Metabolic remodeling is an underlying theme in diseases such as heart disease, type 2
diabetes, and cancer, where cells deviate from their typical fuel utilization profile. The pyruvate
dehydrogenase complex (PDC) is the gatekeeper for aerobic glucose utilization and its
downregulation is greatly associated with these diseases. As such, recent efforts have focused
on the activation of PDC as a therapeutic. Currently, efforts have focused only on inhibition of
the pyruvate dehydrogenase kinase (PDK), which inhibits PDC by phosphorylation. However,
activation of the pyruvate dehydrogenase phosphatase (PDP) as a therapeutic, which de-
phosphorylates PDC to recover activity, has been ignored. Furthermore, it is well-known that
PDC activity is inhibited by NADH but efforts to therapeutically target this regulatory mechanism
have been overlooked. As a more comprehensive PDC activation strategy, we look to expand
PDK inhibitors, identify PDP small molecule activators, and optimize quinone compounds that
have been shown to recover PDC from NADH inhibition as a new therapeutic strategy for
metabolic disease. Secondly, reactive oxygen species (ROS) generated from the mitochondrial
electron transport chain (mETC) are thought to be enhanced by PDC activation but the impact
of PDC as a source of ROS and how specific conditions such as NADH/NAD and ATP demand,
as in exercise, influence ROS homeostasis is unclear. Site-specific quantification of ROS
generation requires detailed enzymatic simulations in various conditions to be elucidated. The
aims of this project are to 1) virtually screen and experimentally validate PDC activators and 2)
apply mathematical modeling to determine effects of PDC activation on site-specific mETC ROS
generation. We obtained 15M drug-like virtual compounds from the ZINC database to
individually screen all PDK isozymes (1-4) at three known inhibition sites: lipoamide, ATP/ADP,
and pzf3 to find isozyme specific and pan inhibitors. This virtual compound set will also be used
to identify PDP activators by virtually screening a hybrid crystal/computationally derived PDPc-
PDPr complex. We will use the NADH binding domain of PDC (E3), in a structure guided
approach, to identify optimal quinone analogs for PDC. Detailed enzyme kinetics models of
PDC, TCA cycle, and mETC will be integrated to simulate various oxidative states including
reductive stress and exercising conditions to quantify site-specific mitochondrial ROS
production. We believe that our proposal addresses significant gaps in strategies to activate
PDC as a therapeutic and will provide a quantitative understanding of the influence of PDC
activity on mitochondrial ROS production.