Shelagh Wilson. KEY WORDS: fl agonist; acetyl-coa carboxylase.

Transcription

1 Bioscience Reports, Vol. 9, No. 1, 1989 Effect of the fl-adrenoceptor Agonist BRL on Fatty Acid Synthesis and on the Activities ofpyruvate Dehydrogenase and Acetyl-CoA Carboxylase in Adipose Tissues of the Rat Shelagh Wilson Received June 7, 1988 BRL is a thermogenic fl-adrenoceptor agonist which stimulates lipolysis and fatty acid oxidation in vivo. It also stimulates insulin secretion, and hence promotes glucose utilisation in vivo. The effect of this agent on white and brown adipose tissue of the rat was investigated. BRL increased the rate of fatty acid synthesis in vivo in white adipose tissue by 135% but reduced the rate of fatty acid synthesis in vivo in brown adipose tissue by 78%. The increase was abolished in white adipose tissue of streptozotocin-diabetic rats, indicating that the effect involved a rise in circulating insulin levels. The reduction in fatty acid synthesis in brown adipose tissues was associated with a reduction in the activity of acetyl-coa carboxylase in the tissue consistent with a direct fl-adrenoceptor-mediated effect. BRL also increased the proportion of pyruvate dehydrogenase in its active form in vivo in brown adipose tissue and this increase was abolished in streptozotocin-diabetic rats. These findings illustrate different sensitivities of white and brown adipose tissues to combined fl-adrenergic and insulin stimulation. KEY WORDS: fl agonist; acetyl-coa carboxylase. brown adipose tissue; fatty acid synthesis; pyruvate dehydrogenase; INTRODUCTION It is generally accepted that fatty acid synthesis in both white and brown adipose tissues of the rat is stimulated in vivo by insulin. The mechanism involves parallel increases in the proportions of acetyl-coa carboxylase and pyruvate dehydrogenase which exist in their active forms (Stansbie et al., 1976, McCormack and Denton, 1977). In contrast, catecholamines generally (but not invariably) inhibit fatty acid synthesis in adipose tissues, and this mechanism is thought to involve fl-adrenoceptor mediated inactivation of acetyl-coa carboxylase activity (Denton and Halperin, 1968; Saggerson and Greenbaum, 1970; Beecham Pharmaceuticals Research Division, Biosciences Research Centre, Great Burgh, Yew Tree Bottom Road Epsom, Surrey KT18 5XQ /89/ /0 ~ 1989 Plenum Publishing Corporation

2 112 Wilson Brownsey et al., 1979; Agius and Williamson, 1980, Gibbins et al., 1985; but see Saggerson, 1972 and Shimazu and Takahishi, 1980). The effect of catecholamines on the activity of adipose tissue pyruvate dehydrogenase is unclear, as both increases (Gibbins et al., 1985; Kilgour and Vernon, 1986) and decreases (Coore et al., 1971; Smith and Saggerson, 1978) have been reported depending on the particular catecholamine, tissue, or condition investigated. BRL is a fl-adrenoceptor agonist which via its metabolite BRL exhibits low potency at classical/31 and f12 receptors but selectively stimulates the atypical ("f13") receptor in brown and white adipose tissues (Arch et al., 1984a; Wilson et al, 1984; Arch, 1989). In vivo, BRL stimulates lipolysis, fatty acid oxidation and brown fat thermogenesis (Arch et al., 1984b; Wilson et al., 1987). BRL also stimulates insulin secretion and hence promotes glucose utilisation in vivo, although lower doses are necessary for these effects compared with those required to produce a thermogenic effect in vivo (Sennitt et al., 1985). Thermogenic doses of BRL might thus be expected to have opposing effects on fatty acid synthesis in adipose tissues in vivo--a direct fl-adrenoceptormediated inhibitory effect and an indirect, insulin-mediated stimulatory effect. The aim of the present study was to determine the effect of BRL on fatty acid synthesis in both brown and white adipose tissues in vivo in order to study the response of these tissues to combined fl-adrenoceptor and insulin stimulation. In addition, changes in the activities of acetyl-coa carboxylase and pyruvate dehydrogenase in brown adipose tissue were investigated. MATERIALS AND METHODS Male Sprague-Dawley rats ( g) were housed at 23~ under a controlled light cycle (light h) and allowed free access to a standard rodent diet (Oxoid rodent breeding diet, H. C. Styles, Bewdley, Worcs., UK). Drugs were administered via oral (p.o.), intraperitoneal (i.p.), subcutaneous (s.c.) or