Cutaneous melanoma is one of the most common types of cancer, and the incidence is increasing worldwide.1 Histopathologically, the diagnosis is based on features such as asymmetry of the lesion, pagetoid spread of melanocytes up in the epidermis, no maturation of the melanocytes toward the bottom of the lesion, deep mitosis, and lymphocytic response.2 Distinguishing melanoma from benign naevi can be a challenge, and a clear marker of melanoma has yet to be identified.
Several studies suggest that increased telomerase activity plays a critical role in the development of melanoma.3–5 Telomeres are tandem repeats of 5′-TTAGGG-3′ sequences located at the end of the human chromosomes, where they protect the chromosomes from damage and end-to-end fusion.6–8 Each cell division causes the telomeres to shorten because the DNA polymerase cannot replicate the end of a linear template (the “end-replication problem”).9 This causes the cell to go into senescence or induce apoptosis, which limits the proliferative potential of normal cells.10,11 Cancer cells are characterized by an infinite division potential,12 and in ∼90% of all cancers, this is due to reactivation of the ribonucleoprotein telomerase.13 Cancer cells without reactivated telomerase can maintain telomere length by an alternative lengthening of the telomeres, involving copying telomeric template DNA by homologous recombination.14
Telomerase is a reverse transcriptase that is capable of lengthening the telomeres by adding telomeric DNA to the ends of chromosomes.15–17 The essential components of telomerase are the human telomerase reverse transcriptase (hTERT or TERT) and human telomerase RNA (hTR/hTER/TERC). Additional proteins are associated with the complex either constantly or transiently.18 hTR acts as a template for the reverse transcription19–21 and hTERT is the catalytic subunit with the reverse transcription activity.22
Because telomerase is mostly present in cancerous tissue, and only modestly in healthy tissue, telomerase or its components could be useful to distinguish malignant tissue from benign tissue. hTR is present in normal tissue to a greater extent than hTERT, and is also present in cells with no detectable telomerase activity,23 hence, hTERT is potentially a better marker for malignancy than hTR. Therefore, a sensitive method for in situ detection of hTERT in tumor tissue is desirable for histopathologic purposes in tumor tissue. Detection of the hTERT protein using immunohistochemistry lacks sensitivity, and a reliable antibody toward the protein has been a challenge to find.24 For this reason, it is possibly more reliable to detect the mRNA of hTERT instead of the protein. However, conventional in situ hybridization for the detection of hTERT mRNA can give a significant background staining of the tumor microenvironment, which can compromise the results. RNAscope could, instead, be a good alternative for detecting hTERT mRNA in formalin-fixed paraffin-embedded (FFPE) tissue. RNAscope is a novel in situ hybridization technology for detecting RNA based on the Advanced Cell Diagnostics–patented probe design. A pair of oligonucleotide probes are used to hybridize to the target mRNA molecule, which makes it possible to detect small amounts of mRNA in FFPE tissue. Using RNAscope, the background noise is reduced significantly, giving more clear results.25
The aim of this study was to establish an RNAscope assay in our laboratory, to detect the mRNA of hTERT in melanomas and benign naevi using RNAscope, and to compare the expression of hTERT mRNA between the malignant and the benign tissue. Furthermore, we wanted to compare the hTERT mRNA expression with the Ki67 proliferation index and the Breslow thickness.
MATERIALS AND METHODS
This study was conducted after approval by the Regional Committee on Health Research Ethics (1-10-72-178-15).
We wanted to examine RNAscope by performing the assay on both cell lines and tissue. Therefore, we developed the assay on a telomerase positive and a telomerase-negative cell line to establish positive and negative controls. Furthermore, we treated a telomerase positive melanoma with RNase to exclude the possibility of detecting DNA in the tissue. In addition, we performed the assay on Hodgkin’s lymphoma tissue, because of the characteristic Hodgkin’s Reed-Sternberg (HRS) cells, to show that hTERT mRNA is present in the neoplastic cells and not in the bystander cells. After establishing the method, we performed the assay on the study population.
