- Email: [email protected]
Inhibition of ketogenesis by ketone bodies in fasting humans
Metabolism Clinical and Experimental VOL. XXIV,
NO. 9
SEPTEMBER
1975
Inhibition of Ketogenesis by Ketone Bodies in Fasting Humans E. 0. Balasse and M. A. ,Neef Although there exists some indirect evidence that circulating ketone bodies might inhibit their own production rate, the direct demonstration of this homeostotic feed-bock phenomenon is still locking. The present work aims ot demonstroting the opemtion of this control mechanism in human fasting ketosis. Six obese subiects, who fasted 2-23 days, were given o primed constant i.v. infusion of 3-“C-ocetoacetate for 4 hr. After o control period of 2 hr, unlabeled sodium ocetoacetate was administered 9s o primed constant i.v. infusion at the rote of 0.688-l .960 mmol/min until the end of the study. During both periods, the mtes of inflow of ketones were estimated from the specific activity of totol ketones measured under neor isotopic steody state conditions. During the control period, total ketone concentration amounted to 3.98-9.65 pmol/ml and production rates of total ketones ranged between 1.450 and 2.053 mmol/min. The levels of free fotty acids, glycerol, glucose, and insulin averoged
respectively 1.30 pmol/ml, 0.11 fimol/ml, 74 mg/lOO ml, and 5.2 &t/ml. The administration of exogenous ketones during the second phase of the study induced o 47%-92% increase in total ketone levels. During this period, the endogenous production of ketones (colculoted OS the difference between total inflow mte and ocetoocetote infusion mte) omounted only to 67%-90% of control values. Among other factors, this inhibition of ketogenesis wos probably portiolly related to the direct ontilipolytic effect of infused ketones. Indeed, there was o concomitant fall in FFA ond in glycerol levels averaging respectively 13.5% and 17.3%, without significant changes in peripheml insulin concentrations. Our results demonstrott thot during fosting, circulating ketone bodies exert on inhibitory influence on the rote of ketogenesis. This mechanism might play an important role in preventing the development of uncontrolled hyperketonemio during starvation.
From the Metabolic Unit and the Laboratory of Experimental Medicine, University of Brussels, Brussels, Belgium. Receivedfor publication April 21, 1975. Presented in part at the 10th Anntcal Meeting of the European Association for the Study of Diabetes held in Jerusalem, Israel, September 1974. Supported by the Fends de la Recherche Scientifique Medicale (grant 20193) and by the Minis&e Belge de la Politique Scientifique. within the framework of the Association Euratom, University of Brussels, University of Pisa. Reprint requests should be addressed to E. 0. Balasse. M. D.. Laboratory of Experimental Medicine, University of Brussels. Boulevard de Waterloo 115. B-1000 Brussels, Belgium. o 1975 by Grune & Stratton, Inc. Mdobolirm, Vol. 24, No. 9 (September),
1975
999
BALASSE
1000
AND
NEEF
S
INCE approximately 10 years, it has been repeatedly observed that the infusion of ketone bodies reduce plasma FFA levels in various animal species including man. ‘-I2 Since plasma FFA are the primary ketogenic substrate, it has been speculated that the ketone-induced inhibition of adipose tissue lipolysis might restrain ketogenesis.‘,6-9J2 If this were true, this feed-back mechanism should play an important homeostatic role during starvation by preventing the development of uncontrolled hyperketonemia. However, direct evidence for the operation of this feed-back cycle has never been provided. The present experiments demonstrate that in fasting man, the rate of ketogenesis, as measured by an isotopic technique, is indeed inhibited by an exogenous supply of ketones.
MATERIALS
AND METHODS
The study was performed on 6 obese nondiabetic female subjects aged 17-46 yr, hospitalized for therapeutic starvation. Their body weight varied between 115% and 164% of ideal body weight at the time of the study. During the fast, the subjects had a daily intake of one multivitamin capsule (Maxivit, Roerig, Belgium) and were allowed to drink calorie-free beverages (water, black coffee, or tea) ad libitum until 12 hr before the test. Only water was allowed thereafter. The experiments were performed after various durations of fast ranging from 2 to 23 days. An indwelling Cournand needle was placed in a brachial artery for blood sampling and two venous cannulae were inserted in the opposite arm for infusions. The studies were conducted according to the following protocol (Fig. 1). After a first arterial blood specimen had been obtained, sodium 3-‘4C-acetoacetate was infused at a constant rate (0.41-0.57 &/min) for 4 hr. A priming dose of label corresponding to the amount infused in 80 min was given at the start of the infusion. After a basal period of 2 hr. a primed constant infusion of unlabeled sodium acetoacetate (AcAc) was superimposed to the isotope infusion. The rate of delivery ranged between 0.688 and 1.960 mmol/min, the priming dose corresponding to 80 times the amount infused per min. Four 25 ml blood samples were collected at 15 min intervals during the last 45 min of each period and were placed in heparinized tubes and immediately chilled. Samples of 5 ml blood were mixed with perchloric acid and neutralized I3 and the resulting protein-free filtrate was kept on ice until further use. The rest of the blood was centrifuged and the plasma was analyzed for content of FFA,14 glucose,‘5*‘6 and immuno-reactive insulin (IRI).” Glycerol, AcAc, and @-hydroxybutyrate (BOHB) concentrations were estimated in the perchloric filtrate using enzymatic methods.‘8Y’9 14C in ketone bodies was measured by the method of Mayes and Felts.” This method permits the determination of 14C activity separately in AcAc and /3OHB; “C-@GHB is converted enzymatically to 14C-AcAc; 14C-AcAc is decarboxylated to acetone and CO2 which are trapped separately in a double-well flask and assayed for 14C in a liquid scintillation spectrometer. This original method was slightly modified along two lines: (1) since the AcAc used in these experiments was labeled on the carbonyl carbon, only the labeled acetone produced was measured; (2) the procedure was adapted to handle larger volumes of perchloric filtrate, so that ultimately the radioactivity measured corresponded to that contained in 0.5 ml of blood instead of 0.050.06 ml of blood for the original method. The recovery of 14C-AcAc and DL-‘4C-/30HB added to nonradioactive blood (obtained from the patient at the start of the study) was determined with each set of assays and found to be 63x-89% (mean 77%) for AcAc and 76x-89% (mean 81%) for @GHB. Recoveries were consistent within an experiment. For BOHB, recovery was calculated taking into account the fact that the assay is specific for the D(-) isomer so that only half of the added 14C was expected to be recovered. Values for ketone radioactivity in blood samples were corrected accordingly. Any “C-acetone which could be present in the perchloric filtrate before determination of 3-14C-AcAc and 3-‘4C-mHB, was removed by prior ventilation of the filtrate with a stream of nitrogen for at least 1 hr.” The determinations of unlabeled and labeled AcAc were done on the day of each experiment owing to the instability of the compound.
