Voltar
Artigo Acesso aberto Revisado por pares
BIBTEX RIS

Cellubrevin Is a Resident Protein of Insulin-sensitive GLUT4 Glucose Transporter Vesicles in 3T3-L1 Adipocytes

1995; Elsevier BV; Volume: 270; Issue: 14 Linguagem: Inglês

10.1074/jbc.270.14.8233

ISSN

1083-351X

Autores

Allen Volchuk, Robert Sargeant, Satoru Sumitani, Zhi Liu, Lijing He, Amira Klip,

Tópico(s)

Parkinson's Disease Mechanisms and Treatments

Resumo

Insulin stimulates glucose transport in muscle and fat cells by inducing translocation of GLUT4 glucose transporters from a storage site to the cell surface. The mechanism of this translocation and the identity of the storage site are unknown, but it has been hypothesized that transporters recycle between an insulin-sensitive pool, endosomes, and the cell surface. Upon cell homogenization and fractionation, the storage site migrates with light microsomes (LDM) separate from the plasma membrane fraction (PM). Cellubrevin is a recently identified endosomal protein that may be involved in the reexocytosis of recycling endosomes. Here we describe that cellubrevin is expressed in 3T3-L1 adipocytes and is more abundant in the LDM than in the PM. Cellubrevin was markedly induced during differentiation of 3T3-L1 fibroblasts into adipocytes, in parallel with GLUT4, and the development of insulin regulated traffic. In response to insulin, the cellubrevin content decreased in the LDM and increased in the PM, suggesting translocation akin to that of the GLUT4 glucose transporter. Vesicle-associated membrane protein 2 (VAMP-2)/synaptobrevin-II, a protein associated with regulated exocytosis in secretory cells, also redistributed in response to insulin. Both cellubrevin and VAMP-2 were susceptible to cleavage by tetanus toxin. Immunopurified GLUT4-containing vesicles contained cellubrevin and VAMP-2, and immunopurified cellubrevin-containing vesicles contained GLUT4 protein, but undiscernible amounts of VAMP-2. These observations suggest that cellubrevin and VAMP-2 are constituents of the insulin-regulated pathway of membrane traffic. These results are the first demonstration that cellubrevin is present in a regulated intracellular compartment. We hypothesize that cellubrevin and VAMP-2 may be present in different subsets of GLUT4-containing vesicles. Insulin stimulates glucose transport in muscle and fat cells by inducing translocation of GLUT4 glucose transporters from a storage site to the cell surface. The mechanism of this translocation and the identity of the storage site are unknown, but it has been hypothesized that transporters recycle between an insulin-sensitive pool, endosomes, and the cell surface. Upon cell homogenization and fractionation, the storage site migrates with light microsomes (LDM) separate from the plasma membrane fraction (PM). Cellubrevin is a recently identified endosomal protein that may be involved in the reexocytosis of recycling endosomes. Here we describe that cellubrevin is expressed in 3T3-L1 adipocytes and is more abundant in the LDM than in the PM. Cellubrevin was markedly induced during differentiation of 3T3-L1 fibroblasts into adipocytes, in parallel with GLUT4, and the development of insulin regulated traffic. In response to insulin, the cellubrevin content decreased in the LDM and increased in the PM, suggesting translocation akin to that of the GLUT4 glucose transporter. Vesicle-associated membrane protein 2 (VAMP-2)/synaptobrevin-II, a protein associated with regulated exocytosis in secretory cells, also redistributed in response to insulin. Both cellubrevin and VAMP-2 were susceptible to cleavage by tetanus toxin. Immunopurified GLUT4-containing vesicles contained cellubrevin and VAMP-2, and immunopurified cellubrevin-containing vesicles contained GLUT4 protein, but undiscernible amounts of VAMP-2. These observations suggest that cellubrevin and VAMP-2 are constituents of the insulin-regulated pathway of membrane traffic. These results are the first demonstration that cellubrevin is present in a regulated intracellular compartment. We hypothesize that cellubrevin and VAMP-2 may be present in different subsets of GLUT4-containing vesicles. INTRODUCTIONInsulin stimulates glucose uptake into muscle and fat cells by recruiting glucose transporters (predominantly the GLUT4 isoform) from an intracellular storage site to the cell surface(1Cushman S. Wardzala L. J. Biol. Chem. 1980; 255: 4758-4762Abstract Full Text PDF PubMed Google Scholar, 2James D.E. Strube M. Mueckler M. Nature. 