Expression, Purification, and Reconstitution of Receptor for Pituitary Adenylate Cyclase-activating Polypeptide
1998; Elsevier BV; Volume: 273; Issue: 25 Linguagem: Inglês
10.1074/jbc.273.25.15464
ISSN1083-351X
AutoresTetsuya Ohtaki, Kazuhiro Ogi, Yasushi Masuda, Kaoru Mitsuoka, Yoshinori Fujiyoshi, Chieko Kitada, Hidekazu Sawada, Haruo Onda, Masahiko Fujino,
Tópico(s)Hypothalamic control of reproductive hormones
ResumoHuman pituitary adenylate cyclase-activating polypeptide (PACAP) receptor was expressed in Sf9 insect cells and Chinese hamster ovary (CHO) cells. The recombinant receptor in Sf9 cell membranes had low affinity for125I-PACAP27 (K d = 155.3 pm) and was insensitive to guanosine 5′-O-3-thiotriphosphate (GTPγS), whereas the receptor in CHO membranes had a high affinity (K d = 44.4 pm) and was GTPγS sensitive. The receptor in Sf9 membranes was converted to a high affinity state (K d = 20–40 pm) following solubilization with digitonin. A large quantity (2 mg from 8 liters of insect cells) of the purified PACAP receptors (B max = 23.9 nmol/mg of protein) were obtained in a digitonin-induced high affinity state (K d = 17.3 pm) using biotinylated ligand affinity chromatography. The apparent molecular weight of the purified receptor (M r = 48,000) was smaller than that of the receptor from CHO cells (M r = 58,000) due to differences in asparagine-linked sugar chains. The purified receptor reverted to a low affinity state (K d = 182.6 pm) upon reconstitution into lipid vesicles, however, the receptor reconstituted with Gs protein had a high affinity (K d = 40.2 pm) and was GTPγS sensitive. [35S]GTPγS binding to the reconstituted Gs protein was enhanced by PACAP27 and PACAP38 (EC50 = 42.5 and 9.4 pm, respectively) but not by antagonist PACAP(6–38), indicating that the purified receptor was functionally active. Human pituitary adenylate cyclase-activating polypeptide (PACAP) receptor was expressed in Sf9 insect cells and Chinese hamster ovary (CHO) cells. The recombinant receptor in Sf9 cell membranes had low affinity for125I-PACAP27 (K d = 155.3 pm) and was insensitive to guanosine 5′-O-3-thiotriphosphate (GTPγS), whereas the receptor in CHO membranes had a high affinity (K d = 44.4 pm) and was GTPγS sensitive. The receptor in Sf9 membranes was converted to a high affinity state (K d = 20–40 pm) following solubilization with digitonin. A large quantity (2 mg from 8 liters of insect cells) of the purified PACAP receptors (B max = 23.9 nmol/mg of protein) were obtained in a digitonin-induced high affinity state (K d = 17.3 pm) using biotinylated ligand affinity chromatography. The apparent molecular weight of the purified receptor (M r = 48,000) was smaller than that of the receptor from CHO cells (M r = 58,000) due to differences in asparagine-linked sugar chains. The purified receptor reverted to a low affinity state (K d = 182.6 pm) upon reconstitution into lipid vesicles, however, the receptor reconstituted with Gs protein had a high affinity (K d = 40.2 pm) and was GTPγS sensitive. [35S]GTPγS binding to the reconstituted Gs protein was enhanced by PACAP27 and PACAP38 (EC50 = 42.5 and 9.4 pm, respectively) but not by antagonist PACAP(6–38), indicating that the purified receptor was functionally active. Pituitary adenylate cyclase-activating polypeptide (PACAP) 1The abbreviations used are: PACAP, pituitary adenylate cyclase-activating polypeptide; VIP, vasoactive intestinal polypeptide; GTPγS, guanosine 5′-O-3-thiotriphosphate; BSA, bovine serum albumin; BIGCHAP,N,N-bis(3-d-gluconamidopropyl)cholamide; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; biotin-(AC5)2-OSu, 5-[5-(N-succinimidyloxycarbonyl)penthylamido]hexyl-d-biotinamide; CHO, Chinese hamster ovary; PAGE, polyacrylamide gel electrophoresis. 