Abstract
Loss-of-function studies (LOF) have historically been the main functional approach to identify and study genes which drive the biology of Multiple Myeloma (MM) or other neoplasias. We hypothesized that substantial complementary data can be derived from systematically examining the impact of transcriptional activation of endogenous loci of different genes. We thus performed genome-scale CRISPR gene activation (CRISPRa) screens in 10 genotypically diverse MM cell lines (dCas9-P65-HSF transcriptional activation system; Calabrese genome-scale sgRNA library) to identify genes with significant sgRNA enrichment reflecting enhanced tumor cell fitness. These experiments identified that MM driver genes are highly cell line-dependent, with no universal “hit” but with many recurrent (e.g. in ≥4/10 lines) drivers, including key transcription factors/cofactors (e.g. POU2F2, POU2AF1, IRF4, MYC, HOXA3); growth factor signaling mediators, (e.g. IGF1R, IRS1); Ras family members (e.g. KRAS in MM1S cells); NFκB signaling (e.g. RELA, TIFA), PKC pathway (PRKCE), cell surface receptors (e.g. CD48, HTR1D, IGF1R, UPK3B), solute carrier transporter genes (e.g. SLC38A1). No correlation is currently observed between the pattern of driver genes for the individual MM cell lines and their respective molecular subtype (e.g., t(4;14), t(14;16), t(11;14), etc). Some MM-preferential dependencies identified by CRISPR knockout (KO) studies (de Matos Simoes et al Nat Cancer 2023) were identified by CRISPRa as recurrent MM drivers (e.g. POU2AF1, IRF4) or drivers with limited role in individual line(s) (e.g., IKZF3 and IKZF1 in MM1S cells). Notably, many top MM drivers are not essential for MM cells based on CRISPR KO studies. Most top MM drivers were not identified in similar CRISPRa studies as recurrent drivers for 17 non-MM cell lines from other hematologic neoplasias or solid tumors. With some notable exceptions (e.g., IRF4, POU2AF1), most other MM drivers are not identified in MM patient samples among the molecules with most frequent mutations, gene amplification or overexpression (e.g. vs normal plasma cells) at transcript or protein level. However, transcript upregulation (log2FC≥1.0) of at least one MM driver gene is observed in ~85% of paired MM samples after relapse vs. before treatment (MMRF CoMMpass study), suggesting that many CRISPRa-driver genes can contribute, alone or in concert, to enhanced MM cell fitness in clinical relapses. We validated the functional relevance of several MM driver genes (e.g. POU2F2, EGLN1, SLC38A1) with individual sgRNAs for CRISPR activation and/or cDNA overexpression (vs. isogenic controls) in competition experiments. We further probed the mechanistic basis for the role of POU2F2 as one of the top positive regulators of MM cell fitness and performed RNA sequencing analyses of MM1S cells harboring CRISPRa-based upregulation of POU2F2 expression. We observed upregulation of a distinct cluster of genes (including HOXA3, USP32, HTR1D) that are identified in our genome-scale CRISPRa studies as driver genes specifically in MM1S cells or ≥4 MM lines of our analyses, as well as several genes that are essential for MM survival (e.g., U2AF1, ZNF492, HYPK, PAM16, H2BC11). These results indicate that genome-scale CRISPRa studies provide data which are complementary to those derived from LOF studies. Indeed, our CRISPRa studies validated that MM cell fitness can be enhanced by upregulation of some genes that are prominently essential for MM cells in LOF studies (e.g. IRF4, POU2AF1) but also identified many other promising, previously understudied, regulators of MM cell biology which are not essential for baseline survival/growth of MM cells but can induce growth when further activated. Upregulation of one or more MM driver genes is recurrently observed in MM clinical relapses, which raises intriguing hypotheses about how potential cooperative interactions between these genes contribute to the risk of MM relapses. More broadly, the results indicate that CRISPRa studies can provide novel insights into the biology of MM cells and identify previously understudied genes with therapeutic implications towards suppressing the transition of MM cells to states of advanced biological aggressiveness.
