- Email: [email protected]
Identification and characterization of the glucagon receptor from adipose tissue
EL-SEVIER
Molecular and Cellular Endocrinology 101 (1994) 257-261
Identification
and characterization of the glucagon receptor from adipose tissue
Victoria Iwanij a,*, Therese M. Amos a, Charles J. Billington b a Department of Genetics and Cell Biology, University of Minnesota, I445 Gortner Avenue, St. Paul, MN 55108, USA b Department of Medicine, Vniversi@ of Minnesota, and Metabolic, Endocrine Section, VA Medical Center, Minneapolis, MN 55417, USA (Received 5 November 1993; accepted 7 January 1994)
Abstract ‘251-glucagon was directly cross-linked to its receptor in isolated adipocyte plasma membranes using a UV irradiation procedure. This investigation resulted in identification of an adipocyte glucagon receptor complex of 62 kDa, present both in white and brown adipose tissues. The specificity of labeling was shown by interference of unlabeled hormone with incorporation of radioactive glucagon into 62 kDa species. Treatment of adipose plasma membranes with N-glycanase resulted in appearance of intermediate species, indicating that the glucagon receptor is modified with several N-linked oligosaccharide chains similarly to the hepatic glucagon receptor. Peptide mapping of the affinity labeled adipose membranes with Stuphylococcus aureus V8 protease generated three distinct receptor fragments identical to that of the hepatic receptor. Overall, the biochemical characterization of the rat adipocyte glucagon receptor indicates that it closely resembles the hepatic glucagon receptor. Key words:
Glucagon; Receptor; Adipocyte; Brown fat; Liver; Affinity labeling
1. Introduction Glucagon, a 29 amino acid peptide hormone, is secreted by the pancreas to maintain plasma glucose concentrations during fasting. In the liver, glucagon stimulates glycogen breakdown and gluconeogenesis, while in adipose tissue glucagon stimulates lipolysis with the concomitant release of glycerol and non-estrified fatty acids. Increased availability of free fatty acids activates hepatic and muscle P-oxidation resulting in higher levels of acetyl-CoA that in turn further stimulate gluconeogenesis (reviewed by Selby and Sherratt, 1989). In addition to its regulation of lipolysis, glucagon also may be involved in regulation of the thermogenesis of brown fat in a manner similar to the adrenergic hormones. Adipose tissue represents an important target tissue for the glucagon action, however it has been less extensively investigated than the liver. Birnbaumer and Rod-
bell (1969) have shown that the biological effects of glucagon in adipose tissue result from an increase of
the adenylyl cyclase activity. Further studies (Rodbell et al., 1971; Cuatrecasas, 1971; Desbuquois et al., 1973; Desbuquois and Laudat, 1974; Livingstone et al., 1974) demonstrated the presence of specific glucagon binding sites in dissociated fat cells and isolated adipocyte plasma membranes. However, to date, biochemical studies that characterize adipocyte glucagon receptor are lacking. Several laboratories have used an affinity labeling approach to identify the hepatic glucagon receptor as a 62 kDa glycoprotein that contains four N-linked oligosaccharide chains and intramolecular disulfide bonds (Johnson et al., 1981; Demoliou-Mason and Epand, 1982; Iyengar and Herberg, 1984; Iwanij and Hur, 1985). We have used a direct affinity crosslinking approach, developed in our laboratory, to obtain structural information about the adipose tissue glucagon receptor.
2. Materials
and methods
1. Preparation of the adipocyte ghosts and plasma membranes: Isolation of adipocytes using collagenase diges* Corresponding
author. Tel.: 612-624-4942; Fax: 612-625-5754.
