Caveolin-1-ablated mice survive in cold by nonshivering thermogenesis despite desensitized adrenergic responsiveness

Mattsson CL, Csikasz RI, Shabalina IG, Nedergaard J, Cannon B. Caveolin-1-ablated mice survive in cold by nonshivering thermogenesis despite desensitized adrenergic responsiveness. Am J Physiol Endocrinol Metab 299: E374–E383, 2010. First published June 8, 2010; doi:10.1152/ajpendo.00071.2010.—Caveolin-1 (Cav1)ablated mice display impaired lipolysis in white adipose tissue. They also seem to have an impairment in brown adipose tissue function, implying that Cav1-ablated mice could encounter problems in surviving longer periods in cold temperatures. To investigate this, Cav1ablated mice and wild-type mice were transferred to cold temperatures for extended periods of time, and parameters related to metabolism and thermogenesis were investigated. Unexpectedly, the Cav1-ablated mice survived in the cold. There were no differences between Cav1ablated and wild-type mice with regard to food intake, in behavior related to shivering, or in body temperature. The Cav1-ablated mice had a halved total fat content independently of acclimation temperature. There was no difference in brown adipose tissue uncoupling protein-1 (UCP1) protein amount, and isolated brown fat mitochondria were thermogenically competent but displayed 30% higher thermogenic capacity. However, the 3-adrenergic receptor amount was reduced by about one-third in the Cav1-ablated mice at all acclimation temperatures. Principally in accordance with this, a higher than standard dose of norepinephrine was needed to obtain full norepinephrine-induced thermogenesis in the Cav1-ablated mice; the higher dose was also needed for the Cav1-ablated mice to be able to utilize fat as a substrate for thermogenesis. In conclusion, the ablation of Cav1 impairs brown adipose tissue function by a desensitization of the adrenergic response; however, the desensitization is not evident in the animal as it is overcome physiologically, and Cav1-ablated mice can therefore survive in prolonged cold by nonshivering thermogenesis.


METHODS
Animals.The study was approved by the Animal Ethics Committee of the North Stockholm region.Cav1-ablated animals [originally generated by the group of Lisanti in 2001 (24)] were purchased from Jackson Laboratories and backcrossed to the C57BL/6 strain for at least seven generations.Before the start of the experiment, all mice were housed at 22°C with a 12:12-h light-dark cycle, with free access to food (R70, Lactamin) and water.At the start of the experiment, 4-mo-old female mice were single caged and either kept at 22°C or transferred to 30°C (thermoneutrality) or to 4°C (with an initial 2 wk at 18°C) for a total of 7 wk.Body weight and body temperature were measured every 7th day for 41 days, and body composition was measured on days 0, 14, 35, and 41.Food intake was measured during week 5; food was weighed when supplied, and remaining food (including spillage, etc.) was subtracted.
The body composition of the mice was measured with a magnetic resonance imaging (MRI) technique (EchoMRI-700/100 Body Composition Analyzer, Echo Medical Systems).The body temperature was measured rectally with a BAT-12 Microprobe Thermometer (AgnTho's).For behavior, Cav1-ablated and wild-type mice were videotaped with a normal video camera (Canon, Digital IXUS 60) during week 5.Each mouse in its home cage was videotaped for 1.5 min at the specific acclimated temperature, and parameters connected by 10.220.33.2 on April 1, 2017 http://ajpendo.physiology.org/Downloaded from to shivering were recorded (e.g., time spent in nest and time spent not moving).
Indirect calorimetry.For indirect calorimetry, the INCA System was used (Somedic, Hörby, Sweden) (1).Before the start of an experiment, the equipment was calibrated with reference gases (18% and 25% O 2 in N2).The airflow was 1 l/min.Metabolic rate was examined during week 6.To obtain resting metabolic rates, the mice were placed during daytime (i.e., during the inactive phase of the mice) in the 4-liter metabolic chambers at 30°C in their home cages.Every 2nd minute for 3 h, O 2 consumption and CO2 production were recorded and the respiratory quotient (RQ) value was calculated (volume CO 2/volume O2).The resting metabolic rate (RMR) was defined as the average of the three lowest 2-min points and the mean metabolic rate (MMR) as the average of the last 60 min of the measurements (i.e., the average oxygen consumption during the last hour)."Agitation" metabolism was defined as the area under the curve (AUC) from 30 min to 60 min minus RMR.
