- Email: [email protected]
Pharmacological characterization of VIP receptors in human lung membranes
Peptides, Vol. 9, pp. 33%345. ~ Pergamon Press plc, 1988. Printed in the U.S.A.
0196-9781/88 $3.00 + .00
Pharmacological Characterization of VIP Receptors in Human Lung Membranes P. ROBBERECHT, M. WAELBROECK, P. DE N E E F , J. C. C A M U S , D. H. C O Y * A N D J. C H R I S T O P H E 2
Department o f Biochemistry and Nutrition, Medical School UniversitO Libre de Bruxelles, B-IO00 Brussels, Belgium and *Department of Medicine, Medical School Tulane University, New Orleans, LA 70112 R e c e i v e d 26 M a y 1987 ROBBERECHT, P., M. WAELBROECK, P. DE NEEF, J. C. CAMUS, D. H. COY AND J. CHRISTOPHE. Pharmacological characterization of V1P receptors in human lung membranes. PEPTIDES 9(2) 33%345, 1988.--The ability of VIP, PHI, secretin, helodermin, and seven N-terminally D-amino monosubstituted V1P and PHI analogs to occupy (125I)iodo-VIPlabeled receptors and to activate adenylate cyclase was tested on human lung membranes purified by the method of Schachter et al. Best fitted Kd, Kact and % of max. values suggested the coexistence, in near equal proportions, of two classes of VIP-preferring binding sites coupled to adenylate cyclase that showed similar decreasing affinity for: VIP > (D-AIa4)-VIP > (D-Asp3)-VIP = (D-Ser2)-VIP > (D-His')-VIP > PHI > (D-Phe~)-VIP > (D-Phe~)-VIP. (D-Arg2)-VIP was a non-selective agonist. A third receptor type, coupled to adenylate cyclase and showing high affinity for secretin and helodermin but not for VIP, was also detected. VIP receptors
Human lung membranes
preparations. To achieve this aim, we compared the ability of VIP, PHI, secretin, helodermin and N-terminally modified VIP and PHI analogs to occupy VIP receptors and to activate adenylate cyclase in purified human lung membranes.
T H E neurotransmitter VIP (Vasoactive Intestinal Peptide) is a physiological regulator of the respiratory tract that could serve as a helpful pharmacological agent in pulmonary disease when applied exogenously. Indeed: (1) nerve endings containing VIP are associated to airways smooth muscle, submucosal glands, and blood vessels in the trachea and stem bronchi [11,33]; (2) VIP nerves are closely related with bronchial smooth muscle and with the tunica adventitia of pulmonary arteries [9]; (3) specific VIP receptors are identiffed by their binding characteristics in pulmonary membranes from several mammalian species including man [4, 10, 15, 23, 31] and by their morphological properties in alveolar capillaries, pulmonary smooth muscle, alveolar walls and bronchial epithelium in rat [15] and man [16]; (4) pulmonary VIP receptors are efficiently coupled to adenylate cyclase [4, 22, 23, 32]; (5) VIP may mediate the nonadrenergic relaxation of guinea pig tracheal smooth muscle [17]; (6) VIP relaxes isolated guinea pig tracheal muscle dose-dependently [2]; (7) VIP causes bronchodilation and protects against histamine-induced bronchoconstriction in asthmatic patients [18]; (8) local levels of VIP increase in rat lung exposed to asbestos [7], and (9) VIP receptors are found in human lung tumor cells in culture [13]. It was thus felt of interest to examine whether VIP analogs (synthetic or natural) might act as superagonist or antagonist on (sub)classes of VIP receptors in human lung
METHOD
Human Lung Specimens Normal lung tissue from 6 male patients (35-55 years old) suffering from bronchopulmonary cancer was obtained during segmentectomy or lobectomy. As soon as possible after tissue devascularization, the tumor was dissected out and the remaining healthy tissue was frozen on dry-ice then stored in liquid nitrogen until use.
Preparation of Lung Membranes After thawing of the tissue, we used the method described by Schachter et al. [31] with only one modification consisting in the omission of leupeptin from the homogenizing medium. Aliquots of crude membranes were stored in liquid nitrogen.
Binding Studies on Lung Membranes Synthetic VIP was radioiodinated by the chloramine-T technique and purified as previously described [3,24]. (~zsI)iodo-VIP with specific radioactivity of 200/~Ci//~g and
'Aided by Grant 3.4571.85 from The Fonds de la Recherche Scientifique M6dicale (Belgium), Grants ROI-AM-17010-9 and AM-31670 from the National Institutes of Health (USA) and a Grant from the Queen Elisabeth Medical Foundation (Belgium). ZRequests for reprints should be addressed to J. Christophe, Department of Biochemistry and Nutrition, Medical School, Universit6 Libre de Bruxelles, Boulevard de Waterloo 115, B-1000 Brussels, Belgium.
339
340
ROBBERECHTETAL.
LUNG
® 100-
~I~. = ~ , ~ , f -
Helodermin
"'~,.~ X N~x , ~,
\\{Secret in
50O Z D 0 cn
u~cu ~--I
O"
r~
,
,
,
,
• ""-
a.i
--
a
--
!