tion of rat epididymal fat pads and the lysis of cells to
0303-7207/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0303-7207(94)00022-2
258
I/: Iwanij et al. /Molecular
and Cellular Endocrinology
obtain cell ghosts was carried out according to the protocol of Rodbell (19721. Alternatively, adipocyte membrane preparation was carried out as follows: epididymal fat pads were removed from the animals, cut into small pieces with scissors, suspended in portions of two pads per 25 ml of homogenization buffer containing 1 mM bicarbonate solution and protease inhibitors (1 mM EDTA, pepstatin 5 pg/ml, leupeptin 5 pg/ml, and Trasylol 100 U/ml), then subjected to homogenization in a Polytron homogenizer for 10 seconds at setting 5. Homogenate was filtered through 4 layers of cheese cloth and residual tissue was re-homogenized as described above. Filtrate was combined and centrifuged in Sorval SS-34 rotor at 20000 rpm for 20 min. Collected pellet was resuspended in a small volume of homogenization buffer and prepared for plasma membrane flotation as described by Neville (1968) for the liver plasma membranes. This protocol was scaled down for the isolation of plasma membranes from rat brown fat (approximately 0.5 of tissue). Membranes collected from the top of the gradients were washed by centrifugation and stored in liquid nitrogen prior to the experiment. 2. Preparation of the liver plasma membranes was carried out as described by Neville (1968), and modified by Rodbell (1972). 3. Binding and cross-linking of ‘251-glucagon to adipocyte and liver plasma membranes were carried out as described previously (Iwanij and Hur, 1985). About 100-200 pg of membranes were incubated in 1 ml of binding buffer containing protease inhibitors with ‘251-glucagon (2-5 x lo6 cpm) for 30 min at 4°C. Membranes were washed 3 times by centrifugation in Eppendorf centrifuge, resuspended in 0.3 ml of 50 mM phosphate buffer, pH 7.4, and subjected to UV irradiation for 20 min. Membranes were collected by centrifugation and separated by SDS-PAGE according to Laemmli (1970) and analyzed by autoradiography. 4. The treatments of adipose and liver membranes with N-glycanase (Genzyme, MA) and Staphylococcus aureus V8 protease (Sigma) were carried out as described by Iwanij and Vincent (1990). 5. Immunobloting procedure was carried out as described previously (Iwanij and Vincent, 19901, using procedure of Towbin et al. (1979). Briefly, membranes 5-20 pg were separated using mini gel apparatus (Hoeffer Scientific) and transferred electrophoretically onto the nitrocellulose. Blots were pretreated with 2% gelatin containing blocking solution, incubated with primary antibodies, extensively washed and then probed with secondary antibodies conjugated to the alkaline phosphatase (Promega, anti-mouse IgGI. 6. Miscellaneous procedures: Iodination of glucagon was described by Lin et al. (19751, and separation of iodinated glucagon from unincorporated iodine was described by Iwanij and Hur (1985). The protein concen-
101 (1994) 257-261
tration was determined by the fluorescamine dure of Udenfriend et al. (1972).
proce-
3. Results
The goal of our investigation was to identify and characterize the glucagon receptor present in the adipose tissue.’ To this end, we have used direct UV irradiation as a means of cross-linking glucagon to the receptor sites. Adipocyte ghosts isolated from enzymatically dissociated adipocytes (Rodbell, 1972) or plasma membranes prepared directly from epididymal fat (EF) pads were incubated with ‘251-glucagon, washed and irradiated with UV light as described above. Membranes were then analyzed by SDS-PAGE and autoradiography. A single radioactive band was observed in the EF plasma membranes (Fig. 1A) and lysed adipocyte ghosts (results not shown). Labeling of this band was abolished in the presence of an excess of unlabeled glucagon indicating that the band represents the glucagon receptor. The labeled adipocyte receptor
A
A
B
C
Fig. 1. UV-induced cross-linking of 1251-glucagonto adipocyte plasma membrane isolated from rat epididymal and brown fat. Membrane preparation and the cross-linking was carried out as described in the Materials and methods section. Panel A: lane A, liver plasma membranes (100 pg); lane B, epididymal fat plasma membranes (200 &; lane C, epididymal fat plasma membranes (200 fig) for which binding of ‘=I-glucagon was carried out in the presence of 2 mM unlabeled glucagon. Panel B: lane A, liver plasma membranes (50 Kg); lane B, brown fat plasma membranes GXJO&; lane C, brown fat plasma membranes (200 pg) for which binding of %glucagon was carried out in the presence of 2 PM unlabeled glucagon. Longer time exposure was needed to visualize BAT receptor as compared with liver tissue receptor. Arrowhead points to the labeled glucagon receptor band.
259
K Iwanij et al. /Molecular and Cellular Endocrinology 101 (1994) 257-261
band was significantly less intense than the labeled hepatic band, that observation being consistent with the lesser density of adipocyte receptor sites as previously reported (Desbuquois and Laudat, 1974). The adipocyte glucagon receptor co-migrated with the hepatic glucagon receptor on the SDS polyacrylamide gels indicating that they represent similar or identical proteins. Having established that our cross-linking procedure is sensitive enough to detect the lower concentrations of the adipose glucagon receptor, we investigated whether brown adipose tissue (BAT) also contains the glucagon receptor. We isolated plasma membranes from brown fat and were able to detect reproducible and specific binding of ‘251-glucagon. When we performed glucagon binding and UV cross-linking experiments with membranes isolated from BAT (Fig. lB), the results were identical to those obtained with the EF membranes. To further characterize the adipocyte glucagon receptor, we have carried out immunoblot experiments using monoclonal antibodies prepared against the hepatic receptor (Iwanij and Vincent, 1990). As shown in Fig. 2, Mab 6ElO recognized a single band in both EF and BAT plasma membranes. The molecular weight of the stained protein species from fat was the same as for the hepatic glucagon receptor. In accord with the affinity labeling experiments, the intensity of staining of t,he adipose receptor band was significantly lower than that in the liver preparations. Previous analysis of the hepatic glucagon receptor indicated that the receptor contains four N-linked
Fig. 2. Immunoblot analysis of membranes isolated from rat epididyma1 and brown fat. Membrane preparation and immunoblotting was carried out as described in the Materials and methods section. Immunoblot was probed with monoclonal antibody 6E10; lanes L, liver plasma membranes (10 pg); lane EF, epididymal fat plasma membranes (20 pg); lane BF, brown fat plasma membranes (20 pg). Arrowhead points to the immunoreactive band corresponding to the glucagon receptor.
