MiR-375 Regulation of LDHB Plays Distinct Roles in Polyomavirus-Positive and -Negative Merkel Cell Carcinoma

MicroRNA-375 (miR-375) is deregulated in multiple tumor types and regulates important targets involved in tumorigenesis and metastasis. This miRNA is highly expressed in Merkel cell carcinoma (MCC) compared to normal skin and other non-MCC skin cancers, and its expression is high in Merkel cell polyomavirus (MCPyV)-positive (MCPyV+) and low in MCPyV-negative (MCPyV−) MCC tumors. In this study, we characterized the function and target of miR-375 in MCPyV+ and MCPyV− MCC cell lines. Ectopic expression of miR-375 in MCPyV− MCC cells resulted in decreased cell proliferation and migration, as well as increased cell apoptosis and cell cycle arrest. However, in MCPyV+ MCC cells, inhibition of miR-375 expression reduced cell growth and induced apoptosis. Additionally, the expression of lactate dehydrogenase B (LDHB), a known target of miR-375, was inversely correlated with miR-375. Silencing of LDHB reduced cell growth in MCPyV− cell lines, while its silencing in MCPyV+ cell lines rescued the cell growth effect mediated by miR-375 inhibition. Together, our results suggest dual roles of miR-375 and LDHB in MCPyV and non-MCPyV-associated MCCs. We propose that LDHB could be a therapeutic target in MCC and different strategies should be applied in virus- and non-virus-associated MCCs.


miR-375 Regulates LDHB Expression in MCC Cells
We next assessed whether miR-375 could regulate LDHB expression in MCC cell lines. We ectopically expressed miR-375 using an expression plasmid (miR-375 OE) in the three MCPyV− cell lines and silenced miR-375 using miRNA sponge (miR-375sp) in two MCPyV+ cell lines. Transfection with miR-375 OE increased miR-375 levels in all three MCPyV− cell lines, while inhibition of miR-375 (miR-375sp) reduced its levels in both WaGa and MKL-1 cells (Figure 2A). Furthermore, miR-375 over-expression reduced and its inhibition increased LDHB mRNA and protein levels ( Figure 2B,C). Taken together, these observations indicate that LDHB is a target of miR-375 in MCC.
Given that miR-375 is one of the MCC-specific miRNAs and its differential expression between MCPyV+ and MCPyV− MCC tumors, we sought to determine whether miR-375 plays distinct roles in these two tumor entities.

Over-Expression of miR-375 Inhibits Cell Growth and Migration in MCPyV− MCC Cell Lines
To determine the effect of miR-375 on tumor phenotypes, we ectopically expressed miR-375 in the MCPyV− cell lines using miRNA mimic or expression plasmid and investigated its effect on cell growth, cell cycle, cell migration, and apoptosis. Using RT-qPCR, we validated increased miR-375 levels in cells transfected with miR-375 mimic or expression plasmid (miR-375 OE) ( Figure 3A). Using WST-1 assay, we observed a decrease of cell growth after 48 h (MCC 14/2) or 72 h (MCC13 and MCC26) of transfection of miR-375 mimic ( Figure 3B). Similar to the effect with miR-375 mimic, MCC14/2 cells stably transfected with miR-375 OE reduced cell growth after 48 h, as evaluated by WST-1 and trypan blue exclusion assays ( Figure 3C). The results support similar effect in miR-375 over-expressing cells using either miR-375 mimic or expression plasmid.

miR-375 Regulates LDHB Expression in MCC Cells
We next assessed whether miR-375 could regulate LDHB expression in MCC cell lines. We ectopically expressed miR-375 using an expression plasmid (miR-375 OE) in the three MCPyV− cell lines and silenced miR-375 using miRNA sponge (miR-375sp) in two MCPyV+ cell lines. Transfection with miR-375 OE increased miR-375 levels in all three MCPyV− cell lines, while inhibition of miR-375 (miR-375sp) reduced its levels in both WaGa and MKL-1 cells (Figure 2A). Furthermore, miR-375 over-expression reduced and its inhibition increased LDHB mRNA and protein levels ( Figure 2B,C). Taken together, these observations indicate that LDHB is a target of miR-375 in MCC.
Given that miR-375 is one of the MCC-specific miRNAs and its differential expression between MCPyV+ and MCPyV− MCC tumors, we sought to determine whether miR-375 plays distinct roles in these two tumor entities.

