family members, we performed bioinformatics analysis, and CTDSPL was identified as a candidate target gene of the
miR-
181 family members through three publicly available algorithms (TargetScan, miRanda and Pictar). Additionally, it had been previously reported and confirmed that
miR-
181a can bind to the 3′-UTR of
CTDSPL mRNA. However, there are no relevant, precise studies regarding
miR-
181 [
6]. Here, we found that all the
miR-
181 family members have nearly the same binding site within the 3’-UTR of
CTDSPL. Additionally, we found five different binding sites (Fig.
2a). A firefly luciferase assay indicated that all
miR-
181 family members could inhibit CTDSPL expression through directly binding to the 3’-UTR of
CTDSPL. Consequently, this work confirmed that the
miR-
181 family members are highly conserved and share a common function of direct binding to the 3’-UTR of
CTDSPL.
The biological functions of the
miR-
181 family have been discussed in different tumors with different underlying biological processes.
miR-
181a was first identified and recognized as a contributor to hematopoietic lineage commitment and differentiation [
19,
20]. In breast cancer,
miR-
181a could prevent and reverse drug resistance via binding to the 3’-UTR of BCRP [
21]. It has been reported that
miR-
181a/b (
miR-
181a and
miR-
181b) suppress the translation of p300/CBP-associated factor (
PCAF) mRNA, a process relevant to the epigenetic fine-tuning of epithelial inflammatory processes in liver epithelial cells [
22]. There is another report that the loss of
miR-181a-1/b-1 dampens the induction of experimental autoimmune encephalomyelitis and reduces basal TCR signaling in peripheral T cells and their migration from lymph nodes to pathogenic sites [
23].
miR-
181c inhibits glioblastoma cell invasion, migration and mesenchymal transition by targeting the TGF-beta pathway and is associated with metastatic brain cancer and high-grade osteosarcoma [
24‐
26].
miR-
181d acts as a glioma suppressor by targeting K-ras and Bcl-2 [
27]. Compared with
miR-
181a,
miR-
181c and
miR-
181d,
miR-
181b was confirmed to be the most effective miRNA of the
miR-
181 family members [
4,
6]. Studies have demonstrated that
miR-
181b is overexpressed in several cancers, such as colorectal cancer, acute lymphocytic leukemia (ALL), acute promyelocytic leukemia and hepatocellular carcinoma [
28,
29]. Furthermore,
miR-
181b expression was found to be strongly associated with clinical response to S-1 in colon cancer patients [
30]. On the other hand, reduced
miR-
181b expression has been observed in several primary human cancers, including gastric, lung, and prostate cancer, and acute myeloid leukemia and chronic lymphocytic leukemia [
28,
31‐
33]. However, research regarding the
miR-
181 family is very scarce. They had been observed in glioblastoma, which could reverse mesenchymal transition by targeting
KPNA4 [
34]. At present, there are no investigations exploring their functions and mechanisms in UM pathogenesis and development. Here, microarray technology indicated there are different expression profiles of
miR-
181a,
miR-
181b,
miR-
181c, and
miR-
181d in UM.
miR-
181b1 and
miR-
181b2 were significantly overexpressed in melanoma tissues, while there were no significant changes in
miR-
181c and
miR-
181d. We also explored the function of these four miRNAs in UM cells. In accordance with the microarray results, we found
miR-
181b was extremely upregulated in most UM cell lines and that
miR-
181a was also upregulated in most UM cell lines. However,
miR-
181c and
miR-
181d were not upregulated in most UM cell lines, except for a slight increase of
miR-
181c in OCM1 and
miR-
181d in OCM1a. Given the different expression patterns of the
miR-
181 family members, we can assume that they function differently in different UM tissues and cell lines. Interestingly, there was no downregulation of any
miR-
181 family members in the UM cells compared with the control group. Thus, there is expression specificity of
miR-
181 family members in UM, but a lack of universality. In our study,
miR-
181b was especially interesting.
miR-
181b displayed the highest degree of expression in melanoma tissues and UM cell lines, and several previous studies have revealed that
miR-
181b is more active [
4,
6] and has an intimate relationship with human malignant tumors, including hepatocellular carcinoma, colorectal gastric, lung, and prostate cancer, and ALL, acute myeloid leukemia, and chronic lymphocytic leukemia [
28,
29,
31‐
33].
