Search

The pyruvate dehydrogenase complex (PDC) is an important gatekeeper enzyme connecting glycolysis to the tricarboxylic acid (TCA) cycle and oxidative phosphorylation (OXPHOS). Thereby, it has a strong impact on the glycolytic flux as well as the metabolic phenotype of a cell. PDC activity is regulated via reversible phosphorylation of three serine residues on the pyruvate dehydrogenase (PDH) E1α subunit. Phosphorylation of any of these residues by the PDH kinases (PDKs) leads to a strong decrease in PDC activity. Under hypoxia, the inactivation of the PDC has been described to be dependent on the hypoxia-inducible factor 1 (HIF-1)-induced PDK1 protein upregulation. In this study, we show in two hepatocellular carcinoma cell lines (HepG2 and JHH-4) that, during the adaptation to hypoxia, PDH is already phosphorylated at time points preceding HIF-1-mediated transcriptional events and PDK1 protein upregulation. Using siRNAs and small molecule inhibitor approaches, we show that this inactivation of PDC is independent of HIF-1α expression but that the PDKs need to be expressed and active. Furthermore, we show that reactive oxygen species might be important for the induction of this PDH phosphorylation since it correlates with the appearance of an altered redox state in the mitochondria and is also inducible by H2O2 treatment under normoxic conditions. Overall, these results show that neither HIF-1 expression nor PDK1 upregulation is necessary for the phosphorylation of PDH during the first hours of the adaptation to hypoxia.

The endogenous cannabinoid 2-arachidonoyl glycerol (2-AG) is an anti-fibrotic lipid mediator that induces apoptosis in hepatic stellate cells (HSCs), but not in hepatocytes. However, the exact molecular mechanisms of this selective induction of HSC death are still unresolved. Interestingly, the inducible isoform of cyclooxygenase, COX-2, can metabolize 2-AG to pro-apoptotic prostaglandin glycerol esters (PG-GEs). We analyzed the roles of COX-2 and endocannabinoid-derived PG-GEs in the differential susceptibility of primary activated HSCs and hepatocytes toward 2-AG-induced cell death. HSCs displayed significant COX-2 expression in contrast to hepatocytes. Similar to 2-AG, treatment of HSCs with PGD2-GE dose-dependently induced cell death independently from cannabinoid receptors that was accompanied by PARP- and caspase 3-cleavage. In contrast to 2-AG, PGD2-GE failed to induce significant ROS formation in HSCs, and depletion of membrane cholesterol did not rescue HSCs from PGD2-GE-induced apoptosis. These findings indicate differential engagement of initial intracellular signaling pathways by 2-AG and its COX-2-derived metabolite PGD2-GE, but similar final cell death pathways. Other PG-GEs, such as PGE2-or PGF2alpha-GE did not induce apoptosis in HSCs. Primary rat hepatocytes were mainly resistant against 2-AG- and PGD2-GE-induced apoptosis. HSCs, but not hepatocytes were able to metabolize 2-AG to PGD2-GE. As a proof of principle, HSCs from COX-2(-/-) mice lacked PDG2-GE production after 2-AG treatment. Accordingly, COX-2(-/-) HSCs were resistant against 2-AG-induced apoptosis. In conclusion, the divergent expression of COX-2 in HSCs and hepatocytes contributes to the different susceptibility of these cell types towards 2-AG-induced cell death due to the generation of pro-apoptotic PGD2-GE by COX-2 in HSCs. Modulation of COX-2-driven metabolization of 2-AG may provide a novel physiological concept allowing the specific targeting of HSCs in liver fibrosis.

Protein kinase CK2 plays an essential role in cell viability in lower and higher eukaryotes. As a global regulator it phosphorylates and thereby regulates a broad array of cellular targets including a large number of transcription factors. Here, we have identified the CCAAT/enhancer binding protein delta (C/EBPdelta) as a new substrate for CK2. Using point mutants of C/EBPdelta the major phosphorylation site for CK2 was mapped to serine 57, which is located within the transactivation domain of C/EBPdelta. For proper functioning as a transcription factor C/EBPdelta has to be translocated into the nucleus where it forms heterodimers with other members of the C/EBP family of proteins and ATF4. Here, we found that CK2 phosphorylation does neither influence the subcellular localization of C/EBPdelta nor its interaction with C/EBPbeta, but rather does CK2 phosphorylation modulate the transcriptional activity of C/EBPdelta. Moreover, we found that CK2 bound to C/EBPdelta, which might help to target CK2 to the transcriptional machinery where it can phosphorylate other transcription factors or co-activators.