Reprogramming glutamine metabolism enhances BCMA-CAR T-cell fitness and therapeutic efficacy in multiple myeloma

Abstract

Glutamine dependence of cancer cells reduces local glutamine availability, which hinders antitumor T-cell functionality and facilitates immune evasion. We thus speculated that glutamine deprivation might be limiting efficacy of chimeric antigen receptor (CAR) T-cell therapies in patients with cancer. We have seen that antigen-specific T cells are unable to proliferate or produce interferon gamma (IFN-γ) in response to antigen stimulation when glutamine concentration is limited. Using multiple myeloma (MM) as a glutamine-dependent disease model, we found that murine CAR T cells selectively targeting B-cell maturation antigen (Bcma) in MM cells were sensitive to glutamine deprivation. However, CAR T cells engineered to increase glutamine uptake by expression of the glutamine transporter Asct2 exhibited enhanced proliferation and responsiveness to antigen stimulation, increased production of IFN-γ, and heightened cytotoxic activity, even under conditions of low glutamine concentration. Mechanistically, Asct2 overexpression reprogrammed the metabolic fitness of CAR T cells by upregulating the mechanistic target of rapamycin complex 1 gene signature, modifying the solute carrier transporter repertoire, and improving both basal oxygen consumption rate and glycolytic function, thereby enhancing CAR T-cell persistence in vivo. Accordingly, expression of Asct2 increased the efficacy of Bcma-CAR T cells in syngeneic and genetically engineered mouse models of MM, which prolonged mouse survival. In patients, higher-level expression of ASCT2 by MM cells predicted poor outcome to combined immunotherapy and BCMA-CAR T-cell therapy. Our results indicate that reprogramming glutamine metabolism may enhance antitumor CAR T-cell functionality in MM. This approach may also be effective for other cancers that depend on glutamine as a key energy source and metabolic hallmark.

Graphical abstract

Figure 1.

Asct2 overexpression on T cells enhances the function of CD8⁺ and CD4⁺ T cells cultured under low and high Gln concentrations. (A) Diagrams of retroviral vectors designed to overexpress Asct2 in T cells. (B) Transduction efficiency (Thy 1.1⁺) in CD8⁺ and CD4⁺ T cells assessed by flow cytometry. (C) Asct2 expression on the transduced T cells’ membrane measured by flow cytometry. (D) ³H-labeled Gln uptake by transduced CD4⁺ and CD8⁺ T cells. (E-F) Proliferation and IFN-γ secretion of Ctrl and Asct2-overexpressing CD4⁺ (E) and CD8⁺ (F) T cells activated with anti-CD3/CD28 beads under different Gln concentrations (0-2 mM). (G-J) Proliferation, IFN-γ secretion, and specific B16OVA lysis of Ctrl and CD8⁺ Asct2–OT-I T cells (G-H), and CD4⁺Asct2–OT-II T cells (I-J) under different Gln concentrations. (K-L) Seahorse metabolic assay was used to measure the OCR and the ECAR of Ctrl and Asct2-overexpressing CD8⁺ OT-I T cells (K), and CD4⁺ OT-II T cells (L). Data are representative of 2 to 3 independently repeated experiments. Data represent mean ± standard error of the mean (SEM) and were analyzed using 2-way analysis of variance (ANOVA) with Bonferroni multiple comparisons test (∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .01; ∗P < .05). AntiA/Rote, antimycin/rotenone; APC, allophycocyanin; cpm, counts per minute; Ctrl, control; LTR, long termi; MFI, mean fluoresecence intensity; Oligom, oligomycin; unstim, unstimulated.

Figure 2.

Asct2 overexpression on OT-I T cells enhances the efficacy of ACT immunotherapy. (A) Gln concentration measured in the TIF and sera of B16OVA-challenged mice. (B) Experimental design to evaluate antitumor activity of ACT. (C-E) Tumor growth measured over time (C-D) and OS of mice (E) after treatment with Asct2-CD8⁺ OT-I or with Ctrl CD8⁺ OT-I T cells (n = 9-11 mice per group). (F) Tumor weight measured at day 14 after treatment. (G) Number of Ag-specific T cells infiltrating the tumor measured by flow cytometry using SIINFEKL tetramers. (H) Percentage of expression of LAG3 and PD-1 measured in CD45.1⁺ tumor-infiltrating CD8⁺ OT-I and Asct2–OT-I T cells measured by flow cytometry. (I) Number of IFN-γ–producing Ag-specific T cells measured in the spleen of untreated mice or treated with Ctrl CD8⁺ OT-I or Asct2–OT-I T cells in response to SIINFEKL peptide. Data are representative of 2 independent experiments. Data represent mean ± SEM and were analyzed using Student t test, 2-way ANOVA, and 1-way ANOVA with Bonferroni multiple comparisons test (∗∗∗P < .001; ∗∗P < .01; ∗P < .05). MOS, median overall survival; PD-1, programmed cell death protein 1; SC, subcutaneous; TIF, tumor interstitial fluid; unstim, unstimulated.

