Liraglutide acutely suppresses glucagon, lipolysis and ketogenesis in type 1 diabetes

In view of the occurrence of diabetic ketoacidosis associated with the use of sodium-glucose transport protein-2 inhibitors in patients with type 1 diabetes (T1DM) and the relative absence of this complication in patients treated with liraglutide in spite of reductions in insulin doses, we investigated the effect of liraglutide on ketogenesis. Twenty-six patients with inadequately controlled T1DM were randomly divided into 2 groups of 13 patients each. After an overnight fast, patients were injected, subcutaneously, with either liraglutide 1.8 mg or with placebo. They were maintained on their basal insulin infusion and were followed up in our clinical research unit for 5 hours. The patients injected with placebo maintained their glucose and glucagon concentrations without an increase, but there was a significant increase in free fatty acids (FFA), acetoacetate and β-hydoxybutyrate concentrations. In contrast, liraglutide significantly reduced the increase in FFA, and totally prevented the increase in acetoacetate and β-hydroxybutyrate concentrations while suppressing glucagon and ghrelin concentrations. Thus, a single dose of liraglutide is acutely inhibitory to ketogenesis.

Cytokine-mediated inhibition of ketogenesis is unrelated to nitric oxide or protein synthesis

Cytokines play an important role in the lipid disturbances commonly associated with sepsis. Ketogenesis is inhibited during sepsis, and tumor necrosis factor alpha (TNF alpha) and interleukin-6 (IL-6) have been suggested to mediate this impairment, irrespective of the ketogenic substrate (fatty acid or branched chain ketoacid). However, the underlying mechanism of cytokine action is still unknown. First we investigated the possible role of the induction of nitric oxide (NO) synthesis, using rat hepatocyte monolayers. Hepatocytes were incubated for 6 h, with either alpha -ketoisocaproate (KIC) (1 mM) or oleic acid (0.5 mM) in the presence or absence of TNF alpha (25 microg/L) and IL-6 (15 microg/L). In some experiments, cells were incubated with NO synthase (NOS) inhibitors. The ketone body (beta -hydroxybutyrate and acetoacetate) production and nitrite production were measured in the incubation medium. Our results indicated no involvement of nitric oxide in the inhibitory action of cytokines on ketogenesis. Secondly, we showed that cycloheximide (10(-4)M) did not counteract the cytokine-mediated ketogenesis decrease; hence, the effects of cytokines on ketogenesis are not protein synthesis-dependent. The cytokine-mediated inhibition of ketogenesis is therefore unrelated to either NO production or protein synthesis.

Role of fat-derived substrates in the regulation of gluconeogenesis during fasting

To determine the role of fat-derived substrates in the regulation of glucose metabolism during fasting, glucose turnover, urea nitrogen production, alanine conversion to glucose, and substrate oxidation rates were measured in 34 normal 4-day-fasted volunteers treated with the antilipolytic drug acipimox or placebo for 8 h. The approximately 50% inhibition of lipolysis induced by acipimox increased glucose concentration and production, respectively, by approximately 35 and approximately 30%, whereas the protein breakdown and the amount of alanine converted to glucose were increased, respectively, by approximately 70 and approximately 85%. Insulin levels were reduced by approximately 40%, cortisol levels doubled, and growth hormone concentration increased sevenfold. The relative contribution of free fatty acid (FFA) and ketone body lowering to the observed response was evaluated in nine acipimox-treated subjects in whom ketone body concentration was clamped with an intravenous beta-hydroxybutyrate infusion. The results of these experiments suggest that, during fasting, both FFA and ketone bodies tend to suppress gluconceogenesis and to protect the protein stores. FFA seem to exert their effects mainly through their ability to modulate the hormonal milieu (especially insulin), whereas ketone bodies seem to act mainly by other mechanisms. Thus the widespread view according to which FFA exert a stimulatory role on gluconeogenesis does not apply to the fasting state in vivo.

