The Interplay of MicroRNA-34a, LGR4, EMT-Associated Factors, and MMP2 in Regulating Uveal Melanoma Cells

Abstract

Purpose: MicroRNA-34a (miR-34a) has been implicated in many biological processes. It is downregulated in uveal melanoma, and introduction of miR-34a inhibits the proliferation and migration of uveal melanoma cells. Leucine-rich repeat-containing G protein-coupled receptor 4 (LGR4) is a novel target of miR-34a identified first in retinal pigment epithelial cells. In this study, we sought to evaluate the interaction of miR-34a and LGR4 in uveal melanoma and its downstream mechanisms.

Methods: The expression of LGR4, epithelial–mesenchymal transition (EMT)-associated factors, and matrix metalloproteinase 2 (MMP2) in uveal melanoma cells was assessed by immunoblotting and immunofluorescence analysis. MicroRNA-34a mimic molecules, LGR4 small interfering RNA (siRNA), or MMP2-specific siRNA were transiently transfected into uveal melanoma cells. In vitro scratch and Transwell assays were used to evaluate the migratory and invasive potential of the resultant uveal melanoma cells.

Results: LGR4 is upregulated in uveal melanoma cells. Introduction of miR-34a significantly decreased the expression level of LGR4. Transfection with miR-34a or knockdown of LGR4 attenuated the aggressiveness of uveal melanoma cells. In addition, there was a decrease in the expression of mesenchymal markers N-cadherin, vimentin, and Snail following miR-34a introduction or knockdown of LGR4. Finally, MMP2 was found to be a downstream effector for miR-34a and LGR4 that regulates the migration and invasion of uveal melanoma cells.

Conclusions: MicroRNA-34a negatively controls LGR4, thereby inhibiting the migration and invasion of uveal melanoma cells. Ultimately, both miR-34a and LGR4 impact the aggressiveness of uveal melanoma with alterations in the markers of the EMT. MMP2 is a downstream effector that influences the metastasis seen with uveal melanoma cells.

Uveal melanoma is the most common primary intraocular malignancy in adults, representing approximately 85% of ocular melanomas.¹ Uveal melanoma is a rare cancer representing only approximately 3% to 5% of the total melanoma cases registered in the United States as reported by the SEER Database. 

Risk factors for development of uveal melanoma include Caucasian ethnicity, light eye color, dysplastic nevus syndrome, ocular melanocyte ptosis, and a germline mutation from BRCA1-associated protein 1 (BAP1).^2–5 Since uveal melanoma has a different origin in comparison to cutaneous melanoma, the molecular profile for uveal melanoma also differs significantly. Unlike cutaneous melanoma, uveal melanoma has a GNAQ/GNA11 mutation leading to proliferation of cells.^6,7 In contrast, cutaneous melanomas frequently harbor BRAF mutations, PTEN mutations, TP53 mutations, and CDKN2A mutations. Mutations associated with G protein alpha subunit GNAQ or GNA11 are observed in over 80% of primary uveal melanomas and are a possible target for treatment.³ 

With the introduction of significant new therapies for cutaneous melanoma including PD-1 inhibition and targeting of BRAF mutation/MEK pathways,^8–12 there has been significant progress in melanoma research. However, uveal melanoma remains a disease with limited treatment options. 

Recent developments in uveal melanoma research have revealed other pathways that target tumor progression, including the microRNA network dysregulation that has been shown to promote cell cycle progression, resistance to apoptosis, and systemic metastasis. One of the microRNAs of particular interest is microRNA-34a (miR-34a), which is a gene product from chromosomal locus 1p36.22 that is also a tumor suppressor in many types of tumor.^13–15 This has been implicated previously in normal biological processes such as cell proliferation, inhibition, cell cycle arrest, and senescence. 

