- Email: [email protected]
Effects of ciglitazone on energy balance, thermogenesis and brown fat activity in the rat
Molecular and Cellular Endocrinology, Elsevier Scientific Publishers Ireland,
57 (1987) 253-257 Ltd.
253
MCE 01661
Effects of ciglitazone on energy balance, thermogenesis and brown fat activity in the rat Nancy J. Rothwell, Department
of Physiology,
St. George’s Hospilal Medical School, Cranmer Terrace, Tooting, London S WI 7 ORE, U.K. (Received
Key words: Ciglitazone;
Insulin;
Michael J. Stock and Alison E. Tedstone
Thermogenesis;
17 December
1986; accepted
Brown adipose
tissue;
12 February
Energy
1987)
balance
Summary Young male rats were treated with vehicle or ciglitazone (150 mg/kg/day, intragastric) for 8 or 14 days. Drug treatment did not affect food intake but reduced body weight and energy gains over 14 days, and significantly depressed energetic efficiency. Energy expenditure and resting oxygen consumption (VO,), when corrected for body size, were elevated in ciglitazone-treated rats, but the difference in VO, was abolished by treatment of the animals with a /3-adrenergic antagonist (propranolol). The acute thermic response (postprandial rise in VO,) to a fat meal was similar for both groups, but the response to carbohydrate ingestion was greater in ciglitazone-treated rats (18%) than controls (11.5%). The mass of interscapular brown adipose tissue was not affected by drug treatment, but its protein content was increased and its thermogenic activity (mitochondrial purine nucleotide binding) was elevated by 25% after chronic treatment with ciglitazone. These results indicate that ciglitazone enhances thermogenesis via sympathetic activation of brown adipose tissue, probably as a result of improved insulin sensitivity.
Introduction Insulin is generally described as an anabolic hormone promoting synthesis and deposition of lipid, protein and carbohydrate (glycogen), and in excess quantities it may produce obesity. However, there is now increasing evidence to suggest that insulin is involved in energy dissipation and can increase sympathetic activity and thermogenesis. Regulatory increases in metabolic rate occur in response to acute ingestion of food or chronic hyperphagia, and this diet-induced thermogenesis
Address for correspondence: Professor M.J. Stock, Department of Physiology, St. George’s Hospital Medical School, Cranmer Terrace, Tooting, London SW17 ORE, U.K. 0303-7207/87/$03.50
0 1987 Elsevier Scientific
Publishers
Ireland.
(DIT), which can be distinguished from the obligatory energy costs of nutrient digestion, absorption and assimilation, results from sympathetic activation of heat production, mainly in brown adipose tissue (BAT, see Rothwell and Stock, 1983, 1984, for reviews). DIT allows young, lean animals to avoid obesity during periods of hyperphagia, but the level of thermogenesis and BAT activity are severely inhibited by streptozotocin-induced diabetes (Rothwell and Stock, 1981) or by subdiaphragmatic vagotomy (Andrews et al., 1985), but restored to normal by insulin replacement. Acute inhibition of insulin release with d&oxide also attenuates the thermic response to a single meal of carbohydrate, but not fat (Rothwell et al., 1983, 1985). Conversely, several groups have demonstrated that Ltd
254
chronic infusion of insulin in rats activates BAT via the sympathetic nervous system (Seydoux et al., 1984; Rothwell and Stock, 1984) although with very high doses of insulin these thermogenic actions may be masked by anabolic effects of the hormone. Insulin is also implicated in the genetic obesity seen in laboratory rodents, where there is increased energetic efficiency associated with impaired thermogenesis and brown fat activity. Mercer and Trayhurn (1984) have shown that the diminished thermogenic activity in obese (ob/ob) mice may be related to insulin resistance, and have further demonstrated that amelioration of this insulin resistance with the oral hypoglycaemic drug ciglitazone largely restores the defective BAT response to cold in the obese mutant to the level of lean animals (Mercer and Trayhurn, 1986). We have therefore investigated the effects of this drug ciglitazone on energy balance, BAT activity and thermic responses to fat or carbohydrate in normal, non-obese rats. The results are consistent with the idea of an insulin involvement in thermogenesis and suggest that improving insulin sensitivity with drugs such as ciglitazone enhances DIT. Materials and methods The animals used in this study were male, Sprague-Dawley SPF rats (Charles River, Kent, U.K.) aged 35-40 days (140-150 g), which were caged in pairs at 24°C with free access to water and a commercial pelleted stock diet (PRD, Christopher Hill Group, Dorset, U.K.) supplying 27% of energy as protein, 9% as fat and 64% as carbohydrate. Metabolizable energy (ME) intake of the diet was assessed from the weight of food consumed and its ME density (12.0 kJ/g) determined in previous feeding trials. Rats were treated daily at 09.00 h with either ciglitazone suspended in water (150 mg/kg/day by gastric intubation) or vehicle (controls). Experiment 1 ME intake and body weight were recorded daily in eight control and eight ciglitazone-treated rats for 8 days. On day 5 and 7 the treatment was delayed until 15.00 h and resting oxygen consumption (VO,) was measured (10.00-14.00 h) in
