Uncoupling protein-1 is not leaky

Biochimica et Biophysica Acta 1797 (2010) 773–784

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a b i o

Uncoupling protein-1 is not leaky Irina G. Shabalina, Mario Ost, Natasa Petrovic, Marek Vrbacky 1, Jan Nedergaard, Barbara Cannon ⁎ The Wenner-Gren Institute, Stockholm University, SE-106 91 Stockholm, Sweden

a r t i c l e

i n f o

Article history: Received 17 January 2010 Received in revised form 23 March 2010 Accepted 8 April 2010 Available online 14 April 2010 Keywords: Uncoupling protein 1 Brown adipose tissue mitochondria Cold acclimation Thermogenesis Basal proton leak Medium tonicity

a b s t r a c t The activity of uncoupling protein-1 (UCP1) is rate-limiting for nonshivering thermogenesis and diet-induced thermogenesis. Characteristically, this activity is inhibited by GDP experimentally and presumably mainly by cytosolic ATP within brown-fat cells. The issue as to whether UCP1 has a residual proton conductance even when fully saturated with GDP/ATP (as has recently been suggested) has not only scientific but also applied interest, since a residual proton conductance would make overexpressed UCP1 weight-reducing even without physiological/pharmacological activation. To examine this question, we have here established optimal conditions for studying the bioenergetics of wild-type and UCP1(−/−) brown-fat mitochondria, analysing UCP1mediated differences in parallel preparations of brown-fat mitochondria from both genotypes. Comparing different substrates, we find that pyruvate (or palmitoyl-L-carnitine) shows the largest relative coupling by GDP. Comparing albumin concentrations, we find the range 0.1–0.6% optimal; higher concentrations are inhibitory. Comparing basic medium composition, we find 125 mM sucrose optimal; an ionic medium (50– 100 mM KCl) functions for wild-type but is detrimental for UCP1(−/−) mitochondria. Using optimal conditions, we find no evidence for a residual proton conductance (not a higher post-GDP respiration, a lower membrane potential or an altered proton leak at highest common potential) with either pyruvate or glycerol3-phosphate as substrates, nor by a 3–4-fold alteration of the amount of UCP1. We could demonstrate that certain experimental conditions, due to respiratoty inhibition, could lead to the suggestion that UCP1 possesses a residual proton conductance but find that under optimal conditions our experiments concur with implications from physiological observations that in the presence of inhibitory nucleotides, UCP1 is not leaky. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Isolated mitochondria possess a proton leak, giving rise to socalled basal or state-4 respiration. The nature of this leak has been much discussed [1,2]. Presently it is discussed that (some of) this leak is mediated via specific mitochondrial membrane proteins that – at least under experimental conditions – possess the ability to transport protons. Notably, isoform 1 of the adenine nucleotide transporter (ANT) has been demonstrated to be such a proton transporter, at least when it is idle [3,4]. It has been recently suggested [5] that UCP1 (uncoupling protein 1, the protein that mediates the thermogenic process in brown adi-

Abbreviations: ANT, adenine nucleotide translocase; BSA, bovine serum albumin; EDTA, ethylenediamine tetraacetic acid; FCCP, carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone; GDP, guanosine 5′-diphosphate; UCP1, uncoupling protein 1 ⁎ Corresponding author. The Wenner-Gren Institute, The Arrhenius Laboratories F3, Stockholm University, SE-106 91 Stockholm, Sweden. Tel.: +46 8 164120; fax: +46 8 156756. E-mail address: [email protected] (B. Cannon). 1 Visiting researcher from Department of Bioenergetics, Institute of Physiology, Academy of Sciences of the Czech Republic, Videnska 1083, 142 20 Prague 4, Czech Republic. 0005-2728/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.bbabio.2010.04.007

pose tissue [6,7]) is also able to contribute to the basal conductance of (brown-fat) mitochondria. However, a fundamental tenet in brownfat mitochondrial bioenergetics has been that although UCP1 is innately active in isolated mitochondria, the purine nucleotide GDP (as well as other purine nucleotide di- and triphosphates) can fully inhibit the activity of UCP1 [8]; it is principally this observation that made it possible to identify UCP1. Thus, the generally accepted model has been that in unstimulated brown-fat cells, cytosolic nucleotides (mainly ATP) constantly totally inhibit UCP1 activity, and only when the brown-fat cells are stimulated with norepinephrine is UCP1 activated, probably through a mechanism involving a direct interaction of released fatty acids with UCP1, in a process that shows functional competitivity with GDP/(ATP) [7,9]; the exact role of the fatty acids is still not clarified, e.g. it is not known whether they are just overcoming the inhibition caused by the nucleotides or whether they are involved in the functional mechanism of UCP1 [6]. In contrast to this generally accepted scenario, Parker et al. [5] suggest that even when UCP1 is incubated with saturating concentrations of GDP/(ATP), UCP1 possesses a significant residual proton conductance. The question as to the presence or not of such a residual conductance is an issue of much more than academic interest. Therapeutically, there is presently a renewed interest in the possibility of

774

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

affecting the propensity for obesity in humans (or of ameliorating an obese condition) through methods involving brown adipose tissue and UCP1 [10–16]. Based on a tenet that UCP1 should possess a residual proton conductance, efforts directed at merely increasing the amount of UCP1 could be promoted, following the expectation that this UCP1 in itself would be uncoupling and therefore weightreducing. However, if UCP1 does not possess such a residual proton conductance, a mere increase in UCP1 content would be metabolically without interest, and therapeutic solutions with the combined goal of not only recruiting but also of activating brown adipose tissue/ UCP1 would be those to advocate. In the present study, we examine the suggestion that UCP1 possesses a residual proton conductance. We find that this is not the case, we therefore conclude that bioenergetic studies concur with physiological studies in refuting the presence of a residual conductance associated with UCP1. 2. Materials and methods 2.1. Animals UCP1-ablated mice (progeny of those described in [17]) were backcrossed to C57Bl/6 for 10 generations and after repeated intercrossing (every 10 generations) were maintained as UCP1(−/−) and UCP1(+/+) (wild-type) strains. The mice were fed ad libitum (R70 Standard Diet, Lactamin), had free access to water, and were kept on a 12:12 h light:dark cycle, routinely at normal (24 °C) animal house temperature. For experiments on warm- or cold-acclimated animals, adult male mice were divided into age-matched (7–8-week-old) groups, one per cage, and acclimated at 30 °C, or successively acclimated to cold by first placing them at 18 °C for 4 weeks with the following 4– 6 weeks at 4 °C (the intermediate 18 °C step is required to allow for survival of the UCP1-KO animals at 4 °C [18]). The experiments were approved by the Animal Ethics Committee of the North Stockholm region. 2.2. Brown adipose tissue collection and mitochondrial isolation Mice were anaesthetised for 1 min with a mixture of 79% CO2 and 21% O2 and decapitated. BAT depots were pooled from 3–5 mice and placed in ice-cold medium containing 250 mM sucrose and freed of white fat and muscle, weighed, and used for isolation of brown-fat mitochondria. Preparations from wild-type and UCP1(−/−) mice were generally made and run in parallel. Brown-fat mitochondria were isolated in general as described for wild-type brown-fat mitochondria in [19] with some modifications required for optimisation of conditions for UCP1-ablated mitochondria. The brown adipose tissue was finely minced with scissors and homogenised in a Potter homogeniser with a Teflon pestle. Throughout the isolation process, tissues were kept at 0–2 °C. Mitochondria were isolated by differential centrifugation. Brown adipose tissue homogenates were centrifuged at 8 500 g for 10 min at 2 °C using a Beckman J2-21 M centrifuge. The resulting supernatant, containing floating fat, was discarded. The pellet was resuspended in ice-cold medium containing 250 mM sucrose. The resuspended homogenate was centrifuged at 800 g for 10 min, and the resulting supernatant was centrifuged at 8 500 g for 10 min. The resulting mitochondrial pellet was resuspended in 100 mM KCl, 20 mM K+-Tes (pH 7.2), 1 mM EDTA, 0.6% fatty-acid-free BSA and centrifuged again at 8 500 g for 10 min. In some experiments, as specified in figure legends, BSA was not present in the final centrifugation. The final mitochondrial pellets were resuspended by hand homogenisation in a small glass homogeniser in the same medium. The concentration of mitochondrial protein was measured using fluorescamine [20] with BSA as a standard.