A
B
C
D
Fig. 3. Immunoblot analysis of the deglycosylated adipocyte glucagon receptor. EF and liver plasma membranes were treated with Nglycanase, separated on the same SDS gel, transferred on the same sheet of nitrocellulose, and probed with 6ElO monoclonal antibody. Lane A, N-glycanase treated adipocyte plasma membranes (30 pg); lane B, untreated adipocyte plasma membranes (20 pg); lane C, N-glycanase treated liver plasma membranes (20 pg); lane D, untreated liver plasma membranes (10 cl.g).
oligosaccharides branches (Iyengar and Herberg, 1984; Iwanij and Hur, 1985). Therefore, we have investigated whether the adipose glucagon receptor is also N-glycosylated. Adipose and liver plasma membranes were treated with N-glycanase, an enzyme that cleaves Nlinked oligosaccharide residues as outlined in the Materials and methods section. After the treatment, membranes were separated by SDS-PAGE, transferred onto nitrocellulose and probed with Mab 6ElO. As shown in Fig. 3, treatment of the EF membranes with Nglycanase resulted in the loss of the 62 kDa immunoreactive band. That band was replaced by two closely spaced doublets. Intermediate bands of identical size were observed also in liver plasma membrane preparations. This result indicated that the adipose glucagon receptor contains multiple N-linked oligosaccharides similar to those of the hepatic receptor. In our previous work (Iwanij and Vincent, 19901, we have noted that digestion of affinity labeled hepatic glucagon receptor with V8 protease results in a characteristic pattern of three sharply defined radioactive bands. We, therefore, carried out treatment of the affinity labeled EF membranes with V8 protease. The autoradiogram shown in Fig. 4 indicates that the V8 protease digest of adipose receptor resulted in the
260
V: Iwunij et al. /Molec&r
AB
and Cellular Endocrinology
CD
Pig. 4. Proteolytic mapping of the adipocyte glucagon receptor with S. aureus V8 protease. Affinity labeled adipocyte and liver membranes were digested with V8 protease, separated by the same SDS gel and analyzed by autoradiography. Adipocyte membranes lanes required longer exposure time to visualize the digestion fragments. .Lane A, liver plasma,~embrane untreated control (100 pg); lane 8, Iiver plasma membranes (100 pgI digested with V8 protease; lane C, adipocyte plasma membranes (200 ~g) digested with V8 protease; lane D, adipocyte plasma membrane (150 kg) untreated control. Open circles denote position of the receptor fragments.
generation of three fragments: one 59-60 kDa band, and a doublet of about 25-30 kDa. These fragments co-migrated on the SDS gels with the proteolytic bands generated from the hepatic receptor (Fig. 4). These results indicate that both hepatic and adipose receptors are cleaved into identical sets of proteolytic fragments.
