Exp. Eye Res. (1990) 51, 317-323

Characterization

of Vasoactive

ANNE Departments

P.SWEDLUND”AND

of “Internal Medicine, Yale University School

(Received

Intestinal Retina

29 September

STEVEN

Peptide

Receptors

in

A. ROSENZWEIGbc*

bOphthalmology and Visual Science, and “Cell Biology. of Medicine, New Haven, CT 06510, U.S.A.

1989 and accepted

in revised

form 17 December

7989)

Vasoactive intestinal peptide (VIP) is a neuropeptide having a wide range of effects on a large number of tissues. To gain insight into the role VIP plays in retinal function, VIP receptors in bovine retinal membranes were analyzed in competition binding assays and by affinity labeling studies and compared to VIP receptors in rat liver membranes. In both membrane preparations, high affiity VIP binding sites (KD - 1 nM) were detected. Secretin and glucagon, each having close structural homology to VIP, were found to have negligible effects on [‘251]VIP binding in retina. In contrast, secretin (I&, = 70 nM) was modestly effective in inhibiting [‘251]VIP binding to rat liver membranes. Affinity labeling analysis revealed a VII? binding site of 59 kDa in both bovine retinal and rat liver membranes. Digestion of affinitylabeled receptor proteins with endoglycosidase F generated final cleavage products of approx. 45 kDa for both receptors. These results indicate that the retina expresses a high affinity, highly selective VIP receptor thereby supporting a specific function for VIP in this tissue. Key words: VIP: VIP Receptor; neuropeptide; gut Hormone; neuropeptide receptor: retina: neuronal receptor subtype: endoglycosidase F: affinity labeling. 1. Introduction Vasoactive intestinal peptide is a 28-residue polypeptide hormone synthesized by neurons in the central and peripheral nervous systems (Larsson et al., 19 76). Its physiological effects include stimulation of acinar cell secretion in pancreas and parotid, fluid and electrolyte secretion in the intestine, stimulation of glycogenolysis in neuronal and hepatic tissue, relaxation of intestinal, bronchial, cerebrovascular and splanchnic vascular smooth muscle and stimulation of anterior pituitary cell secretion (Said, 1984). More recently, VIP has been found to also have a role in immune modulation in the intestine (Stanisz, Befus and Bienenstock, 1986). VIP’s cellular effects have been shown to be mediated by CAMP via the stimulation of adenylate cyclase in its target cells (Fahrenkrug, 1979; Inoue et al., 1985). However, calcium-dependent changes have also been detected in response to VIP action in a number of tissues Magistretti and Schorderet. 1984 ; Scott and Baum, 1985). In the neural retina, VIP-like immunoreactivity has been detected in amacrine cells (Loren, Tornqvist and Alunets, 1980). In retinal homogenates and isolated horizontal cells, VIP has been shown to cause depolarization by a mechanism that does not correlate with its adenylate cyclase stimulating activity (Lasater, Watling and Dowling, 1983). While VIP’s ability to stimulate glycogenolysis in the retina is poor (Scott and Baum, 1985), VIP has been shown to be a potent * For correspondence at: Department of Ophthalmology Visual Science, Yale University School of Medicine, 330 Street, P.O. Box 3333, New Haven, CT 06510, U.S.A. 00144835/90/090317+07

$03.00/O

and Cedar

stimulator of glycogenolysis (Magistretti et al., 1981) and glucose utilization (McCulloch and Kelly, 198 3) in the cortex. In the central nervous system, VIP binding sites have been localized to the ends of tracts staining positively for VIP in the hippocampus, amygdala, geniculate nuclei and ventral hypothalamus (Besson et al., 1984). Biochemical information regarding neuronal VIP receptor structure is presently lacking. To understand better the apparent differences in VIP’s mechanism of action in neuronal and peripheral tissues the isolation and characterization of VIP receptors and their specific effector molecules is required. In this report we describe the binding characteristics and molecular weight of VIP receptors in bovine retina. This receptor was shown to have a high affinity and specificity for VIP, only weakly recognizing secretin. This was in contrast to rat hepatic VIP receptors, which interact with secretin with relatively high affinity. Affinity cross-linking experiments revealed the specific labeling of a 59-kDa protein in bovine retinal membranes and rat liver membranes. Endoglycosidase F digestion of these proteins resulted in final products of 45-kDa. These data indicate that VIP receptors in neuronal and peripheral target tissues exhibit a difference in ligand binding selectivity, but are structurally homologous.

