ABSTRACT: RNA-binding proteins (RBPs) have been shown to control mRNA processing, stability, transport, editing and translation. Application of large-scale quantitative technologies has facilitated genome-wide identification of RBPs and linked their defects to human diseases and carcinogenesis. We have recently conducted transcriptome analysis comparing normal cervical tissues with HPV-positive cervical cancer tissues by using two different RNA-seq platforms. As the results, 614 differentially expressed protein-coding transcripts were identified, which are enriched in cancer related pathways and consist of 95 known RBPs. TaqMan RT-qPCR was used to verify altered expression of 26 genes with a cohort of 72 cervical samples, including 24 normal cervical, 25 CIN 2-3, and 23 cervical cancer tissues. LY6K, FAM83A, CELSR3, ASF1B, IQGAP3, SEMA3F, CLDN10, MSX1, CXCL5, ASRGL1, ELAVL2, GRB7, KHSRP, NOVA1, PTBP1 and RNASEH2A were identified being novel candidates in association with cervical lesion progression and carcinogenesis. We further demonstrated that HPV16 or HPV18 infection leads to altered expression of 8 RBP genes (CDKN2A, ELAVL2, GRB7, HSPB1, KHSRP, NOVA1, PTBP1, and RNASEH2A) in human vaginal and foreskin keratinocytes. While both viral E6 and E7 decrease NOVA1 expression, viral E7 was found to increase the expression of both RNASEH2A and PCNA in an E2F1 dependent manner, two key factors closely associated with the progression of cervical neoplasia. By illustration of the first comprehensive cervical genome expression atlas, we have identified the altered expression of many novel genes due to HPV infection, which could be biomarkers for better diagnosis and treatment of cervical cancer.