2.3. Oxygen consumption Oxygen consumption rates were monitored with a Clark-type oxygen electrode (Yellow Springs Instrument Co., USA) in a sealed chamber at 37 °C, as described [21]. Brown-fat mitochondria (0.25 mg protein/ml) were incubated in a medium consisting of 125 mM sucrose, 20 mM K+-Tes (pH 7.2), 2 mM MgCl2, 1 mM EDTA, 4 mM KPi, 0.1% fatty-acid-free BSA (routinely). The BSA concentration was varied in some experiments as indicated in the figure legends. In some experiments, a salt-based medium consisting of 100 mM KCl, 20 mM K+-Tes (pH 7.2), 2 mM MgCl2, 1 mM EDTA, 4 mM KPi, 0.1% fatty-acid-free BSA was also tested. Experiments designed to test the effect of an ionic hypotonic medium were performed either in our standard medium for respiration or in a medium similar to the one described in Parker et al. [5]: 50 mM KCl, 4 mM KPi, 5 mM Hepes, 1 mM EGTA, 4% BSA. Respiratory activity of mitochondria was measured in the presence of 5 mM pyruvate plus 3 mM malate or of 5 mM glycerol-3-phosphate or of 5 mM succinate in the presence of 2 µg/ml rotenone. The respiration was initiated by addition of substrate. Basal respiration was routinely measured as the residual respiration following addition of 1 mM GDP (2 or 3 mM GDP had no additional effects to those of 1 mM GDP). In some experiments, specifically designed to reproduce the conditions described in [5], media were supplemented with 1 mg/ml oligomycin and 2.5 µM carboxyatractyloside. These are specified in figure legends. Maximal oxygen consumption rates were obtained by addition of FCCP to a final concentration of 1.4 µM–26 µM, dependent on the concentration of BSA in the medium. 2.4. Measurement of mitochondrial membrane potential Mitochondrial membrane potential measurements were performed with the dye safranin O [22]. The changes in absorbance of safranin O were followed at 37 °C in an Aminco DW-2 dual-wavelength spectrophotometer at 511–533 nm with a 3-nm slit. Signals were recorded every 0.5 s via a PowerLab/ADInstrument. The data were stored and analysed using the Chart v4.1.1 program. In some experiments, the Olis® modernised DW-2 spectrophotometer with Olis GlobalWorks™ software were used. Calibration curves were made for each mitochondrial preparation in K+-free medium and were obtained from traces in which, after the addition of 3 µM valinomycin the extramitochondrial K+, [K+]out, was altered by addition of KCl in a 0.1–20 mM final concentration range. The change in absorbance caused by each addition was plotted against [K+]out. The intramitochondrial K+, [K+]in, was estimated by extrapolation of the line to the zero uptake point, as described in [22]. The absorbance readings were used to calculate the membrane potential (mV) by the Nernst equation (61 mV • log([K+]in/[K+]out)). 2.5. Immunoblotting Immunoblot analysis was performed principally as in [21]. Aliquots of freshly isolated mitochondrial suspensions were stored under nitrogen at − 80 °C after supplementation with protease inhibitor cocktail (Complete Mini, Roche). Protein concentrations of the thawed mitochondrial samples were re-quantified using the Lowry method. Mitochondrial protein was loaded on 15% SDS-polyacrylamide gel (10 µg/lane), run for 3 h in 1× Tris-glycine-SDS running buffer (BioRad). Protein samples were transferred by electroblotting to a PVDF membrane. The membrane was blotted with UCP1-antibodies, prepared in rabbit from the C-terminal decapeptide of mouse UCP1, dilution 1:3000. After incubation with primary antibodies, the membranes were washed in PBS + 0.1% Tween 20 buffer and incubated with secondary antibodies, conjugated with HRP (Cell Signalling, dilution 1:2000). After incubation with secondary antibodies, the membrane was washed in PBST and

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

775

incubated with detection reagents (ECL Plus Kit, Amersham Biosciences) and the chemiluminescence signal was detected with a CCD camera (Fuji). Quantifications were performed with the Image Gauge v. 3.45 (FujiFilm) software. 2.6. Chemicals Fatty-acid-free bovine serum albumin (BSA), fraction V, was from Roche Diagnostics GmbH (Germany). Rotenone, FCCP (carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone), oligomycin, GDP (guanosine 5′-diphosphate) (sodium salt), pyruvic acid (sodium salt), glycerol-3-phosphate, succinic acid (disodium salt) and EDTA (ethylenediamine tetraacetic acid) were all from Sigma-Aldrich Co. (St. Louis, MO, USA). Carboxyatractyloside was from CalbiochemNovobiochem. Fluorescamine (4-phenyl spiro-[furan-2(3H),1-phthalan]-3,3′-dione) was from Fluka Chemie Gmbh. GDP was dissolved in 20 mM Tes (pH 7.2) and the pH of the solution readjusted to 7.2. FCCP was dissolved in 95% ethanol and diluted in 50% ethanol; oligomycin was dissolved in 95% ethanol. Ethanol in a final concentration of 0.1% did not in itself have any effects on the parameters measured. 2.7. Statistics All data are expressed as means ± standard errors. Statistical analysis was performed by Student's t-test using KaleidaGraph version 4.04 software. 3. Results 3.1. No residual conductance of UCP1 Due to the physiological significance and therapeutic implications of the suggestion in [5] that UCP1 possesses residual proton conductance, we examined respiratory control and membrane potential under standard conditions for brown-fat mitochondria isolated from wild-type mice and UCP1(−/−) mice. We initially used glycerol-3-phosphate as substrate, as did in [5]. As exemplified in Fig. 1A (heavy trace), the addition of glycerol-3-phosphate in itself led to a very high rate of oxygen consumption in brown-fat mitochondria from wild-type mice (i.e. in brown-fat mitochondria possessing UCP1). As expected, this high rate of respiration (thermogenesis) was inhibited by GDP. The respiratory inhibition by GDP could be overcome by the addition of FCCP, an artificial uncoupling agent (Fig. 1A). The important issue is the quantitative outcome when this experiment is performed with brown-fat mitochondria from UCP1 (−/−) mice. A typical experiment is also shown in Fig. 1A (thin trace). As seen, in these mitochondria, the addition of glycerol-3phosphate only led to a low rate of oxygen consumption. This rate was not affected by GDP, but FCCP addition resulted in a high level of respiration (Fig. 1A). Qualitatively, this is exactly what would be expected (and what has principally been shown earlier [23,24]). Quantitatively, the important outcome was that the basal (GDPinsensitive) respiration was not different between brown-fat mitochondria from UCP1-possessing, wild-type mitochondria and mitochondria from UCP1(−/−) mice, neither in the example trace shown in Fig. 1A, nor in a compilation of similar experiments (Fig. 2C,D,E). It may also be noted that the levels of FCCP-stimulated glycerol-3phosphate respiration were not different between wild-type and UCP1(−/−) mitochondria (Figs. 1A, 2C,D,E). Although respiration (thermogenesis) is the physiologically important parameter, it may bioenergetically be suggested that the same respiratory rate could be obtained in wild-type and UCP(−/−) mitochondria if the latter had a lower proton leak but a higher membrane potential. We therefore performed parallel membrane potential measurements. As shown in Fig. 1B, the initial (after addition of

Fig. 1. Oxygen consumption and membrane potential of brown-fat mitochondria isolated from wild-type or UCP1(−/−) mice. (A, B) Representative recordings from oxygen consumption (A) or membrane potential (B) measurements of brown-fat mitochondria isolated from wild-type mice (heavy lines) or UCP1-ablated mice (thin lines). Additions were 0.25 mg brown-fat mitochondria (BAT), 5 mM glycerol-3phosphate (g3p), 2 mM GDP (GDP) and 1.0–1.4 µM FCCP (FCCP). (C) Proton leak (measured as oxygen consumption rate) at ≈ 183 mV, the average highest membrane potential common for both types of mitochondria (n = 2).