Over-Expression of miR-375 Inhibits Cell Growth and Migration in MCPyV− MCC Cell Lines
To determine the effect of miR-375 on tumor phenotypes, we ectopically expressed miR-375 in the MCPyV− cell lines using miRNA mimic or expression plasmid and investigated its effect on cell growth, cell cycle, cell migration, and apoptosis. Using RT-qPCR, we validated increased miR-375 levels in cells transfected with miR-375 mimic or expression plasmid (miR-375 OE) ( Figure 3A). Using WST-1 assay, we observed a decrease of cell growth after 48 h (MCC 14/2) or 72 h (MCC13 and MCC26) of transfection of miR-375 mimic ( Figure 3B). Similar to the effect with miR-375 mimic, MCC14/2 cells stably transfected with miR-375 OE reduced cell growth after 48 h, as evaluated by WST-1 and trypan blue exclusion assays ( Figure 3C). The results support similar effect in miR-375 over-expressing cells using either miR-375 mimic or expression plasmid.   Cell cycle analysis in MCC13 and MCC14/2 revealed that miR-375 mimic-treated cells had a subtle increase (7-14%) of cells in G1 or G2 phases compared with the negative control cells, respectively ( Figure 4A). Wound healing scratch assays revealed that ectopically expressed miR-375 retarded wound closure compared with the negative control at 18 h or 27 h ( Figure 4B). To determine the effect on apoptosis, we evaluated the cleavage products of Poly (ADP-ribose) polymerase (PARP) (an apoptotic marker) from cells over-expressing miR-375 or negative control using Western blot analysis. As shown in Figure 4C, we observed that the 89-kDa cleavage product of PARP was increased in all three MCC cell lines over-expressing miR-375 compared to miRNA mimic control, suggesting that miR-375 expression induces cell apoptosis in MCPyV− MCC cells. Cell cycle analysis in MCC13 and MCC14/2 revealed that miR-375 mimic-treated cells had a subtle increase (7-14%) of cells in G1 or G2 phases compared with the negative control cells, respectively ( Figure 4A). Wound healing scratch assays revealed that ectopically expressed miR-375 retarded wound closure compared with the negative control at 18 h or 27 h ( Figure 4B). To determine the effect on apoptosis, we evaluated the cleavage products of Poly (ADP-ribose) polymerase (PARP) (an apoptotic marker) from cells over-expressing miR-375 or negative control using Western blot analysis. As shown in Figure 4C, we observed that the 89-kDa cleavage product of PARP was increased in all three MCC cell lines over-expressing miR-375 compared to miRNA mimic control, suggesting that miR-375 expression induces cell apoptosis in MCPyV− MCC cells.       ), which recognizes the full-length (116 kDa) and apoptosis-associated cleaved (89 kDa) forms. GAPDH was used as a loading control. * p < 0.05 and ** p < 0.01 by paired Student's t-test. ns = not significant.

Inhibition of miR-375 Expression Reduces Cell Growth and Induces Apoptosis in MCPyV+ MCC Cells
We suppressed miR-375 expression in two MCPyV+ MCC cell lines using miR-375sp ( Figure 5A). Using WST-1 and trypan blue exclusion assays, we observed that suppression of miR-375 led to decreased cell growth in both WaGa and MKL-1 cell lines ( Figure 5B). To further examine whether Cancers 2018, 10, 443 7 of 16 the reduction of cell growth was due to apoptosis, we determined the apoptotic effect using Annexin V and caspase-3 activity assays. For Annexin V assay, we observed that suppression of miR-375 increased the number of apoptotic cells by 13% (p = 0.016) compared to the vector control-transfected cells ( Figure 5C). Concordantly, we also observed increased of caspase-3 activity upon suppression of miR-375 (2.7-fold, p = 0.001; Figure 5D). Together, our results suggest that miR-375 suppression inhibited cell growth via apoptosis in MCPyV+ MCC cells.