1989; 338: 83-87Crossref PubMed Scopus (669) Google Scholar). In spite of the wide documentation of this phenomenon through biochemical and morphological techniques(3Zorzano A. Wilkinson W. Kotliar N. Thoidis G. Wadzinkski B.E. Ruoho A.E. Pilch P.F. J. Biol. Chem. 1989; 264: 12358-12363Abstract Full Text PDF PubMed Google Scholar, 4Slot J.W. Geuze H.J. Gigengack S. Lienhard G.E. James D.E. J. Cell Biol. 1991; 113: 123-135Crossref PubMed Scopus (709) Google Scholar), the identity of the intracellular organelle endowed with glucose transporters and the mechanism of its incorporation into the plasma membrane remain largely unknown.Several scenarios have been proposed to explain the intracellular traffic of glucose transporters. It has been proposed that regulation of GLUT4 intracellular traffic may share characteristics of the process of regulated secretion (a phenomenon involving fusion of specialized exocytic vesicles with the plasma membrane that occurs only in response to a stimulus). Support for the regulated exocytotic pathway is provided by the targeting of transfected GLUT4 glucose transporters to secretory granules in neuroendocrine PC12 cells(5Hudson A.W. Fingar D.C. Seidner G.A. Griffiths G. Burke B. Birnbaum M.I. J. Cell Biol. 1993; 122: 579-588Crossref PubMed Scopus (53) Google Scholar), by the abundant expression of Rab 3D in cells where glucose transport is regulated by insulin(6Baldini G. Hohl T. Lin H.Y. Lodish H.F. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 5049-5052Crossref PubMed Scopus (195) Google Scholar), and by the colocalization of the adipocyte GLUT4 transporter with proteins that are thought to be exclusive to synaptic and secretory vesicles of neuroendocrine cells(7Cain C.C. Trimble W.S. Lienhard G.E. J. Biol. Chem. 1992; 267: 11681-11684Abstract Full Text PDF PubMed Google Scholar). These proteins, named vesicle-associated membrane proteins or VAMPs 1The abbreviations used are: VAMPvesicle-associated membrane proteinPVDFpolyvinylidine difluoridePBSphosphate-buffered salinePMplasma membranesHDMhigh density microsomesLDMlow density microsomesTMtotal membranePAGEpolyacrylamide gel electrophoresisSCAMPsecretory carrier associated membrane protein. /synaptobrevins, consist of two isoforms, I and II, both of 18 kDa, and are found in synaptic vesicles of neuronal cells(8Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar), secretory granules of pancreatic neuroendocrine (9Baumert M. Maycox P.R. Navone F. De Camilli P. Jahn R. EMBO J. 1989; 8: 379-384Crossref PubMed Scopus (396) Google Scholar), and pancreatic exocrine cells(10Braun J.E.A. Fritz B.A. Wong S.M.E. Lowe A.W. J. Biol. Chem. 1994; 269: 5328-5335Abstract Full Text PDF PubMed Google Scholar, 11Gaisano H.Y. Sheu L. Foskett J.K. Trimble W.S. J. Biol. Chem. 1994; 269: 17062-17066Abstract Full Text PDF PubMed Google Scholar). VAMPs are thought to be involved in docking/fusion of the vesicles/granules with the plasma membrane(8Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar). An antibody raised to the common domain of all VAMPs was recently shown to react with polypeptides of molecular mass of 17 and 18 kDa in GLUT4-containing membranes of rat white adipocytes(7Cain C.C. Trimble W.S. Lienhard G.E. J. Biol. Chem. 1992; 267: 11681-11684Abstract Full Text PDF PubMed Google Scholar), and with yet a third band of lower molecular weight(12Laurie S.M. Cain C.C. Lienhard G.E. Castle J.D. J. Biol. Chem. 1993; 268: 19110-19117Abstract Full Text PDF PubMed Google Scholar).It has also been suggested that glucose transporters recycle constitutively through the endocytic pathway (a phenomenon involving continuous formation of clathrin coated vesicles, their incorporation into endosomes and subsequent fusion of endosome-derived vesicles with the plasma membrane). The participation of the endocytic pathway in the intracellular traffic of glucose