Exploring cancer transcriptomes with long-read Nanopore sequencing

Cancer progression is associated with aberrant co- and post-transcriptional processing of RNA transcripts in both tumour and immune cells. This results in complex transcriptional profiles characterised by alternative splicing, 3’UTR and poly(A) tail length alterations, and chemical modifications. Despite advances, accurate profiling of transcript isoform usage and modifications remains challenging due to the limited read length offered by traditional RNA sequencing technologies. Here, we use long-read Nanopore direct RNA sequencing (DRS) and cDNA sequencing (PCS) to explore the transcriptomic profiles of clear cell renal cell carcinoma (ccRCC). Archival nephrectomy tumour samples from patients who experienced disease recurrence and controls were successfully analysed, demonstrating feasibility of the approach. Differential gene expression and transcript usage analysis identified changes in abundance and isoform usage associated with ccRCC recurrence. Gene expression-based cell-type deconvolution showed loss of tumour infiltrating immune cells, specifically CD8+ T cells in the recurrent tumours. Remarkably, using reference-guided transcriptome reconstruction methods, thousands of unannotated transcripts isoforms were identified, including isoforms of clinically important tumour immune checkpoints. DRS analysis of the ccRCC tumour cell line RCC4 revealed that some of these novel isoforms identified were indeed expressed in cancer cells and that exposure to inflammatory cytokines could lead to isoform switch. Notably, differential alterations in poly(A) tail length were also observed in novel and annotated transcripts in response to the exposure of cytokines. Finally, the role of mRNA N6-Methyladenosine (m6A) was explored in ccRCC tumour cell line via genetic perturbation of the m6A writer complex (METTL3 and WTAP) with CRISPR-Cas9 gene editing and siRNA-mediated transient depletion. Overall, our study demonstrates the feasibility of Nanopore long-read sequencing in tumour samples that uncovers cancer transcriptomes at single-transcript resolution and reveals the existence of multiple disease-associated alterations concurrently occurring in the mRNA of clinically relevant targets.