AMPK promotes antitumor immunity by downregulating PD-1 in regulatory T cells via the HMGCR/p38 signaling pathway

Abstract

Background: AMP-activated protein kinase (AMPK) is a metabolic sensor that maintains energy homeostasis. AMPK functions as a tumor suppressor in different cancers; however, its role in regulating antitumor immunity, particularly the function of regulatory T cells (Tregs), is poorly defined.

Methods: AMPKα1^fl/flFoxp3^YFP-Cre, Foxp3^YFP-Cre, Rag1^-/-, and C57BL/6 J mice were used for our research. Flow cytometry and cell sorting, western blotting, immuno-precipitation, immuno-fluorescence, glycolysis assay, and qRT-PCR were used to investigate the role of AMPK in suppressing programmed cell death 1 (PD-1) expression and for mechanistic investigation.

Results: The deletion of the AMPKα1 subunit in Tregs accelerates tumor growth by increasing the expression of PD-1. Metabolically, loss of AMPK in Tregs promotes glycolysis and the expression of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), a key enzyme of the mevalonate pathway. Mechanistically, AMPK activates the p38 mitogen-activated protein kinase (MAPK) that phosphorylates glycogen synthase kinase-3β (GSK-3β), inhibiting the expression of PD-1 in Tregs.

Conclusion: Our study identified an AMPK regulatory mechanism of PD-1 expression via the HMGCR/p38 MAPK/GSK3β signaling pathway. We propose that the AMPK activator can display synergic antitumor effect in murine tumor models, supporting their potential clinical use when combined with anti-PD-1 antibody, anti-CTLA-4 antibody, or a HMGCR inhibitor.

Keywords: AICAR; AMPK; Compound C; HMGCR; PD-1; Tregs; Tumor.

Fig. 1

AMPK deficiency in Tregs promotes tumor growth. B16F10 melanoma cells were injected subcutaneously into C57BL/6 J mice and analyzed after 22 days. A Analysis of total LKB1 and AMPK in Tregs obtained from tumor-free and tumor-bearing C57BL/6 J mice by western blotting. B Detection of the mRNA expression of the prkaa1 gene by real-time PCR in Tregs isolated from tumor-free and tumor-bearing C57BL/6 J mice. The C tumor volume and D Tumor weight in WT and AMPK^fl/flFoxp3-Cre mice injected s.c. with B16F10 melanoma cells. E Flow cytometric analysis of the percentage of CD4⁺ and CD8⁺ T cells; CD4⁺Foxp3⁺ Tregs and Tregs cellularity in the tumors of WT and AMPK^fl/flFoxp3-Cre mice. F Analysis of IFN-γ-producing CD4⁺ and CD8⁺ T cells percentage in tumors from WT and AMPK^fl/flFoxp3-Cre mice by flow cytometry. G Flow cytometry analysis of GZB-producing CD8⁺ T cells in tumors from WT and AMPK^fl/flFoxp3-Cre mice. H Evaluation of in vivo cytolytic activity in the spleen and draining lymph node of WT and AMPK^fl/flFoxp3-Cre mice. The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 2

AMPK deficiency in Tregs upregulates the expression of PD-1. Bar diagram representation of flow cytometric analysis of relative mean fluorescence intensities (MFI) of PD-1, CTLA-4, Nrp1, ICOS, GITR, OX40, CD73, and CD39 in Tregs A gated from splenocytes of tumor-free mice and B gated from tumor-infiltrated lymphocytes. MFI of PD-1, Nrp1, and ICOS in splenic Tregs from WT mice after treatment with C compound C or D AICAR for 24 h. E Immunoblot analysis of PD-1 in Tregs treated with AICAR and compound C at the indicated doses and times of stimulation with anti-CD3/CD28. F T cell suppression assays with WT and AMPK-KO Tregs. G Bar diagram representation of flow cytometric analysis of IL-10-expressing WT and AMPK-KO Tregs and qRT-PCR results of the Il-10 mRNA levels. H MFI of TGF-β-expressing WT and AMPK- KO Tregs and qRT-PCR results of the Tgf-β mRNA levels. The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3