KETOGENESIS
12
-
10 -
6 LOOD KETONES
a-
pm01 /ml 842o1100 KETONE SPECIFIC
900
ACTIVITY dpm /p mol
700
*c*c
+
500 300
INFLOW RATE
OOHB
OOHB
1
l-•-•-• o-
=
0-Q
/
F, I ‘. I I
‘\ ‘.
‘m
-____-___v
----_--_-_--_-_*-*--,~
---_-_________o_*
3
OF 2
TOTAL KETONES mmol/min
1 0 1
PLASMA
1.4
/-+..._
FFA pmol /ml
1.2
I 1
PLASMA
75
GLUCOSE mg /lOOml
65
1 , 0
Fig. 1.
Average
experimental
=--
.
I I
1.0 1
---f------
------k
I
I
I
I1
O”
mrulk
1
8,
MINUTES
120
..__/“/
’
-
!
190
-
1
I
I
240
in six subjects.
Blood pH was measured at the end of the control period and at the end of the AcAc infusion period, in four of the six patients studied. All analyses were performed in duplicate. Prepumfion ofmarerials for injecfion. The unlabeled and the labeled sodium AcAc were prepared, respectively, from unlabeled ethyl AcAc (Merck, Darmstadt, Germany) and chromatographically pure ethyl 3-t4C-AcAc (The Radiochemical Center, Amersham, England, specific
BALASSE
1002
AND NEEF
activity 12.7 mCi/mmol), according to previously described techniques.‘3 The radiochemical purity of the prepared 3-‘4C-AcAc solutions was determined according to Mayes and Felts” and found to vary between 93% and 97%. The radioactivity was mixed with a small volume (2-3 ml) of 0.5 M AcAc in order to minimize its degradation.” After preparation, the solutions were sterilized through a Millipore filter (0.22 p) and kept frozen at -2O’C until use. Immediately before use, the unlabeled AcAc was diluted with distilled water so as to obtain a concentration varying between 222 and 631 pmol/ml and the labeled AcAc was diluted with s/aline. All infused solutions had a pH of 7-7.5 and were kept chilled with ice during the course of the infusions. Calcularions. During the last 45 min of both phases of the experiments, the specific activity (S.A.) of total ketone bodies was calculated for each time point as follows: S.A. of total ketones (dpmlpmol) = [14C-AcAc (dpm/ml) + “C-fiOHB (dpm/ml)]/[AcAc (pmol/ml) + j3OHB
OtmoWt)l (I). Although on an average, the S.A. of total ketones had almost attained the steady state during both periods (Fig. l), a significant rise was nevertheless observed in several experiments. Therefore, for both periods of each experiment, the reduction rate of total ketones was calculated according to the equation proposed by Steele 3? which allows for correction for the change in S.A. with time: R, = [I - (c x V x dS.A./dr)]/[Sx.] (II) where R, is the rate of appearance (or inflow) of ketones (pmol/min), I is the infusion rate of the tracer (dpm/min), c is the average concentration of total ketones over the studied period (rmol/ml), V is the “operational” volume of distribution of ketones (ml) which was shown to approximate 200/, of body weight in fasted subjects,” dS.A./df represents the increment in S.A. as a function of time and was calculated by linear regression using the four values obtained during the last 45 min of each period, and S.A. is the average S.A. of total ketones over the studied period (dpm/pmol). During the basal period, only the isotope was infused and the rate of appearance of total ketones thus represents the rate of ketogenesis. During the second phase, the rate of appearance of total ketones represents the sum of the exogenous AcAc inflow and of the endogenous ketone production. The latter was thus obtained by subtracting the exogenous AcAc inflow from the rate of appearance of total ketones. The rationale for using the S.A. of total ketones instead of that of AcAc in calculating the inflow rate of ketones under our experimental conditions will be outlined under the discussion section. Statistical significance of the observed changes was tested by paired comparison (Student’s t test) of the mean data obtained in each subject during the two experimental periods. RESULTS
During the basal period, the AcAc and POHB concentrations were steady in all subjects (Fig. 1) and total ketone concentrations ranged between 3.98 and 9.65 rmol/ml (Table 1). POHB levels largely exceeded those of AcAc, the pOHB/AcAc ratio varying between 3.23 and 4.05. The S.A. of ketone bodies almost plateaued during the last 45 min of the basal period but the S.A. of /3OHB did not exceed 41%-52x of that of AcAc. Ketogenesis calculated according to equation II varied between 1.450 and 2.053 mmol/min and seemed roughly independent of the duration of the fast, at least for the studied fasting periods. The administration of unlabeled AcAc (at rates equivalent to 43%-98% of basal ketone production) induced a marked rise in both the AcAc and /3OHB concentrations which plateaued during the last 45 min of the infusion (Fig. 1). Total ketone concentrations amounted to 7.41-16.44 pmol/ml during this period (Table 1). The POHB/AcAc ratio was systematically decreased by comparison with the basal period, varying between 1.47 and 2.44. The rise in ketone concentration was associated with a fall in ketone S.A. indicating that total ketone inflow was increased. The difference between total inflow rate and ex-
135
115
164
135
128
DEF.
LEN.
DUF.
DEL.
KAM.
23
18
14
7
5
2
Dllp
6.94 11.02 9.65 16.44
5.41 7.77 7.74 10.85
3.25 1.91 5.59
0.688 0
1.365
1.418
2.053
1.230
1.614
1.345
13.78
8.19
1.53
7.15
5.59
1.250 1.993
8.05
1.450
5.58
0
1.960
5.17
1.57
2.88
0.784 0
5.43
4.17
1.27
0
1.316
1.836
7.30 10.80
5.99 7.66
1.31 3.14
0
0.904
1.436
7.41
4.93
2.48
0.998
1.588
mmol/min
3.98
3.04
Blood’
0.94
pmol/ml
0
mmoljmin
31
24
33
14
28
10
Control
1.80
1.94
0.146
0.153
0.080
0.099 0.95
0.090
1.03
0.140
0.068
0.070
0.069
0.074
0.071
0.108
Bbod
Irmol/ml’
Glycerol
0.95
1.35
0.88
1.01
1.07
1.36
0.96
1.16
ml Plasma
pmol/’
FFA
82
87
77
81
72
78
63
67
62
61
5.0
4.8
4.0
5.5
6.1
13.0
3.0
3.0
1.5
1.8
3.0
3.0 59
PklSfllO
fiU/ml*
68
ml’
IRI
Plasma
mg/lW
GIUCOW
the AcAc infusion period.
each subject: upper line: mean of four values obtained during the last 45 min of the control period, lower line: mean of four values obtained during the last 45 min of
126
KRU.