The human cervical carcinoma epitheliod cell line, HeLa (telomerase positive), and the human osteosarcoma cell line, U2OS (telomerase-negative), were cultured in the Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 1% penicillin streptomycin (Pen Strep), and 0.1% fungilin. Cells from two T-175 cell culture flasks with 90% confluent cells were harvested using trypsin and transferred to a 15 mL centrifuge tube. After washing with 1×phosphate-buffered saline, the cells were centrifuged and the phosphate-buffered saline was removed. The cell pellet was then treated with 3 drops of human plasma and 2 drops of thrombin. When the cells had formed a soft ball, they were fixed in formalin for 24 hours and then embedded in paraffin to simulate how the tissue is normally treated. The cells were cut in a 4 μm thickness on a microtome and embedded on the slides.
Archival FFPE tissue specimens of 3 Hodgkin’s lymphomas positive for Epstein-Barr Virus diagnosed at Odense University Hospital, Herlev hospital, and Næstved Hospital in 2008 and 2009, and of a primary cutaneous melanoma diagnosed at Aarhus University Hospital in 2011 was used for establishing the method.
Our study population consisted of archival FFPE tissue samples from 17 patients with primary cutaneous melanoma (12 superficial spreading melanomas and 5 nodular melanomas) and from 13 patients with benign naevi (11 compound naevi and 2 dermal naevi), all diagnosed at Aarhus University Hospital in 2011. The cohort was previously used for the detection of hTR.26 Melanomas with a Clarks level IV and a Breslow thickness of at least 1 mm were included. Clinical information about the patients is shown in Table 1.
To eliminate the possibility of detection of DNA in the tissue, a 4-μm section from one melanoma was cut on microtome and mounted on Superfrost Plus slides (Thermo Fisher Scientific, Waltham, MA), and treated with a saturating amount of RNase A/T1 Mix (Thermoscientific #EN0551) before performing RNAscope. The slide was baked in a dry oven for 60 minutes at 60°C, deparaffinized by incubating in xylene followed by 99% ethanol, and then pretreated according to the manufacturer’s instructions. In brief, the pretreatment consisted of submerging the slides into heated 1× Pretreat 2 reagent followed by wash in Tissue-Tek and distilled water. The RNase was added and the slide was loaded into the slide drawers on the Ventana Discovery Ultra and the RNAscope was performed.
The cohort tissue samples were cut in a series of 3 sections. The middle sections were used for RNAscope, and the adjacent sections were stained for melan-A using immunohistochemistry. Another section was used for Ki67/melan-A double staining.
The tissue samples were cut in 5 μm thickness using a microtome and mounted on Superfrost Plus slides along with a section of HeLa cells. The tissue samples were dried for 1 hour at 60°C. For the detection of mRNA of hTERT in the specimens, the fully automated RNAscope assay was performed according to the manufacturer’s instructions.25 This consisted of finding the appropriate heating time, protease time, and amplification time for each slide. Generally, the heating time varied from 16 to 24 minutes, the protease time was 24 minutes, and the amplification time was 1 hour. The assay was performed on a Discovery Ultra automation system (Ventana medical systems, Tucson, AZ) using RNAscopeVS 2.5 Reagent Kit-RED with the Probe-Hs-TERT (catalog no. 605519) (Advanced Cell Diagnostics, Newark, CA) specific to the region between nucleotides 2164 and 3231. For quality control of each sample, RNAscope control probes targeting a common housekeeping gene, PPIB (positive control), and a bacterial gene, DapB (negative control), were included. The slides were manually dehydrated and mounted.
