ARTICLE IN PRESS Prostaglandins, Leukotrienes and Essential Fatty Acids 70 (2004) 309–319 Review The therapeutic implications of ketone bodies: the effects of ketone bodies in pathological conditions: ketosis, ketogenic diet, redox states, insulin resistance, and mitochondrial metabolism Richard L. Veech* Laboratory of Membrane Biochemistry and Biophysics, National Institutes of Alcoholism and Alcohol Abuse, 12501 Washington Ave., Rockville, MD 20850, USA Received 10 August 2003; accepted 1 September 2003 Abstract The effects of ketone body metabolism suggests that mild ketosis may offer therapeutic potential in a variety of different common and rare disease states. These inferences follow directly from the metabolic effects of ketosis and the higher inherent energy present in d-b-hydroxybutyrate relative to pyruvate, the normal mitochondrial fuel produced by glycolysis leading to an increase in the DG0 of ATP hydrolysis. The large categories of disease for which ketones may have therapeutic effects are: (1) diseases of substrate insufficiency or insulin resistance, (2) diseases resulting from free radical damage, (3) disease resulting from hypoxia. Current ketogenic diets are all characterized by elevations of free fatty acids, which may lead to metabolic inefficiency by activation of the PPARsystem and its associated uncoupling mitochondrial uncoupling proteins. New diets comprised of ketone bodies themselves or their esters may obviate this present difficulty. Published by Elsevier Ltd. 1. Metabolic effects of ketone body metabolism that the effects of ketone metabolism in heart would mimic those in brain, which was not analyzed in this The therapeutic potentials of mild ketosis flow detailed manner for a number of technical reasons, most directly from a thorough understanding of their meta- prominently the inhomogeneous nature of the tissue and bolic effects, particularly upon mitochondrial redox its lack of quantifiable outputs. states and energetics and upon substrate availability. A detailed metabolic control strength analysis of The data on metabolic effects of ketone body metabo- glycolysis in heart under the four conditions led to lism presented here has been published previously [1,2]. several major conclusions [1]. Firstly, the control of flux It presents studies of the isolated working rat heart through the glycolytic pathway was context dependent perfused with 11 mM glucose alone, glucose plus 1 mM and shifted from one enzymatic step to another acetoacetate and 4 mM d-b-hydroxybutyrate, gluco- depending upon the conditions. There was not one key se+100 nM insulin or the combination of glucose, ‘‘rate controlling’’ reaction, but rather control was ketone bodies and insulin. The isolated working distributed among a number of steps, including some perfused heart was studied because of the relative enzymes that were very close to equilibrium. The homogeneity of the tissue and the simplicity of its absence of a single dominant rate controlling step in a output, the number of parameters which could be pathway calls into question the assumptions on which accurately measured, particularly O2 consumption many pharmaceutical discovery programs have been relative to actual hydraulic work output of the heart. based [3]. Secondly, when perfused with glucose alone, In our analysis of disease states, it has been assumed there was consistent glycogen breakdown, whereas with addition of ketones, insulin or the combination, *Tel.: +1-301-443-4620; fax: 1-301-443-0930. glycogen synthesis occurred. Addition of either ketones E-mail address: [email protected] (R.L. Veech). or insulin, increased intracellular [glucose] and the 0952-3278/$ - see front matter Published by Elsevier Ltd. doi:10.1016/j.plefa.2003.09.007 ARTICLE IN PRESS 310 R.L. Veech / Prostaglandins, Leukotrienes and Essential Fatty Acids 70 (2004) 309–319 glycolytic intermediates in the first half of the glycolytic working perfused heart [2]. The proton gradient created pathway from 2 to 8 fold. Thirdly, addition of either by the redox energy of the respiratory chain then powers ketones or insulin leads to an increase in the measurable the transport of protons from cytosol back into hydraulic work of the heart, but a net decrease in the mitochondria through the F1 ATPase complex in an rate of glycolysis. Associated with the increase in efficient and reversible process [7,8]. For each electron hydraulic work and decrease in glycolytic rate addition pair transported up the respiratory chain, 3 ATPs are of ketones or insulin increased the free cytosolic [ATP]/ generated. Since the maximum energy available from the [ADP]  [Pi] ratio three to five fold and a similar change redox reactions of the chain is À178 kJ, the energy in the [phosphocreatine]/[creatine] ratio