GOT1 regulates CD8+ effector and memory T cell generation

Abstract

T cell activation, proliferation, function, and differentiation are tightly linked to proper metabolic reprogramming and regulation. By using [U-¹³C]glucose tracing, we reveal a critical role for GOT1 in promoting CD8⁺ T cell effector differentiation and function. Mechanistically, GOT1 enhances proliferation by maintaining intracellular redox balance and serine-mediated purine nucleotide biosynthesis. Further, GOT1 promotes the glycolytic programming and cytotoxic function of cytotoxic T lymphocytes via posttranslational regulation of HIF protein, potentially by regulating the levels of α-ketoglutarate. Conversely, genetic deletion of GOT1 promotes the generation of memory CD8⁺ T cells.

Keywords: CP: Metabolism; GOT1; HIF; NADH/NAD; effector and memory CD8(+) T cell; glucose; glutamate; serine.

Figure 1.. Increased metabolic flux from α-ketoglutarate to glutamate upon CD8⁺ T cell activation is mediated by increased expression of GOT1

(A) Naive or 24-h plate-bound anti-CD3 and soluble anti-CD28 activated OT-I CD8⁺ T cells were pulsed with [U-¹³C]glucose for 4–6 h before collecting for mass spectrometry analysis. Percent contribution of carbon flux from [U-¹³C]glucose to glutamate is shown. (B) Graph describing the intersection of glutamine and glucose metabolism. (C) OT-I CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 in the presence of vehicle (Veh), AOA (250 μM), or EGCG (500 μM) for 24 h. Cells were then pulsed with [U-¹³C]glucose for 4–6 h before collecting for mass spectrometry analysis. Percent contribution of carbon flux from [U-¹³C]glucose to glutamate is shown. *comparison between Veh and AOA, ⁺comparison between Veh and EGCG. (D and E) 24-h plate-bound anti-CD3 and soluble anti-CD28 activated WT and GOT1^−/− OT-I CD8⁺ T cells were pulsed with [U-¹³C]glucose (D) or [U-¹³C]glutamine (E) for 4–6 h before collecting for mass spectrometry analysis. Abundance of (left) and percent contribution (right) of carbon flux from [U-¹³C]-glucose (D) or [U-¹³C]glutamine (E) to glutamate are shown. (F) Isolated OT-I or P14 CD8⁺ T cells were activated using plate-bound anti-CD3 and soluble anti-CD28. Immunoblot measurements of GOT1 and GLUD1 at indicated time points are shown. Actin served as the loading control. (G) In vitro effector (IL-2) and memory (IL-7/15) OT-I or P14 CD8⁺ T cells were generated as previously described. Immunoblot measurements of GOT1 and GLUD1 (left) and statistical analysis (right) are shown. Actin served as the loading control. (H) Naive (T_n) Thy1.1⁺ P14 CD8⁺ T cells were adoptively transferred into WT recipients and infected with LCMV Armstrong. Thy1.1⁺ CD8⁺ T cells were FACS sorted on day 7 as effector T cells (T_eff) and on day 60 as memory T cells (T_mem). Immunoblot measurements of GOT1 and GLUD1 (left) and statistical analysis (right) are shown. Actin served as the loading control. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, not significant. Two-way ANOVA with Sidak’s multiple comparisons test (A, C, D, and E) and paired t test (G and H). Data are representative of at least three independent experiments. Data are represented as mean ± SD. Also see Figure S1.