*For
Weight
Subject
From
Endoganous ProductionR&e of Total Ketones
R-‘y
AcAc
BOHR
AcAc
Percent Inhibition
Infused
ACAC
Percent of Id&
Duration of Fast
Ketone Bodies Production Rates in Fasted Obese Subiects
Table 1. Effect of an Infusion of Sodium Acetoacetate on Blood Metabolite Concentration and
1004
BALASSE
AND NEEF
ogenous AcAc infusion rate provides a measurement of the residual rate of ketogenesis which was lo%-33% lower than that observed during the control period (p < 0.005). This inhibition of ketogenesis by ketone bodies was observed whatever the duration of the fast. The infusion of AcAc inhibited adipose tissue lipolysis as evidenced by a significant fall in the average concentration of FFA (13.5% f 3.6%; p < 0.02) and of glycerol (17.3 f 6.0%; p < 0.05). It should be noticed that, on an average, FFA (Fig. 1) and glycerol levels tended to return towards basal value at the end of the infusion period. In fact, this trend was apparent in only three of the six subjects tested, whereas the other subjects maintained their lowered FFA levels until the end of the ketone infusion. The degree of inhibition of ketogenesis was not significantly related to the fall in FFA levels (p > 0.1). A lowering of blood glucose was noticed (6% + 2%; p < 0.05) but there was no significant change in arterial IRI levels (p > 0.05). In four of the six subjects studied, the infusion of sodium AcAc induced only minimal changes in arterial pH, its mean value rising from 7.36 to 7.40. DISCUSSION
The estimation of ketone bodies production rate using a constant infusion of 14C-AcAc meets with difficulties owing to the absence of isotopic equilibration between blood AcAc and POHB. This situation has been observed by others in ratsz2s2’ and in man.” The liver is believed to be the main site of interconversion between ketones and McGarry et al.22 and Batesz3 provided strong evidence that the disequilibrium found in the isotopic steady state is due to a limited capacity of the liver to carry out the equilibration of infused isotope between AcAc and POHB. Under these conditions, the S.A. of total ketones is assumed to represent the S.A. that would have been attained for both ketones if hepatic equilibration had not been limiting. The best estimate of total ketone body turnover would appear, therefore, to be derived from total ketone body S.A. and this technique of calculation has been employed by McGarry et al. in rats’* and by Reichard et al. in humans.2’ During the basal period, the plasma (blood) concentrations of FFA, glycerol, and glucose were in the accepted normal range for subjects starved 2-23 days but ketone levels tended to be slightly higher than those observed by others.24,25 The rates of production of ketones (1.450-2.053 mmol/min) were significantly higher than those published by Reichard et al. (0.674-1.553 mmol/min) although these authors used an isotopic technique similar to ours.*’ The reasons for these differences are not obvious. Most of the obese subjects studied by this latter group 2’.24,25had a much greater excess body weight2’~24*25and had higher IRI levels24*z5than those studied here and this might account, at least partly, for the observed differences. It should be mentioned that in another group of eight obese subjects fasted 3-23 days studied in our laboratory, rates of ketogenesis measured using either constant infusions or single injections of 14C-AcAc, were of the same order of magnitude (1.670-2.946 mmol/min) as those observed here. Whatever mechanism may be responsible for the higher ketone turnover rates found in our subjects, it should not interfere with the main significance of this work which is based on the comparison of rates of production obtained in the same subject under two different experimental conditions.
KETOGENESIS
1005
The administration of AcAc induced a rise in ketonemia and a significant and sustained drop in ketone production rate (Table 1). This effect was observed whatever the duration of the fast. The decrease in ketone production rate is likely to result, at least partly, from the observed fall in the levels of FFA which are known to be the major ketogenic substrate. Although an insulinotropic effect of ketone bodies can undoubtedly be observed in man under apthe fall in FFA found in our experiments cannot propriate conditions, ‘“~1’*26,27 be explained on this basis, peripheral IRI levels remaining steady or even decreasing in some patients during the AcAc infusion. Under these conditions, the drop in FFA was probably related to the direct antilipolytic properties of ketone bodies first demonstrated in vitro by Bjorntorp2’ and confirmed by many others.*‘” A fall in FFA levels independent of insulin release following ketone infusions in man has been previously described.7,8J2 The participation of a ketone-induced inhibition of glucagon secretion32333 in the observed antilipolytic effect of ketones cannot be ruled out in these experiments. The antiketogenic effect of AcAc infusions can probably not be explained solely on the basis of their antilipolytic action for the two following reasons: (1) in three of the six subjects tested, the antilipolytic effect was transient although ketogenesis remained depressed until the end of the ketone infusion; (2) there was no correlation between the fall in FFA levels and the corresponding degree of inhibition of ketone production; for instance, the marked decrease in ketogenesis observed in subjects DEL and KAM (Table 1) was not associated with a significant drop in FFA levels. Additional mechanism(s) should therefore be considered in order to explain the observed reduction in ketogenesis; one possibility is that, despite the absence of IRI changes in peripheral blood, the ketone infusion could have stimulated insulin release in the portal vein2’ thereby inhibiting hepatic ketogenesis independently of changes in the amounts of FFA delivered to the liver.” The observed hypoglycemic effect of ketones is confirmatory of many other studies.7-9~‘2*27The data of the literature indicate that this effect may be accompanied, but need not depend upon the release of insulin.7p8*‘2 The influence of exogenous AcAc on endogenous ketone production has been studied before by Bergman et al.” in sheep and by Bates23 in rats. The first group of authors found comparable rates of endogenous AcAc production in normal and AcAc infused sheep. Their experimental design has the disadvantage of comparing two different groups of animals whereas our subjects served as their own control in a single experiment. Bates noted that infused AcAc inhibits ketogenesis but she thought that the calculated reduction in endogenous appearance of AcAc could result from a methodological artefact.23 In conclusion, these data represent the first direct demonstration that circulating ketones exert a negative feedback on the rate of ketogenesis in fasting man. This effect is probably partially mediated through the direct antilipolytic action of ketones on adipose tissue, but our data suggest that it is not the only involved factor. The physiological implications and teleological significance of such a mechanism in maintaining ketone homeostasis during starvation are obvious and have already been emphasized by several authors including US.‘*~-~J*,~~ The demonstration that infused ketone bodies reduce FFA levels in pancreatectomized dogs* and in insulin-dependent diabetic humans” suggest
BALASSE
1006
AND
NEEF
that diabetic ketosis is another situation in which ketone bodies tend to limit their own production rate. It is clear, however, that in this case, the feed-back control mechanism is insufficient to maintain ketone bodies at concentrations compatible with life. ACKNOWLEDGMENT We gratefully
acknowledge Mrs D. Calay,
nurse of the Metabolic Unit, and Dr. E. Van Thiel, Department of Cardiology, for their help in performing these studies. Thanks are also due to the technicians of the Department of Cardiology for blood pH determinations, and to Mrs. B. Noel for typing the manuscript.