2. Materials

and Methods

Radioiodination of VIP VIP was iodinated using the chloramine-T method described by Martin et al., (1986). One milliCurie of 0 1990 Academic Press Limited

A. P SWEDLUND

318

Na’““l (Amersham. Arlington Heights, II,) was reacted with 5 Alg of VIP (dissolved in 0.1 M acetic acid) and 14.2 nM chloramine-T (4 ~l,lof a 1 mg mll’ solution in 05 M NaPO, buffer, dissolved immediately before addition) for approx. 20 sec. The reaction was terminated by addition of sodium metabisullite at a concentration of 42.1 nM (4 pug of a 2 mg ml-l solution). Following acidification with 250 /tl of 0.1% trifluoroacetic acid, the entire mixture was applied to a C,, Sep-pak (Waters Assoc., Milford, MA). Free iodine was first removed by washing with 0.1 y0 trifluoroacetic acid and the iodinated peptide was then eluted with 60% acetonitrile in 0.1 y0 trifluoroacetic acid. A reduced volume of the eluate was purified by reverse-phase HPLC on a Vydac C, column (The Nest Group, Southboro, MA) using a Gilson gradient HPLC equipped with a Rheodyne injection valve and a l-ml loop. Radiolabeled peptides were eluted at a flow rate of 1 ml min-’ using a linear gradient of acetonitrile. Buffer A consisted of a 0.1 o/o trifluoroacetic acid solution and Buffer B of 60% acetonitrile-0.1 y0 trifluoroacetic acid. The column was equilibrated at 55 % Buffer A-45 % Buffer B (i.e. 27% acetonitrile). followed by gradient development over 30 min. beginning 1 min post sample injection to 300/, Buffer A-70% Buffer B (i.e. 42% acetonitrile). The major peak obtained was designated as Peak I (first major radioactive peak retention time t, = 34 min after unlabeled VIP t,, = 13 min and unlabeled. oxidized VIP f,, = 11.4 min). and represents ([‘“SI]iodo-Tyr’o, O-Met”)VIP as determined by Martin et al., (1986) and Marie et al., ( 19 8 5). Bovine serum albumin (12 5 pg) was added to all fraction tubes (0.5 ml). The fraction collector was set to automatically collect fractions 11-40. Fractions were scanned for radioactivity in a Nuclear Chicago (Chicago, IL) y-spectrometer. Peak fractions were pooled, and the volume reduced either under a stream of N, or under vacuum in a Speed-Vat, (Savant Instrument Co.. Farmingdale, NY). The samples were neutralized with 0.5 M sodium phosphate, pH 7.5. and stored at - 20% Because all of the VIP in the iodination reaction was oxidized on methionine 17 and the ([‘Z”I]iodo-Tyr10,0-Met17)VIP was resolved from this form of the peptide. we assigned a specific activity of > 2000 Ci mmol-’ to the monoiodinated VIP, based on the specific activity of lz51. Preparation of Membrane Fractions Retinal membranes were prepared from cow eyes by scraping the retinas from the pigment epithelium and placing them in 42 % sucrose (approximately 1.2 ml per retina) followed by vigorous shaking (approx. 300 times). This suspension was layered over 42 y0 sucrose in SW 27 tubes. and overlayered by a buffer containing 100 mM Tris, 2 mM MgCl, and 5 mM DTT, and centrifuged for I hr at 8 5 000 g. The resulting retinal pellet was depleted of rod outer segment membranes. The retinal pellets were homogenized using six strokes