Norepinephrine (NE)-induced thermogenesis was examined during week 6, at least 1 day after the basal/resting metabolic rate.Mice were anesthetized with pentobarbital [35-40 mg/kg ip (mice at 30°C: 35, mice at 22°C: 37.5, and mice at 4°C: 40)], and the mice in their home cages were placed in the chambers at 33°C to obtain basal values for 14 min (the higher ambient temperature is necessary as the thermoregulatory centers become inactive during anesthesia).After this, the mice were removed from the chambers and injected subcutaneously with NE, either with the standard dose (1 mg/kg) or, where indicated, with a higher dose (2.5 mg/kg).The mice were returned to the chambers for ϳ1 h.After the experiments, the mice were placed at 30°C overnight and then returned to the room of their specific acclimation temperature.Anesthetic basal metabolic rate was defined as the mean of the last seven determinations before NE injection and the response to NE as the mean of the three highest points after NE injection minus the anesthetic basal metabolic rate.
Tissue and protein analysis.During week 7, the animals were killed, and interscapular brown adipose tissue and inguinal white adipose tissue were quantitatively dissected out.Samples were directly frozen in liquid nitrogen and stored at Ϫ80°C.The left depot of the brown adipose tissue was homogenized in RIPA buffer [50 mM Tris•HCl, 150 mM NaCl, 1 mM EDTA, and 1% (wt/vol) Triton X-100; volume was 20 times tissue weight] with freshly added protease inhibitor (Complete Mini, Roche).The protein concentration was determined by the Lowry method, and the total protein content in the tissue and the relative protein content of the tissue were calculated.
The quantified values are presented both as the values per milligram of protein loaded and as the total amount of each specific protein in the tissue.Mice acclimated to 22°C were set to 100%.
Bioenergetic characterization.Brown fat mitochondria were prepared from Cav1-ablated and wild-type (C57BL/6) mice acclimated at normal temperature (22°C) and examined as described in References 27 and 28.Briefly, the mitochondria were isolated by differential centrifugation.Oxygen consumption rates were monitored with a Clark-type oxygen electrode (Yellow Springs Instruments) in a sealed chamber at 37°C. Brown fat mitochondria were incubated in a medium consisting of (in mM) 125 sucrose, 20 K ϩ -Tes (pH 7.2), 2 MgCl2, 1 EDTA, and 4 KPi, with 0.1% fatty acid-free BSA.
Statistical analysis.All data are presented as means Ϯ SE.For graphs and statistical analysis, GraphPad Prism was used.For statistical analysis, a two-way ANOVA was used.A Bonferroni posttest was used as indicated in the figures.If the interaction in the two-way ANOVA is not significant, it is not indicated.

RESULTS
To investigate whether Cav1-ablated mice can survive in cold, Cav1-ablated mice were exposed to thermoneutrality, room temperature (22°C), and cold (4°C).Different parameters in connection with metabolism were measured and monitored.At the end of the experiments, parameters related to brown adipose tissue function were analyzed.
Cav1-ablated mice do survive in cold.At the start of the experiment (day 0), mice from normal room temperature (22°C) were either retained at normal room temperature or transferred to thermoneutrality (30°C) or to cold (first 2 wk at 18°C, thereafter 4°C).All the Cav1-ablated mice survived in the cold.We therefore examined the mice further to understand this process.
The body weight of the mice was rather stable; neither the ablation of Cav1 nor acclimation temperature affected the growth of the mice (Fig. 1, A-C).On the indicated days, the body composition was measured by MRI, and fat per lean mass was calculated (Fig. 1, D-F).On day 0 (before transfer), the Cav1ablated mice had a lower fat-to-lean ratio than the wild-type mice, in agreement with Razani et al. (23).The reduced fat/lean mass was maintained throughout the 41 days at all three different acclimation temperatures (Fig. 1, D-F).The fat content itself also showed the same pattern, but the Cav1-ablated mice had slightly more lean mass than the wild-type mice throughout the experiment (Supplemental Fig. S1). 1 In Fig. 1G, the fat-to-lean ratio on day 41 has been compiled and plotted against acclimation temperature.As seen, the higher the acclimation temperature, the higher the fat-to-lean ratio.This was true for both Cav1-ablated and wild-type mice.The lower fat mass in the Cav1-ablated animals was also reflected in a lower amount of dissectable inguinal white adipose tissue (Fig. 1H).