transporters is based on morphological and kinetic criteria. GLUT4 proteins can be detected by immunocytochemistry in clathrin-coated vesicles and early endosomes of white (13Robinson L.J. Pang S. Harris D.S. Heuser J. James D.E. J. Cell Biol. 1992; 117: 1181-1196Crossref PubMed Scopus (256) Google Scholar) and brown (4Slot J.W. Geuze H.J. Gigengack S. Lienhard G.E. James D.E. J. Cell Biol. 1991; 113: 123-135Crossref PubMed Scopus (709) Google Scholar) adipocytes, although it is still debated whether insulin augments (13Robinson L.J. Pang S. Harris D.S. Heuser J. James D.E. J. Cell Biol. 1992; 117: 1181-1196Crossref PubMed Scopus (256) Google Scholar) or diminishes (14Chakrabarti R. Buxton J. Joly M. Corvera S. J. Biol. Chem. 1994; 269: 7926-7933Abstract Full Text PDF PubMed Google Scholar) the amount of GLUT4 in these structures. GLUT4 immunolabeling has also been demonstrated in endosomes filled with extracellular material(4Slot J.W. Geuze H.J. Gigengack S. Lienhard G.E. James D.E. J. Cell Biol. 1991; 113: 123-135Crossref PubMed Scopus (709) Google Scholar). The presence of the glucose transporter in endosomes could reflect its routing to the lysozomes, or alternatively its storage in a compartment capable of recycling. It has been shown that GLUT4 cycles to and from the cell membrane in the basal state(15Yang J. Holman G.D. J. Biol. Chem. 1993; 268: 4600-4603Abstract Full Text PDF PubMed Google Scholar, 16Kandror K. Pilch P.F. J. Biol. Chem. 1994; 269: 138-142Abstract Full Text PDF PubMed Google Scholar), perhaps through endocytosis signals in its primary sequence(17Corvera S. Chawla A. Chakrabarti R. Joly M. Buxton J. Czech M.P. J. Cell Biol. 1994; 126: 979-989Crossref PubMed Scopus (103) Google Scholar, 18James D.E. Piper R.C. Slot J.W. Trends Cell Biol. 1994; 4: 120-127Abstract Full Text PDF PubMed Scopus (94) Google Scholar). Consistent with this view, disruption of the clathrin coat by low K+ causes accumulation of GLUT4 polypeptides at the cell surface(19Nishimura H. Zarnowski M.J. Simpson I.A. J. Biol. Chem. 1993; 268: 19246-19253Abstract Full Text PDF PubMed Google Scholar). Insulin augments the rate of appearance of GLUT4 glucose transporters at the cell surface (15Yang J. Holman G.D. J. Biol. Chem. 1993; 268: 4600-4603Abstract Full Text PDF PubMed Google Scholar, 16Kandror K. Pilch P.F. J. Biol. Chem. 1994; 269: 138-142Abstract Full Text PDF PubMed Google Scholar) and reduces their rate of endocytosis(20Jhun B.H. Rampal A.L. Liu H. Lachaal M. Jung C.Y. J. Biol. Chem. 1992; 267: 17710-17715Abstract Full Text PDF PubMed Google Scholar, 21Czech M.P. Buxton J.M. J. Biol. Chem. 1993; 268: 9187-9190Abstract Full Text PDF PubMed Google Scholar). Upon insulin removal, surface photolabeled GLUT4 glucose transporters internalize, and a fraction of them can re-emerge at the cell surface upon a second exposure to insulin(22Satoh S. Nishimura H. Clark A.E. Kozka I.J. Vannucci S.J. Simpson I.A. Quon M.J. Cushman S.W. Holman D.G. J. Biol. Chem. 1993; 268: 17820-17829Abstract Full Text PDF PubMed Google Scholar). Collectively, these observations suggest that GLUT4 continuously recycles to and from the cell surface and that the rate of this process is altered by insulin (18James D.E. Piper R.C. Slot J.W. Trends Cell Biol. 1994; 4: 120-127Abstract Full Text PDF PubMed Scopus (94) Google Scholar).Recently, a protein belonging to the synaptobrevin family was purified and found to be localized in recycling endosomes(23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). This protein, named cellubrevin, is thought to be a marker of endocytic vesicles rather than of specialized secretory vesicles. In contrast to the circumscribed tissue expression of VAMP-1 and VAMP-2, the 17-kDa cellubrevin is widely distributed in a variety of tissues(23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). This protein is present in coated vesicles isolated from hepatocytes (23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar) and colocalizes with the transferrin receptor in CV-1 (23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar) and Chinese hamster ovary cells(24Galli T. Chilcote T. Mundigl O. Binz T. Niemann H. De Camilli P. J. Cell