REFERENCES 1. Madison LL, Mebane D, Unger RH, et al: The hypoglycemic action of ketone. II. Evidence for a stimulatory feedback of ketones on the pancreatic beta cells. J Clin Invest 43: 408415, 1964 2. Bjijrntorp P, Scherstin T: Effect of &hydroxybutyrate on lipid mobilization. Am J Physiol212:683-687, 1967 3. Balasse E, Couturier E, Franckson JRM: Influence of sodium &hydroxybutyrate on glucose and free fatty acid metabolism in normal dogs. Diabetologia 3:488-493, 1967 4. Metzger P, Franken P, Balasse EO: Permissive role of glucose on the insulinotropic effect of ketone bodies in vivo. Horm Metab Res 5:313-315, 1973 5. Balasse EO: Inhibition of free fatty acid oxidation by acetoacetate in normal dogs. Em J Clin Invest 1:155-160, 1970 6. Hawkins RA, Alberti KGMM, Houghton CRS, et al: The effect of acetoacetate on plasma insulin concentration. Biochem J 125: 541-544,197l 7. Balasse E, Ooms HA: Changes in the concentrations of glucose, free- fatty acids, insulin and ketone bodies in the blood during sodium @hydroxybutyrate infusions in man. Diabetologia 4:133-135, 1968 8. Senior B, Loridan L: Direct regulatory effect of ketones on lipolysis and on glucose concentrations in man. Nature 219:83-84, 1968 9. Jenkins DJA: Ketone bodies and the inhibition of free-fatty-acid release. Lancet 2:338340, 1967 IO. Jenkins DJA, Hunter WM, Goff DV: Ketone bodies and evidence for increased insulin secretion. Nature 227:384-385, 1970 I I. Owen OE, Reichard GA Jr, Markus H, et al: Rapid intravenous sodium acetoacetate infusion in man. Metabolic and kinetic responses. J Clin Invest 52:2606-2616, 1973 12. Binkiewicz A, Sadeghi-Nejad A, Hochman H, et al: An effect of ketones on the con-
centrations of glucose and of free fatty acids in man independent of the release of insulin. Pediatrics 84:226-23 1, 1974 13. Balasse EO, Have1 RJ: Evidence for an effect of insulin on the peripheral utilization of ketone bodies in dogs. J Clin Invest 50~801-8 13. 1971 14. Trout DL, Estes EH, Friedberg SJ: Titration of free fatty acids of plasma: A study of current methods and a new modification. J Lipid Res 1:199-202, 1960 15. Hoffman WS: A rapid photoelectric method for the determination of glucose in blood and urine. J Biol Chem 1205-55, 1937 16. Technicon Auto Analyzer Methodology Manual. Chauncey, N.Y ., Technicon Instruments 17. Herbert V, Lau KS, Gottlieb CW, et al: Coated charcoal immunoassay of insulin. J Clin Endocr 25:1375-1384, 1965 18. Chernick SS: Determination of glycerol in acyl-glycerols, in Lowenstein J (ed): Methods in Enzymology, vol 14. New York, Academic Press, 1969, p 627 19. Williamson DH, Mellanby J, Krebs HA: Enzymic determination of D(-)&hydroxybutyric acid and acetoacetic acid in blood. Biochem J 82:90-96, 1962 20. Mayes PA, Felts JM: Determination of “C-labeled ketone bodies by liquid-scintillation counting. Biochem J 102230-235, 1967 21. Reichard GA, Owen OE, HafT AC, et al: Ketone-body production and oxidation in fasting obese humans. J Clin Invest 53:5088515, 1974 22. McGarry JD, Guest MJ, Foster DW: Ketone body metabolism in the ketosis of starvation and alloxan diabetes. J Biol Chem 245:4382-4390, 1970 23. Bates MW: Kinetics of ketone body metabolism in fasted and diabetic rats. Am J Physiol221:984-991, 1971 24. Owen OE, Felig P, Morgan AP, et al:
1007
KETOGENESIS
Liver and kidney
metabolism
during
prolonged
starvation. J Clin Invest 48574-582, 1969 25. Owen OE, Reichard GA Jr: Human forearm metabolism during progressive starvation. J Clin Invest 50~1536-1545, 1971 26. Pi-Sunyer FX, Sethi SS: Stimulation of insulin secretion by sodium beta-hydroxybutyrate in man. Diabetes 21:373, 1972 27. Balasse EO, Ooms HA, Lambilliotte JP: Evidence for a stimulatory effect of ketone bodies on insulin secretion in man. Horm Metab Res 2:371-372, 1970 28. Bjorntorp P: The effect of beta-hydroxybutyric acid on glycerol outflow from adipose tissue in vitro. Metabolism 15:191-193, 1966 29. Devecerski M, Pierce CE, Frawley TF: Effect of ketone acids on glucose and fat metabolism in adipose tissue of the rat. Metabolism 17:877-884, 1968 30. Hellman DE, Senior B, Goodman HM: Antilipolytic effects of p-hydroxybutyrate. Metabolism 18:906-915, 1969 31. Hotta N, Sirek A, Sirek OV: Inhibitory
effect of beta-hydroxybutyrate on lipolysis stimulated by dihydroergotamine and growth hormone in vitro. Can J Physiol Pharmacol 49:87-91, 1971 32. Edwards JC, Taylor KW: Fatty acids and the release of glucagon from isolated guinea-pig islets of Langerhans incubated in vitro. Biochim Biophys Acta 215:310-315, 1970 33. Marliss EB, Wollheim CB, Blonde1 B, et al: Insulin and glucagon release from monolayer cell cultures of pancreas from newborn rats. Eur J Clin Invest 3: 16-26, 1973 34. Bieberdorf FA, Chernick SS, Scow RO: Effect of insulin and acute diabetes on plasma FFA and ketone bodies in the fasting rat. J Clin Invest 49168551693, 1970 35. Bergman EN, Kon K, Katz ML: Quantitative measurements of acetoacetate metabolism and oxidation in sheep. Am J Physiol 205:658-662, 1963 36. Steele R: Influence of glucose loading and of injected insulin on hepatic glucose output. Ann NY Acad Sci 82:420-430, 1959
NO. 9
SEPTEMBER
1975
Inhibition of Ketogenesis by Ketone Bodies in Fasting Humans E. 0. Balasse and M. A. ,Neef Although there exists some indirect evidence that circulating ketone bodies might inhibit their own production rate, the direct demonstration of this homeostotic feed-bock phenomenon is still locking. The present work aims ot demonstroting the opemtion of this control mechanism in human fasting ketosis. Six obese subiects, who fasted 2-23 days, were given o primed constant i.v. infusion of 3-“C-ocetoacetate for 4 hr. After o control period of 2 hr, unlabeled sodium ocetoacetate was administered 9s o primed constant i.v. infusion at the rote of 0.688-l .960 mmol/min until the end of the study. During both periods, the mtes of inflow of ketones were estimated from the specific activity of totol ketones measured under neor isotopic steody state conditions. During the control period, total ketone concentration amounted to 3.98-9.65 pmol/ml and production rates of total ketones ranged between 1.450 and 2.053 mmol/min. The levels of free fotty acids, glycerol, glucose, and insulin averoged
respectively 1.30 pmol/ml, 0.11 fimol/ml, 74 mg/lOO ml, and 5.2 &t/ml. The administration of exogenous ketones during the second phase of the study induced o 47%-92% increase in total ketone levels. During this period, the endogenous production of ketones (colculoted OS the difference between total inflow mte and ocetoocetote infusion mte) omounted only to 67%-90% of control values. Among other factors, this inhibition of ketogenesis wos probably portiolly related to the direct ontilipolytic effect of infused ketones. Indeed, there was o concomitant fall in FFA ond in glycerol levels averaging respectively 13.5% and 17.3%, without significant changes in peripheml insulin concentrations. Our results demonstrott thot during fosting, circulating ketone bodies exert on inhibitory influence on the rote of ketogenesis. This mechanism might play an important role in preventing the development of uncontrolled hyperketonemio during starvation.
