利用嵌合eCLIP揭示头颈癌中血统相关的非编码RNA

Introduction: Chimeric enhanced cross-linking and immunoprecipitation (eCLIP) is a high-throughput method that enables direct mapping of Argonaute (Ago)-mediated RNA interactions. To our knowledge, this method has never been used to investigate ancestral differences in cancer outcomes. We performed chimeric eCLIP in African and European-ancestry head and neck cancer cell lines, with the goal of identifying canonical and non-canonical RNA interactions that may underscore differences in tumor behavior and therapeutic response related to ancestry. Methods: We abstracted gene expression data from patients in the Cancer Genome Atlas (TCGA) head and neck cancer cohort with different ancestral backgrounds. Differential expression analysis (edgeR) was used to identify differentially expressed genes (FDR < 0.05, -1 < log((FC)) > 1) and KEGG pathway analysis was performed to identify pathway enrichment (p adj. < 0.05). miRNA-target predictions were generated using miRnet. Admixture analysis was used to validate ancestry from head and neck cancer cell lines. Chimeric eCLIP protocol was performed on these cell lines. Briefly, Ago proteins were covalently cross-linked to their native RNA interaction partners. Ago immunoprecipitation was performed. RNA fragments were phosphorylated and ligated to generate RNA-RNA chimeras. Libraries were reverse transcribed, PCR amplified, and size selected to limit adapter dimer contamination. Paired-end sequencing (150 bp reads) was performed via Illumina NovaSeq. Results: Higher expressed genes in patients with self-reported African ancestry were enriched for drug metabolic pathways (p= 8.31E -10 ), while lower expressed were enriched for actin cytoskeletal function (p= 3.22E -52 ). miRnet predicted genes associated with drug metabolism and actin cytoskeletal function to be regulated by miR-200, miR-99, and miR-143/145 family members, which play a role in cell growth and survival, epithelial-mesenchymal transition, and cytoskeletal integrity. Preliminary eCLIP data supports this prediction, highlighting miRNA-mRNA pairings such as miR-100-5p/ABCF2, miR-200c-3p/CDK6, and miR-141-3p/ITGB8. Conclusion: We present evidence of ancestry-related noncoding RNA regulation associated with head and neck cancer. Future analyses will involve differential binding analysis, pathway enrichment analysis, and genetic overlay with SNPs to determine associations with ancestry. Overall, this methodology may be used to uncover pathways in other cancers with differing outcomes. ChatGPT was used for language editing and text refinement of this abstract.