AND

S.A

ROSENZWElG

of a motor-driven tetlon pestle at 2000 rpm in 0.2 5 M sucrose containing 5 mM EDTA. 1 mM phenylmethylsoybean trypsin sulfonylfluoride (PMSF). 001% inhibitor (STI). 1 mM o-phenanthroline and 0.01% leupeptin. The homogenate was then centrifuged at 5000 g in a refrigerated table top centrifuge for 10 min. and the supernatant obtained pelleted at 30 000 g for 30 min in a Beckman Type 3 5 rotor. The final retinal membrane pellets were resuspended in 25 rnh< HEPES. pH 7.4, containing 104 mM NaCl, 5 m&t MgCI,. 1 mM PMSF. 1 mM bacitracin, and 0.2(,y1 bovine serum albumin (HMS), and were stored at - 81PC. Rat retinal membranes were similarly prepared except that rod outer segments were not removed before homogenization as described above. Rat liver plasma membranes were prepared from the livers of 12 S-150 g male rats (Sprague-Dawley, CAMM Research Laboratories, Wayne, NJ) according to the method of Neville ( 1968) to step 11. All steps were carried out at 4°C. Competition Binding Analysis Membranes (approx. 50 kg protein) from bovine retina and rat liver were incubated in triplicate with equal amounts of radiolabeled VIP (SO PM) for 15 min at room temperature (t,,, = 10 min) in the presence of the indicated concentrations of unlabeled peptide. The mixtures were then placed on ice and centrifuged at 10000 g for 3 min. The pellets were then resuspended in ice-cold incubation buffer and recentrifuged. The washed pellets were then counted in a yspectrometer. Membrane-associated counts measured in the presence of 1 ,uM VIP (i.e. non-specific binding) were subtracted from all values. The data presented represent the means from at least two experiments for each membrane preparation analyzed. The Kiso indicates the amount of unlabeled VIP necessary to inhibit 50 % of the maximum radioligand binding to receptor. Afinity

Labeling of VIP Receptors

Membranes (approx. 100 ,ug protein) were incubated with [‘ZsI)VIP (1 nM) at 21°C for 15 min in a total reaction volume of 200 ,~l in the presence of the indicated concentrations of unlabeled VIP. Following washing and pelleting, the membranes were crosslinked using disuccinimidyl suberate (DSS) (500 ,uM). The cross-linking reaction was terminated by addition of 0.2 M Tris, pH 7.4 (10 ltl). The peileted membranes were resuspended in SDS-sample buffer containing 0.1 M DTT and the proteins resolved on a 10 % acrylamide SDS gel. The gel was fixed, stained, dried and autoradiographed using XAR-5 X-ray film and a DuPont Cronex Lightning Plus intensifying screen for 48 hr at -80°C. prior to development.

RETINAL

VIP

RECEPTORS

319

--

Endoglycosidase F Digestion of Affinity Labeled Receptors Membranes (approx. 50 ,ug of protein) were affinity labeled as described, solubilized in SDS sample buffer containing O-1 M DTT, and resolved on thin (O-75 mm) 10 % acrylamide SDS gels. The gels were dried without staining and flxation and the specifically labeled proteins identified by autoradiography. The specific bands were excised from the dried gel, placed in separate tubes and rehydrated in 0.18 ml endo F digestion buffer (0.1 M sodium phosphate, pH 6.1, with 50 mM EDTA, 1% NP-40, 0.1% SDS and 1% 2mercaptoethanol). Accessibility of enzyme to substrate was enhanced by sequentially freeze-thawing the tubes three times in dry ice-methanol and a 3 7°C bath for 10 min each. Twenty microliters of endo+Nacetylglucoaminidase F (endo F) (1 unit plll) was added to each tube (buffer added to controls) and incubation was allowed to proceed for 24 hr at 37°C. Incubations were terminated by the addition of 50 ~1 of SDS sample buffer containing 0.1 M DTT. The samples were electrophoresced on a second thick (1.5 mm) 10 Y0acrylamide SDS gel with a long stacker and analyzed by autoradiography. Endo F was prepared in the laboratory by a modification (Rosenzweig, Madison and Jamieson, 1984) of the method of Elder and Alexander (1982).

3. Results and Discussion Radioiodination of VIP As reported by Martin et al., ( 19 8 6) and Marie et al., (1985) we routinely obtained two major peaks and an occasional third peak of iodinated VIP by HPLC on a C, column (Fig. 1). The major peak obtained (denoted as Peak I) and that used in the analyses presented has been previously identified as ([1251]iodo-Tyr10,0Met”)VIP. Total cpm bound to membranes for this ligand represented 610% of the total added counts, with specific binding being approx. 65%. Under the radioiodination conditions used, all of the VIP in the reaction was oxidized on the single methionyl residue at position 17, based on monitoring the reaction products eluting from the column at 214 nm. Thus, the non-iodinated VIP in the reaction eluted from the column with a retention time of 11.4 min, corresponding to oxidized standard prepared with either chloramine-T, iodogen or hydrogen peroxide. Because of this, we were assured that the ([1251]iodo-Tyr10,0Met17)VIP obtained was not contaminated with native VIP thereby enabling us to assign a specific radioactivity of 3 2000 Ci mmol-* to the radioligand. Competition Binding Analyses VIP receptor affinity in bovine retinal membranes was examined by competition binding studies and the

Peak I /

?