showing that available for the synthesis of each of the 3 ATPs can not both ketones or insulin increased the energy of the exceed À59.2 kJ/mol. The energy of the hydrolysis of 0 phosphorylation state of heart significantly compared to ATP, DGATP; in heart, liver and red cell under 6 perfusion with glucose alone. These data clearly show conditions ranged from À54 to À58 kJ/mol, implying that addition of either ketone bodies or insulin, that the overall process of electron transport and markedly improved the energy status of working oxidative phosphorylation is a remarkably efficient perfused heart. How ketone bodies could increase the process. hydraulic efficiency of heart by 28% could not be The electrons liberated from mitochondrial NADH explained by the changes in the glycolytic pathway by the NADH dehydrogenase complex or complex I, are alone, but rather by the changes that were induced carried within the mitochondria by co-enzyme Q, where in mitochondrial ATP production by ketone body they are transferred to cytochrome C, by CoQH2- metabolism. cytochrome C reductase, complex III. The difference The mitochondrial processes of ATP generation between the redox potential of the mitochondrial NAD derive their energy from the respiratory electron couple and the co-enzyme Q couple, DEQ=NADH; transport chain, where the electrons from substrates determines the energy of the proton gradient generated enter at different catalytic centers and travel up through by mitochondria. This in turn determines the energy of 0 various redox couples within the chain to ultimately hydrolysis of ATP, DGATP; which is generated by the + combine with H and O2 to form H2O. The respiratory mitochondrial F1 ATPase [2]. chain begins with the NADH multienzyme complex, The ketone bodies, acetoacetate and d-b-hydroxybu- whose substrate is the free mitochondrial NADH. The tyrate are in near-equilibrium with the free mitochon- redox potential of the free mitochondrial [NAD+]/ drial [NAD+]/[NADH] ratio in a reaction catalyzed by [NADH] couple is about À0.28 V [4] while that of the d-b-hydroxybutyrate dehydrogenase [9] It would also [O2]/[H2O] couple is +1.2 V [5]. The total energy appear that the [succinate]/[[fumarate] couple is in near available from the movement of electron up the equilibrium with the mitochondrial [Q]/[QH2] couple in respiratory chain is therefore determined by the a reaction catalyzed by succinate dehydrogenase [2]. difference between the variable redox potential of the When ketone bodies are metabolized in heart, the mitochondrial NAD couple and the O2 couple, which is mitochondrial NAD couple is reduced while the constant at all O2 concentrations and is given by mitochondrial Q couple is oxidized increasing the redox DG0 ¼ÀnFDE; span, DEQ=NADH; between site I and site III, making more energy available for the synthesis of ATP, and where n ¼ 2 electrons and F ¼ 96:485 kJ/mol/V. hence an increase in the DG0 of ATP hydrolysis. This in This means 2 electrons traveling up the electron turn is observable in the 28% increase in the hydraulic transport system in the redox reaction efficiency of the working perfused rat heart. þ 1 - þ The fundamental reason why the metabolism of NADHmito þ H þ 2O2 H2O þ NADmito ketone bodies produce an increase of 28% in the can yield À178 kJ/2 moles of electrons. hydraulic efficiency of heart compared with a heart The energy gained from mitochondrial electron metabolizing glucose alone is that there is an inherently transport is transferred to the pumping of protons from higher heat of combustion in d-b-hydroxybutyrate than the mitochondrial to cytosolic phase [6] at sites I, III and in pyruvate, the mitochondrial substrate which is the IV creating an electrochemical proton gradient between end product of glycolysis (Table 1). the two phases where the energy of the proton gradient If pyruvate were burned in a bomb calorimeter, it is would liberate 185.7 kcal/mole of C2 units, whereas the 0 þ þ d DG ½H cyto=½H mito combustion of -b-hydroxybutyrate would liberate 243.6, or 31% more calories per C unit than ¼ RT ln½Hþ =½Hþ þ FE ; 2 cyto mito mito=cyto pyruvate. Metabolizing d-b-hydroxybutyrate in per- where the major energy component is the electric fused working heart creates a 28% increase in the potential between the mitochondrial and cytosolic hydraulic efficiency of heart when compared to the phases, which ranges between À120 and À140 mV in metabolism of the end product of glycolysis, pyruvate. ARTICLE IN PRESS R.L. Veech / Prostaglandins, Leukotrienes and Essential Fatty Acids 70 (2004) 309–319 311 Table 1 In addition to their effects on mitochondrial ener- Heats of combustion of common non-nitrogenous energy substrates getics, the metabolism of ketone bodies has other effects o o Substrate DH kcal/mol