Dermot H. Williamson

The role of ketone bodies in caloric homeostasis

Summary In order to obtain information on the factors which control the rate of ketone body utilization in rat tissues the activities of the enzymes involved in the utilization of ketone bodies—3-hydroxybutyrate dehydrogenase, 3-oxoacid CoA-transferase and acetoacetyl-CoA thiolase—were assayed in various tissues. The activities of the transferase (which initiates acetoacetate utilization) were highest in kidney and heart. In submaxillary gland and adrenals the activities were about one-quarter of those in kidney and heart. In brain they were about one-tenth and they were lower still in lung, spleen, skeletal muscle and adipose tissue. No activity was measurable in liver. The activities of the thiolase were roughly parallel to those of the transferase, with the exception of liver and adrenals. The high activity in the latter two tissues may be connected with the additional function of thiolase in the production of acetyl-CoA from fatty acids for the tricarboxylic acid cycle and for steroid synthesis. The activities of the two enzymes in tissues of mouse, hamster, guinea pig and sheep were similar to those of rat tissues, except that the activities were rather low in sheep heart and brain. Starvation, alloxan diabetes or fat feeding—conditions where the rates of ketone body utilization are increased—did not appreciably change the activity of the transferase. Thiolase activity increased in kidney and heart on fat feeding. The activity of 3-hydroxybutyrate dehydrogenase in rat brain did not change during starvation. The findings indicate that the activities of the ketone body utilizing enzymes are not major mechanisms controlling the variations in the rate of ketone body utilization in starvation or alloxan diabetes. The controlling factor in these situations is the concentration of ketone bodies, in particular acetoacetate, in plasma and tissue. At birth the activities of 3-hydroxybutyrate dehydrogenase and of the transferase in rat brain were about two-thirds of those of adult rat brain. The activities rose rapidly throughout the suckling period and at the time of weaning reached values about three times higher than those for adult brain. Later they gradually declined. The thiolase activity remained fairly constant during the suckling period. In rat kidney and rat heart the activities of the three enzymes at birth were less than one-third of those at maturity. They gradually rose and after 5 weeks approached, but never exceeded, the adult value. Throughout the suckling period the total ketone body concentration in the blood was about 6 times higher than in adult fed rats and the concentration of free fatty acids in the blood was 3–4 times higher. Thus, the rate of ketone body utilization in the brain of the suckling rat is determined by both greater amounts of key enzymes and higher concentrations of ketone bodies. The latter may be specially directed to the brain because of the low activities of the relevant enzymes in kidney and heart of suckling rats. The concentration dependence of ketone body utilization has also been demonstrated by measurements of the arterio-venous differences of ketone bodies across the rat brain in vivo. On starvation the arterio-venous difference rose about 6-fold. When, by acetoacetate infusion, the ketone body concentration in the blood of fed rats was raised to that of starved rats, the arterio-venous difference was about the same as in starvation. Both in adult and suckling rats the arterio-venous differences were roughly proportional to the acetoacetate concentration in the blood. Experiments on human volunteers show that post-exercise ketosis is preceded by a rise of the plasma free fatty acid concentration. The degree of post-exercise ketosis is much greater in untrained subjects than in trained athletes. Post-exercise ketosis is probably the consequence of the persistence of raised plasma free fatty acid levels after cessation of exercise.

Integration of metabolism in tissues of the lactating rat

Lactation is characterized by a number of physiological changes including increased cardiac output [ 11, hypertrophy of liver [ 11, heart [ 11, intestine [2,3] and mammary gland [ 11, and an increase in dietary intake [2] (for details see table 1). In addition, there are profound alterations to the metabolism of the tissues of the lactating rat. The net result of these changes is that a high proportion of the substrates available in the circulation (glucose, triacylglycerols, non-esterified fatty acids and ketone bodies) are ‘directed’ [4] to the gland for the production of milk. The aim of this contribution is to describe some of the metabolic changes that are known to occur during lactation in the rat and to speculate on the ‘signals’ (hormones or substrates) which may be involved. Lactation in other species will not be dealt with except where the data are pertinent to the situation in the rat. This contribution is not intended to be’a comprehensive review of the area or the literature and merely contains a selection of topics and questions which interest the writer. Information on the integration of tissue metabolism in the lactating rat is not only of fundamental interest, but is also relevant to the study of dietaryinduced obesity, because despite a large increase in dietary intake (table 1) the lactating rat does not

ENZYMES OF KETONE-BODY UTILISATION IN HUMAN BRAIN

Abstract The enzymes concerned in the utilisation of ketone bodies—3-hydroxybutyrate dehydrogenase, 3-oxoacid CoA-transferase, and acetoacetyl-CoA thiolase—have been measured in human brain obtained at necropsy from subjects aged 0-70 years. The enzymes are present in all regions of the brain examined, and their activities are high enough to account for the uptake of ketone bodies observed by other workers in human subjects undergoing a prolonged fast.