Biol. 1994; 125: 1015-1024Crossref PubMed Scopus (195) Google Scholar).In the present study, we investigate whether the endocytic marker cellubrevin is expressed in mouse 3T3-L1 cells, if its levels change during differentiation from fibroblasts into adipocytes, and whether this protein is involved in insulin-regulated intracellular traffic. We describe that, like GLUT4, cellubrevin translocates to the plasma membrane in response to insulin. Furthermore, cellubrevin is found to be a resident protein of GLUT4-containing vesicles, and immunopurified cellubrevin-containing vesicles contain GLUT4. Finally, we show that the VAMP isoform expressed in adipocytes is VAMP-2. These results suggest that there are GLUT4 vesicles containing cellubrevin and VAMP-2. The possibility that there are two types of intracellular GLUT4-containing vesicles is discussed, vis a vis the possible participation of the endosomal system and of vesicular exocytosis in the regulation of intracellular traffic of the GLUT4 transporter.EXPERIMENTAL PROCEDURESMaterialsCell culture medium, serum, supplements and reagents were obtained from Life Technologies, Inc. Polyvinylidene difluoride (PVDF, 0.2-μm pore size) membranes were obtained from Bio-Rad. The immunoprecipitating anti-GLUT4 antiserum was prepared by immunizing rabbits with synthetic peptides corresponding to the 12 C-terminal amino acids of the GLUT4 protein as described by James et al.(2James D.E. Strube M. Mueckler M. Nature. 1989; 338: 83-87Crossref PubMed Scopus (669) Google Scholar). The following antibodies were used for immunoblotting. Anti-GLUT4 antiserum (East Acres Biologicals, Southbridge, MA) was diluted 1:1000. Affinity-purified anti-cellubrevin antiserum D204 (kind gift from Dr. T. C. Südhof, Howard Hughes Medical Institute, Dallas, TX) (23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar) was used at a 1:1000 dilution. The anti-cellubrevin antibody does not cross-react with either VAMP-1 or VAMP-2(23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). Two anti-VAMP-2 antibodies were used. One was raised to a peptide corresponding to the unique N-terminal sequence of VAMP-2 (kind gift from Dr. W. Trimble, University of Toronto) that was previously shown to react specifically and uniquely with this protein and not with either VAMP-1 or cellubrevin(25Volchuk A. Mitsumoto Y. He L. Liu Z. Habermann E. Trimble W. Klip A. Biochem. J. 1994; 304: 139-145Crossref PubMed Scopus (69) Google Scholar), and one raised to a fusion protein encompassing residues 1-96, which includes the middle domain common to other synaptobrevins as well as the N terminus sequence unique to VAMP-2. As discussed in this study, this antibody reacted only very weakly with cellubrevin. Anti α1 Na+/K+-ATPase monoclonal antibody 6H (a kind gift from Dr. M. Caplan, Yale University, New Haven, CT), was used at a 1:500 dilution. Fluorescein isothiocyanate-conjugated anti-rabbit IgG was from Jackson Immunoresearch Laboratories Inc, West Grove, PA. Goat serum was from Life Technologies, Inc. Slow Fade™ was from Molecular Probes, Inc., Eugene, OR. Recombinant VAMP-1 and VAMP-2 (glutathione S-transferase derivatives of the cytosolic portion of each synaptobrevin) were a gift from Dr. W. Trimble, University of Toronto. Purified tetanus toxin light chain was a kind gift from Dr. Ernst Habermann, Justus-Leibig Universitat, Giesen, Germany.ImmunofluorescenceIndirect immunofluorescence was performed according to the method of Piper et al.(26Piper R.C. Hess L.J. James D.E. Am. J. Physiol. 1991; 260: C570-C580Crossref PubMed Google Scholar) with slight modifications. 