Building a novel TRUCK by harnessing the endogenous IFN-gamma promoter for cytokine expression

Abstract

Despite the remarkable success of chimeric antigen receptor (CAR) T therapy in hematological malignancies, its efficacy in solid tumors remains limited. Cytokine-engineered CAR T cells offer a promising avenue, yet their clinical translation is hindered by the risks associated with constitutive cytokine expression. In this proof-of-concept study, we leverage the endogenous interferon (IFN)-γ promoter for transgenic interleukin (IL)-15 expression. We demonstrate that IFN-γ expression is tightly regulated by T cell receptor signaling. By introducing an internal ribosome entry site IL15 into the 3' UTR of the IFN-γ gene via homology directed repair-mediated knock-in, we confirm that IL-15 expression can co-express with IFN-γ in an antigen stimulation-dependent manner. Importantly, the insertion of transgenes does not compromise endogenous IFN-γ expression. In vitro and in vivo data demonstrate that IL-15 driven by the IFN-γ promoter dramatically improves CAR T cells' antitumor activity, suggesting the effectiveness of IL-15 expression. Last, as a part of our efforts toward clinical translation, we have developed an innovative two-gene knock-in approach. This approach enables the simultaneous integration of CAR and IL-15 genes into TRAC and IFN-γ gene loci using a single AAV vector. CAR T cells engineered to express IL-15 using this approach demonstrate enhanced antitumor efficacy. Overall, our study underscores the feasibility of utilizing endogenous promoters for transgenic cytokines expression in CAR T cells.

Keywords: CAR; IFN-γ; IL-15; antitumor activity; chimeric antigen receptor; endogenous promoter; interferon-γ; interleukin-15.

Graphical abstract

Figure 1

Determination of IFN-γ expression kinetics and evaluation of 3′ UTR targeting gRNAs (A) Schematic of the experiment designed to investigate the kinetics of IFN-γ expression upon antigen stimulation. (B) Representative result of IFN-γ intracellular staining (left) and graph based on three independent experiments (right). (C) Schematic of the experiment designed to investigate the shutting down kinetics of IFN-γ expression upon antigen removal. (D) Representative result of IFN-γ intracellular staining (left) and graph based on three independent experiments (right). (E) Four gRNAs were designed targeting IFN-γ 3′ UTR (up) and insertion/deletion (indel) efficiency were determined with online sequence trace decomposition tool (TIDE). (F) Investigate the impact of 3′ UTR indel on IFN-γ expression. T cells were edited with four gRNAs separately, and 48 h later, the IFN-γ production capacity was measured with intracellular staining (N = 3). FC, flow cytometry; MFI, mean fluorescence intensity; NS, not significant.

Figure 2

Hijacking of IFN-γ promoter for GFP expression through the integration of GFP into IFN-γ 3′ UTR (A) Schematic illustration of the GFP knock-in strategy targeting the IFN-γ 3′ UTR via HDR. (B) Knock-in experiment was conducted using PCR amplicon as the donor template. 3 days later, cells were restimulated with CD3/28 beads, and GFP expression was determined after 24 h stimulation. (C) The HDR template for sgRNA4 was delivered by AAV6, and GFP expression was determined before and after stimulation. (D) Co-expression of GFP and IFN-γ in GFP engineered cells were determined by intracellular staining. (E) Cells were stimulated with CD3/28 beads for 24 h, and the levels of IFN-γ in the supernatant were determined using enzyme-linked immunosorbent assay (ELISA) (N = 3). (F) GFP-engineered T cells were stimulated with CD3/28 beads, and GFP expression at different time points were determined by flow cytometry. CTL, control.

Figure 3

IL-15 is co-expressed with IFN-γ in antigen-stimulation-dependent manner (A) Schematic illustration of the IL15GFP knock-in strategy targeting the IFN-γ 3′ UTR via HDR. (B) IL15GFP engineered T cells were restimulated with CD3/28 beads for 24 h, and knock-in efficiency was determined by measuring GFP expression. (C) IL15GFP engineered T cells were restimulated with CD3/28 beads for different durations, and IL-15 levels in the supernatant were determined by ELISA. (D) Schematic diagram of experiment design (left), and IL-15 concentrations in the supernatants collected at different time points. (E) Co-expression of GFP and IFN-γ in IL15GFP engineered cells was determined by intracellular staining. (F) Cells were stimulated with CD3/28 beads for 24 h, and the levels of IFN-γ in the supernatant were determined using ELISA (N = 3). (G and H) GFP and IL15GFP engineered T cells were cultured for an additional 9 days after electroporation, and cell proliferation (G) and phenotype were determined (H). FMO, fluorescence minus one; ND, not detected.