substrate glycerol-3-phosphate) safranin O absorbance was lower in wild-type mitochondria as compared with UCP1(−/−) mitochondria, as expected and as shown earlier [21,23,24]. After calibration of each mitochondrial preparation, the membrane potential was calculated to be 124 ± 3 mV for wild-type mitochondria and 185 ± 4 mV for UCP1(−/−) mitochondria. Addition of GDP increased absorbance in wild-type mitochondria so that the wild-type signal level reached the level of UCP1(−/−) mitochondria (Fig. 1B). Notably, GDP addition did not change the signal in UCP1(−/−) mitochondria. In the presence of GDP, the calculated membrane potential values became equal (≈183 mV) in the two types of

776

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

Fig. 2. Effect of different substrates on parameters of oxygen consumption of brown-fat mitochondria isolated from wild-type or UCP1(−/−) mice. (A, B) Representative recordings from oxygen consumption measurements of brown-fat mitochondria isolated from wild-type mice (heavy lines) or UCP1-ablated mice (thin lines). Mitochondrial respiration was supported by (A) 5 mM succinate (in the presence of 2 µg/ml rotenone) or (B) 5 mM pyruvate. Additions were as in Fig. 1 except that 0.5 mg brown-fat mitochondria was used. (C, D) Oxygen consumption rates of wild-type (C) or UCP1-ablated (D) brown-fat mitochondria supported by different substrates. The values in (C) and (D) represent the means ± SE of 5–6 independent mitochondrial preparations of each genotype, examined principally as in Figs. 1A and 2A,B. The same mitochondrial preparations were examined with all three substrates, for direct comparisons, with alternating preparations from wild-type or UCP1(−/−) mice. (E) UCP1-dependent oxygen consumption rates were estimated as rates of wild-type mitochondria (C) minus rates of UCP1-ablated mitochondria (D).

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

mitochondria (formally 186 ± 4 mV for wild-type mitochondria and 180 ± 2 mV for UCP1(−/−) mitochondria). Thus, the hypothesis of a low proton leak combined with an increased membrane potential in the UCP1(−/−) mitochondria could be ruled out. Accordingly, Fig. 1C shows the proton leak rate (measured as oxygen consumption rate) at ≈183 mV, which is thus the highest membrane potential common for both types of mitochondria; there is thus no difference between the mitochondria possessing UCP1 and those without. Thus, under these standard experimental conditions, it is evident that UCP1 is not leaky, i.e. there are no indications that it displays residual proton conductance when inhibited by GDP. 3.2. Oxidative capacity in brown-fat mitochondria is dependent on type of substrate It has been discussed [5] that there could be differences in the ability of UCP1 to display a residual proton conductance, depending on the substrate used. Although this earlier article [5] described experiments with glycerol-3-phosphate (with which substrate – as seen in Section 3.1 – we found no evidence indicating residual conductance), we nonetheless found it necessary for a comprehensive study to examine whether the use of other substrates would reveal such a conductance. 3.2.1. Succinate We first tested another complex-II-linked substrate: succinate. As seen in Fig. 2A, succinate addition to wild-type brown-fat mitochondria had only a small effect on respiration. This low rate was not inhibited by GDP (there was even a tendency to an increased respiration after GDP) and the rate was not stimulated by FCCP. Similar low rates of respiration were observed in UCP1-ablated mitochondria. Thus, succinate is a very poor substrate for brown-fat mitochondria, and with succinate as substrate, no thermogenic effect of UCP1 is observable. The reason for the apparently low capacity for succinate oxidation is presumably the very low ability of brown-fat mitochondria to take up succinate: the mitochondrial dicarboxylate carrier is very poorly expressed and its activity is thus rate-limiting for respiration [25,26] (a possibility that the carrier nevertheless is active under conditions of a high ΔpH [27] would not seem to fit with uptake studies in energised mitochondria [25]). It may be noted that this inability of brown-fat mitochondria to oxidise succinate (and other Krebs-cycle substrates) could affect the interpretation of earlier published results on the UCP1-dependence of various functions of brownfat mitochondria. Additionally, it emphasises that addition of succinate to tissue pieces or isolated cells from brown adipose tissue cannot lead to interpretable results concerning the degree of coupling of the mitochondria within the cells. Indeed, it has been demonstrated [28] that succinate respiration of isolated brown-fat cells is due to contaminating mitochondrial fragments in the preparation, since such fragments are capable of oxidising succinate without the limitations of membrane permeability or a mitochondrial membrane potential (similar conclusions apply in this respect to the addition of glycerol-3phosphate to cells or tissue pieces [28]). 3.2.2. Pyruvate We also examined the bioenergetics of brown-fat mitochondria supported by a complex-I-linked substrate, pyruvate (in the presence of malate) (Fig. 2B). Both wild-type and UCP1(−/−) mitochondria respiring on pyruvate behaved qualitatively similarly to mitochondria oxidising glycerol-3-phosphate (Fig. 1A). A high level of UCP1-dependent thermogenesis was observed only in wild-type mitochondria, whereas basal respiration and FCCP-stimulated respiration were not different between mitochondria from wild-type and UCP1(−/−) mice (Fig. 2B,C,D,E), in agreement with our earlier published results [4,9]. Thus, also with pyruvate as substrate, we were unable to verify any residual respiration in GDP-inhibited UCP1-possessing brown-fat mitochondria.

777

3.2.3. Comparisons between substrates When the respiratory responses to the three substrates are compared (Fig. 2C,D,E), several issues may be commented. Firstly, the basal rate of respiration (GDP-rate) is notably higher when complexII-linked substrates (glycerol-3-phosphate, succinate) are used, as compared to when a complex-I-linked substrate is used (pyruvate). Due to the lower number of protons per oxygen released from complex-II substrates, a somewhat higher (≤50%) oxygen consumption rate is expected from complex-II substrates than from complex-I substrates, if the proton flux through the membrane is the same and limiting in both cases. However, the rate is much higher than that (it is about ≈5-fold higher), and other explanations may be forwarded for this phenomenon (e.g. [29]). Secondly, the UCP1-dependent oxygen consumption (black bars in Fig. 2E) in glycerol-3-phosphatesupported mitochondria is ≈50% higher than with the complex-Ilinked substrate pyruvate, in accordance with the proton-transporting capacity of UCP1 being the rate-limiting factor in both cases. Thirdly, it is easy to form the impression that the GDP-inhibitory effect is “larger” in pyruvate-respiring mitochondria than in glycerol-3-phosphaterespiring mitochondria; this is a misunderstanding caused by the high basal respiration with glycerol-3-phosphate. Fourthly, there is no significant effect of UCP1 ablation on any mitochondrial parameter except that of UCP1-dependence, as expected (Fig. 2E). Fifthly, specifically, there was no decrease in the GDP-inhibited rate due to the absence of UCP1 for any of the substrates, in contrast to what was observed under other conditions [5]. Thus, it is essential to identify suitable conditions for bioenergetic examination of brown-fat mitochondria from both wild-type and UCP1(−/−) mitochondria. 3.3. Bioenergetics of brown-fat mitochondria are dependent on BSA concentration In the study [5], a very low effect of GDP on “basal” respiration in wild-type brown-fat mitochondria can be observed, i.e. the very marked decrease in respiration caused by the addition of GDP (Fig. 1A–C) was almost absent in their hands. A robust effect of GDP addition (a more than 2-fold inhibition of initial respiration, even with glycerol-3-phosphate as substrate) has always been considered a main characteristic of brown-fat mitochondria. When examining the experimental conditions used in [5], two features were striking. We observed that the mitochondria were examined in the presence of 4% BSA (bovine serum albumin) and were incubated in an ionic hypotonic medium. While this level of albumin is approximately that found in the blood and a concentration often used for experiments with e.g. brown-fat cells [30–32], it is an unusually high concentration for brown-fat mitochondrial studies. To investigate the effect of this particular condition – and to ensure optimal BSA concentrations for experiments with brown-fat mitochondria from wild-type and UCP1(−/−) mice – we examined the effect of a broad range of BSA concentrations on the bioenergetics parameters of these mitochondria. Traces as those shown on Fig. 2B were performed in the presence of various concentrations of BSA. As pyruvate showed the highest relative coupling efficiency with GDP, we chose this as substrate. Values for the initial pyruvatestimulated, the GDP-inhibited (basal) and the FCCP-stimulated oxygen consumption are presented in Fig. 3A,B. Concerning the wild-type mitochondria, it may first be noted that the total respiratory capacity (i.e. the top curve, FCCP) was bell-shaped, with three components. At low BSA concentrations (0–0.1%), increasing amounts of BSA removed inhibitory components (probably fatty acids), and the total respiratory capacity therefore increased. In the concentration range between 0.1–1.2%, BSA had no additional effects, as compared with the beneficial effect observed at 0.1%. However, at concentrations exceeding 1%, BSA had an inhibitory effect on respiratory capacity, for unknown reasons. Already at 3% BSA, the respiratory capacity was more than halved.