!
®
is
~,~-.,
(D-ARG2) -rIP .~
(D-PHE21-VlP
\
\ \ /
co
-11
-10
-9
-8
-7
-6
-5
[PEPTIDE CONCENTRATION](10g M) FIG. 1. Inhibition of (l~al)iodo-VIP binding to human lung membranes by VIP (©), PHI (A), secretin (0) and helodermin (&) (upper panel A) and by VIP (©), (D-AIa4)-VIP (T), (D-Asp3)VIP (A), (D-Ser2).VIP (A), (D-His~)-VIP (0), {D-Phe2)-VIP ([~), (D-Arg~)-VlP (m) and (D-Phe4)-PHI (V) (lower panel B). The results were expressed as a percentage of tracer specifically bound in the absence of unlabeled peptide and were the means of four experiments performed in duplicate.
0.4 iodine atom/molecule on an average was stored at -20°C. B i n d i n g o f (12~l)iodo-VIP. Binding studies were carried out in a total volume of 0.120 ml in a 50 mM Tris/maleate buffer (pH 7.4) containing 5 mM MgC12, 0.5 mg/ml bacitracin, 100 kallikrein inhibitor unit/ml Trasylol, 1% bovine serum albumin, 20-30 pM (125I)iodo-VIP, increasing concentrations of unlabeled peptide and 10 to 15/zg tissue protein. Incubation was conducted at 37°C for 10 min, i.e., until binding equilibrium was attained. The incubation was terminated by rapid filtration under reduced pressure through Whatman GF/C filters (Maidstone, UK) pre-soaked overnight in 0.5% polyethyleneimine [31] in order to reduce non-specific tracer binding. The filters were washed 3 times with 3 ml ice-cold buffer containing 50 mM Tris, 0.5 mM EDTA, 0.2% bovine serum albumin (pH 7.4) and counted in a Gamma Spectrometer. Non-specific binding was defined as ('~5I)iodo-VIP bind-
ing in the presence of 1/zM VIP and never exceeded 20% of the total bound radioactivity. Adenylate
Cyclase Assay
Adenylate cyclase activity was determined according to the Salomon et al. procedure [30] as previously detailed [32]. T e n / z M GTP was included in all assays. Cyclic A M P production was always linear and proportional to the amount of tissue protein added. Peptides and Chemicals
VIP, PHI, VIP analogs and PHI analogs were assembled by solid-phase methods and purified by ion-exchange and partition chromatography as previously described [6]. Natural helodermin was prepared as detailed in [34]. Synthetic
VIP RECEPTORS IN H U M A N L U N G
K d
341
TABLE 1 VALUES AND PROPORTIONS OF ('2~I)IODO-VIPLABELED RECEPTORS FOR OCCUPANCY BY VIP AND VIP ANALOGSIN HUMAN LUNG MEMBRANES
Peptide
Kdl (nM)
% of Sites
VIP Secretin PHI Helodermin (D-Hisl)-VIP (D-Ser2)-VIP (D-Asp'~)-VIP (D-Ala4)-VIP (D-Phe2)-VIP (D-Arg2)-VIP (D-Phe4)-VIP
0.08 3 3 1 1 0.7 0.5 0.1 10 200 100
43 8 43 10 43 43 43 40 40 100 40
Kd2 (nM) 2.5 300 100 50 20 20 20 10 100 . . 2000
% of Sites 57 32 57 30 57 57 57 60 60 . 60
Kd3 (nM) - -
% of Sites - -
1000 -2000 ------
60 -60 ------
--
--
.
The values were deduced from the best fit of data presented in Fig. 1. porcine secretin was a generous gift from D. W. Krnig (Hoechst Aktiengesellschaft, Frankfurt, FRG). Carrier-free N a 1~5I (600-800 mCi/ml) and (8-all)cyclic AMP (20-30 Ci/mole) were purchased from Amersham International (Amersham, Bucks, GB). (c~-3ZP)ATP (30 Ci/mmole) was from New England Nuclear (Boston, MA). Bovine serum albumin (fraction V), bacitracin, ATP, cyclic AMP, GTP, and polyethyleneimine were obtained from Sigma Chemical Co. (St Louis, MO). Kallikrein inhibitor (Trasylol) was a gift from Bayer (Brussels, Belgium). All other reagents were of the highest purity degree available. All peptides were dissolved in 0.01 M sodium phosphate buffer (pH 7.5), enriched with 0.9% sodium chloride, 0.1% sodium azide and 0.05% Tween 20. This buffer reduced peptide binding to stock tubes and, when diluted in the media used for binding studies or adenylate cyclase assay, was uninfluential on the parameters investigated yet minimized peptide binding to the test tubes. The results of binding studies were analyzed assuming that the ligand recognized one, two or three subclasses of receptors, using the program of Richardson and Humrich [21]. RESULTS
General Characteristics of (125I)iodo-VIPBinding The present data fully confirmed previous results from Schachter et al. [31] showing that tracer binding was rapid at 37°C: steady state binding was attained after 5 min already then maintained for at least 15 more min. This binding was rapidly and totally reversible in the presence of GTP (data not shown). During the first 15 min of incubation, tracer degradation did not exceed 15% as measured by trichloroacetic acid precipitation (it is acknowledged that this relatively low percentage probably represented a minimal figure of tracer degradation). The cations magnesium (5 to 20 mM) and manganese (0.25 to 2.5 mM) but not calcium (0.5 to 5.0 mM) increased tracer binding to a maximum of 50% whereas EDTA and EGTA were without effect (data not shown).
