RPS15 and TP53 Co-Mutation Drives B Cell Malignancy through Altered Translation and MYC Activation in a Murine Model
Catherine Gutierrez,Tamara Ouspenskaia,Doris Fu,Peyton Waddicor,Anat Biran,Aviv Liani,Gregory Lazarian,Elisa Ten Hacken,Elizabeth Witten,Fara Faye Regis,Heather Joyal,Leah Billington,Fabienne Lucas,Mei Zheng,Blake Tye,Maria Hernandez-Sanchez,Miguel Quijada Álamo,Shuqiang Li,Binyamin A. Knisbacher,Ziao Lin,Aziz Al'Khafaji,Lili Wang,Kenneth J. Livak,Donna S. Neuberg,Florence Cymbalista,Gad Getz,Stirling Churchman,Ruben D. Carrasco,Sichen Shao,Aviv Regev,Catherine J. Wu
DOI: https://doi.org/10.1182/blood-2020-138900
IF: 20.3
2020-11-05
Blood
Abstract:Amongst the novel putative drivers identified by large-scale sequencing studies of chronic lymphocytic leukemia (CLL) is the ribosomal protein RPS15. Mutated in 5.3% of CLL, it co-occurs with heterozygous TP53 alterations in 36% of RPS15-mutated samples. Mutation of this mediator of ribosome maturation and translation is associated with poor disease prognosis and enriched in cohorts with del(17p) and relapsed CLL, suggesting a role in disease progression and therapeutic resistance. However, the impact of RPS15 mutation on B cell function and CLL development, in the presence or absence of TP53 mutation, has yet to be characterized. To this end, we developed overexpression HG3 CLL cell lines modeling four common RPS15 mutations (G134R, H137Y, S138F, and S139F) and a conditional knock-in mouse model of the S138F mutation with and without heterozygous Trp53 deletion (generated by crossing Rps15 and Trp53 mutant mice with Cd19-Cre mice). To characterize the impact of RPS15 mutation on transcription, we performed RNA-sequencing on splenic B cells from 3-month-old Rps15WT, Rps15Het and Rps15Hom mice (3 per cohort). We identified 255 and 670 upregulated and 596 and 777 downregulated genes in the Rps15MT vs Rps15WT mice (Rps15Het and Rps15Hom, respectively; log2FC>0.5, p<0.05). Gene set enrichment analysis (GSEA) revealed strong enrichment for MYC target genes that was also evident upon RNA-sequencing of the HG3 RPS15-S138F MT vs WT overexpression lines, and of 3 primary untreated CLLs with heterozygous RPS15 mutation (compared to 3 RPS15WT CLLs of similar genetic background). Pathway analysis of differentially expressed signatures across murine, cell line and primary CLL models revealed a common enrichment in translational machinery, such as mRNA splicing/processing, rRNA processing, and snRNP assembly (normalized enrichment score>1, nominal p-value<0.05). To evaluate whether RPS15 mutant proteins incorporate into ribosomes, we performed polysome profiling of the HG3 lines. All overexpressed RPS15-WT and MT proteins were observed to integrate into the small ribosomal subunit and mature ribosomes, potentially impacting translation. Next, ribosome profiling of HG3 RPS15-WT and S138F cells revealed 2,334 genes with differential translation efficiency (TE) between RPS15-S138F vs WT cells and 2,425 genes between RPS15-S138F vs WT in TP53 knock-out cells (log2FC>0.5, p<0.05). GSEA of differentially translated genes in RPS15 MT- vs WT cells revealed a strong enrichment for TP53-related genes, consistent with the activation of stress pathways by RPS15 mutant expression. RPS15 MT- vs WT cells with TP53-deletion, however, exhibited a strong increase in TE of MYC target genes and components of the ribosomal machinery. This finding suggests that loss of TP53 surveillance allows RPS15 MT cells to induce MYC-mediated changes in mRNA processing and translation - potentially setting the stage for oncogenesis. To determine whether Rps15 mutation can drive CLL-like disease, we engineered 6 novel mouse lines with B cell restricted expression of alterations through crossing with CD19-Cre mice: Rps15WT, Rps15Het, and Rps15Hom mutant mice alone or co-expressing Trp53 deletion. We detected circulating CLL-like (B220+CD5+) cells in 5 of 30 (17%) Rps15Het mice by 20 months of age, but not in 30 age-matched Rps15WT mice. We also detected CLL-like cells in 6 of 30 (20%) Trp53+/- mice by 17 months, indicating that Trp53 deletion alone can induce CLL-like disease. Interestingly, we found CLL-like cells in 2 of 30 Rps15Het/Trp53+/- mice as early as 15 months of age. The cohorts of Rps15Hom and Rps15Hom/Trp53+/- mice, however, have been monitored for 18 months of age with no disease occurrences, indicating that a double dosage of Rps15 mutation may be detrimental to disease formation. Altogether, Rps15 heterozygous mutation can drive CLL development in mice, and our early data hint that co-mutation with Trp53 may shorten the latency of CLL-like disease. Overall, RPS15 mutant protein can incorporate into the ribosome and induce changes in mRNA translation, resulting in MYC activation predominantly in the context of TP53 loss. Our mouse studies indicate that mut-Rps15 drives CLL development, with a more aggressive disease course when combined with Trp53 deletion. Our results collectively suggest that RPS15 and TP53 co-mutation drives CLL development through translational dysregulation and MYC-mediated signaling. Disclosures Neuberg: Pharmacyclics: Research Funding; Celgene: Research Funding; Madrigak Pharmaceuticals: Current equity holder in publicly-traded company. Getz:Broad Institute: Patents & Royalties: MuTect, ABSOLUTE, MutSig, MSMuTect, MSMutSig, POLYSOLVER and TensorQTL; Pharmacyclics: Research Funding; IBM: Research Funding; Scorpion Therapeutics: Consultancy, Current equity holder in publicly-traded company, Other: Founder. Wu:BionTech: Current equity holder in publicly-traded company; Pharmacyclics: Research Funding.
hematology