Figure 3.

Asct2 overexpression enhances Bcma-CAR T-cell activity. (A) Schematic representation of second-generation Bcma-CAR and PSMA-CAR constructs (upper). Schematic representation of second-generation of Asct2 Bcma-CAR construct (lower). (B) Proliferation and IFN-γ production of Bcma and PSMA-CAR T cells in response to Bcma-, Taci-, or ovalbumin-coated plates (as an irrelevant protein). (C) Bcma expression in MM cell lines 5080 and 9275 and Ctrl lymphoma cell line LY5026. (D) IFN-γ production and percentage of lysis induced by Bcma and PSMA-CAR T cells in response to these MM cell lines (5080 and 9275) or to B16OVA melanoma cells. (E) CAR expression in T lymphocytes transduced with retrovirus expressing Bcma or Asct2 BCMA-CAR constructs. Representative flow cytometry plots are shown together with histograms representing the mean fluorescence intensity in RV-transduced CD4⁺ and CD8⁺ T cells. (F) Asct2 expression measured by flow cytometry. (G) ³H-labeled Gln uptake by CD4⁺ and CD8⁺Bcma-CAR T cells and Asct2 Bcma-CAR T cells (purified by flow cytometric cell sorting). (H) IFN-γ secretion of CD4⁺ or CD8⁺Bcma-CAR T cells and Asct2 Bcma-CAR T cells in response to Bcma-coated plates under different Gln concentrations. (I) Specific lysis of 5080 MM cells by Bcma-CAR T cells and Asct2 Bcma-CAR T cells under different Gln concentrations. (J-K) Seahorse metabolic assay was used to measure the OCR and the ECAR of Ctrl and Asct2-overexpressing CD4⁺ (J) and CD8⁺ (K) Bcma-CAR T cells. Data are representative of 3 independent experiments. Data represent mean ± SEM, and were analyzed using 2-way ANOVA and 1-way ANOVA with Bonferroni multiple comparisons test (∗∗∗∗P < .0001; ∗∗∗P < .001; ∗∗P < .01; ∗P < .05). Ab, antibody; AntiA/Rote, antimycin A/rotenone; APC, allophycocyanin; cpm, counts per minute; Ctrl, control; FSC-A, forward scatter area; H&L, heavy and light; IgG, immunoglobulin G; LTR, long terminal repeat; MFI, mean fluoresecence intensity; Ova, ovalbumin; SC, subcutaneous; stim, stimulated; Taci, transmembrane activator and CAML interactor; unstim, unstimulated.

Figure 4.

Transcriptomic analysis of Asct2 Bcma-CAR T cells and Bcma-CAR T cells before and after BCMA antigen stimulation. (A) Expression of Slc1a5 (Asct2). (B) Number of differentially expressed genes between Asct2 Bcma-CAR T cells and Bcma-CAR T cells before and after BCMA antigen stimulation. (C) PC analysis of gene expression profiles. (D) Gene signatures upregulated in Asct2 Bcma-CAR T cells compared with Bcma-CAR T cells, before and after BCMA antigen stimulation. (E) Heat map showing differential expression of SLC family members in Asct2 Bcma-CAR T cells compared with Bcma-CAR T cells after antigen stimulation. Data represent mean ± SEM, and were analyzed using Student t test (∗∗P < .01). Exp, expression; NES, normalized enriched score; PC, principal component; unstim, unstimulated.

Figure 5.