4. Discussion Adipose tissue is an important target tissue for the action of the pancreatic peptide glucagon. Glucagon has been shown to increase plasma concentration of free fatty acids in vivo and to stimulate release of glycerol and free fatty acids from intact adipose tissue and dissociated adipocytes (Lefebvre, 1972). Interestingly, the adipocyte responsiveness to giucagon varies with the physiological state of the cell; it was shown to decline during adipocyte aging (Livingston et al., 1974) and to increase in experimentally induced diabetes (Chatzipanteli and Saggerson, 1983). The effect of glucagon on adipose tissue metabolism .has been extensively studied using dissociated adipocyte cells. The characteristic lipolytic response following glucagon binding to its receptor is believed to be mediated by an increase in CAMP levels. Some investigators have noted a lower sensitivity of adipose tissue as
lOI(1994)
257-261
compared to liver, and have suggested the presence of adipose tissue specific receptor subtypes (reviewed in Lefebvre, 1972). However, detailed experiments carried out by Desbuquois and Laudat 0974) that included determination of the binding constants, effects of nucleotides on hormone binding, sensitivity to proteolytic enzymes, and phospholipase A digestion all indicated that the properties of the adipose glucagon binding sites closely resemble those of the liver, and suggested that the observed differences arose from a variations in the adipocyte dissociation procedures. Our investigation of the structural properties of the adipose glucagon receptor supports the conciusions of Desbuquois and Laudat (1974). In order to identify and characterize the adipose tissue receptor, we have applied a direct UV irradiation affinity cross linking approach. This investigation resulted in the identification of the glucagon-receptor complex of 62 kDa estimated MW present both in epididymal and brown fat. The size of the affinity labeled rat adipose glucagon receptor is identical to the previously characterized rat liver receptor. Immunobloting experiment with antiglucagon receptor antibodies also identified the presence of the major immunoreactive band in the membranes isolated from epididymal and brown fat, identical in size to the band observed in liver plasma membranes. The intensity of the radiolabeled and immunoreactive band detected in the adipose membranes is significantly lower than that one present in liver indicating lower levels of the glucagon receptor in‘ fat tissue as previously reported by Desbuquois and Laudat (1974). Treatment of the adipose plasma membranes with N-glycanase, an enzyme that cleaves N-linked oligosaccharides, resulted in the appearance of several intermediate species indicating that the adipose glucagon receptor is a glycoprotein with a level of glycosylation similar to that of the hepatic receptor. We have carried out structural analysis of the affinity labeled EF glucagon receptor by proteolytic treatment with S. aureus V8 protease. Autoradiographic analysis showed the presence of three distinct radioactive fragments identical in size to ones obtained with the hepatic receptor. Overall structural characterization described in this report indicates that glucagon receptor present in rat epidid~al and brown adipose tissues shows a striking similarity to the hepatic receptor and most probably represent identical structures.
5. Acknowledgements This work was supported in part by funds from the American Diabetes Association, Minnesota Chapter grant, the American Cancer Society grant BC-688, and University of Minnesota institutional ACS grant IN-13-
K Iwanij et al. /Molecular and Cellular Endocrinology I01 (1994) 257-261
29-2 (to V.I.). We thank Dr. Norman Kagan for critical reading of this manuscript.
6. References Birnbaumer, L. and Rodbell, M. (1969) J. Biol. Chem. 244, 34773482. Chatzipanteh, K. and Saggerson, D. (1983) FEBS Lett. 155, 135-138. Cuatrecasas, P. (1971) J. Biol. Chem. 246, 6532-6542. Demohou-Mason, C. and Epand (1982) Biochemistry 21, 1996-2004. Desbuquois, B., Laudat, M-H. and Laudat, P. (1973) Biochem. Biophys. Res. Commun. 53, 1187-1194. Desbuquois, B. and Laudat, M-H. (1974) Mol. Cell. Endocrinol. 1, 355-370. Iwanij, V. and Hur, K.C. (1985) Proc. Natl. Acad. Sci. USA 82, 325-329. Iwanij, V. and Vincent, A.C. (1990) J. Biol. Chem. 265,21302-21308.
261
Iyengar, R. and Herberg, J.T. (1984) J. Biol. Chem. 259, 5222-5229. Johnson, G.L., MacAndrew, V.I. and Pilch, P.F. 0981) Proc. Natl. Acad. Sci. USA 78, 875-878. Laemmli, U.K. (1970) Nature 227, 680-685. Lefebvre, P.J. (1972) in Glucagon (Lefebvre, P.J. and Unger, R.H., eds.), pp. 109-121, Pergamon Press, Oxford. Lin, M.C., Wright, D.E., Hruby, V.J. and Rodbell, M. (1975) Biochemistry, 14, 1559-1563. Livingston, J.N., Cuatrecasas, P. and Lockwood, D.H. (1974) J. Lipid Res. 15, 26-32. Neville, D.M. (1968) Biochim. Biophys. Acta 154, 540-552. Rodbell, M., Krans, H.M.J., Pohl, S.L. and Birnbaumer, L. (1971) J. Biol. Chem. 246, 1861-1871. Rodbell, M. (1972) in Cyclic Nucleotide Research (Chasin, ed.), pp. 101-125, Marcel Dekker, New York. Selby, P.L. and Sherratt, H.S.A. (1989) Trends Pharmacol. Sci. 10, 495-500. Towbin, H., Staehlin, T. and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350-4354. Udenfriend, S., Stein, S., Bohlen, P., Dairman, W., Leimgruber, W., and Weigeli, M. (1972) Science 178, 871-872.
