• cell cultures;
  • confocal fluorescence microscopy;
  • dihydroethidium;
  • free radicals;
  • tetramethyl rhodamine ethyl ester;
  • UCP5


  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Outside the nervous system, members of the mitochondrial uncoupling protein (UCP) family have been proposed to contribute to control of body temperature and energy metabolism, and regulation of mitochondrial production of reactive oxygen species (ROS). However, the function of brain mitochondrial carrier protein 1 (BMCP1), which is highly expressed in brain, remains to be determined. To study BMCP1 expression and function in the nervous system, a high-affinity antibody to BMCP1 was generated and used to analyze tissue expression of BMCP1 protein in mouse. BMCP1 protein was highly expressed in heart and kidney, but not liver or lung. In the nervous system, BMCP1 was present in cortex, basal ganglia, substantia nigra, cerebellum, and spinal cord. Both BMCP1 mRNA and protein expression was almost exclusively neuronal. To study the effect of BMCP1 expression on mitochondrial function, neuronal (GT1-1) cell lines with stable overexpression of BMCP1 were generated. Transfected cells had higher State 4 respiration and lower mitochondrial membrane potential (ψm), consistent with greater mitochondrial uncoupling. BMCP1 expression also decreased mitochondrial production of ROS. These data suggest that BMCP1 can modify mitochondrial respiratory efficiency and mitochondrial oxidant production, and raise the possibility that BMCP1 might alter the vulnerability of brain to both acute injury and to neurodegenerative conditions.

Abbreviations used

brain mitochondrial carrier protein 1


bovine serum albumin


carbonylcyanide m-chlorophenylhydrazone






carbenylcyanide p-(trifluoromethoxy) phenylhydrazone


glial fibrillary acidic protein


mitochondrial processing peptidase


normal goat serum


open reading frame


phosphate-buffered saline


reactive oxygen species


tetramethylrhodamine ethyl ester


uncoupling protein


white adipose tissue.

BMCP1 (brain mitochondrial carrier protein 1; also referred to as UCP5), and the other mitochondrial uncoupling proteins (UCPs) are members of the superfamily of mitochondrial carriers which transport small molecules across the inner mitochondrial membrane. The first identified uncoupling protein, UCP1, dissipates the mitochondrial proton gradient by transporting H+ across the inner membrane, thereby uncoupling electron transport from ATP production (Nicholls 1977; for review see also Pecqueur et al. 2001a; Stuart et al. 2001). UCP1 has been established as a key source of fatty acid-dependent heat generation by brown adipose tissue in newborns (Nicholls and Locke 1984; Enerback et al. 1997) and has been linked to the physiological response to cold stress into adulthood (Rothwell and Stock 1979; Enerback et al. 1997). Two additional UCP family members, the ‘novel UCPs’ (UCP2 and UCP3), have significant sequence similarity to UCP1 (57% and 55%, respectively). UCP3 is found predominantly in skeletal muscle and brown adipose tissue (Boss et al. 1997), while UCP2 is expressed in multiple organs, including brain, where it is present primarily in certain hypothalamic nuclei (Fleury et al. 1997). These novel UCPs have been proposed to contribute to fatty acid metabolism and the physiological response to calorie restriction (reviewed by Pecqueur et al. 2001a; also see Stuart et al. 2001). UCP4 and BMCP1 have recently been added to the UCP family because of their sequence similarity to UCP1 (30% for BMCP1, Sanchis et al. 1998; 34% for UCP4, Mao et al. 1999), and because overexpression of these proteins in HEK293 cells was found to decrease mitochondrial membrane potential (Mao et al. 1999; Yu et al. 2000). Both show expression of message in brain, but their function in the nervous system has not been determined.

Recently, a role for UCPs in regulating mitochondrial reactive oxygen species (ROS) production has been advanced. An early study on mitochondria isolated from various tissues, employing GDP as a pharmacological inhibitor of UCP activity, indicated that inhibition of UCPs might increase H2O2 production (Negre-Salvayre et al. 1997). More recently, macrophages from UCP2-deficient mice were shown to generate higher levels of ROS (Arsenijevic et al. 2000). The site of increased ROS production was not determined. However, two additional studies indicate that UCPs specifically modulate mitochondrial ROS production. Overexpression of UCP1 in endothelial cells resulted in a specific reduction in mitochondrial ROS (Nishikawa et al. 2000), and UCP3 knockout mice exhibited enhanced superoxide radical production from mitochondria isolated from skeletal muscle (Vidal-Puig et al. 2000).

Disruption of mitochondrial energy metabolism and altered mitochondrial free radical production have been proposed to contribute to neurodegenerative diseases (Beal 1992; Rapoport et al. 1996; Melov et al. 1999; Wallace 1999). Uncoupling proteins could be a link between these two processes in the nervous system physiologically, and under injury or disease conditions. To investigate the function of BMCP1 in brain, and the role of BMCP1 in modifying neuronal mitochondrial function, we generated a high-affinity antibody to BMCP1, and evaluated BMCP1 protein expression in mouse tissues, brain slices and neural cell cultures. In addition, the effect of BMCP1 expression on mitochondrial function was investigated by generating GT1-1 hypothalamic neurons which overexpress BMCP1. Mitochondrial respiration was assessed by measuring mitochondrial oxygen consumption in BMCP1-transfected cells, and mitochondrial membrane potential (ψm), was evaluated using confocal fluorescence microscopy and the membrane potential-sensitive fluorescent probe, tetramethylrhodamine ethyl ester (TMRE). The effect of BMCP1 expression on mitochondrial superoxide production was determined by confocal fluorescence imaging of dihydroethidium (DHE) oxidation. Our studies indicate that BMCP1 is highly expressed in neurons throughout the brain, and can modify mitochondrial respiratory function, ψm, and free radical production in neurons.

Experimental procedures

  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Cortical cell cultures

Murine pure neuronal, pure astrocyte, and mixed astrocyte–neuronal cultures were prepared and maintained as described previously (Dugan et al. 1995). Cultures were periodically screened for contaminating cells; pure neuronal cultures contained < 0.3% glial fibrillary acid protein (GFAP)-positive cells, and no other cell types. Astrocyte cultures contained < 0.1% oligodendroglia (O1 or Gal-C positive cells), and < 0.5% microglia. All experiments involving animals were approved by the Animal Studies Committee of the Washington University School of Medicine.