The antitumor efficacy of ACT immunotherapy with Bcma-CAR T cells was improved by Asct2 overexpression. (A) Gln concentration measured in the BM of healthy mice or mice challenged with 5080 MM cell line (25 days after tumor challenge). (B) Experimental design to evaluate antitumor activity of CAR T cells in a MM murine model based on the IV injection of 5080 MM cells and percent of survival of 5080 MM-challenged mice treated with 1 × 10⁶ CAR T cells (1:1 ratio of CD4⁺ and CD8⁺ CAR T cells; number of mice per group n = 10). (C) Total number of tumor cells (B220⁺GFP⁺) found in the BM of mice treated with Asct2 Bcma-CAR T cells and Bcma-CAR T cells. (D-F) Total numbers of CAR T cells measured in the BM (D, F), and the spleen (E) of mice treated with Asct2 Bcma-CAR T cells or Bcma-CAR T cells at day 14 (D-E), and at day 21 (F) after CAR T-cell transfer. (G-J) Asct2 Bcma-CAR T cells exert antitumor activity in a genetic model of MM. (G) Gln concentration in the BM aspirates and the sera of 170-day-old MICγt1 mice compared with the Gln levels in BM or serum of healthy mice. (H) Experimental design for testing CAR T-cell immunotherapy against the MICγt1 MM genetic model. Fc γ-globulin fraction levels and survival curves after CAR T-cell treatment (n = 7 mice per group; survival data for untreated animals corresponds to a cohort of 50 mice). (I) Number of tumor cells in both the spleen and the BM of mice treated with Bcma-CAR T cells and Asct2 Bcma-CAR T cells compared with untreated mice. (J) CAR T cells in the spleen and in the BM of MICγt1 mice treated with Bcma or Asct2 Bcma-CAR T cells analyzed in the CD4⁺ and CD8⁺ compartments. Data are representative of 2 to 3 independent experiments. Data represent mean ± SEM and were analyzed using Student t test, 2-way ANOVA, and 1-way ANOVA with Bonferroni multiple comparisons test. Statistical analysis for survival curves is a log-rank test (∗∗P < .01; ∗P < .05; ∗∗∗P < .001). ACT, adoptive cell therapy; FC, fold change; rIL, recombinant interleukin.

Figure 6.

ASCT2 expression is associated with a poor prognosis in MM. Correlation between ASCT2 expression and clinical outcomes in MM in PFS (A) and OS (B). (C) ASCT2 messenger RNA expression levels in NBs, CBs, CCs, MEMs, TPCs, and BMPCs, as well as in MM aspirates and MM cell lines. (D) Proliferation of human MM cell lines in culture medium with decreasing concentrations of Gln. Percentage of proliferation with respect to that found in culture medium supplemented with Gln at 2 mM. (E) ASCT2 expression in CD4⁺ and CD8⁺ T cells in patients who experienced a CR compared with those who did not respond to therapy. Data represent mean ± SEM and were analyzed using 1-way ANOVA with Bonferroni multiple comparisons test (∗P < .05; ∗∗∗P < .001). BMPC, BM plasma cell; CB, centroblast; CCs, centrocytes; CR, complete response; MEM, memory B cells; NB, naïve B cells; NS, no significance; PD, progressive disease; TPC, tonsillar plasma cell; TPM, transcripts per million.

Figure 7.

ASCT2 overexpression enhances antihuman BCMA-CAR T-cell activity. (A) Schematic diagrams of lentiviral vectors used to overexpress ASCT2 in human BCMA-CAR T cells. (B) Expression levels of the CAR construct, measured by BFP fluorescence. (C) Human ASCT2 expression detected on CAR T cells by flow cytometry. (D) Proliferation of CAR T cells in response to BCMA-coated plate stimulation at different Gln concentrations, measured by [³H]-thymidine incorporation. (E) Phenotypic analysis of CD4⁺ and CD8⁺ CAR T cells at the time of manufacture. (F) Asct2 silencing in H929 MM cells using 3 different single-guide RNAs (TadCBEd). (G) Measurement of [³H]-Gln uptake in WT and ASCT2-silenced H929 MM cells. (H) Cytotoxic activity of BCMA-CAR T cells against WT and ASCT2-silenced H929 MM cells. Data are representative of 2 independent experiments. Data represent mean ± SEM and were analyzed using 2-way ANOVA with Bonferroni multiple comparisons test (∗∗∗P < .001). BFP, blue fluorescent protein; cpm, counts per million; KO, knockout; neg, negative; SSC-A, side scatter area; T_CM, T central memory; T_EM, T effector memory; T_EMRA, T effector memory recently reactivated; T_SCM, T stem cell memory; WT, wild-type.