were preacclimated to 28~ for 3-4 days with free access to food (R3, Ewos) and water ..... has been shown to bind specifically to the/3-adrenergic receptor (Levin.
Bioscience Reports, Vol. 6, No. 7, 1986
and ] -Adrenergic Control of Thermogenin mRNA Expression in Brown Adipose Tissue Anders Jacobsson, 1 Jan Nedergaard and Barbara Cannon Received July 3, 1986
KEY WORDS: brown adipose tissue; thermogenin; uncoupling protein; gene expression; adrenergic effects. By the use of an earlier characterised cDNA clone, CIN-1, corresponding to a sequence of the mRNA coding for the brown-fat specific "uncoupling" protein, thermogenin, the amount of thermogenin mRNA found in the brown adipose tissue of mice was quantitatively investigated under different physiological and pharmacological conditions. It was found that a 4 hr cold stress led to a 7-fold increase in the amount of thermogenin mRNA; injection of norepinephrine had a significant but smaller effect. Most notably, isoprenaline (fi-agonist) and phenylephrine (~-agonist) had in themselves no effect, but when injected together were able to increase the mRNA level synergistically. In 4 hr cold-stressed mice, norepinephrine, isoprenaline and cholera toxin could all further potentiate the effect of the cold stress itself on the mRNA level. Insulin and the glucocorticoid dexamethasone both had weak stimulatory effects on the mRNA level. It is concluded that an increase in intracellular cAMP levels is a necessary and perhaps sufficient stimulus for the increase in thermogenin gene expression. However, at least under in vivo conditions, this increase requires stimulation of both ~- and ]3adrenergic pathways.
INTRODUCTION A major challenge in brown adipose tissue research today is the question of the regulation of the expression of the brown-fat specific mitochondrial "uncoupling" protein, thermogenin, in such diverse physiological states as perinatal development The Wenner-Gren Institute, BiologihusF3, Universityof Stockholm,S-106 91 Stockholm,Sweden. 1 To whom correspondenceshould be addressed. 621 0144-8463/86/0700-0621505.00/0 ~2 1986Plenum Publishing Corporation
Jacobsson, Nedergaard,and Cannon
622
(Nedergaard et al., 1986), acclimation to cold (Himms-Hagen, 1986), and adaptation to certain diets (Rothwell and Stock, 1986). The recent isolation in several laboratories of cDNA clones corresponding to the mRNA encoding for thermogenin (Bouillaud et al., 1985; Jacobsson et al., 1985; Ridley et al., 1986) has therefore permitted a more direct approach to the question of the regulation of thermogenin synthesis in brown adipose tissue [for general reviews see Nedergaard and Lindberg (1982), Cannon and Nedergaard (1985), and Trayhurn and Nicholls (1986)]. In order to study this recruitment process, we have examined here the interaction between pharmacological agents (primarily adrenergic agents), and a short-term cold stress by measuring the level of thermogenin mRNA in the tissue. We conclude that under in vivo conditions a simultaneous ~- and fi-adrenergic stimulation is a requirement for stimulation of thermogenin gene expression, and that apparently an increase in cytosolic cAMP is able to initiate gene expression. Some of these results have earlier been published in abstract form (Jacobsson et al., 1986a,b).
MATERIALS AND METHODS Animals
Groups of 16 adult mice (about 8 weeks old with a body weight of about 30 g) of the NMRI strain were obtained from a local supplier (Eklunds, Stockholm). The mice were preacclimated to 28~ for 3-4 days with free access to food (R3, Ewos) and water. On the day of experiment, the mice were divided into two groups: eight mice remained at 28~ throughout and eight were transferred to 5~ Immediately at the transfer (if not otherwise stated), 4 mice in each group were injected with saline and 4 mice were injected with the agent under study. Four hours later, the mice were killed and the interscapular brown adipose tissue was excised. For studies with inhibitory agents (propranolol, cycloheximide and actinomycin D), the agents or the saline were injected 1 hr prior to the initiation of the cold stress, which was continued for 4 hr.