Generation of anti-BMCP1 antibody and western blot analysis of tissue expression

Rabbits were injected with a 19-amino acid peptide corresponding to the sequence between transmembrane domain 1 and 2 of mouse BMCP1. Inoculation and harvesting were performed by Alpha Diagnostic International (ADI, San Antonio, TX, USA). Specificity of the BMCP1 antisera was tested by western blot analysis. Tissue and cell culture samples for protein analysis were extracted in modified RIPA lysis buffer [50 mm Na phosphate, pH 7.4, 1% Igepal CA-630, 0.5% sodium dodecyl sulfate (SDS), 150 mm NaCl, 1 mm EGTA, 1 mm dithiothreitol (DTT), 20 µg/mL phenylmethylsulfonyl fluoride, 1 µg/mL leupeptin, 1 µg/mL pepstatin A, and 2 µg/mL aprotinin]. Protein concentrations were measured by a BCA Protein Assay Kit (Pierce, Rockford, IL, USA). Proteins were separated by SDS polyacrylamide gel electrophoresis (SDS–PAGE) through 12% polyacrylamide gels, transferred overnight to nitrocellulose membranes, and incubated with primary antibody overnight at 4°C. BMCP1 antibodies were used at 1 : 2000 dilution (for Ab #3647) or 1 : 250 (for affinity-purified Ab #3644), and cytochrome c antibody (#65981 A, PharMingen, San Diego, CA, USA) was used at 1 : 500 dilution. Immunoreactive BMCP1 and cytochrome c bands were visualized by incubation with the appropriate anti-IgG secondary antibody conjugated to horseradish peroxidase (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and bound secondary antibody was visualized using the SuperSignal ULTRA Chemiluminescent Kit (Pierce) and subsequent exposure to X-ray film. Certain lots were stripped using a western blot recycling kit from Alpha Diagnostics (San Antonia, TX, USA). Anti-β-actin was purchased from Santa Cruz Biotechnology.

Analysis of BMCP1 mRNA by RT-PCR

Total RNA was extracted from fresh tissue using Trizol RNA isolation reagent (Gibco-BRL, Grand Island, NY, USA). RT-PCR was performed using mouse-specific BMCP1 primers (left: 5′-ATTGCCCCTGCGTTACTAAG-3′ and right: 5′-AGCCAAACCA CAGGTGAAAC-3′), which produced a 426-bp PCR product.

Subcellular fractionation

Subcellular fractions for Western blot analysis were prepared as described (Fujimura et al. 1999). Briefly, cells were rinsed with phosphate-buffered saline (PBS), harvested, and homogenized in sucrose buffer containing protease inhibitor cocktail using 40 strokes of an all-glass Dounce homogenizer. After centrifugation at 400 g for 5 min, the supernatant was centrifuged at 13 500 g for 5 min. Pellets, representing a mitochondrially enriched fraction, were collected and the supernatant was further centrifuged at 200 000 g for 20 min. The supernatant from this centrifugation represented the cytosolic fraction. After protein assay by BCA reagent, proteins were used for SDS–PAGE and immunoblotting. Purity of the mitochondria-enriched fraction was confirmed by enrichment of cytochrome oxidase activity.


Cytochrome c antibody (#65971 A) was purchased from PharMingen. NeuN antibody was obtained from Chemicon (Temeculah, CA, USA). Cell cultures were rinsed with PBS, fixed in 4% paraformaldehyde in PBS for 30 min, and permeabilized in 0.25% Triton X-100–PBS for 7 min. Cultures were blocked in 10% normal goat serum (NGS)/PBS for 2 h and incubated in primary antibody for 2 h at room temperature, or overnight at 4°C. Immunoreactivity was visualized by incubating with the appropriate secondary antibody conjugated to Alexa 488 or 568 fluorescent labels on a laser scanning fluorescence confocal microscope (Noran) equipped with Kr-Ar laser and MetaMorph image acquisition software (Universal Imaging Corp.).

Construction of BMCP1-transfected cell lines

The partial sequence of mouse BMCP1 cDNA containing the open reading frame was amplified from mouse brain total RNA. The first cDNA strands were synthesized from 1 µg of total RNA using reverse transcriptase and a random primer. PCR was performed using the primers BMCP1–5F (5′-GTTTCACCTGTGGTTTGGCTG-3′) and BMCP1–3R (5′-GGGAAGAGGCAACACAAATGC-3′), generating a 1160-bp PCR product. PCR products were cloned using pGEM-T vector systems (Promega, Madison, WI, USA) according to the manufacturer's protocol. EcoRI/XbaI fragments were ligated into the pcDNA3 vector at 4°C overnight and sequenced. One day after plating, an SV40 immortalized GnRH-secreting neuronal line, GT1-1 cells (Zhou et al. 1994) were transfected with the plasmid using lipofectamine reagent according to manufacturer's protocol (Gibco-BRL). G418 (Gibco-BRL) was used to select for transfected cells. Cells were maintained in G418. Plasmids without the BMCP1 sequence (pcDNA3) were used to obtain control cells.

Measurement of mitochondrial oxygen consumption in BMCP1 cell lines

Oxygen consumption was measured using an Clark-type oxygen electrode, OxygraphTM (Hansatech, Norfolk, UK) in digitonin-permeabilized GT1-1 cells as described (Murphy et al. 1996). GT1-1 cell lines were gently scraped off the culture plates, and checked for viability by trypan blue staining of a small aliquot. Cells (4 × 106) were then added to the oxymetry chamber in a solution containing 20 mm HEPES, pH 7.1, 250 mm sucrose, 10 mm MgCl2, 2 mm phosphate, 4 µm rotenone and 10 mm succinate. Digitonin-permeabilization was carried out by adding digitonin (0.001% per 106 cells) to the chamber under constant stirring. ADP (0.5 mm) was added to measure State 3 (phosphorylating) respiration. Oligomycin (2.5 µg/mL) was added to inhibit the F0F1-ATPase and determine State 4 (resting) respiration. In initial experiments, this concentration of oligomycin was determined to provide complete inhibition of ADP-dependent respiration (i.e. via the F0F1-ATPase), and to hyperpolarize the mitochondrial resting potential. The maximal uncoupled respiratory rate was obtained by adding 1 µm CCCP (carbonylcyanide m-chlorophenylhydrazone) to the chamber. Oxygen utilization traces and rate determinations were obtained using OxygraphTM software. The effect of linoleic acid (3 µm) or bovine serum albumin (BSA; 3–8 µm) on State 4 respiration was determined by injecting these reagents into the oxymetry chamber after oligomycin. ATP content in each cell lines was measured using a luciferin : luciferase assay (Calbiochem, San Diego, CA, USA) as per instructions.