Molecular and Cellular Endocrinology 101 (1994) 257-261
Identification
and characterization of the glucagon receptor from adipose tissue
Victoria Iwanij a,*, Therese M. Amos a, Charles J. Billington b a Department of Genetics and Cell Biology, University of Minnesota, I445 Gortner Avenue, St. Paul, MN 55108, USA b Department of Medicine, Vniversi@ of Minnesota, and Metabolic, Endocrine Section, VA Medical Center, Minneapolis, MN 55417, USA (Received 5 November 1993; accepted 7 January 1994)
Abstract ‘251-glucagon was directly cross-linked to its receptor in isolated adipocyte plasma membranes using a UV irradiation procedure. This investigation resulted in identification of an adipocyte glucagon receptor complex of 62 kDa, present both in white and brown adipose tissues. The specificity of labeling was shown by interference of unlabeled hormone with incorporation of radioactive glucagon into 62 kDa species. Treatment of adipose plasma membranes with N-glycanase resulted in appearance of intermediate species, indicating that the glucagon receptor is modified with several N-linked oligosaccharide chains similarly to the hepatic glucagon receptor. Peptide mapping of the affinity labeled adipose membranes with Stuphylococcus aureus V8 protease generated three distinct receptor fragments identical to that of the hepatic receptor. Overall, the biochemical characterization of the rat adipocyte glucagon receptor indicates that it closely resembles the hepatic glucagon receptor. Key words:
Glucagon; Receptor; Adipocyte; Brown fat; Liver; Affinity labeling
1. Introduction Glucagon, a 29 amino acid peptide hormone, is secreted by the pancreas to maintain plasma glucose concentrations during fasting. In the liver, glucagon stimulates glycogen breakdown and gluconeogenesis, while in adipose tissue glucagon stimulates lipolysis with the concomitant release of glycerol and non-estrified fatty acids. Increased availability of free fatty acids activates hepatic and muscle P-oxidation resulting in higher levels of acetyl-CoA that in turn further stimulate gluconeogenesis (reviewed by Selby and Sherratt, 1989). In addition to its regulation of lipolysis, glucagon also may be involved in regulation of the thermogenesis of brown fat in a manner similar to the adrenergic hormones. Adipose tissue represents an important target tissue for the glucagon action, however it has been less extensively investigated than the liver. Birnbaumer and Rod-
bell (1969) have shown that the biological effects of glucagon in adipose tissue result from an increase of
the adenylyl cyclase activity. Further studies (Rodbell et al., 1971; Cuatrecasas, 1971; Desbuquois et al., 1973; Desbuquois and Laudat, 1974; Livingstone et al., 1974) demonstrated the presence of specific glucagon binding sites in dissociated fat cells and isolated adipocyte plasma membranes. However, to date, biochemical studies that characterize adipocyte glucagon receptor are lacking. Several laboratories have used an affinity labeling approach to identify the hepatic glucagon receptor as a 62 kDa glycoprotein that contains four N-linked oligosaccharide chains and intramolecular disulfide bonds (Johnson et al., 1981; Demoliou-Mason and Epand, 1982; Iyengar and Herberg, 1984; Iwanij and Hur, 1985). We have used a direct affinity crosslinking approach, developed in our laboratory, to obtain structural information about the adipose tissue glucagon receptor.
2. Materials
and methods
1. Preparation of the adipocyte ghosts and plasma membranes: Isolation of adipocytes using collagenase diges* Corresponding
author. Tel.: 612-624-4942; Fax: 612-625-5754.