closed-circuit respirometers (Stock, 1975) for 2 h before and 2-3 h after gastric intubation with either carbohydrate (40 kJ cornflour in 4 ml water) or fat (40 kJ corn oil). The animals had been conditioned to daily intubation and this procedure has only transient effects on VO,. Values were taken only when the animals were resting (see Stock, 1979) and have been corrected for body size (kg’.“). On the ninth day, the rats did not receive ciglitazone and were killed at 10.00 h by cervical dislocation and the interscapular BAT depot dissected, weighed and homogenised in 0.2 M sucrose. Mitochondria were prepared (Slinde et al., 1975) and the activity of the thermogenic mitochondrial proton conductance pathway (Nicholls, 1979) was assessed from the binding of [3H]guanosine diphosphate (2 PM GDP, 10 Ci/mmol; Amersham International, Bucks, U.K.) with or without excess (200 PM) unlabelled nucleotide (Brooks et al., 1982). Tissue and mitochondrial protein content was determined by a dye-reagent method (BioRad, Watford, U.K.). Experiment 2 Eight rats were treated with ciglitazone and eight with vehicle alone (as above) for 14 days, and a further six weight-matched rats were killed at the start of the experiment for determination of initial body energy content (Bo group). On day 7, resting VO, was measured for 2 h before and 2 h after injection of the adrenergic antagonist propranolol (20 mg/kg s.c.). At the end of this experiment, the rat carcasses were frozen, chopped and freeze-dried. Body energy content was determined by bomb calorimetry on triplicate, homogenised samples and body energy gain calculated from the final body energy content minus the initial energy content, assessed from the energy density of the Bo group. Energy expenditure was calculated from the difference between energy intake and gain, and gross efficiency as the gain per unit intake. Net energetic efficiency was estimated from the body energy gain per unit intake above maintenance, using the interspecific mean value of 420 kJ/ kg0-75/day for maintenance (Kleiber, 1961). Values have been presented as means + SEM. Significant differences between groups were as-
255
sessed using Student’s t-test for matched matched data and two-tailed probabilities.
or un-
2
POSTPRANDIAL TRAGASTRIC IN CONTROL (EXPT. 1)
Results In Experiment 1, ciglitazone caused a slight, but insignificant reduction in body weight gain (Table 1) and did not significantly affect energy intake, even when values were corrected for the slight difference in body size (ME intake control = 785 + 20, ciglitazone = 730 + 15 kJ/kg0-7s/ day). However, feed efficiency (g weight gain per unit energy intake) was reduced by almost 20% in ciglitazone-treated rats. During the course of the experiment, resting VO, measured at thermoneutrality did not differ significantly between groups either before or after feeding (Table 2). Gastric intubation with either carbohydrate or fat stimulated metabolic rate in all rats, with peak values occurring at about 60-90 and 80-120 min for the two nutrients, respectively. Fat and carbohydrate elicited similar thermic responses in control rats (approximately 12% increases in VOZ) but carbohydrate feeding produced an increase in VO, in ciglitazone-treated rats that was larger than in control animals and greater than the response to fat (Table 2). Measurements at the end of the study revealed that the interscapular BAT mass
TABLE
TABLE
RESPONSES (VO,) TO 40 kJ INMEALS OF FAT OR CARBOHYDRATE AND CIGLITAZONE-TREATED RATS
Mean values + SEM (n = 8). Control
a
Ciglitazone
a
Carbohydrate Preprandial VO, Postprandial VO, Increment (8 Increase)
14.8 f0.3 16.5 f0.3 1.7kO.2 (11.5_+1.3)
14.15 0.5 16.8 + 0.4 2.5fO.l ** (18.0+0.5 **)
Far Preprandial VO, Postprandial VO, Increment (% Increase)
15.3 +0.3 17.1 io.4 1.8kO.2 (11.8fl.O)
14.8 + 0.4 16.7 f 0.6 1.9+0.2 + (13.05 1.2 ++ )
’ ml/min/kg’ 75. * *P -c 0.001 compared to control values. + P < 0.05, ++ P < 0.001 compared to carbohydrate (paired t-test).