Melan-A staining was performed on the cohort using an indirect immunoenzymatic technique, and a sequential immunoenzymatic double staining technique was performed on the melanoma sections, on Benchmark XT (Ventana) with anti-Ki67 combined with anti-melan-A. The 3-μm sections of the tissue samples, mounted on Superfrost Plus slides, were dried for 1 hour at 60°C. Standard settings and reagent kits of Benchmark XT were used in deparaffinization, rehydration, antigen retrieval (CC1 mild), and endogenous peroxidase blocking. For the melan-A stain the monoclonal mouse antibody Melan-A (Melan-A M2-7C10, dilution 1:50; Cell Marque, Rocklin, CA) was incubated for 32 minutes at room temperature followed by Ventanas XT ultraView Universal Alkaline Phosphatase Red Detection kit. Slides were counterstained with Mayer’s hematoxylin and bluing reagent. For the double staining the monoclonal rabbit antibody Ki67 (clone 30-9; Ventana) was incubated for 16 minutes at room temperature, followed by Ventanas ultraView Universal DAB detection kit. Monoclonal mouse antibody Melan-A (Melan-A M2-7C10, dilution 1:50; Cell Marque) incubated 40 minutes at room temperature followed by Ventanas XT ultraView universal alkaline phosphatase red detection kit. The slides were manually dehydrated and mounted.
Images of the whole slides were captured by Nanozoomer (Hamamatsu Phototonics KK, Hamamatsu City, Japan) at a magnification of ×40, z-stacking in 5 layers of 0.2 μm and −1 as the set off for the RNAscope slides, and regular ×20 for the melan-A and Ki67/melan-A immunohistochemical slides. The slides were imported to Visiopharm Integrator System (Visiopharm, Hoersholm, Denmark), where the RNAscope slides and the melan-A IHC slides were aligned (ie, slides were digitally superimposed). This virtual double staining made it possible to identify the tumor regions, and consequently the tumor cells, with the melan-A slide, and the hTERT mRNA in this region was counted on the RNAscope-slide. The manual count of the hTERT mRNA signals was executed with the aid of a counting frame. Approximately 30 counting frames were randomly distributed throughout the tumor and placed in a melanocyte-rich area assessed by the melan-A stain. The amount of melanocyte nuclei and hTERT mRNA signals were counted on the RNAscope-slide. Ten slides were recounted by the same observer.
The Ki67 index of each Ki67/melan-A double-stained melanoma sample was automatically quantified by the Visiopharm Integrator System (Visiopharm), as described elsewhere,27 within a manual outline of the dermal compartment. The Ki67 index was calculated based on the area of Ki67-positive nuclei of melan-A-positive cells out of the total melan-A-positive tumor area.
Statistical analyses were performed in Stata 14.1 (StataCorp, College Station, TX). The nonparametric Mann-Whitney rank test was used to calculate the significance level of the difference in hTERT mRNA expression between the melanomas and the benign naevi. The Spearman rank correlation was used to asses the association between hTERT mRNA expression and Breslow thickness, and to asses the association between hTERT mRNA expression and the Ki67 index. A 2-way receiver operating characteristic was consulted to choose the most optimal cutoff.
Establishing the Method
No hTERT mRNA signals were found in the RNase-treated melanoma. When not treated with RNase the same melanoma showed hTERT mRNA expression mainly in the nuclei of the cells.
We saw no hTERT mRNA signals in the telomerase-negative U2OS cell line. The telomerase positive HeLa cell line showed hTERT mRNA expression. The signals were present in both the cytoplasm and in the nucleus of the cells, and not all cells had a signal. In addition, we saw hTERT mRNA signals in the nucleus of the neoplastic HRS cells and only very few signals in the cytoplasm.
Detection of hTERT mRNA in the Cohort
We found significantly (P<0.001) more hTERT mRNA signals per neoplastic cell in melanoma compared with benign naevi. The signals were localized to the nuclei of the cells, and rarely in the cytoplasm (Fig. 1). Figure 2 illustrates the difference in hTERT mRNA expression per neoplastic cell in melanomas and benign naevi.
The Spearman rank correlation showed an association between the amount of hTERT mRNA per neoplastic cell in melanoma and Breslow thickness (ρ=0.56, P=0.0205), as depicted in Figure 3. In addition, a high Ki67 proliferation index correlated with a high hTERT mRNA expression. This is shown in Figure 4 and is supported by a Spearman’s ρ of 0.72 (P=0.001).