In our study,
miR-
181b was significantly upregulated in most UM cell lines, specifically VUP, SP6.5, OCM1 and 92-1, but was not upregulated in MUM2b or OCM1a cells. We then transfected MUM2b and OCM1a cells with a
miR-
181b overexpression plasmid. Compared with the control group, the cell cycle was at a later stage in miR-181b-overexpressing MUM2b and OCM1a cells. Bioinformatic analyses and dual-luciferase reporter assays demonstrated and confirmed that
CTDSPL was the target of the
miR-
181 family members. A significant downregulation of CTDSPL and upregulation of E2F1 in MUM2b-over-
miR-
181b and OCM1a-over-
miR-
181b cells was confirmed. The underlying molecular pathway responsible for the effects of
miR-
181b in UM cell survival might control the G0/G1 to S phase transition through the repression of CTDSPL.
CTDSPL is an important phosphatase-like tumor suppressor gene located at 3p21.3, and belongs to the small C-terminal domain phosphatase family, which modulates the RB/E2F1 signaling pathway and results in cell cycle arrest at the G1/S boundary [
13,
35]. Previous studies have also reported that CTDSPL removes the phosphate group from serines 807 and 811 in its substrate, pRB, and thereby induces the formation of the RB/E2F1 complex [
36,
37]. It had been reported that
miR-100 regulates myeloid differentiation by targeting
CTDSPL [
38]. However, the involvement of
CTDSPL in the regulation of cell growth in UM cells has not yet been studied. Our data demonstrated that overexpressed
miR-
181b knocked down CTDSPL expression and resulted in an accelerated G1/S transition in UM cells. These results indicate that overexpressed
miR-
181b inactivates the phosphatase CTDSPL protein and that this inactivation may be a common step that is required for UM progression.
In this work,
miR-
181b was found to control the G0/G1 to S phase transition by repressing CTDSPL and regulating E2F1 expression in most UM cells, except for MUM2b and OCM1a cells. This finding suggests that
miR-181b expression is specifically increased or unchanged, without downregulation, in all UM cells, but there is lack of universality in UM. There were other reports that CTDSPL can directly bind to Rb just like the mechanism demonstrated by Beniaminov et al. [
39,
40]. It showed a new method called surface plasmon resonance (SPR) to detect the direct interaction. However, in our co-immunoprecipitation assay, no interaction between CTDSPL and RB1 was found in OCM1a cell line (Additional file
1). Previous studies showed that cyclic phosphorylation/dephosphorylation of the pRb protein plays an important role in decreasing ppRb levels, ultimately blocks the G1/S progression. In our study, CTDSPL is another important regulator of ppRb level through phosphatase activity [
14,
41,
42], showing as another parallel pathway, without relationship exists between cyclins and CTDSPL, at least in OCM1a cell line (Additional file
1).
miR-
181 family members have been reported as potential therapeutic targets for myeloid dysplastic syndrome and acute myeloid leukemia [
6]. Although
miR-
181c and -
181d mimics and inhibitors could promote and inhibit CTDSPL expression through the same binding site in the
CTDSPL gene,
miR-181c and
miR-
181d were not upregulated in most UM cells except for a slight increase in miR-181c in OCM1 cells and miR-181d in OCM1a cells. This finding suggests that
miR-
181c and
miR-
181d do not play leading roles in UM cells but rather that
miR-
181c and
181d have support and backup functions for
miR-
181b, through binding to the 3’-UTR of
CTDSPL and inhibiting its expression. In other words, while one of the
miR-
181 family members may be the primary functional miRNA in one tumor, the other
miR-
181 family members may assist it. Finally,
miR-
181 family members, and especially
miR-
181b overexpression, could be used as therapeutic targets for UM. Our results characterize a new role for
miR-
181 family members. It is regrettable that
miR-
181 family members were not detected in the limited UM tissues available, except via the microarray chip. Going forward, we hope to detect the expression of
miR-
181 in additional UM tissues. The prognostic and therapeutic value of
miR-
181 family members, especially
miR-
181b, needs to be confirmed in UM patients and other tumor types in the future. Moreover,
miR-
181 expression would be ideally detectable in the blood of UM patients, which would be useful in the future for UM patient diagnosis and prognosis.