From the Metabolic Unit and the Laboratory of Experimental Medicine, University of Brussels, Brussels, Belgium. Receivedfor publication April 21, 1975. Presented in part at the 10th Anntcal Meeting of the European Association for the Study of Diabetes held in Jerusalem, Israel, September 1974. Supported by the Fends de la Recherche Scientifique Medicale (grant 20193) and by the Minis&e Belge de la Politique Scientifique. within the framework of the Association Euratom, University of Brussels, University of Pisa. Reprint requests should be addressed to E. 0. Balasse. M. D.. Laboratory of Experimental Medicine, University of Brussels. Boulevard de Waterloo 115. B-1000 Brussels, Belgium. o 1975 by Grune & Stratton, Inc. Mdobolirm, Vol. 24, No. 9 (September),
1975
999
BALASSE
1000
AND
NEEF
S
INCE approximately 10 years, it has been repeatedly observed that the infusion of ketone bodies reduce plasma FFA levels in various animal species including man. ‘-I2 Since plasma FFA are the primary ketogenic substrate, it has been speculated that the ketone-induced inhibition of adipose tissue lipolysis might restrain ketogenesis.‘,6-9J2 If this were true, this feed-back mechanism should play an important homeostatic role during starvation by preventing the development of uncontrolled hyperketonemia. However, direct evidence for the operation of this feed-back cycle has never been provided. The present experiments demonstrate that in fasting man, the rate of ketogenesis, as measured by an isotopic technique, is indeed inhibited by an exogenous supply of ketones.
MATERIALS
AND METHODS
The study was performed on 6 obese nondiabetic female subjects aged 17-46 yr, hospitalized for therapeutic starvation. Their body weight varied between 115% and 164% of ideal body weight at the time of the study. During the fast, the subjects had a daily intake of one multivitamin capsule (Maxivit, Roerig, Belgium) and were allowed to drink calorie-free beverages (water, black coffee, or tea) ad libitum until 12 hr before the test. Only water was allowed thereafter. The experiments were performed after various durations of fast ranging from 2 to 23 days. An indwelling Cournand needle was placed in a brachial artery for blood sampling and two venous cannulae were inserted in the opposite arm for infusions. The studies were conducted according to the following protocol (Fig. 1). After a first arterial blood specimen had been obtained, sodium 3-‘4C-acetoacetate was infused at a constant rate (0.41-0.57 &/min) for 4 hr. A priming dose of label corresponding to the amount infused in 80 min was given at the start of the infusion. After a basal period of 2 hr. a primed constant infusion of unlabeled sodium acetoacetate (AcAc) was superimposed to the isotope infusion. The rate of delivery ranged between 0.688 and 1.960 mmol/min, the priming dose corresponding to 80 times the amount infused per min. Four 25 ml blood samples were collected at 15 min intervals during the last 45 min of each period and were placed in heparinized tubes and immediately chilled. Samples of 5 ml blood were mixed with perchloric acid and neutralized I3 and the resulting protein-free filtrate was kept on ice until further use. The rest of the blood was centrifuged and the plasma was analyzed for content of FFA,14 glucose,‘5*‘6 and immuno-reactive insulin (IRI).” Glycerol, AcAc, and @-hydroxybutyrate (BOHB) concentrations were estimated in the perchloric filtrate using enzymatic methods.‘8Y’9 14C in ketone bodies was measured by the method of Mayes and Felts.” This method permits the determination of 14C activity separately in AcAc and /3OHB; “C-@GHB is converted enzymatically to 14C-AcAc; 14C-AcAc is decarboxylated to acetone and CO2 which are trapped separately in a double-well flask and assayed for 14C in a liquid scintillation spectrometer. This original method was slightly modified along two lines: (1) since the AcAc used in these experiments was labeled on the carbonyl carbon, only the labeled acetone produced was measured; (2) the procedure was adapted to handle larger volumes of perchloric filtrate, so that ultimately the radioactivity measured corresponded to that contained in 0.5 ml of blood instead of 0.050.06 ml of blood for the original method. The recovery of 14C-AcAc and DL-‘4C-/30HB added to nonradioactive blood (obtained from the patient at the start of the study) was determined with each set of assays and found to be 63x-89% (mean 77%) for AcAc and 76x-89% (mean 81%) for @GHB. Recoveries were consistent within an experiment. For BOHB, recovery was calculated taking into account the fact that the assay is specific for the D(-) isomer so that only half of the added 14C was expected to be recovered. Values for ketone radioactivity in blood samples were corrected accordingly. Any “C-acetone which could be present in the perchloric filtrate before determination of 3-14C-AcAc and 3-‘4C-mHB, was removed by prior ventilation of the filtrate with a stream of nitrogen for at least 1 hr.” The determinations of unlabeled and labeled AcAc were done on the day of each experiment owing to the instability of the compound.