response
was not significantly affected by ciglitazone treatment, but the tissue protein content was elevated, and the thermogenic activity, assessed from mitochondrial GDP binding was increased by 25% in the ciglitazone-treated rats (Table 1). In the second experiment, which was of a longer duration, ciglitazone again failed to affect ME intake (Table 3) but caused a significant reduc-
1
BODY WEIGHT, ENERGY INTAKE AND INTERSCAPULAR BAT IN CONTROL AND CIGLITAZONE-TREATED RATS (EXPT. 1) Mean values + SEM (n = 8). Control Final body weight (g) Body weight gain (g) ME intake (kJ) Feed efficiency (g gain/MJ eaten) Interscapular BAT Mass (mg) Protein (mg) (W) Mitochondrial (pmol/mg
GDP binding protein)
+ 4 + 4
1630
+45
33.2i
1695 1.7
200 i13 17.4* 0.9 s.5* 0.4
48
* P c 0.05, * * P i 0.001 compared
+
187 46
1
to control
4 3 rt 3 f35
26.8+
1.5 *
231 +13 20.9i 0.7 * 9.2+ 0.3
60
3
BODY WEIGHT AND ENERGY TROL AND CIGLITAZONE-TREATED
k
values.
BALANCE IN CONRATS (EXPT. 2)
Mean values* SEM (n = 8). See Methods gross and net efficiency.
Cightazone
194 54
TABLE
3**
Control Initial body weight (g) Final body weight (g) Body weight gain (g) Energy intake (kJ) (kJ/kg0.75/day) Body energy gain (kJ) Energy expenditure (kJ) (kJ/kg0-75/day) Gross efficiency (S) Net efficiency (W)
147 248 101
Cightazone * 2 & 5 + 4
4045 +60 900 +_lO 750 _+30 3 290 ? 30 735 + 5 18.5+ 0.7 34.9& 1.8
* P < 0.05, * * P i 0.01 compared
for calculation
147 236 89
f 3 + 3 *2*
3 860 +10 895 +20 585 +30 ** 3215 +45 760 +lO * 15.1* 0.7 ** 28.5* 1.6 *
to control
values.
of
256
cl
Before After
CONTROL
Propranolol
ClGLlTAZONE
Fig. 1. Resting oxygen consumption (VO,, ml/min/kg0.75) of control and ciglitazone-treated rats, before and after a single injection of propranolol (20 mg/kg). Mean values, bars denote SEM(n=8).
tion in body weight gain (12%) and body energy gain (22%). The slightly lower energy density of gain (control = 7.4 + 0.2, ciglitazone-treated = 6.5 f 0.3 kJ/g) suggests that the reduced body energy content was due to less fat being deposited in ciglitazone-treated rats. Total energy expenditure over the 1Cday experiment did not differ between groups, but values corrected for body size (kg0.75) were elevated in the treated rats, and both gross and net energetic efficiency were reduced by ciglitazone. Resting VO, in ciglitazone-treated rats from this experiment was increased by 12.4% (Fig. l), but this difference was abolished by injection of the fl-adrenergic antagonist propranolol.
Discussion Ciglitazone has been shown to improve insulin resistance in genetically obese rodents, but is claimed to have little effect in lean animals (Chang et al., 1983a-c), and its mechanism of action is largely unknown. In addition, Mercer and Trayhurn (1986) have now demonstrated that this drug normalizes the acute cold-induced increase in brown fat activity of obese mutants, suggesting that insulin resistance of brown adipose tissue is an important factor in the impaired thermogenic responsiveness of obese mice. The results of the present study extend these findings to normal, lean rats, and demonstrate
that ciglitazone stimulates thermogenesis and BAT activity in the absence of a cold stimulus, and this causes significant changes in energy balance. In both experiments, ciglitazone caused slight reductions in weight gain, which were larger and statistically significant in the second (longer) experiment. Drug treatment did not affect energy intake, so it is presumed to have exerted its action on body weight via effects on energetic efficiency, i.e. thermogenesis. This was evident in Experiment 2, where energy expenditure and resting VO, (corrected for body size) were both enhanced by ciglitazone, and energetic efficiency was reduced. The effects of propranolol on VO, indicate that these increases in metabolic rate were mediated by the sympathetic nervous system and may therefore have involved activation of thermogenesis in BAT. The high rates of heat production in brown fat are due mainly to uncoupling of oxidative phosphorylation via a unique mitochondrial proton conduction pathway (Nicholls, 1979). Mitochondrial GDP binding is used routinely to assess the activity of this pathway, and levels