Fig. 2

CAR-TIM3 NK-92 cells effectively reduce AML tumor burden and biodistribution in vivo. A Timeline for inoculation of 1.5 × 10⁶ Luc-labeld TIM3-overexpressed U937 cells, two doses of 1.5 × 10⁶ NK cells, and tumor imaging in NOD SCID gamma (NSG) mice. B Representative in vivo bioluminescence imaging of mice taken at days 7, 10, and 15 after tumor inoculation. C Quantification of whole-body signals in different mice (n = 6 per group) after receiving WT or CAR-TIM3 NK-92 cells normalized to their initial signals before NK cell treatment at day 7. D Relative AML tumor burden was calculated by normalization of signals (C) in CAR-TIM3 NK-92 group to those of WT NK-92 group obtained on the same day. *p < 0.05, **p < 0.01 vs WT NK-92 group; Mann–Whitney U-test. E (left) Representative ex vivo bioluminescence imaging of isolated organs obtained from NSG mice at the end of experiment. (right) Quantification of signals indicated a significant reduction of tumor burden in liver of mice receiving CAR-TIM3 NK-92 cells. **p < 0.01 vs WT NK-92 group; Mann–Whitney U-test. F Flow cytometric analysis of AML cell engraftment in isolated bone marrow based on the detection of human TIM3⁺CD45⁺ cells. Representative plots (left) and quantification (right) are shown. **p < 0.01 vs WT NK-92 group; Mann–Whitney U-test. G, H CAR-TIM3 NK cells improved NK cytotoxicity in part by reducing NK cell exhaustion. G Flow cytometric analysis of surface TIM3 and PD-1 in WT and CAR-TIM3 NK-92 cells transduced with TIM3 transgene (O/E TIM3) or vector control (mock) to first identify the functional role of TIM3 in NK exhaustion. H (left) Percentages of total cell death of PKH67-labeled Kasumi-1 cells as evaluated by annexin-V/7-AAD assay after exposure to indicated NK cells at E:T ratio of 1:1 for 4 h. (right) Percentages of TIM3 and PD-1 coexpression in indicated NK-92 cells upon exposure to Kasumi-1 cells. ***p < 0.001 vs indicated NK cells; Mann–Whitney U-test