MicroRNA-124a Is Epigenetically Regulated and Acts as a Tumor Suppressor by Controlling Multiple Targets in Uveal Melanoma

Abstract

Purpose.: MicroRNA-124a (miR-124a), an abundant microRNA in the central neuron system, has been linked to tumor progression. Here, we investigated the role of miR-124a in uveal melanoma development.

Methods.: Expression of miR-124a in uveal melanoma cells was examined using real time RT-PCR. The effect of miR-124a on cell proliferation, migration, and invasion was analyzed using MTS assay, flow cytometry, and transwell experiments. The ability of miR-124a to repress tumor growth was tested in vivo. Target genes of miR-124a were first predicted by bioinformatics, confirmed using a luciferase assay, and their expression determined by Western blotting. DNA methylation and histone modification of miR-124a was analyzed by methylation-specific PCR and ChIP assay. Finally, epigenetic drugs were used to alter the expression of miR-124a.

Results.: miR-124a expression was downregulated in both uveal melanoma cells and clinical specimens. Transient transfection of miR-124a into uveal melanoma cells inhibited cell growth, migration, and invasion. Moreover, introduction of miR-124a suppressed in vivo growth of tumor. Potential targets of miR-124a were found to include CDK4, CDK6, cyclin D2, and EZH2. Knockdown of EZH2 by siRNA resulted in inhibition of uveal melanoma cell migration and invasion. In addition, miR-124a expression was found to be regulated via epigenetic mechanisms, with its expression restored when cells were treated with a DNA hypomethylating agent, 5-aza-2′-deoxycytidine, and a histone deacetylase inhibitor, trichostatin A.

Conclusions.: Our results demonstrated that miR-124a could function as a potent tumor suppressor by regulation of multiple targets, and was epigenetically silenced in the development of uveal melanoma.

Introduction

Uveal melanoma is an uncommon disease that behaves in a clinically distinct manner from cutaneous melanoma. Studies into posterior uveal melanomas, which comprise of choroidal and ciliary body melanomas, reveal that these tumors have genetic variations from cutaneous melanomas and a clinical course distinct from cutaneous melanomas. ¹ Uveal melanoma has a propensity for hematogenous metastases to the liver in up to 50% of patients, with many having subclinical evidence of metastases early on. Genetically, they are different from cutaneous melanomas, having been associated with the presence of GNAQ/GNA11 mutations ^2,3 and BAP1 mutations. ⁴ As a corollary, some new biologic agents effective against metastatic melanomas are active only against tumors that harbor BRAF (V600E), mutation found from metastatic cutaneous melanomas. ⁵ In exploring causes and potential therapeutic avenues in uveal melanoma, we have concentrated and previously identified several microRNAs (miRNAs) that directly affect uveal melanoma progression through miRNA inactivation. ^6,7 Various techniques of miRNA silencing coupled with downstream upregulation of intracellular proteins confirmed our hypothesis that miRNA has a role in the suppression of uveal melanoma. ^6,7  

miRNAs were first discovered in 1993 as a developmental modulator in the nematode, Caenorhabditis elegans . ⁸ Soon, with the understanding that miRNAs can have differential spatial and temporal expression, they were found to be excellent regulators of gene expression through binding of complementary sites on their target transcripts. ⁹ miRNAs were then naturally associated with cancer development. ¹⁰ Various modes of regulation were identified to control the expression pattern of miRNAs. ¹¹ The disruption of this fine control via epigenetic mechanisms can also lead to cancer development. There are three main avenues to control epigenetic mechanisms, namely DNA hypermethylation, global genomic hypomethylation, and histone modification changes. ¹² Several previously demonstrated tumor suppressor miRNAs, including miR-124a and miR-137, were found to be targets of epigenetic regulation. ¹³ These miRNAs target oncogenes, such as CDK6, but can be suppressed through epigenetic regulation of DNA, such as hypermethylation. ^11,14  

Our laboratory has identified various miRNAs, including miR-34a, miR-137, and miR-182, that have activity in regulating uveal melanoma development. ^6,7,15 Specifically, we have demonstrated that miR-137 is regulated by epigenetic mechanisms in the previous study. ⁷ This prompted us to explore other epigenetically regulated miRNAs in uveal melanoma. Recently, our attention has been redirected toward miR-124a, one of the most abundant miRNAs in the central nervous system and the retina. ¹⁶ Predicted targets of miR-124a are found to be involved in organ development and retinal development, making this an excellent avenue for further investigation. ¹⁶ Herein, we aimed to evaluate the function of miR-124a in the development of posterior uveal melanoma, which has a neural crest origin. 