Injections
The mice were injected i.p. with solutions of the following agents. All solutions were made in 0.9% NaC1, and normally 100#1 was injected per mouse. 3 #mol norepinephrine per kg body weight (25 #g of arterenol bitartrate (Sigma) injected per mouse); 3 #tool phenylephrine (18 #g of the hydrochloride (Sigma)); 3 #mol isoprenaline (25 #g of isoproterenol bitartrate (Sigma)); 1 mg cholera toxin (25 #g (Sigma, pfs) dissolved in saline containing 1% fatty-acid-free bovine serum albumin; controls received saline plus albumin); 77 #tool propranolol (0.7 mg of DLpropranolol hydrochloride (Sigma)); 0.4 IU insulin (0.01 IU Insulin solution (KabiVitrum); l g glucose (25mg); 5.5rag dexamethasone (160#g Decadrone (MSD)); 30mg cycloheximide (0.9rag (Sigma)); 0.4mg actinomycin D (12#g Cosmegen (MSD) dissolved in saline).
Adrenergic Control of ThermogeninExpression
623
Preparation of RNA RNA was prepared as earlier described (Jacobsson et al., 1985). The interscapular brown adipose tissue from each animal was processed separately. The RNA preparation was routinely checked for degradation on an agarose minigel under UV light and checked for protein contamination; if the ratio A260/A280 was lower than 1.8, the preparation was discarded. The amount of RNA was calculated from the absorbance at 260 nm based on an extinction coefficient of 0.025 cm- 1 (mg RNA)- i. The total amount of RNA obtained from the brown fat of one mouse was in the range 50-100 #g.
Northern Blots Northern blots were performed essentially as described by Derman et al. (1981). Equal amounts of total RNA (7 #g) were applied to a 1.25 % agarose gel in a buffer consisting of 10 mM sodium phosphate (pH 6.5) with 3 % formaldehyde, and run for 5 hr at 100 V. After electrophoresis, the RNA was blotted overnight in 20 x SSC (i.e. 22 % NaC1 and 11% sodium citrate) to Zetaprobe blotting membrane (Biorad). These filters were baked for 2 hr at 80~ and stored in sealed plastic envelopes.
Hybridization The hybridization was performed, principally as described by Derman et al. (1981). The filters were first prehybridized at 42~ in a plastic envelope in a waterbath for at least 4 hr in 10 ml of a solution containing 50% formamide, 5 x SSC, 5 x Denhardt's solution, 50mM sodium phosphate (pH 6.5), 0.1 ~o SDS, 20#g/ml salmon sperm DNA (Sigma, phenol extracted), and 50 #g/ml of each poly A and poly C (Sigma). After the prehybridization, the filter papers were incubated for 48 hr at 42~ in 7 ml of the same medium (but with 2 x Denhardt's solution) with the thermogenin cDNA probe CIN-1 earlier identified (Jacobsson et al., t985). The probe was [32p]dCTP labelled by the use of the nick translation kit from BRL. The probe had a specific activity of about 108 cpm/#g DNA and about 500,000 dpm were added per mt hybridization liquid. The filters were then washed 4 times for 15 rain at room temperature in 2 x SSC (pH 7.0) with 0.2 % SDS, and then twice at 50~ for 1 hr in 0.1 x SSC with 0.2 % SDS and air dried at room temperature for 1 hr. The filters were then applied to X-ray films (Kodak X-omat XAR-5). After exposure (routinely overnight at - 80~ the resulting autoradiographs were measured in a laser scanning densitometer (LKB 2202 Ultrascan). The results were expressed as a % intensity, compared to that seen with a standard preparation of brown-fat RNA, run on each gel. This standard had been obtained as a pooled total RNA preparation from mice which had been exposed to cold for 17 hr.
