Background: Calcium/Calmodulin-dependent protein kinase IV (CaMK4) is a serine kinase expressed in CD4 T cells whom activity is increased in patients with active systemic lupus erythematosus (SLE). We have shown that CaMK4 negatively impacts T regulatory (Treg) cells differentiation and function in SLE, but the underlying mechanism remains unclear. Recent, data suggest that CaMK4 impacts cellular metabolism.

Objectives: Investigate how CaMK4 impacts Treg cell metabolism and its potential effect on Treg cell function.

Methods: We harvested CD62L ⁺ CD4 ⁺ T cells from wild-type (WT) or Camk4 ^{-/ -} mice and differentiated them in vitro into Treg (iTreg) cells. We assessed iTreg metabolism using Seahorse XF analyzer and mass spectrometry (metabolomics). Gene expression was assessed at the mRNA (RT-qPCR) and at the protein level (Western Blot). Phosphofructokinase activity was assessed by a colorimetric assay (Abcam). In vitro gene knockdown was conducted by transfecting a guide RNA (gRNA) in CRISPR/Cas9-expressing T cells. Treg cell function was evaluated by in vitro immunosuppressive assay and in vivo by the adoptive transfer of T conventional T and iTreg cells (8:1 ratio) in Rag1 ^{-/ -} mice to induce inflammatory colitis. The relevance of CaMK4 in SLE was evaluated in vivo using a T-cell specific knockdown of CaMK4 in the B6.lpr mouse model, and in humans by culturing SLE patient T cells with KN-93, a CaMK4 specific inhibitor.

Results: iTreg cells from Camk4 ^{-/ -} mice had decreased glycolysis and increased mitochondrial metabolism compared to WT mice. Metabolomics studies suggested decreased activity of the rate-limiting glycolysis enzyme phosphofructokinase platelet-type (PFKP). While PFKP mRNA and protein levels were similar between WT and Camk4 ^{-/ -} iTreg, we found that PFKP activity was significantly decreased in Camk4 ^{-/ -} iTreg, suggesting post-transcriptional control of PFKP activity. Mechanistically, immunoprecipitation experiments confirmed that CaMK4 interacted with PFKP, and phosphoproteomic study suggested that CaMK4 phosphorylated serine residue 539 of PFKP, a site known to control PFKP activity. Excitingly, PFKP’s endproduct fructose 1,6-biphosphate negatively regulates the activation of the mitochondrial metabolism masterswitch AMPK, therefore linking decreased PFKP activity/glycolysis with increased mitochondrial metabolism in Camk4 ^{-/ -} Treg. To confirm the importance of PFKP in Treg biology, we confirmed that PFKP knockdown significantly improved iTreg function in vitro (p < 0.01) and in vivo using an adoptive CD4+ T cell transfer in to Rag1 ^{-/ -} mice (colitis model). Interestingly, iTreg lacking PFKP were transferred Rag1 ^{-/ -} mice were less likely to lose FoxP3 expression and to produce IL-17A, demonstrating higher Treg stability in an inflammatory environment. On a translational basis, lupus-prone B6. lp r mice with a T-cell specific CaMK4 knockdown displayed significantly less lupus manifestations. In human SLE, CD4 ⁺ T cells had higher PFKP activity compared to healthy donors, and PFKP activity correlated with the SLE disease activity index (SLEDAI, r= 0.47; p <0.05). Finally, culture of SLE CD4+ T cells with KN-93 led to a significant decrease in PFKP activity (p < 0.001).

Conclusion: Increased CaMK4 activity in human SLE mediates Treg dysfunction and instability by altering PFKP activity. Restoring normal Treg metabolism by inhibition of CaMK4 or its downstream target PFKP represents a novel strategy for the treatment of SLE.

Acknowledgements: Marc Scherlinger is financially supported by the Societe Françaises de Rhumatologie (SFR), Philippe, Monahan & Arthurs Sachs foundations.

Disclosure of Interests: None declared