P-355: A novel genomewide RNA-Targeting CRISPR/Cas13 Screen Identifies a Plasma Cell-Specific long non-coding RNA (LncRNA) Essential for Myeloma Cell Growth

Introduction: Long non-coding RNAs (lncRNAs) are abundant RNA molecules that outnumber protein-coding genes in the human genome. They play significant roles in biological processes by interacting with proteins and other nucleic acids. Our previous work using RNA-seq data and a CRISPRi screen identified a series of lncRNAs as independent predictors of clinical outcome and confirmed their functional significance in promoting MM cell growth. However, CRISPRi focuses just on lncRNAs with defined transcription start sites.

Methods: To find tumor-promoting lncRNAs in a genomewide unbiased manner, we have exploited the RNA-targeting activity of the CRISPR-Cas13d endonuclease, which enables the selective knockdown of lncRNA transcripts of any genomic origin. We have used it to perform viability screens in five MM cell lines, targeting > 6,000 lncRNAs whose expression was identified using RNA-seq data from CD138+ MM cells from 360 MM patients. We developed a pooled lentiviral library comprising 60,000 sgRNAs, with 9 of them per lncRNA and > 500 non-targeting sgRNAs as negative controls. This library was infected at a low MOI ( < 0.3) into five MM cell lines (AMO1, H929, KMS11, OPM2, R8226) expressing Cas13d. After three weeks, we employed MAGeCK robust rank aggregation (RRA) algorithm to identify sgRNAs that displayed either depletion or enrichment within the MM cell population.

Results: We identified 155 lncRNAs essential for MM cell proliferation. A novel lncRNA named MYND1, was one of the most abundant lncRNAs in MM and normal plasma cells (median TPM>25), whereas it is significantly lower or absent in 54 normal tissues and cell types. RT-qPCR confirmed the plasma-cell specificity of MYND1 and its higher expression in MM cells compared to plasma cells. Furthermore, using subcellular qRT-PCR, we detected MYND1 mostly in the nucleus. Higher MYND1 expression predicted a worse clinical outcome in the newly diagnosed MM patients (n=360) enrolled in the IFM/DFCI clinical trial 2009 (NCT01191060). Knocking down MYND1 using antisense oligonucleotides in three MM cell lines (AMO1, H929, and KMS11) caused significant time-dependent inhibition of MM cell viability. Following MYND1 depletion in AMO1 cells, RNA-seq and GSEA identified significant modulation of expression of splice variants and inhibition of spliceosome-related gene signatures. MYND1 shares half of its sequence with U1 small nuclear RNA, a key component of the spliceosome complex. We confirmed the extensive interaction of MYND1 with spliceosome components HNRNPC, HNRNPH2, and HNRNPH3 in MM cells using ChIRP-MS and in vitro using RPPD-MS.

Conclusions: The functional impact of modulating MYND1 on spliceosome proteins is currently under investigation. Furthermore, ongoing nucleotide pairing analysis and structure simulations are being conducted to determine the optimal MYND1 targeting strategy. Our preliminary study suggests that MYND1 supports splicing activity, thus promoting MM cell growth and survival, and is a potential therapeutic target.