tion of rat epididymal fat pads and the lysis of cells to
0303-7207/94/$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved SSDI 0303-7207(94)00022-2
258
I/: Iwanij et al. /Molecular
and Cellular Endocrinology
obtain cell ghosts was carried out according to the protocol of Rodbell (19721. Alternatively, adipocyte membrane preparation was carried out as follows: epididymal fat pads were removed from the animals, cut into small pieces with scissors, suspended in portions of two pads per 25 ml of homogenization buffer containing 1 mM bicarbonate solution and protease inhibitors (1 mM EDTA, pepstatin 5 pg/ml, leupeptin 5 pg/ml, and Trasylol 100 U/ml), then subjected to homogenization in a Polytron homogenizer for 10 seconds at setting 5. Homogenate was filtered through 4 layers of cheese cloth and residual tissue was re-homogenized as described above. Filtrate was combined and centrifuged in Sorval SS-34 rotor at 20000 rpm for 20 min. Collected pellet was resuspended in a small volume of homogenization buffer and prepared for plasma membrane flotation as described by Neville (1968) for the liver plasma membranes. This protocol was scaled down for the isolation of plasma membranes from rat brown fat (approximately 0.5 of tissue). Membranes collected from the top of the gradients were washed by centrifugation and stored in liquid nitrogen prior to the experiment. 2. Preparation of the liver plasma membranes was carried out as described by Neville (1968), and modified by Rodbell (1972). 3. Binding and cross-linking of ‘251-glucagon to adipocyte and liver plasma membranes were carried out as described previously (Iwanij and Hur, 1985). About 100-200 pg of membranes were incubated in 1 ml of binding buffer containing protease inhibitors with ‘251-glucagon (2-5 x lo6 cpm) for 30 min at 4°C. Membranes were washed 3 times by centrifugation in Eppendorf centrifuge, resuspended in 0.3 ml of 50 mM phosphate buffer, pH 7.4, and subjected to UV irradiation for 20 min. Membranes were collected by centrifugation and separated by SDS-PAGE according to Laemmli (1970) and analyzed by autoradiography. 4. The treatments of adipose and liver membranes with N-glycanase (Genzyme, MA) and Staphylococcus aureus V8 protease (Sigma) were carried out as described by Iwanij and Vincent (1990). 5. Immunobloting procedure was carried out as described previously (Iwanij and Vincent, 19901, using procedure of Towbin et al. (1979). Briefly, membranes 5-20 pg were separated using mini gel apparatus (Hoeffer Scientific) and transferred electrophoretically onto the nitrocellulose. Blots were pretreated with 2% gelatin containing blocking solution, incubated with primary antibodies, extensively washed and then probed with secondary antibodies conjugated to the alkaline phosphatase (Promega, anti-mouse IgGI. 6. Miscellaneous procedures: Iodination of glucagon was described by Lin et al. (19751, and separation of iodinated glucagon from unincorporated iodine was described by Iwanij and Hur (1985). The protein concen-
101 (1994) 257-261
tration was determined by the fluorescamine dure of Udenfriend et al. (1972).
proce-
3. Results
The goal of our investigation was to identify and characterize the glucagon receptor present in the adipose tissue.’ To this end, we have used direct UV irradiation as a means of cross-linking glucagon to the receptor sites. Adipocyte ghosts isolated from enzymatically dissociated adipocytes (Rodbell, 1972) or plasma membranes prepared directly from epididymal fat (EF) pads were incubated with ‘251-glucagon, washed and irradiated with UV light as described above. Membranes were then analyzed by SDS-PAGE and autoradiography. A single radioactive band was observed in the EF plasma membranes (Fig. 1A) and lysed adipocyte ghosts (results not shown). Labeling of this band was abolished in the presence of an excess of unlabeled glucagon indicating that the band represents the glucagon receptor. The labeled adipocyte receptor
A
A
B
C
Fig. 1. UV-induced cross-linking of 1251-glucagonto adipocyte plasma membrane isolated from rat epididymal and brown fat. Membrane preparation and the cross-linking was carried out as described in the Materials and methods section. Panel A: lane A, liver plasma membranes (100 pg); lane B, epididymal fat plasma membranes (200 &; lane C, epididymal fat plasma membranes (200 fig) for which binding of ‘=I-glucagon was carried out in the presence of 2 mM unlabeled glucagon. Panel B: lane A, liver plasma membranes (50 Kg); lane B, brown fat plasma membranes GXJO&; lane C, brown fat plasma membranes (200 pg) for which binding of %glucagon was carried out in the presence of 2 PM unlabeled glucagon. Longer time exposure was needed to visualize BAT receptor as compared with liver tissue receptor. Arrowhead points to the labeled glucagon receptor band.
259
K Iwanij et al. /Molecular and Cellular Endocrinology 101 (1994) 257-261
band was significantly less intense than the labeled hepatic band, that observation being consistent with the lesser density of adipocyte receptor sites as previously reported (Desbuquois and Laudat, 1974). The adipocyte glucagon receptor co-migrated with the hepatic glucagon receptor on the SDS polyacrylamide gels indicating that they represent similar or identical proteins. Having established that our cross-linking procedure is sensitive enough to detect the lower concentrations of the adipose glucagon receptor, we investigated whether brown adipose tissue (BAT) also contains the glucagon receptor. We isolated plasma membranes from brown fat and were able to detect reproducible and specific binding of ‘251-glucagon. When we performed glucagon binding and UV cross-linking experiments with membranes isolated from BAT (Fig. lB), the results were identical to those obtained with the EF membranes. To further characterize the adipocyte glucagon receptor, we have carried out immunoblot experiments using monoclonal antibodies prepared against the hepatic receptor (Iwanij and Vincent, 1990). As shown in Fig. 2, Mab 6ElO recognized a single band in both EF and BAT plasma membranes. The molecular weight of the stained protein species from fat was the same as for the hepatic glucagon receptor. In accord with the affinity labeling experiments, the intensity of staining of t,he adipose receptor band was significantly lower than that in the liver preparations. Previous analysis of the hepatic glucagon receptor indicated that the receptor contains four N-linked
Fig. 2. Immunoblot analysis of membranes isolated from rat epididyma1 and brown fat. Membrane preparation and immunoblotting was carried out as described in the Materials and methods section. Immunoblot was probed with monoclonal antibody 6E10; lanes L, liver plasma membranes (10 pg); lane EF, epididymal fat plasma membranes (20 pg); lane BF, brown fat plasma membranes (20 pg). Arrowhead points to the immunoreactive band corresponding to the glucagon receptor.
