Comparison of the miRNA profiles in HPV-positive and HPV-negative tonsillar tumors and a model system of human keratinocyte clones

BACKGROUND: Better insights into the molecular changes involved in virus-associated and -independent head and neck cancer may advance our knowledge of HNC carcinogenesis and identify critical disease biomarkers. Here we aimed to characterize the expression profiles in a matched set of well-characterized HPV-dependent and HPV-independent tonsillar tumors and equivalent immortalized keratinocyte clones to define potential and clinically relevant biomarkers of HNC of different etiology. METHODS: Fresh frozen tonsillar cancer tissues were analyzed together with non-malignant tonsillar tissues and compared with cervical tumors and normal cervical tissues. Furthermore, relative miRNAs abundance levels of primary and immortalized human keratinocyte clones were evaluated. The global quantitation of miRNA gene abundance was performed using a TaqMan Low Density Array system. The confirmation of differentially expressed miRNAs was performed on a set of formalin-fixed paraffin-embedded tumor samples enriched for the tumor cell fraction by macrodissection. RESULTS: We defined 46 upregulated and 31 downregulated miRNAs characteristic for the HPV-positive tonsillar tumors and 42 upregulated miRNAs and 42 downregulated miRNAs characteristic for HPV-independent tumors. In comparison with the expression profiles in cervical tumors, we defined miR-141-3p, miR-15b-5p, miR-200a-3p, miR-302c-3p, and miR-9-5p as specific for HPV induced malignancies. MiR-335-5p, miR-579-3p, and miR-126-5p were shared by the expression profiles of HPV-positive tonsillar tumors and of the HPV immortalized keratinocyte clones, whereas miR-328-3p, miR-34c-3p, and miR-885-5p were shared by the miRNA profiles of HPV-negative tonsillar tumors and the HPV-negative keratinocytes. CONCLUSIONS: We identified the miRNAs characteristic for HPV-induced tumors and tonsillar tumors of different etiology, and the results were compared with those of the model system. Our report presents the basis for further investigations leading to the identification of clinically relevant diagnostic and/or therapeutic biomarkers for tumors of viral and non-viral etiology. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12885-016-2430-y) contains supplementary material, which is available to authorized users.