The effect of cold exposure (4°C) or prolonged norepinephrine infusion on the activity and mRNA levels of glycerokinase (GyK) was investigated in rat interscapular brown adipose tissue (BAT). Cold exposure for 12 and 24 h induced increases of 30% and 100%, respectively, in the activity of BAT GyK, which was paralleled by twofold and fourfold increase in enzyme mRNA levels. BAT hemidenervation resulted in reductions of 50% and 30% in GyK activity and in mRNA levels, respectively, in denervated pads from rats kept at 25°C, and suppressed in these pads the cold-induced increases in both GyK activity and mRNA levels. The increase in GyK activity induced by cold exposure was not affected by phenoxybenzamine, but was markedly inhibited by previous administration of propranolol or actinomycin D. BAT GyK activity did not change significantly after 6 h of continuous subcutaneous infusion of norepinephrine (20 μg/h), but increased twofold and fourfold after 12 and 24 h, with no further increase after 72 h of infusion. Norepinephrine infusion also activated mRNA production, but the effect was comparatively smaller than that on enzyme activity. β-Adrenergic agonists also stimulated GyK activity with the following relative magnitude of response: CL316243 (β3) > isoproterenol (non-selective) > dobutamine (β1). In vitro rates of incorporation of glycerol into glyceride-glycerol were increased in BAT from rats exposed to cold. The data suggest that in conditions of a sustained increase in BAT sympathetic flow there is a stimulation of GyK gene expression at the pretranslational level, with increased enzyme activity, mediated by β-adrenoreceptors, mainly β3.
- cold exposure
- norepinephrine infusion
- brown adipose tissue hemidenervation
- β-adrenergic agonists and antagonists
- actinomycin D
- glyceride-glycerol synthesis
activation of brown adipose tissue (BAT) thermogenesis requires the hydrolysis of endogenous triacylglycerols to produce fatty acids (FA), which are both substrates and uncoupling messengers for BAT mitochondria (13). Therefore, maintenance of adequate stores of triacylglycerols, through esterification, via glycerol-3-phosphate (G3P), of newly synthesized or preformed FA seems to be essential for the normal functioning of BAT. One of the sources of G3P for acylation is the glycerol produced by hydrolysis of stored triacylglycerols or taken up by the tissue from the circulation that is directly phosphorylated to G3P by glycerokinase (GyK). However, despite the relatively high levels of GyK in BAT (3), there is little information concerning the physiological control of this enzyme. An increased activity of GyK has been detected in BAT from rats submitted to a prolonged period of cold exposure (2), which is a well-known activator of sympathetic activity. Correspondingly, in experiments with rats adapted to a high-protein, carbohydrate-free diet, we found that the reduced BAT thermogenesis (4) and sympathetic activity (6) in these animals are accompanied by a marked decrease in the activity of BAT GyK (3, 15). Very recently (15) we showed that the marked increases in GyK activity induced by prolonged (10 days) or short (12 h) cold exposure were completely suppressed by previous BAT hemidenervation. Furthermore, we showed that BAT denervation induced in rats fed a balanced diet a 50% decrease in the activity of GyK of denervated pads and produced in rats adapted to the high-protein diet a further reduction of the already low levels of enzyme activity (15). These findings clearly indicate that an adequate sympathetic flow to BAT is required not only for the maintenance of normal levels of GyK activity but also for the enzyme response to situations in which the sympathetic activity is increased. In the above study (15), although the activity of BAT GyK changed in parallel with the changes in tissue sympathetic flux in the different experimental conditions, BAT GyK responses to the modifications of tissue sympathetic flux were very slow compared with the known rapid responses of triacylglycerols hydrolysis and thermogenesis activation, suggesting that the sympathetic nervous system acts by controlling the expression of the enzyme. One of the objectives of the present experiments was to test this hypothesis, investigating the changes in the activity and in BAT GyK mRNA levels induced by BAT denervation in rats kept at ambient temperature (25°C) or exposed to cold (4°C) for 12 and 24 h. Another objective of the present work was to investigate the effect of prolonged periods of norepinephrine and β-adrenergic agent infusion on the activity and mRNA levels of BAT GyK.