A
B
C
D
Fig. 3. Immunoblot analysis of the deglycosylated adipocyte glucagon receptor. EF and liver plasma membranes were treated with Nglycanase, separated on the same SDS gel, transferred on the same sheet of nitrocellulose, and probed with 6ElO monoclonal antibody. Lane A, N-glycanase treated adipocyte plasma membranes (30 pg); lane B, untreated adipocyte plasma membranes (20 pg); lane C, N-glycanase treated liver plasma membranes (20 pg); lane D, untreated liver plasma membranes (10 cl.g).
oligosaccharides branches (Iyengar and Herberg, 1984; Iwanij and Hur, 1985). Therefore, we have investigated whether the adipose glucagon receptor is also N-glycosylated. Adipose and liver plasma membranes were treated with N-glycanase, an enzyme that cleaves Nlinked oligosaccharide residues as outlined in the Materials and methods section. After the treatment, membranes were separated by SDS-PAGE, transferred onto nitrocellulose and probed with Mab 6ElO. As shown in Fig. 3, treatment of the EF membranes with Nglycanase resulted in the loss of the 62 kDa immunoreactive band. That band was replaced by two closely spaced doublets. Intermediate bands of identical size were observed also in liver plasma membrane preparations. This result indicated that the adipose glucagon receptor contains multiple N-linked oligosaccharides similar to those of the hepatic receptor. In our previous work (Iwanij and Vincent, 19901, we have noted that digestion of affinity labeled hepatic glucagon receptor with V8 protease results in a characteristic pattern of three sharply defined radioactive bands. We, therefore, carried out treatment of the affinity labeled EF membranes with V8 protease. The autoradiogram shown in Fig. 4 indicates that the V8 protease digest of adipose receptor resulted in the
260
V: Iwunij et al. /Molec&r
AB
and Cellular Endocrinology
CD
Pig. 4. Proteolytic mapping of the adipocyte glucagon receptor with S. aureus V8 protease. Affinity labeled adipocyte and liver membranes were digested with V8 protease, separated by the same SDS gel and analyzed by autoradiography. Adipocyte membranes lanes required longer exposure time to visualize the digestion fragments. .Lane A, liver plasma,~embrane untreated control (100 pg); lane 8, Iiver plasma membranes (100 pgI digested with V8 protease; lane C, adipocyte plasma membranes (200 ~g) digested with V8 protease; lane D, adipocyte plasma membrane (150 kg) untreated control. Open circles denote position of the receptor fragments.
generation of three fragments: one 59-60 kDa band, and a doublet of about 25-30 kDa. These fragments co-migrated on the SDS gels with the proteolytic bands generated from the hepatic receptor (Fig. 4). These results indicate that both hepatic and adipose receptors are cleaved into identical sets of proteolytic fragments.