KETOGENESIS
12
-
10 -
6 LOOD KETONES
a-
pm01 /ml 842o1100 KETONE SPECIFIC
900
ACTIVITY dpm /p mol
700
*c*c
+
500 300
INFLOW RATE
OOHB
OOHB
1
l-•-•-• o-
=
0-Q
/
F, I ‘. I I
‘\ ‘.
‘m
-____-___v
----_--_-_--_-_*-*--,~
---_-_________o_*
3
OF 2
TOTAL KETONES mmol/min
1 0 1
PLASMA
1.4
/-+..._
FFA pmol /ml
1.2
I 1
PLASMA
75
GLUCOSE mg /lOOml
65
1 , 0
Fig. 1.
Average
experimental
=--
.
I I
1.0 1
---f------
------k
I
I
I
I1
O”
mrulk
1
8,
MINUTES
120
..__/“/
’
-
!
190
-
1
I
I
240
in six subjects.
Blood pH was measured at the end of the control period and at the end of the AcAc infusion period, in four of the six patients studied. All analyses were performed in duplicate. Prepumfion ofmarerials for injecfion. The unlabeled and the labeled sodium AcAc were prepared, respectively, from unlabeled ethyl AcAc (Merck, Darmstadt, Germany) and chromatographically pure ethyl 3-t4C-AcAc (The Radiochemical Center, Amersham, England, specific
BALASSE
1002
AND NEEF
activity 12.7 mCi/mmol), according to previously described techniques.‘3 The radiochemical purity of the prepared 3-‘4C-AcAc solutions was determined according to Mayes and Felts” and found to vary between 93% and 97%. The radioactivity was mixed with a small volume (2-3 ml) of 0.5 M AcAc in order to minimize its degradation.” After preparation, the solutions were sterilized through a Millipore filter (0.22 p) and kept frozen at -2O’C until use. Immediately before use, the unlabeled AcAc was diluted with distilled water so as to obtain a concentration varying between 222 and 631 pmol/ml and the labeled AcAc was diluted with s/aline. All infused solutions had a pH of 7-7.5 and were kept chilled with ice during the course of the infusions. Calcularions. During the last 45 min of both phases of the experiments, the specific activity (S.A.) of total ketone bodies was calculated for each time point as follows: S.A. of total ketones (dpmlpmol) = [14C-AcAc (dpm/ml) + “C-fiOHB (dpm/ml)]/[AcAc (pmol/ml) + j3OHB
OtmoWt)l (I). Although on an average, the S.A. of total ketones had almost attained the steady state during both periods (Fig. l), a significant rise was nevertheless observed in several experiments. Therefore, for both periods of each experiment, the reduction rate of total ketones was calculated according to the equation proposed by Steele 3? which allows for correction for the change in S.A. with time: R, = [I - (c x V x dS.A./dr)]/[Sx.] (II) where R, is the rate of appearance (or inflow) of ketones (pmol/min), I is the infusion rate of the tracer (dpm/min), c is the average concentration of total ketones over the studied period (rmol/ml), V is the “operational” volume of distribution of ketones (ml) which was shown to approximate 200/, of body weight in fasted subjects,” dS.A./df represents the increment in S.A. as a function of time and was calculated by linear regression using the four values obtained during the last 45 min of each period, and S.A. is the average S.A. of total ketones over the studied period (dpm/pmol). During the basal period, only the isotope was infused and the rate of appearance of total ketones thus represents the rate of ketogenesis. During the second phase, the rate of appearance of total ketones represents the sum of the exogenous AcAc inflow and of the endogenous ketone production. The latter was thus obtained by subtracting the exogenous AcAc inflow from the rate of appearance of total ketones. The rationale for using the S.A. of total ketones instead of that of AcAc in calculating the inflow rate of ketones under our experimental conditions will be outlined under the discussion section. Statistical significance of the observed changes was tested by paired comparison (Student’s t test) of the mean data obtained in each subject during the two experimental periods. RESULTS
During the basal period, the AcAc and POHB concentrations were steady in all subjects (Fig. 1) and total ketone concentrations ranged between 3.98 and 9.65 rmol/ml (Table 1). POHB levels largely exceeded those of AcAc, the pOHB/AcAc ratio varying between 3.23 and 4.05. The S.A. of ketone bodies almost plateaued during the last 45 min of the basal period but the S.A. of /3OHB did not exceed 41%-52x of that of AcAc. Ketogenesis calculated according to equation II varied between 1.450 and 2.053 mmol/min and seemed roughly independent of the duration of the fast, at least for the studied fasting periods. The administration of unlabeled AcAc (at rates equivalent to 43%-98% of basal ketone production) induced a marked rise in both the AcAc and /3OHB concentrations which plateaued during the last 45 min of the infusion (Fig. 1). Total ketone concentrations amounted to 7.41-16.44 pmol/ml during this period (Table 1). The POHB/AcAc ratio was systematically decreased by comparison with the basal period, varying between 1.47 and 2.44. The rise in ketone concentration was associated with a fall in ketone S.A. indicating that total ketone inflow was increased. The difference between total inflow rate and ex-
135
115
164
135
128
DEF.
LEN.
DUF.
DEL.
KAM.
23
18
14
7
5
2
Dllp
6.94 11.02 9.65 16.44
5.41 7.77 7.74 10.85
3.25 1.91 5.59
0.688 0
1.365
1.418
2.053
1.230
1.614
1.345
13.78
8.19
1.53
7.15
5.59
1.250 1.993
8.05
1.450
5.58
0
1.960
5.17
1.57
2.88
0.784 0
5.43
4.17
1.27
0
1.316
1.836
7.30 10.80
5.99 7.66
1.31 3.14
0
0.904
1.436
7.41
4.93
2.48
0.998
1.588
mmol/min
3.98
3.04
Blood’
0.94
pmol/ml
0
mmoljmin
31
24
33
14
28
10
Control
1.80
1.94
0.146
0.153
0.080
0.099 0.95
0.090
1.03
0.140
0.068
0.070
0.069
0.074
0.071
0.108
Bbod
Irmol/ml’
Glycerol
0.95
1.35
0.88
1.01
1.07
1.36
0.96
1.16
ml Plasma
pmol/’
FFA
82
87
77
81
72
78
63
67
62
61
5.0
4.8
4.0
5.5
6.1
13.0
3.0
3.0
1.5
1.8
3.0
3.0 59
PklSfllO
fiU/ml*
68
ml’
IRI
Plasma
mg/lW
GIUCOW
the AcAc infusion period.
each subject: upper line: mean of four values obtained during the last 45 min of the control period, lower line: mean of four values obtained during the last 45 min of
126
KRU.
