Background Autoreactive CD4⁺ T cells are essential participants in the pathogenesis of Systemic Lupus Erythematosus (SLE). Immune substrate utilisation, including glucose metabolism, plays a central role in dictating the effector functions of CD4⁺ T cells. We hypothesised that 1) SLE T cells have metabolic defects that enhance their pro-inflammatory functions, and 2) Inhibiting glycolytic metabolism in CD4⁺ T cell may normalise CD4⁺ T cell functions and reduce disease symptoms in SLE mice and in CD4⁺ T cells from SLE patients.

Materials and methods We utilised four models of spontaneous lupus, B6.NZM2410.Sle1.Sle2.Sle3 Triple Congenic (TC), BWF1, BXSB.YAA and B6.lpr that differ in their genetic background as well as mechanisms of autoimmune activation. C57BL/6 (B6) served as a non-autoimmune control strain. CD4⁺ T cells obtained from lupus-prone mice and controls, as well as from SLE patients and healthy controls (HC) were treated with metabolic inhibitors, including metformin, which inhibits mitochondrial complex I and activates AMPK, and the glycolytic inhibitor 2-Deoxy-D-Glucose (2-DG). Lupus-prone mice were treated with these drugs, either before or after disease onset. Glycolysis, oxygen consumption, activation and effector subset distribution were measured in CD4⁺ T cells. Disease progression was assessed by measuring standard lupus biomarkers. Gene profiling was performed on CD4⁺ T cells from SLE patients and HCs.

Results CD4 T cells from lupus mice and patients have a significantly higher metabolism as well as an enhanced mTOR activity as compared to controls. A combination of inhibitors and gene expression studies showed that glucose oxidation rather than aerobic glycolysis is a major metabolic pathway of lupus T cells. In vitro, both metformin and 2-DG blocked IFNγ production and metformin increased IL-2 production. In vivo, a combined treatment with metformin and 2-DG normalised T cell metabolism and reversed disease phenotypes in all four lupus mouse models. Remarkably, the number of T_FH cells, which correlates with disease severity in patients, was normalised by the combination treatment in every model. Further, excessive IFNγ production by CD4 T cells from SLE patients was also normalised by metformin. Finally, CD4⁺ T cells from lupus patients showed a metabolic signature with over-expression of genes promoting glucose utilisation, mitochondrial oxidative phosphorylation and sterol synthesis.

Conclusions The combination of a glucose inhibitor with metformin restores T cell function and reverses disease across diverse mouse models of SLE, and metformin treatment normalises the function of T cells from SLE patients. We propose that T cell metabolism may serve as a biomarker of disease activity and provides a novel target for immune intervention in SLE.

Acknowledgements This work was supported by NIH grants R01 AI045050 and ALR-TIL 0000075018 to L. Morel.