Fig. 2

The transcriptomic mechanism that 28z CAR-NK outperform BBz CAR-NK cells. (A) Schematic diagram of BBz and 28z CAR interactome analysis by Co-IP coupled with LC-MS/MS assay. (B) Venn diagram of the interacting proteins of BBz and 28z CAR. (C) Schematic diagram of RNA sequencing analysis of BBz and 28z CAR-NK. (D) KEGG pathway enrichment bubble chart of 28z CAR-NK92MI versus BBz CAR-NK92MI. (E) Differential genes Venn diagram of 28z versus BBz in CAR-NK92MI and CAR-YTS cells. (F-I) Effect of MAP3K8 knockdown (F, G) and inhibition (H, I) on cytotoxicity of 28z CAR-NK92MI to Huh7^{EGFR + CD133+} cells. For MAP3K8 inhibition, 28z CAR-NK92MI be pretreated with 2.5 µM MAP3K8 inhibitor (Coti-2) for 1 h, then cocultured with target cells for 6 h in the presence of 2.5 µM Coti-2. (J and K) Effect of MAP3K8 overexpression on cytotoxicity of BBz CAR-NK92MI to Huh7^{EGFR + CD133+} cells. (L) Untransduced NK92MI, BBz CAR-NK92MI, and MAP3K8-overexpressed BBz CAR-NK92MI cells were cocultured with Huh7^{EGFR + CD133+} cells at an E: T ratio of 2:1 for 24 h, then the secreted IFN-γ, TNF-α, Granzyme B, and Granulysin were measured by CBA. (M) Diagram of experimental design of Huh7 xenograft model. M-NSG mice were subcutaneously injected with 2 × 10⁶ Huh7 cells with Matrigel (day 0). Once the tumor volume reached nearly 100 mm³ (day 10), mice were randomly grouped and injected (i.v.) with 1 × 10⁷ CAR-NK92MI cells. 20,000 U/mouse recombinant human IL-2 was injected (i.v.) one day after the CAR-NK92MI infusion. n = 6 mice per group. Body weight (N) and tumor volume (O) were monitored daily. Tumors (P) were dissected from the mice at the end point and the tumor weights (Q) were recorded. All results are presented as mean ± SD. The differences were analyzed by one-way or two-way ANOVA analysis. *p < 0.05; **p < 0.01; ***p < 0.001; n.s., not significant