Materials and Methods

Cell Culture and Tumor Specimens

The human uveal melanoma cell lines M17, M21, M23, and SP6.5, isolated from Caucasian patients with primary choroidal melanoma, were kindly provided by Drs Dan-Ning Hu and Guy Pelletier (Research Center of Immunology, Quebec, Canada) ^17,18 and grown as previously described. ^6,7 HEK-293 cells were purchased from ATCC (Manassas, VA). The primary human uveal melanocytes um95 were isolated from a Chinese donor at Wenzhou Medical College (Wenzhou, China), and cultured as described. ¹⁹ Six primary uveal melanoma specimens, as well as their adjacent uveal tissue counterparts, were obtained from patients treated at the Eye Hospital, Wenzhou Medical College, and the First Affiliated Hospital of School of Medicine, Zhejiang University (Hangzhou, China), with documented informed consent in each case. All uveal melanomas studied here were choroidal melanomas. After surgery, tumor specimens and the adjacent uveal tissues were first snap-frozen in liquid nitrogen and then stored at −80°C for subsequent analysis. All studies and procedures involving human tissue were approved by the Wenzhou Medical college ethics committee, and performed in compliance with the Helsinki Declaration and national laws. 

Quantitative RT-PCR

Total RNA was extracted from cells with Trizol reagent (Invitrogen, Carlsbad, CA) and the integrity was confirmed; 10 ng of total RNA were used for cDNA synthesis by the Taqman MicroRNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA), and miR-124a expression level was quantified by the Taqman MicroRNA Assay (Applied Biosystems), according to the manufacturer's instructions. Real-time RT-PCR was performed using the Applied Biosystems 7500 Fast Real-Time PCR System, the expression of miR-124a was normalized to the expression of U6 small nuclear RNA (snRNA), and relative expression levels were calculated as previously reported. ²⁰  

DNA Methylation Analysis

Methylation-specific PCR was performed to evaluate the DNA methylation status of promoter regions in three genomic loci: has-miR-124a-1, has-miR-124a-2, and hsa-miR-124a-3. First, genomic DNA was extracted from uveal melanoma cell lines and clinical samples, followed by treatment with sodium bisulfite. Specific primers were designed to amplify the methylated DNA region in three miR-124a promoters. Real-time PCR was performed to quantify DNA methylation status by measuring SYBR-Green incorporation (SYBR-Green PCR Master Mix; Applied Biosystems) on the Applied Biosystems 7500 Fast Real-Time PCR System. The modified, unmethylated sequence of the housekeeping gene β-actin was used as a reference. 

DNA Demethylation and Deacetylation Assays

The demethylation and deacetylation assays were performed as previously described. ⁷ Briefly, M17, M23, and SP6.5 cell lines were seeded in a six-well plate. Then 1 or 5 μM 5-aza-2′-deoxycytidine (5-aza-dC) (Sigma-Aldrich, St. Louis, MO), was added prior to culturing the cells for 48 hours. Alternatively, trichostatin A (TSA, 100 ng/mL; Sigma-Aldrich), was added and cells were cultured for 12 hours. For the combination study, 1 or 5 μM 5-aza-dC was present for the first 48 hours and TSA (100 ng/mL) was then added before culturing for another 12 hours. The media containing 5-aza-dC were changed every 24 hours. Then real-time RT-PCR assay for miR-124a was performed as previously described. 