Peak II

E :

VIF 5

k d

I

3

Fraction

40

50

number

FIG. 1. El&ion profile of radioiodination products chromatographed on a C, column. [1z51]VIP was prepared, as described in Materials and Methods, and purified by reversephase HPLC on a C, column. Fractions (0.5 min per fraction = 0.5 ml) 1140 were collected and 5 ~1 from each was scanned in a y-spectrometer for radioactive content (solid line). The dotted line indicates the acetonitrile gradient used to elute the peptides from the column.

results compared with VIP receptors in rat liver membranes. As shown in Fig. 2 (A) bovine retinal VIP receptors had a high affinity (Ki,, of approx. 1 nM) for VIP. This affinity was identical to that found for VIP receptors in rat liver membranes (K,,, of approx. 1 nM) and depicted in Fig. 2 (B). The specificity of this binding was examined by measuring the abilities of secretin and glucagon to compete with l”sI-VIP binding to retinal and hepatic membranes. Secretin is a peptide hormone bearing significant amino acid sequence homology to VIP and which has been reported to share binding sites with VIP in the guinea pig pancreas (Bissonnett et al., 1984). Interestingly, secretin had a negligible effect on [‘251]VIP binding in bovine retinal membranes whereas it was modestly effective in rat liver membranes (Kis, = 70 nM). Glucagon. which shares approx. 25% sequence homology with VIP, was relatively ineffective in competing with [1251]VIP for receptor binding in membranes from either the retina or liver. At concentrations up to 10m4 M, it produced less than 10% inhibition of [1251]VIP binding in retinal membranes and approx. 17 y0 inhibition in liver membranes. (Fig. 2). AfJinity Cross-linking

of VIP Receptors

In order to obtain biochemical information on the structure of the VIP receptor in retina, VIP binding siteswere affinity labeled with DSS ascross-linker, and the affinity labeled membranes were analyzed by SDS-PAGE under reducing conditions. As shown in Fig. 3, [1251]VIPlabeled a 59-kDa protein in bovine retinal membranes. Inclusion of cold VIP during binding resulted in the inhibition of radioligand binding and completely abolished the labeling of this protein at 0.1 pM.

320

A P SWEDLUND

0 VIP

0

0 Secrelin

AND

10-g .j ._-

10-7

S.A.

ROSENZWEIG

10-6

VIP, M

i

+Glucagon

1169760 66m b x i

45-

31-

FIG. 3. Affinity labelingof VIP receptorsin bovine retinal membranes.Bovine retinal membranes were incubated with

Peptide,

Molar

FIG. 2. VIP competitionbinding in bovine retinal and rat

liver membranes.Membranesfrom bovine retina and rat liver were preparedasdescribedin Materials and Methods. Incubations were carried out for 15 min at room temperature, at which time all manipulationswere carried out on ice. Shown are the specific counts bound. The data plottedrepresentsthe meansof triplicate determinationsand are representativeof three experiments.A, Bovine retinal membranes:B, rat liver membranes. When rat liver plasma membranes were affinity labeled, a major component of 59 kDa was labeled (Fig. 4). For comparison, VIP receptors on rat retinal membranes were also analyzed by affinity crosslinking. As shown in Fig. 4, a S6-kDa protein was specifically labeled in these membranes; approx. 3 kDa lessthan the rat liver and bovine retinal VIP receptors. A minor component of approx. 66 kDa was also detected in some experiments and probably represents the labeling of BSA present in the binding buffer. It should be noted that a minor 24-kDa component was labeled in bovine and rat retinal and in rat liver membranes. This protein was invariably detected and likely represents the non-specific affinity labeling of soybean trypsin inhibitor present in the cross-linking buffer.