intravenous (i.v.) routes. Rats were made diabetic by administration of streptozotocin (60 mg kg -~ body weight i.v. in 10mM sodium citrate, ph4.5) four days prior to use. Diabetes was confirmed 24 h later by determination of blood glucose levels. Rats with blood glucose levels of less than 15 mm were not used. Diabetic rats were maintained on slow-release insulin (Insulatard, The Wellcome Foundation, Crewe, UK, 2U per rat per day s.c. at 1800 h) until use. Control rats received vehicle throughout. All experiments took place between 0900 and 1100 h and fully-fed rats were housed at 28-29~ (the lower limit of the thermoneutral range) one hour prior to and during each experiment. Rates of fatty acid synthesis in vivo were determined by measuring the incorporation of tritium from 3H20 into tissue fatty acids. Rats were treated with BRL (5 mg. kg -1 p.o.) or water, 30 min later were treated with 3H20 (5 mci per rat i.p.) then after a further 30 min were killed by cervical dislocation and samples of blood, interscapular brown adipose tissue and epididymal white adipose tissue were removed rapidly. Rates of fatty acid synthesis were determined as described by Stansbie et al. (1976). The dose of BRL given

3 Effect of BRL on Fatty Acid Synthesis 113 was sufficient to stimulate thermogenesis in vivo maximally (64 + 7% increase in metabolic rate one hour after dosing (Wilson et al., 1987). For the determination of enzyme activities rats were dosed with BRL (5mg. kg -1 p.o.) or water, 30min later were anaesthetised (sodium pentobarbitone 60 mg-kg -1 i.p.) then after a further 30rain the interscapular brown adipose tissue was rapidly dissected and frozen in liquid nitrogen. Activities of pyruvate dehydrogenase and of acetyl-coa carboxylase were measured as described by Stansbie et al. (1976). Secretion of insulin in response to BRL was determined in anaesthetised rats (sodium pentobarbitone, 60 mg-kg -1 i.p.). The carotid artery was exposed, cannulated and kept patent with heparin (5 U 9 m1-1 in 0.9% w/v NaCI). An initial blood sample (1 ml) was removed, then rats were given BRL (5 mg. kg -~ i.p.). A further blood sample (1 ml) was removed after 60 min. Plasma for subsequent assay of insulin (Hales and Randle, 1963) was stored at -20~ Energy expenditure of control and diabetic rats was determined by indirect calorimetry as described by Arch and Ainsworth (1983). Rats were dosed with BRL (5 mg- kg -~ p.o.) or water. Unless otherwise stated, all chemicals were obtained from Sigma Chemical Co. (Poole, Dorset, UK). Radiochemicals were obtained from Amersham International (Little Chalfont, Bucks., UK). Pyruvate dehydrogenase phosphate phosphatase, arylamine acetyltransferase and p-(p-aminophenylazo) benzene sulphonic acid were kindly provided by Dr R. M. Denton, Department of Biochemistry, University of Bristol. BRL was synthesised in these laboratories and was dosed as the hemi-fumarate salt (Arch et al., 1984a). All results are expressed as means +SEM. Statistical significance was determined using "Student's" t-test (two-tailed). RESULTS BRL increased the rate of fatty acid synthesis in white adipose tissue by 135% during the minute period after dosing (Table 1). In contrast, the compound produced a 78% decrease in the rate of fatty acid synthesis in brown adipose tissue of the same animals over the same period of time (Table 1). The contribution of changes in insulin secretion to the observed changes in fatty acid synthesis was determined using rats made diabetic with streptozotocin 4 days before use and supplemented with slow-release insulin until the day before use. There was no significant difference between diabetic and control rats in their body weights or food intake (not shown). However, diabetic rats displayed a small reduction in basal plasma insulin levels (Table 2) and reduced basal rates of fatty acid synthesis (Table 1) suggesting that insulin replacement was not complete. Nevertheless, diabetic rats exhibited a normal thermogenic response to BRL even though plasma insulin levels did not rise significantly in response to the drug (Table 2). BRL had no effect on fatty acid synthesis in white adipose tissue of diabetic rats, although a reduction in fatty acid synthesis in brown adipose tissue was still observed (Table 1). The reduction in fatty acid synthesis in brown adipose tissue of intact (i.e.