Imaging of mitochondrial ψ

GT1-1 cells with stable overexpression of BMCP1 or the pcDNA3 vector were grown on coverslip dishes (Mat-Tek) until ∼ 60% confluent. TMRE (100 nm), a potentiometric fluorescent indicator of mitochondrial ψ, was loaded into cells for 20 min. TMRE fluorescence (Ex λ 568 nm, Em λ > 590 nm) was analyzed by confocal fluorescence microscopy. The concentration of TMRE used for these experiments is below the minimal unquenching concentration (∼ 120 nm) in GT1-1 cells (L. L. Dugan, unpublished data), so loss of ψm results in decreasing TMRE fluorescence. After applying a masking function to the images to eliminate non-mitochondrial pixels, TMRE fluorescence was analyzed in individual cells by an observer blind to the transfection status.

Analysis of mitochondrial ROS formation

GT1-1 cells were loaded with 1 µg/mL DHE (Molecular Probes, Eugene, OR, USA) for 1 h, and culture dishes were then placed on the stage of the confocal microscope. Fields of cells were randomly selected under phase contrast optics, and then fluorescence images were obtained using excitation λ = 488 nm and emission λ > 590 nm (590 nm long-pass filter). Frame-averaged confocal images were digitized at 640 × 480 pixels using MetaMorph (Universal Imaging Corp.) image acquisition software, and an observer blind to the identity of the cell line determined average fluorescence pixel intensity for each cell. To calculate the percentage of DHE oxidation that derived from mitochondrial electron transport chain, carbonylcyanide p-(trifluoromethoxy) phenylhydrazone (FCCP) (3 µm) was included at the time of DHE addition to depolarize mitochondria and block mitochondrial production of ROS (Budd et al. 1997).


  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Tissue distribution of BMCP1 expression

To date, BMCP1 protein expression has not been evaluated due to lack of antibodies to this recently described UCP. To allow protein expression and distribution to be assessed, rabbit polyclonal antibodies were raised against two sequences of mouse BMCP1: (1) a 19 amino acid sequence between TM1 and TM2, and (2) a 14 amino acid sequence between TM2 and TM3, regions that lack homology to other proteins, including other UCPs, by BLAST search analysis. Specificity of the BMCP1 antisera were assessed by western blot analysis, and antibodies from rabbits #3647 (TM1-TM2 sequence), and #3644 (TM2-TM3 sequence; affinity-purified) demonstrated high-affinity labeling of a 36-kDa band, the expected size of BMCP1 (Fig. 1). Staining density of the BMCP1 band was decreased by co-incubation with the antigenic BMCP1 peptide (Fig. 1a). Because of the low aqueous solubility of the peptide, the highest concentration that could be tested was 100 µg/mL. By 12% SDS–PAGE, the 32 kDa UCP2 protein band was easily differentiated from the 36-kDa BMCP1 band (Fig. 1b). To further confirm the specificity of the antibody, we compared immunostaining of our ‘standard’ BMCP1 antibody (#3647) with a second antibody raised against an alternative epitope of BMCP1 (#3644); there was perfect concordance of the 36-kDa protein band recognized by each antibody (Fig. 1c). We then performed BLAST protein database analysis for sequences similar to those used to generate our two antibodies, but found only one protein with partial sequence overlap, a putative 30-kDa protein corresponding to a cDNA found in mouse testes. We specifically found no sequence similarity between our antigenic peptides and other members of the mitochondrial carrier protein family. Like UCP2, UCP1 and UCP3 are 32 kDa proteins. However, as these UCPs are not expressed in brain (Ricquier and Bouillaud 2000), we performed Western blot analysis on adipose tissue (UCP1), and skeletal muscle (UCP3), but failed to detect a 32-kDa protein using the anti-BMCP1 antibody in these tissues (Fig. 2b), indicating that the BMCP1 antibody does not cross-react with the smaller UCP1-3 proteins.


Figure 1. Characterization of the anti-BMCP1 antibody. Mouse brain proteins were separated by SDS–PAGE and western blots were performed using a polyclonal BMCP1 antibody raised against the sequence between transmembrane domains 1 and 2. (a) Blot showing specific staining of a band at 36 kDa, the expected size of BMCP1. Protein loading was 20 µg per lane and antibody was used at 1 : 2000. Co-incubation with the BMCP1 antigenic peptide (+ lane) decreased BMCP1 staining intensity compared with the lane without peptide (– lane). (b) Western immunoblot for UCP2 demonstrated the smaller UCP2 band at 32 kDa using a polyclonal UCP2 antibody (UCP2-JSK). (c) Immunostaining of BMCP1 protein by the standard antibody (#3647) was confirmed by stripping and re-staining the blot using a second anti-BMCP1 antibody (#3644), raised against an alternative sequence between TM2 and TM3.

Download figure to PowerPoint


Figure 2. Expression of BMCP1 in mouse tissues. (a) BMCP1 mRNA was determined by RT-PCR using mouse BMCP1-specific primers followed by agarose gel electrophoresis, and protein was analyzed by western immunoblot, in various tissues from mouse. (b) Western blot for BMCP1 in white adipose tissue (WAT), and skeletal muscle (SkM). (c) BMCP1 protein in cortex, basal ganglia (BG), cerebellum and spinal cord (SC). Twenty micrograms of protein were used per lane.

Download figure to PowerPoint

Because UCP4 is expressed in brain and has an expected size similar to BMCP1 (36 kDa; Mao et al. 1999), it is a candidate for recognition by the BMCP1 antibody. However, the peptide sequence used to generate the BMCP1 antibody shares only four out of 19 amino acids with UCP4 in the homologous region, and three of these amino acids are also shared with UCP1, UCP2, and UCP3. Furthermore, we detected a strong 36 kDa protein band in heart and kidney (Fig. 2a), tissues which express BMCP1 message, but not UCP4 (Mao et al. 1999). Taken together, the results suggest that there is no substantial cross-reactivity of the BMCP1 antibody with other members of the uncoupling protein family.

We then compared BMCP1 mRNA and protein expression in several tissues. It has previously been noted that for UCP2, another uncoupling protein found in brain, expression of mRNA and protein often fail to correlate (Pecqueur et al. 2001b). This was recently found to reflect the presence an upstream open reading frame (ORF) in the UCP2 message that strongly inhibits UCP2 translational efficiency (Pecqueur et al. 2001b). Because this upstream ORF is also present in BMCP1, we felt it was important to compare BMCP1 mRNA and protein levels in various mouse tissues (Fig. 2). BMCP1 mRNA and protein were expressed in brain, heart and kidney (Fig. 2a), roughly mirroring results of previous northern blot analysis in mouse (Sanchis et al. 1998). We found that brain exhibited the highest level of BMCP1 protein expression (Fig. 2a). However, the level of BMCP1 protein expression in heart, assessed by densitometry, was substantially higher than in kidney, even though kidney demonstrated higher levels of message by both by RT-PCR (Fig. 2a) and northern blot (Sanchis et al. 1998). Skeletal muscle had modest levels of BMCP1 protein, but only trace expression was seen in white adipose tissue (WAT) (Fig. 2b). We failed to detect BMCP1 mRNA or protein expression in liver and lung (Fig. 2a).