Chromatin Immunoprecipitation Assay

Chromatin immunoprecipitation (ChIP) assay was performed using the ChIP Assay Kit (Millipore, Billerica, MA) following the manufacturer's instructions. Briefly, immunoprecipitated DNA fractions were isolated from uveal melanoma cell lines and clinical samples, using the antibody against acetylated histone 3 (AcH3; Millipore) or the antibody against acetylated histone 4 (AcH4; Millipore). Specific primers were designed to amplify the miR-124a promoter region in three genomic loci. Real-time PCR was performed to quantify histone acetylation status by measuring SYBR-Green incorporation (SYBR-Green PCR Master Mix; Applied Biosystems) on the Applied Biosystems 7500 Fast Real-Time PCR System. 

Cell Proliferation Assay

M23 and SP6.5 cells were plated at 3 × 10³ cells per well in 96-well plates (Costar, High Wycombe, UK) for each transfection. Transfections were performed using Lipofectamine 2000 (Invitrogen). For each well, 50 nM of miR-124a precursor molecule (Ambion, Austin, TX) or a negative control precursor miRNA (Ambion) was transfected. After 24-hour culture, cell proliferation was assessed using the CellTiter 96 AQueous assay kit (Promega, Madison, WI) according to the manufacturer's instructions. 

Flow Cytometry Analysis

Forty-eight hours after transfection with 50 nM of miRNAs, M23 and SP6.5 cells (1 × 10⁵) were stained with propidium iodide using the Cycle Test Plus DNA Reagent Kit (Becton Dickinson, San Jose, CA) and then analyzed for DNA content with a flow cytometer (FACScaliber; Becton Dickinson). 

Transwell Migration and Matrigel Invasion Assays

M23 and SP6.5 cells were grown to approximately 60% confluence and transfected with 50 nM of miRNAs, and harvested by trypsinization in 24 hours. To measure cell migration, 8-mm pore size culture inserts (Transwell; Costar) were placed into the wells of 24-well culture plates. To measure cell invasion, 8-mm pore size Matrigel inserts (Becton Dickinson) were used. The Matrigel inserts were rehydrated before use. In both experiments, 400 μL of Dulbecco's modified Eagle medium (DMEM) containing recombinant human hepatocyte growth factor (HGF; 20 ng/mL; R&D Systems, Minneapolis, MN) were added in the lower chamber; 1 × 10⁵ cells were then added to the upper chamber. After 24 hours of incubation, the number of cells that had migrated through the pores was quantified by counting 10 independent visual fields under the microscope (Zeiss, Oberkochen, Germany) using a ×20 objective. EZH2-specific small interfering RNA (siRNA; Ambion) and negative control siRNA (Ambion) were used to downregulate EZH2 expression in uveal melanoma cells; 50 nM of EZH2-specific siRNA or negative control was transfected into M23 and SP6.5 cells with Lipofectamine 2000 (Invitrogen). Transwell migration and Matrigel invasion assays were carried out, as described above. 

Luciferase Reporter Assays

The 3′ untranslated region (UTR) of human CDK4, CDK6, cyclin D2, or EZH2 was amplified from human genomic DNA and individually cloned into pMIR-REPORT vector (Ambion) by directional cloning. Seed regions were mutated to remove all complementarity to nucleotides 7 to 8 of miR-124a by using the QuickchangeXL Mutagenesis Kit (Stratagene, La Jolla, CA). HEK-293 cells were cotransfected with 0.4 μg of firefly luciferase reporter vector and 0.02 μg of the control vector containing Renilla luciferase, pRL-SV40 (Promega), using Lipofectamine 2000 (Invitrogen) in 24-well plates (Costar). Each transfection was carried out in four wells. For each well, 50 nM of miRNAs was cotransfected with the reporter construct. Luciferase assays were performed 24 hours after transfection using the Dual Luciferase Reporter Assay System (Promega). Firefly luciferase activity was normalized to Renilla luciferase activity. 

Western Blot Analysis

M23 and SP6.5 cells (1 × 10⁵) were seeded and grown in DMEM with 10% fetal bovine serum in six-well plates for 24 hours. After transfection, the cells were washed and Western blot was carried out as previously described. ^6,7 The membranes were immunoblotted with antibodies for total cyclin D2, EZH2, E2F1, phosphorylated-cdc2 (threonine 161), CDK4, CDK6, and phosphorylated-Rb (serine 795) (Cell Signaling Technology, Beverly, MA). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a loading control. 