4 '114 Wilson Table 1. Effect of BRL on rates of fatty acid synthesis in brown and white adipose tissue of intact and streptozotocin-treated rats Treatment Rate of fatty acid synthesis White adipose Brown adipose tissue tissue Control 5.1 5: BRL : 1.1" * Streptozotocin- 2.6 :t: diabetic +BRL : : 31 Rates of fatty acid synthesis are expressed as #mol of 31-I incorporated into lipid per h per g wet weight tissue. Each value is the mean of 6 determinations. * p < vs control t P < 0.05 vs streptozotocin treatment alone non-diabetic) rats was associated with a similar (75%) reduction in the activity of acetyl-coa carboxylase in the tissue. Activity was measured in tissue obtained one hour after dosing (Table 3). The decrease in activity was apparent both in fresh extracts of tissue ("Initial" activity) and after incubation of the extracts with 20 mm citrate. This procedure is thought to promote polymer 177 and partial activation of the enzyme but does not overcome the effects of fl-adrenoceptormediated inactivation (Brownsey et al., 1979; Gibbins et al., 1985). BRL also produced a reduction in the initial activity of the enzyme in diabetic (streptozotocin-treated) rats, but this reduction failed to reach statistical significance after incubation of extracts with citrate (Table 3). In contrast to its effects on the activity of acetyl-coa carboxylase, BRL increased the proportion of pyruvate dehydrogenase in its active form in brown adipose tissue (Table 3). Pyruvate dehydrogenase activity was measured in fresh extracts of tissue ("Initial" activity) and after treatment of extracts with Table 2. Effect of pre-treatment with streptozotocin on the insulin secretion and thermogenic response to BRL Plasma concentration Increase in energy Time after of insulin (#U - m1-1 of expenditure administration of human equivalents) J. min -1 % BRL (rain) Control 48 5: : 77* : Streptozotocindiabetic t 40 5: The increase in energy expenditure is the difference between BRL and water-dosed rats. Each value is the mean of 6 determinations. * p < 0.05 vs concentration at time 0. t P = 0.05 vs control rats.

5 Effect of BRL on Fatty Acid Synthesis 115 Table 3. Effect of BRL on activities of acetyl-coa carboxylase and pyruvate dehydrogenase in brown adipose tissue of intact and streptozotocin-treated rats Activity (U per g wet weight) Actyl-CoA carboxylase Pyruvate dehydrogenase After citrate Initial/Total Treatment Initial treatment Initial Total % Control i 0.8 +BRL :4.5 Streptozotocin : ,7 5:1.0 diabetic +BRL : 0.002t ,80 5: unit of enzyme activity is defined as the amount that will catalyse the formation of 1/~mol of product per rain at 30~ Each value is the mean of 7 determinations. * p < vs control. t P < 0.05 vs streptozotocin treatment alone. pyruvate dehydrogenase phosphate phosphatase in the presence of Mg Z+ and Ca 2+. Under these latter conditions the phosphorylated (inactive) form of the enzyme is converted into the active non-phosphorylated form, thus giving a measure of 'total' activity (McCormack and Denton, 1977). BRL had no effect on total activity of the enzyme (Table 3). Pre-treatment of rats with streptozotocin abolished the marked activiation of pyruvate dehydrogenase produced by BRL (Table 3). DISCUSSION Hormonal effects on fatty acid synthesis in brown and white adipose tissues have usually been regarded as being qualitatively similar. Thus, insulin stimulates lipogenesis in both tissues by mechanisms which involve parallel increases in the activities of acetyl-coa carboxylase and pyruvate dehydrogenase, whilst catecholamines inhibit lipogenesis in both tissues (Saggerson and Greenbaum, 1970; Agius and Williamson, 