During week 5, the food intake of the mice was measured.It increased with lower acclimation temperature as expected (14,31), but there was no difference between Cav1-ablated and wild-type animals (Fig. 1I), in agreement with Razani et al. (23).
Metabolic rates of Cav1-ablated mice.To examine whether Cav1 ablation had an effect on resting metabolism, Cav1ablated and wild-type animals were placed during their inactive phase in metabolic chambers for 3 h at 30°C, and oxygen consumption and carbon dioxide production were measured.In Fig. 2, A-C, the oxygen consumption rates for mice acclimated to 30, 22, and 4°C, respectively, are shown.From these curves, RMR and MMR per lean body mass were calculated and plotted against acclimation temperature (Fig. 2, D and E).The mice acclimated to 4°C had slightly increased RMR and MMR, in agreement with Golozoubova and colleagues (16,18).Most noticeable, however, was that there were no differences in RMR or MMR between wild-type and Cav1-ablated mice.
The second half-hour of measurement can be taken as a measure of agitation; this was not markedly different between temperatures, but there was an effect of genotype, with the Cav1-ablated mice being somewhat less agitated (Fig. 2F).This behavioral response may be said to be different from the observations of Gioiosa et al. (15), who found that Cav1ablated mice were more active with respect to "crossing frequency," but more similar to those of Trushina et al. (29), who found that the Cav1-ablated mice had reduced activity in an open cage.In contrast, Razani et al. (23) could not see any differences in the activity of the Cav1-ablated mice.
RQ is shown in Fig. 2, G-I.RQ values did not differ between Cav1-ablated and wild-type mice, indicating that they used the same substrate for their resting metabolism (Fig. 2J).
Cav1-ablated mice display diminished thermogenic response to norepinephrine.Cohen et al. (9) observed that Cav1-ablated mice have impaired brown adipose tissue function.As cold acclimation-recruited nonshivering thermogenesis is fully dependent on brown adipose tissue-derived heat (4,16,18), this implies that Cav1-ablated mice would have to utilize other mechanisms to survive longer periods in cold.
Acclimation to cold in wild-type mice is associated with a recruitment of brown adipose tissue and the successive replacement of shivering thermogenesis with nonshivering thermogenesis (4).The recruitment of brown adipose tissue is also reflected in a successive augmentation of the metabolic effect of adrenergic stimulation (NE injection) of the mice; this metabolic response mimics pharmacologically the sympathetic stimulation of brown adipose tissue occurring when the mice are in constant cold; the response must therefore be examined at thermoneutral temperature, where this sympathetic response is absent.We therefore examined nonshivering thermogenesis capacity in Cav1-ablated mice acclimated to different temperatures by injecting NE at the established dose of 1 mg/kg.In Fig. 3,  A-C, oxygen consumption is plotted against time for the three acclimation temperatures.In Fig. 3D, NE-induced oxygen consumption is plotted against acclimation temperature.As expected (4,16), the lower the acclimation temperature, the higher the NE-induced oxygen.Qualitatively, this was true for both Cav1-ablated and wild-type mice.However, the Cav1ablated mice did not demonstrate as high NE-induced thermogenesis as the wild-type animals, principally as implied by Cohen et al. (9).Basically, this is in agreement with the fact that brown adipocytes isolated from Cav1-ablated mice when tested in primary culture also show a diminished adrenergic response (21).It would seem possible that some of the pharmacological "thermogenic" response that is induced in nonbrown fat tissues through this procedure is also diminished in the Cav1-ablated mice (cf.results at 30°C).However, as the increase in response caused by acclimation to cold is entirely due to recruitment of brown adipose tissue (16), and as this parameter is substantially reduced in the Cav1-ablated mice, the outcome of this experiment is that a diminished nonshiv- The RQ after NE injection was calculated (Fig. 3, E-H).Directly after the injection, the wild-type animals changed transiently to the combustion of more lipids, as indicated by the drop in RQ from Ϸ0.80 to Ϸ0.75.This transition to a higher degree of lipid combustion was not seen in the Cav1-ablated animals, in which the RQ after NE injection was significantly higher than in the wild-type mice (Fig. 3H).Thus the Cav1ablated mice were not able to combust as much lipids after NE injection as the wild-type mice, again principally in agreement with a deterioration of the ability of brown adipose tissue to respond to adrenergic stimulation (9).By the end of the recording, there was no difference in RQ between Cav1ablated and wild-type mice in this respect.