Treatment with anti-PD-1 antibody reduces tumor growth in AMPK^fl/flFoxp3-Cre mice. B16F10 melanoma tumor cells were injected subcutaneously into AMPK-WT and AMPK^fl/flFoxp3-Cre mice. Anti-mouse PD-1 was administered i.p. at 200 μg/mouse at 8, 11, 14, and 17 days after the injection of cancer cells. Tumors were analyzed at day 20 post-tumor transplantation. The A volume and B weight of the tumors from WT and AMPK^fl/flFoxp3-Cre mice treated with or without anti-PD-1 antibody. C Percentage of CD4⁺ and CD8⁺ T cells and CD4⁺Foxp3⁺ Tregs in the tumors of WT and AMPK^fl/flFoxp3-Cre mice treated with or without anti-PD-1 antibody. D Flow cytometry analysis of the percentage of GZB- and IFN-γ-producing CD8 ⁺ T cells in lymphocytes isolated from tumor tissues. E Bar diagram representation of flow cytometric analysis of the percentage of Foxp3⁺IL-10⁺ T cells. F Flow cytometric analysis and MFI of ICOS and Nrp1 in the tumors of WT and AMPK^fl/flFoxp3-Cre mice with or without anti-PD-1 treatment. The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 4

AMPK activation negatively regulates HMGCR in Treg cells. A Analysis of the indicated genes in Tregs isolated from WT and AMPK^fl/flFoxp3-Cre mice by RT PCR. B Immunoblot analysis of the indicated proteins in WT and AMPK-KO Tregs after 1 h of resting and 4 h of stimulation with anti-CD3/CD28. C Detection of the phosphorylated and dephosphorylated form of HMGCR from WT Tregs treated with AICAR at indicated doses after 4 h of stimulation with anti-CD3/CD28. D Immunoprecipitation of the lysates of WT Tregs using anti-AMPK and anti-HMGCR antibodies. E The volume and weight of B16F10 tumors from WT mice after treatment with PBS, AICAR, statin, and AICAR + statin combination therapy. F The volume and weight of B16F10 tumors in Rag1^−/− mice following PBS treatment or co-treatment with AICAR and statin. G MFI of PD-1 in Tregs from tumor-bearing WT mice after the indicated treatments. The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 5

AMPK activates p38 MAPK through the inhibition of HMGCR to regulate the expression of PD-1. A Western blot analysis of the expression of PD-1 in treatment with the AMPK modulators: AICAR and compound C and the mevalonate pathway byproducts: cholesterol, GGPP, and mevalonate. B Immunoblot analysis of phosphorylated and total ERK, JNK, and p38 MAPK in Tregs from WT and AMPK^fl/flFoxp3-Cre mice after 1 h of resting and 3 h of stimulation with anti-CD3/CD28. Western blot analysis of phosphorylated and total p38 MAPK in WT Tregs treated with or without C AICAR or D compound C at the indicated doses. E Immunoblot analysis of phosphorylated and total ERK and JNK in WT Tregs treated with AICAR, compound C, or statin. F Immunoblot analysis of phosphorylated and total form of p38 in Tregs treated with different concentrations of statin. G Immunoblot analysis of PD-1 in Tregs treated with SB203580 at the indicated doses. H MFI of PD-1 in WT Tregs with or without SB203580 treatment. The I tumor volume and J weight of B16F10 melanoma tumors in WT mice treated with PBS or SB203580. K MFI of PD-1 in Tregs isolated from tumor tissues after treatment with PBS or SB203580. The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 6

AMPK suppresses the activity of GSK3β through p38 MAPK activation to regulate the expression of PD-1. A Analysis of phosphorylated and total GSK3β and β-catenin in WT and AMPK-KO Tregs stimulated with or without anti-CD3/CD28 stimulation for 4 h. Analysis of the expression of GSK3β and β-catenin by western blotting in activated Tregs after B AICAR and C compound C treatment at the indicated doses. D Immunoblot analysis of the indicated proteins in activated Tregs treated with the p38 MAPK inhibitor SB203580 at the indicated doses. E Immunoprecipitation assay of the lysates of WT Tregs using anti-p38 and anti-GSK3β antibodies. Flow cytometric analysis of F PD-1 and G T-bet in Tregs from WT and AMPK^fl/flFoxp3-Cre mice after i.p. treatment with or without the GSK3β inhibitor SB216763 daily for 7 days (2 μg/kg). The data are presented as the mean ± standard deviation (SD); n = 5 mice per group. *P < 0.05; **P < 0.01; ***P < 0.001