AMPK Controls Glycolytic Compensation in Gvhd-Causing T Cells

Abstract

Background: The heightened metabolic requirements of alloreactive T cells following allogeneic stem cell transplantation represents a promising area for therapeutic intervention. We have shown that during the initiation of acute graft-versus-host-disease (aGVHD), murine T cells upregulate activity of AMP-activated protein kinase (AMPK), a cellular energy sensor that regulates oxidative and mitochondrial metabolism. Further, transfer of allogeneic T cells deficient in AMPK (AMPK KO) increased recipient survival and decreased disease severity while still preserving leukemia clearance. In the current studies, we demonstrate a novel role for AMPK in controlling glycolytic compensation, both in vitro and vivo, and provide a mechanism as to why cells lacking AMPK are unable to perform this vital function. Together, these data highlight the potential clinical utility in targeting metabolic decision-making to curtail T cell-mediated inflammatory responses, with significant therapeutic implications for the prevention and treatment of aGVHD. Methods: AMPK-deficient murine T cells from AMPKa1/a2 dKO C57BL/6J mice were transplanted into allogeneic B6D2F1 recipients and recovered on day 7 post-transplant. To delete AMPK in primary human T cells, T cells were isolated from healthy donors and electroporated with Cas 9 ribonucleoprotein complexes targeting the AMPKa1 locus. Glycolytic compensation was assessed by measuring extracellular acidification rates (ECAR) using the Agilent Seahorse analyzer and a mitochondrial stress kit. Intracellular cytokine expression for IFNgamma was assessed in day 7 murine T cells following re-stimulation with F1 splenocytes in the presence of Brefeldin A. Aldolase A and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) enzymatic assays were run on cell lysates with Biovision-Abcam kits according to manufacturer instructions. To measure GAPDH activity, T cells were positively selected for CD45.1 from day 7 recipient spleens, followed by cell lysis and detection of enzymatic activity. Results: AMPK KO murine T cells recovered on day 7 were unable to mediate a compensatory increase in glycolysis following inhibition of the electron transport chain (Figure 1A), (-6.499±5.156 vs 29.11±7.064, p<0.0001). Human T cells lacking AMPK gave similar results, with glycolytic compensation impaired both in vitro and following human T cell expansion in vivo in a xenogeneic model of GVHD. Immunoprecipitation of proteins from day 7 murine T cells, using an antibody specific to phosphorylated AMPK targets, recovered lower levels of multiple glycolytic enzymes including aldolase, enolase, pyruvate kinase M, and GAPDH (Figure 1B). Functionally, murine T cells lacking AMPK exhibited impaired aldolase activity following anti-CD3/CD28 stimulation and a decrease in GAPDH activity on day 7 post-transplant. Importantly, these changes in glycolysis correlated with an impaired ability of AMPK KO T cells to produce significant amounts of IFNγ upon antigenic re-stimulation with F1 splenocytes. Discussion: Our results demonstrate that in addition to influencing oxidative metabolism, AMPK also plays a role in glycolytic metabolism during GVHD in both murine and human T cells. Further, our data suggest that the more prominent lesion in human T cells post-transplant was in glycolysis , a result supported by the novel observation that a set of glycolysis-related proteins are immunoprecipitated together at much lower rates in the absence of AMPK. These findings suggest a model in which AMPK phosphorylation provides a nidus for important interactions among glycolytic enzymes and that in the absence of AMPK, these interactions are lessened, and glycolysis is consequently reduced. Important to our understanding of GVHD pathogenesis, rates of T cell glycolysis can be directly linked to IFNγ production, where reduced glycolytic activity frees GAPDH to sequester IFNγ transcripts and thereby limit IFNγ protein translation. Altogether, these results document a prominent role for AMPK in controlling T cell glycolytic activity during GVHD development and offer an additional mechanistic explanation for the decreased severity of GVHD seen following transplantation of AMPK KO T cells. Consistently similar results in human xenogeneic T cells provides a further strong rationale for development of AMPK inhibition as a clinically translatable strategy for GVHD prevention and treatment.