MicroRNAs execute their functions through regulating the expression of their downstream targets. We previously identified a novel miR-34a target, LGR4 (leucine-rich repeat-containing G protein-coupled receptor 4) in retinal pigment epithelium cells.¹³ LGR4, also known as GPR48 (G protein-coupled receptor 48), has been implicated in embryonic development, cell motility, and tumor metastasis.^14,16–19 LGR4 was found to be necessary for prostate cancer metastasis epithelial–mesenchymal transition (EMT).²⁰ EMT is a term originally coined to describe cellular behavior observed by Elizabeth Hay during gastrulation but has recently gained significant attention in tumor cell metastasis.^21,22 A common feature of the transition to a mesenchymal state is the loss of apical–basal polarity and stable junctions followed by migration of cells. Therefore, cells collectively lose cellular binding through suppression of epithelial markers such as E-cadherin and gain expression of mesenchymal markers such as N-cadherin, vimentin, and Snail.²² The complex interplay between miR-34a, LGR4, and the EMT pathway is further complicated by interactions with matrix metalloproteinases. In the midst of these complex interactions, matrix metalloproteinase 2 (MMP2) and other MMPs have been shown to promote EMT, which is involved in cancer metastasis.²³ This study aimed to evaluate and demonstrate the interplay between miR-34a, LGR4, MMP2, and the EMT pathways leading to uveal melanoma progression. 

Materials and Methods

Cell Culture

The human uveal melanoma cell lines M17, M23, and SP6.5 were kind gifts from Dan-Ning Hu (Icahn School of Medicine at Mount Sinai, New York, NY, USA) and Guy Pelletier (Research Center of Immunology, Montreal, Quebec, Canada). The cells were isolated from Caucasian patients with primary choroidal melanoma and cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA) at 37°C in a humidified incubator containing 5% CO₂.²⁴ The human primary uveal melanocyte cells U-96 were isolated from a Chinese donor at Wenzhou Medical University and cultured as previously described.²⁵ All studies and procedures involving human tissue were approved by the Wenzhou Medical University ethics committee, and performed in compliance with the Declaration of Helsinki and national laws. 

Immunofluorescence Staining

For LGR4 expression analysis, U-96, M17, M23, and SP6.5 cells were directly seeded on coated glass cover slides in six-well plates and grown for 24 hours. For EMT molecule detection, M17 and SP6.5 cells were seeded and then transfected with 100 nM LGR4-specific small interfering RNA (siRNA) (5′-GGUAAGAAACUCCUAAUUAUU; GenePharma, Shanghai, China), miR-34a mimics, or negative control for 48 hours with jetPRIME kit (Polyplus Transfection, Illkirch, France) according to the manufacturer's instructions. Cells were then fixed with freshly prepared 4% paraformaldehyde solution for 1 hour at room temperature, followed by permeabilization with 0.4% Triton X-100 in Tris-buffered saline (TBS) for 10 minutes. After blocking with goat serum (Boster, Wuhan, China) for 1 hour at room temperature, cells were then incubated with the primary antibodies at 4°C overnight. Cy3-conjugated secondary antibodies (Beyotime, Shanghai, China) were added to the cells for 1 hour, and the cell nuclei were displayed by staining with 4′,6-diamidino-2-phenyl-indole (DAPI). Cells were mounted with fluorescent mounting medium (Beyotime) and the images were captured using a fluorescence microscope (Imager Z1; Zeiss, Oberkochen, Germany). The primary antibodies for LGR4 were purchased from Abcam (Cambridge, MA, USA); antibodies for E-cadherin and vimentin were from Santa Cruz (Dallas, TX, USA), and antibodies for N-cadherin and Snail were from Cell Signaling Technology (Danvers, MA, USA). All primary and secondary antibodies were used at 1:100 dilution. The mean fluorescence intensity of EMT marker staining was measured with ImageJ 1.52i software (National Institutes of Health, Bethesda, MD, USA), and then normalized to the value of the mock group. 

In Vitro Scratch Assay

M17 and SP6.5 cells were grown to approximately 60% confluence and then transfected with 100 nM LGR4-specific siRNA, MMP2-specific siRNA (5′-GGAGAGCUGCAACCUGUUU, GenePharma), miR-34a mimics, or negative control. Forty-eight hours after transfection, the cell monolayers were scratched with a 200-μL pipette tip and floating cells were washed away immediately. Cells were cultured in freshly prepared serum-free medium supplemented with 10 ng/mL human hepatocyte growth factor (HGF; R&D Systems, Minneapolis, MN, USA). Images were taken after scratching at 24, 48, and 72 hours. The migratory potential was assessed by comparing the migration rate toward the gap. 