1The abbreviations used are: PACAP, pituitary adenylate cyclase-activating polypeptide; VIP, vasoactive intestinal polypeptide; GTPγS, guanosine 5′-O-3-thiotriphosphate; BSA, bovine serum albumin; BIGCHAP,N,N-bis(3-d-gluconamidopropyl)cholamide; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate; biotin-(AC5)2-OSu, 5-[5-(N-succinimidyloxycarbonyl)penthylamido]hexyl-d-biotinamide; CHO, Chinese hamster ovary; PAGE, polyacrylamide gel electrophoresis. was first discovered in 1989 as a novel hypothalamic hormone that increases adenylate cyclase activity in pituitary cells (1Miyata A. Arimura A. Dahl R.R. Minamino N. Uehara A. Jiang L. Culler M.D. Coy D.H. Biochem. Biophys. Res. Commun. 1989; 164: 567-574Crossref PubMed Scopus (1664) Google Scholar). PACAP exists in two carboxyl-terminal-amidated forms: PACAP38 with 38 amino acid residues (1Miyata A. Arimura A. Dahl R.R. Minamino N. Uehara A. Jiang L. Culler M.D. Coy D.H. Biochem. Biophys. Res. Commun. 1989; 164: 567-574Crossref PubMed Scopus (1664) Google Scholar) and PACAP27 with the same amino-terminal 27 residues (2Miyata A. Jiang L. Dahl R.R. Kitada C. Kubo K. Fujino M. Minamino N. Arimura A. Biochem. Biophys. Res. Commun. 1990; 170: 643-648Crossref PubMed Scopus (855) Google Scholar). Molecular cloning studies revealed that the structure of PACAP is highly conserved among rat, sheep, and humans (3Kimura C. Ohkubo S. Ogi K. Hosoya M. Itoh Y. Onda H. Miyata A. Jiang L. Dahl R.R. Stibbs H.H. Arimura A. Fujino M. Biochem. Biophys. Res. Commun. 1990; 166: 81-89Crossref PubMed Scopus (293) Google Scholar, 4Ogi K. Kimura C. Onda H. Arimura A. Fujino M. Biochem. Biophys. Res. Commun. 1990; 173: 1271-1279Crossref PubMed Scopus (229) Google Scholar). PACAP has structural similarity to peptides in the secretin/glucagon peptide family, especially to vasoactive intestinal polypeptide (VIP) (1Miyata A. Arimura A. Dahl R.R. Minamino N. Uehara A. Jiang L. Culler M.D. Coy D.H. Biochem. Biophys. Res. Commun. 1989; 164: 567-574Crossref PubMed Scopus (1664) Google Scholar). PACAP is distributed in the central nervous system and various peripheral organs (5Arimura A. Somogyvári-Vigh A. Miyata A. Mizuno K. Coy D.H. Kitada C. Endocrinology. 1991; 129: 2787-2789Crossref PubMed Scopus (564) Google Scholar) and elicits a wide variety of biological functions, such as neuroprotective action against gp120-induced cell death (6Arimura A. Somogyvari-Vigh A. Weill C. Fiore R.C. Tatsuno I. Bay V. Brenneman D.E. Ann. N. Y. Acad. Sci. 1994; 739: 228-243Crossref PubMed Scopus (177) Google Scholar), protection of cerebellar granule neurons from apoptosis (7Cavallaro S. Copani A. D'Agata V. Musco S. Petralia S. Ventra C. Stivala F. Travali S. Canonico P.L. Mol. Pharmacol. 1996; 50: 60-66PubMed Google Scholar), secretion of pituitary hormones (1Miyata A. Arimura A. Dahl R.R. Minamino N. Uehara A. Jiang L. Culler M.D. Coy D.H. Biochem. Biophys. Res. Commun. 1989; 164: 567-574Crossref PubMed Scopus (1664) Google Scholar, 8Culler M.D. Paschall C.S. Endocrinology. 1991; 129: 2260-2262Crossref PubMed Scopus (171) Google Scholar), secretion of interleukin-6 from astrocytes or folliculo-stellate cells (9Tatsuno I. Somogyvari-Vigh A. Mizuno K. Gottschall P.E. Hidaka H. Arimura A. Endocrinology. 1991; 129: 1797-1804Crossref PubMed Scopus (155) Google Scholar, 10Matsumoto H. Koyama C. Sawada T. Koike K. Hirota K. Miyake A. Arimura A. Inoue K. Endocrinology. 1993; 133: 2150-2155Crossref PubMed Scopus (73) Google Scholar), secretion of catecholamines from chromaffin cells (11Watanabe T. Masuo Y. Matsumoto H. Suzuki N. Ohtaki T. Masuda Y. Kitada C. Tsuda M. Fujino M. Biochem. Biophys. Res. Commun. 1992; 182: 403-411Crossref PubMed Scopus (163) Google Scholar) or adrenal glands (12Watanabe T. Shimamoto N. Takahashi A. Fujino M. Am. J. Physiol. 