*For
Weight
Subject
From
Endoganous ProductionR&e of Total Ketones
R-‘y
AcAc
BOHR
AcAc
Percent Inhibition
Infused
ACAC
Percent of Id&
Duration of Fast
Ketone Bodies Production Rates in Fasted Obese Subiects
Table 1. Effect of an Infusion of Sodium Acetoacetate on Blood Metabolite Concentration and
1004
BALASSE
AND NEEF
ogenous AcAc infusion rate provides a measurement of the residual rate of ketogenesis which was lo%-33% lower than that observed during the control period (p < 0.005). This inhibition of ketogenesis by ketone bodies was observed whatever the duration of the fast. The infusion of AcAc inhibited adipose tissue lipolysis as evidenced by a significant fall in the average concentration of FFA (13.5% f 3.6%; p < 0.02) and of glycerol (17.3 f 6.0%; p < 0.05). It should be noticed that, on an average, FFA (Fig. 1) and glycerol levels tended to return towards basal value at the end of the infusion period. In fact, this trend was apparent in only three of the six subjects tested, whereas the other subjects maintained their lowered FFA levels until the end of the ketone infusion. The degree of inhibition of ketogenesis was not significantly related to the fall in FFA levels (p > 0.1). A lowering of blood glucose was noticed (6% + 2%; p < 0.05) but there was no significant change in arterial IRI levels (p > 0.05). In four of the six subjects studied, the infusion of sodium AcAc induced only minimal changes in arterial pH, its mean value rising from 7.36 to 7.40. DISCUSSION
The estimation of ketone bodies production rate using a constant infusion of 14C-AcAc meets with difficulties owing to the absence of isotopic equilibration between blood AcAc and POHB. This situation has been observed by others in ratsz2s2’ and in man.” The liver is believed to be the main site of interconversion between ketones and McGarry et al.22 and Batesz3 provided strong evidence that the disequilibrium found in the isotopic steady state is due to a limited capacity of the liver to carry out the equilibration of infused isotope between AcAc and POHB. Under these conditions, the S.A. of total ketones is assumed to represent the S.A. that would have been attained for both ketones if hepatic equilibration had not been limiting. The best estimate of total ketone body turnover would appear, therefore, to be derived from total ketone body S.A. and this technique of calculation has been employed by McGarry et al. in rats’* and by Reichard et al. in humans.2’ During the basal period, the plasma (blood) concentrations of FFA, glycerol, and glucose were in the accepted normal range for subjects starved 2-23 days but ketone levels tended to be slightly higher than those observed by others.24,25 The rates of production of ketones (1.450-2.053 mmol/min) were significantly higher than those published by Reichard et al. (0.674-1.553 mmol/min) although these authors used an isotopic technique similar to ours.*’ The reasons for these differences are not obvious. Most of the obese subjects studied by this latter group 2’.24,25had a much greater excess body weight2’~24*25and had higher IRI levels24*z5than those studied here and this might account, at least partly, for the observed differences. It should be mentioned that in another group of eight obese subjects fasted 3-23 days studied in our laboratory, rates of ketogenesis measured using either constant infusions or single injections of 14C-AcAc, were of the same order of magnitude (1.670-2.946 mmol/min) as those observed here. Whatever mechanism may be responsible for the higher ketone turnover rates found in our subjects, it should not interfere with the main significance of this work which is based on the comparison of rates of production obtained in the same subject under two different experimental conditions.
KETOGENESIS
1005
The administration of AcAc induced a rise in ketonemia and a significant and sustained drop in ketone production rate (Table 1). This effect was observed whatever the duration of the fast. The decrease in ketone production rate is likely to result, at least partly, from the observed fall in the levels of FFA which are known to be the major ketogenic substrate. Although an insulinotropic effect of ketone bodies can undoubtedly be observed in man under apthe fall in FFA found in our experiments cannot propriate conditions, ‘“~1’*26,27 be explained on this basis, peripheral IRI levels remaining steady or even decreasing in some patients during the AcAc infusion. Under these conditions, the drop in FFA was probably related to the direct antilipolytic properties of ketone bodies first demonstrated in vitro by Bjorntorp2’ and confirmed by many others.*‘” A fall in FFA levels independent of insulin release following ketone infusions in man has been previously described.7,8J2 The participation of a ketone-induced inhibition of glucagon secretion32333 in the observed antilipolytic effect of ketones cannot be ruled out in these experiments. The antiketogenic effect of AcAc infusions can probably not be explained solely on the basis of their antilipolytic action for the two following reasons: (1) in three of the six subjects tested, the antilipolytic effect was transient although ketogenesis remained depressed until the end of the ketone infusion; (2) there was no correlation between the fall in FFA levels and the corresponding degree of inhibition of ketone production; for instance, the marked decrease in ketogenesis observed in subjects DEL and KAM (Table 1) was not associated with a significant drop in FFA levels. Additional mechanism(s) should therefore be considered in order to explain the observed reduction in ketogenesis; one possibility is that, despite the absence of IRI changes in peripheral blood, the ketone infusion could have stimulated insulin release in the portal vein2’ thereby inhibiting hepatic ketogenesis independently of changes in the amounts of FFA delivered to the liver.” The observed hypoglycemic effect of ketones is confirmatory of many other studies.7-9~‘2*27The data of the literature indicate that this effect may be accompanied, but need not depend upon the release of insulin.7p8*‘2 The influence of exogenous AcAc on endogenous ketone production has been studied before by Bergman et al.” in sheep and by Bates23 in rats. The first group of authors found comparable rates of endogenous AcAc production in normal and AcAc infused sheep. Their experimental design has the disadvantage of comparing two different groups of animals whereas our subjects served as their own control in a single experiment. Bates noted that infused AcAc inhibits ketogenesis but she thought that the calculated reduction in endogenous appearance of AcAc could result from a methodological artefact.23 In conclusion, these data represent the first direct demonstration that circulating ketones exert a negative feedback on the rate of ketogenesis in fasting man. This effect is probably partially mediated through the direct antilipolytic action of ketones on adipose tissue, but our data suggest that it is not the only involved factor. The physiological implications and teleological significance of such a mechanism in maintaining ketone homeostasis during starvation are obvious and have already been emphasized by several authors including US.‘*~-~J*,~~ The demonstration that infused ketone bodies reduce FFA levels in pancreatectomized dogs* and in insulin-dependent diabetic humans” suggest
BALASSE
1006
AND
NEEF
that diabetic ketosis is another situation in which ketone bodies tend to limit their own production rate. It is clear, however, that in this case, the feed-back control mechanism is insufficient to maintain ketone bodies at concentrations compatible with life. ACKNOWLEDGMENT We gratefully
acknowledge Mrs D. Calay,
nurse of the Metabolic Unit, and Dr. E. Van Thiel, Department of Cardiology, for their help in performing these studies. Thanks are also due to the technicians of the Department of Cardiology for blood pH determinations, and to Mrs. B. Noel for typing the manuscript.
