Increased Ketogenesis Related to Insulin Deficiency in Isolated Hepatocytes from NIDDM Model Rats
Summary To investigate the hepatic ketone body metabolism in N I D D M , we studied the ketone body production rates in hepatocytes from newly developed non-obese N I D D M model rats. N I D D M model rats were prepared by intraperitoneal injection of streptozotocin at 2 or 5 days of age (STZ 2, STZ 5 respectively). After 10—15 weeks, ketone body production rates in hepatocytes isolated from these rats were compared with those from control rats as well as ketotic rats made by intravenous injection of streptozotocin into adult rats. Basal ketone body production rates from 0.3 mM [U-14C] palmitate in hepatocytes from control, STZ 2, STZ 5 and ketotic rats were 11.7 + 0.98, 14.9 + 0.72, 16.0 + 0.45, 22.8 + 2.32 nmole-palmitate/mg.prot/hr, respectively. These rates were stimulated by 1 (ug/ml of glucagon in control, STZ 2 and STZ 5 rats (14.1 ±0.99, 18.6+1.36, 18.7 + 0.69 nmole-palmitate/mg.prot/hr, respectively), but not in ketotic rats (22.8 + 2.07 nmole-palmitate/mg-prot/hr). The similar effects were observed by 1 (ug/ml of epinephrine. The basal ketone body production rates were negatively correlated to both hepatic glycogen contents and plasma IRI levels. Considering these parameters together, the extent of metabolic derangement in STZ 2 and STZ 5 rats was between that in control and ketotic rats. These results indicate that the derangements of hepatic ketone body production are related to the severity of insulin deficiency and suggest that the enhanced hepatic ketogenesis contributes in part to the elevated plasma ketone body levels in non-obese N I D D M . Key words N I D D M Model Rats - Freshly Isolated Hepatocytes — Ketone Bodies — Glucagon — Insulin
mellitus (NIDDM) have been believed to be ketosis resistant, however, we previously reported that the subjects with N I D D M also showed higher levels of plasma ketone body when poorly controlled, and that the levels of plasma ketone body were normalized only by insulin therapy (Harano, Kosugi, Hyousu, Suzuki, Hidaka, Kashiwagi, Uno and Shigeta 1984). In IDDM subjects, it is believed that the elevation of free fatty acids in plasma contributes to the elevated plasma ketone body levels as extrahepatic supply of ketogenic substrates, furthermore the enhanced hepatic ketone body production plays a major part of the elevated plasma ketone body levels (Foster and McGarry 1989; McGarry, Wright and Foster 1975). Although it is necessary to investigate the hepatic ketogenesis in N I D D M in relation to both the supply of substrates and hepatic activity of ketogenesis, there have been few reports about the hepatic ketone body production in NIDDM, and no reports in non-obese N I D D M because of the lack of the adequate animal model for non-obese N I D D M . When streptozotocin is given to neonatal rats, a transient period of hyperglycemia lasting only a few days has been observed (Portha, Levacher, Picon and Rosselin 1974). Bonner-Weir et al. further observed the long-term effects of this early injury by streptozotocin upon the relationship of (B-cell number and function to glucose tolerance (Bonner-Weir, Trent, Honey and Weir 1981; Weir, Clore, Zwachinski and Bonner-Weir 1981), and reported this model was a useful tool for the study of metabolism in non-obese N I D D M . To investigate the ketone body metabolism in non-obese N I D D M , we measured hepatic glycogen contents as well as ketone body production from [U-14C]palmitate in isolated hepatocytes from the non-obese N I D D M model rats, and compared with those from ketotic rats as a model of IDDM. Materials and Methods 1.