MATERIAL AND METHODS
Male Wistar rats weighing 180–220 g and maintained at 25 ± 2°C on a 12-h light-dark cycle were used in all experiments. They were fed a commercial nutritionally balanced diet containing (% wt/wt) 19% protein, 50% carbohydrate, and 5% lipid. In the cold-exposure experiments the animals were housed individually for a period of 12 or 24 h in a cold room (4°C), also maintained on a 12:12-h light-dark cycle. In some experiments, rats exposed to cold for 12 h were injected intraperitoneally 30 min before and 3, 6, and 9 h after cold exposure with either a β-adrenergic antagonist, propranolol (25 mg/kg body wt), or an α-adrenergic antagonist, phenoxybenzamine (10 mg/kg body wt). For tissue removal, the animals were always killed by cervical dislocation in the fed state between 8:00 and 10:00 AM.
Unilateral denervation of BAT.
Surgical hemidenervation was performed under ether anesthesia as described (15). The animals were used for the experiments 6 days after surgery. In earlier experiments (15) we found that the norepinephrine content of the denervated side, measured as described (9), was reduced to <2% of values in the control, innervated side.
Infusion of norepinephrine and β-adrenergic agonists.
Norepinephrine was infused subcutaneously at the rate of 1 μl (20 μg)/h for 6, 12, 24, or 72 h, with the help of an osmotic minipump (Alzet, model 2002), which was placed in the subcutaneous tissue laying directly over the interscapular BAT. Norepinephrine solution contained 10 mM sodium metabisulfite as an antioxidant. Control rats were infused with saline containing sodium metabisulfite. The same technique was used to infuse continuously for 3 days, at a rate of 1 μl/h, a saline solution containing one of the following β-adrenergic agonists (μg · kg body wt−1 · μl−1): dobutamine (80), isoproterenol (160), and CL316243 (40).
In vitro glyceride-glycerol and glyceride-FA synthesis.
The rats were killed and the interscapular BAT was removed, cleaned free of fat and muscle, and cut into small pieces of ∼5 mg. Portions of 200 mg were incubated for 2 h at 37°C in 2 ml of Krebs-Henseleit bicarbonate buffer, pH 7.4, containing [U-14C]glycerol (1 mM, 1 μCi). After incubation, the tissue fragments were washed several times with 0.9% NaCl and placed in 2:1 chloroform-methanol. The procedure used for isolation and counting of 14C-labeled glyceride-glycerol and glyceride-FA was as previously described (5).
Measurement of BAT GyK activity.
Tissue was homogenized in ice-cold 1% KCl and l mM EDTA. After centrifugation of the homogenate for 10 min at 2,000 g, 4°C, and removal of the top fat layer, the activity of GyK was measured as previously described (15). Protein content of homogenates was determined by the method of Lowry et al. (16).
Total RNA extraction.
Total RNA was isolated from 0.1-g portions of frozen BAT using Trizol reagent (GIBCO-BRL) and quantified spectrophotometrically. The integrity of RNA was evaluated by gel electrophoresis (17).
Primers for GyK cDNA (forward 5′-ATTCATGGCTTATTGGGAA-3′ and reverse 5′-TGGAATATACAGA- ATGTCTGC-3′) were based on the rat GyK gene sequence (GenBank Accession #Q63060). Primers for β-actin cDNA (forward 5′-TGGAATCCTGTGGCATCCATGAAC-3′ and reverse 5′-TAAAACGCAGCTCAGTAACAGTCG-3′) were based on β-actin gene sequence (GenBank Accession #Q63060). Total RNA was reverse-transcribed using the GyK and β-actin primers through the Thermoscript RT System (Invitrogen). PCR was then performed in a total volume of 50 μl containing 1 μM of GyK or β-actin primers, 2.5 units of Taq DNA polymerase, reaction buffer, and 1.5 mM of MgCl2 (Invitrogen). The fragments were amplified through 30 cycles of PCR consisting of denaturation at 94°C for 1 min, annealing at 50°C for 1 min, and extension at 72°C for 1.2 min. The length of the amplified fragments was 800 and 350 bp for GyK and β-actin, respectively. The amplified products were cloned directly into pGEM T Easy Vector (Promega), sequenced by the dideoxynucleotide chain termination method and analyzed on an ABI377 automated sequencer.