1980; Gibbins et al., 1985; Stansbie et al., 1976). The mechanism of the inhibitory effect of catecholamines, at least in white adipose tissue, is thought to involve /3-adrenoceptor-mediated phosphorylation and inactivation of acetyl-coa carboxylase (Lee and Kim, 1978; Brownsey et al., 1979). The inhibitory mechanism in brown adipose tissue remain to be confirmed but is thought to be similar (Gibbins et at., 1985). Administration of BRL in order to achieve combined/6-adrenoceptor and insulin stimulation in vivo produced opposite effects on fatty acid synthesis in white and brown adipose tissues of the same animal. An increase in fatty acid synthesis was observed in white adipose tissue which was abolished by prior treatment with streptozotocin, suggesting that the increase was mediated by the rise in circulating insulin levels produced by BRL (Tables 1 and 2). A reduction in fatty acid synthesis was observed in brown adipose tissue and this was probably largely due to a reduction in the proportion of acetyl-coa carboxylase in its active form (Tables 1 and 3). The inactivation was still seen

6 116 Wilson after treatment of extracts of the tissue with citrate, consistent with fladrenoceptor-mediated inactivation of the enzyme (Brownsey et al., 1979; Gibbins et al., 1985). The predominance of the stimulatory, insulin-mediated effect in white adipose tissue contrasts with the predominance of the inhibitory, fi-adrenoceptormediated effect in brown adipose tissue and this presumably reflects different sensitivities Of the tissues to each agent. This in turn may reflect the different metabolic functions of the two tissues: white adipose tissues is regarded mainly as a site for synthesis and storage of lipid whilst brown adipose tissue also has important thermogenic functions (Foster and Frydman, 1979). It should be noted that the dose of BRL given in the current studies (5 mg- kg -1) was high, sufficient to stimulate thermogenesis maximally and rapidly deplete brown adipose tissue of its store of lipid (Wilson et al., 1987). Administration of a much lower dose of BRL (0.8 mg. kg-1), sufficient to produce an increase in insulin levels but only low or insignificant increases in thermogenesis (Sennitt et al., 1985) has been found to increase the rate of fatty acid synthesis in brown adipose tissue of mice (M. A. Cawthorne, personal communication). The absence of any significant effect of BRL on the activity of pyruvate dehydrogenase in brown adipose tissue of diabetic rats (Table 3) suggests that the increase seen in intact rats is mediated by a rise in circulating insulin levels. The tissue thus still retains responsiveness to at least one of the effects of insulin when subjected to apparently overriding /6-adrenoceptor stimulation. Moreover, the absence of a significant effect of BRL on the activity of pyruvate dehydrogenase in diabetic rats suggests that fl-adrenoceptor stimulation does not produce any direct effect on the enzyme in vivo. A number of conflicting reports exist concerning the effect of catecholamines on the activity of adipose tissue pyruvate dehydrogenase: in white adipose tissue in vitro, adrenaline (probably acting via fi-adrenoceptors to promote lipolysis) produces a decrease in activity, an effect most notable in the presence of insulin (Coore et al., 1971; Smith and Saggerson, 1978). In vivo, however, administration of noradrenaline at doses sufficient to stimulate thermogenesis (and hence, presumably, lipolysis) in brown adipose tissue increases the activity of the enzyme in both white and brown adipose tissues (Gibbins et al., 1985; Kilgour and Vernon, 1986). The effect in white adipose