In Vivo Tumor Growth Assay

The pre-miRNA expression construct lenti-miR-124a and pCDH-CMV-MCS-EF1-copGFP control vector were purchased from System Biosciences (Mountain View, CA). The lentivirus was produced according to the manufacturer's instructions. M23 cells and SP6.5 cells were infected with lentivirus expressing miR-124a or a negative control. Female nude mice at 6 weeks of age were used for xenograft studies. All animal treatments were carried out in strict accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and approval of the Wenzhou Medical College Animal Care and Use Committee. M23 cells (5 × 10⁶) or SP6.5 cells (5 × 10⁶) expressing miR-124a or the negative control were inoculated subcutaneously into the flanks of nude mice. All mice were killed 4 weeks later. Tumor size was measured with a caliper, and the volume was calculated using the following formula: (L × W ²) × 0.5, where L is length, and W is width. ²¹  

Statistical Analysis

All data are shown as the mean ± SEM. Differences between cells transfected with miR-124a and a negative control were analyzed using the Student's t-test. Statistical significance was accepted at P less than 0.05. 

Results

miR-124a Is Downregulated in Uveal Melanoma Cells and Clinical Specimens

To determine if miR-124a was involved in the regulation of tumorigenesis of uveal melanoma cells, we first compared miR-124a expression in normal melanocytes and uveal melanoma cells. Real-time RT-PCR was performed to detect miR-124a expression in uveal melanoma cell lines, including M17, M23, and SP6.5, as well as the normal uveal melanocytes um95. miR-124a was expressed in uveal melanocytes (Fig. 1A). In contrast, expression of miR-124a was significantly decreased in all the uveal melanoma cell lines examined (Fig. 1A). Expression pattern of miR-124a in six human uveal melanoma specimens also showed downregulation in comparison with adjacent uveal tissues, used as a normal control (Fig. 1B). 

Figure 1

 

miR-124a expression is downregulated in uveal melanoma cells. (A) Real-time RT-PCR analysis was performed to detect the expression of miR-124a in uveal melanoma cell lines including M17, M23, and SP6.5, as well as primary uveal melanocytes um95. (B) Real-time RT-PCR analysis was performed to detect the expression of miR-124a in six uveal melanoma specimens and normal tissues. The value for miR-124a in normal tissues was set at 1, and the relative amount of miR-124a in the tumors was plotted as fold induction. N, normal tissues; T, tumors. U6 snRNA was used as an internal control. *Differences in miR-124a expression between uveal melanocytes and uveal melanoma cells were significant, P < 0.01.

miR-124a Inhibits Uveal Melanoma Cell Proliferation, Migration, and Invasion

Transfection with miR-124a caused a dramatic inhibition of M23 and SP6.5 cell growth when compared with that of control over a 5-day interval (Fig. 2A). The decrease in cell number was statistically significant between cells transfected with miR-124a and cells transfected with a negative control at day 5 (88.82% ± 1.33% decrease in M23 cells and 84.20% ± 1.14% decrease in SP6.5 cells, P < 0.01, n = 3). Complementary to the finding that miR-124a inhibited cell proliferation, miR-124a was found to cause increased G1 cell cycle arrest in these cells. M23 cells transfected with miR-124a showed 86.22% G1 arrest in comparison with 51.8% for negative control. SP6.5 cells transfected with miR-124a showed 83.04% G1 arrest in comparison with 50.48% with negative control (Fig. 2B). Furthermore, ectopic expression of miR-124a dramatically suppressed colony formation as compared with that of negative control (Fig. 2C). 