Total RNA (20 μg) was transferred to nitrocellulose membrane (Millipore) after gel electrophoresis. Prehybridization (4 h) and hybridization (16 h) were carried out at 65°C. After labeling with [32P]dCTP using random primer method, the amplified RT-PCR fragments were used as probe for GyK and β-actin transcripts. Membranes were washed as described by Sambrook et al. (17) and exposed to Hyperfilm XAR-5 Kodak at −70°C. The autoradiograms were quantified using the program ImageQuant (version 4.2, Molecular Dynamics). Levels of GyK mRNA were normalized to β-actin levels to account for differences in total mRNA. No significant difference was found in the relation β-actin mRNA/total mRNA in any of the experimental conditions. The results are expressed in arbitrary units. To compare the results, the data are related to controls, assigning value 1 to their means.
Results are expressed as means ± SE, and differences between means were analyzed by ANOVA, with P < 0.05 as the criterion of significance.
Figure 1 A shows the RT-PCR product obtained using the pair of the glycerokinase primers described in material and methods. The amino acid sequence exhibited 100% identity to the mice and human GyK (GenBank Gi6680139 and P32169, respectively). With the use of this cDNA, only one hybridization signal for BAT GyK mRNA was detected corresponding to a size of 2.1 kb in length (Fig. 1 B).
Figure 2 shows the results of measurements of activity and mRNA levels of GyK in rats previously submitted to BAT hemidenervation. As shown in Fig. 2 A, cold exposure induced marked increases in the levels of BAT GyK mRNA. After 12 and 24 h at 4°C, mRNA levels in BAT were about two and four times higher, respectively, than values in BAT from rats kept at 25°C. Figure 2 A also shows that hemidenervation reduced by ∼30% mRNA levels in rats maintained at ambient temperature and suppressed in rats kept at 4°C the cold-induced increase in GyK mRNA levels. The data in Fig. 2 B show that in innervated, intact pads, cold exposure for 12 h induced a 30% increase in the activity of GyK. After 24 h at 4°C, the activity of the enzyme was twice as high as that found in intact BAT pads from rats kept at ambient temperature. Figure 2 B also shows, in confirmation of previous results (15), that BAT hemidenervation resulted in a 50% reduction in the activity of GyK in denervated pads from animals kept at 25°C and suppressed in denervated pads from rats kept at 4°C the cold-induced increase in the activity of the enzyme. GyK mRNA levels in intact pads from hemidenervated rats before and after cold exposure did not differ significantly from levels in BAT from nonoperated rats (data not shown). To verify the effect of inhibition of mRNA production on the cold-induced increase in BAT GyK activity, rats were injected intraperitoneally with actinomycin D (1.5 mg/kg body wt), a well-known transcription inhibitor, 15 min before being exposed to cold for 12 h. It was found that after this period, the activity of the enzyme was significantly lower (P < 0.05) in animals injected with the drug (nmol · mg protein−1 · min−1, 6.1 ± 1.1, 7 rats) than in saline injected controls (9.7 ± 0.7, 8 rats).
The results of the experiments with prolonged infusion of norepinephrine are shown in Fig. 3. The data show that the levels of BAT GyK mRNA increased ∼30% after 12 h, twofold after 24 h, and almost threefold after 72 h of norepinephrine infusion. The data in Fig. 3 also show that 6 h of norepinephrine infusion did not affect significantly the activity of BAT GyK. After 12 and 24 h of infusion there was an increase in the activity of the enzyme that was more marked than the increase produced in mRNA levels. Thus GyK activity increased approximately twofold and fourfold after 12 and 24 h, respectively. No further increase was obtained by prolonging the period of infusion, BAT GyK activity showing after 72 h the same fourfold increase observed after 24 h.