Rescue of Cav1-ablated mice in cold is not based on behavior.The apparent outcome of the above experiment implied that Cav1-ablated animals should not be able to induce nonshivering thermogenesis to the same extent as wild-type mice.This raised the question as to how the Cav1-ablated mice were able to survive in the cold.We could see no difference in body temperature between Cav1-ablated and wild-type animals [after 41 days at 4°C, the mean body temperature of the wild-type mice was 37.6 Ϯ 0.1°C and that of the Cav1-ablated mice was 37.5 Ϯ 0.1°C (not statistically significant)].Thus if the brown adipose tissue were not functioning properly, some other mechanism should be present to compensate.The behavior of the mice acclimated to the different temperatures was examined, but no behavior attributable to the absence of nonshivering thermogenesis could be seen, i.e., no shivering, and the mice were not spending more time sitting still or in the nest (data not shown) [the video by Xue et al. (33) shows the conspicuous difference between the behavior in the cold of mice capable of nonshivering thermogenesis or not; no such differences were observable in the Cav1-ablated mice].
Brown adipose tissue in Cav1-ablated mice.Since NEinduced oxygen consumption in the cold is fully dependent on brown adipose tissue function (16), we investigated the brown adipose tissue more closely to localize the site of the decreased capacity.Razani et al. (23) found that brown adipose tissue wet weight is enlarged in Cav1-ablated mice on a high-fat diet.We found that Cav1-ablated mice on chow also had a greater brown adipose tissue wet weight (Fig. 4A).There was no difference in tissue protein concentration (mg protein/mg wet wt) between the Cav1-ablated mice and the wild-type mice (Fig. 4B).In wild-type mice, the total protein amount in the tissue was increased by decreasing acclimation temperatures [as expected (20)], and the same could be seen for the Cav1ablated animals (Fig. 4C).Notably, the Cav1-ablated animals had significantly higher total tissue protein content than the wild-type mice at all temperatures, implying that brown adipose tissue in Cav1-ablated mice is active (it would otherwise be expected to be more lipid filled and thus have a lower protein concentration).
We investigated whether there were abnormalities in the levels of specific proteins.As expected, the Cav1-ablated mice did not have Cav1, while the wild-type mice did (Fig. 4, D and  G).As the level of Cav1 per milligram of protein was not changed (Fig. 4D), wild-type mice in cold had a higher total Cav1 protein amount than mice at higher acclimation temperatures (Fig. 4G).We examined whether any compensatory change in the expression level of the Cav2 or Cav3 genes in brown adipose tissue was observable; this was not the case (not shown).

E378 Cav1-ABLATED MICE SURVIVE IN COLD
There was no difference in the level of UCP1 protein per milligram of total protein between wild-type and Cav1-ablated animals at any acclimation temperature (Fig. 4E).Thus the total UCP1 protein amount increased dramatically with lowered acclimation temperature [as expected for the wild-type mice (20)], leading, if anything, to a slightly higher total amount of UCP1 in the Cav1-ablated animals (Fig. 4H).Therefore, a reduced UCP1 amount could not account for the decreased level of NE-induced thermogenesis observed in the Cav1-ablated mice (Fig. 3).
In rodents, the thermogenic response to NE in brown adipocytes is mediated via ␤ 3 -ARs (34,35).The level of ␤ 3 -AR protein per milligram of protein was not affected by temperature, but the level in Cav1-ablated mice was only half of that in wild-type mice (Fig. 4F) and the total amount of ␤ 3 -AR in the Cav1-ablated animals was therefore significantly lower (by ϳ1/3) than that in the wild-type animals at any temperature (Fig. 4I).Thus the amount of UCP1 protein was not different between Cav1-ablated mice and wild-type mice, but the amount of ␤ 3 -AR protein was decreased in the Cav1-ablated mice.The diminished ability of the Cav1-ablated mice to respond to NE injection therefore either could be related to an inability of the UCP1 in the brown fat mitochondria to respond adequately to stimulation or could be a consequence of the lowered amount of ␤ 3 -AR.