Figure 2

 

Ectopic miR-124a induces G1-arrest and inhibits cell proliferation, migration, and invasion. (A) MTS cell proliferation assay was carried out on days 1 to 5 as indicated after lipofectamine transfection of uveal melanoma cells M23 and SP6.5 with either miR-124a (50 nM) or a negative control (NC) scrambled oligonucleotide. The data at each time point are expressed as the mean value ± SEM of the results obtained from triplicates in one experiment. Results represent those obtained in three separate experiments. (B) M23 and SP6.5 cells were collected 48 hours after transfection with miR-124a or NC, stained with propidium iodide, and analyzed by flow cytometry. A total of 10,000 cells were evaluated in each sample. The most representative results in three independent experiments are depicted. (C) M23 and SP6.5 cells transfected with miR-124a or NC were seeded at low density. After 7 days, colony formation was determined by staining with crystal violet. Typical results in three independent experiments are shown. M23 and SP6.5 were transfected with miR-124a or NC for 24 hours and plated on either culture or Matrigel inserts in DMEM medium containing 20 ng/mL of HGF to assess the number of migratory or invasive cells. The number of cells that had migrated through the culture insert pores (D) or had invaded through the Matrigel insert pores (E) was quantified. Results represent those obtained in three experiments. *Differences in cell migration or invasion between miR-124a and negative control transfected cells were significant, P < 0.01.

In addition, cells transfected with miR-124a were assessed for their ability to migrate and invade through transwell experiments. As shown in Figure 2D, the HGF-induced migration was significantly decreased when comparing miR-124a transfected cells with negative control transfected cells (192 ± 17 vs. 406 ± 28 in M23 cells, and 136 ± 10 vs. 227 ± 15 in SP6.5, P < 0.01, n = 3). Figure 2E shows that HGF-induced invasiveness was also significantly hampered following miR-124a transfection (79 ± 11 vs. 172 ± 16 in M23 cells, and 46 ± 6 vs. 97 ± 8 in SP6.5, P < 0.01, n = 3). 

Overexpression of miR-124a Suppresses Tumor Growth In Vivo

We next investigated if overexpression of miR-124a could repress tumor growth in vivo. M23 and SP6.5 cells infected with lentivirus expressing miR-124a or a negative control were injected subcutaneously into the flanks of nude mice. After 4 weeks, the averaged tumor volumes were significantly lower in cells infected with lentivirus expressing miR-124a, as compared with the control (Fig. 3). The results demonstrated that overexpression of miR-124a suppresses the growth of uveal melanoma in vivo as well. 

Figure 3

 

Introduction of miR-124a in uveal melanoma cells suppresses tumor growth in nude mice. (A) Representative photographs of nude mice 4 weeks after inoculation with miR-124a or control lentivirus-infected uveal melanoma cells. (a) Inoculation with M23 cells; (b) inoculation with SP6.5 cells. (B) Average volume of tumors derived from M23 or SP6.5 cells infected with miR-124a or control lentivirus in nude mice. *Differences in tumor volume between miR-124a and control infected cells were significant, n = 6 each, P < 0.01 for both M23 and SP6.5 cell inoculation.

CDK4, CDK6, Cyclin D2, and EZH2 Are Targets of miRNA-124a

To identify gene targets of miR-124a, we searched public algorithms, TargetScan (http://www.targetscan.org), for theoretical target genes whose downregulation could mediate the observed effects of miR-124a. CDK4 (cyclin-dependent kinase 4), CDK6 (cyclin-dependent kinase 6), cyclin D2, and EZH2 (enhancer of zeste homolog 2) were all predicted targets, of which CDK6 and EZH2 have previously been validated as subject to control by miR-124a in the development of colon cancer ¹⁶ and hepatocellular carcinoma. ²² We found by bioinformatics analysis that the CDK4 3′-UTR contains one target sequence for miR-124a at position 130–136. CDK6 contains four target sequences at positions 1533–1539, 1648–1654, 7788–7795, and 8004–8010. Cyclin D2 contains two target sequences at positions 3810–3816 and 4728–4734. EZH2 contains one target sequence at position 36–42 (Fig. 4A). To further confirm target specificity between miR-124a and the predicted genes, we performed a luciferase assay with a vector containing the putative target gene 3′ UTR downstream of a luciferase reporter gene. To test if miR-124a directly targets CDK4, CDK6, cyclin D2, and EZH2 genes, we cloned the wild-type 3′ UTR of each gene into a luciferase reporter vector, exemplified by Figure 4B. We then transfected each resulting reporter construct (pLuc-CDK4 3′ UTR, pLuc-CDK6 3′ UTR, pLuc-CCND2 3′ UTR, and pLuc-EZH2 3′ UTR) into HEK293 cells, along with miR-124a or a negative control. The luciferase activity assays at 24 hours after transfection demonstrated that miR-124a suppressed luciferase reporter activity to 44.66% ± 5.78%, 42.02% ± 3.64%, 65.02% ± 4.39%, and 52.44% ± 2.74% using 3′ UTR of CDK4, CDK6, cyclin D2, and EZH2, respectively (Fig. 4C). Mutational reporter constructs of each of the four genes showed these mutations would eliminate the suppression of the luciferase reporter activity (Fig. 4C). 