A sensitivity of 88.2% and a specificity of 76.9% was found based on the best cutoff which was estimated to be 0.0052 hTERT mRNA signals per neoplastic cell.
Our study showed that hTERT mRNA can be reliably detected by RNAscope in FFPE melanoma. Only 2 other studies have used the assay to detect hTERT mRNA in meningioma and spitzoid melanoma.28,29
In addition, we found that there is a statistically significant abundance of hTERT mRNA in melanomas in comparison with benign naevi. This is consistent with what others have found using immunohistochemistry.30 Three melanomas (of the 17) expressed no hTERT mRNA, whereas 2 benign naevi (of the 13) expressed the same amount of hTERT mRNA as some of the melanomas. Because of this overlap, hTERT mRNA is not an obvious diagnostic marker of melanoma. Arguably, hTERT mRNA could be used in cases where there is doubt about the diagnosis, as hTERT mRNA is mainly expressed in melanomas, and only in very few benign naevi, although it would not be enough to give a definite diagnosis.
In our study, we demonstrated that the probe truly targets RNA and not DNA by using RNase treatment, and we showed that the signals were indeed hTERT mRNA, by performing the assay on the telomerase-negative U2OS cell line. These investigations emerged from the observation that we mostly saw the hTERT mRNA signals in the nuclei of the cells. Because of what we know about mRNA, we expected to see the signals mostly in the cytoplasm. Others have made the same observation when using RNAscope to detect E6/E7 mRNA on FFPE oropharyngeal squamous cell carcinoma tissue.31 It is possible that we detected pre-mRNA at sites of active transcription, that is, the nucleus. Others have hypothesized this as well.31,32 A potential problem is that we cannot be sure that the pre-mRNA will become the functioning telomerase protein, but because hTERT transcription is the primary step in the regulation of telomerase,33 we would not detect any mRNA if it was not to be translated into hTERT protein. However, alternative splice sites for hTERT mRNA have been detected, and so far it is believed that only the wild-type transcript will translate into the functioning protein. The RNAscope probe only detects part of the transcript (from approximately exon 5 to exon 15), but 4 of the most studied alternative splice sites are located within this region.34 This means, that the RNAscope probe probably detects the alternative splice variants as well as the wild-type hTERT mRNA. As a result, it is uncertain whether the detected mRNA will become the functioning protein. Nevertheless, in the case of using hTERT mRNA to differentiate between benign and malignant, this is possibly of little significance.
We found that the cells usually contained a single signal equivalent to a single mRNA molecule. Some cells, however, expressed >1 signal, as much as 10 signals in a cell was observed, and some cells did not express any hTERT mRNA. This is consistent with other studies.35,36 The variability in expression may be explained by different cell cycle stages, the fragile nature of RNA, or it could be due to the sectioning. We saw that the Hodgkin’s lymphomas expressed hTERT mRNA in the HRS cells (the neoplastic cells) and not in other cells, which corresponds to the fact that hTERT mRNA is present in tumor cells, and absent in normal cells.37
Breslow thickness is regarded as the most important indicator of prognosis, as the survival rate for melanoma decreases with an increase in thickness.38 We found that hTERT mRNA expression is associated with Breslow thickness. Carvalho et al5 demonstrated that telomerase activity increases with increasing Breslow thickness, which agrees with our findings, if we assume that hTERT mRNA is a surrogate marker of telomerase activity. This correlation between hTERT mRNA expression and the thickness of the melanoma makes biological sense, as you would assume that a thick melanoma has benefitted from an increased telomerase activity for the vivid proliferation.