3T3-L1 cells were grown on glass coverslips, differentiated into adipocytes 2 days before reaching confluence, washed 2 times with serum-free Dulbecco's modified Eagle's medium, and fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. Excessive fixative was quenched by incubation in PBS containing 100 mM glycine for 15 min. Cells were then permeabilized with 0.1% Triton X-100 in PBS for 15 min at room temperature, washed 3 times with PBS, and incubated with 10% goat serum in PBS for 20 min. After three more washes in PBS, coverslips were incubated with anti-cellubrevin antibody (dilution 1:10 in PBS) at 4°C overnight. Cells were then washed 3 × 15 min in PBS, incubated for 60 min with 6 μg/ml fluorescein isothiocyanate-conjugated donkey anti-rabbit IgG at room temperature, and mounted in Slow Fade™. Coverslips were examined with an Olympus VANOX AHBT3 fluorescence microscope.Cell Culture and Subcellular Fractionation3T3-L1 fibroblasts, kindly provided by Dr. D. Lane (Johns Hopkins School of Medicine, Baltimore, MD), were differentiated into adipocytes as described previously(27Student A.K. Hsu R.Y. Lane M.D. J. Biol. Chem. 1980; 255: 4745-4750Abstract Full Text PDF PubMed Google Scholar). Cells from two 10-cm dishes/condition were pretreated for 3 h in serum-free Dulbecco's modified Eagle's medium and then incubated with or without 100 nM insulin for 20 min at 37°C and fractionated according to Piper et al.(26Piper R.C. Hess L.J. James D.E. Am. J. Physiol. 1991; 260: C570-C580Crossref PubMed Google Scholar) to obtain plasma membranes (PM), high density microsomes (HDM), and low density microsomes (LDM). The protein content of the fractions was determined by the Bio-Rad procedure. Total membranes (TM) from 3T3-L1 fibroblasts or adipocytes were prepared as follows. Monolayers were rinsed twice with ice-cold homogenization buffer (255 mM sucrose, 0.5 mM phenylmethylsulfonyl fluoride, 1 μM pepstatin A, 1 μM leupeptin, 10 μM E-64, 1 mM EDTA, and 20 mM Na-HEPES, pH 7.4), scraped vigorously with a rubber policeman into 4 ml of the same buffer, and homogenized with 30 strokes of a Teflon pestle in a glass homogenizer at 1200 rotations/min. The homogenate was centrifuged at 1000 × g for 3 min to pellet the nuclei and large cellular debris. The supernatant was centrifuged at 245,000 × g for 90 min to sediment the total membranes. The membrane samples were resuspended in homogenization buffer.Treatment with Tetanus ToxinMembrane fractions derived from 3T3-L1 cells were freeze dried and resuspended in 20 μl of potassium glutamate buffer (138 mM potassium glutamate, 20 mM HEPES, 8 mM MgCl2, 0.285 mM CaCl2, 1 mM EGTA, 1 mM dithiothreitol, pH 7.15) containing 0.5% Triton X-100 with or without the indicated concentrations of tetanus toxin light chain prepared as in (28Weller U. Dauzenroth M.E. Meyer Zu Heringdorf D. Habermann E. Eur. J. Biochem. 1989; 182: 649-656Crossref PubMed Scopus (65) Google Scholar). The samples were incubated at 37°C in a water bath under constant agitation for 1 h; a brief vortex was given at the 30-min interval. The samples were then boiled for 3 min in 20 μl of 2 × concentrated Laemmli sample buffer and immediately resolved by SDS-PAGE. Proteins were then electrotransferred and immunoblotted.Immunoprecipitation of GLUT4-containing Vesicles and of Cellubrevin-containing VesiclesGLUT4-containing vesicles were immunoprecipitated by a modification of the protocol of Laurie et al.(12Laurie S.M. Cain C.C. Lienhard G.E. Castle J.D. J. Biol. Chem. 1993; 268: 19110-19117Abstract Full Text PDF PubMed Google Scholar). Briefly, 6 μl of anti-GLUT4 serum or preimmune serum were incubated with 100 μl of sheep anti-rabbit IgG magnetic beads (M-280 Dynabeads, Dynal Inc, Great Neck, NY) in 100 mM potassium phosphate buffer, pH 7.4 (KP) containing 5 mg/ml bovine serum albumin for 5-6 h at 4°C under constant rotation then washed 3 times with 0.5 ml of KP-bovine serum albumin. Adipocyte monolayers were fractionated as above using 5 ml of homogenization buffer/10-cm diameter dish to remove nuclei, mitochondria, plasma membranes, and heavy microsomes but without sedimenting the LDM. The supernatant (500 μl containing approximately 350 μg protein of LDM), adjusted to 100 mM KP, was added to the beads and incubated 16-18 h at 4°C under constant rotation. The supernatant was removed, the beads were washed 3 times with KP without bovine serum albumin, and the resulting supernatants were combined with the first one. Pooled supernatants were centrifuged at 200,000 × g for 60 min, and the sedimented membranes were resuspended in 2 × concentrated Laemmli sample buffer (29Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (206024) Google Scholar) containing 8 M urea (sample named Supernatant). The material bound