Aberrant mRNA processing is known to drive the pathogenesis of chronic lymphocytic leukemia (CLL). Recurrent gene mutations in the RNA splicing factor SF3B1 and widespread RNA intronic polyadenylation impact genome-wide gene expression and inactivate tumor suppressors, respectively. Nevertheless, how mRNA processing is regulated and exerts its function in CLL remain elusive.
To comprehensively characterize the role of mRNA processing in CLL, we performed RNA sequencing (RNA-seq) and Tandem Mass Tag (TMT) proteomics using normal and CLL B cells derived from healthy donors (n=5) and untreated CLL patients (n=22). We detected 328 proteins differentially expressed between normal and CLL B cells (|Log2FC|>0.58, q<0.05). Gene set enrichment analysis (GSEA) revealed that proteins involved in RNA metabolism (transcription, splicing, modification, 3'end processing, nuclear export, decay) were upregulated in CLL, while those impacting translation were downregulated. These findings were validated by immunoblotting in an independent set of samples (n=10). However, we observed no significant gene expression changes of RNA metabolism at the transcript level, indicating that regulation of these proteins occurred post-transcriptionally.
Since N6-methyladenosine (m6A) is the most abundant RNA internal modification and has emerged as a key regulator for RNA metabolism, we sought to determine whether m6A is dysregulated in CLL cells. With an m6A dotblot assay and HPLC-MS, we consistently detected increased level of m6A in mRNA from CLL cells compared with normal B cells. As one of the most upregulated proteins in CLL, METTL3 writes m6A and promotes translation efficiency through its writer and reader functions, respectively. When we knocked down (KD) METTL3 in CLL cell lines (HG3, MEC1) as well as in primary CLL cells, we observed significant cell death and growth disadvantage in CLL compared to control cells, highlighting METTL3 is essential for CLL survival.
We next examined whether KD of METTL3 affects m6A and RNA translation using m6A dotblot and O-propargyl-puromycin run on assays. Loss of METTL3 had subtle impact on m6A levels but it significantly decreased protein translation (t test, p<0.01) in all the cell lines tested (HG3, MEC1, JeKo-1, Mino). To define the target protein that METTL3 affects, we performed an integrated Ribosome profiling and RNA-seq analysis using HG3 and Mino cells with or without METTL3. At both transcriptome and translatome levels, loss of METTL3 significantly decreased genes enriched in the mTORC1 pathway, which has an essential role in translation (Metascape, hypergeometric test, q<0.05). Furthermore, it also decreased the translation efficiencies of genes involved in mRNA processing, DNA synthesis, and cell cycle pathways. This observation suggests that upregulation of METTL3 in CLL cells may regulate protein translation of the RNA metabolism pathway.
Since m6A at the stop codon region is critical for METTL3 regulating protein translation, we performed MAZTER sequencing to determine m6A modification sites in normal and CLL B cells derived from healthy donors (n=5) and untreated CLL patients (n=11). We identified 214 genes with significant differential m6A modification at the stop codon region (delta cleavage efficiency>0.1, Wilcoxon rank-sum test, p<0.1, within DRACA motif) between normal and CLL B cells. These genes were highly enriched for mRNA processing (Metascape, q=0.017), supporting our notion that METTL3 may modulate protein expression of mRNA processing genes by recognizing m6A modification via its reader function in CLL. Consistent with its role in regulating protein expression, we detected downregulation of splicing factors (SF3A1, SF3A2, SF3B1, U2AF1) in various METTL3 KD cell lines (HG3, MEC1, JeKo-1, Mino) at only protein level but not transcription level. These data link METTL3 upregulation with RNA metabolism protein enrichment in CLL.
Altogether, our integrated analysis uncovered a novel regulatory axis of METTL3 in CLL biology. We demonstrated that CLL cells have an increased m6A modification and upregulation of METTL3 at the protein level, resulting in translation of RNA metabolism related genes through its reader function by the recognition of m6A modification. Our results collectively suggest METTL3 as a central regulator for mRNA processing in CLL and provide a rationale for targeting METTL3 in this disease.
Disclosures
Brown: Janssen, Teva: Speakers Bureau; Gilead, Loxo, Sun, Verastem: Research Funding; Abbvie, Acerta, AstraZeneca, Beigene, Invectys, Juno/Celgene, Kite, Morphosys, Novartis, Octapharma, Pharmacyclics, Sunesis, TG Therapeutics, Verastem: Consultancy. Rosen:Seattle Genetics: Consultancy; NeoGenomics: Consultancy; Aileron Therapeutics: Consultancy; Novartis: Consultancy; Pebromene: Consultancy; Celgene: Speakers Bureau; paradigm Medical Communications: Speakers Bureau; Abbvie: Speakers Bureau. Siddiqi:TG Therapeutics: Research Funding; Janssen: Speakers Bureau; Seattle Genetics: Speakers Bureau; Oncternal: Research Funding; BeiGene: Consultancy, Research Funding; Kite, a Gilead Company: Consultancy, Research Funding; Juno: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; Pharmacyclics: Consultancy, Research Funding, Speakers Bureau; AstraZeneca: Consultancy, Research Funding, Speakers Bureau.
Signatures of Dermal Fibroblasts from RDEB Pediatric Patients