[““I]VIP for 1 5 min at room temperature in the presence of the indicated amounts of unlabeled VIP. Following washing of the membranesto removenon-specificallyboundpeptide, bound ligand was cross-linked to its receptor with DSS for 5 min on ice. CnreactedDSS was quenchedby the addition of 10 mM Tris, pH 7.4 and the cross-linked membranes were cotlected by centrifugation and processed for SDS gel

electrophoresisand autoradiography. Shown is the autoradiograph

of the dried gel.

The discrepancy in molecular sizes between our affinity labeling data and that reported by others is unclear. In previous analyses of rat hepatic VIP receptors proteins of 48 kDa and 86 kDa (Couvineau and Laburthe. 198 5) and of 77 kDa and 53 kDa (Nguyen, Williams and Gray, 1986) were described. Similarly, binding sites of 73 kDa (major) and 33 kDa (minor) have been reported for the VIP receptor in intestinal epithelial cells (Laburthe, Breant and Rouyer-Fessard, 1984) where VIP has been shown to elicit a cyclase-related stimulation of electrogenic C1~ secretion (Amiranoff et al.. 1978). In colonic epithelial cells, [1251]VIPwas shown to label a 67-kDa protein (Muller et al., 1985). In human lymphoblasts, in which a VIP-stimulatable cyclase activity has been demonstrated (O’Dorisio et al., 1981). a 47-kDa binding protein has been identified by cross-linking (Wood and O’Dorisio. 19851. Thus, no general consensusof VIP receptor molecular weight has been deduced for its peripheral target cells. Our present data suggest that, unlike retinal cholecystokinin receptors (Bone and Rosenzweig, 1988). VIP receptors in the retina have essentially identical

RETINAL

VIP

RECEPTORS

321

retina

liver +

-

VIP

+

t-

59 kDa

e

56 kDa

FIG. 4. Affinity labeling of VIP receptors in rat retinal and liver membranes. Membranes from rat retina and rat liver were affinity labeled and processed as described in the legend to Fig. 2. Receptors of 56 kDa and 59 kDa. respectively, were labeled in rat retina and liver.

molecular weights to their counterparts in the periphery. Since the difference between neuronal and peripheral receptor. subtypes for insulin and cholecystokinin receptors is due to differential glycosylation (i.e. less glycosylation in neuronal cells), it was important to test whether VIP receptors exhibit differences in glycosylation. In order to determine whether the core protein molecular weights of the retinal and hepatic receptors were the same, affinity labeled proteins were treated with endo F and their digestion profiles compared. As shown in Fig. 5, the 59-kDa bovine retinal receptor was digested to two products of 52 and 45 kDa. The 56-kDa receptor in rat retina was digested to products of 48 and 45 kDa. The 59-kDa receptor in rat liver was digested to 47- and 45-kDa products (Table I), Thus, complete digestion led to a maximal decrease in molecular weight of approximately 11-14 kDa representing the apparent loss of two N-linked oligosaccharide chains. These results suggest that differences in glycosylation may be responsible for the disparities in molecular weights among the receptor proteins analyzed, but do not rule out additional structural differences, perhaps on the endodomains of the receptor(s) including primary sequence heterogeneity. As already mentioned, the differences in sensitivity of the hepatic VIP receptor to secretin may either be due to glycosylation or other structural modifications. In this regard. the hepatic receptor appears to contain two approx. 59 kDa

t

+

endo

F

45 kDo

FIG. 5. Endoglycosidase F digestion of affinity labeled VIP receptors. Af6nity labeled VIP receptors were resolved on a 10% acrylamide SDS gel and the radioactive proteins excised, rehydrated and digested with endo F, as described in Materials and Methods. A, Bovine retina; B, rat liver; C, rat retina. The size of the final, N-linked carbohydrate-free, digestion products obtained was 45 kDa.

23

Elm

51

322

A P SWEDLUND

TABLE I

Moleculur weights of afinity labeledVIP receptors (kDa) Glycosylated Bovine Retina Rat Liver Rat Retina

59 59 56

Deglycosylated -45 45 45

Values shown are uncorrected for the molecular weight of covalently associated [1251]VIP(3300 kDa). These values are based on at least three experiments and are representative of the data shown in Fig. 5. components (Fig. 5 ), perhaps representing two distinct binding proteins with one having a high sensitivity toward secretin and being absent in retinal membranes. It will be important to determine the functional properties of retina, VIP receptors in comparison to hepatic VIP receptors, particularly in regard to the stimulation of cellular effector systems such as adenylate cyclase and phospholipase C as was done for glucagon (Wakelam et al., 1986). These analyses may further delineate the apparent existence of VIP receptor subtypes in these tissues.