to the pelleted magnetic beads was equally solubilized in Laemmli sample buffer containing 8 M urea (sample named Pellet). The entire volumes of Supernatant and Pellet were resolved by SDS-PAGE. For immunopurification of cellubrevin-containing vesicles, a similar procedure was employed except that the anti-cellubrevin antiserum was used as primary antibody and an irrelevant serum was used for the nonimmune serum control.PAGE and ImmunoblottingMembrane proteins were separated by SDS-PAGE (29Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (206024) Google Scholar) in 13 and 14% polyacrylamide gels (as indicated in the figure legends) and transferred electrophoretically to PVDF membranes. Immunoblotting was carried out as described previously(25Volchuk A. Mitsumoto Y. He L. Liu Z. Habermann E. Trimble W. Klip A. Biochem. J. 1994; 304: 139-145Crossref PubMed Scopus (69) Google Scholar). Bound monoclonal antibodies were detected with 125I-labeled sheep-anti-mouse IgG. Polyclonal antibodies with 125I-labeled protein A or Enhanced Chemiluminescence (Amersham Corp.) as indicated in the figure legends. Scanning of x-ray films was done using a PDI model DNA35 scanner with the version 1.3 of the Discovery Series one-dimensional gel analysis software.RNA Isolation and Northern Blot AnalysisTotal RNA was extracted from 3T3-L1 fibroblasts or adipocytes and from rat brain (as control), using the single-step RNA isolation with acid guanidinium thiocyanate-phenol chloroform(30Chomczynski P. Sacchi N. Anal. Biochem. 1987; 162: 156-169Crossref PubMed Scopus (62983) Google Scholar), quantitated by its 260/280 UV absorbance and electrophoresed under denaturing conditions in 1.2% (w/v) agarose gels containing 8% (v/v) formaldehyde. RNA was then transferred onto nytron membranes and baked for 90 min at 80°C in a vacuum oven. Equal loading of all samples containing 20 μg of RNA (3T3-L1 fibroblasts and adipocytes) was confirmed by the ethidium bromide fluorescence incorporated into nucleic acids. Where indicated, mRNA was isolated using the Dynabeads mRNA direct kit (Dynal, Oslo, Norway) according to the manufacturer's instructions. The nytron membranes containing RNA or mRNA samples were pre-hybridized overnight at 42°C with 200 μg/ml salmon sperm DNA in 6 × SSPE (1 × SSPE: 0.15 M NaCl, 10 mM NaH2PO4, 1 mM EDTA, pH 7.4), 10 × Denhardt's solution, 0.5% SDS. Hybridization was carried out in 50% formamide, 6 × SSPE, 0.5% SDS, 5% dextran sulfate, 100 μg/ml salmon sperm DNA for 16 h at 37°C by adding [α-32P]dCTP-labeled VAMP-1 cDNA (5.5 × 108 cpm/μg DNA) or VAMP-2 cDNA (5.7 × 108 cpm/μg DNA), each labeled by the random primer method. The VAMP-1 probe corresponded to nucleotides 1-1470 of the rat VAMP-1 sequence(31Elferink L.A. Trimble W.S. Scheller R.H. J. Biol. Chem. 1989; 264: 11061-11064Abstract Full Text PDF PubMed Google Scholar), and the VAMP-2 probe corresponded to nucleotides 1-432. Both were the kind gift of Dr. W. Trimble, University of Toronto. Following hybridization, nytron membranes were washed 3 × 5 min with 1 × SSC, 0.1% SDS at room temperature and 2 × 1 h with 0.1 × SSC, 0.5% SDS at 65°C prior to exposure to DuPont autoradiographic film at −70°C.RESULTSExpression of Cellubrevin during Differentiation of 3T3-L1 CellsFig. 1 examines the expression of cellubrevin in total membranes of undifferentiated 3T3-L1 fibroblasts and differentiated 3T3-L1 adipocytes. The first lane (Br) shows that the anti-cellubrevin antibody does not react with proteins present in rat brain homogenates, suggesting that it does not cross-react with either VAMP-1 or VAMP-2, abundant proteins of brain stem and brain, respectively(32Trimble W.S. J. Physiol. 1993; 87: 107-115Google Scholar). Lanes labeled FTM (fibroblast total membranes) contained total membranes of 3T3-L1 fibroblasts, in which the 17 kDa band of cellubrevin was not detectable. Overexposure of the x-ray films revealed a very faint band in this fraction in the absence of tetanus toxin light chain (lane "[minus]"), and this weak band totally disappeared in fibroblast total membranes treated with the toxin in vitro (lane "+"). Lanes labeled ATM (adipocyte total membranes) show the significantly higher abundance of the 17 kDa band in total membranes of fully differentiated 3T3-L1 adipocytes compared with undifferentiated fibroblasts. Treatment with tetanus toxin light chain in vitro completely eliminated the immunoreactive 17 kDa band. This strongly supports the notion that the reactive band is indeed cellubrevin, since the clostridium endopeptidase cleaves VAMPs/synaptobrevins and cellubrevin but not other known proteins(8Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar, 32Trimble W.S. J. Physiol. 