of binding were significantly elevated in ciglitazone-treated rats. Mercer and Trayhurn (1986) found no effect of ciglitazone on GDP-binding in lean mice, but a 50% increase in obese mice at 22°C. This discrepancy between the two studies may result from the fact that our rats, although normally considered lean, have a relatively high fat content (over 10%) and may have developed a slight degree of insulin resistance. Certainly by this age (40 days), brown fat activity and the thermic responses to diet begin to decline somewhat from levels seen at, or soon after weaning (see Rothwell and Stock, 1983, 1984 for reviews). Adaptive DIT results largely from sympathetic activation of BAT, and in many animals the acute thermic response to food is also partly due to sympathetic activation of heat production (Rothwell et al., 1982; Rothwell and Stock, 1983). The acute thermic response to carbohydrate is dependent on insulin release, whereas the effect of fat does not require an increase in insulin levels and is not affected by treatment with diazoxide to inhibit insulin release (Rothwell et al., 1985). This probably explains why ciglitazone treatment enhanced the thermic response to carbohydrate but not to fat, and suggests that ciglitazone should exert its
251
greatest effects on energy balance in animals consuming a high carbohydrate diet. Ciglitazone causes amelioration of insulin resistance independently of changes in body weight (Mercer and Trayhurn, 1986) probably acting at the level of the insulin receptor or post-receptor mechanism (Chang et al., 1983a; Kobayashi et al., 1983). In the present study, the effects of ciglitazone on overall insulin sensitivity were not studied for several reasons. The first is that the primary object was to determine whether this drug could affect thermogenic activity in normal, lean animals before assessing its mode of action. Secondly, and unlike genetically obese rodents, differences in whole body insulin sensitivity in these young ciglitazone-treated animals are likely to be small (Chang et al., 1983a-c), and therefore difficult to detect and could obscure more marked changes in specific tissues related to the activation of thermogenesis. However, having established an enhanced thermogenic responsiveness in normal animals treated with ciglitazone, the next step will be to determine its mode and site of action. Acknowledgements We are grateful to Ian Connolley, Mike Lacey and Nick Busbridge for their excellent technical assistance, and to Dr. A.Y. Chang (Upjohn, Kalamazoo, U.S.A.) for the gift of ciglitazone. This work was supported by a grant from ICI plc. NJR is supported by a Royal Society University Research Fellowship.
References Andrews, P.L.R., Rothwell, N.J. and Stock, M.J. (1985) J. Physiol. 362, 1-12. Brooks, S.L., Rothwell, N.J. and Stock, M.J. (1982) Q. J. Exp. Physiol. 67, 259-268. Chang, A.Y., Wyse, B.M., Gilchrist, B.J., Peterson, T. and Diani, A.R. (1983a) Diabetes 32, 803-838. Chang, A.Y., Wyse, B.M. and Gilchrist. B.J. (1983b) Diabetes 32, 839-845. Chang, A.Y., Wyse, B.M. and Gilchrist, B.J. (1983~) Diabetologia 25, 514-520. Kleiber, M. (1961) The Fire of Life. J. Wiley & Sons, New York, p. 454. Kobayashi, M., Iwasaki, M., Ohgaku, S., Maegawa, H., Watanabe, N. and Shigeta, Y. (1983) FEBS Lett. 163, 50-53. Mercer, SW. and Trayhum, P. (1984) Biosci. Rep. 4, 933-940. Mercer, S.W. and Trayhum, P. (1986) FEBS Lett. 95, 12-16. Rothwell, N.J. and Stock, M.J. (1981) Metabolism 30, 673-678. Rothwell, N.J. and Stock, M.J. (1983) Metabolism 32, 371-376, Rothwell, N.J. and Stock, M.J. (1984) In: Recent Advances in Physiology, Vol. 10, Ed.: P.F. Baker (Churchill Livingstone, Edinburgh) pp. 349-384 Rothwell, N.J., Saville, ME. and Stock, M.J. (1982) Int. J. Obesity 6, 53-59. Rothwell, N.J., Stock, M.J. and Warwick, B.P. (1983) J. Physiol. 340, 63P. Rothwell, N.J., Stock, M.J. and Warwick, B.P. (1985) Metabolism 34, 43-47. Seydoux, J., Trimble, E.R., Bouillaud, F., AssimacopoulosJeannet. F., Bas, S., Ricquier, D., Giacobino, J.P. and Girardier, L. (1984) FEBS Lett. 166, 141-145. Slinde, E., Pederson, J.J. and Flatmark, T. (1975) Anal. Biothem. 65, 581-585. Stock, M.J. (1985) J. Appl. Physiol. 39, 849-850.