Introduction It is well known that subjects with poorly controlled insulin dependent diabetes mellitus (IDDM) frequently show markedly elevated plasma ketone body levels, and in some cases develop ketoacidosis (Foster and McGarry 1989). Subjects with non-insulin dependent diabetes Horm. metab. Res. 24 (1992) 258 - 262 © Georg Thieme Verlag Stuttgart • New York
Animals
Non-obese NIDDM model rats were prepared by intraperitoneal injection of streptozotocin 100 mg/kg to Sprague-Dawley rats at the age of 2 or 5 days (STZ 2, STZ 5 respectively). Male rats were selected and were fed on a solid laboratory chow (Kurea, Japan) until 10—15 weeks of age (300—400 g of body weight) for experiment. Ketotic rats were made by intravenous injection of streptozotocin (100 mg/kg) to normal Sprague-Dawley rats weighing 250—350 g (10—13 weeks of age), and they were treated with 8 units/day insulin for 7—10 Received: 3 Apr. 1991
Accepted: 19 Oct. 1991
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
T. Aoki1, H. Hidaka1 , K. Kosugi1, H. Kojima1 , Y. Nakajima1 , T.Nakamura1, Y. Harano2 and Y. Shigeta1 Third Department of Medicine, Shiga University of Medical Science, Ohtsu, Shiga, Japan Department of Atherosclerosis and Metabolism, National Cardiovascular Center Hospital, Suita, Osaka, Japan
Ketogenesis in Hepatocytes from NIDDMModel
Horm. metab. Res. 24 (1992)
259
Characteristics of the experimental rats. n
Body Weight (g)
Control
5
344 ±29
STZ2
5
334 ±21
Plasma Glucose (mg/dl)
5
Ketotic
5
IRI (uU/ml)
Total Ketone Bodies (MM)
OHBA (uM)
AcAc (uM)
Glycogen Contents (ug-glucose/ mg-protein)
5.7 ±0.8
284 ±25
20.9 ±1.9
385 ±31
324 ±33
60 ±10
140.5 ±13.4
+ 80
±5
++
++
+ *
8.7 ±0.9
1055 ±387
858 ±334
**
**
**
.
..
9166 ±1951
4044 ±1556
9.7 ±1.1
9.8 ±0.8
..
+ .,
++
415 ±13.8
13.3 ±1.5
432 ±69
,,
**
**
P
++ 337 ±28
**
407 ±44
290 + 9.4
.
++ 417 ±31
++ . 7.5 ±0.7
+ ..
324 ±20
265 ±22
FFA (u.Eq/1)
201 ±6.8
+
STZ 5
Glycated Hb (%)
466 ±66.2
9.4 ±0.5
1192 ±172
**
8.3 ±1.0
13210 ±3260
197 ±51
++ 142.4 ±14.6 ++ 89.0 ±22.0
The body weights on the day of STZ infusion in ketotic rats are in the parenthesis. Results are expressed as Mean ± SEM. Hb: hemoglobin, FFA: free fatty acids, OHBA: 3-hydroxybutyrate, AcAc: acetoacetate. *p < 0.05; **p < 0.01 vs control by t-test. + p < 0.05; + + p < 0.01 vs ketotic by t-test.
days (Novolin U , Novo, Denmark). On the day before experiments, ketotic rats did not receive insulin. Rats with ketonuria (determined by Ketostics ) on the day of experiment were used. Other groups of the rats except for ketotic did not show ketonuria. 2. Isolation of
hepatocytes
Under anesthesia with intraperitoneal injection of pentobarbital (6.5 mg/100 g body weight), hepatocytes were isolated as described previously (Kojima, Harano, Kosugi, Nakano and Shigeta 1986) with a cell viability over 90 % by Trypan Blue exclusion. Immediately after abdominal opening, 1 ml of blood was drawn from inferior caval vein. Following centrifugation, aliquots of plasma were stored at — 20 °C. For the determination of ketone body concentrations, the plasma was stored at — 70 °C. Red blood cells were stored at -20 °C after adding saline for the measurement of glycated hemoglobin. Plasma ketone body concentrations in unanesthetized rats were also determined using plasma taken from the rats by decapitation. 3. Ketone body production in isolated hepatocytes
rates
Ketone body production rates in hepatocytes were determined after 60 min incubation at 37 °C with 5.5 mM of glucose 6 and 0.25 x 10 cells/ml of hepatocytes saturated with 95% O2, 5% CO2 (final volume 2.0 ml). The reaction was started by adding 0.2 ml of [U- C]palmitate-albumin complex, and was terminated by adding 1 ml of 10% perchloric acid. After centrifugation, 0.5 ml of the supernatant was taken for the measurements of ketone body production rates (Christiansen 1977; Demaugre, Buc, Girard and Leroux 1983).
