Multicolor microRNA FISH effectively differentiates tumortypes

摘要

MicroRNAs (miRNAs) are excellent tumor biomarkers because of their cell-typespecificity and abundance. However, many miRNA detection methods, such as real-timePCR, obliterate valuable visuospatial information in tissue samples. To enable miRNAvisualization in formalin-fixed paraffin-embedded (FFPE) tissues, we developedmulticolor miRNA FISH. As a proof of concept, we used this method to differentiatetwo skin tumors, basal cell carcinoma (BCC) and Merkel cell carcinoma (MCC), withoverlapping histologic features but distinct cellular origins. Using sequencing-basedmiRNA profiling and discriminant analysis, we identified the tumor-specific miRNAsmiR-205 and miR-375 in BCC and MCC, respectively. We addressed three majorshortcomings in miRNA FISH, identifying optimal conditions for miRNA fixation andribosomal RNA (rRNA) retention using model compounds and high-pressure liquidchromatography (HPLC) analyses, enhancing signal amplification and detection byincreasing probe-hapten linker lengths, and improving probe specificity usingshortened probes with minimal rRNA sequence complementarity. We validated our methodon 4 BCC and 12 MCC tumors. Amplified miR-205 and miR-375 signals were normalizedagainst directly detectable reference rRNA signals. Tumors were classified usingpredefined cutoff values, and all were correctly identified in blinded analysis. Ourstudy establishes a reliable miRNA FISH technique for parallel visualization ofdifferentially expressed miRNAs in FFPE tumor tissues.