Adequate thermogenic responses of brown fat mitochondria from Cav1-ablated mice.To examine the functionality of brown fat mitochondria from Cav1-ablated mice, we isolated mitochondria from these mice and compared their function with that of wild-type brown fat mitochondria.This examination was particularly essential because in electron microscopy studies, Cohen et al. ( 9) reported a major morphological alteration of the brown fat mitochondria, with clear implications for effects on mitochondrial function.Although the integrity of the outer and inner membrane of the mitochondria was examined by Cohen et al. and found to be unchanged in Cav1-ablated mice, direct studies of thermogenic parameters have not been performed in these mitochondria.As seen in the traces shown in Fig. 5A, the wild-type brown fat mitochondria responded as expected: the addition of substrate (here palmitoyl-CoA ϩ carnitine) resulted in a spontaneous high rate of thermogenesis (oxygen consumption).That this was due to UCP1 activity was confirmed by the fact that it was inhibitable by GDP and that this inhibition could be overcome by an artificial uncoupler (FCCP).As also seen in Fig. 5A, the corresponding traces in the mitochondria from Cav1-ablated mice were qualitatively similar, despite the major morphological alterations observed (9).Thus the UCP1 in the Cav1-ablated mice was functional.Additionally, in the direct comparison shown here, the Cav1-ablated mice showed an enhanced rate of respiration before GDP addition; in a series of four preparations from each strain, this difference was consistently observed (Fig. 5B), while there was no difference in the basal respiratory rate (the rate after GDP addition) or in the total oxidative capacity (the rate after FCCP addition).When similar experiments were performed with pyruvate as substrate, similar results were obtained (Fig. 5C).The inhibition of respiration occurring after GDP addition (the delta GDP effect) may be interpreted as showing the activity of UCP1; this parameter is depicted in Fig. 5D: an increased UCP1 activity was observed in the brown fat mitochondria from the Cav1ablated mice.A probable explanation for an increased UCP1 activity would be an increased amount of UCP1 per mitochon-drion; the compilation in Fig. 5E demonstrates this result, especially as the levels of respiratory chain marker (Cox1) and total mitochondria marker (VDAC) in the preparation were unchanged.In Fig. 5F, the delta GDP respiration is shown as a function of UCP1 amount in each mitochondrial preparation.The activity is directly predicted by the amount of UCP1; thus brown fat mitochondria from Cav1-ablated mice have an increased thermogenic potential due to an increased UCP1 amount.
Accordingly, the diminished response to NE injection observed in the Cav1-ablated mice would not seem to be due to alterations in mitochondrial function.
Adrenergic response is desensitized in Cav1-ablated mice.As noted above, the level of ␤ 3 -AR protein was decreased in the Cav1-ablated mice.This means that a given dose of NE would lead to a lower response, provided that the signaling pathway was not already maximally stimulated.To investigate the possibility that a higher dose of NE could rescue the signal intensity, Cav1-ablated mice acclimated to 4°C were injected with a higher dose of NE (2.5 mg/kg body wt) compared with the standard dose (1 mg/kg).Again the oxygen consumption of anesthetized mice before and after the NE injection was measured and plotted against time (Fig. 6A).The NE-induced oxygen consumption in Cav1-ablated and wild-type mice is plotted in Fig. 6B as a function of NE dose used.As was shown above, with the standard dose the Cav1-ablated mice displayed a lower response than the wild-type mice.With the higher NE dose, there was not a significant alteration in the thermogenic response in the wild-type mice, demonstrating the adequacy of the standard dose.However, with the higher NE dose, the response in the Cav1-ablated mice was enhanced and there was now no difference in NE-induced thermogenesis between wildtype and Cav1-ablated animals.
RQ was also plotted against time (Fig. 6C), and a comparison of the RQ value directly after the NE injection was made between the standard dose and the higher dose (Fig. 6D).As was shown above, there was no effect of NE on the RQ value in the Cav1-ablated mice with the standard dose (Fig. 3, E-G).However, with the higher dose, both the Cav1-ablated and wild-type animals now responded with a markedly decreased RQ ratio (Fig. 6D), demonstrating that both wild-type and Cav1-ablated mice changed toward lipid combustion as an acute effect of NE injection.

DISCUSSION
In the present study, we investigated the cold tolerance of Cav1-ablated mice.We found that Cav1-ablated mice were able to survive in cold through nonshivering thermogenesis, but the mice displayed desensitized adrenergic responsiveness.Thus the genetic absence of Cav1, which resulted in adrenergic desensitization, could apparently be overcome physiologically.