Figure 2.. GOT1 supports serine biosynthesis and promotes CD8⁺ T cell proliferation under serine-restricted conditions

(A) Abundance of non-essential amino acids (NEAAs) from targeted metabolomics analysis of 24-h activated WT and GOT1^−/− OT-I CD8⁺ T cells. (B) 24-h plate-bound anti-CD3 and soluble anti-CD28 activated WT or GOT1^−/− OT-I CD8⁺ T cells were pulsed with [U-¹³C]glucose for 4–6 h before collecting for mass spectrometry analysis. Percent contribution of carbon flux from [U-¹³C]glucose to serine is shown. (C) Flow cytometry analysis of proliferation of WT or GOT1^−/− P14 CD8⁺ T cells cultured under 0, 0.1, and 0.3 mM serine conditions at indicated time points. (D) Flow cytometry analysis of proliferation of separately cultured or co-cultured WT and GOT1^−/− P14 CD8⁺ T cells in serine-free media at indicated time points. (E) Targeted metabolomics analysis of blank media and media after 10 h of WT OT-I CD8⁺ T cell culture. Volcano plot (left) and abundance of pyruvate (right) are shown. (F) WT and GOT1^−/− P14 CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 without or with 1 mM pyruvate (Pyr) in serine-free media. Flow cytometry analysis of proliferation at indicated time points is shown. (G) Abundance of NADH (left) and NAD⁺ (middle) from targeted metabolomics analysis of 24-h activated WT and GOT1^−/− OT-I CD8⁺ T cells without or with 0.3 mM serine. NAD⁺/NADH ratios are shown (right). (H) WT and GOT1^−/− P14 CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 without or with 1 mM α-ketobutyrate (αKB) in serine-free media. Flow cytometry analysis of proliferation at indicated time points is shown. (I) WT and GOT1^−/− P14 CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 without or with 1 mM sodium formate, 200 μM hypoxanthine, or 50 μM adenine in serine-free media. Flow cytometry analysis of proliferation is shown. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, not significant. Two-way ANOVA with Sidak’s multiple comparisons test (A, B, and G). Data are representative of three independent experiments. Data are represented as mean ± SD. Also see Figure S2.

Figure 3.. GOT1 posttranslationally regulates HIF1α expression and the cytotoxic function of CTLs

(A) Abundance of α-ketoglutarate from targeted metabolomics analysis of 24-h activated WT and GOT1^−/− OT-I CD8⁺ T cells. (B) WT and GOT1^−/− OT-I CD8⁺ T cells were activated using plate-bound anti-CD3 and soluble anti-CD28 for 24 h. Immunoblot measurements of HIF1α, c-MYC, and mTORC1 activity indicated by p-S6K and p-S6 are shown. Total S6K, total S6, and actin served as the loading control. (C) Real-time PCR analysis of Hif1α mRNA levels in 24-h plate-bound anti-CD3 and soluble anti-CD28 activated WT and GOT1^−/− OT-I CD8⁺ T cells. (D) WT and GOT1^−/− OT-I CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 without (Veh) or with 1 mM dimethyl-succinate (dSuc) for 24 h. Immunoblot measurements of HIF1α are shown. Tubulin served as the loading control. (E) WT and GOT1^−/− OT-I CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 under normoxia for 21 h. Cells were then incubated under normoxia or hypoxia for an additional 3 h. Immunoblot measurements of HIF1α are shown. Tubulin served as the loading control. (F) Immunoblot measurements of HIF1α and hydroxy-HIF1α at proline 564 site in 24-h activated WT and GOT1^−/− OT-I CD8⁺ T cells. Tubulin served as the loading control. (G and H) WT and GOT1^−/− OT-I CD8⁺ T cells were activated with plate-bound anti-CD3 and soluble anti-CD28 for 24 h followed by 2 h of vehicle (Veh) or MG132 (10 μM) treatment. (G) Immunoblot measurements of HIF1α and c-MYC are shown. Tubulin served as the loading control. (H) Compared with vehicle control, fold increase after MG132 treatment was analyzed. (I) OT-I or P14 CTLs were generated as previously described. Immunoblot measurements of HIF1α in WT and GOT1^−/− CTLs are shown. Tubulin served as the loading control. (J) Flow cytometry comparison of perforin production between WT and GOT1^−/− OT-I CTLs alone or co-cultured with EL4-OVA for 6 h. Flow plots (left panel) and statistical analysis of geometric mean fluorescence intensity (gMFI, right) are shown. (K) Percentages of killed EL4-OVA by WT or GOT1^−/− CTLs at indicated effector:tumor ratios. (L) WT and GOT1^−/− OT-I CTLs were adoptively transferred into B16-OVA tumor bearing mice on day 11 after inoculation. Measurements of tumor volume are shown. (M) Naive WT CD8⁺ T cells were isolated and electroporated with control sgRNA (Ctrl) or sgRNAs targeting Vhl (g1, g2, g3) complexed with Cas9 endonuclease. Cells were recovered and activated with plate-bound anti-CD3 and soluble anti-CD28 and expanded in IL-2 until day 6. Immunoblot measurements of VHL and HIF1α are shown. Tubulin served as the loading control. (N and O) Naive WT and GOT1^−/− OT-I CD8⁺ T cells were isolated and electroporated with control sgRNA (Ctrl) or sgRNAs targeting Vhl (g2, g3) complexed with Cas9 endonuclease. Cells were activated and expanded in IL-2 until day 6, and co-incubated without or with EL4-OVA at 1:1 ratio to assess perforin production (N), or at effector:tumor 4:1 ratio to assess the percentages of tumor cells killed (O). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, not significant. Unpaired t test (A and C), paired t test (H), two-way ANOVA with Sidak’s multiple comparisons test (J, K, and L), two-way ANOVA with Tukey’s multiple comparisons test (N), one-way ANOVA with Tukey’s multiple comparisons test (O). Data are representative of at least three independent experiments. Data are represented as mean ± SD. Also see Figure S3.