4. Hepatic glycogen contents Hepatic glycogen contents were determined using the glucose oxidase method (DIAGLUCA , Toyobo, Japan) after hydrolyzing by amyloglucosidase (EC.3.2.1.3) (Keppler and Decker 1974). 5. Assay kits and sources Ketone bodies, Ketone Test Sanwa (Sanwa Kagaku Kenkyusho, Japan) (Harano, Kosugi, Hyousu, Uno, Ishikawa and
Shigeta 1983); free fatty acids, Determiner NEFA 755 (Kyowa Medics, Japan); glycated hemoglobin, Glyc-Affin GHb (Seikagaku Kogyo Co. Ltd., Japan); plasma immunoreactive insulin (IRI), Insulin kit "Daiichi" S (Daiichi Radioisotope, Kenkyusho, Japan); protein, Bio-Rad Protein Assay (Nippon Bio-Rad Laboratories, Tokyo, Japan), y-globulin as a standard. [U-14C]palmitate, Amersham (Japan); collagenase class II, Worthington (U. S. A.); fatty acid free bovine serum albumin, sodium palmitate and amyloglucosidase, Sigma (U. S. A.); human insulin and porcine glucagon were gifts from Novo (Denmark). 6. Data expression and
statistics
All results were expressed as mean ± SEM, and statistical comparisons were made with the Student's t-test.
Results 1. Characteristics of the experimental rats Glycated hemoglobin levels in STZ 5 and ketotic rats were significantly higher than the control rats. Diabetic rat groups had significantly lower IRI levels than control rats, and ketotic rats had lower hepatic glycogen contents compared with other rats. Although plasma free fatty acid levels in STZ 5 and control were not significantly different because of the wide distribution, the plasma levels in STZ 2 and ketotic rats were approximately 1.5 times and 4 times of those in control rats, respectively (Table 1). In unanesthetized rats, plasma total ketone body levels in control, STZ 2, STZ 5 and ketotic rats were 194 ± 2 0 , 269 + 38, 422 ± 6 5 and 16189 ±2667 uM, respectively (in each group, mean + SEM, n = 5), and the values observed in STZ 5 and ketotic rats were significantly higher than that in control rats (for STZ 5: p < 0.05, for ketotic: p < 0.01). The higher levels of plasma ketone bodies in the anesthetized animals especially in control rats might be accounted for in part by the influence of the experimental procedures including the anesthesia.
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
Table 1
Rats
260 Horm. metab. Res. 24 (1992)
Ketone body production rates in isolated hepatocytes from diabetic rats. Without hormones Palmitate concentration (mM) 0.075 0.15 0.3
Control
6.43
9.50
11.7
±
+
±
0.71 STZ 2
STZ5
7.01
11.5
14.1*
14.1*
0.98
0.99
1.33
14.9*
18.6#
20.0
14.5
+
±
±
+
+
+
0.83
0.80
0.72
12.6**
16.0"
±
±
9.10**
±
Ketotic
0.80
With hormones Palmitate concentration (0.3 mM) Glucagon Epinephrine Insulin (1 u.g/ml) (1 ug/ml) 1 ug/ml)
±
±
1.36
1.72
18.7**
19.0**
±
0.20
0.11
0.45
14.6**
17.8**
22.8**
22.8
21.6
±
±
±
+
+
2.07
1.56
1.24
1.52
2.32
0.69
±
0.75
12.1
+ 1.05
1.80 16.4
±
0.21 21.1
±
1.57
Ketone body production rates are expressed as nmole.palmitate/mg.prot/hr, and the values are Means±SEM. *p < 0.05; **p < 0.01 vs control at same palmitate concentration without hormones, by t-test. p < 0.05; p < 0.01 vs palmitate concentration 0.3 mM without hormones, by paired t-test.