Transwell Migration and Matrigel Invasion Analysis

M17 and SP6.5 cells were transfected with 100 nM LGR4- and MMP2-specific siRNA, miR-34a mimics, and miR-34a inhibitor or corresponding negative control for 48 hours as indicated and then harvested by trypsinization. To measure cell migration, 8-μm pore size culture inserts (Transwell; Costar, High Wycombe, UK) were placed into the wells of 24-well culture plates to separate the upper and the lower chambers. DMEM (400 μL) supplemented with 10 ng/mL human HGF was added to the lower chambers. 1 × 10⁵ cells were seeded to the upper chamber in most circumstances except that 5 × 10⁴ cells were used for miR-34a inhibitor function assay. The cells that migrated through the pores were fixed with 4% paraformaldehyde and stained with crystal violet solution after 24 hours. Cell numbers in five random vision fields were counted under the microscope (Zeiss) using a ×20 objective. For the invasion analysis, BioCoat Matrigel Invasion Chambers (BD Biosciences, San Jose, CA, USA) were used, and cells were treated the same as in the Transwell migration assay. 

Immunoblotting

Cells were grown in six-well plates and transfected as described above. Seventy-two hours after transfection, cells were lysed in lysis buffer (Beyotime) supplemented with protease inhibitor cocktail (Sigma-Aldrich Corp., St. Louis, MO, USA). Equal amounts of cell lysates were loaded to SDS polyacrylamide gels for electrophoresis. Proteins were transferred onto nitrocellulose membranes (GE Healthcare, Pittsburgh, PA, USA) and then blocked with 5% nonfat milk in PBS with 0.05% Tween 20. Membranes were incubated with primary antibodies overnight at 4°C and then treated with peroxidase-conjugated secondary antibodies (Santa Cruz). Enhanced chemiluminescence detection kit (Pierce, Rockford, IL, USA) was used to develop the membrane. β-actin was probed as a loading control and its primary antibodies were purchased from Santa Cruz. All the primary antibodies were used at 1:1000 dilution except for β-actin at 1:2000, and the secondary antibodies were diluted at 1:2000. The band density was analyzed with ImageJ 1.52i software and then normalized to the level of the mock group. 

Statistical Analysis

All data were expressed as mean ± SD. Differences between mock and experimental groups were evaluated with Student's t-test, and P < 0.05 was considered significant. 

Results

LGR4 Is Upregulated in Uveal Melanoma Cell Lines

Since LGR4 has been shown to be involved in the tumorigenesis and regulation of many cancers,^26–28 we first detected the expression level of LGR4 in uveal melanoma cell lines. The Western blot results showed that LGR4 is highly expressed in uveal melanoma cell lines M23, M17, and SP6.5 compared with normal uveal melanocyte U-96 cells (Fig. 1A). To further confirm this, immunofluorescence assays were performed and the results showed that the fluorescence intensity in M17 and SP6.5 cells was much higher than in U-96 cells (Fig. 1B). These results suggest that LGR4 is upregulated in uveal melanoma cell lines. 

Figure 1

 

LGR4 is highly expressed in uveal melanoma cell lines. (A) The expression level of LGR4 in uveal melanocyte cell line U-96 and uveal melanoma cell lines M23, M17, and SP6.5 was detected with immunoblotting. β-actin was detected as a loading control. (B) U-96, M17, and SP6.5 cells were fixed and subjected to immunofluorescence analysis for LGR4 expression. The nuclei were probed with DAPI. Images are representative of at least three independent experiments. Magnification: ×200.

LGR4 Is a Direct Target of miR-34a in Uveal Melanoma Cells

We previously reported that miR-34a is downregulated in uveal melanoma cells and inhibits the proliferation and migration of uveal melanoma cells.²⁹ We also showed that LGR4 is a direct target of miR-34a in RPE cells.¹³ Here we showed that transfection of miR-34a to M17 and SP6.5 cells decreased LGR4 expression level, similar to transfection of LGR4-specific siRNA (Fig. 2). The expression level of β-actin remained unchanged. This indicates that miR-34a directly regulates the expression of LGR4 in uveal melanoma cells. 