Figure 4 shows the results of the experiments with adrenergic antagonists and agonists. The data in Fig.4 A show that administration of propranolol, a nonspecific β-adrenergic antagonist, caused a 30% reduction in the increase in BAT GyK activity induced by 12 h of cold exposure. The cold-induced increase in enzyme activity was not affected by treatment with phenoxybenzamine, an α-adrenergic antagonist. Figure4 B shows the results of experiments in which β-adrenergic agonists were infused for 72 h. Infusion of dobutamine, a β1-agonist, induced a twofold increase in the activity of BAT GyK activity. The stimulatory effect of isoproterenol, a non-selective β-agonist, and of CL316243, a specific β3-agonist, was more marked, inducing approximately fourfold and sixfold increases, respectively, in the activity of the enzyme. No additive effect was obtained when these two compounds were administered together (Fig. 4 B).
The experiments in Fig. 5 were performed to verify if the metabolism of glycerol by BAT was affected by the cold-induced increase in the activity of GyK. The data show that exposure of rats to cold for 24 h induced a marked increase in the rate of incorporation of glycerol into glyceride-glycerol. Incorporation into glyceride-FA also increased in cold-exposed rats, but rates in both experimental groups were much smaller than rates of incorporation into glyceride-glycerol.
As detailed in the introduction, we recently provided strong evidence indicating that an adequate sympathetic flow to BAT is required for the maintenance of normal levels of GyK activity and for the enzyme response to situations, such as cold exposure, which markedly increase BAT sympathetic flow (15). In the present study we demonstrate that the activity of BAT GyK is also stimulated by norepinephrine. The stimulating effect of the catecholamine was detected only after 12 h of infusion, in agreement with our previous observation that the response of BAT GyK activity to changes in sympathetic flow is relatively slow, suggesting that inducers of enzyme expression, rather than allosteric factors, are involved in the activation of this enzyme (14). This hypothesis is supported by the data of the present work, which clearly show that in all situations examined the decrease (after denervation) or increase (after cold exposure or prolonged norepinephrine infusion) in BAT GyK activity was accompanied by parallel changes in the enzyme mRNA levels. The importance of an enhancement of mRNA production was also supported by the finding that previous administration of actinomycin D, a transcription inhibitor, markedly reduced the increase in enzyme activity induced by cold exposure. Although mRNA levels of proteins are not always directly associated with enzyme protein content, the magnitude of the effects observed, and the fact they were accompanied by parallel changes in GyK activity, strongly suggest concomitant changes in the expression of enzyme protein.
The increase in GyK mRNA levels induced by prolonged norepinephrine infusion (Fig. 3) was smaller than that induced by cold exposure (Fig.2 A). It is possible that the norepinephrine administered systemically, contrary to cold exposure, was not sufficient to produce the high concentrations of the neurohormone at the synaptic cleft required by the β3-adrenoreceptors (see below). Also, metabolic/hormonal changes induced by cold exposure, such as increased levels of thyroid hormones, could contribute to activate the expression of BAT GyK in cold-stressed rats. Independently of comparisons with enzyme mRNA levels in cold-exposed animals, another finding in the experiments with norepinephrine infusion was that, despite the relative small increase of transcripts levels, the catecholamine infusion markedly stimulated the activity of BAT GyK. The enzyme activity had already attained high maximal levels after 24 h of infusion, when mRNA levels were still increasing (Fig. 3). This raises the possibility of an additional action of norepinephrine at posttranscriptional level, such as an increase in enzyme protein half-life, which might precede its effect on mRNA production. Clearly, further experiments are needed to clarify these points, including Western blot analyses; as in the present study, only maximal activities of the enzyme, apparently in optimal conditions, were measured.