778

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

The respiratory rate after GDP addition (lowest curve, GDP), i.e. the basal respiration, showed another pattern. In the range 0–0.02% BSA, the rate decreased. This was probably also due to the removal of fatty acids. Fatty acids may uncouple these mitochondria in two ways: they may be general uncouplers of the mitochondria (see UCP1(−/−) results Fig. 3B), perhaps through interaction with ANT2 [4,33], and additionally, they may counteract the inhibitory effect of GDP, in a functionally competitive manner, as earlier demonstrated [9]. At BSA concentrations 0.02–2%, no additional effects were observed, whereas above 2% BSA, an inhibition also developed. The initial rate after pyruvate addition, i.e. the rate mainly due to UCP1 activity, was limited by the respiratory capacity (i.e. equal to the FCCP rate) at low and at very high BSA concentrations. In the range 0.1–2% BSA, a tendency to a reduction in the pyruvate rate was observable. Brown-fat mitochondria from UCP1(−/−) mice showed principally qualitatively and quantitatively the same picture (Fig. 3B). We could therefore conclude that optimal conditions for examination of both types of mitochondria were to use BSA concentrations at 0.1% or slightly exceeding that, up to about 0.6% BSA. The use of BSA concentrations above 1% was detrimental for both for UCP1 activity and notably for general mitochondrial function. The above study was performed with pyruvate as substrate. However, glycerol-3-phosphate and 4% BSA were used in [5]. As glycerol3-phosphate could behave differently from pyruvate in these respects,

we performed additional experiments to reproduce these conditions (glycerol-3-phosphate and 4% BSA) [5] and compared this to 0.6% BSA. Similarly to what was observed with pyruvate, wild-type mitochondria incubated in the presence of this high BSA concentration exhibited significant decreases in all three examined respiratory rates (substrate-stimulated, basal (GDP-insensitive) and maximal (FCCP-induced)) (Fig. 3C). It may particularly be noted that the use of this high concentration of BSA decreased the substrate-stimulated rate more than the GDP-inhibited rate, diminishing the ratio between these (i.e. an apparently lower GDP effect). Similarly to what was observed with pyruvate (Fig. 3B), UCP1(−/−) brown-fat mitochondria supported by glycerol-3-phosphate showed a dramatic decrease in FCCP-induced respiration and small decreases in substratestimulated and basal (GDP-insensitive) respiration when incubated in the presence of 4% BSA (Fig. 3D). 3.4. UCP1 ablation changes the tolerance of brown-fat mitochondria to the tonicity of the incubation medium The presence of UCP1 in brown-fat mitochondria has, as was classically observed, very pronounced effects on the osmotic behaviour of the mitochondria [34]. Apparently, during isolation, the mitochondria lose osmotic support through the activity of UCP1, and when incubated under regular osmotic conditions for other types of mitochondria, respiration is severely inhibited. Incubation in hypo-

Fig. 3. Effects of various concentrations of BSA on parameters of brown-fat mitochondrial oxygen consumption. (A, B) Effect of BSA concentration on the bioenergetic parameters of brown-fat mitochondria. Experiments were performed principally as illustrated in Fig. 2B, except that the medium for preparation of mitochondria contained 0.2% BSA and an additional final centrifugation was performed using medium without BSA, as in [9]. The incubation medium was supplemented with the indicated BSA concentrations. The rates of oxygen consumption after the successive addition of 5 mM pyruvate, 1 mM GDP and FCCP are depicted for wild-type in (A) and for UCP1(−/−) in (B). FCCP was added in a concentration range of 0.8–24 µM depending on the BSA concentration in the medium. Results are combined from 3 mitochondrial preparations in each figure. (C, D) Representative recordings from oxygen consumption measurements of wild-type (C) or UCP1(−/−) (D) mitochondria supported by 5 mM glycerol-3-phosphate in incubation media containing 0.6% BSA (heavy line) or 4% (thin line). 1 mM GDP was added. FCCP was added successively to reach the indicated final concentration. Mitochondria were prepared and incubated in the standard sucrose medium (as described in Materials and methods) supplemented by 1 µg/ml oligomycin and 2.5 µM carboxyatractyloside as in [5].

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

osmotic media is one means to obtain full respiratory capacity [34]. This is exemplified in Fig. 4A, where the extreme inhibition that occurs at “regular” osmolarity (250 mM sucrose) is observed. Although 125 mM sucrose is thus optimal for UCP1-possessing brown-fat mitochondria, this may not necessarily be so for UCP1 (−/−) mitochondria (that probably do not lose osmotic support and thus maintain an expanded matrix). However, also UCP1(−/−) mitochondria respired better in 125 mM sucrose than in 250 mM sucrose (Fig. 4A), although the inhibitory effect of 250 mM sucrose was clearly not as marked as it was in wild-type mitochondria. Similar, although less pronounced, effects were seen in liver mitochondria (Fig. 4A), in agreement with [35]. We also tested an incubation medium with normal osmolarity but based on KCl (100 mM), that has been shown to be adequate for the functionality of wild-type brown-fat mitochondria [19,34]. As expected, this medium was very effective for wild-type mitochondria. We found, however, that such a medium was inhibitory for UCP1ablated mitochondria (Fig. 4A). Thus, an iso-osmotic KCl medium worked well for wild-type mitochondria but not for UCP1(−/−) mitochondria. In the study [5], the brown-fat mitochondria were examined in an ionic hypotonic medium, mainly consisting of 50 mM KCl, which, according to the experiments above, could be detrimental for UCP1-

779

ablated mitochondria. We therefore examined whether such conditions with respect to osmolarity affected the outcome. First, we compared the respiratory function of wild-type brown-fat mitochondria in our standard medium with that in the ionic hypotonic medium supplemented with 4% BSA used in [5] (Fig. 4B). A reduction in both the initial and in the FCCP-stimulated rates was observed in mitochondria incubated in the Parker et al. medium (Fig. 4B). Since the presence of 4% BSA induced a similar inhibition (Fig. 3C), almost no additional effect of the ionic hypotonic medium was observed in these mitochondria and it was not possible to discern if the hypotonicity was detrimental (Fig. 4B). It should be noted that the basal respiration observed in these experiments (in the presence of GDP) was ≈100 nmol O2 • min− 1 • mg protein− 1 both under our standard conditions and under those of Parker et al. [5]. The UCP1(−/−) brown-fat mitochondria were affected by the Parker et al. medium to a much more pronounced degree (Fig. 4C). Since the effect of 4% BSA was not as detrimental as ionic hypotonicity in these mitochondria (Fig. 3D) the decreases in substrate-stimulated and basal (GDP-insensitive) respiration were mainly due to the ionic hypotonic medium. Particularly the decrease in the GDPinsensitive basal respiration was significant. In our standard medium, the GDP-insensitive basal rate was ≈100 nmol O2 • min− 1 • mg protein− 1, the same level as that seen for the wild-type mitochondria.