4. Discussion Adipose tissue is an important target tissue for the action of the pancreatic peptide glucagon. Glucagon has been shown to increase plasma concentration of free fatty acids in vivo and to stimulate release of glycerol and free fatty acids from intact adipose tissue and dissociated adipocytes (Lefebvre, 1972). Interestingly, the adipocyte responsiveness to giucagon varies with the physiological state of the cell; it was shown to decline during adipocyte aging (Livingston et al., 1974) and to increase in experimentally induced diabetes (Chatzipanteli and Saggerson, 1983). The effect of glucagon on adipose tissue metabolism .has been extensively studied using dissociated adipocyte cells. The characteristic lipolytic response following glucagon binding to its receptor is believed to be mediated by an increase in CAMP levels. Some investigators have noted a lower sensitivity of adipose tissue as
lOI(1994)
257-261
compared to liver, and have suggested the presence of adipose tissue specific receptor subtypes (reviewed in Lefebvre, 1972). However, detailed experiments carried out by Desbuquois and Laudat 0974) that included determination of the binding constants, effects of nucleotides on hormone binding, sensitivity to proteolytic enzymes, and phospholipase A digestion all indicated that the properties of the adipose glucagon binding sites closely resemble those of the liver, and suggested that the observed differences arose from a variations in the adipocyte dissociation procedures. Our investigation of the structural properties of the adipose glucagon receptor supports the conciusions of Desbuquois and Laudat (1974). In order to identify and characterize the adipose tissue receptor, we have applied a direct UV irradiation affinity cross linking approach. This investigation resulted in the identification of the glucagon-receptor complex of 62 kDa estimated MW present both in epididymal and brown fat. The size of the affinity labeled rat adipose glucagon receptor is identical to the previously characterized rat liver receptor. Immunobloting experiment with antiglucagon receptor antibodies also identified the presence of the major immunoreactive band in the membranes isolated from epididymal and brown fat, identical in size to the band observed in liver plasma membranes. The intensity of the radiolabeled and immunoreactive band detected in the adipose membranes is significantly lower than that one present in liver indicating lower levels of the glucagon receptor in‘ fat tissue as previously reported by Desbuquois and Laudat (1974). Treatment of the adipose plasma membranes with N-glycanase, an enzyme that cleaves N-linked oligosaccharides, resulted in the appearance of several intermediate species indicating that the adipose glucagon receptor is a glycoprotein with a level of glycosylation similar to that of the hepatic receptor. We have carried out structural analysis of the affinity labeled EF glucagon receptor by proteolytic treatment with S. aureus V8 protease. Autoradiographic analysis showed the presence of three distinct radioactive fragments identical in size to ones obtained with the hepatic receptor. Overall structural characterization described in this report indicates that glucagon receptor present in rat epidid~al and brown adipose tissues shows a striking similarity to the hepatic receptor and most probably represent identical structures.
5. Acknowledgements This work was supported in part by funds from the American Diabetes Association, Minnesota Chapter grant, the American Cancer Society grant BC-688, and University of Minnesota institutional ACS grant IN-13-
K Iwanij et al. /Molecular and Cellular Endocrinology I01 (1994) 257-261
29-2 (to V.I.). We thank Dr. Norman Kagan for critical reading of this manuscript.
6. References Birnbaumer, L. and Rodbell, M. (1969) J. Biol. Chem. 244, 34773482. Chatzipanteh, K. and Saggerson, D. (1983) FEBS Lett. 155, 135-138. Cuatrecasas, P. (1971) J. Biol. Chem. 246, 6532-6542. Demohou-Mason, C. and Epand (1982) Biochemistry 21, 1996-2004. Desbuquois, B., Laudat, M-H. and Laudat, P. (1973) Biochem. Biophys. Res. Commun. 53, 1187-1194. Desbuquois, B. and Laudat, M-H. (1974) Mol. Cell. Endocrinol. 1, 355-370. Iwanij, V. and Hur, K.C. (1985) Proc. Natl. Acad. Sci. USA 82, 325-329. Iwanij, V. and Vincent, A.C. (1990) J. Biol. Chem. 265,21302-21308.
261
Iyengar, R. and Herberg, J.T. (1984) J. Biol. Chem. 259, 5222-5229. Johnson, G.L., MacAndrew, V.I. and Pilch, P.F. 0981) Proc. Natl. Acad. Sci. USA 78, 875-878. Laemmli, U.K. (1970) Nature 227, 680-685. Lefebvre, P.J. (1972) in Glucagon (Lefebvre, P.J. and Unger, R.H., eds.), pp. 109-121, Pergamon Press, Oxford. Lin, M.C., Wright, D.E., Hruby, V.J. and Rodbell, M. (1975) Biochemistry, 14, 1559-1563. Livingston, J.N., Cuatrecasas, P. and Lockwood, D.H. (1974) J. Lipid Res. 15, 26-32. Neville, D.M. (1968) Biochim. Biophys. Acta 154, 540-552. Rodbell, M., Krans, H.M.J., Pohl, S.L. and Birnbaumer, L. (1971) J. Biol. Chem. 246, 1861-1871. Rodbell, M. (1972) in Cyclic Nucleotide Research (Chasin, ed.), pp. 101-125, Marcel Dekker, New York. Selby, P.L. and Sherratt, H.S.A. (1989) Trends Pharmacol. Sci. 10, 495-500. Towbin, H., Staehlin, T. and Gordon, J. (1979) Proc. Natl. Acad. Sci. USA 76, 4350-4354. Udenfriend, S., Stein, S., Bohlen, P., Dairman, W., Leimgruber, W., and Weigeli, M. (1972) Science 178, 871-872.