57 (1987) 253-257 Ltd.
253
MCE 01661
Effects of ciglitazone on energy balance, thermogenesis and brown fat activity in the rat Nancy J. Rothwell, Department
of Physiology,
St. George’s Hospilal Medical School, Cranmer Terrace, Tooting, London S WI 7 ORE, U.K. (Received
Key words: Ciglitazone;
Insulin;
Michael J. Stock and Alison E. Tedstone
Thermogenesis;
17 December
1986; accepted
Brown adipose
tissue;
12 February
Energy
1987)
balance
Summary Young male rats were treated with vehicle or ciglitazone (150 mg/kg/day, intragastric) for 8 or 14 days. Drug treatment did not affect food intake but reduced body weight and energy gains over 14 days, and significantly depressed energetic efficiency. Energy expenditure and resting oxygen consumption (VO,), when corrected for body size, were elevated in ciglitazone-treated rats, but the difference in VO, was abolished by treatment of the animals with a /3-adrenergic antagonist (propranolol). The acute thermic response (postprandial rise in VO,) to a fat meal was similar for both groups, but the response to carbohydrate ingestion was greater in ciglitazone-treated rats (18%) than controls (11.5%). The mass of interscapular brown adipose tissue was not affected by drug treatment, but its protein content was increased and its thermogenic activity (mitochondrial purine nucleotide binding) was elevated by 25% after chronic treatment with ciglitazone. These results indicate that ciglitazone enhances thermogenesis via sympathetic activation of brown adipose tissue, probably as a result of improved insulin sensitivity.
Introduction Insulin is generally described as an anabolic hormone promoting synthesis and deposition of lipid, protein and carbohydrate (glycogen), and in excess quantities it may produce obesity. However, there is now increasing evidence to suggest that insulin is involved in energy dissipation and can increase sympathetic activity and thermogenesis. Regulatory increases in metabolic rate occur in response to acute ingestion of food or chronic hyperphagia, and this diet-induced thermogenesis
Address for correspondence: Professor M.J. Stock, Department of Physiology, St. George’s Hospital Medical School, Cranmer Terrace, Tooting, London SW17 ORE, U.K. 0303-7207/87/$03.50
0 1987 Elsevier Scientific
Publishers
Ireland.
(DIT), which can be distinguished from the obligatory energy costs of nutrient digestion, absorption and assimilation, results from sympathetic activation of heat production, mainly in brown adipose tissue (BAT, see Rothwell and Stock, 1983, 1984, for reviews). DIT allows young, lean animals to avoid obesity during periods of hyperphagia, but the level of thermogenesis and BAT activity are severely inhibited by streptozotocin-induced diabetes (Rothwell and Stock, 1981) or by subdiaphragmatic vagotomy (Andrews et al., 1985), but restored to normal by insulin replacement. Acute inhibition of insulin release with d&oxide also attenuates the thermic response to a single meal of carbohydrate, but not fat (Rothwell et al., 1983, 1985). Conversely, several groups have demonstrated that Ltd
254
chronic infusion of insulin in rats activates BAT via the sympathetic nervous system (Seydoux et al., 1984; Rothwell and Stock, 1984) although with very high doses of insulin these thermogenic actions may be masked by anabolic effects of the hormone. Insulin is also implicated in the genetic obesity seen in laboratory rodents, where there is increased energetic efficiency associated with impaired thermogenesis and brown fat activity. Mercer and Trayhurn (1984) have shown that the diminished thermogenic activity in obese (ob/ob) mice may be related to insulin resistance, and have further demonstrated that amelioration of this insulin resistance with the oral hypoglycaemic drug ciglitazone largely restores the defective BAT response to cold in the obese mutant to the level of lean animals (Mercer and Trayhurn, 1986). We have therefore investigated the effects of this drug ciglitazone on energy balance, BAT activity and thermic responses to fat or carbohydrate in normal, non-obese rats. The results are consistent with the idea of an insulin involvement in thermogenesis and suggest that improving insulin sensitivity with drugs such as ciglitazone enhances DIT. Materials and methods The animals used in this study were male, Sprague-Dawley SPF rats (Charles River, Kent, U.K.) aged 35-40 days (140-150 g), which were caged in pairs at 24°C with free access to water and a commercial pelleted stock diet (PRD, Christopher Hill Group, Dorset, U.K.) supplying 27% of energy as protein, 9% as fat and 64% as carbohydrate. Metabolizable energy (ME) intake of the diet was assessed from the weight of food consumed and its ME density (12.0 kJ/g) determined in previous feeding trials. Rats were treated daily at 09.00 h with either ciglitazone suspended in water (150 mg/kg/day by gastric intubation) or vehicle (controls). Experiment 1 ME intake and body weight were recorded daily in eight control and eight ciglitazone-treated rats for 8 days. On day 5 and 7 the treatment was delayed until 15.00 h and resting oxygen consumption (VO,) was measured (10.00-14.00 h) in
