MicroRNAs have been implicated in various skin cancers, including melanoma, squamous cell carcinoma, and basal cell carcinoma; however, the expression of microRNAs and their role in Merkel cell carcinoma (MCC) have yet to be explored in depth. To identify microRNAs specific to MCC (MCC-miRs), next-generation sequencing (NGS) of small RNA libraries was performed on different tissue samples including MCCs, other cutaneous tumors, and normal skin. Comparison of the profiles identified several microRNAs upregulated and downregulated in MCC. For validation, their expression was measured via qRT-PCR in a larger group of MCC and in a comparison group of non-MCC cutaneous tumors and normal skin. Eight microRNAs were upregulated in MCC: miR-502-3p, miR-9, miR-7, miR-340, miR-182, miR-190b, miR-873, and miR-183. Three microRNAs were downregulated: miR-3170, miR-125b, and miR-374c. Many of these MCC-miRs, the miR-183/182/96a cistron in particular, have connections to tumorigenic pathways implicated in MCC pathogenesis.
Merkel cell carcinoma (MCC) is a primary neuroendocrine carcinoma of the skin of uncertain origin. Although not as prevalent as other skin cancers, MCC is aggressive and has a high mortality rate, with an overall five-year survival of sixty percent [
Unfortunately, while the incidence of MCC increases, our knowledge of these tumors remains limited. Several factors are implicated in its pathogenesis, including UV radiation exposure, an associated polyomavirus (MCPyV), and immunosuppression [
With average lengths of roughly twenty-two nucleotides, these molecules serve as guides of the RNA-induced silencing complex (RISC). MicroRNAs regulate the expression of genes by binding to partially complementary target sites in mRNA transcripts and inhibiting their translation [
The study of microRNAs holds much promise for improving the diagnosis and treatment of cancer. Recent progress in our understanding of the role of microRNAs in disease has excited oncology. An effective antiviral therapy was recently developed based on microRNA biology, and microRNAs are promising new tools and targets in cancer research [
Indeed, microRNAs have been demonstrated to play significant roles in the pathogenesis of other skin cancers, such as squamous cell carcinoma (SCC) and melanoma [
Frozen tissue samples of various skin cancers (MCC, melanoma, SCC, and BCC), normal skin, and normal lymph node were obtained from the Cooperative Human Tissue Network (CHTN) and stored at −80°C. Formalin-fixed paraffin-embedded (FFPE) tissue samples of various skin cancers (MCC, melanoma, SCC, and BCC) and normal skin were obtained from Vanderbilt Pathology and Dermatopathology (Nashville, TN). Clinical information corresponding to the FFPE MCC samples is provided (Table
MCC sample data. Clinical information corresponding to each of the FFPE MCC samples utilized in qRT-PCR analysis.
# | Age (yrs) | Sex | Race | P/M | Metastasis/invasion/recurrence | Clinical information |
---|---|---|---|---|---|---|
1 | 57 | F | W | P | 6/9 axillary LNs+, distant LN+ | N/A |
2 | 57 | F | W | P | Soft tissue involvement, 2/6 LNs+, distant LN+ | N/A |
3 | 63 | M | W | P | Lumbar spinal cord involvement highly suspected based on FDG-PET/CT imaging | Immunosuppression regimen for renal transplant: mycophenolate mofetil, prednisone, tacrolimus |
4 | 64 | M | W | P | N/A | Immunosuppression regimen for renal transplant 2° granulomatosis with polyangiitis: mycophenolate mofetil, prednisone, tacrolimus; history of multiple SCC |
5 | 65 | M | W | M | Submandibular gland with soft tissue involvement, 6/24 LNs+, local recurrence | N/A |
6 | 70 | M | W | M | Distant (supraclavicular) LN+ | History of renal cancer |
7 | 71 | M | W | M | Parotid gland with intraparotid LN involvement, local recurrence | N/A |
8 | 72 | M | W | P | Local recurrence | History of colon cancer |
9 | 72 | M | W | M | Thyroid, parotid gland involvement, 3/23 LNs+ | Concurrent papillary thyroid carcinoma |
10 | 76 | M | W | P | Distant (cervical) LN+ | Concurrent SCC, history of lung cancer |
11 | 76 | M | W | P | 6/37 axillary LNs+ | N/A |
12 | 78 | M | W | P | Parotid gland involvement, |
N/A |
13 | 79 | M | W | P | N/A | History of bladder cancer |
14 | 79 | M | W | M | Salivary gland, deep soft tissue surrounding large arteries and skeletal muscle involvement, |
N/A |
15 | 80 | F | W | M | Multifocal extranodal tumor invasion, soft tissue, and sternocleidomastoid muscle involvement, 11/15 LNs+ | History of chronic lymphocytic leukemia |
16 | 82 | M | W | P | Parotid gland involvement, |
History of colon cancer, laryngeal cancer, multiple SCC |
17 | 82 | F | W | P | Parotid gland involvement, |
History of breast cancer |
18 | 82 | M | W | M | Parotid gland with invasion of right upper neck soft tissue, 2/11 LNs+, distant LN+ | History of colon cancer, laryngeal cancer, multiple SCC |
19 | 85 | M | W | P | N/A | History of acute myeloid leukemia, in remission |
20 | 85 | M | W | M | Rectum involvement | History of rectal cancer, prostate cancer |
LN: lymph node; M: metastasis; P: primary; W: white.