Figure 2

 

MicroRNA-34a targets LGR4 in uveal melanoma cells. M17 and SP6.5 cells were transfected with LGR4-specific siRNA, miR-34a mimics, or a negative control (NC). The expression level of LGR4 was detected by Western blot. Images are representative of at least three independent experiments.

Transfection with miR-34a or Knockdown of LGR4 Attenuates the Migration and Invasion of Uveal Melanoma Cells

To elucidate the role of LGR4 in uveal melanoma cells, Transwell and in vitro scratch assays were performed to evaluate the migration potential. In Transwell assay, the number of migrated cells toward the lower chamber was significantly reduced in miR-34a- or LGR4 siRNA-transfected M17 and SP6.5 cells (in M17 cells: mock: 248 ± 18, negative control [NC]: 255 ± 11, siLGR4: 119 ± 8, miR-34a: 79 ± 9; in SP6.5 cells: mock: 440 ± 21, NC: 401 ± 18, siLGR4: 216 ± 13, miR-34a: 224 ± 19 cells/vision field, n = 3, Fig. 3A). Using in vitro scratch assay, miR-34a- or LGR4 siRNA-transfected M17 and SP6.5 cells migrated much more slowly toward the gap than mock and NC-transfected cells, especially after 48 and 72 hours (Fig. 3B). 

Figure 3

 

MicroRNA-34a and LGR4 interference inhibit the migratory and invasive potential of uveal melanoma cells. M17 and SP6.5 cells were transfected as indicated. Transwell analysis (A) and in vitro scratch assay (B) were performed. Cells that migrated to the lower chamber in five independent vision fields were counted and quantified (A). (C) The invasive ability was evaluated with Matrigel Transwell assay. Migrated cells were quantified by counting five independent vision fields under a microscope. Images are representative of at least three independent experiments. All the data are expressed as mean ± SD. *P < 0.05, magnification: ×100.

We further addressed if miR-34a and LGR4 can regulate the invasion ability of uveal melanoma cells with Matrigel Transwell analysis. Significantly fewer cells migrated to the lower chamber in miR-34a- or LGR4 siRNA-transfected M17 and SP6.5 cells (in M17 cells: mock: 387 ± 15, NC: 369 ± 14, siLGR4: 118 ± 8, miR-34a: 107 ± 9; in SP6.5 cells: mock: 142 ± 11, NC: 132 ± 8, siLGR4: 48 ± 5, miR-34a: 46 ± 7 cells/vision field, n = 3, Fig. 3C). 

MicroRNA-34a and LGR4 Alter the EMT Process of Uveal Melanoma Cells

To illustrate the mechanism underlying miR-34a and LGR4 mediation in uveal melanoma cells, we detected the expression of EMT markers with Western blot (Fig. 4A) and immunofluorescence analysis (Figs. 4B, 4C). Increase of epithelial marker E-cadherin was observed in both miR-34a- and LGR4 siRNA-transfected M17 and SP6.5 cells. The expression of mesenchymal markers N-cadherin, vimentin, and Snail was significantly decreased accordingly. The expression level of β-actin remained unchanged. These data indicate that miR-34a and LGR4 regulate the EMT process in uveal melanoma cells. 

Figure 4

 

MicroRNA-34a and LGR4 knockdown inhibit the EMT process in uveal melanoma cells. M17 and SP6.5 cells were transfected as indicated, and (A) the expression level of epithelial marker E-cadherin and mesenchymal marker N-cadherin, vimentin, Snail was probed with immunoblot. β-actin was detected as an internal control. The band density was quantified with ImageJ software and normalized to the level of the mock group. (B) The epithelial and mesenchymal markers were probed in fixed uveal melanoma cells. The nuclei were stained with DAPI. Images are representative of at least three independent experiments. Magnification: ×200. (C) The mean fluorescence density in each image in (B) was analyzed with ImageJ software and then normalized to the level of the mock group. *P < 0.05.

MMP2 Is a Downstream Effector of miR-34a and LGR4 and Regulates the Migration and Invasion of Uveal Melanoma Cells

The importance of MMP2 in cancer progression has been well studied, especially related to its ability to degrade type IV collagen, which is the most abundant component of the basement membrane.²³ After having confirmed that either miR-34a or LGR4 knockdown can inhibit the migration and invasion of uveal melanoma cells, we sought to detect the expression level of MMP2 in miR-34a- and LGR4 siRNA-transfected cells. Introduction of miR-34a or LGR4-specific siRNA significantly decreased the expression level of MMP2 (Fig. 5A). Similarly, the expression of MMP2 was reduced by MMP2-specific siRNA in both M17 and SP6.5 cells (data not shown). 

