Prime editing of the endogenous IL-2Rβ locus is a feasible strategy for generating oIL-2–responsive CAR T cells. (A) Experimental validation of PE3-mediated functional oIL-2Rβ editing with the optimized pegRNA. To achieve ideal editing efficiency, we evaluated 5 pegRNAs with different lengths of RTT and PBS (supplemental Table 1) using SeAx, an IL-2–dependent cutaneous T-cell lymphoma cell line. The impact of each pegRNA on editing efficiency was assayed using next generation sequencing (NGS). (B) Human oIL-2 promotes the specific expansion of the human primary T cells expressing the edited oIL-2Rβ. The nonedited and edited oIL-2Rβ human primary T cells were cultured in the media with 20nM of human WT IL-2 or oIL-2. To monitor cell expansion, cells were counted every other day over an 11-day period. Growth trends were visualized by comparing mock editing (left) and orthogonal editing (right) conditions, as shown in the corresponding growth curves. (C) Western blot analysis of human oIL-2–induced signal pathways through the edited oIL-2Rβ. The edited and nonedited T cells were stimulated with oIL-2 or WT IL-2 for 20 minutes followed by western blotting with indicated antibodies. Actin served as the loading control. (D) Human oIL-2 selectively expands the T cells toward oIL-2Rβ⁺ cells. The activated human primary T cells were edited with a 25% oIL-2Rβ editing efficiency. After 7 days of culture, the editing efficiency increased to >50% in the presence of oIL-2 and halved in the presence of WT IL-2. (E) A diagram of the in vivo evaluation of prime-editing edited oIL-2Rβ⁺ CAR T cells (PE ortho/LV-CAR-T) antileukemic activity with human oIL-2 treatment. NSG mice were engrafted with 0.8 × 10⁶ to 1 × 10⁶ click green beetle (CBG)-labeled CD19⁺ Nalm6 leukemic cells on day 0 (Nalm6-LUC). A total of 1 × 10⁶ PE orthoIL-2Rβ–CAR-T cells were injected on day 5 following BLI on day 4. phosphate-buffered saline or 40 000 IU of oIL-2 was administered via intraperitoneal route once a day for 14 days starting on day 5. Tumor burden was assessed via BLI twice per week, and CAR T-cell expansion was examined weekly. (F) Individual BLI intensity of Nalm6-LUC was determined for each mouse. Blue indicates PE ortho/LV-CAR-T; red indicates PE ortho/LV-CAR-T + 40 000 IU orthohIL-2. (G) Body weight of individual mice during the in vivo experiment from panel F. Mouse body weight was normalized to the body weight on day 0 for each mouse. Blue indicates PE ortho/LV-CAR-T; red indicates PE ortho/LV-CAR-T + 40 000 IU orthohIL-2. (H) Off-target evaluation for prime editing using NGS. Three-step polymerase chain reactions were performed with the genomic DNA from the 3 edited donors using the primers (supplemental Figure 5). Paired-end 2 × 250-bp reads were sequenced using MiSeq (Illumina) at 10.4pM along with 20.5% PhiX (University of Pennsylvania). The bar graph shows the indel percent of on- and off-target edits analyzed by next-generation sequencing (NGS). NSG, NOD scid gamma; orthohIL-2, orthogonal human interleukin-2; p-ERK, phosphorylated extracellular signal-related kinase; p-STAT5, phosphorylated signal transducer and activator of transcription 5.