Figure 4

Manufacturing of anti-claudin18.2 CAR T cells engineered to express IL-15 (A) Schematic diagram of anti-claudin18.2 CAR construct. (B) T cells were activated for 24 h and subsequently transduced with anti-claudin18.2 CAR virus at MOIs of 1, 2, 3, and 4. CAR expression was determined on day 6 post transduction. (C and D) T cells were co-cultured with HGC-27-035 at effector-to-target (E:T) ratios of 3:1 and 1:1 for 48 h, and cytotoxicity (C) and IFN-γ expression (D) of CAR T cells were measured using xCELLigence RTCA and ELISA. (E) Manufacturing process of IL-15-secreting CAR T cells. (F) Viral transduction and T cell activation were conducted simultaneously on day 0, and the transduction efficiency of the CAR from three different healthy donors was determined by flow cytometry. (G) The knock-in efficiency of GFP and IL15GFP was determined by measuring GFP expression after 24 h stimulation with CD3/28 beads. (H) The percentage and intensity of GFP in GFP and IL15GFP engineered T cells were determined at different time points during the manufacturing process using flow cytometry. (I) The frequency of GFP-positive cells was determined before and after fluorescence-activated cell sorting enrichment using flow cytometry. (J) T cells were co-cultured with the HGC-27-035 tumor cells for 24 h, and the levels of IL-15 in the supernatant were determined by ELISA. NT, non-transduced T cells.

Figure 5

IL-15 driven by IFN-γ promoter enhances CAR T cell antitumor activity in vitro and in vivo (A) Schematic diagram of the cytotoxicity assay for CAR T cells under repeated antigen stimulation. (B) Assessment of tumor-killing activity of CAR T cells by co-culture CAR T cells with HGC-27-035 tumor cells for 48 h. (C and D) Assessment of tumor-killing activity of T cells using repeated antigen stimulation assay. Tumor growth from the last round of co-culture (C) and cytokines in the supernatants (D) were determined using xCELLigence RTCA and ELISA, respectively. (E) T cell subsets after the last round co-culture was determined based on the expression of CD45RA and CD62L. (F) Schematic diagram of HGC-27-035 xenograft mouse model. (G) Tumor growth based on tumor volume (mm³ ± SD), with a sample size of n = 6 mice per group. (H) The body weight of each treatment groups. (I) Kaplan-Meier survival curves for each group, with 6 mice per group, and survival comparison conducted using the Gehan-Breslow-Wilcoxon test. DPBS, Dulbecco’s PBS. ∗∗p < 0.01.

Figure 6

IL-15 driven by IFN-γ promoter enhances antitumor activity of CAR T cells constructed via CAR and IL-15 double knock-in (A) Schematic diagram of the DKI-IL15 AAV vector. LA, left homology arm; RA, right homology arm; P2A, porcine teschovirus-1 2A; pA, poly A tail. (B) Manufacturing process of double knock-in CAR T cells. (C) The efficiency of CAR knock-in in double knock-in cells was determined using flow cytometry. (D) The integration of IL-15 at the IFN-γ locus was confirmed using PCR. (E) Cells were stimulated with phorbol 12-myristate 13-acetate (PMA)-ionomycin, and the levels of IL-15 in the supernatant were determined at different timepoints using ELISA. (F) Nectin4-targeting CAR T cells were manufactured as (B) and later subjected to multiple-round killing assay. After the final round of killing, supernatant was collected, and IL-15 levels were determined by ELISA. (G) We inoculated 5 × 10⁶ Nectin4-expressing HGC-27 cells (designated HGC-27-180) subcutaneously. (H and I) After 7 days, once the tumor was established, 5 × 10⁶ T cells were infused, and tumor growth (H) and survival (I) were determined (N = 6). DKI, double knock-in; ∗∗p < 0.01.