All cell lines were cultured in DMEM with 10% FCS and Pen-Strep.
Total RNA, including microRNAs, was isolated from frozen tissue and cell culture with the miRNeasy Mini Kit and from FFPE tissue with the miRNeasy FFPE Kit (Qiagen; Hilden, Germany), according to the manufacturer’s protocols. The concentration and integrity of the extracted total RNA were estimated by Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA, USA) and Agilent 2100 Bioanalyzer (Applied Biosystems, Carlsbad, CA, USA), respectively. RNA samples with a RNA Integrity Number (RIN) value of at least 7.0 or higher was used for further processing.
Approximately 1
At least 15 million, 50 bp, SE reads were generated from each sample. Further downstream analysis of the sequenced reads from each sample was performed as per our unique in-house pipeline. Briefly, quality control checks on raw sequence data from each sample will be performed using FastQC (Babraham Bioinformatics, London, UK). Raw reads were then imported on a commercial data analysis platform CLCbio (CLCbio, MA, USA). Adapter trimming was done to remove ligated adapter from 3′ end of the sequenced reads with only one mismatch allowed; poorly aligned 3′ ends were also trimmed. Sequences shorter than 15 nucleotides length were excluded from further analysis. Trimmed Reads were then used to extract and count the small RNA which were then annotated with microRNAs in miRBase release 18 database. Samples were grouped as per their types identifiers and quantification of miRNA abundance was done. Differential expression of miRNA was calculated on the basis of their fold change (default cut-off ≥±2.0) between mapped counts observed between individual groups.
To validate the NGS data, the expression of microRNAs was analyzed via quantitative real-time reverse transcription polymerase chain reaction (qRT-PCR) with the Rotor-Gene SYBR Green PCR Kit, miScript Primer Assays, and miScript Universal Primer, following reverse transcription of total RNA with the miScript II RT Kit (Qiagen; Hilden, Germany).