Acknowledgments We thank Dr Peter Stein for providing rod outer segmentdepleted bovine retinas, Mr Joseph Albert for assistance with the HPLC purification of [1251]VIP, and Mr Robert Brown for photography. We thank Dr Cecil C. Yip (University of Toronto) for critical review of this manuscript. This work was supported by NIH grants EY-06581 and DK-34389, a Juvenile Diabetes Foundation International Research Award and a Ronald McDonald Children’s Charities Research Award to S. A. Rosenzweig; Core Grant EY-00785, and a Research to Prevent Blindness unrestricted grant to the Department of Ophthalmology and Visual Science. A. P. Swedlund was supported by training grant AM-0701 7.

References Amiranoff, B., Laburthe, M., DuPont, C. and Rosselin, G. (19 78). Characterization of a vasoactive intestinal peptide-sensitive adenylate cyclase in rat intestinal epithelial cell membranes. Biochim. Biophys. Acta 544, 474-81. Besson, J.. Dussaillant, M., Marie, J. C., Rostene, W. and Rosselin. G. (1984). In vitro autoradiographic localization of vasoactive intestinal polypeptide (VIP) binding sites in the rat central nervous system. Peptides 5, 3 39-40. Bissonnett. B. M.. Collen M. J.. Adachi, H., Jensen. R. T. and Gardner, J. I$ (1984). Receptors for vasoactive intestinal peptide and secretin on rat pancreatic acini. Am. J. Phys. 246, G710-7. Bone, E. A. and Rosenzweig, S. A. ( 1988). Characterization of choiecystokinin receptors in toad retina. Peptides 9, 373-81. Christophe, J. P., Conlon, T. P and Gardner, J. D. (1976) Interaction of porcine vasoactive intestinal peptide with dispersed pancreatic acinar cells from guinea pig. Binding of radioiodinated peptide. 1. Biol. Chem. 251, 4629-34.

AND

S.A.