RESULTS Using the previously characterised clone CIN-1 (Jacobsson et al., 1985), we
624
Jacobsson, Nedergaard, and Cannon
estimated the amount of thermogenin mRNA present in the brown adipose tissue of mice under different physiological and pharmacological conditions. We previously demonstrated that it is possible to observe a cold-induced increase in the amount of thermogenin mRNA in mice after only a few hours of cold stress (Jacobsson et al., 1985); similar results have also been obtained in rats with other clones for thermogenin (Ricquier et al., 1984). In the present investigation we used a 4 hr test period. A typical autoradiograph of Northern blots of the cDNA clone CIN-1 hybridized with RNA from control and cold-stressed mice is shown in Fig. 1A. As seen, exposure to cold for only 4 hr led to a large increase in the amount of thermogenin mRNA found in the tissue. The test period used here was not long enough to increase the amount of thermogenim mRNA sufficiently to show the (much weaker) extra band at 18S which has been observed in earlier investigations (Bouillaud et al., 1985; Jacobsson et al., 1985; Ridley et al. 1986). We have here compared this physiologically-induced increase in thermogenin
A. Northern blot 2 8 o 28O+NE
5~
5~
S
B. Densitometric tracings 28QC
28~
+ NE
5~
5~
+ NE
Fig. 1. Estimation of thermogenin mRNA levels. A. autoradiograph of Northern blots of brown-adipose-tissue RNA preparations hybridized with a 32P-labelled cDNA probe for thermogenin mRNA. RNA was individually prepared from the brown adipose tissue of each of 16 mice which had either been injected with saline and had remained at 28~ for 4 hr (4 mice), or which had been injected with norepinephrine and had remained at 28~ (4 mice), or which had been exposed to a 4 hr cold stress (5~ and saline injected (4 mice), or which had been both cold-stressed and norepinephrine injected (4 mice). S indicates a preparation of brown-adipose-tissue RNA, used to standardize the response between experimental series. B. Densitometric tracings of lanes from A.
Adrenergic Control of Thermogenin Expression
625
gene expression with that which can be obtained upon injection of pharmacological agents.
Effects of Adrenergic Agents We initially examined the ability of norepinephrine to induce thermogenin m R N A expression in brown adipose tissue. We used an amount of norepinephrine which has been shown to be sufficient to maximally stimulate thermogenesis in rodents (Jansky et al., 1969), and which is also able to increase the activity oflipoprotein lipase (Carneheim et al., 1984) and thyroxine deiodinase (Jones et aI., 1986) in the tissue to the same extent as is a 4 hr cold stress. It was clear from the autoradiographs obtained (Fig. 1) that norepinephrine in itself was able to increase the amount ofthermogenin m R N A , but not to the extent seen after only 4 hr cold stress. Even by visual inspection of the autoradiographs it was apparent that norepinephrine also had a positive effect on the amount of thermogenin m R N A found in the cold-stressed animals. For further quantitative evaluation of the results, the autoradiographs were analysed in a laser scanning densitometer. Typical densitometric tracings of some of the lanes presented in Fig. 1A are seen in Fig. lB. As is also evident from the autoradiographs, a minor but reproducible degradation of m R N A occurs. It is noteworthy that a specific degradation product of about t kb is invariably formed, indicating the presence of a specific nuclease-hypersensitive site. The area under each curve was related to that obtained with a standard preparation of brown fat RNA (obtained from a larger RNA preparation from mice cold-exposed for 17 hr). The results are shown in Fig. 2A. This quantitative analysis confirmed that norepinephrine injections in themselves led to a statistically significant doubling of the thermogenin m R N A level, but that the cold stress was much more efficient and led to a 7-fold increase in thermogenin m R N A amount. Further, and very interestingly, it is seen that norepinephrine was able to further potentiate the increase in thermogenin m R N A expression already induced by the 4 hr cold stress.
A. Norepinephrine
B. Isoprenaline
C. Phenylephrine
lOO. %
D. Propranolol
E. Cholera toxin
% _100
80
60
~176
4O
40
2O
0 --
+
--
+
1280128~ -- + +
--
+
-
Fig. 2. Interaction between adrenergic agents and cold exposure on the levels of thermogenin mRNA in mouse brown adipose tissue. Mice were injected with the indicated agents and exposed to a 4 hr cold stress, as described in Materials and Methods and as illustrated for the norepinephrine experiment in Figure 1. The values (in percent of the standard RNA preparation) are means + SE from 4 preparations, each made from one mouse. *, ** and *** indicate significanteffectsof agents (P < 0.05, 0.01 and 0.001, respectively;Student's t-test). The effectof cold stress as such has not been formallyindicated; as is evident, it was highlysignificantin every case (P < 0.01).