Inhibition of hepatic ketogenesis by tumor necrosis factor-alpha in rats

Tumor necrosis factor-alpha (TNF-alpha) stimulates hepatic lipogenesis. Therefore, it could play a role in the control of ketogenesis. To test this hypothesis, we measured simultaneously free fatty acids (FFA; [1-13C]palmitate) and ketone body (KB; [3,4-13C2]acetoacetate) kinetics, before and after intraperitoneal injection of saline or TNF-alpha, in postabsorptive rats or rats starved for 24 h. In both groups of rats, TNF-alpha injection did not modify insulinemia and induced a moderate increase of FFA concentrations and appearance rates (P < 0.05). Despite increased FFA availability, ketogenesis was impaired after TNF-alpha injection, as shown by lower KB concentrations and appearance rates; this effect was more important in postabsorptive than in starved rats. The percentage of FFA flux used for ketogenesis was decreased by TNF-alpha in the postabsorptive group (P < 0.05) and starved (P < 0.05) rats. In both groups, maximal liver acetyl-coenzyme A carboxylase activity and estimated phosphorylation state were not modified by TNF-alpha injection, but hepatic concentrations of citrate were increased (P < 0.05). This increased citrate level could be related to a mobilization of glucose stored as glycogen since liver glycogen was decreased by TNF-alpha injection (P < 0.05). In conclusion, TNF-alpha injection in rats decreased hepatic ketogenesis. This action could be related to an increased mobilization and utilization of carbohydrate stores.

Differential effects of interleukin-1 and tumor necrosis factor on ketogenesis

To determine the role of cytokines in mediating the decrease in ketones associated with infection, we studied the effect of endotoxin (LPS), interleukin-1 (IL-1), and tumor necrosis factor (TNF) on serum and hepatic ketone body levels (KB), serum free fatty acids (FFA), and hepatic malonyl-CoA levels. LPS decreased serum and hepatic KB in C57Bl/6 (LPS sensitive) mice, whereas it had little effect in C3H/HeJ (LPS resistant) mice, whose macrophages lack the ability to produce IL-1 and TNF in response to LPS, suggesting that IL-1 and TNF may mediate this effect. IL-1 and TNF decreased serum KB in both strains of mice. As seen with LPS, IL-1 decreased hepatic KB, whereas TNF had no such effect. LPS, IL-1, and TNF increased hepatic malonyl-CoA levels. TNF acutely raised serum FFA, whereas LPS and IL-1 did not. Postulating that the TNF-induced increase in FFA overrides the inhibitory effect of malonyl-CoA on fatty acid oxidation and ketogenesis, we used R-2-phenylisopropyladenosine to block TNF-induced lipolysis and demonstrated that in the absence of increased fatty acid flux, TNF inhibits KB formation. As seen with LPS, IL-1, but not TNF, decreased KB in the fasting state. These data suggest that IL-1 and TNF may mediate the antiketogenic effect of infection and that IL-1 has properties closest to that of LPS.

[Potentiation of the antiketogenic effect of exogenous glucose with the help of the new beta-oxidation inhibitor mildronate--3-(2,2,2-trimethylhydrazine)propionate]

Antiketogenic effect of exogenous glucose (2 g/kg, per os, 1 hr before death) was potentiated after preadministration of mildronate 3-(2,2,2-trimethylhydrazinium) propionate into rats either kept on usual ration or fasting within 48 hrs at a dose of 200 and 400 mg/kg, per os, during 10 days. Mildronate is inhibitor of carnitine-dependent metabolism of fatty acids affecting at the step of gamma-butyrobetaine turnover into carnitine. The drug inhibitory influence studied appears to be realized via activation of the glycolytic pathway of glucose metabolism specific for inhibitors of beta-oxidation.