Fig. 4. The effect of medium tonicity on oxygen consumption of brown-fat mitochondria isolated from wild-type or UCP1-ablated mice. (A) Effects of medium composition on oxidative capacity (FCCP response) of brown-fat mitochondria from wild-type and UCP1(−/−) mice respiring on pyruvate. In the graph, 125 mM sucrose indicates the standard medium containing 125 mM sucrose (composition as described in Materials and methods), and 250 mM sucrose indicates the medium containing 250 mM sucrose instead of 125 mM. 100 mM KCl indicates the standard ionic medium (composition as described in Materials and methods). Experiments were performed principally as illustrated in Fig. 2B, except that for preparation of mitochondria, a medium with 0.2% BSA was used and an additional final centrifugation was performed using medium without BSA, as in [9]. 1.0–1.4 µM FCCP was added. Mean FCCP response in the medium containing 125 mM sucrose was set to 100% in each type of mitochondria. *P b 0.05 and **P b 0.01 — statistically significant differences from 125 mM sucrose. (B, C) Representative recordings from oxygen consumption measurements of wild-type (B) or UCP1-ablated (C) mitochondria, either prepared and incubated in the standard sucrose medium (described in Materials and methods) supplemented by 0.6% BSA (heavy line) or in the ionic medium from [5] containing 4% BSA (thin line). Additions were as in Fig. 3C, including 1 µg/ml oligomycin and 2.5 µM carboxyatractyloside.

780

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

However, this rate in the Parker et al. [5] medium was only ≈60 nmol O2 • min− 1 • mg protein− 1 (Fig. 4C). We can therefore conclude that the observation [5] of a lower respiratory rate in the presence of GDP in brown-fat mitochondria from UCP1(−/−) mice than from wild-type mice is secondary to the medium used. The lower rate observed (in our experiments 60 nmol O2 • min− 1 • mg protein− 1 as compared to 100) was interpreted as being directly due to the absence of UCP1 in the mitochondria, whereas in our hands it appears to be entirely due to a general inhibitory effect of the experimental conditions used for the UCP1 (−/−) mitochondria. As this observation (or the corresponding decreased proton leak) was the main argument that the residual proton conductance should be a consequence of the presence of UCP1 [5], it can be concluded that this suggestion could not be verified. Here we find that it was not observable under standard conditions for brown-fat mitochondria, and we were also able to identify the experimental background for the result: the respiratory inhibition caused by low osmolarity. 3.5. Modulation of UCP1 content by acclimation temperature does not alter basal respiratory rate In the above experiments, we examined brown-fat mitochondria from wild-type and UCP1(−/−) mice. However, it could be postulated that the use of brown-fat mitochondria from UCP1(−/−) mice may result in brown-fat mitochondria that are qualitatively different from brown-fat mitochondria that possess UCP1. Therefore, for quantitative elucidation of the influence of UCP1 on the basal bioenergetic features of brown-fat mitochondria, we also physiologically (in contrast to the genetic ablation of the UCP1 gene) modulated the UCP1 content of the mitochondria, by acclimating the mice to warm and cold environments (cf. [36,37]). As expected, the UCP1 content was elevated 3–4-fold in brown-fat mitochondria from cold-acclimated wild-type mice as compared to warm-acclimated mice (Fig. 5A,B). To examine whether the innate UCP1 activity is directly proportional to UCP1 amount, we estimated the level of GDP-inhibitable respiration in brown-fat mitochondria respiring on either pyruvate or glycerol-3-phosphate, in experiments as exemplified in Fig. 5C. We defined the UCP1-mediated respiration as the GDP-inhibitable respiration. As seen both for pyruvate (Fig. 5D) and for glycerol-3-phosphate (Fig. 5E), the ΔGDP respiration (Fig. 5F) was about 2–3-fold higher in brown-fat mitochondria from cold-acclimated than from warm-acclimated mice. Thus, there is reasonable proportionality between UCP1 amount and UCP1 activity (cf. [37]). If the suggestion [5] that UCP1 possesses a residual proton conductance were correct, it would be expected that the GDP-inhibited basal rate would be increased by cold acclimation. This was not the case (Fig. 5D,E). The observations concerning the unaltered GDP-insensitive, basal respiratory rate between brown-fat mitochondria from warm-acclimated and cold-acclimated mice have some further bearings. The fatty acyl composition of phospholipids has been implicated as a determinant of mitochondrial proton leak [38]. The alterations of phospholipid composition caused by cold acclimation are UCP1-independent [39] but they could have functional significance for both basal and adaptive thermogenesis. However, as we do not observe any alteration in basal respiratory rate due to cold acclimation, the phospholipid alterations occurring here do not have any effect on basal proton conductance. Additionally, as UCP1-mediated respiration is reasonably proportional to UCP1 amount (Fig. 5B and F), the present study also establishes that the cold-induced phospholipid alterations do not cause major alterations in the efficiency of UCP1. The morphology of wild-type brown-fat mitochondria is significantly different at 30 °C and at 4 °C. In a warm environment, the mitochondria partially lose their highly developed inner membrane

cristae surface [40]. It could be assumed that this alteration in total mitochondrial surface could alter the proton permeability of the mitochondria, especially as it has been discussed that the inner surface area could be a determinant of total mitochondrial proton leak [1,2,38]. The present studies imply that this parameter does not affect basal proton leak. 3.6. Proton leak was not different in warm-acclimated mice of the two genotypes In a further extension of the experimental conditions above, we have also examined GDP-insensitive, basal respiration in brown-fat mitochondria from wild-type and UCP1(−/−) mice acclimated to 30 °C (thermoneutrality), a condition which, as indicated above, allows low but regulated expression of UCP1 in wild-type mice, but leads to mitochondria with altered morphology compared to those from coldacclimated animals. Although the wild-type mitochondria only had low UCP1 expression (Fig. 5), it was sufficient to allow for significant UCP1-dependent respiration both with pyruvate and with glycerol-3-phosphate as substrates (Figs. 5 and 6). For both substrates, the UCP1(−/−) mitochondria from 30 °C-acclimated mice showed no GDP-sensitive respiration. The rates of GDP-insensitive oxygen consumption were similar in both genotypes (Fig. 6A,B). Thus, again, the presence of UCP1 did not lead to any higher residual respiration than did its absence. Data from parallel membrane potential measurement are shown in Fig. 6C. As the amount of UCP1 present in wild-type mitochondria from 30 °C-acclimated mice was low (Fig. 5A,B), the difference in initial potential (after glycerol-3-phosphate addition) between wildtype (159 ± 5 mV) and UCP1(−/−) (182 ± 4 mV) mitochondria was only about 23 mV (vs 63 mV in mitochondria from mice at 24 °C, Fig. 1B). However, GDP changed the absorbance in wild-type mitochondria to the level observed in UCP1(−/−) mitochondria (Fig. 6C). In the presence of GDP, both mitochondria types exhibited identical membrane potentials (180 ± 5 mV for wild-type mitochondria and 179 ± 4 mV for UCP1(−/−) mitochondria). Thus, even under these conditions, there was no indication of a residual proton leak caused by UCP1. 4. Discussion In the present study, we have examined the tenet [5] that UCP1 should possess a residual proton conductance, even when it is fully inhibited by GDP. We did not find support for this proposal, and have noted that the experimental conditions used in the earlier report inhibited mitochondrial respiration, such that results indicating that UCP1 possesses such a residual activity may be obtained. The study necessitated the reexamination [35] of optimal conditions for examining the bioenergetics of brown-fat mitochondria from both wildtype and from UCP1(−/−) mice, and this enabled explanation of the results obtained earlier [5]. Studies on cells and intact animals also support an absence of residual proton conductance. 4.1. Establishing optimal conditions for examining brown-fat mitochondria 4.1.1. Substrate We have here directly compared the suitability of different substrates for examination of the bioenergetics of brown-fat mitochondria (and for brown-fat mitochondrial studies in general) (Fig. 2). As has been earlier demonstrated, succinate – a preferentially used substrate in many types of experiments on mitochondria from other organs – is totally unsuited for studies of brown-fat mitochondria, due to its low permeability. Glycerol-3-phosphate is oxidised on the outside of the mitochondrial inner membrane and has therefore no