INITIAL EFFECT OF INJURY ON KETONE BODIES AND OTHER BLOOD METABOLITES

The arterial and venous concentrations of ketone bodies and other metabolites were measured in twelve adults, from 2 to 24 hours after injury. Two groups could be distinguished, with or without hyperketonaemia (defined as more than 0.2 mmol per litre) in the 2-hour blood-sample. In the hyperdetonaemic group the concentrations of alanine, pyruvate, and lactate and the urinary nitrogen excretion were lower throughout the first 24 hours than in the non-hyperdetonaemic group. These preliminary results indicate that hyperketonaemia after trauma is associated with decreased protein breakdown.

Stimulation of [1-14C]oleate oxidation to 14CO2 in isolated rat hepatocytes by vasopressin: effects of Ca2+.

Vasopressin and glucagon exert a number of rapid effects on the liver, which include stimulation of ~ycogenolysis f 1,2], activation of phospho~lase [3,4] and enhancement of gluconeogenesis [5,6]. It seems likely that these hormones exert their effects in different ways, since unlike glucagon, vasopressin does not increase the hepatic cyclic AMP concentration [7] and increases glycolytic flux [8]. Glucagon (or cyclic AMP) stimulates ketogenesis [9]. This has been suggested to result from a decrease in the concentmtion of m~onyl-CoA which inhibits carnitine acyltransferase I (EC 2.3.1.2 1 ), an enzyme required for the entry of long-chain acyl-CoA into the mitechondrion for oxidation [lO,l l]. The decrease in malonyl-CoA is associated with ambition of lipogenesis [IO]. Vasopressin was demonstrated to inhibit ketogenesis in hepatocytes from fed rats when oleate is the added substrate [8]. However, vasopressin does not stimulate lipogenesis [8,12]. This suggests that the antiketogenic action of vasopressin does not result from an increase in the malonyl-CoA concentration. The ~tiketo~enic effect might in part result from the increased conversion of oleoyl-CoA into esterified products [8] and consequently less oxidation of oleoyl-CoA to acetyl-CoA. A possible effect of vasopressin on the complete oxidation of oleate was not investigated in [8]. This communication indicates that there is an intramitochondrial site of action of vasopressin, namely increased oxidation of [ 1 -r4C] oleate to CO,. This action of vasopresain is dependent on Ca2+ and may contribute to the antiketogenic action of the hormone.

Tumor growth and lipid metabolism during lactation in the rat

Implantation of the Walker 256 carcinoma in lactating rats 2-3 days after parturition had no effect on maternal food intake or pup weight gain over the next 8-9 days. The rate of mammary gland lipogenesis in vivo, which is an index of glucose utilization by the gland, was similar in control and post-partum implanted rats. The accumulation of 14C-lipid in the mammary tissue after an oral load of [1-14C]triolein was also not altered by the presence of the tumor, nor was there evidence for hypertriglyceridaemia. This suggests that the activity of lipoprotein lipase in mammary tissue is not sensitive to the tumor as it appears to be in adipose tissue of non-lactating rats. In contrast, implantation of the tumor 1-2 days before parturition resulted in a faster rate of tumor growth, decreased maternal food intake and decreased pup weight gain compared to either control rats or rats with tumor implanted post-partum. In addition, the rate of mammary gland lipogenesis was decreased by 70% and that of the carcass by 50%. This decrease in lipogenesis is likely to be due to the relative hypophagia in the pre-partum implanted group. The 14C-lipid accumulation in mammary tissue after oral [1-14C]triolein tended to be lower in the pre-partum group but this was not statistically significant. It is concluded that the marked effects on lactation of pre-partum implantation of the tumor are due to effects of the tumor or its presence on the differentiation of the gland around parturition. The alternative explanation that the pre-partum tumor implantation suppresses the stimulus for physiological hyperphagia during lactation is less likely, because this does not occur with the post-partum implantation. The role of putative humoral factors in these effects of the Walker 256 carcinoma in lactation is discussed.