Effect of epinephrine on the relationship between nonesterified fatty acid availability and ketone body production in postabsorptive man: evidence for a hepatic antiketogenic effect of epinephrine

The effect of epinephrine (EPI) on the transformation of nonesterified fatty acids (NEFA) into ketone bodies (KB) in normal subjects was determined by measuring simultaneously NEFA ([1-13C]palmitic acid) and KB ([3-13C]- or [3,4-13C2]acetoacetate) kinetics at different NEFA levels in the presence of basal (control test) or increased (EPI infusion test) EPI concentrations. During the control test the initial (postabsorptive state) concentrations and turnover rates of NEFA and KB were 476 +/- 47 (+/- SEM) and 4.30 +/- 0.17 mumol kg-1 min-1 (NEFA) and 126 +/- 17 and 2.49 +/- 0.07 mumol kg-1 min-1 (KB). The fraction of NEFA converted into KB was between 11.5-14.6%. Raising NEFA levels to about 650 mumol L-1 (iv infusion of a triglyceride emulsion) resulted in an increase in this fraction to between 26-30.3% (P less than 0.01). When NEFA concentrations were next abruptly raised to high levels (near 3 mmol L-1) by heparin injection this fraction returned to near the initial values (15-19.2%). During the EPI infusion test the initial (postabsorptive) concentrations and turnover rates of NEFA and KB as well as the fraction of NEFA converted into KB (10.5-11.5%) were comparable to the initial values of the control test. Intravenous infusion of EPI (10 ng kg-1 min-1) raised NEFA between 600 and 750 mumol L-1, comparable to values during the triglyceride test, but the fraction of NEFA converted into KB remained between 8.2-12% (P less than 0.05 vs. control test); when NEFA then were raised to even higher values (near 2.5 mmol L-1) by the infusion of a triglyceride emulsion and the injection of heparin, this fraction decreased to between 4-8% (P less than 0.05 vs. initial values of the EPI test and P less than 0.05 vs. the control test). In conclusion, 1) the fraction of NEFA converted into KB appears to depend in part on the NEFA concentration; and 2) the net effect of EPI infusion was to decrease the fraction of NEFA converted into KB.

Antiketogenic action of fructose, glyceraldehyde, and sorbitol in the rat in vivo

The purpose of this study was to compare the metabolism and antiketogenic properties of fructose, glyceraldehyde, and sorbitol. Fructose, glyceraldehyde, and sorbitol were readily metabolized and exhibited an antiketogenic effect in both blood and liver when injected intramuscularly to starved (forty-eight hours) rats. Sorbitol had the most pronounced antiketogenic effect and produced an 80 to 90 per cent decrease in the blood ketone bodies sixty minutes after administration. Fructose and glyceraldehyde were equally effective and produced about a 60 to 70 per cent decrease in ketone bodies. Fructose, glyceraldehyde, and sorbitol caused a significant decrease in the concentration of hepatic ketone bodies. In liver, sorbitol was found to be most effective in its antiketogenic action. The concentration of plasma free fatty acids remained unchanged after injection of all three antiketogenic substrates. Fructose, glyceraldehyde, or sorbitol caused increased blood lactate and pyruvate concentrations, and fructose was the most effective of the three substrates. Fructose administration resulted in a significant decrease in hepatic lactate/pyruvate and beta-OH-butyrate/acetoacetate concentration ratios, whereas sorbitol caused an increase in the concentration ratio of these two substrat pairs. Decreases in blood and liver ketone body levels were associated with lowering of liver acetyl-CoA concentration . However, the decrease in hepatic acetyl-CoA produced upon the administration of antiketogenic substrates was not pronounced. Sorbitol administration resulted in the most pronounced increase in hepatic alpha-glycerophosphate concentration. Fructose or glyceraldehyde also caused an increase in alpha-glycerophosphate content. Administration of each of the three antiketogenic substrates produced an increase in hepatic dihydroxyacetone phosphate concentration. All three antiketogenic compounds increased liver glycogen and blood glucose concentrations. No significant changes were observed in hepatic ATP, ADP, or AMP concentrations sixty minutes after the injections of any of the antiketogenic substrates. Although decreased liver acetyl-CoA levels were associated with the antiketogenic effects of the compounds tested, the increased liver alpha-glycerophosphate content best explains the differences between fructose or glyceraldehyde and sorbitol.