1995; 269: E903-E909Crossref PubMed Google Scholar), and insulin release (13Yada T. Sakurada M. Ihida K. Nakata M. Murata F. Arimura A. Kikuchi M. J. Biol. Chem. 1994; 269: 1290-1293Abstract Full Text PDF PubMed Google Scholar). The biological actions of PACAP are mediated by a PACAP-specific receptor (type I receptor) and a PACAP/VIP-nonselective receptor (type II receptor). The type I PACAP receptor includes the PACAP1receptor (14Pisegna J.R. Wank S.A. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 6345-6349Crossref PubMed Scopus (371) Google Scholar, 15Hashimoto H. Ishihara T. Shigemoto R. Mori K. Nagata S. Neuron. 1993; 11: 333-342Abstract Full Text PDF PubMed Scopus (247) Google Scholar, 16Hosoya M. Onda H. Ogi K. Masuda Y. Miyamoto Y. Ohtaki T. Okazaki H. Arimura A. Fujino M. Biochem. Biophys. Res. Commun. 1993; 194: 133-143Crossref PubMed Scopus (151) Google Scholar, 17Morrow J.A. Lutz E.M. West K.M. Fink G. Harmar A.J. FEBS Lett. 1993; 329: 99-105Crossref PubMed Scopus (120) Google Scholar, 18Svoboda M. Tastenoy M. Ciccarelli E. Stiévenart M. Christophe J. Biochem. Biophys. Res. Commun. 1993; 195: 881-888Crossref PubMed Scopus (87) Google Scholar, 19Ogi K. Miyamoto Y. Masuda Y. Habata Y. Hosoya M. Ohtaki T. Masuo Y. Onda H. Fujino M. Biochem. Biophys. Res. Commun. 1993; 196: 1511-1521Crossref PubMed Scopus (109) Google Scholar, 20Spengler D. Waeber C. Pantaloni C. Holsboer F. Bockaert J. Seeburg P.H. Journot L. Nature. 1993; 365: 170-175Crossref PubMed Scopus (1115) Google Scholar, 21Pisegna J.R. Wank S.A. J. Biol. Chem. 1996; 271: 17267-17274Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar) and a novel variant PACAPR-TM4 (22Chatterjee T.K. Sharma R.V. Fisher R.A. J. Biol. Chem. 1996; 271: 32226-32232Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar). There are two alternatively spliced exons, rat hip and hop (20Spengler D. Waeber C. Pantaloni C. Holsboer F. Bockaert J. Seeburg P.H. Journot L. Nature. 1993; 365: 170-175Crossref PubMed Scopus (1115) Google Scholar) or human SV-1 and SV-2 (21Pisegna J.R. Wank S.A. J. Biol. Chem. 1996; 271: 17267-17274Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar), in the PACAP1 receptor gene, resulting in the possible existence of five splicing variants in the PACAP1receptor (20Spengler D. Waeber C. Pantaloni C. Holsboer F. Bockaert J. Seeburg P.H. Journot L. Nature. 1993; 365: 170-175Crossref PubMed Scopus (1115) Google Scholar, 21Pisegna J.R. Wank S.A. J. Biol. Chem. 1996; 271: 17267-17274Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar). All of these receptors belong to the G protein-coupled receptor superfamily and are subdivided structurally into the secretin/glucagon receptor family (23Laburthe M. Couvineau A. Gaudin P. Maoret J.-J. Rouyer-Fessard C. Nicole P. Ann. N. Y. Acad. Sci. 1996; 805: 94-109Crossref PubMed Scopus (123) Google Scholar) that is distinguished from rhodopsin-type receptors.All G protein-coupled receptors have seven hydrophobic segments that probably form transmembrane α-helices. Direct evidence for the arrangement of transmembrane domains was obtained from the two-dimensional crystallography of rhodopsin (24Unger V.M. Hargrave P.A. Baldwin J.M. Schertler G.F.X. Nature. 1997; 389: 203-206Crossref PubMed Scopus (480) Google Scholar), providing valuable information for molecular modeling of other G protein-coupled receptors. More precise modeling requires elucidating the structures of another receptors. In particular, receptors in the secretin/glucagon receptor family are predicted to have a different arrangement in the transmembrane domains (25Donnelly D. FEBS Lett. 1997; 409: 431-436Crossref PubMed Scopus (69) Google Scholar). On the other hand, structural biology directly clarifying the three-dimensional structure of a G protein-coupled receptor has been hindered by several difficulties in the purification and crystallization of the receptor protein. Most G protein-coupled receptors exist at very low level in tissue membranes. Thus, it is essential to develop an expression system that can produce a large amount of the recombinant receptor. Parker et al.