Artesunate Inhibits Graft-Versus-Host Disease in Mice Via a Mechanism of Inducing Mitochondrial Calcium Overloading in Activated T Cells

Despite pharmacological prophylaxis using calcineurin (CN) inhibitors (i.e., cyclosporin A and Tacrolimus), graft-versus-host disease (GVHD) remains a major barrier to the success of allogeneic hematopoietic stem cell transplantation (allo-HSCT). Upon antigen stimulation, activated T cell receptor (TCR) signaling triggers rapid Ca²⁺ influx. This induces both CN-mediated NFAT activation, and increase of mitochondrial Ca²⁺ content, a major driver of metabolic activity. However, mitochondrial Ca²⁺ overload triggers opening of the mitochondrial permeability transition pore and cell death. We hypothesize that pharmacologically increasing mitochondrial Ca²⁺ load may decrease T cell survival capability, thereby reducing the GVH reaction. If this can be accomplished, it may lead to new strategies for inhibition of GVHD, which is conceptually different from the use of CN inhibitors. To test this hypothesis, we employed a high throughput drug screening system with the proliferation and cytotoxicity of TCR-activated human T cells as the readout, and screened the NPL-800 library (https://www.timtec.net) composed of 800 pure natural compounds. With a stringent criterion in consideration of dose-dependent effect, 26 compounds stood out for reducing the count of activated human T cells with a reduction rate of at least 30% at both 1.0uM and 10.0uM. Positive hits included inhibitors of DNA synthesis, the Na-K-ATPase and mitochondrial metabolism. We were particularly interested in artesunate (ART), which is a derivative of artemisinin that has been used for treating malaria in patients. While artemisinin acts by inhibiting sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) in P. falciparum malaria, which causes passive endoplasmic reticulum (ER) Ca²⁺ depletion and the subsequent cytosolic Ca²⁺ influx, ART did not inhibit SERCA in T cells. Ex vivo culture assays showed that ART dose-dependently reduced the survival of TCR-activated murine T cells. This inhibitory effect of ART was abrogated by inhibiting Ca²⁺ influx using BTP2, a potent inhibitor of store-operated Ca²⁺ channels. Furthermore, treatment of murine CD8 T cells with ART induced significant increases in mitochondrial Ca²⁺ loading upon TCR activation. These data suggest that inhibition of T cell survival by ART was dependent on TCR activation-induced Ca²⁺ influx and associated with enhanced mitochondrial Ca²⁺ uptake. We examined the impact of ART on GVHD in Balb/c mice receiving C57BL/6 (B6) mouse T cell-depleted bone marrow (TCD-BM) and CD4⁺ T cells. Intraperitoneal injection of ART (10 mg/kg, every other day) from day 1 to day 28 after transplantation reduced clinical signs of GVHD in these recipients and significantly improved their overall survival. Similar inhibition effects of ART on GVHD were observed in miHA-mismatched B6 anti-Balb/b and haplo-identical B6 anti-BDF1 mouse models of GVHD. Further investigations showed that in vivo administration of ART caused significant decreases in the number of host-reactive donor T cells in the spleen and liver of Balb/c mice 7 days after transfer of B6 TCD-BM plus CD4⁺ T cells. ART treatment did not affect the capacity of donor T cells to produce effector cytokines (e.g., IFN-g and TNF-α) in individual cells. Importantly, in vivo administration of ART preserved anti-leukemia activity of donor T cells and did not impair the reconstitution of hematopoiesis and lymphocytes. Collectively, our findings indicate that pharmacologically increasing mitochondrial Ca²⁺ loading may have significant implications in the development of novel strategies to prevent GVHD and other T cell-mediated inflammatory disorders in a broad context. Since ART therapy has been clinically approved, this work could be immediately translated into patients.