The qRT-PCR analysis was performed in technical replicates according to the manufacturer’s instructions using the Rotor-Gene SYBR Green PCR Master Mix (Qiagen; Hilden, Germany). The packaged operating software was utilized for instrument control, data acquisition, and raw data analysis. The plates were run in relative quantification (
Amplification curves were analyzed using the packaged operating software, and assays were inspected for distinct melting curves. In addition, only assays detected with
Sequencing of small RNA libraries was performed for the following frozen tissue samples: three MCCs, one melanoma, one SCC, one BCC, and one normal skin. The MCC and melanoma were lymph node metastases, while the SCC and BCC were primary cutaneous lesions. Comparison of the sequencing profiles identified several microRNAs upregulated and downregulated in MCC versus other tissues (Table
MCC-miR candidates identified via NGS. Lists of top fifteen microRNAs upregulated and downregulated in MCC (
Upregulated in MCC |
Upregulated in MCC |
Downregulated in MCC |
Highly expressed in MS-1 | ||||
---|---|---|---|---|---|---|---|
microRNA | Fold change2 | microRNA | Fold change2 | microRNA | Fold change2 | microRNA | RPKM |
miR-885 | 234.3 | miR-183 | 54.6 | miR-455 | −100.0 | miR-182 | 441,774 |
miR-1252 | 159.4 | miR-182 | 44.3 | miR-146a | −33.3 | miR-183 | 406,019 |
miR-190b | 72.3 | miR-96 | 26.4 | miR-125b-2 | −16.7 | miR-10b | 368,383 |
miR-876 | 69.7 | miR-7-2 | 9.0 | miR-224 | −16.7 | miR-30d | 354,071 |
miR-873 | 62.2 | miR-7-1 | 8.4 | miR-125b-1 | −16.7 | let-7i | 302,976 |
miR-1468 | 42.5 | miR-769 | 6.0 | miR-452 | −12.5 | miR-30a | 256,918 |
miR-3065 | 33.8 | miR-708 | 5.8 | miR-27a | −8.3 | miR-21 | 231,060 |
miR-3074 | 19.9 | miR-93 | 5.7 | miR-503 | −6.7 | miR-26a | 230,375 |
miR-1250 | 15.6 | miR-106b | 5.7 | miR-34a | −5.3 | miR-9-2 | 152,761 |
miR-502 | 15.2 | miR-9-2 | 4.7 | miR-378d-2 | −4.8 | miR-20a | 118,839 |
miR-660 | 14.4 | miR-532 | 4.7 | miR-24-2 | −4.0 | miR-532 | 109,596 |
miR-501 | 9.3 | miR-9-3 | 4.7 | miR-193a | −4.0 | miR-93 | 108,909 |
miR-708 | 9.2 | miR-9-1 | 4.7 | miR-378i | −3.6 | miR-340 | 89,562 |
miR-532 | 8.2 | miR-340 | 4.0 | miR-22 | −3.0 | miR-7-1 | 73,329 |
miR-500a | 7.6 | miR-192 | 3.5 | miR-34c | −2.9 | miR-96a | 60,009 |
To validate the next generation sequencing (NGS) data, several microRNAs were evaluated via qRT-PCR in larger cohorts of FFPE tissue samples. The MCC cohort consisted of a mixture of primary cutaneous lesions and metastases (Table
Validation of MCC-miRs via qRT-PCR. Eight microRNAs were confirmed to be upregulated in MCC versus other tumors and normal skin: miR-190b, miR-9, miR-7, miR-182, miR-183, miR-873, miR-502-3p, and miR-340. The tumor group consists of melanoma (
To assess whether these microRNAs are specific tumor markers for MCC, the expression of each of the eight MCC-miRs was evaluated via qRT-PCR in several frozen MCC lymph node metastases and compared to a human tissue panel consisting of twelve different organs (Figure
MCC-miRs are specific for MCC. Three MCC-miRs were confirmed via qRT-PCR to be upregulated in frozen MCC samples versus a human tissue panel consisting of twelve different body organs: miR-182, miR-183, and miR-190b. Error bars refer to SEM.
To assess the potential of these findings for future functional studies, NGS of small RNA libraries was performed for the patient derived MCPyV-positive cell line, MS-1 (Table
MCC-miRs are highly expressed in MS-1. Four MCC-miRs were confirmed via qRT-PCR to be upregulated in the MCC cell line, MS-1, versus sixteen other non-MCC cell lines: miR-182, miR-183, miR-190b, and miR-340. Error bars refer to SEM.