By comparing the hTERT mRNA expression with the Ki67 index in the melanomas, we found that there is a correlation between hTERT mRNA expression and proliferation. This is consistent with results from a study on papillary thyroid carcinoma by Chou et al.39 The elevated hTERT mRNA expression in melanomas with a high Ki67 proliferation index is associated with cancer cells’ dependence on telomerase to proliferate. These findings suggest that hTERT mRNA could be a prognostic marker of melanoma. To examine this, a study comparing the hTERT mRNA expression and the patients’ outcomes needs to be performed.
In conclusion, we have used the novel in situ hybridization technique, RNAscope, for detection of hTERT mRNA in melanoma and benign naevi. We conclude that RNAscope is a useful method for mRNA detection in FFPE tissue, and that the expression of hTERT mRNA is significantly more abundant in melanoma compared with benign naevi. For the distinction between malignant and benign melanocytic lesions, hTERT mRNA could be used as a supplement to other markers, but cannot be used as an independent diagnostic marker. Furthermore, we can conclude that the expression of hTERT mRNA in melanoma is associated with Breslow thickness and the Ki67 proliferation index, indicating that hTERT mRNA is a potential prognostic marker of melanoma. Further studies need to be performed to validate these results.
1. Fitzmaurice C, Dicker D, Pain A, et al. The global burden of cancer 2013. JAMA Oncol. 2015;1:505–527.
2. Smoller BR. Histologic criteria for diagnosing primary cutaneous malignant melanoma. Mod Pathol. 2006;19(S2):S34–S40.
3. Ramirez RD, D’Atri S, Pagani E, et al. Progressive increase in telomerase activity from benign melanocytic conditions to malignant melanoma. Neoplasia. 1999;1:42–49.
4. Miracco C, Margherita De Santi M, Schurfeld K, et al. Quantitative in situ evaluation of telomeres in fluorescence in situ hybridization-processed sections of cutaneous melanocytic lesions and correlation with telomerase activity. Br J Dermatol. 2002;146:399–408.
5. Carvalho L, Lipay M, Belfort F, et al. Telomerase activity in prognostic histopathologic features of melanoma. J Plast Reconstr Aesthet Surg. 2006;59:961–968.
6. Muller H. The remaking of chromosomes. Collecting Net. 1938;13:181–198.
7. McClintock B. The stability of broken ends of chromosomes in Zea mays
. Genetics. 1941;26:234–282.
8. Moyzis RK, Buckingham JM, Cram LS, et al. A highly conserved repetitive DNA sequence, (TTAGGG)n, present at the telomeres of human chromosomes. Proc Natl Acad Sci U S A. 1988;85:6622–6626.
9. Watson JD. Origin of concatemeric T7 DNA. Nat New Biol. 1972;239:197–201.
10. Olovnikov AM. A theory of marginotomy. J Theor Biol. 1973;41:181–190.
11. Harley CB, Futcher AB, Greider CW. Telomeres shorten during ageing of human fibroblasts. Nature. 1990;345:458–460.
12. Hanahan D, Weinberg Robert A. Hallmarks of cancer: the next generation. Cell. 2011;144:646–674.
13. Kim N, Piatyszek M, Prowse K, et al. Specific association of human telomerase activity with immortal cells and cancer. Science. 1994;266:2011–2015.
14. Cesare AJ, Reddel RR. Alternative lengthening of telomeres: models, mechanisms and implications. Nat Rev Genet. 2010;11:319–330.
15. Greider CW, Blackburn EH. Identification of a specific telomere terminal transferase activity in Tetrahymena extracts. Cell. 1985;43 (pt 1):405–413.
16. Greider CW, Blackburn EH. The telomere terminal transferase of tetrahymena is a ribonucleoprotein enzyme with two kinds of primer specificity. Cell. 1987;51:887–898.
17. Morin GB. The human telomere terminal transferase enzyme is a ribonucleoprotein that synthesizes TTAGGG repeats. Cell. 1989;59:521–529.
18. Meena J, Lenhard Rudolph K, Günes CGhadimi BM, Ried T. Telomere dysfunction, chromosomal instability and cancer. Recent Results in Cancer Research (Vol 200). New York: Springer (LLC); 2015:61–79.
19. Greider CW, Blackburn EH. A telomeric sequence in the RNA of Tetrahymena telomerase required for telomere repeat synthesis. Nature. 1989;337:331–337.
20. Shippen-Lentz D, Blackburn EH. Functional evidence for an RNA template in telomerase. Science. 1990;247:546–552.
21. Yu G-L, Bradley JD, Attardi LD, et al. In vivo alteration of telomere sequences and senescence caused by mutated Tetrahymena telomerase RNAs. Nature. 1990;344:126–132.
22. Lingner J, Hughes TR. Reverse transcriptase motifs in the catalytic subunit of telomerase. Science. 1997;276:561–567.
23. Avilion AA, Piatyszek MA, Gupta J, et al. Human telomerase RNA and telomerase activity in immortal cell lines and tumor tissues. Cancer Res. 1996;56:645–650.
24. Yan P, Benhattar J, Seelentag W, et al. Immunohistochemical localization of hTERT protein in human tissues. Histochem Cell Biol. 2004;121:391–397.
25. Wang F, Flanagan J, Su N, et al. RNAscope
: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues. J Mol Diagn. 2012;14:22–29.
26. Vagner J, Steiniche T, Stougaard M. In-situ hybridization-based quantification of hTR: a possible biomarker in malignant melanoma. Histopathology. 2015;66:747–751.
27. Nielsen PS, Riber-Hansen R, Raundahl J, et al. Automated quantification of MART1-verified Ki67 indices by digital image analysis in melanocytic lesions. Arch Pathol Lab Med. 2012;136:627–634.
28. Abedalthagafi MS, Bi WL, Merrill PH, et al. ARID1A and TERT promoter mutations in dedifferentiated meningioma. Cancer Genet. 2015;208:345–350.
29. Wu G, Barnhill RL, Lee S, et al. The landscape of fusion transcripts in spitzoid melanoma and biologically indeterminate spitzoid tumors by RNA sequencing. Mod Pathol. 2016;29:359–369.
30. Fullen DR, Zhu W, Thomas D, et al. hTERT expression in melanocytic lesions: an immunohistochemical study on paraffin-embedded tissue. J Cutan Pathol. 2005;32:680–684.
31. Ukpo OC, Flanagan JJ, Ma XJ, et al. High-risk human papillomavirus E6/E7 mRNA detection by a novel in situ hybridization assay strongly correlates with p16 expression and patient outcomes in oropharyngeal squamous cell carcinoma. Am J Surg Pathol. 2011;35:1343–1350.
32. Brown J, Buckle VBridger JM, Volpi EV. Detection of nascent RNA transcripts by fluorescence in situ hybridization. Fluorescence In Situ Hybridization (FISH) (Vol 659). New York: Humana Press; 2010:33–50.
33. Zhang F, Cheng D, Wang S, et al. Human specific regulation of the telomerase reverse transcriptase gene. Genes. 2016;7:7.
34. Bollmann FM. Physiological and pathological significance of human telomerase reverse transcriptase splice variants. Biochimie. 2013;95:1965–1970.
35. Ducrest AL, Szutorisz H, Lingner J, et al. Regulation of the human telomerase reverse transcriptase gene. Oncogene. 2002;21:541–552.
36. Akincilar SC, Unal B, Tergaonkar V. Reactivation of telomerase in cancer. Cell Mol Life Sci. 2016;73:1659–1670.
37. Yi X, Shay JW, Wright WE. Quantitation of telomerase components and hTERT mRNA splicing patterns in immortal human cells. Nucleic Acids Res. 2001;29:4818–4825.
38. Thompson JF, Scolyer RA, Kefford RF. Cutaneous melanoma. Lancet. 2005;365:687–701.
39. Chou S-J, Chen C-M, Harn H-J, et al. Detection of hTERT mRNA relates to Ki-67 labeling index in papillary thyroid carcinoma. J Surg Res. 2001;99:75–83.