626
Jacobsson, Nedergaard, and Cannon
Adrenergic Classification In order to investigate the adrenergic subtype involved in the activation of thermogenin expression, we have compared different adrenergic agonists, injected in the same quantity as norepinephrine. The relatively fl-selective agonist isoprenaline was apparently unable to significantly affect the thermogenin mRNA level in control mice, but, like norepinephrine, it could potentiate the cold-induced increase in thermogenin mRNA (Fig. 2B). The relatively c~-selective agonist phenylephrine was also unable to affect thermogenin mRNA level in control animals, and further, it did not significantly increase the level in cold-stressed animals (Fig. 2C). The fl-selective antagonist propranolol had no significant effect on the mRNA level in the control state, but it had a marked inhibitory effect on the cold-induced increase in thermogenin mRNA level (Fig. 2D). A similar observation has been made in rats by Ricquier et al. (1984). (We have been unable to study the effect of comparable amounts of c~-antagonists due to severe detrimental effects of general a-blockade in animals exposed to cold.) The activator of adenylate cyclase, cholera toxin, was able to increase thermogenin mRNA expression both in control and in cold-stressed animals (Fig. 2E). Thus, in mice kept at 28~ only the dual (~- and fl-adrenergic) agonist norepinephrine, and the direct activator of adenylate cyclase cholera toxin, were able to increase thermogenin expression. However, in the physiologically stimulated tissue, this ability was extended to isoprenaline. A very noteworthy synergistic effect was also observed: the absolute increase in amount of thermogenin mRNA caused by the agents was always higher in the cold-stressed than in the control mice.
The Synergistic Effect of ~- and fi-Adrenergic Stimulation It is clear from the above that the ability of a fl-adrenergic stimulus to lead to thermogenin gene expression demands a simultaneous s-stimulation. In this respect, an interesting parallel can be drawn to the observations of Ma and Foster (1984). In whole animal studies (generally similar to those presented here) Ma and Foster observed that the amount of cAMP in the blood leaving the tissue could only be increased if the animals were simultaneously a- and fi-adrenergically stimulated. This synergism between ~- and fl-stir/aulation was also seen when whole animal heat production (oxygen consumption) was followed. Generally similar results have been obtained by Harris et al. (1986). We therefore carried out tests to establish whether there would also be synergistic effects of selective ~- and fl-adrenergic stimulation on the level of thermogenin mRNA. As seen in Fig. 3, this was the case. Isoprenaline or phenylephrine, when injected alone, were again without effect on the mRNA level, but when injected simultaneously, the two agents markedly increased the thermogenin mRNA level. There is thus a very marked synergistic interaction between c~- and fi-adrenergic stimulation on thermogenin mRNA expression in brown fat.
Adrenergic Control of Thermogenin Expression 100.
627
%
80. P < 0.001 60.
Fig. 3. Synergistic effect of ~- and /%adrenergic stimulation on thermogenin mRNA expression. The experiment was performed principally as illustrated in Figures 1 and 2, except that phenylephrine was increased to 4.4 #mol per kg body weight. The result here was evaluated with a slot blot procedure performed with a Schleicher and Schuell Minifold II slot blot as described by the manufacturers except that Zetaprobe was used, Three #g RNA was applied to each slot.
P < 0.001 40, 20 O.
PHE ISO PHE
+
|SO
Effects of Insulin and Glucocortieoids With regard to non-adrenergic agents, it appears from Fig. 4A that insulin had only a weak and, in itself, insignificant effect on thermogenin mRNA expression, both in control and cold-stressed animals. However, when the data were evaluated as a twodimensional analysis of variance, the effect of insulin was found to be significant (P < 0.05). It must be stressed that these experiments were performed with fed animals which probably already had an elevated insulin level, and it is of course likely that under other feeding conditions insulin could have a more marked effect on thermogenin gene expression. That insulin apparently had an effect even under the conditions used here is thus an indication that insulin, either directly (i.e. working on the brown fat cell itself), or indirectly (i.e. working via higher centers and sympathetic stimulation), can initiate thermogenin gene expression and thus lead to an elevation of thermogenin synthesis and amount (Seydoux et al., 1984). This effect of insulin may perhaps also be the reason behind the elevation of the amount of thermogenin seen, e.g. in cafeteria-fed rats (Brooks et al., 1980; Nedergaard et al., 1984), as has been discussed by Rothwell and Stock (1981). When an attempt was made to increase insulin levels physiologically by injection of glucose (as suggested, e.g. by McCormack and Denton (1977)), no effect on the mRNA level was seen (Fig. 4B). However, surprisingly, glucose injection into coldstressed animals led to a diminution of the cold-induced increase in mRNA level. The interaction between dietary and thermal stimuli in the regulation of thermogenin mRNA expression is being further examined in our laboratory. The synthetic glucocorticoid dexamethasone could increase the expression, but A. Insulin
B. Glucose
C. Dexamethesone
1(]0.