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

781

Fig. 5. UCP1 content and oxygen consumption of brown-fat mitochondria isolated from mice acclimated to 30 °C or 4 °C. (A) Representative immunoblot analysis of UCP1 content. (B) Immunoblots (as that in A) were quantified. The mean amount of UCP1 in wild-type mice was set to 100% in each blot and the levels of UCP1 in other mitochondria expressed relative to this. The values represent the means ± SE of 4 independent mitochondrial preparations for each group. ***P b 0.001 — statistically significant difference between 30 °C and 4 °C. (C) Representative recordings from oxygen consumption measurements of wild-type mitochondria isolated from mice acclimated to 30 °C (thin line) or 4 °C (heavy line). Additions were as in Fig. 2B. (D, E) Oxygen consumption rates of brown-fat mitochondria from wild-type mice, respiring on pyruvate (D) or glycerol-3-phosphate (E). The values represent the means ± SE of 5–6 independent mitochondrial preparations of each group, examined principally as in (C). **P b 0.01 and ***P b 0.001 — statistically significant differences between 30 °C and 4 °C. (F) The GDP effect expressed as the mean difference between the rates before and after GDP addition, i.e. the UCP1-mediated oxygen consumption. **P b 0.01 and ***P b 0.001 — statistically significant differences between 30 °C and 4 °C.

permeation problems, and it is therefore a reasonable complex-IIlinked substrate to use in brown-fat mitochondria (as early pointed out [41]), but it shares with succinate the problem that it can be oxidised by mitochondrial fragments. The high apparent state-4 rate (“basal rate”) also makes the relative coupling effect of GDP small (about a 50% reduction at the most). Pyruvate (in the presence of malate) is an excellent complex-I-coupled substrate. Due to its very

low “basal rate”, relative coupling effects of GDP are high (some 80% inhibition may be seen), it has a very high permeability over the mitochondrial membrane [25], and the oxidative capacity of brownfat mitochondria for pyruvate is very high, so that UCP1 activity is the rate-limiting step. Additionally, it can be used for studies in brown-fat cells, as it readily enters the cells and does not in itself, in contrast to fatty acids, alter the degree of coupling of the cell [30,42,43]. As

782

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

ponents of the respiratory chain. Additionally, in early experiments with brown-fat mitochondria it was demonstrated that, if fatty acids were not removed by oxidation, the presence of albumin in the isolation and incubation medium was essential for obtaining regulated thermogenic activity [46,47]. There are therefore good reasons to include albumin in the incubation medium. Here we show that the albumin range 0.1–0.6% is optimal, but higher amounts quickly become detrimental (Fig. 3). 4.1.3. Medium tonicity We confirmed here that a hypo-osmotic sucrose medium (125 mM), which allows matrix volume expansion [34], is necessary for brown-fat mitochondrial function, and we found that this medium could also be used with brown-fat mitochondria from UCP1 (−/−) mice. However, an isotonic ionic medium (100 mM KCl) – which is also adequate for and allows matrix volume expansion in wild-type brown-fat mitochondria – is detrimental for UCP1(−/−) brown-fat mitochondria, as is the hypotonic ionic medium (50 mM KCl) used in [5]. 4.2. Interpretation of data implying residual proton conductance of UCP1

Fig. 6. No residual proton leak in UCP1-containing mitochondria from mice acclimated to 30 °C. (A, B) Oxygen consumption rates supported by pyruvate + malate (A) or glycerol-3-phosphate (B). The values represent the means ± SE of 5–6 independent mitochondrial preparations of each group, examined principally as in Fig. 5C. **P b 0.01 — statistically significant difference between wild-type and UCP1-ablated mice. C) Representative recordings from membrane potential (B) measurements of brown-fat mitochondria isolated from wild-type mice (heavy lines) or UCP1-ablated mice (thin lines). Additions were as in Fig. 1B.

we have shown earlier, palmitoyl-carnitine (or palmitoyl-CoA plus carnitine) are also very good substrates for brown-fat mitochondria [44,45]. The complex-I-linked substrate glutamate, often used in other mitochondrial studies (in the presence of malate), does not work with brown-fat mitochondria, again for permeability reasons [25]. 4.1.2. Albumin Fatty acids interact with mitochondria in general and specifically with brown-fat mitochondria in complex ways. In general, fatty acids may “uncouple” any mitochondrial preparation and may inhibit com-

Previous studies of brown-fat mitochondria from wild-type warmand cold-acclimated mice and UCP1(−/−) mice have directly or indirectly implied that there is no residual proton conductance in GDPinhibited UCP1 [9,24,37,43,48,49]. A recent study [5] suggested the presence of such a conductance, but we demonstrate here that the results leading to this suggestion were influenced by components of the incubation medium. Thus, the previous conclusion remains. Furthermore, earlier proton conductance measurements performed in wild-type and UCP1(−/−) mitochondria found no essential differences between the two genotypes [48] while similar studies in warm- (very low UCP1) and cold-acclimated (high UCP1) guinea pigs even found a lower proton conductance in the mitochondria from the cold-acclimated animals than in those from the warm-acclimated ones [37]. In neither of these studies was the inhibitor of the adenine nucleotide translocase (ANT), carboxyatractyloside, used. Since ANT is believed to be a component of the basal proton leak [3,4], this could influence the results. However, this component appeared to be somewhat larger in the UCP1(+/+) mitochondria [4], and the absence of carboxyatractyloside therefore cannot affect the conclusion that UCP1 does not possess a residual proton conductance. In the present experiments, carboxyatractyloside was present in the critical experiments comparing the Parker medium with the standard medium (Fig. 4B,C), as was oligomycin. This thus failed to explain the differences. It was correctly observed in [5] that in an earlier publication [9], our data could be interpreted as implying a residual proton conductance in UCP1. These data could only be interpreted in this way for glycerol-3-phosphate, not for pyruvate. These early experiments were performed with non-optimised BSA concentrations, which may contribute to the results obtained at that time. We would therefore ascribe these earlier data to a difference at that time in the mitochondrial preparation and incubation conditions. 4.3. Lack of evidence of residual UCP1 proton conductance in cells and intact animals The data presented here are limited to the outcome of experiments with isolated mitochondria. However, data from biological systems at different levels of organisation support the conclusion reached here. Thus, in isolated brown-fat cells from wild-type and UCP1(−/−) mice, there is no difference between the unstimulated rates of oxygen consumption; only after norepinephrine addition is the presence of UCP1 revealed in the wild-type (by a high respiratory rate being induced) [43]. Similarly, INS-1 cells transfected with UCP1 showed no