tissue is claimed to be blocked by inhibition of either fi- and o:l-adrenoceptors (Kilgour and Vernon, 1986). Further work is required to reconcile these various observations, and probably awaits the development of a viable preparation of brown adipose tissue in vitro. The increase in pyruvate dehydrogenase activity coupled with the reductions in acetyl-coa carboxylase activity and lipogenesis seen in brown fat of intact rats suggests that only a small proportion of any glucose metabolised in the tissue via pyruvate dehydrogenase was converted to fatty acids under the conditions of maximal thermogenic stimulation by BRL This further suggests that oxidation of glucose may have increased, and this may have provided an additional source of energy during thermogenesis. However, it is unlikely that an increase in glucose oxidation would contribute a significant amount of energy to the thermogenic response to BRL since the response was unimpaired in

7 Effect of BRL on Fatty Acid Synthesis 117 streptozotocin-diabetic rats (Table 2). These results are consistent with those of Ma and Foster (1986) who have shown that (in cold-acclimated rats) the bulk of glucose taken up by brown adipose tissue undergoing maximal thermogenic stimulation is released from the tissues as pyruvate and lactate, leaving only a small proportion to be oxidised via pyruvate dehydrogenase. In summary, administration of BRL which produces combined ~-adrenoceptor and insulin stimulation in vivo had opposite effects on the rate of fatty acid synthesis in white and brown adipose tissues, and opposite effects on the activities of acetyl-coa carboxylase and pyruvate dehydrogenase in brown adipose tissue. ACKNOWLEDGEMENT I am very grateful to Dr R. M. Denton and Dr J. Gibbins for their helpful discussions and provision of material. REFERENCES Agius, L. and Williamson, D. H. (1980) Biochem. J. 190: Arch, J. R. S., Ainsworth, A. T., Cawthorne, M. A., Piercy, V., Sennitt, M. V., Thody, V. E., Wilson, C. and Wilson, S. (1984a) Nature (London) 309: Arch, J. R. S. and Ainsworth, A. T_ (1983) Am. J. Clin. Nutr. 38: Arch, J. R. S., Air, swarth, A. T., Ellis, R. D. M., Piercy, V., Thody, V. E., Thurlby, P. L., W~son, C., Wilson, S. and Young, P. (1984b) Int. J. Obesity 8:Suppl. l, [-ll. Arch, J. R. S. (1989) Proceedings of Nutrition Society (in press). Brownsey, R. W. and Denton, R. M. (1982) Biochem. J. 202: Brownsey, R. W., Hughes, W. A. and Denton, R. M. (1979) Biochem. J. 184: Coore, H. G., Denton, R. M., Martin, B. R. and Randle, P. J. (1971) Biochem. J. 125: Denton, R. M. and Halperin, M. L. (1968) Biochem. J. 110: Foster, D. O. and Frydman, M. L. (1979) Can. J. Physiol. Pharmac. 57: Gibbins, J. M., Denton, R. M. and McCormack, J. G. (1985) Biochem. J. 228: Hales, C. N. and Randle, P. J. (1963) Biochem. J. 88:137. Kilgour, E. and Vernon, R. G. (1986) Biochem. Soc. Trans. 14: Lee, K-H. and Kim, K-H. (1978) J. Biol. Chem. 253: Ma. S. W. Y. and Foster, D. O. (1986) Can. J. Physiol.Pharmacol. 64: McCormack, J. G. and Denton, R. M. (1977) Biochem. J. 1(~: Saggerson, E. D. (1972) Biochem. J_ 128: Saggerson, E. D. and Greenbaum, A. L. (1970) Biochem. J. 119: ~ Sennitt, M. V., Arch, J. R. S., Levy, A. L., Simson, D. L., Smith, S. A. and Cawthorne, M. A. (1985) Biochem. Pharm. 34: I Shimazu, T. and Takahishi, A. (1980) Nature (London) 284: Smith, S. J. and Saggerson, E. D. (1978) Biochem. J. 174: Stansbie, D., Brownsey, R. W., Crettaz, M. and Denton, R. M. (1976) Biochem. J. 160: Wilson, C., Wilson, S., Piercy, V., Sennitt, M. V. and Arch, J. R. S. (1984) European J. Pharm. 100: Wilson, S., Thurlby, P. L. and Arch, J. R. S. (1987) Can. J. Physiol. Pharmacol. 65: