DNA-methylation patterns imply a common cellular origin of virus- and UV-associated Merkel cell carcinoma
November 23, 2021
November 23, 2021
There has been a long-standing controversy as to what type of cell, when infected with the Merkel cell polyomavirus, ultimately turns into what we know as Merkel cell carcinoma. Some studies indicate the cell of origin might be fibroblasts from the dermis, keratinocytes from the epidermis, or even B lymphocytes from the immune system. Most people agree though that mature Merkel cells are not likely the cell of origin, in part because they cannot seem to divide.
This study led by Gravemeyer and Becker was a collaboration between scientists in Germany & Michigan. Their data suggests that both the virus-induced and the sunlight-induced forms of MCC arise from ‘epidermal’ cells similar to keratinocytes.
Merkel cell carcinoma (MCC) is a neuroendocrine tumor either induced by integration of the Merkel cell polyomavirus into the cell genome or by accumulation of UV-light-associated mutations (VP-MCC and UV-MCC). Whether VP- and UV-MCC have the same or different cellular origins is unclear; with mesenchymal or epidermal origins discussed. DNA-methylation patterns have a proven utility in determining cellular origins of cancers. Therefore, we used this approach to uncover evidence regarding the cell of origin of classical VP- and UV-MCC cell lines, i.e., cell lines with a neuroendocrine growth pattern (n = 9 and n = 4, respectively). Surprisingly, we observed high global similarities in the DNA-methylation of UV- and VP-MCC cell lines. CpGs of lower methylation in VP-MCC cell lines were associated with neuroendocrine marker genes such as SOX2 and INSM1, or linked to binding sites of EZH2 and SUZ12 of the polycomb repressive complex 2, i.e., genes with an impact on carcinogenesis and differentiation of neuroendocrine cancers. Thus, the observed differences appear to be rooted in viral compared to mutation-driven carcinogenesis rather than distinct cells of origin. To test this hypothesis, we used principal component analysis, to compare DNA-methylation data from different epithelial and non-epithelial neuroendocrine cancers and established a scoring model for epithelial and neuroendocrine characteristics. Subsequently, we applied this scoring model to the DNA-methylation data of the VP- and UV-MCC cell lines, revealing that both clearly scored as epithelial cancers. In summary, our comprehensive analysis of DNA-methylation suggests a common epithelial origin of UV- and VP-MCC cell lines.