To determine whether there were differences in protein expression among brain regions, BMCP1 protein content was determined in specific regions, and in spinal cord. Previous in situ hybridization analysis demonstrated BMCP1 message in cortex, hippocampus, and thalamus (Sanchis et al. 1998). We also found that protein expression was abundant in cortex (Fig. 2c), thalamus (data not shown) and hippocampus (Fig. 3), as well as cerebellum, basal ganglia and spinal cord (Fig. 2c). To identify which cell types in the nervous system express BMCP1, immunocytochemistry was carried out on brain slices using antibodies to BMCP1 and to the neuron-specific marker, NeuN (Fig. 3). BMCP1 immunoreactivity co-localized with NeuN in all brain regions examined. BMCP1 immunoreactivity was typically seen surrounding the NeuN positive nucleus, as expected for a protein with a mitochondrial location (Fig. 3i). Only a few cells per slice appeared to be immunoreactive for BMCP1, but not NeuN (Fig. 3i). There were also occasional NeuN-positive cells that were not clearly associated with BMCP1 immunostaining (Fig. 3i). We also performed RT-PCR and western blot analyses on neuronal and astrocyte cultures (Fig. 4). BMCP1 message was abundant in neurons, but was undetectable in astrocytes (Fig. 4a). Similarly, BMCP1 protein was at least 20-fold higher, by densitometry, in neurons than astrocytes on western blot, and was up-regulated with age in culture (Fig. 4b). Immunostaining of neurons and astrocytes for BMCP1 also demonstrated strong BMCP1 immunoreactivity in neurons but minimal staining in astrocytes (Fig. 4c).


Figure 3.  Imaging of BMCP1 expression in brain. Brain slices were immunostained for BMCP1 (green), or the intranuclear neuron-specific marker, NeuN (red). Co-localization of the BMCP1 and NeuN signals appears as yellow. (a) Overview of the pattern of BMCP1 immunoreactivity in mouse brain. The inset shows a high-power view of the corpus collosum (CC). (b–e) CA1 region of hippocampus. (e) Reduction of staining after incubation of the antibody with the BMCP1 peptide. (f–i) Cortex. (i) High power image showing co-localization of NeuN and BMCP1 in cortex. The arrowheads point to NeuN-positive cells which appear to lack BMCP1 immunoreactivity. The arrow at far right designates one of the rare cells showing staining for BMCP1 but not NeuN. (j–l) Substantia nigra. The bar in (b) = 100 µm for images (a–h); the bar in (i) = 50 µm. Bar in (l) = 50 µm, for (j–l).

Download figure to PowerPoint


Figure 4. Analysis of BMCP1 expression in astrocytes and neurons. (a) BMCP1 message was assessed by RT-PCR in cultured cortical astrocytes and pure cortical neuronal cultures (PNC) at 14 days in culture, and in adult mouse brain. (b) BMCP1 protein was also measured in astrocytes and neurons at various days in culture. Adult mouse cortex was included as a control. Equal protein loading (20 µg for all except day 21 astrocytes (25 µg) and adult cortex (5 µg)) was confirmed by stripping and reprobing the blots for β-actin. (c) Immunocytochemistry and confocal fluorescence imaging of BMCP1 immunoreactivity was performed on neurons and astrocytes in ‘mixed’ cortical cultures (neurons on a layer of astrocytes). Images of both cell types are from the same culture. Bar = 20 µm.

Download figure to PowerPoint

Mitochondrial localization and lack of processing of BMCP1

BMCP1 is presumed to be a mitochondrial protein based on the presence of three mitochondrial targeting sequences that it shares with UCP1, UCP2, and UCP3. However BMCP1, unlike the smaller UCP isoforms, also possesses a hydrophobic 22 amino acid N-terminal sequence which might modify targeting of the protein to mitochondria; in fact, a BMCP1 construct with an N-terminal FLAG-tag failed to localize to mitochondria (Yu et al. 2000), indicating that the BMCP1 N-terminus is critical for mitochondria targeting, or that BMCP1 is not a mitochondrial protein. There is precedent for the latter possibility, as other members of mitochondrial carrier protein superfamily concentrate in peroxisomes rather than mitochondria (Weber et al. 1997). To determine the intracellular location of BMCP1 protein, we performed western blot analyses on subcellular fractions prepared from neuronal cultures (Fig. 5a). BMCP1 was found in mitochondria but not cytosol in pure neuronal cultures, as well as neurons in mixed astrocyte–neuronal co-cultures. Only the 36-kDa BMCP1 protein band was present in mitochondria, indicating that BMCP1 does not undergo cleavage by the mitochondrial processing peptidase (MPP), which removes N-terminal targeting sequences from proteins after import into mitochondria (Neupert 1994). Immunostaining of cortical neurons with both BMCP1 and cytochrome c also confirmed that BMCP1 co-localized with mitochondrial cytochrome c (Fig. 5b).


Figure 5. Subcellular localization of BMCP1. To define the intracellular location of BMCP1 protein, western blots (a) were run on subcellular fractions prepared from pure neuronal cultures and neurons in mixed cultures. The total cell homogenate [H], the mitochondrial (enriched) fraction [M], and the cytosolic fraction [C] were compared. Protein loading was 30 µg per lane for mixed cultures, and 20 µg for neuronal cultures. BMCP1 protein was present in the total homogenate and in the mitochondria-enriched fraction, but not in the cytosolic fraction. (b) Fluorescence imaging of cortical neuron immunostained for BMCP1 (red), and cytochrome c (green). Overlay of the fluorescence images shows colocalization of the two signals (yellow). Bar = 10 µm.

Download figure to PowerPoint

Effects of BMCP1 overexpression on mitochondrial respiration and ψm

To study the effect of BMCP1 on mitochondrial function in neurons, GT1-1 cells with stable overexpression of BMCP1 were generated. Non-transfected parental GT1-1 cells, or GT1-1 cells transfected with the pcDNA vector, served as controls. GT1-1 cells, which are SV40 immortalized hypothalamic neurons, retain key neuronal properties such as expression of synaptic vesicle proteins and neuronal structural proteins, production of gonadotropin-releasing hormone, and the presence of functional γ-aminobutyric acid type A receptor complexes and spontaneous action potentials (Zhou et al. 1994). We found that there was basal expression of BMCP1 protein in the parental GT1-1 line (data not shown) and in GT1-1 cells transfected with vector alone (line 3B; Fig. 6a). However, after screening, several transfected lines were found to have three- to fourfold greater expression of BMCP1 by western blot analysis and densitometry (Fig. 6a). Two lines generated from independent clones, 10B and 10C, were used for further studies.