ROSENZWEIG

C‘ouvineau. A. and I,aburthe. M. I I Y85). The ral liver vasoactive intestinal peptide binding site. Molecular characterization of covalent cross-linking and evidence for differences from the intestinal receptor. Biochm. /. 225,4/3-9. Elder. 1. H. and Alexander, S. I 1982). Endo-/j-l-acetylglucosaminidase F : endoglycosidase from Flewhnrtuium mningoseptiwn that cleaves both highmannose and complex glycoproteins. Proc,. N&l. Acad. Sci. U.S.A. 79, 45403. Fahrenkrug. J. ( 1979). Vasoactive intestinal polypeptide: measurement, distribution and putative neurotransmitter function. Digmtion 19. 149-69. Inoue. Y.. Kaku, K.. Kaneko. T., Yanaihara, N. and Kanno. T. (19851. Vasoactive intestinal peptide binding to specific receptors on rat parotid acinar cells induces amylase secretion accompanied by intracellular accumulation of cyclic adenosine 3’-5’-monophosphate. f:ndor’rinolog:\ 11, 686-92. Laburthe. M.. Breant, B. and Rouyer-Fessard. C’. (19841. Rilolecular identification of receptors for vasoactive intestinal peptide in rat intestinal epithelium by covalent cross-linking. Evidence for two classes of binding sites with different structural and functional properties. /hr. 1. Biochcltrl. 139. 18 l-l 87. Larsson. L. I., Fahrenkrug, 1.. Schaffalitzky de Muckadell. 0.. Sundler, F.. Hakanson. R.. and Rehfeld, J. F. (1976). Localization of vasoactive intestinal polypeptide (VIP) to L.entral and peripheral neurons. Proc. Nntl. Acad. Sci. U.S.A. 73, 3 197-200. Lasater. E. M., Watling. K. J. and Dowling. J. E. ( 198 3 I. Vasoactive intestinal peptide alters membrane potential and cyclic nucleotide levels in retinal horizontal cells. Science 221. 1070-Z. Loren, I.. Tornqvist, K. and Alunets. J. (1980). VIP ivasoactive intestinal polypeptide)-immunoreactive neurons in the retina of the rat. Cell. Tissw Kes. 210, 167-70. Magistretti, P. J.. Morrison, j. H., Shoemaker, W. J.. Sapin. V. and Bloom, F. E. (1981). Vasoactive intestinal polypeptide induces glycogenolysis in mouse cortical slices : a possible regulatory mechanism for the local control of energy metabolism. Pm. Natl. Acad. Sci. L’.S.A. 38. 6535-9. Magistretti, P. J. and Schorderet. M. (1984). VIP and noradrenaline act synergistically to increase cyclic AMP in cerebral cortex. Nature 308. 280-2. Marie. J. C.. Hui Bon Hoa. D., Jackson, R.. Hejblum. G. and Rosselin. G. (1985). The biological relevance of HPLC purified vasoactive intestinal polypeptide monoiodinated at tyrosine IO or tyrosine 22. Regul. Pept. 12. 113-23. Martin. J. L., Rose. K., Hughes, G. J. and Magistretti. I’. J. ( 1986). [mono[‘251]iodo-Tyri0, metO”]-vasoactive intestinal polypeptide. Preparation, characterization, and use for radioimmunoassay and receptor binding. 1. viol. Chem. 261, 5320-7. McCulloch. J. and Kelly. P. A. T. ( 1983). A functional role for vasoactive intestinal polypeptide in anterior cingulate cortex. Nature 304, 438-40. Muller. J.-M.. Luis, J.. Fantini. J., Abadie. B.. Giannellinni. F.. Marvaldi. J. and Pichon, J. (1985). Covalent crosslinking of vasoactive intestinal peptide (VIP) to its receptor in intact colonic adenocarcinoma cells in culture (HT 29). Eur. 1. Biochorn. 151. 41 1-i. Neville. D. M. (I 968). Isolation of an organ specific protein antigen from cell-surface membrane of rat liver. Biochem. Biophys Acta. 154, 540-52. Nguyen. T. D.. Williams. J. A. and Gray, G. M. (1986). Vasoactive intestinal peptide receptor on liver plasma

RETINAL

VIP

323

RECEPTORS

membranes : Characterization

as a glycoprotein. Bio-

chemistry 25, 361-8.

O’Dorisio,M. S., Hernina, N. S., O’Dorisio,T. M. and Balcerzak, S.P. (1981). Vasoactiveintestinal polypeptide modulation of lymphocyte adenylate cyclase. J. Immunof. 127, 22514. Robberecht,P.. Waelbroeck.M., Camus,J. C., deNeef,P. and Christophe,J. (1984). Importanceof disulfidebondsin receptorsfor vasoactive intestinal peptideand secretin in rat pancreaticplasmamembranes.Biochim.Biophys. Acta 73, 271-8. Rosenzweig,S.A., Madison,L. D. andJamieson,J. D. (1984). Analysis of cholecystokinin-binding proteins using endo-P-N-acetylglucosaminidase F. J. Cell Biol. 99, 1110-6. Said, S.I. (1984). Vasoactiveintestinal polypeptide(VIP): Current status.Peptides5, 143-50.

Scott, J. and Baum. B.J. (1985). Involvement of cyclic AMP and calcium in exocrine protein secretioninduced by vasoactive intestinal polypeptidein rate parotid cells. Biochim.Biophys.Acta 847, 255-62. Stanisz, A. M., Befus, D. and Bienenstock, J. (1986). Differential effects of vasoactive intestinal peptide. substanceP, and somatostatin on immunoglobulin synthesis and proliferations by lymphocytes from Peyer’spatches,mesentericlymph nodesand spleen.J. lmmunol. 136. 152-6. Wakelam,M. J. 0.. Murphy, G. J., Hruby, V. J. and Houslay, M. D. (1986). Activation of two signal-transduction systems in hepatocytes by glucagon. Nature 323. 68-71. Wood, C. L. and O’Dorisio,M. S. (1985). Covalent crosslinking of vasoactive intestinal polypeptide to its receptorson intact human lymphoblasts.J. Biol. Chem. 260. 1243-7.

23-2