2. Ketone body production rates from [U-14C] palmitate in rat hepatocytes Ketone body production rates from 0.3 mM [U-I4C]palmitate in hepatocytes from control, STZ 2, STZ 5 and ketotic rats were 11.7 ±0.98, 14.9 + 0.72, 16.0 + 0.45 and 22.8 ±2.32 nmole-palmitate/mg-prot/hr, respectively (Table 2). The rates were also increased in the cells isolated from diabetic rats at lower concentrations of palmitate, suggesting that the enhancement of ketogenesis in diabetic rat hepatocytes was due to the increase of the maximal capacity of hepatic ketone body production. 1 u.g/ml glucagon, a maximal concentration for the stimulation of hepatic ketogenesis (Christiansen 1977), increased ketone body production rates significantly in the hepatocytes from control, STZ 2, STZ 5 rats, but not from ketotic rats (21.4 ±4.33, 22.8 ±6.71, 17.4±3.66 and 0.3±3.5%, respectively). The stimulation of the production rates by 1 ug/ml of epinephrine was similar (% increase; 23.4 ±5.2% in control, 34.1 ±11.1% in STZ 2, 19.6 ± 2.4% in STZ 5 and 1.9 ± 2.8% in ketotic). 3. Relationship between the ketone body production rates and plasma IRI levels Ketone body production rates in the hepatocytes were negatively correlated with plasma IRI (Fig. 1), and hepatic glycogen contents (r = — 0.672, p < 0.05). When glycogen contents and plasma IRI were expressed as a logarithmic function, more linear correlation was obtained (for glycogen contents; r = — 0.795, IRI; r = — 0.637). In either case, the values of STZ 2 and STZ 5 were between those of control and those of ketotic rats. Although weak, there was a positive correlation between plasma glucose levels and hepatic ketone body production rates (r = 0.446, p < 0.1).
Fig. 1 Correlations between plasma IRI and plasma glucose levels (A), hepatic ketone body production rates (B). Control (closed circles), STZ 2 (open circles), STZ 5 (crossed marks), ketotic rats (closed triangles).
Discussion In this report, we studied the ketogenic capacity of the isolated hepatocytes from the NIDDM model rats with mild impairment of insulin secretion that showed no sig-
nificant differences in fasting plasma glucose levels as well as body weights compared with those of normal rats (Weir et al. 1981; Tsuji, Taminato, Usami, Ishida, Kitano, Fukumoto, Koh, Kurose, Yamada, Yano, Seino and Imura 1988). The he-