Figure 6. Mitochondrial function in GT1-1 cells with stable overexpression of BMCP1. BMCP1-transfected lines were evaluated for levels of BMCP1 protein expression. Two lines, 10B and 10C, from independent clones, showed threefold higher expression of BMCP1 than the 3B vector control (a). Mitochondrial membrane potential (ψm) was compared in the BMCP1 and control cell lines using TMRE and fluorescence confocal microscopy (b). Values are average TMRE mitochondrial fluorescence pixel intensity ± SEM, n > 40 cells per condition. *p < 0.05 by anova and Student–Newman–Keuls. State 4 mitochondrial respiration was compared in digitonin-permeabilized BMCP1 cells, parental GT1-1 cells, and the vector controls (c,d). State 4 respiration (c) was induced by addition of oligomycin (2.5 µg/mL), a concentration which fully inhibited ADP-dependent O2 consumption (not shown). Values are mean O2 consumption ± SEM, *p < 0.05 by anova and Student–Newman–Keuls; n for each group is listed above each bar. (d) The ability of BSA to improve coupling of mitochondrial respiration was determined by adding BSA to cells in state 4 respiration (after oligomycin). The percentage change in state 4 respiration after BSA is shown for vector control (3B) and BMCP1 (10B, 10C) cells. Values are mean ± SEM. *p < 0.05 by anova and Student–Newman–Keuls, n = 6 (ctrl), n= 7 (BMCP1). (e) Representative oxymetry trace from a second set of experiments showing effects of linoleic acid (LA) and BSA on mitochondrial respiration (O2 consumption as nmol/106 cells) in digitonin-permeabilized GT1 cell lines. Arrows indicate time of addition ADP (0.5 mm), oligomycin, linoleic acid (LA, 3 µm), bovine serum albumin (BSA, 5 µm), or the uncoupler, CCCP (1 µm).

Download figure to PowerPoint

We examined mitochondrial membrane potential in these cell lines using the potential-sensitive probe, TMRE (Kowaltowski et al. 2000; Ward et al. 2000), and confocal microscopy. BMCP1 cells were found to have lower basal ψm(Fig. 6b). Although the magnitude of reduction was relatively small (12%), it is similar to the degree of depolarization (15%) reported to almost completely abolish mitochondrial H2O2 production (Skulachev 1998). Although such small changes in ψm are generally not believed to affect ATP production significantly (Skulachev 1998), we assayed ATP in GT1-1 cells. Values were (in pmol ATP/106 cells; n = 4 for each group) 238 ± 3 (parental cell line), 245 ± 5 (3B vector control), 250 ± 5 (BMCP1 line 10B), 263 ± 10 (BMCP1 line 10C). Values were not significantly different by anova.

Mitochondrial respiratory function was then analyzed in digitonin-permeabilized GT1-1 cells. Permeabilization of the plasma membrane by digitonin capitalizes on the specific interaction of digitonin with cholesterol to form membrane pores. This technique has previously been used to measure the effects of bcl-2 on mitochondrial respiration and calcium uptake in GT1-7 cells (Murphy et al. 1996), and had several advantages over the use of isolated mitochondria for our studies. Digitonin-permeabilization allows mitochondria to retain their association with specific intracellular compartments and cytoskeletal elements. We felt this feature was especially important for studies on neurons, which possess highly distinct populations of mitochondria that differ in composition, physiology and association with cellular compartments (Jung et al. 2000). With appropriate titration of the digitonin concentration, this method has been reported to provide mitochondria with better respiratory coupling than in isolated mitochondrial preparations (Becker et al. 1980; Moreadith and Fiskum 1984; Murphy et al. 1996). Digitonin-permeabilized GT1-1 cells had respiratory control ratios (RCR = state 3/state 4 respiration) with succinate of > 7.0, indicating good preservation of mitochondrial respiratory function after digitonin treatment. Our initial studies using conventional isolation techniques indicated that it would not be feasible to prepare sufficient mitochondria from the GT1-1 cell lines for our studies, whereas the yield of functional mitochondria using digitonin permeabilization was much higher.

Activation of the prototypic UCP1 by fatty acids has been shown to increase state 4 respiration (Nicholls and Locke 1984). In order to determine whether fatty acids would also activate uncoupling by BMCP1, we studied the effect of a fatty acid, linoleic acid, or BSA (to remove lipid activators of BMCP1) on state 4 respiration in the GT1-1 cell lines. We found that, even in the absence of added fatty acids, BMCP1 transfected cells demonstrated higher state 4 respiration than either vector controls or the parental GT1-1 cell line (Fig. 6c). State 4 respiration was reversed by addition of BSA, and this decrease in state 4 was significantly greater in the BMCP1 cell lines than in either vector controls or the parental GT1-1 cells (Fig. 6d). In addition, BMCP1-mediated uncoupling of mitochondrial respiration appeared to be increased by addition of linoleic acid (3 µm). When exposure to linoleic acid was terminated by addition of BSA, state 4 respiration was rapidly reversed to levels that were similar in the BMCP1-transfected cells and vector controls. These data suggest that BMCP1 may be activated by low concentrations of fatty acids, and possibly other endogenous lipids, and that these effects are reversed upon removal of the activating lipid(s).

Reduction of superoxide production in neurons with stable overexpression of BMCP1

One function proposed for the uncoupling protein homologs is regulation of mitochondrial free radical production. Using DHE oxidation as a selective, but not specific, marker for superoxide generation (Bindokas et al. 1996; Budd et al. 1997; Benov et al. 1998), we found that superoxide production in BMCP1-transfected cells was 25% lower than controls (Fig. 7). This decrease actually represented the majority of the superoxide production that derived from the mitochondrial electron transport chain, as mitochondrial depolarization by FCCP did not further decrease DHE oxidation in BMCP1 cells.


Figure 7.  Effects of BMCP1 overexpression on mitochondrial reactive oxygen species production. Cells were loaded with DHE (1 µg/mL), and oxidation of DHE to ethidium was determined 1 h later in control (3B) or BMCP1-transfected cells (10B, 10C) by fluorescence confocal microscopy. (a) Confocal images of DHE oxidation in control and BMCP1 cells in which fluorescence pixel intensity is represented by a linear pseudocolor scale. Bar = 20 µm (b). Quantitative analysis of DHE oxidation in untreated cells (light gray bars), or cells treated with FCCP to depolarize mitochondria and inhibit mitochondrial free radical production (dark gray bars). *p < 0.05 by anova and Student–Newman–Keuls post hoc test compared with untreated control (3B) cells, n > 80 cells per group.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Mitochondrial energy impairment and altered free radical production are believed to be important mechanisms in neurodegenerative disorders (Beal 1992; Melov et al. 1999; Wallace 1999). The mitochondrial uncoupling proteins are of special interest in this context because they appear to regulate both mitochondrial metabolic efficiency (Brand et al. 1994; Rolfe et al. 1999; Ricquier and Bouillard 2000) and free radical generation. Although this association has been shown most convincingly for UCP1, growing evidence suggests that UCP2 and UCP3 are capable of carrying out these functions as well (reviewed by Ricquier and Bouillaud 2000; Rial and Gonzalez-Barroso 2001; Stuart et al. 2001). However, although UCP2 message is found in brain, protein expression appears to be localized primarily to the hypothalamus and is present at extremely low levels (Pecqueur et al. 2001b). BMCP1, in contrast, has been shown to have a wide distribution of message in brain (Sanchis et al. 1998), and might therefore play a role throughout the nervous system. However, little is yet known about its function in the CNS.