Acute response to cold.The acute response of a warmacclimated mammal (mouse) to cold is initially to shiver, to increase shivering thermogenesis.Not until after Ϸ3 wk of acclimation to cold has nonshivering thermogenesis been fully recruited (4).A mouse at normal animal house environmental temperatures (18 -22°C) may be considered partly cold accli-mated, as this temperature is considerably below its thermoneutral zone and it has constantly increased metabolism by Ϸ50% to counteract heat loss at these temperatures (cf.food intake data in Fig. 1I); it has therefore some capacity for nonshivering thermogenesis, which is adequate to compensate this degree of cold.If such a mouse were exposed to additional cold (4°C), it would acutely use all its available capacity for nonshivering thermogenesis, and in addition it would shiver to obtain the requisite additional heat.
Cohen et al. found that Cav1-ablated and wild-type mice tolerated an acute exposure to cold equally well.However, under the severe conditions of cold plus fasting, Cav1-ablated mice were no longer able to defend body temperature, whereas wild-type mice still managed.Particularly, the Cav1-ablated mice showed no increase in serum nonesterified fatty acids (NEFA) levels ( 9), in agreement with earlier observations of a reduced ability of the white adipose tissue of Cav1-ablated mice to respond to lipolytic stimuli (7).Thus, apparently, during the initial phase in the cold, when shivering is important for heat production, there is a need for energy transfer to muscle.This energy may be supplied in the form of food (and given food, wild-type and Cav1-ablated mice tolerate the cold equally well) or in the form of fatty acids supplied by lipolysis from white adipose tissue.Since lipolysis was defective (and food absent), it would seem that the Cav1-ablated mice could not maintain shivering at a level adequate to maintain body temperature.Thus the major problem for Cav1-ablated mice in acute cold without food probably resides mainly in an inadequate shivering response probably related to inadequate substrate supply.
In contrast, under the conditions examined here, the mice were not deprived of food and could therefore keep their Fig. 6.Desensitized adrenergic responses.Anesthetized mice (n ϭ 6 -7) were injected with NE (2.5 mg/kg sc) as in METHODS.A: oxygen consumption before and after NE injection.B: increase in oxygen consumption by NE: comparison between the standard and higher doses (1 mg/kg vs. 2.5 mg/kg).Standard dose data are the same data as in Fig. 3D at 4°C.Dose ***, genotype ***, interaction **.Bonferroni posttest showed only a difference between the genotypes at 1 mg/kg (***) and not at 2.5 mg/kg (ns).C: RQ before and after NE injection.D: RQ directly after NE injection (34 -44 min): comparison between the standard and higher doses (1 mg/kg vs. 2.5 mg/kg).Standard dose data are the same data as in Fig. 3H at 4°C.Dose ns, genotype *.Bonferroni posttest showed only a difference between the genotypes at 1 mg/kg (*) and not at 2.5 mg/kg (ns).Symbols are defined as in Fig. 1. euthermic body temperature by shivering.As mice can survive cold even in the complete absence of UCP1-mediated nonshivering thermogenesis (18,30), their survival as such does not indicate the availability of adequate nonshivering thermogenesis.However, we found that, given sufficient adrenergic stimulus, Cav1-ablated mice could mount a thermogenic response equal to that of wild-type mice.The mice had therefore successively recruited their brown adipose tissue and become capable of exchanging shivering for nonshivering thermogenesis, even in the absence of Cav1.
Reduced level of ␤ 3 -adrenergic receptors and adrenergic desensitization.We found that the amount of ␤ 3 -ARs was markedly reduced in the brown adipose tissue of Cav1-ablated mice.A decreased amount of ␤ 3 -AR mRNA levels has been seen in detrusor muscle of Cav1-ablated mice (32), and decreased ␤ 3 -agonist sensitivity has been found in different systems (11), including cultured brown adipocytes (21).No effect of Cav1 gene ablation on ␤ 3 -AR level was seen in white adipose tissue (7).
There are marked similarities between what we observe here and what we have earlier observed concerning adrenergic sensitivity in mice ablated of all functional thyroid hormone receptors.In both cases, the mice were able to withstand cold, but the standard NE injection test demonstrated a reduced thermogenic response.However, this response could in both cases be overcome by a higher adrenergic stimulus (17).It cannot be excluded that these observations are related, because the Cav1-ablated mice also demonstrate a reduced level of triiodothyronine (26).