In agreement with the notion that α-adrenoreceptors (ARs) play only a minor role in BAT thermogenic/metabolic functions (7), the cold-induced increase in BAT GyK activity was not affected by phenoxybenzamine, an α-antagonist, but was partially inhibited by propranolol, a nonselective β-antagonist. Although the three β-agonists used activated the enzyme, the highest levels of BAT GyK activity were obtained with CL316243, a specific β3-agonist, and the lowest with dobutamine, a β1-agonist. The enzyme levels in BAT from rats infused with isoproterenol, a nonspecific β-agonist (Fig. 4 B), were similar to those in animals infused with norepinephrine, another mixed β-agonist, infused for the same period (Fig. 3). A similar relative magnitude of response to these compounds has been found for stimulation of lipolysis and respiration of brown adipocytes (1). The present data provide strong evidence of a direct sympathetic control of BAT GyK activity and seem to support the view that the low-affinity, desensitization-resistant β3-AR may represent the physiological important receptors for the high concentrations of norepinephrine in the synaptic cleft. Especially during the relatively prolonged period of stimulation of the present experiments, the high-affinity β1-ARs are probably desensitized. β2-AR have been shown to be very scarce in typical brown adipocytes, most of them being present in other types of cell of BAT (8).
The increased in vitro rates of glyceride-glycerol synthesis by fragments of BAT from rats exposed to cold (Fig. 5) show that the activity of BAT GyK, which is usually associated with the recycling of glycerol produced by hydrolysis of stored triacylglycerols, may also be important for the esterification of glycerol taken up by the tissue from the circulation. During cold exposure there is an activation of white adipose tissue lipolysis (12) with increased release of glycerol to the plasma.
Recently (10, 11), strong evidence has been obtained that GyK has also a physiological role in two tissues considered as lacking significant levels of the enzyme, skeletal muscle and white adipose tissue. Thus it has been shown (11) in in vivo experiments with labeled glycerol infusion that plasma glycerol is an important precursor of G3P for triacylglycerol synthesis in skeletal muscle. While this paper was being prepared for publication, it was reported (9) that thiazolidinediones, antidiabetic agents that reduce free FA mobilization, markedly induce adipocyte GyK gene expression and stimulate glycerol incorporation into triacylglycerols.
The increase in the activity and expression of GyK in situations of sustained activation of BAT thermogenesis, and the consequent increased production of glycerol-3-P (G3P) for FA acylation, emphasize the importance of maintaining adequate triacylglycerol reserves for the normal functioning of BAT. However, the control of this process, in particular the supply of G3P, has been little studied. The only other traditionally recognized source of G3P, besides glycerol recycled from triacylglycerol hydrolysis and directly phosphorylated by GyK, is glucose, via dihydroxyacetone in the glycolytic pathway. We demonstrated (5) that BAT can also generate G3P from noncarbohydrate sources, via glyceroneogenesis, forming phosphoenolpyruvate via the dicarboxylic shuttle and subsequently G3P by a partial reversion of glycolysis. In fact, G3P generation through glyceroneogenesis is very active in BAT, being responsible for the majority of glyceride-glycerol synthesized by BAT in vivo (5).
BAT metabolism is controlled by a complex interaction of neural (sympathetic) and endocrine/metabolic factors. The present data and our previous studies suggest that the generation of G3P by GyK is controlled mainly or exclusively by the sympathetic outflow to the tissue, whereas BAT glyceroneogenesis is predominantly regulated by hormonal/metabolic factors. This was clearly seen in our experiments with rats adapted to a high-protein, carbohydrate-free diet. In these animals, reduced BAT thermogenesis (4) and sympathetic activity (6) are accompanied, in agreement with the present results, by very low levels of BAT GyK activity (3,15). In contrast, BAT glyceroneogenic activity is increased (5), probably because of the markedly reduced glycolytic flux (3) and reduced availability of dihydroxyacetone-P for G3P production. The findings above illustrate the need of further studies on the changes in the flux of G3P-generating pathways and the role of neural and hormonal/metabolic factors in different physiological conditions for a better understanding of the functional organization of BAT.
We thank V. D. Galban, N. M. Z. Resano, and E. Filippin for technical assistance.
This work was supported by grants from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 01/10050–8) and from the Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPQ 513296/96).
Address for reprint requests and other correspondence: R. H. Migliorini, Dept. of Biochemistry and Immunology, School of Medicine, 14049–900 Ribeirão Preto, SP, Brazil (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2003 the American Physiological Society