Previous studies on BMCP1 have described expression of message, but because of lack of antibodies to BMCP1, no characterization of BMCP1 protein has been performed to date. To allow BMCP1 protein expression to be assessed, we developed and characterized an anti-mouse BMCP1 polyclonal for these studies which appeared to lack cross-reactivity with the smaller UCP1, UCP2 and UCP3 proteins, or with another more distantly related UCP homolog, UCP4. BMCP1 protein expression was abundant in brain, heart and kidney, was moderate in skeletal muscle, and was minimal or absent in adipose tissue, lung and liver. For liver, our results are consistent with the lack of BMCP1 expression in liver found on northern blot analysis of B6D2 mice (Sanchis et al. 1998), but contrast with results from a second group (Yu et al. 2000), who reported substantial BMCP1 message in liver from both C57B6 and FVB-N mice. This second group did not observe lung expression of BMCP1 in their mice, whereas Sanchis et al. (1998) documented expression in lung from both B6D2 mice and rat. These results suggest that strain differences strongly influence the pattern of tissue expression of BMCP1, and indicate that the effect of species and strain will need to be considered in evaluating the regulation of BMCP1 expression and function.

A dissociation between message and protein has been reported for UCP2, which can exhibit substantial mRNA expression without demonstrable protein synthesis (Pecqueur et al. 2001b). The lack of correlation between message and protein for UCP2 has been shown to reflect the presence of an alternative upstream ORF in the UCP2 message, also present in BMCP1, which dramatically reduces translational efficiency of UCP2 (Pecqueur et al. 2001b). However, we found a reasonable correlation between message and protein expression in the tissues examined, suggesting that steady-state levels of BMCP1 message provide a fairly reliable index of BMCP1 protein expression. However, we also found evidence that translational efficiency of BMCP1 may differ between tissues, e.g. heart versus kidney. Although our data indicate that the disparity between mRNA expression and protein synthesis may not be as profound for BMCP1 as for UCP2, it leaves open the question of whether post-transcriptional regulation of BMCP1 protein expression will be more important under conditions of stress or injury. The impact of this alternative ORF in the BMCP1 message will be an important area of future investigation.

To assess the cellular localization of BMCP1 protein expression, brain slices were evaluated for BMCP1 immunoreactivity. Strong BMCP1 immunostaining was observed in cortex, hippocampus, substantia nigra, and cerebellum, with less seen in spinal cord. Dual labeling of slices with antibodies to BMCP1 and the neuronal nuclear marker, NeuN, demonstrated that immunoreactivity of BMCP1 co-localized almost completely with cells expressing NeuN in all regions examined. In cortex and hippocampus, we observed only rare cells which expressed BMCP1 but not NeuN, suggesting that BMCP1 is primarily neuronal. This was further supported by western blot analysis of proteins from neuronal and astrocyte cultures, which showed protein expression in neurons but not astrocytes. Message for BMCP1 was also found in neurons, but not astrocytes. In brain slices, occasional NeuN-positive cells which appeared to lack BMCP1 expression were observed. It is possible that these represent neurons that do not express detectable levels of BMCP1. Further investigation of BMCP1 expression in different neuronal populations may reveal links between synaptic activity, and metabolic regulation in brain.

We then evaluated the subcellular location of BMCP1 protein. Despite evidence that the original uncoupling protein, UCP1, is localized to the inner mitochondrial membrane, a mitochondrial location for other UCP homologs is less well established. A recent study in yeast found that UCP3 was primarily extramitochondrial (Winkler et al. 2001), although these authors proposed that this might be an artifact of overexpression of UCPs in yeast. However, other members of the mitochondrial carrier protein family localize to peroxisomes instead of mitochondria (Weber et al. 1997). Using both subcellular fractionation and immunocytochemistry to determine whether BMCP1 co-localized with cytochrome c, we confirmed that in cultured neurons, BMCP1 protein is localized to mitochondria. Our results also indicated that BMCP1 does not undergo cleavage during mitochondrial import, despite the presence of an N-terminal sequence that appears to be required for mitochondrial targeting (Yu et al. 2000).

To allow analysis of the functional effects of BMCP1 expression on neuronal mitochondria, a neuronal line (GT1-1) was used to generate cells with stable overexpression of BMCP1. Polarigraphic studies of O2 consumption were performed on digitonin-permeabilized GT1-1 cells. We employed this approach instead of using isolated mitochondrial preparations because digitonin-permeabilization results in mitochondria with better respiratory coupling (Becker et al. 1980; Moreadith and Fiskum 1984), and because of the difficulty in isolating sufficient mitochondria from cell cultures. We found that cells overexpressing BMCP1 demonstrated a greater degree of uncoupling (state 4 respiration) than either the vector control cell line, or the parental GT1-1 line. As it is known that lipids, including non-esterified fatty acids, can activate mitochondrial uncoupling by other UCPs (Nicholls 1977; Nicholls and Locke 1984; reviewed by Rial and Gonzalez-Barroso 2001), we evaluated the effect on state 4 respiration of removing lipids. GT1-1 cells expressing BMCP1 showed enhanced uncoupling (state 4 respiration). Fatty acid-free BSA was added to remove endogenous lipid activators of BMCP1, and decreased state 4 respiration to a greater extent in BMCP1 cells than in the vector control cells. Conversely, BMCP1-mediated uncoupling of mitochondrial respiration was increased by low concentrations of the unsaturated fatty acid, linoleic acid (3 µm). Although this concentration of linoleic acid is somewhat higher than that required to activate UCP1 in isolated mitochondria, which is 100 nm (Gonzalez-Barroso et al. 1998), in digitonin-permeabilized cells, fatty acids are likely to be incorporated into other cellular membranes, including the plasma membrane, thereby decreasing the effective concentration at the mitochondrial membrane. Uncoupling proteins 1–3 are also regulated (inhibited) by purine nucleotides ATP, GTP, ADP and GDP through a high-affinity nucleotide binding site (reviewed by Klingenberg and Echtay 2001). ATP is believed to be the principal ligand. We did not explore the effect of purine nucleotides on BMCP1 in the current studies, however, because of the difficulty in removing endogenous ATP from the nucleotide binding site, which would result in ‘masking’ of the binding site and underestimation of the inhibitory effect of added nucleotides (Klingenberg and Echtay 2001). In addition to fatty acids and purine nucleotides, a number of other physiological regulators of UCP1, UCP2 and UCP3 have been identified (Echtay et al. 2000, 2001). However, factors which regulate BMCP1 remain to be determined. Future studies on BMCP1 will focus on the physiological and pharmacological regulation of BMCP1 activity.