Figure 4

 

CDK4, CDK6, cyclin D2, and EZH2 are targets of miR-124a. (A) Specific locations of the binding sites were marked with red color and CDK4, CDK6, cyclin D2 (CCND2), EZH2 3′UTR were marked with blue color. Alignment between the predicted miR-124a target sites and miR-124a, the conserved 7 to 8 bp “seed” sequence for miR-124a:mRNA pairing is indicated. (B) Diagram depicting the pMIR luciferase reporter constructs, containing a CMV promoter, which was used to verify the putative miR-124a binding sites. (C) HEK293 cells were cotransfected with miR-124a, pLuc-CDK4 3′UTR, pLuc-CDK6 3′UTR, pLuc-CCND2 3′UTR, or pLuc-EZH2 3′UTR along with a pRL-SV40 reporter plasmid. After 24 hours, the luciferase activity was measured. Values are presented as relative luciferase activity after normalization to Renilla luciferase activity. Results represent those obtained in three separate experiments. *Differences in luciferase activity between miR-124a and negative control transfected cells were significant, P < 0.01. (D) miR-124a downregulates the expression of CDK4, CDK6, cyclin D2 (CCND2), EZH2, and other cell cycle–related proteins including E2F1, phosphorylated-cdc2 (p-cdc2) and phosphorylated-Rb (p-Rb), in both M23 and SP6.5 cells. Cell lysates were prepared and used for Western blot analysis with multiple antibodies. GAPDH was used as an internal control.

To examine the target proteins affected by miR-124a in uveal melanoma cells, we next determined the expression of targets and other cell cycle–related proteins by Western blot. Primary targets of miR-124a, including CDK4, CDK6, cyclin D2, and EZH2, were dramatically suppressed by miR-124a, as expected (Fig. 4D). All these targets were commonly upregulated in uveal melanoma cells (See Supplementary Material and Supplementary Fig. S1). Moreover, ectopic miR-124a also downregulated cell cycle regulatory proteins, such as E2F1, phosphorylated-cdc2, and phosphorylated-retinoblastoma protein in both M23 and SP6.5 cells (Fig. 4D). These results demonstrated that introduction of miR-124a suppressed CDK4, CDK6, cyclin D2, and EZH2, as well as cell cycle–related gene expressions, thus inhibiting cell proliferation. 

Downregulation of EZH2 Inhibits Uveal Melanoma Cell Migration and Invasion

Because EZH2 is a direct target of miR-124a, we next investigated the effect of EZH2 on uveal melanoma cells. EZH2-specific siRNA was used to decrease the expression of EZH2 in both M23 and SP6.5 cells (Fig. 5A). As indicated in Figure 5B, HGF-induced migration was significantly decreased when comparing EZH2 siRNA-transfected cells with negative control transfected cells (206 ± 18 vs. 403 ± 36 in M23 cells, and 142 ± 9 vs. 225 ± 14 in SP6.5, P < 0.01, n = 3). Figure 5C shows that HGF-induced invasiveness was also significantly hampered following EZH2 siRNA transfection (92 ± 9 vs. 178 ± 14 in M23 cells, and 52 ± 6 vs. 109 ± 10 in SP6.5, P < 0.01, n = 3). 