REFERENCES 1. Madison LL, Mebane D, Unger RH, et al: The hypoglycemic action of ketone. II. Evidence for a stimulatory feedback of ketones on the pancreatic beta cells. J Clin Invest 43: 408415, 1964 2. Bjijrntorp P, Scherstin T: Effect of &hydroxybutyrate on lipid mobilization. Am J Physiol212:683-687, 1967 3. Balasse E, Couturier E, Franckson JRM: Influence of sodium &hydroxybutyrate on glucose and free fatty acid metabolism in normal dogs. Diabetologia 3:488-493, 1967 4. Metzger P, Franken P, Balasse EO: Permissive role of glucose on the insulinotropic effect of ketone bodies in vivo. Horm Metab Res 5:313-315, 1973 5. Balasse EO: Inhibition of free fatty acid oxidation by acetoacetate in normal dogs. Em J Clin Invest 1:155-160, 1970 6. Hawkins RA, Alberti KGMM, Houghton CRS, et al: The effect of acetoacetate on plasma insulin concentration. Biochem J 125: 541-544,197l 7. Balasse E, Ooms HA: Changes in the concentrations of glucose, free- fatty acids, insulin and ketone bodies in the blood during sodium @hydroxybutyrate infusions in man. Diabetologia 4:133-135, 1968 8. Senior B, Loridan L: Direct regulatory effect of ketones on lipolysis and on glucose concentrations in man. Nature 219:83-84, 1968 9. Jenkins DJA: Ketone bodies and the inhibition of free-fatty-acid release. Lancet 2:338340, 1967 IO. Jenkins DJA, Hunter WM, Goff DV: Ketone bodies and evidence for increased insulin secretion. Nature 227:384-385, 1970 I I. Owen OE, Reichard GA Jr, Markus H, et al: Rapid intravenous sodium acetoacetate infusion in man. Metabolic and kinetic responses. J Clin Invest 52:2606-2616, 1973 12. Binkiewicz A, Sadeghi-Nejad A, Hochman H, et al: An effect of ketones on the con-
centrations of glucose and of free fatty acids in man independent of the release of insulin. Pediatrics 84:226-23 1, 1974 13. Balasse EO, Have1 RJ: Evidence for an effect of insulin on the peripheral utilization of ketone bodies in dogs. J Clin Invest 50~801-8 13. 1971 14. Trout DL, Estes EH, Friedberg SJ: Titration of free fatty acids of plasma: A study of current methods and a new modification. J Lipid Res 1:199-202, 1960 15. Hoffman WS: A rapid photoelectric method for the determination of glucose in blood and urine. J Biol Chem 1205-55, 1937 16. Technicon Auto Analyzer Methodology Manual. Chauncey, N.Y ., Technicon Instruments 17. Herbert V, Lau KS, Gottlieb CW, et al: Coated charcoal immunoassay of insulin. J Clin Endocr 25:1375-1384, 1965 18. Chernick SS: Determination of glycerol in acyl-glycerols, in Lowenstein J (ed): Methods in Enzymology, vol 14. New York, Academic Press, 1969, p 627 19. Williamson DH, Mellanby J, Krebs HA: Enzymic determination of D(-)&hydroxybutyric acid and acetoacetic acid in blood. Biochem J 82:90-96, 1962 20. Mayes PA, Felts JM: Determination of “C-labeled ketone bodies by liquid-scintillation counting. Biochem J 102230-235, 1967 21. Reichard GA, Owen OE, HafT AC, et al: Ketone-body production and oxidation in fasting obese humans. J Clin Invest 53:5088515, 1974 22. McGarry JD, Guest MJ, Foster DW: Ketone body metabolism in the ketosis of starvation and alloxan diabetes. J Biol Chem 245:4382-4390, 1970 23. Bates MW: Kinetics of ketone body metabolism in fasted and diabetic rats. Am J Physiol221:984-991, 1971 24. Owen OE, Felig P, Morgan AP, et al:
1007
KETOGENESIS
Liver and kidney
metabolism
during
prolonged
starvation. J Clin Invest 48574-582, 1969 25. Owen OE, Reichard GA Jr: Human forearm metabolism during progressive starvation. J Clin Invest 50~1536-1545, 1971 26. Pi-Sunyer FX, Sethi SS: Stimulation of insulin secretion by sodium beta-hydroxybutyrate in man. Diabetes 21:373, 1972 27. Balasse EO, Ooms HA, Lambilliotte JP: Evidence for a stimulatory effect of ketone bodies on insulin secretion in man. Horm Metab Res 2:371-372, 1970 28. Bjorntorp P: The effect of beta-hydroxybutyric acid on glycerol outflow from adipose tissue in vitro. Metabolism 15:191-193, 1966 29. Devecerski M, Pierce CE, Frawley TF: Effect of ketone acids on glucose and fat metabolism in adipose tissue of the rat. Metabolism 17:877-884, 1968 30. Hellman DE, Senior B, Goodman HM: Antilipolytic effects of p-hydroxybutyrate. Metabolism 18:906-915, 1969 31. Hotta N, Sirek A, Sirek OV: Inhibitory
effect of beta-hydroxybutyrate on lipolysis stimulated by dihydroergotamine and growth hormone in vitro. Can J Physiol Pharmacol 49:87-91, 1971 32. Edwards JC, Taylor KW: Fatty acids and the release of glucagon from isolated guinea-pig islets of Langerhans incubated in vitro. Biochim Biophys Acta 215:310-315, 1970 33. Marliss EB, Wollheim CB, Blonde1 B, et al: Insulin and glucagon release from monolayer cell cultures of pancreas from newborn rats. Eur J Clin Invest 3: 16-26, 1973 34. Bieberdorf FA, Chernick SS, Scow RO: Effect of insulin and acute diabetes on plasma FFA and ketone bodies in the fasting rat. J Clin Invest 49168551693, 1970 35. Bergman EN, Kon K, Katz ML: Quantitative measurements of acetoacetate metabolism and oxidation in sheep. Am J Physiol 205:658-662, 1963 36. Steele R: Influence of glucose loading and of injected insulin on hepatic glucose output. Ann NY Acad Sci 82:420-430, 1959