Mitochondria in GT1-1 cells transfected with BMCP1 had a slight decrease in mitochondrial membrane potential (ψm), but no drop in ATP levels. Recently, the idea has been advanced that extremely polarized mitochondria generate substantially more ROS than slightly depolarized mitochondria (Skulachev 1998), i.e. under conditions where the mitochondrial are slightly depolarized, ATP production would still be supported, but superoxide and/or H2O2 production would be decreased. In keeping with this idea, neurons transfected with BMCP1 had lower levels of superoxide radical production than controls. The modest overexpression of BMCP1 we achieved in transfected GT1-1 cells was sufficient to abolish the majority of superoxide production which derived from mitochondria. This is consistent with a previous study on UCP1-transfected endothelial cells which also found that mitochondria superoxide production could be suppressed by relatively low levels of UCP expression (Nishikawa et al. 2000)

In summary, our data suggest that BMCP1 is neuronal protein that is widely expressed throughout the CNS, and is localized to neuronal mitochondria. We found that BMCP1 can lower mitochondrial membrane potential without affecting cellular ATP levels, and can enhance uncoupling of mitochondrial respiration in a manner that appeared to be lipid-dependent. Finally, expression of BMCP1 decreased mitochondrial superoxide radical production in neurons. Taken together, these data suggest that BMCP1 might play a role in modifying neuronal mitochondrial function, although whether BMCP1 contributes to damage or rescue of CNS tissue after injury remains an open question, and will be the focus of future studies.


  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This work was supported by a Paul Beeson Physician Scholars Award from the American Federation for Aging Research (LLD) and a grant from the National Institutes of Health (NS41796). We thank Dr Mark Goldberg and Suzanne Underhill for assistance with immunostaining of oligodendroglia.