1993; 87: 107-115Google Scholar). The endopeptidase did not alter the protein profile of the membranes as assessed by Coomassie Blue staining or the content of other membrane proteins such as the GLUT4 glucose transporter (results not shown). These results demonstrate that the abundance of cellubrevin, per mg of protein of total membranes, sharply increases during differentiation of 3T3-L1 fibroblasts into adipocytes (also shown more clearly in the last two lanes of Fig. 3).Figure 3:Effect of insulin on the subcellular distribution of α1 Na+/K+-ATPase, GLUT4 glucose transporters, and cellubrevin in 3T3-L1 adipocytes. The following membrane fractions were isolated from control (C) or insulin-treated (I) 3T3-L1 cells as described under "Experimental Procedures": adipocyte high density microsomes (HDM), adipocyte plasma membranes (PM), adipocyte low density microsomes (LDM), and adipocyte (A) or fibroblast (F) total membranes (TM). Twenty μg of HDM, PM or LDM, and 40 μg of TM were resolved by SDS-PAGE on 13% polyacrylamide gels and electrotransferred to PVDF filters. A, the top portion of the filters was immunoblotted with antibodies to the α1 subunit of the Na+/K+-ATPase (apparent Mr 105,000), and the bottom part was immunoblotted with antibodies to the GLUT4 glucose transporter (apparent Mr 53,000), followed by detection with 125I-labeled sheep-anti-mouse IgG or protein A, respectively. B, parallel samples were resolved on a 14% polyacrylamide gel, electrotransferred, and immunoblotted with anti-cellubrevin antibody detected by the enhanced chemiluminescence procedure.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The last four lanes analyze the subcellular distribution of cellubrevin in purified LDM and PM from 3T3-L1 adipocytes. Lanes labeled "[minus]" show the amount of the 17-kDa protein in untreated membranes, and lanes labeled "+" show its complete disappearance upon in vitro treatment with tetanus toxin light chain. The 17-kDa protein was present in both the LDM and PM fractions, but per mg of protein it was enriched in the LDM. Scans of six independent experiments show that the relative enrichment per mg of protein in LDM to PM was 2.64 ± 0.54 to 1.00. On average, the subcellular fractionation of one confluent 10-cm dish of 3T3-L1 adipocytes yields approximately 160 μg of protein of PM, 40 μg of protein of HDM, and 225 μg of protein of LDM. When this yield of protein in the LDM and PM fractions was considered, the relative recovery of cellubrevin in these fractions was 3.7 to 1.00. Cellubrevin was located largely in these two fractions, since analysis of HDM of these cells showed only minute amounts of the protein (shown later in Fig. 3). Finally, the anti-cellubrevin antibody did not detect any bands in samples of recombinant VAMP-2 or VAMP-1, respectively (results not shown). This confirms the specificity of the antibody, and its tetanus toxin sensitivity underscores the identity of the 17 kDa band seen in 3T3-L1 cell membranes as cellubrevin.This affinity-purified antibody was therefore used to immunolocalize cellubrevin in permeabilized 3T3-L1 adipocytes. The antibody was previously used to detect cellubrevin by immunofluorescence in CV-1 cells(23McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Jahn R. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). Fig. 2 shows a representative micrograph of the distribution of cellubrevin in a 3T3-L1 adipocyte. Panela shows the fluorescence image when focussing near the cell surface, and panelb shows the fluorescence image when focussing at the level of the nucleus. The two images show that immunoreactive cellubrevin was distributed in a pole around the

Referência(s)