(26Parker E.M. Kameyama K. Higashijima T. Ross E.M. J. Biol. Chem. 1991; 266: 519-527Abstract Full Text PDF PubMed Google Scholar) first described that a baculovirus expression system was beneficial for the expression of β-adrenergic and muscarinic receptors at high levels (5–30 pmol/mg). Recombinant β-adrenergic receptors purified from the baculovirus-infected insect cells were functionally active as were the β-adrenergic receptors from turkey erythrocytes (26Parker E.M. Kameyama K. Higashijima T. Ross E.M. J. Biol. Chem. 1991; 266: 519-527Abstract Full Text PDF PubMed Google Scholar). The expression system was also used to produce various G protein-coupled receptors, however, only a few reports (27Doi T. Hiroaki Y. Arimoto I. Fujiyoshi Y. Okamoto T. Satoh M. Furuichi Y. Eur. J. Biochem. 1997; 248: 139-148Crossref PubMed Scopus (39) Google Scholar) succeeded in providing a practical amount of purified receptor for further biochemical or structural studies. This is probably due to difficulty in the solubilization and purification of G protein-coupled receptors.We previously described successful purification of the PACAP1 receptor from bovine brain membranes in a high affinity state (28Ohtaki T. Masuda Y. Ishibashi Y. Kitada C. Arimura A. Fujino M. J. Biol. Chem. 1993; 268: 26650-26657Abstract Full Text PDF PubMed Google Scholar). In the present study we conducted large-scale purification of the recombinant PACAP1 receptor by combining the previously described purification procedures and the baculovirus expression system. The recombinant PACAP1receptor purified in a digitonin-solubilized form retained high affinity for PACAP and was functionally active when reconstituted with Gs protein in lipid vesicles. The purified receptor will likely contribute to the understanding of the regulatory mechanisms and structure of G protein-coupled receptors.DISCUSSIONIn the present study, the human PACAP receptor was overexpressed in Sf9 insect cells under the control of a strong polyhedrin promoter. The expression level (50–150 pmol/mg) was higher than those of other G protein-coupled receptors so far reported (ranging from 0.5 to 100 pmol/mg) (26Parker E.M. Kameyama K. Higashijima T. Ross E.M. J. Biol. Chem. 1991; 266: 519-527Abstract Full Text PDF PubMed Google Scholar, 33Masuda Y. Sugo T. Kikuchi T. Kawata A. Satoh M. Fujisawa Y. Itoh Y. Wakimasu M. Ohtaki T. J. Pharmacol. Exp. Ther. 1996; 279: 675-685PubMed Google Scholar, 38Butkerait P. Zheng Y. Hallak H. Graham T.E. Miller H.A. Burris K.D. Molinoff P.B. Manning D.R. J. Biol. Chem. 1995; 270: 18691-18699Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar, 39Mills A. Allet B. Bernard A. Chabert C. Brandt E. Cavegn C. Chollet A. Kawashima E. FEBS Lett. 1993; 320: 130-134Crossref PubMed Scopus (49) Google Scholar, 40Ng G.Y.K. George S.R. Zastawny R.L. Caron M. Bouvier M. Dennis M. O'Dowd B.F. Biochemistry. 1993; 32: 11727-11733Crossref PubMed Scopus (137) Google Scholar, 41Mulheron J.G. Casañas S.J. Arthur J.M. Garnovskaya M.N. Gettys T.W. Raymond J.R. J. Biol. Chem. 1994; 269: 12954-12962Abstract Full Text PDF PubMed Google Scholar, 42Quehenberger O. Prossnitz E.R. Cochrane C.G. Ye R.D. J. Biol. Chem. 1992; 267: 19757-19760Abstract Full Text PDF PubMed Google Scholar). The presence of a signal sequence and many potential glycosylation sites in the PACAP receptor, favoring the translocation of the receptor to cell membranes, presumably contributes to the high expression level. Furthermore, trimming of the 5′- and 3′-noncoding regions along with the use of pAKKO vector having SRα promoter might increase the expression level of the receptor in CHO/dhfr− cells compared with the previous expression study using CHO-K1 cells and pRc/CMV expression vector (19Ogi K. Miyamoto Y. Masuda Y. Habata Y. Hosoya M. Ohtaki T. Masuo Y. Onda H. Fujino M. Biochem. Biophys. Res. Commun. 