Inhibition of miR-375 Expression Reduces Cell Growth and Induces Apoptosis in MCPyV+ MCC Cells
We suppressed miR-375 expression in two MCPyV+ MCC cell lines using miR-375sp ( Figure  5A). Using WST-1 and trypan blue exclusion assays, we observed that suppression of miR-375 led to decreased cell growth in both WaGa and MKL-1 cell lines ( Figure 5B). To further examine whether the reduction of cell growth was due to apoptosis, we determined the apoptotic effect using Annexin V and caspase-3 activity assays. For Annexin V assay, we observed that suppression of miR-375 increased the number of apoptotic cells by 13% (p = 0.016) compared to the vector control-transfected cells ( Figure 5C). Concordantly, we also observed increased of caspase-3 activity upon suppression of miR-375 (2.7-fold, p = 0.001; Figure 5D). Together, our results suggest that miR-375 suppression inhibited cell growth via apoptosis in MCPyV+ MCC cells.

Silencing of LDHB Rescues Cell Growth Effect Mediated by miR-375 Suppression
To determine whether LDHB plays a role in miR-375 regulation of cell growth, we compared cell growth in miR-375sp-transfected cells with and without silencing of LDHB using two different siRNAs (siLDHB #1 and siLDHB #2). In parallel, we also transfected cells with miR-375sp or vector control only. As shown in Figure 6A, cells transfected with miR-375sp only or together with siCTR had higher LDHB levels than the pcDNA3 vector control. Co-transfection of miR-375sp and siLDHB led to a decrease in LDHB levels compared to cells transfected with miR-375sp and siCTR. Consistently, we observed decreased cell growth upon inhibition of miR-375, in which the effect was rescued by silencing of LDHB ( Figure 6B).

Silencing of LDHB Rescues Cell Growth Effect Mediated by miR-375 Suppression
To determine whether LDHB plays a role in miR-375 regulation of cell growth, we compared cell growth in miR-375sp-transfected cells with and without silencing of LDHB using two different siRNAs (siLDHB #1 and siLDHB #2). In parallel, we also transfected cells with miR-375sp or vector control only. As shown in Figure 6A, cells transfected with miR-375sp only or together with siCTR had higher LDHB levels than the pcDNA3 vector control. Co-transfection of miR-375sp and siLDHB led to a decrease in LDHB levels compared to cells transfected with miR-375sp and siCTR. Consistently, we observed decreased cell growth upon inhibition of miR-375, in which the effect was rescued by silencing of LDHB ( Figure 6B).

Silencing of LDHB Reduces Cell Growth in MCPyV− MCC Cells
In MCPyV− MCC cells, we observed higher LDHB levels ( Figure 1B) and that over-expression of miR-375 reduced cell growth ( Figure 3B,C). We therefore asked whether silencing of LDHB could phenocopy the effect of miR-375 down-regulation. Indeed, silencing of LDHB reduced cell growth and increased apoptosis (as indicated by increased cleaved PARP levels) (Figure 6C,D).