Glycerol metabolism in severe falciparum malaria

Gluconeogenesis and liver blood flow (LBF) in severe falciparum malaria were assessed from the clearance and metabolic response to intravenously administered glycerol (0.3 g/kg) and Indocyanine Green ([ICG] 0.4 mg/kg), respectively. Fasting baseline blood glycerol concentrations (mean +/- SD) were significantly higher in acute malaria (133 +/- 65 mumol/L, n = 14), than in convalescence (65 +/- 31 mumol/L, n = 9, P = .01), but basal triacylglycerol concentrations were similar. Estimated glycerol turnover was also more than twice as high in acute malaria compared with convalescence (1.36 +/- 0.87 v 0.54 +/- 0.15 mumol.min-1.kg-1, P = .015). The increment in plasma glucose (AUC0-55 min) following glycerol infusion was greater during acute malaria compared with convalescence (median [range], +31.6 [-0.9 to +107.6] v +14.5 [-103 to +27.1] mmol.min-L-1, P < .05), but the insulin increments were similar (P = .9), indicating reduced tissue insulin sensitivity. The increment in venous lactate (AUC0-55 min) was higher in severely ill patients (17.2 [-7.8 to +53.4] mmol.min.L-1, n = 10) compared with patients with moderately severe malaria (-3.1 [-8.7 to 3.2] mmol.min-L-1, n = 4, P = .01). LBF estimated from ICG clearance was lower during acute illness than in convalescence (mean +/- SD, 15.5 +/- 2.3 v 18.6 +/- 2.9 mL.min-1.kg-1, P = .007) and correlated inversely with the basal venous lactate concentration (rs = .53, P < .05). LBFs less than 15 mL.min-1.kg-1 were associated with hyperlactatemia, and all four fatal cases had LBFs of less than 12 mL.min-1.kg-1.(ABSTRACT TRUNCATED AT 250 WORDS)

The utilization of ketone bodies by the interscapular brown adipose tissue of the rat

The activities of 3-oxo acid-CoA transferase (EC 2.8.3.5, 13-15 micromol/min per g) and acetoacetyl-CoA thiolase (EC 2.3.1.9, 18-21 micromol/min per g) in interscapular brown adipose tissue of the rat are comparable to the activities reported for heart and kidney. The incorporation of D-3-hydroxy[3-14C]butyrate into lipid in vivo was about 30-fold higher in interscapular brown adipose tissue than in white adipose tissue of virgin rats. In lactating rats, the mammary gland was the major site of ketone body incorporation into lipid and incorporation of D-3-hydroxy-[3-14C]butyrate into lipid in brown adipose tissue was lower than in virgin rats. After an oral load of medium chain triacylglycerol, which inhibits lipogenesis in lactating mammary gland, the incorporation of ketone bodies into lipid was decreased in mammary gland but increased in brown adipose tissue. The rate of oxidation of D-3-hydroxy[3-14C]butyrate by brown adipose tissue slices in vitro was higher than the rate of incorporation into lipid.

Short-term dietary regulation of lipogenesis in the lactating mammary gland of the rat

Short-term (6 hr) withdrawal of chow diet from lactating rats decreases the rate of lipogenesis in mammary gland by 87%. This inhibition is in part explained by a 60% decrease in the extraction of glucose (the major lipogenic precursor) by the mammary tissue. These changes are not accompanied by any significant alteration in the arterial concentrations of glucose, lactate or insulin; the concentration of acetoacetate did increase by about 30%. Removal of food for 6 hr did not alter the activation state of acetyl-CoA carboxylase or the total activity of the enzyme. Glucose utilization by mammary gland acini from short-term starved rats was not depressed although a higher proportion of the glucose appeared as lactate in the medium and consequently less glucose was converted to lipid. Insulin was able to reverse these changes. Glucagon, adrenaline or cAMP did not inhibit glucose utilization or lipogenesis in isolated acini. It is concluded that the inhibition of lipogenesis in mammary gland after short-term withdrawal of food is mainly due to decreased extraction of glucose. The signal for this change does not appear to be an alteration in plasma insulin and it is postulated that there may be an intestinal factor(s) which acts synergistically with insulin.

Effects of starvation, insulin or prolactin deficiency on the activity of acetyl-CoA carboxylase in mammary gland and liver of lactating rats

During lactation in the rat the rate of lipogenesis in vivo in mammary gland is several-fold higher than that in liver and adipose tissue and it alters with the nutritional and hormonal state of the animal [1 ]. Starvation (24 h) decreases the rate in the lactating gland by 98% and this is reversed by refeeding for 2 h [1,2] or by injection of insulin [2]. Conversely, shortterm insulin deficiency induced with streptozotocin [3] inhibits mammary gland lipogenesis [1 ]. These changes in the rates of mammary gland lipogenesis correlate with the alterations in the activation state of pyruvate dehydrogenase in vivo [4-7]. In starvation, inactivation of pyruvate dehydrogenase appears to be the major factor in the control of mammary gland lipogenesis, but there is evidence for an insulin-sensitive step which is after the formation of acetyl-CoA [21. In the other lipogenic tissues (adipose tissue and liver) it is now well established that acetyl-CoA carboxylase is a regulatory enzyme and that its activity can be altered by hormones [8]. In epididymal fat pads, unlike liver, there appears to be coordinate control between the activation state of pyruvate dehydrogenase and acetyl-CoA carboxylase in response to alterations of the plasma insulin concentration [9]. Phosphorylation and inactivation of purified acetyl-Co A carboxylase from lactating rabbit [ 10] and rat mammary gland [11] by endogenous cAMPdependent and independent protein kinases has been demonstrated. Dephosphorylation of the acetyl-CoA