1993; 196: 1511-1521Crossref PubMed Scopus (109) Google Scholar). Butkerait et al. (38Butkerait P. Zheng Y. Hallak H. Graham T.E. Miller H.A. Burris K.D. Molinoff P.B. Manning D.R. J. Biol. Chem. 1995; 270: 18691-18699Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar) suggested that the shortening of the 5′-untranslated sequence to 11 authentic bases might be the reason for their high expression level of the 5-HT1A receptors in insect cells (5–34 pmol/mg) compared with the values reported by others (0.15 and 3 pmol/mg).The native PACAP receptor is believed to be coupled primarily to Gs proteins for activating adenylate cyclase. Additional coupling to another G protein species is suggested by the pleiotropic cellular response to PACAP stimulation such as increases in phosphoinositide turnover (20Spengler D. Waeber C. Pantaloni C. Holsboer F. Bockaert J. Seeburg P.H. Journot L. Nature. 1993; 365: 170-175Crossref PubMed Scopus (1115) Google Scholar, 21Pisegna J.R. Wank S.A. J. Biol. Chem. 1996; 271: 17267-17274Abstract Full Text Full Text PDF PubMed Scopus (168) Google Scholar, 43Miyamoto Y. Habata Y. Ohtaki T. Masuda Y. Ogi K. Onda H. Fujino M. Biochim. Biophys. Acta. 1994; 1218: 297-307Crossref PubMed Scopus (40) Google Scholar) and intracellular calcium ion concentration (11Watanabe T. Masuo Y. Matsumoto H. Suzuki N. Ohtaki T. Masuda Y. Kitada C. Tsuda M. Fujino M. Biochem. Biophys. Res. Commun. 1992; 182: 403-411Crossref PubMed Scopus (163) Google Scholar, 44Delporte C. Poloczek P. de Neef P. Vertongen P. Ciccarelli E. Svoboda M. Herchuelz A. Winand J. Robberecht P. Mol. Cell. Endocrinol. 1995; 107: 71-76Crossref PubMed Scopus (35) Google Scholar). The recombinant receptor in the Sf9 cell membranes, however, was not coupled to endogenous G proteins as evidenced by its low affinity and GTPγS insensitivity in agonist binding. This is most probably due to a deficiency in the appropriate G proteins. The concentrations of immunoreactive Gi2 and Go proteins in High 5 insect cells were estimated as 10.7 and 0.5 pmol/mg, respectively, while those in Sf9 insect cells were below the detection limit (45Wehmeyer A. Schulz R. J. Neurochem. 1997; 68: 1361-1371Crossref PubMed Scopus (12) Google Scholar). This finding strongly suggests that the expression level of Gs-like protein in Sf9 insect cells is also very low. Butkerait et al. (38Butkerait P. Zheng Y. Hallak H. Graham T.E. Miller H.A. Burris K.D. Molinoff P.B. Manning D.R. J. Biol. Chem. 1995; 270: 18691-18699Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar) suggested that the coupling of recombinant receptors to endogenous G proteins in insect cells is related to the receptor expression level. Receptors expressed at a low level, such as the D4 dopamine receptor (5 pmol/mg) (39Mills A. Allet B. Bernard A. Chabert C. Brandt E. Cavegn C. Chollet A. Kawashima E. FEBS Lett. 1993; 320: 130-134Crossref PubMed Scopus (49) Google Scholar), serotonin 5-HT1B receptor (0.5 pmol/mg) (40Ng G.Y.K. George S.R. Zastawny R.L. Caron M. Bouvier M. Dennis M. O'Dowd B.F. Biochemistry. 