Figure 5

 

Downregulation of EZH2 inhibits cell migration and invasion. M23 and SP6.5 cells were transfected with EZH2 siRNA or a negative control (NC). (A) EZH2 expression levels in M23 and SP6.5 cells after transfection with EZH2-specific siRNA or NC were determined by Western blot analysis. GAPDH was used as an internal control. The number of cells that had migrated through the culture insert pores (B) and invaded through the Matrigel insert pores (C) was quantified. Results represent those obtained in three experiments. *Differences in cell migration or invasion between EZH2 siRNA and negative control transfected cells were significant, P < 0.01.

The Expression of miR-124a in Uveal Melanoma Is Regulated by DNA Methylation and Histone Modification

To determine if epigenetic mechanism was responsible for miR-124a regulation in uveal melanoma, we compared the expression levels of miR-124a in different uveal melanoma cells M17, M23, and SP6.5 treated with either the DNA hypomethylating agent, 5′-aza-2′-deoxycytidine (5-aza-dC), or the histone deacetylase inhibitor, trichostatin A (TSA), or combination of both. Expression of miR-124a was increased after treatment with 5-aza-dC with maximal induction observed in cells treated with only 5 μM of 5-aza-dC (Fig. 6A). Expression of miR-124a was also increased after treatment with 100 ng/mL of TSA, especially in SP6.5 cells. Furthermore, the effect of the drug combination seems to be additive. 

Figure 6

 

Epigenetic regulation of miR-124a in uveal melanoma cells and clinical samples. (A) Uveal melanoma cell lines including M17, M23, and SP6.5 were treated with 5-aza-dC (5Aza) at 1 μM or 5 μM alone, TSA (100 ng/mL) alone, or combinations of both. miR-124a expression level was measured by real-time RT-PCR relative to U6 snRNA. Methylation-specific real-time PCR analyses for miR-124a methylation in uveal melanoma cell lines and uveal melanocytes (B) and in clinical specimens (D). The value for miR-124a methylation in normal tissue was set at 1, and the relative amounts of miR-124a methylation in tumor tissues were plotted as fold induction. ChIP and real-time PCR analyses for miR-124a acetylation in uveal melanoma cell lines and uveal melanocytes (C) and in clinical specimens (E). The value for miR-124a acetylation in uveal melanocytes or normal tissues was set at 1, and the relative amounts of miR-124a acetylation in tumor tissues were plotted as fold induction.

miR-124a is transcribed in three genomic loci (miR-124a-1 [8p23.1], miR-124a-2 [8q12.3], and miR-124a-3 [20q13.33]). In the three uveal melanoma cell lines, the corresponding CpG island was much more methylated than in normal uveal melanocytes (Fig. 6B). The hypermethylation status of miR-124a was not only a notable feature of the particular uveal melanoma cell lines, but also observed in the clinical samples (Fig. 6D). Moreover, a remarkable decrease of AcH3 and AcH4 (Figs. 6C, 6E), which are associated with a closed chromatin structure leading to gene silencing, was also observed. These results strongly indicated that DNA methylation and chromatin modification are responsible, at least in part, for the abnormal downregulation of miR-124a in uveal melanoma cells. 

Discussion

Emerging evidence indicates that miRNAs are a major category of regulators that fine tune the expression of protein-coding genes. ⁹ Because the expression of miRNAs is both organ and developmental stage specific, single miRNA can affect and regulate hundreds of genes. ⁹ When miRNA was discovered to be involved in oncogenesis with leukemia, ²³ tremendous interest was generated in identifying specific miRNAs that affect tumor development and modes of regulation for those specific miRNAs. 

