Impact of Hydroxyglutarate on Dendritic Cell Activation

Tumor development is commonly associated with genetic alterations such as activation of oncogenes or the silencing of tumor suppressor genes. Recently point mutations in genes encoding isocitrate dehydrogenase (IDH) have gained substantial interest in glioma classification. IDH mutations are also frequently found in neoplasias including acute myeloid leukemia (AML). Mutated IDH aquires the function to produce the oncometabolite D-2-hydroxyglutarate (D-HG). D-HG levels in the tissue of glioma patients can reach concentrations up to 35mM. Nevertheless, little is known on the impact of D-HG on non-malignant human immune cells. In the present work, we compared the effect of both HG enantiomers, D-HG and L-2-Hydroxyglutarate (L-HG) on dendritic cell (DC) maturation and activation, and found similar effects with both enantiomers. We also analyzed the impact of D-HG on the respiration of DCs and two glioblastoma cell lines. We found that D-HG was taken up by LPS-stimulated DCs. Furthermore, the presence of D-HG during maturation and activation of DCs strongly affected the production of IL-12. Especially when DCs were treated with D-HG during the entire maturation process, the IL-12 production was extremely reduced. To elucidate the mechanism by which D-HG modulates IL-12 production, we studied the principal components of the Toll-like receptor 4 signaling pathway involved in IL-12 p70 production (NF-κB, HIF-α, PI3-kinase pathway components). However, we found no significant effect of D-HG treatment on any of these pathways after 1hr or 24hrs of incubation. As we found an increase of GPR109A expression after LPS stimulation, we studied a possible involvement of this receptor but found no significant relation between IL-12 production and the reported GPR109A ligand, niacin. Cyclic adenosine monophosphate (cAMP) has been described to be involved in DC function. We found significant reduction of IL-12 production by the treatment of DCs with cAMP and the cAMP-inducing-factor forskolin. However, due to technical problems, we could not detect cAMP in DCs. As several reports state a possible relation between cAMP signaling pathways and mitochondrial respiration, we analyzed the impact of D-HG on mitochondrial respiration of LPS-stimulated DCs. We observed an inhibition of routine respiration caused by LPS, whereas pre-treatment with D-HG blocked this LPS effect. As treatment with oligomycin and rotenone (inhibitors of mitochondrial respiration) did not recover IL-12 suppression, the impact of HG on respiration alone is not sufficient to explain the suppressive effect of HG. To further analyze the effect of D-HG on mitochondrial respiration we utilized an inhibitor of mitochondrial calcium efflux, cyclosporine A (CsA). We noticed an increase of IL-12 levels after CsA treatment compared with DCs activated without CsA treatment. Combination of CsA and D-HG in LPS-stimulated DCs partially rescued IL-12 levels caused by D-HG treatment. During high resolution respirometry studies we observed a delayed reduction of routine respiration 2hrs after CsA treatment. Also, we found a modest increase of respiratory chain complex expression in stimulated DCs treated with D-HG (complex I and II). Moreover, lactate secretion studies showed an increase of lactate levels by LPS. D-HG was able to block this increase. These observations point towards the capacity of D-HG to interfere with the metabolic shift from oxidative phosphorylation into glycolysis promoted by LPS. This could also explain the suppressive effect of D-HG on IL-12 production. IDH mutations are associated with glioma development and prognosis. Nevertheless the effect of D-HG on mitochondrial respiration of glioma cell lines is poorly documented. In the present work, we found no consistent effect on mitochondrial respiration by D-HG in two glioma cell lines. Furthermore, we investigated global effects of HG on DCs by RNAseq analyses. Surprisingly most genes were similarily regulated by HG and our control substance di-ketoglutarate most likely due to its close structural similarity. Only 6 genes were upregulated and 3 genes were downregulated by HG. Further investigations have to be performed to investigate the link between the identified genes and DC function