Expression profiling by small RNA-seq for identifying miRNAs associated with pathological classifications in thyroid cancer.

There are no reliable small molecules to accurately differentiate indolent thyroid tumors from more aggressive thyroid cancers. This study aimed to develop new micro RNA (miRNA) markers for diagnosis and recurrence risk stratification of papillary thyroid carcinoma (PTC). Thyroid tumor-specific miRNA profiling was investigated in 34 fresh frozen tissues, which included nontumor (n = 7), noninvasive follicular thyroid neoplasms with papillary-like nuclear features (NIFTP, n = 6) and PTC (n = 21), based on small RNA-seq technique. The PTC samples also consists of a number of pathological sub-cateories including classic (n = 11), tall cell (n = 7), and encapsulated follicular variant (EFV, n = 3) subtypes. By applying unsupervised hierarchical clustering with miRNAs, we found that thyroid tumor samples were well differentiated from normal thyroid tissues. When the histological data were compared among the tumor samples, patients with NIFTP or EFV were found to be differently clustered from those with classic or tall cell, indicating that micro-environment displayed by miRNAs well reflect pathological characteristics. Overall design: miRNA-seq data of 34 samples including nontumor (n = 7), NIFTP (n = 6), classic (n = 11), tall cell (n = 7), and EFV (n = 3) subtypes were generated. Total RNA was isolated using the mirVana miRNA isolation kit (Ambion). Small RNAs (20 - 30 nt) were purified from 15% Novex TBE-Urea gel (Invitrogen) and ligated first with the 5' RNA adaptor and then with the 3' RNA adaptor provided by Illumina TruSeq small RNA sample preparation protocol. In each step, the ligated product was PAGE-gel purified. After first-strand synthesis and 11 cycles of PCR amplification, the product was PAGE-gel purified. Sequencing was performed in single end reads (75 bp) using NextSeq 500 platform (Illumina).