Figure 5.
Genome-wide CRISPR screens identify essential genes for CAR T-cell cytotoxicity and SMAC mimetic function. (A) Schematic of the CRISPR-Cas9 screening approach. (B) Positively selected hits of NALM6 CRISPR screen with CAR T-cell coculture compared with NALM6 only. Genes are ranked by negative log10 of the MAGeCK RRA score for positive selection. (C) Positively selected hits of NALM6 CRISPR screen with CAR T cells plus SMAC mimetic compared with NALM6 only. Genes are ranked by negative log10 of MAGeCK RRA score for positive selection. (D) Validation of the effects of FADD and TNFRSF10B disruption in the B-ALL cell lines NALM6 and RCHACV and the DLBCL cell line SUDHL4 exposed to CAR T cells alone. Bar plots show viability of target cells expressing FADD, TNFRSF10B, or control sgRNAs exposed to CAR T cells for 24 hours (1:4 effector/target ratio) treated with DMSO. (E) Validation of the effects of FADD, TNFRSF10B, RIPK1, and TNFRSF1A disruption in NALM6 and RCHACV exposed to CAR T cells and SMAC mimetic. Bar plots show viability of target cells expressing FADD, TNFRSF10B, RIPK1, TNFRSF1A, or control sgRNAs exposed to CAR T cells for 24 hours (1:4 effector/target ratio) treated with birinapant. In panels D and E, the data are normalized to DMSO-treated target cells without T cells (indicated as dashed line). Bar heights represent average of 6 technical replicates for each sgRNA, dots represent technical replicates, and error bars show standard deviation. P values are calculated using Welch’s t test between the indicated condition and single-guide control. (F) Schematic of proposed mechanism of CAR T-cell cytotoxicity and SMAC mimetic–mediated sensitization to CAR T-cell death receptor–mediated cytotoxicity. Genes within top 50 hits in CRISPR screens are colored in dark blue. *P < .05, **P < .01, ***P < .001, ****P < .0001. ns, not significant (P ≥ .05).