Discussion
MCC are generally divided into MCPyV+ and MCPyV− tumors, depending on their etiologies. While MCPyV+ tumors are the most common MCCs in US and Europe, the MCPyV− tumors are more common in Australia [17,46]. Numerous data indicate anatomical, genetical, and clinical differences between MCPyV+ and MCPyV− MCCs. Anatomically, MCPyV+ tumors are found more frequently on extremities, and MCPyV− tumors are more frequent in the head and neck [17,47]. Molecularly, MCPyV− MCCs harbor high mutation loads associated with ultra-violet (UV) signature, suggesting that UV exposure is the underlying etiology of MCPyV− MCCs [48]. On the contrary, MCPyV+ tumors have low mutation burdens, suggesting that the viral oncoproteins control key processes involved in MCC tumorigenesis [49]. Clinically, MCPyV− tumors are more aggressive, with increased risk of tumor progression and MCC-related death [17,47]. Additionally, MCPyV− and MCPyV+ MCCs may derive from different cell lineages [50]. All these observations support that MCPyV+ and MCPyV− MCCs are distinct tumor entities.
Given substantial differences between MCPyV+ and MCPyV− MCCs, we speculated that miR-375 is functionally distinct between these two tumor types. Indeed, our results support that miR-375 acts as a tumor suppressor in MCPyV− and function as an oncogene in MCPyV+ MCC cell lines. Consistent with our findings, low expression of miR-375 and its tumor suppressor role has been observed in MCPyV− MCC cell lines [41]. In MCPyV+ cell lines, we observed that suppression of miR-375 reduced cell growth and induced apoptosis, indicating that miR-375 is important to maintain cell viability in virus-positive cells. miR-375 is an MCC-specific miRNA and is highly expressed in MCPyV+ tumors and sera; it is thus not surprising that this miRNA plays pivotal roles in this tumor type.
It has also been shown that miR-375 can directly repress the key glycolytic enzyme LDHB [39]. Given that MCPyV small T-antigen can promote glycolysis [61], we speculated that miR-375 regulation of LDHB might be important in MCC tumorigenesis. Here, we demonstrated that LDHB mRNA and protein levels were reduced following over-expression of miR-375 and increased after suppression of miR-375, supporting that LDHB is a target of miR-375 in MCC. Functionally, we showed that silencing of LDHB could phenocopy the anti-survival effect of miR-375 over-expression in MCPyV− MCC cell lines, indicating its oncogenic role. The results are consistent with previous studies supporting that LDHB promotes tumor development and progression [43]. However, in MCPyV+ MCC cell lines, silencing of LDHB could rescue the cell growth inhibition effect mediated by miR-375 suppression, suggesting its role as a suppressor in MCPyV+ MCC. Similarly, reduced LDHB expression levels have also been observed in several cancer types, such as prostate cancer [44] and pancreatic cancer [62]. One common observation between these tumor types and MCPyV small T-antigen-transfected cells is their glycolytic phenotype. One possible explanation for the differential role of LDHB in MCPyV+ and MCPyV− MCC cell lines is that MCPyV+ cell lines rely on aerobic glycolysis, which requires continuous generation of NAD+ from LDHB suppression, while the oxidative cancer cells largely rely on LDHB activity to generate substrates for the Krebs cycle that fuels cellular activities. It is thus tempting to speculate that cellular metabolisms in MCPyV+ and MCPyV− MCC cells are different from one another; MCPyV+ cells are likely glycolytic and MCPyV− cells are oxidative. Given that the MCPyV small T-antigen can promote a pro-glycolytic phenotype, the question arises whether the viral oncoprotein could change the cellular metabolism of the cells that converts LDHB from its oncogenic role to tumor suppressor. Alternatively, the differential roles observed could be due to different cellular contexts rather than an effect of the virus itself. Further investigations are warranted to fully understand cellular metabolism differences between these two groups and whether MCPyV oncoproteins could change cellular metabolism of the cells or the function of LDHB.

MCC Tumor Samples
Twenty-six formalin-fixed paraffin-embedded (FFPE) and 28 frozen tumor samples were collected from the Karolinska University Hospital and Stockholm South General Hospital (Stockholm, Sweden). All samples had been included in our previous studies [18,63]. The study was approved by the Ethics Committee of Karolinska Institutet (2010/1092-31/3), and the use of archival materials was approved by the Karolinska University Hospital Biobank (BbK-00557). All materials were coded. The materials were obtained with written informed consent, except those samples collected prior to 2010, which at that time were covered by a general application of endocrine tumor collection approved by the ethic committee board of the Karolinska Institutet (Dnr. 91:86), and oral informed consent was applied.

RNA Extraction
Total RNA was extracted using mirVana miRNA isolation kit (Applied Biosystem/Ambion, Austin, TX, USA) and the concentrations were measured with the NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) and stored at −80 • C for further use.