Figure 5

 

MMP2 is regulated by miR-34a and LGR4. Downregulation of MMP2 inhibits the migratory and invasive potential of uveal melanoma cells. M17 and SP6.5 cells were transfected with miR-34a mimics or LGR4 siRNA, and (A) the expression level of MMP2 was detected by Western blot. β-actin was probed as a loading control. M17 and SP6.5 cells were transfected with MMP2-specific siRNA, and the migratory potential was analyzed by in vitro scratch assay (B) and Transwell analysis (C). The invasive potential was evaluated with Matrigel Transwell analysis (D). Cell numbers in the lower chamber were counted in at least five vision fields. Images are representative of at least three independent experiments. *P < 0.05, magnification: ×200.

We further explored the physiological role of MMP2 in uveal melanoma cells. In vitro scratch assay (Fig. 5B) showed that knockdown of MMP2 decreased the migratory potential of both M17 and SP6.5 cells, starting after 24 hours and becoming increasingly prominent after 48 and 72 hours. In Transwell assay, significantly fewer cells migrated to the lower chamber in MMP2 knockdown cells (in M17 cells: mock: 260 ± 22, NC: 298 ± 30, siMMP2: 71 ± 9; in SP6.5 cells: mock: 134 ± 16, NC: 149 ± 14, siMMP2: 52 ± 17 cells/vision field, n = 3, Fig. 5C). Similar results were obtained in Matrigel Transwell assay (in M17 cells: mock: 269 ± 47, NC: 236 ± 46, siMMP2: 101 ± 28; in SP6.5 cells: mock: 180 ± 23, NC: 160 ± 30, siMMP2: 31 ± 4 cells/vision field, n = 3, Fig. 5D). These results demonstrated that MMP2 knockdown attenuates both the migratory and invasive potential of uveal melanoma cells. 

Discussion

Uveal melanoma harbors a different set of gene mutations in comparison to cutaneous melanomas. Because of the propensity for development of metastatic disease, there is an urgent need to discover alternative pathways that can be targeted in management of advanced uveal melanoma. Unfortunately, our understanding of the treatment of uveal melanoma is still in its infancy, in comparison to advanced cutaneous melanoma. Uveal melanomas frequently harbor mutations in the GNAQ/GNA11 gene, which would be uncommon in cutaneous melanomas. Other differences include mutation in SF3B1, which does not always lead to metastasis.^30,31 Gene expression profiling has shown that uveal melanomas tend to cluster into two molecular classes with distinct phenotypes. The first class of uveal melanomas, which tend to be low-grade tumors, contains downregulation of genes on chromosome 3 and upregulation of genes on chromosome 8q.³² The second class of tumors, which have a higher rate of proliferation, tends to demonstrate an increased expression of EMT and transformation of polarized epithelial cells into motile invasive cancer.^33,34 One of the new developments in the management of uveal melanoma is the use of microRNA molecules that play a role in gene silencing or posttranscriptional regulation of genes involved in development of uveal melanomas.^35,36 

MicroRNA-34a was found to be a proapoptotic transcriptional target of the p53 tumor suppressor gene effector network.³⁷ MicroRNA-34a has been shown to downregulate c-Met levels, another pro-oncogenic pathway to tumor development.²⁹ MicroRNA-34a is also downregulated in uveal melanoma and can inhibit the proliferation and migration of uveal melanoma cells.²⁹ MicroRNAs function through modulating target gene expression by binding to their 3′-untranslated region (UTR) and one microRNA may target dozens of downstream molecules. To our knowledge, this is the first study to show that miR-34a can directly modulate the expression of LGR4 in uveal melanoma cells. 

