High-throughput mRNA sequencing of CLL cells containing an E571K XPO1 mutations

Purpose: Exportin 1 (XPO1/CRM1) is a key mediator of nuclear export with relevance to multiple cancers, including chronic lymphocytic leukemia (CLL). Whole exome sequencing has identified hot-spot somatic XPO1 point mutations which we found to disrupt highly conserved biophysical interactions in the NES-binding groove, conferring novel cargo-binding abilities and forcing cellular mis-localization of critical regulators. However, the pathogenic role played by change-in-function XPO1 mutations in CLL is not fully understood. Thus, we aimed to identify disrupted cellular pathways in response to the E571K XPO1 mutation, the mutation most frequently cited in CLL patients, via high-throughput RNA-sequencing Results & Conclusion: Multidimensional scaling (MDS) analysis of the top 1000 most variable genes demonstrated distinctly unique clustering along the first dimension for XPO1-E571K patient and XPO1-wt patient samples. This clustering pattern was driven by a number of genes that were either over- or under-expressed in XPO1-E571K patient cells, contributing to a unique gene expression pattern in patients with this mutation signature. Overall design: Methods: We performed unbiased RNA-sequencing (RNA-seq) in CLL patient B-CLL cells containing an E571K XPO1 mutation (n=3) and compared this with XPO1-wt IgVH-U patient CLL cells (n=4). Library preparation and RNA-sequencing was performed and analyzed under protocols developed and approved by the Genomic Services Laboratory at Nationwide Children's Hospital. Downstream pathway analysis of differentially expressed genes was determined via publically available Ingenuity pathway analysis software as previously described. Paired-end 150bp reads were aligned to the human reference genome GRCh38 using HISAT2 and gene counts were generated with featureCounts. Gene expression was quantified as log2 counts per million and differential expression analysis was performed using R limma voom function. Heatmaps were generated using R pheatmap package.