Abstract Non-coding RNAs drive cancer phenotypes and are associated with patients’ outcome. SnoRNAs are required for the 2’-O-methylation (2’-O-Me) and pseudouridylation of ribosomal RNA (rRNA) thus being essential for ribosomal biogenesis and function. So far, the role of rRNA methylation (ribomethylation) for non-small cell lung cancer (NSCLC) pathogenesis is unknown. To investigate the role of ribomethylation we included samples from 92 patients with lung adenocarcinoma (pathological stage IA to IIB) for comprehensive multi-omics profiling. Amongst transcriptomics, small RNA sequencing, proteomics, and whole exome sequencing, ribomethylation as epitranscriptomic dimension was analyzed using RiboMethSeq. We focused the analysis on complete tumor resection samples. Here 47 out of 106 2’-O-Me sites were fully methylated in all tumor samples suggesting a crucial role for methylation at this position. Surprisingly, 59 sites had a dynamic methylation pattern. By combining ribomethylation sequencing data of dynamic sites with whole-exome-sequencing as well as transcriptome and proteome analyses using a multi-omics factor analysis (MOFA), we discovered a 2’-O-Me signature in a subset of NSCLC patients with a high risk of metastasis and poor prognosis. We termed the patient subset, with underlying ribomethylation signature, epitranscriptomic pro-metastatic phenotype (EPROMET). This phenotype was not associated with genetic mutations as analyzed by exome sequencing. Analysis of gene sets in both transcriptomics and proteomics showed an upregulation of secreted and extracellular matrix proteins in EPROMET patients. Alongside, the ribomethylation site with major contribution to the EPROMET signature; 18S-Um799, mediated through SNORD105/105B, was found to be the most dynamic within the dataset. It also showed the largest difference between EPROMET and non-EPROMET patients. For functional validation, lung cancer cell lines with knockout of snoRNAs contributing to EPROMET signature were generated using CRISPR/Cas9. Lung cancer cells lacking the corresponding rRNA modifications migrated slower in vitro, failed to grow at a distant site in vivo and were impaired in metastasis. Functionally, it was shown that the modulated ribomethylation signature effected translation of secreted proteins by altered mRNA binding to ribosomes. This study indicates the presence of a ribomethylome-associated phenotype of NSCLC related to metastasis and poor prognosis. Altered ribomethylation affects the ribosome function towards differential expression patterns of secreted proteins. Taken together, an epitranscriptomic pattern of 2’-O-Me is associated with invasion/migration properties of NSCLC cells and with development of metastasis. The ribomethylome may present a suitable biomarker and a potential therapeutic target. Citation Format: Cornelius Pauli, Daniel Heid, Christian Rohde, Nadja Krall, Sylvain Delaunay, Michael Kienhoefer, Christian Tischer, Michael Allgäuer, Maximilian Felix Blank, Fengbiao Zhou, Michael Kardorff, Michael Thomas, Hauke Winter, Sarah Sandmann, Marc Kriegsmann, Marc Schneider, Thomas Muley, Alexander Brobeil, Nicole Bäumer, Sebastian Bäumer, Simon Raffel, Albrecht Stenzinger, Peter Schirmacher, Junyan Lu, Judith Zaugg, Michaela Frye, Carsten Müller-Tidow. A specific ribomethylome pattern in lung cancer is associated with increased risk of metastasis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2763.
Read full abstract