LGR4 is widely expressed in multiple organs such as intestines, heart, kidneys, cartilage, reproductive tracts, and the nervous system and plays an important role in the development of these organs in mice.^14,16–19 Moreover, a nonsense mutation of LGR4 is strongly related to low bone marrow density, electrolyte imbalance, and increased squamous cell carcinoma of the skin in human.³⁸ LGR4 has also been shown to be elevated in gastric, lung, prostate, breast, and colorectal cancers.^{20,26,39–42} Moreover, LGR4 appears to be a universal tumor promoter, and this property of LGR4 makes it an excellent target for study of oncogenesis. 

In this study, we showed that LGR4 is highly expressed and is directly targeted by miR-34a in uveal melanoma cells compared with primary uveal melanocytes. Furthermore, knockdown of LGR4 or introduction of miR-34a inhibits the migration and invasion of uveal melanoma cells. Since miR-34a has been reported to be downregulated in cancers when LGR4 is upregulated,⁴³ this reverse correlation between miR-34a and LGR4 may indicate a general mechanism that occurs in tumor occurrence and development. From the experiments, we were able to show that miR-34a upregulation with LGR4 downregulation inhibits uveal melanoma cells. We also attempted to inhibit miR-34a in uveal melanoma cell lines even further to see if there would be increase in tumor cell migration and invasion (Supplementary Figs. S1, S2). However, miR-34a inhibition did not result in changes in the aggressiveness of the uveal melanoma cell lines employed in this experiment. We surmised that the minimal expression of miR-34a seen in uveal melanoma cells failed to make the tumor cells even more aggressive, a fact that may be meaningful in design of uveal melanoma targets. 

One of the interesting findings prior to metastasis for a tumor cell is the conversion from a differentiated epithelial cell to a migrating mesenchymal cell. This process is frequently known as the EMT where there is a spectrum or continuum of phenotypic cells during the process of tumor cell metastasis and migration. Because of the potential role in cancer progression, recent emphasis has been placed on studying EMT with the use of molecular markers expressed in epithelial or mesenchymal cells.²² Cells that either acquire or lose E-cadherin, vimentin, and fibronectin are thought to be in the process of the EMT. During our evaluation, we have identified the downstream markers N-cadherin, vimentin, and Snail being downregulated as a result of miR-34a upregulation and LGR4 downregulation. Similarly, loss of cellular polarity needed prior to transition from a mesenchymal structural phenotype will render a repression of E-cadherin. During evaluation of the uveal melanoma cells, we have further confirmed that E-cadherin is upregulated with the use of miR-34a or LGR4 siRNA. 

The process by which these tumor cells gain the ability to invade and metastasize may be related to the upregulation of matrix metalloproteinases.²² One of the targets of miR-34a is Yin Yang-1 (YY-1), which has been shown to suppress the matrix metalloproteases MMP2 and MMP9.⁴⁴ The effects of microRNAs on matrix metalloproteinases as well as fibronectin type III domain containing 3B expression levels have been studied in the past.⁴⁵ In this study, we found that introduction of miR-34a or knockdown of LGR4 can inhibit the expression level of MMP2, which has been shown to be responsible for the migration and invasion of many kinds of tumors. This is the first study to establish a possible causal relationship between LGR4 and expression of MMP2. We also showed that downregulation of MMP2 attenuates the migratory and invasive potential of uveal melanoma cells. Thus, MMP2 downregulation induced by the introduction of miR-34a or LGR4 knockdown may inhibit the tumor cells with regard to gaining the ability for systemic invasion. 

In conclusion, uveal melanoma development and metastasis appear to be a complex interplay of various genes and protein expressions. MicroRNA-34a is near the epicenter of this regulation affecting both LGR4 and various downstream factors. MicroRNA-34a introduction or LGR4 knockdown is able to influence EMT as well as downregulate MMP2. This will hopefully offer another therapeutic avenue in the management of this orphan disease with limited treatment options. 

Acknowledgments

Supported in part by the Zhejiang Provincial Natural Science Foundation of China under Grant No. LY16H120009, the National Natural Science Foundation of China under Grant No. 81100671, and the Project of State Key Laboratory of Ophthalmology, Optometry and Visual Science, Wenzhou Medical University, under Grant No. K171205. 

Disclosure: Q. Hou, None; S. Han, None; L. Yang, None; S. Chen, None; J. Chen, None; N. Ma, None; C. Wang, None; J. Tang, None; X. Chen, None; F. Chen, None; X.D. Dong, None; L. Tu, None