1993; 32: 11727-11733Crossref PubMed Scopus (137) Google Scholar), and 5-HT1A receptor (0.15 pmol/mg) (41Mulheron J.G. Casañas S.J. Arthur J.M. Garnovskaya M.N. Gettys T.W. Raymond J.R. J. Biol. Chem. 1994; 269: 12954-12962Abstract Full Text PDF PubMed Google Scholar), were sensitive to guanine nucleotides in agonist binding, indicating coupling to G protein. On the other hand, the 5HT1Areceptor (5–34 pmol/mg) (38Butkerait P. Zheng Y. Hallak H. Graham T.E. Miller H.A. Burris K.D. Molinoff P.B. Manning D.R. J. Biol. Chem. 1995; 270: 18691-18699Abstract Full Text Full Text PDF PubMed Scopus (146) Google Scholar), the formyl peptide receptor (27 pmol/mg) (42Quehenberger O. Prossnitz E.R. Cochrane C.G. Ye R.D. J. Biol. Chem. 1992; 267: 19757-19760Abstract Full Text PDF PubMed Google Scholar), the β-adrenergic receptor and the muscarinic receptor (30 pmol/mg) (26Parker E.M. Kameyama K. Higashijima T. Ross E.M. J. Biol. Chem. 1991; 266: 519-527Abstract Full Text PDF PubMed Google Scholar), did not exhibit GTPγS sensitivity in agonist binding. Therefore, it is not surprising that most of the PACAP receptors overexpressed in insect cells were not coupled to endogenous G proteins.The recombinant PACAP receptor in the Sf9 cell membranes was purified in a digitonin-solubilized form. The purified receptor had a protein core similar to that purified from CHO cells but with differentN-linked sugar chains. The apparent molecular weight of theN-glycanase digested band (M r = 43,000) was smaller than the calculated molecular weight of the protein core (M r = 51,354) (19Ogi K. Miyamoto Y. Masuda Y. Habata Y. Hosoya M. Ohtaki T. Masuo Y. Onda H. Fujino M. Biochem. Biophys. Res. Commun. 1993; 196: 1511-1521Crossref PubMed Scopus (109) Google Scholar). A possible reason for this discrepancy is proteolytic degradation of the carboxyl terminus region. The carboxyl-terminal amino acid sequence could not, however, be determined in the present study. Another possibility is that strong hydrophobic interactions between membrane spanning α-helices restricted complete unfolding, resulting in a higher mobility than soluble proteins with a similar molecular weight.Glycosidase digestion studies suggested structural differences in theN-linked sugar chains of the PACAP receptors. The PACAP receptor from CHO cells was resistant to endoglycosidase H and endoglycosidase F2. Endoglycosidase H digests high mannose-type and hybrid-type sugar chains (46Trimble R.B. Tarentino A.L. J. Biol. Chem. 1991; 266: 1646-1651Abstract Full Text PDF PubMed Google Scholar). Endoglycosidase F2 cleaves biantennary complex-type sugar chains preferentially, but also cleaves high mannose-type sugar chains at a slower rate (46Trimble R.B. Tarentino A.L. J. Biol. Chem. 1991; 266: 1646-1651Abstract Full Text PDF PubMed Google Scholar). Taken together with the result from digestion by exoglycosidases, it is suggested that the receptor from CHO cells has sialylated tri- or tetraantennary complex-type sugar chains. In contrast, the PACAP receptor from Sf9 cells was digested by endoglycosidase F2, indicating that it has biantennary complex-type and/or high mannose-type sugar chains. The presence of biantennary complex-type sugar chains, however, is somewhat controversial, because exoglycosidase digestion studies indicate that the PACAP receptor from Sf9 cells has only mannosyl residues at non-reducing terminal but does not have NeuAc and GlcNAc residues. The presence of high mannose-type sugar chains is compatible with the result from mannosidase digestion but inconsistent with the resistance to endoglycosidase H. Considering that the PACAP receptor from Sf9 cells has truncated N-linked sugar chains (paucimannosidic N-linked sugar chains) (47Altman F. Trends Glycosci. Glycotechnol. 