%1.100
80
.80
60
.60
40
40
20
-20
0 --
4-
-
+
--
+
-
+
-
+
-
+
.0
Fig. 4. Effects of insulin and glucocorticoids on thermogenin mRNA expression. The experiment was performed and evaluated as described in the legend to Figure 2.
628
Jacobsson, Nedergaard, and Cannon
only in the control state (Fig. 4C). This observation is apparently not in agreement with the general idea that corticoids have a depressing effect on brown adipose tissue activity and thermogenin content (especially in genetically obese rodents) (Holt and York, 1982, 1984; Rothwell et al., 1984), but it does agree with preliminary observations that prolonged dexamethasone treatment can increase thermogenin content (estimated as [3H]GDP binding capacity) (Nedergaard and Cannon, 1984).
Effects of Cycloheximide and Actinomycin D The inhibitor of RNA synthesis, actinomycin D, did not have any effect on the amount of thermogenin mRNA in the control state (Fig. 5A). However, it totally blocked the cold-induced increase. This indicates that the half-life of thermogenin mRNA in the control state exceeds 4 hr and that the increase in thermogenin mRNA content caused by cold stress is not due to a stabilization (increased half-life) of already-synthesized mRNA, but is indeed due to increased transcription of the gene. A. Actinomycin D B. Cycloheximide 100_ % % ~00 80
.8O
60. 40.
~ ~
40
20_
-
+ -
4-
--
+
--
+
0
Fig. 5. Effectsof cycloheximideand actinomycinD on thermogenin mRNA expression. The experiment was performed and evaluated as described in the legend to Figure 2.
Unexpectedly, the protein synthesis inhibitor cycloheximide had in itself, both in control and cold-stressed animals, a stimulatory effect on thermogenin gene expression (Fig. 5B). This effect is difficult to interpret in terms of an effect on protein synthesis. It may however be mentioned that in control experiments, where cycloheximide was tested for its ability to compete with the ligand [3H]CGP-12177 [which in brown adipose tissue has been shown to bind specifically to the/3-adrenergic receptor (Levin and Sullivan, 1986)], it was found that cycloheximide was a good competitor, and that cycloheximide, in concentrations in vitro which approximated those expected here in vivo, substantially inhibited the [3H]CGP-12177 binding. It is not yet clear whether this observation is directly related to the ability of cycloheximide to increase thermogenin mRNA expression demonstrated in the present work.
DISCUSSION We have examined here the ability of different pharmacological agents to mimic or interfere with the ability of a cold stress to induce an increase in thermogenin gene expression in mouse brown adipose tissue. Whereas treatments with insulin or dexamethasone had only a weak (if any) effect on gene expression, adrenergic stimulation was clearly able to increase the expression, but surprisingly not to the extent caused by a cold stress.
Adrenergic Control of ThermogeninExpression
629
The ability of/~-adrenergic agents to elevate the amount of thermogenin mRNA above that seen in the acutely cold-stressed state was an unexpected finding. This potentiating effect of adrenergic treatments on the mRNA level in cold stress occurs principally in contrast to the effect of such treatments on the enzymatic activities of, for example, lipoprotein lipase (Carneheim et al., 1984), and thyroxine deiodinase (Jones et al., 1986), which (in cold-stressed rat) cannot be further elevated by adrenergic stimulation. Although effects on gene expression and enzyme activity are not of course directly comparable, the potentiation observed here could seem to imply that the coldinduced increase in thermogenin gene expression occurs via a pathway other than an increase in adrenergic stimulation (and that an adrenergic stimulation can thus be added on top of this unknown stimulus); this explanation is, however, not in agreement with the generally accepted role of norepinephrine in the acclimation process. Alternatively, the mice do not, or cannot, release norepinephrine to an extent corresponding to that obtainable by injection; this does not, however, agree with the observation that in control mice, norepinephrine could not increase expression as much as the cold stress. One possibility could be that the expression process may be initiated earlier by the injection than by the physiological stimulus.