  1. Top of page
  2. Abstract
  3. Experimental procedures
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Arsenijevic D., Onuma H., Pecqueur C., Raimbault S., Manning B. S., Miroux B., Couplan E., Alves-Guerra M. C., Goubern M., Surwit R., Bouillaud F., Richard D., Collins S., Ricquier D. (2000) Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygen species production. Nat. Genet. 26, 435439.
  • Beal M. F. (1992) Does impairment of energy metabolism result in excitotoxic neuronal death in neurodegenerative illness? Ann. Neurol. 31, 119130.
  • Becker G. L., Fiskum G., Lehninger A. L. (1980) Regulation of free Ca2+ by liver mitochondria and endoplasmic reticulum. J. Biol. Chem. 255, 90099012.
  • Benov L., Sztejnberg L., Fridovich I. (1998) Critical evaluation of the use of hydroethidine as a measure of superoxide anion radical. Free Rad. Biol. Med. 25, 826831.
  • Bindokas V. P., Jordan J., Lee C. C., Miller R. J. (1996) Superoxide production in rat hippocampal neurons: selective imaging with hydroethidine. J. Neurosci. 16, 13241336.
  • Boss O., Samec S., Paoloni-Giacobino A., Rossier C., Dulloo A., Seydoux J., Muzzin P., Giacobino J. P. (1997) Uncoupling protein-3. a new member of the mitochondrial carrier family with tissue-specific expression. FEBS Lett. 408, 3942.DOI: 10.1016/s0014-5793(97)00384-0
  • Brand M. D., Chien L. F., Ainscow E. K., Rolfe D. F., Porter R. K. (1994) The causes and functions of mitochondrial proton leak. Biochim. Biophys. Acta 1187, 132139.
  • Budd S. L., Castilho R. F., Nicholls D. G. (1997) Mitochondrial membrane potential and hydroethidine-monitored superoxide generation in cultured cerebellar granule cells. FEBS Lett. 415, 2124.
  • Dugan L. L., Bruno V. M. G., Amagasu S. M., Giffard R. G. (1995) Glia modulate the response of murine cortical neurons to excitotoxicity: glia exacerbate AMPA neurotoxicity. J. Neurosci. 15, 45454555.
  • Echtay K. S., Winkler E., Klingenberg M. (2000) Coenzyme Q is an obligatory cofactor for uncoupling protein function. Nature 408, 609613.
  • Echtay K. S., Winkler E., Frischmuth K., Klingenberg M. (2001) Uncoupling proteins 2 and 3 are highly active H (+) transporters and highly nucleotide sensitive when activated by coenzyme Q (ubiquinone). Proc. Natl Acad. Sci. USA 98, 14161421.
  • Enerback S., Jacobsson A., Simpson E. M., Guerra C., Yamashita H., Harper M. E., Kozak L. P. (1997) Mice lacking mitochondrial uncoupling protein are cold-sensitive but not obese. Nature 387, 9094.
  • Fleury C., Neverova M., Collins S., Raimbault S., Champigny O., Levi-Meyrueis C., Bouillaud F., Seldin M. F., Surwit R. S., Ricquier D., Warden C. H. (1997) Uncoupling protein-2. a novel gene linked to obesity and hyperinsulinemia. Nat. Genet. 15, 269272.
  • Fujimura M., Morita-Fujimura Y., Kawase M., Copin J. C., Calagui B., Epstein C. J., Chan P. H. (1999) Manganese superoxide dismutase mediates the early release of mitochondrial cytochrome C andsubsequent DNA fragmentation after permanent focal cerebral ischemia in mice. J. Neurosci. 19, 34143422.
  • Gonzalez-Barroso M. M., Fleury C., Bouillaud F., Nicholls D. G., Rial E. (1998) The uncoupling protein UCP1 does not increase the proton conductance of the inner mitochondrial membrane by functioning as a fatty acid anion transporter. J. Biol. Chem. 273, 1552815532.DOI: 10.1074/jbc.273.25.15528
  • Jung C., Higgins C. M., Xu Z. (2000) Measuring the quantity and activity of mitochondrial electron transport chain complexes in tissues of central nervous system using blue native polyacrylamide gel electrophoresis. Anal. Biochem. 286, 214223.
  • Klingenberg M. & Echtay K. S. (2001) Uncouplng proteins. the issues from a biochemist point of view. Biochem. Biophys. Acta 1504, 128143.
  • Kowaltowski A. J., Smail S. S., Russell J. T., Fiskum G. (2000) Elevation of resting mitochondrial membrane potential of neural cells by cyclosporin A, BAPTA-AM, and Bcl-2. Am. J. Physiol. 279, C852C859.
  • Mao W., Yu X. X., Zhong A., Li W., Brush J., Sherwood S. W., Adams S. H., Pan G. (1999) UCP4, a novel brain-specific mitochondrial protein that reduces membrane potential in mammalian cells. FEBS Lett. 443, 326330.DOI: 10.1016/s0014-5793(98)01713-x
  • Melov S., Coskun P. E., Wallace D. C. (1999) Mouse models of mitochondrial disease, oxidative stress, and senescence. Mutat. Res. 434, 233242.
  • Moreadith R. W. & Fiskum G. (1984) Isolation of mitochondria from ascites tumor cells permeabilized with digitonin. Anal. Biochem. 137, 360367.
  • Murphy A. N., Bredesen D. E., Cortopassi G., Wang E., Fiskum G. (1996) Bcl-2 potentiates the maximal calcium uptake capacity of neural cell mitochondria. Proc. Natl Acad. Sci. USA 93, 98939898.
  • Negre-Salvayre A., Hirtz C., Carrera G., Cazenave R., Troly M., Salvayre R., Penicaud L., Casteilla L. (1997) A role for uncoupling protein-2 as a regulator of mitochondrial hydrogen peroxide generation. FASEB J. 11, 809815.
  • Neupert W. (1994) Heinrich Wieland prize lecture. Transport of proteins across mitochondrial membranes. Clin. Invest. 72, 251261.
  • Nicholls D. G. (1977) Cellular mechanisms in brown fat thermogenesis mitochondria. Experientia 33, 11301131.
  • Nicholls D. G. & Locke R. M. (1984) Thermogenic mechanisms in brown fat. Physiol. Rev. 64, 164.
  • Nishikawa T., Du Edelstein D. X. L., Yamagishi S., Matsumura T., Kaneda Y., Yorek M. A., Beebe D., Oates P. J., Hammes H. P., Giardino I., Brownlee M. (2000) Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 404, 787790.
  • Pecqueur C., Alves-Guerra M. C., Gelly C., Levi-Meyrueis C., Couplan E., Collins S., Ricquier D., Bouillaud F., Miroux B. (2001a) Uncoupling protein 2, in vivo distribution, induction upon oxidative stress, and evidence for translational regulation. J. Biol. Chem. 276, 87058712.
  • Pecqueur C., Couplan E., Bouillaud F., Ricquier D. (2001b) Genetic and physiological analysis of the role of uncoupling proteins in human energy homeostasis. J. Mol Med. 79, 4856.
  • Rapoport S. I., Hatanpaa K., Brady D. R., Chandrasekaran K. (1996) Brain energy metabolism, cognitive function and down-regulated oxidative phosphorylation in Alzheimer's disease. Neurodegeneration 5, 473476.DOI: 10.1006/neur.1996.0065
  • Rial E. & Gonzalez-Barroso M. M. (2001) Physiological regulation of the transport activity in the uncoupling proteins UCP1 and UCP2. Biochim. Biophys. Acta 1504, 7081.
  • Ricquier D. & Bouillaud F. (2000) The uncoupling protein homologues: UCP1, UCP2, UCP3, StUCP and AtUCP. Biochem. J. 345, 161179.
  • Rolfe D. F., Newman J. M., Buckingham J. A., Clark M. G., Brand M. D. (1999)Contribution of mitochondrial proton leak to respiration rate in working skeletal muscle and liver and to SMR.Am. J. Physiol.276, C692C699.
  • Rothwell N. J. & Stock M. J. (1979)A role for brown adipose tissue in diet-induced thermogenesis.Nature281, 3135.
  • Sanchis D., Fleury C., Chomiki N., Goubern M., Huang Q., Neverova M., Gregoire F., Easlick J., Raimbault S., Levi-Meyrueis C., Miroux B., Collins S., Seldin M., Richard D., Warden C., Bouillaud F., Ricquier D. (1998) BMCP1, a novel mitochondrial carrier with high expression in the central nervous system of humans and rodents, and respiration uncoupling activity in recombinant yeast. J. Biol. Chem. 273, 3461134615.DOI: 10.1074/jbc.273.51.34611
  • Skulachev V. P. (1998) Uncoupling. new approaches to an old problem of bioenergetics. Biochim. Biophys. Acta. 1363, 100124.
  • Stuart J. A., Cadenas S., Jekabsons M. B., Roussel D., Brand M. D. (2001) Mitochondrial proton leak and the uncoupling protein 1 homologues. Biochim. Biophys. Acta 1504, 144158.
  • Vidal-Puig A. J., Grujic D., Zhang C. Y., Hagen T., Boss O., Ido Y., Szczepanik A., Wade J., Mootha V., Cortright R., Muoio D. M., Lowell B. B. (2000) Energy metabolism in uncoupling protein 3 gene knockout mice. J. Biol. Chem. 275, 1625816266.
  • Wallace D. C. (1999) Mitochondrial diseases in man and mouse. Science 283, 14821488.
  • Ward M. W., Rego A. C., Frenguelli B. G., Nicholls D. G. (2000)Mitochondrial membrane potential and glutamate excitotoxicity in cultured cerebellar granule cells.J. Neurosci.20, 72087219.
  • Weber F. E., Minestrini G., Dyer J. H., Werder M., Boffelli D., Compassi S., Wehrli E., Thomas R. M., Schulthess G., Hauser H. (1997) Molecular cloning of a peroxisomal Ca2+-dependent member of the mitochondrial carrier superfamily. Proc. Natl Acad. Sci. USA 94, 85098514.
  • Winkler E., Heidkaemper D., Klingenberg M., Liu Q., Caskey T. (2001) UCP3 expressed in yeast is primarily localized in extramitochondrial particles. Biochem. Biophys. Res. Commun. 282, 334340.DOI: 10.1006/bbrc.2001.4563
  • Yu X. X., Mao W., Zhong A., Schow P., Brush J., Sherwood S. W., Adams S. H., Pan G. (2000) Characterization of novel UCP5/BMCP1 isoforms and differential regulation of UCP4 and UCP5 expression through dietary or temperature manipulation. FASEB J. 14, 16111618.
  • Zhou J., Holtzman D. M., Weiner R. I., Mobley W. C. (1994) Expression of TrkA confers neuron-like responsiveness to nerve growth factor on an immortalized hypothalamic cell line. Proc. Natl Acad. Sci. USA 91, 38243828.