Pyruvate Dehydrogenase Kinase Modulates Platelet Function and Thrombosis

Background and objective: Pyruvate dehydrogenase kinases (PDK 1-4) are mitochondrial enzymes that regulate the activity of pyruvate dehydrogenase (PDH) complex, which converts pyruvate (end product of glycolysis in cytosol) to acetyl coenzyme A that is further oxidized to produce energy (oxidative phosphorylation; OXPHOS). Like most normal cells, resting platelets rely primarily on OXPHOS to generate ATP, whereas activated platelets exhibit a high level of aerobic glycolysis (conversion of glucose to lactate in the presence of oxygen, a phenomenon referred to as the Warburg effect in tumor cells). It seems paradoxical that why platelets prefer aerobic glycolysis instead of OXPHOS. The rate of ATP generation is faster in aerobic glycolysis compared to OXPHOS, which is well suited for high-energy demand of activated platelets to form blood clot. We tested the hypothesis that manipulating aerobic glycolysis (metabolic reprogramming), by down regulating PDK activity in the mitochondria of platelets, will inhibit platelet function and thereby thrombosis.

Methods: Global knock out or floxed PDK mice does not exists, therefore we used dichloroacetic acid (DCA), a potent inhibitor of PDKs, known to divert metabolism back from aerobic glycolysis to OXPHOS. DCA has been successfully used in clinical trials to treat several type of cancer. In human and murine platelets, using standardized platelet in vitro assays, we determined the mechanistic role of PDK in regulating platelet function. Rate of glycolysis and mitochondrial respiration was measured using the Seahorse XF24 extracellular flux analyzer. Using intravital microscopy, thrombus growth was evaluated in vitro (microfluidics flow chamber) and in vivo (FeCl₃-induced carotid artery thrombosis) models.

Results: Using, extracellular flux analysis, we first confirmed that activated platelets have higher aerobic glycolysis. With both human and mice platelets, we found that DCA (10 and 25 mM) significantly inhibited in vitro platelet aggregation to collagen and convulxin-induced signaling, but not thrombin or ADP-induced signaling in a dose-dependent manner (P<0.05 vs. vehicle treated). DCA also inhibited static adhesion of washed platelets to collagen in dose-dependent manner. Furthermore, we found that DCA significantly inhibited tyrosine phosphorylation of Syk and PLCγ2 in the GPVI signaling pathway with both mice and human platelets (P<0.05 vs. vehicle treated). Metabolic profiling, using extracellular flux analysis, showed that DCA significantly inhibited extracellular acidification rate in convulxin-stimulated platelets in dose-dependent manner. Using microfluidics flow chamber, we found that both mouse and human whole blood treated with DCA formed small thrombi when perfused over collagen for 10 min at a arterial shear rate of 1500s^-1 (P<0.05 vs. vehicle control), suggesting that inhibiting PDK with DCA limits thrombus growth. Finally, wild-type mice pre-treated with DCA (600 mg/Kg body weight) were less susceptible to thrombosis in FeCl₃-induced arterial thrombosis model (P<0.05 vs. control, N=10 mice/group), without altered hemostasis (tail-bleeding assay). The precise mechanism how aerobic glycolysis modulates GPVI signaling pathway is unclear and is currently under investigation.

Conclusion: Our results suggest that PDK present in mitochondria regulates platelet function and thrombus growth, most likely via GPVI signaling pathway.