Requirement for Simultaneous ~- and/~-Stimulation in vivo It was very clear that the adrenergically-induced increase in thermogenin expression occurred only when both the ~- and the/~-components of the adrenergic stimulation were present. There are two possibilities for this synergistic effect of ~- and /?-stimulation: either both have to interact directly on the brown fat cells themselves, or the synergism can be an in vivo effect in some way related to the ability of the stimulus to "reach" the cells. Concerning the possibility that the effect is on the brown fat cells themselves, it is well known that the brown fat cells possess/~-adrenergic receptors linked to cAMP production [for review see Nedergaard and Lindberg (1982)]. There is now also evidence that the cells possess el-adrenergic receptors (Mohell et al., 1983a) linked to IP 3 turnover (Nanberg and Putney, 1986) and Ca 2 + mobilization (Connolly et aI., 1984). However, no synergism between these two receptor types has been shown in vitro, and the effect of e- and /?-adrenergic stimulation on the acute thermogenic response appears to be simply additive (not synergistic), with the a-receptors only being responsible for a minor fraction of the thermogenesis (Mohell et al., 1983b). Thus, in acute in vitro experiments it is not apparently necessary to costimulate the ~pathway in order to obtain the full/?-response and at least 80 % of maximal respiration (Mohell et al., 1983b). This should be contrasted with the situation observed in vivo. Ma and Foster (1984) observed that it was mandatory that both the ~- and the/~-pathways were concurrently stimulated in order to obtain an increase in cAMP efflux from the tissue, an increase in tissue blood flow, and a stimulation of whole animal thermogenesis. Foster (1985) found that this synergism occurred within the tissue itself, but it was not possible for Ma and Foster (1984) to exclude the possibility that the synergistic effect on cAMP efflux was secondary to the elevated blood flow. If these observations do indeed indicate that the ability (at least in vivo) of a selective/?-stimulation to activate
630
Jacobsson, Nedergaard, and Cannon
the adenylate cyclase is dependent upon a simultaneous ~-stimulation, this could be in agreement with an ability of cholera toxin to bypass this convergence point and induce an increase in cAMP and, through this, an increase in thermogenin gene expression (Fig. 2E), without simultaneous c~-adrenergic stimulation. Thus, with regard to the question of the physiological significance of the c~receptors of the brown fat cells, the primary implication of the present results, i.e. that simultaneous ~- and/%stimulation of the cells themselves is necessary to elicit an increased expression of thermogenin mRNA, may be of considerable interest. However, the alternative possibility, i.e. that an increase in intracellular cAMP is a sufficient signal for thermogenin mRNA expression, and that in in vivo experiments the ability of a pure/?-signal to influence the cells and stimulate the cyclase is low, cannot be refuted. The elucidation of this point will require in vitro experiments either with isolated cells or with cell cultures of brown fat (Nbchad et al., 1983), which seem to be able to synthesize thermogenin, although at a much reduced rate (Nedergaard and Cannon, 1984). The nature of such a pathway leading from an increase in cytosolic cAMP level to an increased expression of thermogenin mRNA is presently unknown.
ACKNOWLED GEMENTS
These investigations were supported by a grant from the Swedish Natural Science Research Council to BC and JN. The authors are grateful to Claes Carneheim for advice on the injections. The assistance of Elisabeth Palmer in the receptor binding experiments is gratefully acknowledged.