Our data from primary brown adipocytes matured in cell culture show that at a given degree of adrenergic stimulation, less cAMP is produced in cells from Cav1-ablated mice than in wild-type cells (21).In wild-type cells, thermogenesis is fully stimulated with cAMP levels markedly below those that maximally can be induced with adrenergic stimulation (34).Therefore an increased adrenergic stimulation would be sufficient to overcome the desensitization so that a full thermogenic effect can be observed.It is possible that such an increased adrenergic stimulation would also induce a repression of ␤ 3 -AR gene expression; principally this is what is observed in vitro (2,3).A model would be that the augmented adrenergic stimulation in vivo that would be necessary to achieve normal cAMP levels necessarily would be accompanied by an increased ␣ 1 -adrenergic stimulation, and this pathway may be the one responsible for the repression of ␤ 3 -AR gene expression (2) in Cav1ablated mice.
In both the Cav1-ablated mouse and the thyroid-receptorablated mouse, it would therefore seem that a genetically induced decreased sensitivity to adrenergic stimulation is counteracted by the physiological regulatory system that governs the intensity of the sympathetic stimulation to the tissue.If tissue responsiveness is less than adequate, the intensity of stimulation is apparently augmented so that brown adipose tissue heat production (and production capacity) is always adjusted to exactly counteract heat loss.A similar reaction is observable in UCP1 knockout mice, i.e., there are indexes of augmented adrenergic stimulation as a homeostatic response to a lack of adequate thermogenic response to normal stimulation (13).In those mice, this augmented adrenergic stimulation cannot result in a thermogenic response, but in the Cav1ablated mice the desensitization can be overcome.Thus, be-cause of this physiological regulation, Cav1 is neither essential for survival in prolonged cold nor essential for nonshivering thermogenesis.

Fig. 2 .
Fig. 2. Resting metabolism.The basal oxygen consumption of Cav1-ablated and wild-type mice (n ϭ 8 -12) was analyzed with indirect calorimetry at 30°C for 3 h as described in METHODS.A-C: oxygen consumption (cons.) in mice acclimated to 30, 22, and 4°C, respectively.For clarity, only every 4th minute is shown in graphs.D: resting metabolic rate (RMR) calculated per lean body mass as a function of the different acclimation temperatures.RMR is defined as the mean of the 3 lowest points during the 3 h.Acclimation temperature ***, genotype ns.E: mean metabolic rate (MMR) calculated per lean body mass as a function of the different acclimation temperatures.Mean metabolic rate is defined as the mean oxygen consumption during the last 60 min.Acclimation temperature ***, genotype ns.F: "agitation" metabolism calculated per lean body mass as a function of the different acclimation temperatures."Agitation" metabolism is defined as the area under the curve (AUC) from 30 to 60 min minus RMR.Acclimation temperature ns, genotype *.G-I: respiratory quotient [RQ; volume CO2/volume O2 (V ˙CO2/V ˙O2)] for the mice during all 3 h.J: average RQ for the last 60 min.Acclimation temperature ***, genotype ns.Symbols are defined as in Fig. 1.

Fig. 5 .
Fig. 5. Bioenergetics of brown fat mitochondria from wild-type and Cav1-ablated mice.A: example of oxygen electrode traces with mitochondria from wild-type or Cav1-ablated mice.Amount of BAT mitochondria was 0.6 mg protein.The additions were 20 M palmitoyl CoA (Palm CoA) in the presence of 5 mM carnitine and 2 mM malate; 2 mM GDP and 1.4 M FCCP were added at the end.B: compilation of experiments as shown in A, based on 4 preparations from each strain.C: compilation of experiments as in A but performed with 5 mM pyruvate (pyr) in the presence of 2 mM malate, based on 5 preparations from each strain.D: respiration inhibitable by GDP.E: relative amounts of UCP1, Cav1, the mitochondrial marker voltage-dependent anion channel (VDAC), and the respiratory marker cytochrome-c oxidase subunit 1 (Cox1) in the 2 strains.F: UCP1 activity as a function of UCP1 amount in each mitochondrial preparation.Circles indicate data with pyruvate and squares with palmitoyl CoA; filled symbols represent wild-type and open symbols Cav1-ablated mice.The correlation has an r 2 ϭ 0.44 and a P value of 0.005.Significance symbols are defined as in Fig. 1.AU, arbitrary units.