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

decrease in mitochondrial membrane potential until stimulated with fatty acids (additionally it was observed that transfected UCP2 could not be activated with fatty acids, underlining the unique uncoupling ability of UCP1) [50]. An – erroneous – impression that UCP1 is innately active in cells (i.e. even in the presence of cytosolic purine nucleotides) may arise from results of studies of mice in which UCP1 is artificially ectopically expressed in different organs (white adipose tissue [51], skeletal muscle [52,53], heart [54]). These mice become lean or in other ways show phenotypes implying that a residual proton conductance is associated with UCP1. However, studies in yeast have shown that although the activity of UCP1 is still controlled at low expression levels, the mitochondria may lose their response to GDP with overexpression [55,56]. This has been interpreted to imply that high amounts of UCP1 destroy the integrity of the mitochondria [56,57]. Ectopic (over)expression may thus be equivalent to addition of an artificial uncoupler (FCCP, DNP) in an organ-specific way. The total amount of UCP1 in a cold-acclimated mouse as compared to a thermoneutrality-acclimated mouse is increased some 50-fold [58]. There is a minor (≈10%) increase in mouse basal metabolism (i.e. metabolism measured at thermoneutral temperatures) in coldacclimated mice — but this is seen both in wild-type and UCP1(−/−) mice and is thus not related to UCP1 activity [18,59]. Thus, the 50-fold higher amount of UCP1 has no effect at all on basal metabolic rate. The absence of residual activity evidently makes physiological sense, as a cold-acclimated mouse coming “in from the cold” would risk dying of hyperthermia if its UCP1 had residual proton conductance. Similarly, the (smaller) increase in UCP1 amount associated with adaptation to different diets (diet-induced thermogenesis) is not reflected in any alteration in the basal metabolic rate of the animals [60]. Thus, biochemical and physiological studies concur in the conclusion that in the presence of saturating amounts of inhibitory nucleotides, UCP1 is not leaky. Acknowledgements These studies were supported by the Swedish Research Council and the European Union Collaborative Project ADAPT (contract 201100). The authors are members of the COST Action “Mitofood” network, which supported the visit of M.V. to Stockholm. M.O. was supported by a Leonardo scholarship from the EU. We thank Solveig Sundberg for establishing and verifying mouse strains. References [1] D.F. Rolfe, M.D. Brand, The physiological significance of mitochondrial proton leak in animal cells and tissues, Biosci. Rep. 17 (1997) 9–16. [2] R.K. Porter, Mitochondrial proton leak: a role for uncoupling proteins 2 and 3? Biochim. Biophys. Acta 1504 (2001) 120–127. [3] M.D. Brand, J.L. Pakay, A. Ocloo, J. Kokoszka, D.C. Wallace, P.S. Brookes, E.J. Cornwall, The basal proton conductance of mitochondria depends on adenine nucleotide translocase content, Biochem. J. 392 (2005) 353–362. [4] I.G. Shabalina, T.V. Kramarova, J. Nedergaard, B. Cannon, Carboxyatractyloside effects on brown-fat mitochondria imply that the adenine nucleotide translocator isoforms ANT1 and ANT2 may be responsible for basal and fatty-acid-induced uncoupling respectively, Biochem. J. 399 (2006) 405–414. [5] N. Parker, P.G. Crichton, A.J. Vidal-Puig, M.D. Brand, Uncoupling protein-1 (UCP1) contributes to the basal proton conductance of brown adipose tissue mitochondria, J. Bioenerg. Biomembr. 41 (2009) 335–342. [6] J. Nedergaard, V. Golozoubova, A. Matthias, A. Asadi, A. Jacobsson, B. Cannon, UCP1: the only protein able to mediate adaptive non-shivering thermogenesis and metabolic inefficiency, Biochim. Biophys. Acta 1504 (2001) 82–106. [7] B. Cannon, J. Nedergaard, Brown adipose tissue: function and physiological significance, Physiol. Rev. 84 (2004) 277–359. [8] D.G. Nicholls, The bioenergetics of brown adipose tissue mitochondria, FEBS Lett. 61 (1976) (1976) 103–110. [9] I.G. Shabalina, A. Jacobsson, B. Cannon, J. Nedergaard, Native UCP1 displays simple competitive kinetics between the regulators purine nucleotides and fatty acids, J. Biol. Chem. 279 (2004) 38236–38248. [10] J. Nedergaard, T. Bengtsson, B. Cannon, Unexpected evidence for active brown adipose tissue in adult humans, Am. J. Physiol. 293 (2007) E444–E452.

783

[11] A.M. Cypess, S. Lehman, G. Williams, I. Tal, D. Rodman, A.B. Goldfine, F.C. Kuo, E.L. Palmer, Y.H. Tseng, A. Doria, G.M. Kolodny, C.R. Kahn, Identification and importance of brown adipose tissue in adult humans, N. Engl. J. Med. 360 (2009) 1509–1517. [12] M. Saito, Y. Okamatsu-Ogura, M. Matsushita, K. Watanabe, T. Yoneshiro, J. NioKobayashi, T. Iwanaga, M. Miyagawa, T. Kameya, K. Nakada, Y. Kawai, M. Tsujisaki, High incidence of metabolically active brown adipose tissue in healthy adult humans: effects of cold exposure and adiposity, Diabetes 58 (2009) 1526–1531. [13] W.D. van Marken Lichtenbelt, J.W. Vanhommerig, N.M. Smulders, J.M. Drossaerts, G.J. Kemerink, N.D. Bouvy, P. Schrauwen, G.J. Teule, Cold-activated brown adipose tissue in healthy men, N. Engl. J. Med. 360 (2009) 1500–1508. [14] K.A. Virtanen, M.E. Lidell, J. Orava, M. Heglind, R. Westergren, T. Niemi, M. Taittonen, J. Laine, N.J. Savisto, S. Enerbäck, P. Nuutila, Functional brown adipose tissue in healthy adults, N. Engl. J. Med. 360 (2009) 1518–1525. [15] M.C. Zingaretti, F. Crosta, A. Vitali, M. Guerrieri, A. Frontini, B. Cannon, J. Nedergaard, S. Cinti, The presence of UCP1 demonstrates that metabolically active adipose tissue in the neck of adult humans truly represents brown adipose tissue, FASEB J. 23 (2009) 3113–3120. [16] J. Nedergaard, B. Cannon, The new metabolic world with human brown adipose tissue: therapeutic implications, Cell Metab. 11 (2010) 268–278. [17] S. Enerbäck, A. Jacobsson, E.M. Simpson, C. Guerra, H. Yamashita, M.-E. Harper, L.P. Kozak, Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese, Nature 387 (1997) 90–94. [18] V. Golozoubova, E. Hohtola, A. Matthias, A. Jacobsson, B. Cannon, J. Nedergaard, Only UCP1 can mediate adaptive nonshivering thermogenesis in the cold, FASEB J. 15 (2001) 2048–2050. [19] B. Cannon, J. Nedergaard, Studies of thermogenesis and mitochondrial function in adipose tissues, Methods Mol. Biol. 456 (2008) 109–121. [20] S. Udenfriend, S. Stein, P. Bohlen, W. Dairman, W. Leimgruber, M. Weigele, Fluorescamine: a reagent for assay of amino acids, peptides, proteins, and primary amines in the picomole range, Science 178 (1972) 871–872. [21] I.G. Shabalina, N. Petrovic, T.V. Kramarova, J. Hoeks, B. Cannon, J. Nedergaard, UCP1 and defense against oxidative stress. 4-Hydroxy-2-nonenal effects on brown fat mitochondria are uncoupling protein 1-independent, J. Biol. Chem. 281 (2006) 13882–13893. [22] J. Nedergaard, The relationship between extramitochondrial Ca2+ concentration, respiratory rate, and membrane potential in mitochondria from brown adipose tissue of the rat, Eur. J. Biochem. 133 (1983) 185–191. [23] A. Matthias, A. Jacobsson, B. Cannon, J. Nedergaard, The bioenergetics of brown fat mitochondria from UCP1-ablated mice. Ucp1 is not involved in fatty acid-induced de-energization (“uncoupling”), J. Biol. Chem. 274 (1999) 28150–28160. [24] W.E. Hofmann, X. Liu, C.M. Bearden, M.E. Harper, L.P. Kozak, Effects of genetic background on thermoregulation and fatty acid-induced uncoupling of mitochondria in UCP1-deficient mice, J. Biol. Chem. 276 (2001) 12460–12465. [25] B. Cannon, V.S. Bernson, J. Nedergaard, Metabolic consequences of limited substrate anion permeability in brown fat mitochondria from a hibernator, the golden hamster, Biochim. Biophys. Acta 766 (1984) 483–491. [26] A. Unami, Y. Shinohara, K. Kajimoto, Y. Baba, Comparison of gene expression profiles between white and brown adipose tissues of rat by microarray analysis, Biochem. Pharmacol. 67 (2004) 555–564. [27] P. Jezek, J. Borecky, Inner membrane anion channel and dicarboxylate carrier in brown adipose tissue mitochondria, Int. J. Biochem. Cell Biol. 28 (1996) 659–666. [28] V. Bernson, P. Lundberg, B. Pettersson, Flavoprotein-linked substrate oxidation in preparations of hamster brown adipocytes. A discrimination between internally and externally oxidized substrates, Biochim. Biophys. Acta 587 (1979) 353–361. [29] D.G. Nicholls, The effective proton conduction of the inner membrane of mitochondria from brown adipose tissue. Dependency on proton electrochemical potential gradient, Eur. J. Biochem. 77 (1977) 349–356. [30] I.G. Shabalina, E.C. Backlund, J. Bar-Tana, B. Cannon, J. Nedergaard, Within brownfat cells, UCP1-mediated fatty acid-induced uncoupling is independent of fatty acid metabolism, Biochim. Biophys. Acta 1777 (2008) 642–650. [31] N. Petrovic, I.G. Shabalina, J.A. Timmons, B. Cannon, J. Nedergaard, Thermogenically competent nonadrenergic recruitment in brown preadipocytes by a PPARgamma agonist, Am. J. Physiol. Endocrinol. Metab. 295 (2008) E287–E296. [32] N. Petrovic, T.B. Walden, I.G. Shabalina, J.A. Timmons, B. Cannon, J. Nedergaard, Chronic Peroxisome Proliferator-activated Receptor gamma (PPAR{gamma}) activation of epididymally derived white adipocyte cultures reveals a population of thermogenically competent, UCP1-containing adipocytes molecularly distinct from classic brown adipocytes, J. Biol. Chem. 285 (2010) 7153–7164. [33] V.P. Skulachev, Anion carriers in fatty acid-mediated physiological uncoupling, J. Bioenerg. Biomembr. 31 (1999) 431–445. [34] D.G. Nicholls, H.J. Grav, O. Lindberg, Mitochondria from hamster brown-adipose tissue. Regulation of respiration in vitro by variations in volume of the matrix compartment, Eur. J. Biochem. 31 (1972) 526–533. [35] I.G. Shabalina, A.V. Panov, Control of the energizing of liver mitochondria at the adenine nucleotide carrier level under hypotonic conditions, Biokhimiia 56 (1991) 439–446. [36] R.M. Locke, E. Rial, D.G. Nicholls, The acute regulation of mitochondrial proton conductance in cells and mitochondria from the brown fat of cold-adapted and warm-adapted guinea pigs, Eur. J. Biochem. 129 (1982) 381–387. [37] E. Rial, D.G. Nicholls, The mitochondrial uncoupling protein from guinea-pig brown adipose tissue. Synchronous increase in structural and functional parameters during cold-adaptation, Biochem. J. 222 (1984) 685–693. [38] R.K. Porter, A.J. Hulbert, M.D. Brand, Allometry of mitochondrial proton leak: influence of membrane surface area and fatty acid composition, Am. J. Physiol. 271 (1996) R1550–R1560.