REFERENCES Bouillaud, F., Ricquier, D., Thibault, J., and Weissenbach, J. (1985). Proc. Natl. Acad. Sci. USA 82:445448. Brooks, S. L., Rothwell, N. J., Stock, M. J., Goodbody, A. E., and Trayhurn, P. (1980). Nature 286:274-276. Cannon, B., and Nedergaard, J. (1985). Essays Biochern. 20:110-164. Carneheim, C., Nedergaard, J., and Cannon, B. (1984). Am. J. Physiol. 246 :E327-E333. Connolly, E., Nanberg, E., and Nedergaard, J. (1984). Eur. J. Biochem. 141:187-193. Derman, E., Krauter, K., Walling, L., Weinberger, C., Ray, M., and Darnell Jr., J. E. (1981). Cell 23:731739. Foster, D. O. (1985). Int. J. Obesity 9, Suppl. 2:25-29. Harris, W. H., Moore, L. A., and Yamashiro, S. (1986). Can. J. Physiol. Pharmaeol. 64:133-137. Himms-Hagen, J. (1986). In Brown Adipose Tissue (P. Trayhurn and D. G. Nicholls, Eds), Edward Arnold Publ., London, pp. 000. Holt, S., and Yoki, D. A. (1982). Biochem. J. 208:819-822. Holt, S. J., and York, D. A. (1984). Horm. Metabol. Res. 16:378-379. Jacobsson, A., Nedergaard, J., and Cannon, B. (1986). Fourth Eur. Bioenergetics Conf. In press. Jacobsson, A., Nedergaard, J., and Cannon, B. (1986). lnternatl. Congress Obesity. In press. Jacobsson, A., Stadler, U., Glotzer, M. A., and Kozak, L. P. (1985). J, Biol. Chem. 260:16250-16254. Jansky, L., Bartunkova, R., Kockova, J., Mejsnar, J., and Zeisberger, E. (1969). Fed. Proe. 28 : 1053-1058. Jones, R. G., Henschen, L., Mohell, N., and Nedergaard, J. (1986). Subm. for publ. Levin, B. E., and Sullivan, A. C. (1986). J. Pharmaeol. Exp. Therap. 236:681-688. Ma, S. W. Y., and Foster, D. O. (1984). Can. J. Physiol. Pharmaeol. 62:943-948. McCormack, J. G., and Denton, R. M. (1977). Biochem. J. 166:627-630. Mohell, N., Nedergaard, J', and Cannon, B. (1983b). Eur. J. Pharmacol. 93:183-193. Mohell, N., Svartengren, J., and Cannon, B. (1983a). Eur. J. Pharrnacol. 92:15-25.
Adrenergic Control of Thermogenin Expression
631
Nanberg, E., and Putney Jr., J. W. (1986). FEBS Lett. 195:319-322. Nechad, M., Kuusela, P., Carneheim, C., BjiSrntorp, P., Nedergaard, J., and Cannon, B. (I983). Exp. Cell Res. 149:105-118. Nedergaard, J., and Cannon, B. (1984). In Thermal Physiology (J. R. S. Hales, Eds), Raven Press, New York, pp. 169-173. Nedergaard, J., and Lindberg, O. (1982). Int. Rev. Cytol. 74:I87-286. Nedergaard, J., Raasmaja, A., and Cannon, B. (1984). Biochem. Biophys. Res. Commun. 122:1328-1336. Ricquier, D., Mory, G., Bouillaud, F., Thibault, J., and Weissenbach, J. (1984). FEBS Lett. 178:240-244. Ridley, R. G., Patel, H. V., Parfett, C. L. J., Olynyk, K. A., Reichling, S., and Freeman, K. B. (1986). Biosci. Rep. 6:87-94. Rothwell, N. J., and Stock, M. J. (1981). Metabolism 30:673~678. Rothwell, N. J., and Stock, M. J. (1986). In Brown Adipose Tissue (P. Trayhurn and D. G. NichoUs, Eds), Edward Arnold Publ., London, pp. 000. Rothwell, N. J., Stock, M. J., and York, D. A. (1984). Comp. Biochem. Physiol. 78A:565-569. Seydoux, J., Trimble, E. R., Bouillard, F., Assimacopoulos-Jeannet, F., Bas, S., Ricquier, D., Giacobino, J. P., and Girardier, L. (1984). FEBS Lett. 166:141-145.