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
Table 2
T. Aoki, H. Hidaka, K. Kosugi, H. Kojima, Y. Nakajima, T. Nakamura et al.
Ketogenesis in Hepatocytes from NIDDM Model Rats
An enhanced ketogenesis has been reported in the liver of the diabetic model rats, however, these rats were characterized by extremely impaired insulin secretion. It has been thought that the ketogenesis in the liver at fasting or severe diabetic states could never be regulated by glucagon {McGarry and Foster 1979; Witters and Trasko 1979). The results in this study that the ketone body production was not stimulated in hepatocytes from ketotic rats by glucagon agree with the previous reports. However, in rats with mild hyperglycemia whose hepatic glycogen contents were conserved, ketone body production can be increased by glucagon or epinephrine as in control rats. This result clearly indicates that hepatic ketone body production is regulated by the hormones in mild diabetic states. As shown in Table 2, the ketone body production rates in the hepatocytes from control, STZ 2 and STZ 5 rats stimulated by glucagon never exceeded that of ketotic rats. This might indicate that the ketogenic capacity of liver with a severe diabetes is fully accelerated, and that hepatic ketogenesis could no longer be regulated by the hormones. Although there have been a few reports about the ketone body metabolism in non-obese N I D D M using model rats injected with small amounts of streptozotocin intravenously (50 — 60 mg/kg) (Schein, Alberti and Williamson 1971; Cook and Gamboe 1987), this non-ketotic model rat shows a marked body weight reduction and extremely high levels of plasma ketone bodies ( 2 - 5 mM) (Schein, Alberti and Williamson 1971; Nishio, Kashiwagi, Kida, Kodama, Abe, Saeki and Shigeta 1988). At this point, the rat injected streptozotocin at neonatal period seems to be a better model for nonobese N I D D M (Bonner- Weiretal 1981; Weiretal 1981). The primary defect in human NIDDM has been hypothesized to be impaired insulin actions in peripheral tissues, especially in muscle (DeFronzo, Ferrannini and Koivisto 1983), however, the abnormalities of insulin secretion play important roles in the development of N I D D M , at least in non-obese subjects (Kosaka, Hagura and Kuzuya 1977). The ketone body metabolism in the isolated hepatocytes from the NIDDM rats used in our study could be interpreted as the ketone body metabolism in non-obese N I D D M . Since the ketone body production in isolated hepatocytes was increased with higher fatty acid concentrations in the incubation medium as shown in Table 2, elevated plasma levels of free fatty acids must play some role in the development of mild ketonemia in N I D D M . However, enhanced ketogenesis in the hepatocytes from N I D D M rats should be considered to be one of the factors which induce the elevation of plasma ketone body levels in N I D D M . In addition, the regulation by anti-insulin hormones would also contribute to hyperketonemia in N I D D M , because the ketone body production rates in hepatocytes from NIDDM rats were enhanced by glucagon and epinephrine.
261
Insulin deficiency in the hepatocytes is an important factor for the derangements of hepatic ketone body metabolism, because the ketone body production rates in hepatocytes from control, mild diabetic and severe diabetic rats were increased in this turn, and were negatively correlated to plasma IRI levels. Therefore, the elevated levels in plasma ketone bodies in NIDDM subjects would reflect, at least in part, the absolute and/or relative impairment of insulin action in the liver. Moreover, the derangements of ketone body metabolism are more sensitive to insulin deficiency, at least in hepatocytes, than the derangements of glucose metabolism, since the hepatic glycogen contents in hepatocytes from control rats were not significantly different from those in hepatocytes from STZ 2 and STZ 5 rats. In summary, we showed that the enhanced hepatic ketone body production due to insulin deficiency and its regulation by anti-insulin hormones played an important role for the elevated plasma ketone body levels in NIDDM model rats. And the N I D D M model rat employed in this report was considered to be a useful animal model for the study of nonobese N I D D M with impaired insulin secretion. References Bonner-Weir, S., D. F. Trent, R. N. Honey, G. C. Weir: Responses of neonatal rat islet to streptozotocin: Limited B-cell regulation and hyperglycemia. Diabetes 30: 64-69 (1981) Christiansen, R. Z.: Regulation of palmitate metabolism by carnitine and glucagon in hepatocytes isolated from fasted and carbohydrate refed rats. Biochem. Biophys. Acta 488:249 - 262 (1977) Cook, G. A., M. S. Gamboe: Regulation of carnitine palmitoyltransferase by insulin results in decreased activity and decreased apparent Ki values for malonyl-CoA. J. Biol. Chem. 262: 2050— 2055 (1987) DeFronzo, R. A., E. Ferrannini, V. Koivisto: New concepts in the pathogenesis and treatment of noninsulin-dependent diabetes mellitus. Am. J. Med. 74 (Suppl. 