1996; 8: 101-114Crossref Scopus (18) Google Scholar) rather than high mannose-type sugar chains, this discrepancy could be explained by possible difference between substrate specificity of endoglycosidase H and endoglycosidase F2. Endoglycosidase H scarcely digests some kinds of paucimannosidicN-linked sugar chains such as Manα1→3(Manα1→6)Manβ1→4GlcNAc2 (46Trimble R.B. Tarentino A.L. J. Biol. Chem. 1991; 266: 1646-1651Abstract Full Text PDF PubMed Google Scholar), while the reactivity of endoglycosidase F2 on these sugar chains is not revealed. Direct evidences are required to clarify the structure of N-linked sugar chains in the PACAP receptors.The purified PACAP receptor presented several oligomer bands upon SDS-PAGE. Similar results were found in various G protein-coupled receptors such as rhodopsin (48Borjigin J. Nathans J. J. Biol. Chem. 1994; 269: 14715-14722Abstract Full Text PDF PubMed Google Scholar) and the olfactory receptor (49Nekrasova E. Sosinskaya A. Natochin M. Lancet D. Gat U. Eur. J. Biochem. 1996; 238: 28-37Crossref PubMed Scopus (40) Google Scholar). Pharmacological evidence also suggests that the m2-muscarinic receptor forms a dimeric structure (50Potter L.T. Ballesteros L.A. Bichajian L.H. Ferrendelli C.A. Fisher A. Hanchett H.E. Zhang R. Mol. Pharmacol. 1991; 39: 211-221PubMed Google Scholar). On the other hand, bacteriorhodpsin has been shown to form a trimeric structure in orthorhombic two-dimensional crystals (51Michel H. Oesterhelt D. Henderson R. Proc. Natl. Acad. Sci. U. S. A. 1980; 77: 338-342Crossref PubMed Scopus (93) Google Scholar). The physiological significance of receptor oligomerization observed in SDS-PAGE is still unclear, because high temperatures used during SDS-PAGE sample preparation might promote artificial oligomerization via hydrophobic interactions as described by Sagné et al. (52Sagné C. Isambert M.-F. Henry J.-P. Gasnier B. Biochem. J. 1996; 316: 825-831Crossref PubMed Scopus (72) Google Scholar). In fact, it was reported that the olfactory receptor forms higher oligomers after prolonged boiling (49Nekrasova E. Sosinskaya A. Natochin M. Lancet D. Gat U. Eur. J. Biochem. 1996; 238: 28-37Crossref PubMed Scopus (40) Google Scholar). Receptor oligomerization in physiological conditions should be further examined.The purified receptor had high affinity for PACAP27 and PACAP38 but low affinity for VIP. These ligand binding properties were similar to those of the receptor expressed in CHO cells. It has been proposed that digitonin might stabilize the PACAP receptor in the high affinity state based on the observations that purified PACAP receptor has a high affinity by itself (28Ohtaki T. Masuda Y. Ishibashi Y. Kitada C. Arimura A. Fujino M. J. Biol. Chem. 1993; 268: 26650-26657Abstract Full Text PDF PubMed Google Scholar). This hypothesis was further substantiated by reconstituting the purified PACAP receptor into lipid vesicles. The receptor reconstituted in lipid vesicles alone had low affinity in the absence of digitonin but high affinity in the presence of digitonin. The molecular mechanism of digitonin action, however, is not yet clear. For example, it is not clear whether solubilization into a lipid/digitonin mixed micelle is required for stabilizing the receptor or if the insertion of a small amount of digitonin into membrane bilayers is sufficient. Also, it is not clear whether digitonin acts directly on the receptor protein or acts indirectly by changing the milieu of the membrane or micelle. The small difference found betweenK d values for the purified receptor in a digitonin-solubilized form and the reconstituted receptor in the DG-BSA/TED buffer suggests that complete solubilization is required for the full effect of digitonin. Studies on possible conformational changes in the PACAP receptor induced by digitonin may aid in understanding the G protein-dependent regulation of ligand binding affinity.It should be also noted that the effect of digitonin is different from receptor to receptor. Some G protein-coupled receptors, such as β-adrenergic (53Cubero A. Malbon C.C. J. Biol. Chem. 1984; 259: 1344-1350Abstract Full Text PDF PubMed Google Scholar) and neuropeptide Y (54Sheikh S.P. Hansen A.P. Williams J.A. J. Biol. Chem. 1991; 266: 23959-23966Abstract Full Text PDF PubMed Google Scholar) receptors, were solubilized successfully using digitonin. The corticotropin-releasing factor receptor solubilized with digitonin had high affinity for its ligand and no GTPγS sensitivity (55Grigoriadis D.E. Zaczek R. Pearsall D.M. De Souza E.B. Endocrinology. 