closed-circuit respirometers (Stock, 1975) for 2 h before and 2-3 h after gastric intubation with either carbohydrate (40 kJ cornflour in 4 ml water) or fat (40 kJ corn oil). The animals had been conditioned to daily intubation and this procedure has only transient effects on VO,. Values were taken only when the animals were resting (see Stock, 1979) and have been corrected for body size (kg’.“). On the ninth day, the rats did not receive ciglitazone and were killed at 10.00 h by cervical dislocation and the interscapular BAT depot dissected, weighed and homogenised in 0.2 M sucrose. Mitochondria were prepared (Slinde et al., 1975) and the activity of the thermogenic mitochondrial proton conductance pathway (Nicholls, 1979) was assessed from the binding of [3H]guanosine diphosphate (2 PM GDP, 10 Ci/mmol; Amersham International, Bucks, U.K.) with or without excess (200 PM) unlabelled nucleotide (Brooks et al., 1982). Tissue and mitochondrial protein content was determined by a dye-reagent method (BioRad, Watford, U.K.). Experiment 2 Eight rats were treated with ciglitazone and eight with vehicle alone (as above) for 14 days, and a further six weight-matched rats were killed at the start of the experiment for determination of initial body energy content (Bo group). On day 7, resting VO, was measured for 2 h before and 2 h after injection of the adrenergic antagonist propranolol (20 mg/kg s.c.). At the end of this experiment, the rat carcasses were frozen, chopped and freeze-dried. Body energy content was determined by bomb calorimetry on triplicate, homogenised samples and body energy gain calculated from the final body energy content minus the initial energy content, assessed from the energy density of the Bo group. Energy expenditure was calculated from the difference between energy intake and gain, and gross efficiency as the gain per unit intake. Net energetic efficiency was estimated from the body energy gain per unit intake above maintenance, using the interspecific mean value of 420 kJ/ kg0-75/day for maintenance (Kleiber, 1961). Values have been presented as means + SEM. Significant differences between groups were as-
255
sessed using Student’s t-test for matched matched data and two-tailed probabilities.
or un-
2
POSTPRANDIAL TRAGASTRIC IN CONTROL (EXPT. 1)
Results In Experiment 1, ciglitazone caused a slight, but insignificant reduction in body weight gain (Table 1) and did not significantly affect energy intake, even when values were corrected for the slight difference in body size (ME intake control = 785 + 20, ciglitazone = 730 + 15 kJ/kg0-7s/ day). However, feed efficiency (g weight gain per unit energy intake) was reduced by almost 20% in ciglitazone-treated rats. During the course of the experiment, resting VO, measured at thermoneutrality did not differ significantly between groups either before or after feeding (Table 2). Gastric intubation with either carbohydrate or fat stimulated metabolic rate in all rats, with peak values occurring at about 60-90 and 80-120 min for the two nutrients, respectively. Fat and carbohydrate elicited similar thermic responses in control rats (approximately 12% increases in VOZ) but carbohydrate feeding produced an increase in VO, in ciglitazone-treated rats that was larger than in control animals and greater than the response to fat (Table 2). Measurements at the end of the study revealed that the interscapular BAT mass
TABLE
TABLE
RESPONSES (VO,) TO 40 kJ INMEALS OF FAT OR CARBOHYDRATE AND CIGLITAZONE-TREATED RATS
Mean values + SEM (n = 8). Control
a
Ciglitazone
a
Carbohydrate Preprandial VO, Postprandial VO, Increment (8 Increase)
14.8 f0.3 16.5 f0.3 1.7kO.2 (11.5_+1.3)
14.15 0.5 16.8 + 0.4 2.5fO.l ** (18.0+0.5 **)
Far Preprandial VO, Postprandial VO, Increment (% Increase)
15.3 +0.3 17.1 io.4 1.8kO.2 (11.8fl.O)
14.8 + 0.4 16.7 f 0.6 1.9+0.2 + (13.05 1.2 ++ )
’ ml/min/kg’ 75. * *P -c 0.001 compared to control values. + P < 0.05, ++ P < 0.001 compared to carbohydrate (paired t-test).