Initially, miR-124a was described as a brain-specific miRNA in mammals, the function of which is presumably to help define and maintain neuron cell–specific characteristics. ^24,25 Pre–miR-124a is preferentially expressed in brain, pancreas, kidney, and muscle. ²⁶ miR-124a has also been described to help maintain cell-specific characteristics of retinal cells. ²⁷ Functionally, multiple cellular targets of miRNA-124a have been identified. ²⁸ Neuronal tumor tissue, such as glioblastoma, has depressed miR-124a, which leads to increased tumor cell migration and invasion. ²⁹ Combined, this makes miR-124a a likely gene regulating uveal melanoma development. In this study, we found that miR-124a was downregulated in uveal melanoma cells. Similarly, downregulation of miR-124a was reported in cutaneous melanoma as well. ³⁰ miR-124a regulates CDK6 in medulloblastoma and modulates medulloblastoma cell growth. ³¹ In rheumatoid arthritis, miRNA-124a leads to cell cycle arrest at the G1 phase by targeting CDK2 and monocyte chemoattractant protein 1 (MCP-1), resulting in suppressed synovial cell proliferation. ³² In hepatocellular carcinoma, miR-124a regulates hepatocellular carcinoma cell invasion and metastasis by directly targeting ROCK2 and EZH2. ²² Some of these targets were studied and confirmed in this study, including CDK6 and EZH2 (Fig. 4). EZH2 has been reported to be overexpressed in metastatic prostate cancer, and its knockdown decreased prostate cell invasion. ³³  

Our initial studies showed that introduction of miR-124a slowed the growth of melanoma cell lines and inhibited their migratory capabilities (Fig. 2). This led us to look into the targets of miR-124a. With the aid of bioinformatics, multiple intracellular targets were identified and confirmed using both luciferase assay and Western blot analysis. Among these target genes, CDK4, CDK6, and cyclin D2 are key proteins in the regulation of cell cycle progress. ^34,35 CDK6 has also been identified as a direct target of miR-124a in colon, glioblastoma, and leukemia cells. ^11,13,14 Furthermore, we have identified two other targets of miR-124a, CDK4 and cyclin D2, which play essential roles in cell cycle G1 phase. EZH2, a polycomb group protein serving as histone lysine methyltransferase, has been linked to cutaneous melanoma progression. ^36,37 Specific EZH2 functional assays were performed in this study to confirm that EZH2 can inhibit uveal melanoma cell migration and invasion (Fig. 5). 

Downregulation of subsets of miRNAs is a common finding in certain cancers, suggesting that some of these miRNAs may act as putative tumor suppressor genes. Analysis of a comprehensive collection of human cancer cell lines and primary samples from colon, breast, and lung carcinomas, leukemias, and lymphomas showed a frequent presence of miR-124a hypermethylation. ^11,14 To determine whether the epigenetic silencing of miR-124a had functional cancer relevance, we therefore examined its effect in the regulation of presumed target genes with oncogenic capacity. miR-124a undergoes transcriptional inactivation by CpG island hypermethylation in human tumors from different cell types. ¹¹ Functionally, the epigenetic loss of miR-124a is linked with the activation of CDK6, an oncogenic factor, as well as phosphorylation of the Rb, a tumor suppressor gene. miR-124a is represented in three genomic loci and the corresponding CpG island was shown to be more methylated in uveal melanoma than in uveal melanocytes or normal tissues (Fig. 6). Similarly, the CpG island is methylated in HCT-116 colon cancer cells and unmethylated in normal colon. ¹¹ Moreover, using a ChIP assay and real-time PCR, we were able to compare and confirm that AcH3 and AcH4 were decreased in uveal melanoma cell lines, which led to a closed chromatin structure and gene silencing. These results prove that DNA methylation and histone modification contributed to the transcriptional downregulation of miRNAs in uveal melanomas. 

In conclusion, miR-124a is an important miRNA in the development of uveal melanoma, with extensive intracellular effects that can be reversible with either the introduction of exogenous miR-124a or controlling the epigenetic mechanisms that silence it. 

Supplementary Materials

pdf S1 

Footnotes

 Supported in part by the National Natural Science Foundation of China Grant 81071682, Zhejiang Provincial Natural Science Foundation of China Grant Y2110516, 973 Projects (2011CB504605 and 2012CB22303) from the Ministry of Science and Technology of China, Major Program of Science Foundation of the Affiliated Eye Hospital of Wenzhou Medical College YNZD201002, and Science Foundation of Wenzhou Medical College QTJ11020.

Footnotes

⁶  These authors contributed equally to the work presented here and should therefore be regarded as equivalent authors.

Footnotes

 Disclosure: X. Chen, None; D. He, None; X.D. Dong, None; F. Dong, None; J. Wang, None; L. Wang, None; J. Tang, None; D.-N. Hu, None; D. Yan, None; L. Tu, None