The expression of each of the four microRNAs was also evaluated via qRT-PCR in the MCPyV-negative cell line, MCC13, but with different results (Figure
To support the notion that these microRNAs play a role in the actual tumor cells in lieu of the surrounding tissue, ISH was performed for one of the more highly expressed MCC-miRs, miR-182, on a sample of MCC of the cheek and on a sample of normal skin (Figure
MCC remains one of the least understood cancers of the skin. MicroRNAs are a relatively young field of biomedical research, born in 2000 with the detection of let-7 in humans, with potential for applications in other pathologies [
While these MCC-miRs are highly expressed in MCC, some of them have been demonstrated to play significant roles in the pathogenesis of other cancers as well. In particular, the miR-183/96/182 cluster, at chromosomal locus 7q32, is expressed in a diversity of cancers and may contribute to their pathogenesis by targeting multiple components of the cell cycle, DNA damage response, and homologous recombination pathways, and by enriching pathways associated with metastasis, migration, and epithelial-mesenchymal transition [
Regarding skin cancer, the miR-183 cluster is frequently overexpressed in melanoma. Our work confirms this observation, inasmuch as miR-182 and miR-183 were upregulated in melanoma versus SCC, BCC, and normal skin (although still not as highly expressed as in MCC). In melanoma, overexpression of miR-182 promotes survival, migration, and metastasis by directly repressing the tumor suppressors FOXO3 and microphthalmia-associated transcription factor-M; and expression of miR-182 increases with progression from primary to metastatic melanoma [
The miR-183/96/182 cluster serves essential functions in various noncutaneous carcinomas as well, with much research focused on its role in breast cancer, particularly with invasion and metastasis. For example, in mammary ductal carcinoma
The molecular pathways altered in MCC pathogenesis have yet to be fully characterized, but a literature review reveals that some connections to well-known tumorigenic pathways have been made. For example, multiple research groups have discovered that the PI3K/AKT/mTOR pathway is activated, independent of MCPyV-status, in the majority of human MCCs, identifying it as a potential new therapeutic target [
As an additional note, data also suggest that inactivation of PTEN may play a role in MCC pathogenesis; however, mutation and homozygous deletion screening of the PTEN gene in tumor samples reveals nonsense mutations and homozygous deletions in only a small subset of patients [
Furthermore, some of these MCC-miRs, when employed in combination, could be potentially useful in the diagnosis of MCC. Misdiagnosis is high on the list of issues associated with this cancer; however, unlike BCC, for which it may often be confused [
Recently, Renwick et al. demonstrated that multicolor microRNA FISH can be utilized to effectively differentiate between MCC and BCC in FFPE tissues. The researchers employed miR-205 and miR-375, which were shown to be tumor-specific for BCC and MCC, respectively [
In addition, our results from the tissue panel demonstrate that the MCC-miRs could also potentially serve as markers for tumor metastasis (Figure
Employing microRNA signatures as diagnostic tools has been successfully carried out for other skin cancers. In 2011, Ralfkiaer et al. developed a qRT-PCR-based classifier consisting of three microRNAs capable of differentiating CTCL from other cutaneous pathologies with high accuracy [
Finally, we demonstrated that several of these MCC-miRs are highly expressed in the patient-derived MCC cell line, MS-1. The same findings were not demonstrated with the MCPyV-negative cell line, MCC13; however, we are not the first to experience such findings. In their previously mentioned study, Renwick et al., upon clustering samples via comparison of microRNA profiles, found that MCPyV-positive cell lines (MS-1, MKL-1, MKL-2) clustered in the MCC group, while MCPyV-negative cell lines (MCC13, MCC26, UISO) clustered in the non-MCC group [
This raises the question of which cell line, MS-1 or MCC13, holds more validity as a surrogate of MCC
Based upon these findings, we believe that our results in the MS-1 and MCC13 cell lines corroborate the existing literature that suggests that only the MCPyV-positive cell lines truly mimic MCC. Its cellular, immunohistochemical, and virological features, along with its high expression of MCC-miRs, showcase MS-1 as an attractive candidate for future studies. Further evaluation of the miRNomes of other MCPyV-positive cell lines (MKL-1, MKL-2) would be valuable in supporting this notion.
The present study was approved by the Institutional Review Board of Vanderbilt University prior to initiation (IRB ID 111357).
The authors declare that there is no conflict of interests regarding the publication of this paper.
Support of this work by the American Skin Association via the Medical Student Grant Targeting Melanoma and Skin Cancer Research is gratefully acknowledged. This work was also supported by the Meharry-Vanderbilt-TSU Cancer Research Partnership (NIH/NCI U54 CA91405-10) and by CTSA Award no. UL1TR000445 from the National Center for Advancing Translational Sciences. Its contents are solely the responsibility of the authors and do not necessarily represent official views of the National Center for Advancing Translational Sciences or the National Institutes of Health. In addition, The authors are indebted to Dr. Lloyd E. King, Jr., M.D., Ph.D. of Vanderbilt University Medical Center, for providing them with several of the MCC samples utilized in this study. The authors also thank the Chang-Moore Laboratory at the University of Pittsburgh for providing them with the MS-1 cell line.