response
was not significantly affected by ciglitazone treatment, but the tissue protein content was elevated, and the thermogenic activity, assessed from mitochondrial GDP binding was increased by 25% in the ciglitazone-treated rats (Table 1). In the second experiment, which was of a longer duration, ciglitazone again failed to affect ME intake (Table 3) but caused a significant reduc-
1
BODY WEIGHT, ENERGY INTAKE AND INTERSCAPULAR BAT IN CONTROL AND CIGLITAZONE-TREATED RATS (EXPT. 1) Mean values + SEM (n = 8). Control Final body weight (g) Body weight gain (g) ME intake (kJ) Feed efficiency (g gain/MJ eaten) Interscapular BAT Mass (mg) Protein (mg) (W) Mitochondrial (pmol/mg
GDP binding protein)
+ 4 + 4
1630
+45
33.2i
1695 1.7
200 i13 17.4* 0.9 s.5* 0.4
48
* P c 0.05, * * P i 0.001 compared
+
187 46
1
to control
4 3 rt 3 f35
26.8+
1.5 *
231 +13 20.9i 0.7 * 9.2+ 0.3
60
3
BODY WEIGHT AND ENERGY TROL AND CIGLITAZONE-TREATED
k
values.
BALANCE IN CONRATS (EXPT. 2)
Mean values* SEM (n = 8). See Methods gross and net efficiency.
Cightazone
194 54
TABLE
3**
Control Initial body weight (g) Final body weight (g) Body weight gain (g) Energy intake (kJ) (kJ/kg0.75/day) Body energy gain (kJ) Energy expenditure (kJ) (kJ/kg0-75/day) Gross efficiency (S) Net efficiency (W)
147 248 101
Cightazone * 2 & 5 + 4
4045 +60 900 +_lO 750 _+30 3 290 ? 30 735 + 5 18.5+ 0.7 34.9& 1.8
* P < 0.05, * * P i 0.01 compared
for calculation
147 236 89
f 3 + 3 *2*
3 860 +10 895 +20 585 +30 ** 3215 +45 760 +lO * 15.1* 0.7 ** 28.5* 1.6 *
to control
values.
of
256
cl
Before After
CONTROL
Propranolol
ClGLlTAZONE
Fig. 1. Resting oxygen consumption (VO,, ml/min/kg0.75) of control and ciglitazone-treated rats, before and after a single injection of propranolol (20 mg/kg). Mean values, bars denote SEM(n=8).
tion in body weight gain (12%) and body energy gain (22%). The slightly lower energy density of gain (control = 7.4 + 0.2, ciglitazone-treated = 6.5 f 0.3 kJ/g) suggests that the reduced body energy content was due to less fat being deposited in ciglitazone-treated rats. Total energy expenditure over the 1Cday experiment did not differ between groups, but values corrected for body size (kg0.75) were elevated in the treated rats, and both gross and net energetic efficiency were reduced by ciglitazone. Resting VO, in ciglitazone-treated rats from this experiment was increased by 12.4% (Fig. l), but this difference was abolished by injection of the fl-adrenergic antagonist propranolol.
Discussion Ciglitazone has been shown to improve insulin resistance in genetically obese rodents, but is claimed to have little effect in lean animals (Chang et al., 1983a-c), and its mechanism of action is largely unknown. In addition, Mercer and Trayhurn (1986) have now demonstrated that this drug normalizes the acute cold-induced increase in brown fat activity of obese mutants, suggesting that insulin resistance of brown adipose tissue is an important factor in the impaired thermogenic responsiveness of obese mice. The results of the present study extend these findings to normal, lean rats, and demonstrate
that ciglitazone stimulates thermogenesis and BAT activity in the absence of a cold stimulus, and this causes significant changes in energy balance. In both experiments, ciglitazone caused slight reductions in weight gain, which were larger and statistically significant in the second (longer) experiment. Drug treatment did not affect energy intake, so it is presumed to have exerted its action on body weight via effects on energetic efficiency, i.e. thermogenesis. This was evident in Experiment 2, where energy expenditure and resting VO, (corrected for body size) were both enhanced by ciglitazone, and energetic efficiency was reduced. The effects of propranolol on VO, indicate that these increases in metabolic rate were mediated by the sympathetic nervous system and may therefore have involved activation of thermogenesis in BAT. The high rates of heat production in brown fat are due mainly to uncoupling of oxidative phosphorylation via a unique mitochondrial proton conduction pathway (Nicholls, 1979). Mitochondrial GDP binding is used routinely to assess the activity of this pathway, and levels