Figure 4.. Deletion of GOT1 promotes memory CD8⁺ T cell generation

(A) Flow cytometry comparison of CD62L expression (left) and real-time PCR analysis of Sell mRNA (right) between WT and GOT1^−/− CTLs. (B) WT and GOT1^−/− CTLs with different congenic markers were mixed at 1:1 ratio and co-adoptively transferred into WT recipients. Flow cytometry analysis of percentages of WT and GOT1^−/− CTLs before adoptive transfer (left) or from the spleen and lymph nodes 48-h after adoptive transfer (right). (C–G) GOT1^−/− CD8⁺ T cells show enhanced memory generation in an adoptive transfer model. (C) Experimental scheme. Naive WT and GOT1^−/− OT-I CD8⁺ T cells were adoptively transferred into WT recipients separately and infected with LMOVA. (D) Percentages of antigen specific Thy1.1⁺ cells of CD8⁺ T cells from spleen on day 7. (E) Percentages of KLRG1⁺CD127⁻ and KLRG1⁻CD127⁺ cells of Thy1.1⁺ T cells from spleen on day 7. (F) Percentages of IL-2⁺ cells of Thy1.1⁺ T cells post OVAI peptide stimulation ex vivo from spleen on day 7. (G) Percentages of antigen specific Thy1.1⁺ cells of CD8⁺ T cells from blood on day 70 (left) and from spleen 5 days post Vaccinia-OVA (VacOVA) rechallenge (right). (H–J) GOT1^−/− CD8⁺ T cells show enhanced memory generation in a co-adoptive transfer model. (H) Experimental scheme. WT and GOT1^−/− OT-I CD8⁺ T cells with different congenic markers were mixed at 1:1 ratio, co-transferred into WT recipients, and infected with LMOVA. (I) Flow cytometry analysis of percentages of WT and GOT1^−/− OT-I CD8⁺ T cells before transfer (left) and from blood at indicated time points after adoptive transfer (right). (J) Percentages of WT and GOT1^−/− OT-I CD8⁺ T cells from spleen and lymph nodes on day 60. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; ns, not significant. Unpaired t test (A), paired t test (B, J), Mann-Whitney t test (D, E, F, and G), two-way ANOVA with Sidak’s multiple comparisons test (I). Data are representative of three independent experiments (A–B, D–G, and I–J). Data are represented as mean ± SD.