TaqMan Reverse Transcription-Quantitative PCR (RT-qPCR)
Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) was used to quantify miR-375 and LDHB expressions using the StepOnePlus™ Real-Time PCR system (Life Technologies, Carlsbad, CA, USA). Predesigned TaqMan assays for mature miR-375 (ID_000564), RNU6B (ID_001093), LDHB (Hs00929956_m1) and GAPDH (Hs99999905_m1) were purchased from Applied Biosystems. For mature miR-375 and RNU6B, cDNA was synthesized from 120 ng total RNAs using TaqMan MicroRNA Reverse Transcription Kit (cat. no. 4366597; Applied Biosystems). For mRNAs, 100 ng total RNAs was used for cDNA synthesis using High Capacity cDNA Reverse Transcription kit (cat. no. 4368814; Applied Biosystems). All reactions were performed in triplicate. The relative expression levels of mature miR-375 were normalized to RNU6B, while the LDHB expressions were normalized to GAPDH. The quantification of miR-375 and RNU6B in 26 samples was previously analyzed [18], while the remaining samples were analyzed in this study.

Trypan Blue Exclusion Assay
Cells were stained with 0.4% trypan blue stain (Invitrogen) and analyzed using the TC10 TM automated cell counter (Bio-Rad, Hercules, CA, USA). Total live cells in the miR-375 OE, miR-375sp, and siLDHB-transfected cells were compared to their respective controls.

Cell Cycle Analysis
At 72 h after transfection, 1 × 10 6 cells were washed with PBS and fixed in cold 50% ethanol for 1 h. After washing with PBS and treating with RNase A (0.2 mg/mL; R6513, Sigma-Aldrich, St. Louis, MO, USA) for 1 h at 37 • C, the cells were then stained with 10 µL propidium iodide (1 mg/mL; P4170, Sigma-Aldrich) and kept on ice in the dark. Cell cycle analysis was performed using flow cytometry (Cytomics FC 500; Beckman Coulter, Brea, CA, USA) and FlowJo software version 7.6.2 (Tree Star Inc., Ashland, OR, USA). All experiments were performed independently in triplicate.

Wound Healing Scratch Assay
After 48 h of transfection, a scratch wound was made on the confluent monolayer cells of each treatment group and cultured in low serum (2% FBS) medium. The scratch was imaged in real-time using IncuCyte S3 (Essen BioScience, Ann Arbor, MI, USA). Image J software version 1.43u (http://rsbweb.nih.gov/ij/) was used to process all images for quantification purposes. The wound closure (cell migration) was calculated by fraction of wound at the given time to the wound area at 0 h and normalized to viable cell number of transfected cells plated in parallel. Three independent replicates were included in each experimental group.

Apoptosis Assays
Cell apoptosis was evaluated in WaGa cells after 72 h of transfection with miR-375sp or pcDNA3 using Annexin V FITC Apoptosis kit (cat. no. 640905; BioLegend, San Diego, CA, USA) and Caspase-3 colorimetric assay (#K106; BioVision, Mountain View, CA, USA). All experimental conditions were performed according to the manufacturer's instructions. The Annexin V and propidium iodide-stained cells were analyzed by NovoCyte flow cytometer (ACEA Biosciences, San Diego, CA, USA), and the caspase-3 cleavage products were measured at wavelength 405 nm using a VERSAmax microplate reader (Molecular Devices). All experiments were replicated three times independently.

Statistical Analysis
All analyses were performed using IBM SPSS Statistics version 24.0 (IBM Corp., Armonk, NY, USA) or MS Office Excel 2007. Paired Student's t-test was performed to analyze transfection experiments. Spearman's rank order correlation was used to evaluate correlation between miR-375 and LDHB expressions. All analyses were 2-tailed, and p-values < 0.05 were regarded as significant.

Conclusions
We demonstrate distinct functional roles of miR-375 and LDHB in MCPyV+ and MCPyV− MCCs. Targeting LDHB could be a novel therapy for MCC.