1989; 125: 3068-3077Crossref PubMed Scopus (11) Google Scholar), as observed in the present study. In contrast, the use of digitonin failed in the solubilization of gonadotropin-releasing hormone (56Hazum E. Schvartz I. Waksman Y. Keinan D. J. Biol. Chem. 1986; 261: 13043-13048Abstract Full Text PDF PubMed Google Scholar) and B2 bradykinin (57Fauβner A. Heinz-Erian P. Klier C. Roscher A.A. J. Biol. Chem. 1991; 266: 9442-9446PubMed Google Scholar) receptors. Receptors solubilized with digitonin in high affinity states are easier targets for receptor purification.The PACAP receptor reconstituted with Gsα/Gβγ in lipid vesicles had high affinity for PACAP27, in contrast to that without Gsα/Gβγ. GTPγS, known to inhibit receptor/G protein coupling by destabilizing the G protein trimer, almost completely neutralized the effect of Gsα/Gβγ, demonstrating that the purified PACAP receptor reconstituted in lipid vesicles requires Gsα/Gβγ for expressing high affinity for PACAP27. The B max value also decreased when the PACAP receptor was uncoupled from G protein. The result leads to the hypothesis that there are at least two states in G protein-uncoupled PACAP receptors, a low affinity state (K d = 100–200 pm) and a very low affinity state without detectable affinity for125I-PACAP27. Difference between theB max values in the presence and absence of digitonin represents G protein-uncoupled receptor in a very low affinity state. It should thus be interpreted that the CHO membranes contain G protein-uncoupled spare PACAP receptor as much as G protein-coupled PACAP receptor. The number of the PACAP receptors at the high affinity state (approximately 20–25 pmol/mg) in the CHO membranes reflects the maximum amount of G protein available for coupling to the PACAP receptor.GDP as well as GTPγS or GTP decreased the specific binding of125I-PACAP27 to the reconstituted receptor. The molecular mechanism of GDP action is thought to decrease the dissociation rate of GDP from the G protein α-subunit. Thus, this indicates that high affinity binding of PACAP27 requires the dissociation of GDP, whereas the dissociation of GDP is believed to be promoted by agonist stimulation. In other words, the PACAP receptor has high affinity for PACAP27 when it is interacting with nucleotide-free G protein. Therefore, PACAP binding to the receptor shifts the guanine nucleotide binding equilibrium toward the nucleotide free state, leading to the cooperative PACAP binding and GDP dissociation. A similar result with GDP has been observed in other G protein-coupled receptors such as the muscarinic receptor (58Haga K. Haga T. Ichiyama A. J. Biol. Chem. 1986; 261: 10133-10140Abstract Full Text PDF PubMed Google Scholar). The mathematical solution for the multiequilibrium system explains the observed effect of GDP on ligand binding (59Onaran H.O. Costa T. Rodbard D. Mol. Pharmacol. 1993; 43: 245-256PubMed Google Scholar).The expression of a homogeneous high affinity was attained with a smaller amount of G protein (PACAP receptor:Gsα:Gβγ = 1:4:8) compared with other reconstitution studies. Florio and Sternweis (60Florio V.A. Sternweis P.C. J. Biol. Chem. 1985; 260: 3477-3483Abstract Full Text PDF PubMed Google Scholar) described that approximately a 1000-fold excess of Gop
Referência(s)