of binding were significantly elevated in ciglitazone-treated rats. Mercer and Trayhurn (1986) found no effect of ciglitazone on GDP-binding in lean mice, but a 50% increase in obese mice at 22°C. This discrepancy between the two studies may result from the fact that our rats, although normally considered lean, have a relatively high fat content (over 10%) and may have developed a slight degree of insulin resistance. Certainly by this age (40 days), brown fat activity and the thermic responses to diet begin to decline somewhat from levels seen at, or soon after weaning (see Rothwell and Stock, 1983, 1984 for reviews). Adaptive DIT results largely from sympathetic activation of BAT, and in many animals the acute thermic response to food is also partly due to sympathetic activation of heat production (Rothwell et al., 1982; Rothwell and Stock, 1983). The acute thermic response to carbohydrate is dependent on insulin release, whereas the effect of fat does not require an increase in insulin levels and is not affected by treatment with diazoxide to inhibit insulin release (Rothwell et al., 1985). This probably explains why ciglitazone treatment enhanced the thermic response to carbohydrate but not to fat, and suggests that ciglitazone should exert its
251
greatest effects on energy balance in animals consuming a high carbohydrate diet. Ciglitazone causes amelioration of insulin resistance independently of changes in body weight (Mercer and Trayhurn, 1986) probably acting at the level of the insulin receptor or post-receptor mechanism (Chang et al., 1983a; Kobayashi et al., 1983). In the present study, the effects of ciglitazone on overall insulin sensitivity were not studied for several reasons. The first is that the primary object was to determine whether this drug could affect thermogenic activity in normal, lean animals before assessing its mode of action. Secondly, and unlike genetically obese rodents, differences in whole body insulin sensitivity in these young ciglitazone-treated animals are likely to be small (Chang et al., 1983a-c), and therefore difficult to detect and could obscure more marked changes in specific tissues related to the activation of thermogenesis. However, having established an enhanced thermogenic responsiveness in normal animals treated with ciglitazone, the next step will be to determine its mode and site of action. Acknowledgements We are grateful to Ian Connolley, Mike Lacey and Nick Busbridge for their excellent technical assistance, and to Dr. A.Y. Chang (Upjohn, Kalamazoo, U.S.A.) for the gift of ciglitazone. This work was supported by a grant from ICI plc. NJR is supported by a Royal Society University Research Fellowship.
References Andrews, P.L.R., Rothwell, N.J. and Stock, M.J. (1985) J. Physiol. 362, 1-12. Brooks, S.L., Rothwell, N.J. and Stock, M.J. (1982) Q. J. Exp. Physiol. 67, 259-268. Chang, A.Y., Wyse, B.M., Gilchrist, B.J., Peterson, T. and Diani, A.R. (1983a) Diabetes 32, 803-838. Chang, A.Y., Wyse, B.M. and Gilchrist. B.J. (1983b) Diabetes 32, 839-845. Chang, A.Y., Wyse, B.M. and Gilchrist, B.J. (1983~) Diabetologia 25, 514-520. Kleiber, M. (1961) The Fire of Life. J. Wiley & Sons, New York, p. 454. Kobayashi, M., Iwasaki, M., Ohgaku, S., Maegawa, H., Watanabe, N. and Shigeta, Y. (1983) FEBS Lett. 163, 50-53. Mercer, SW. and Trayhum, P. (1984) Biosci. Rep. 4, 933-940. Mercer, S.W. and Trayhum, P. (1986) FEBS Lett. 95, 12-16. Rothwell, N.J. and Stock, M.J. (1981) Metabolism 30, 673-678. Rothwell, N.J. and Stock, M.J. (1983) Metabolism 32, 371-376, Rothwell, N.J. and Stock, M.J. (1984) In: Recent Advances in Physiology, Vol. 10, Ed.: P.F. Baker (Churchill Livingstone, Edinburgh) pp. 349-384 Rothwell, N.J., Saville, ME. and Stock, M.J. (1982) Int. J. Obesity 6, 53-59. Rothwell, N.J., Stock, M.J. and Warwick, B.P. (1983) J. Physiol. 340, 63P. Rothwell, N.J., Stock, M.J. and Warwick, B.P. (1985) Metabolism 34, 43-47. Seydoux, J., Trimble, E.R., Bouillaud, F., AssimacopoulosJeannet. F., Bas, S., Ricquier, D., Giacobino, J.P. and Girardier, L. (1984) FEBS Lett. 166, 141-145. Slinde, E., Pederson, J.J. and Flatmark, T. (1975) Anal. Biothem. 65, 581-585. Stock, M.J. (1985) J. Appl. Physiol. 39, 849-850.