1): 52-81 (1983) Demaugre, F., H. Buc, J. Girard, J. P. Leroux: Role of the mitochondrial metabolism of pyruvate on the regulation of ketogenesis in rat hepatocytes. Metabolism 32:40-48 (1983) Foster, D. W., J. D. McGarry: Acute complication of diabetes: ketoacidosis, hyperosmolar coma, lactic acidosis. Endocrinology, 2nd Edit, Vol. 2, Chapter 87. Degroot, L. J. (ed.), W. B. Saunders Company, Philadelphia: 1439-1453 (1989) Harano, Y., K. Kosugi, T. Hyousu, M. Suzuki, H. Hidaka, A. Kashiwagi, S. Uno, Y. Shigeta: Ketone bodies as markers for Type 1 (insulin-dependent) diabetes and their value in the monitoring of diabetic control. Diabetologia26:343-348 (1984) Harano, Y., K. Kosugi, T. Hyousu, S. Uno, Y. Ichikawa, Y. Shigeta: Sensitive and simplified method for the differential determination of serum levels of ketone bodies. Clinica. Chim. Acta 134: 327— 337(1983) Keppler, D., K. Decker: Glycogen: Determination with amyloglucosidase. Methods of enzymatic analysis. 2nd English edit., Vol. 3, Bergmeyer, H. U. (ed.), Verlag Chemie Academic Press, Inc. New York: 1127-1131(1974) Kojima, H, Y. Harano, K. Kosugi, T. Nakano, Y. Shigeta: A suppressive role of c-kinase for the stimulation of hepatic ketogenesis by glucagon and epinephrine. FEBS Letters 201:271 —276 (1986) Kosaka, K., R. Hagura, T. Kuzuya: Insulin responses in equivocal and definite diabetes, with special reference to subjects who had mild glucose intolerance but later developed definite diabetes. Diabetes 26:944-952(1977) McGarry, J. D., D. W. Foster: In support of the role of malonyl-CoA and carnitine acyltransferase I in the regulation of hepatic fatty acid oxidation and ketogenesis. J. Biol. Chem. 254: 8163—8168 (1979)
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
patic ketone body production rates in the hepatocytes from N I D D M model rats were between those in the control and in the severe diabetic rats, and were stimulated by both glucagon and epinephrine. However, the hormones did not affect ketogenesis in the ketotic rats. The hepatic ketone body production rates were negatively correlated with plasma IRI levels as well as the hepatic glycogen contents.
Horm. metab. Res. 24 (1992)
Horm. metab. Res. 24 (1992)
T. Aoki, H. Hidaka, K. Kosugi, H. Kojima, Y. Nakajima, T. Nakamura et al.
McGarry, J. D., P. H. Wright, D. W. Foster: Hormonal control of keto-Weir, G. C, E. T. Clore, C. J. Zwachinski, S. Bonner-Weir: Islet secregenesis: Rapid activation of hepatic ketogenie capacity in fed rats tion in a new experimental model for non-insulin-dependent diaby anti-insulin serum and glucagon. J. Clin. Invest. 55: 1202—1209 betes. Diabetes 30:590-595 (1981) (1975) Witters, L. A., C. S. Trasko: Regulation of hepatic free fatty acid meNishio, Y., A. Kashiwagi, Y. Kida, M. Kodama, N. Abe, Y. Saeki, Y. tabolism by glucagon and insulin. Am. J. Physiol. 237 (1): E23Shigeta: Deficiency of cardiac p-adrenergic receptor in streptozoE29 (1979) tocin-induced diabetic rats. Diabetes 37:1181 -1187 (1988) Portha, B., C. Levacher, L. Picon, G. Rosselin: Diabetogenic effect of streptozotocin in the rat during the perinatal period. Diabetes 23: 889-895(1974) Requests for reprints should be addressed to: Schein, P. S., K. G. M. M. Alberti, D. H. Williamson: Effects of streptozotocin on carbohydrate and lipid metabolism in the rat. EnDr. Hideki Hidaka docrinology 89: 827-834 (1971) Third Department of Medicine Tsuji, K., T. Taminato, M. Usami, H. Ishida, N. Kitano, H. Fukumoto, Shiga University of Medical Science G. Koh, T. Kurose, Y. Yamada, H. Yano, Y. Seino, H. Imura: Ohtsu Characteristic features of insulin secretion in the streptozotocinShiga520—21 (Japan) induced NIDDM rat model. Metabolism 37:1040-1044 (1988)
Downloaded by: University of Pennsylvania Libraries. Copyrighted material.
262