784

I.G. Shabalina et al. / Biochimica et Biophysica Acta 1797 (2010) 773–784

[39] A. Ocloo, I.G. Shabalina, J. Nedergaard, M.D. Brand, Cold-induced alterations of phospholipid fatty acyl composition in brown adipose tissue mitochondria are independent of uncoupling protein-1, Am. J. Physiol. Regul. Integr. Comp. Physiol. 293 (2007) R1086–R1093. [40] S. Cinti, The Adipose Organ, Editrice Kurtis s.r.l., Milano, 1999. [41] D.G. Nicholls, V.S.M. Bernson, Interrelationships between proton electrochemical gradient, adenine-nucleotide phosphorylation potential and respiration during substrate-level and oxidative phosphorylation by mitochondria from brown adipose tissue of cold-adapted guinea-pigs, Eur. J. Biochem. 75 (1977) 601–612. [42] B. Cannon, J. Nedergaard, The physiological role of pyruvate carboxylation in hamster brown adipose tissue, Eur. J. Biochem. 94 (1979) 419–426. [43] A. Matthias, K.B. Ohlson, J.M. Fredriksson, A. Jacobsson, J. Nedergaard, B. Cannon, Thermogenic responses in brown fat cells are fully UCP1-dependent. UCP2 or UCP3 do not substitute for UCP1 in adrenergically or fatty scid-induced thermogenesis, J. Biol. Chem. 275 (2000) 25073–25081. [44] J.P. Silva, I.G. Shabalina, E. Dufour, N. Petrovic, E.C. Backlund, K. Hultenby, R. Wibom, J. Nedergaard, B. Cannon, N.G. Larsson, SOD2 overexpression: enhanced mitochondrial tolerance but absence of effect on UCP activity, EMBO J. 24 (2005) 4061–4070. [45] R. Westerberg, J.E. Månsson, V. Golozoubova, I.G. Shabalina, E.C. Backlund, P. Tvrdik, K. Retterstøl, M.R. Capecchi, A. Jacobsson, ELOVL3 is an important component for early onset of lipid recruitment in brown adipose tissue, J. Biol. Chem. 281 (2006) 4958–4968. [46] H.J. Hohorst, J. Rafael, Oxydative Phosphorylierung durch Mitochondrien aus braunem Fettgewebe, Hoppe-Seylers Z, Physiol. Chem. 349 (1968) 268–270. [47] H.J. Grav, J.I. Pedersen, E.N. Christiansen, Conditions in vitro which affect respiratory control and capacity for respiration-linked phosphorylation in brown adipose tissue mitochondria, Eur. J. Biochem. 12 (1970) 11–23. [48] S. Monemdjou, L.P. Kozak, M.E. Harper, Mitochondrial proton leak in brown adipose tissue mitochondria of Ucp1-deficient mice is GDP insensitive, Am. J. Physiol. 276 (1999) E1073–E1082. [49] J. Nedergaard, V. Golozoubova, A. Matthias, I. Shabalina, K. Ohba, K. Ohlson, A. Jacobsson, B. Cannon, Life without UCP1: mitochondrial, cellular and organismal characteristics of the UCP1-ablated mice, Biochem. Soc. Trans. 29 (2001) 756–763.

[50] S. Galetti, A. Sarre, H. Perreten, N. Produit-Zengaffinen, P. Muzzin, F. Assimacopoulos-Jeannet, Fatty acids do not activate UCP2 in pancreatic beta cells: comparison with UCP1, Pflugers Arch. 457 (2009) 931–940. [51] J. Kopecky, G. Clarke, S. Enerback, B. Spiegelman, L.P. Kozak, Expression of the mitochondrial uncoupling protein gene from the aP2 promoter prevents genetic obesity, J. Clin. Invest. 96 (1995) 2914–2923. [52] B. Li, L.A. Nolte, J.S. Ju, D. Ho Han, T. Coleman, J.O. Holloszy, C.F. Semenkovich, Skeletal muscle respiratory uncoupling prevents diet-induced obesity and insulin resistance in mice, Nat. Med. 6 (2000) 1115–1120. [53] E. Couplan, C. Gelly, M. Goubern, C. Fleury, B. Quesson, M. Silberberg, E. Thiaudiere, P. Mateo, M. Lonchampt, N. Levens, C. De Montrion, S. Ortmann, S. Klaus, M.D. Gonzalez-Barroso, A.M. Cassard-Doulcier, D. Ricquier, A.X. Bigard, P. Diolez, F. Bouillaud, High level of uncoupling protein 1 expression in muscle of transgenic mice selectively affects muscles at rest and decreases their IIb fiber content, J. Biol. Chem. 277 (2002) 43079–43088. [54] J. Hoerter, M.D. Gonzalez-Barroso, E. Couplan, P. Mateo, C. Gelly, A.M. CassardDoulcier, P. Diolez, F. Bouillaud, Mitochondrial uncoupling protein 1 expressed in the heart of transgenic mice protects against ischemic-reperfusion damage, Circulation 110 (2004) 528–533. [55] E. Rial, M. Gonzalez-Barroso, C. Fleury, S. Iturrizaga, D. Sanchis, J. Jimenez-Jimenez, D. Ricquier, M. Goubern, F. Bouillaud, Retinoids activate proton transport by the uncoupling proteins UCP1 and UCP2, EMBO J. 18 (1999) 5827–5833. [56] J.A. Stuart, J.A. Harper, K.M. Brindle, M.B. Jekabsons, M.D. Brand, A mitochondrial uncoupling artifact can be caused by expression of uncoupling protein 1 in yeast, Biochem. J. 356 (2001) 779–789. [57] J. Nedergaard, D. Ricquier, L.P. Kozak, Uncoupling proteins: current status and therapeutic prospects, EMBO Rep. 6 (2005) 917–921. [58] M. Ashwell, G. Jennings, D. Richard, D.M. Stirling, P. Trayhurn, Effect of acclimation temperature on the concentration of the mitochondrial “uncoupling” protein measured by radioimmunoassay in mouse brown adipose tissue, FEBS Lett. 161 (1983) 108–112. [59] V. Golozoubova, B. Cannon, J. Nedergaard, UCP1 is essential for adaptive adrenergic nonshivering thermogenesis, Am. J. Physiol. 291 (2006) E350–E357. [60] H.M. Feldmann, V. Golozoubova, B. Cannon, J. Nedergaard, UCP1 ablation induces obesity and abolishes diet-induced thermogenesis in mice exempt from thermal stress by living at thermoneutrality, Cell Metab. 9 (2009) 203–209.