Specificity of VIP Binding Sites and Structural Requirements for Receptor Occupancy The capability of peptides to occupy VIP binding sites
was evaluated by their ability to inhibit (125I)iodo-VIP binding: the data shown in Fig. 1A were expressed in % of tracer bound in the presence of increasing concentrations of unlabeled peptide. VIP binding sites were highly specific as PHI, secretin and helodermin were, respectively, 30-, 500- and 600-fold less potent than VIP based on the peptide concentration required for 50% tracer binding inhibition (Kd). The N-terminal part of VIP was critical for receptor recognition: a systematic substitution of the first four amino acids by their D-isomer led to compounds 2- to 8-fold less potent than VIP, the potency decreasing with D-substitutions closer to the N-terminus. Substitution of serine in position 2 by a large D-Phe or D-Arg residue reduced the peptide potency even more: 50- and 200-fold, respectively. (D-Phe4)-PHI was 60fold less potent than PHI itself.
Evidence for the Presence of Subclasses of VIP Binding Sites All inhibition c u r v e s - - e x c e p t the curve for (D-Arg2)V I P - - s h o w e d a Hill slope factor smaller than one, suggesting that more than one class of binding sites was occupied by tracer and/or unlabeled peptide. The use of a curve fitting program [21] allowed to determine the proportions of sites and corresponding K d values for the peptides capable to occupy these sites. The results in Table 1 indicate the best fit after considering, successively, the potential existence of one, two or three binding sites. The average proportions of high affinity and low affinity binding sites for VIP and PHI analogs were 42% and 58%, respectively. The selectivity, i.e., the capability of peptides to discriminate between both classes of binding sites was inferred from the ratio of Kd values for low and high affinity sites. VIP, (D-Hisl)-VIP, (D-Ser2)-VIP, (D-Asp3)-VIP, PHI and (D-Phe4)-PHI showed comparable selectivity, ranging from 20 to 40. (D-AIa4)-VIP recognized high affinity binding sites preferentially. (DPhe2)-VIP and (D-Arg~)-VIP had a lower selectivity and (DArgZ)-VIP could even be considered as a non-selective ligand. For secretin and helodermin the best fit, when considering two classes of binding sites, gave 25% high affinity sites and 75% low affinity sites, i.e., proportions significantly different from those obtained with VIP and PHI analogs.
ROBBERECHT ET AL.
342
LUNG
® I00-
PHI . ~ vIp
.o.--'" /
/
50 Z o
~, ,~-'"
~
,
He.lodermin
I--
---'" ~ I.u
0 dl,
Iii
®
J
Secretin ,*,
,.
.
.
.
(D-HIS I) -VIP (D-ALA4) -VIP ~ , . ~ ~
100-
iii I-J >.. z iii i-i
tpPO
50
" ~ o-'"
.,"
-I0
//7/
• -""
/r/q-
-9
-8
~ ~'~'-+,7
sp31-/
-7
/
Co- HE21-vI
-6
-5
[PEPTIDE CONCENTRATION] (log M) FIG. 2. Dose-effect curves of activation of human lung membrane adenylate cyclase by VIP (O), PHI (A), secretin (O), helodermin (&) (upper panel A) and by VIP (O), (D-Ala~)-VIP(T), (D-Asp3)-VIP (A), (D-Ser2)-VIP (A), (D-Hisl)-VIP (0), (D-Phe2)-VIP (rq), (D-Arg2)-VIP (11) and (D-Phe4)-PHI (V) (lower panel B). The results were expressed as a percentage of adenylate cyclase activation observed in the presence of 10/xM VIP after subtraction of the basal value and were the means of four experiments performed in duplicate. The mean_+SEM basal activity was 58.3---4.0 pmol cyclic AMP.min l.mg protein -t and the mean±SEM activity in the presence of 10/zM VIP was 138.2±8.0 pmol cyclic AMP.min-Umg protein 1.
This fitting improved, however, when a three classes model was considered (see Table 1 and Discussion).
Adenylate Cyclase Activation All peptides tested were able to stimulate adenylate cyclase but with variable efficacy (Figs. 2A and 2B). (D-Arg2)VIP and (D-Phe2)-VIP, although active, were partial agonists only with an intrinsic activity of 2 7 - 2 ~ as compared to that of VIP. Most curves were compatible with an activation process mediated by more than one class of binding sites. By analogy with the data obtained in binding studies, the results were fitted according to models with one, two or three classes of functional receptors. For the VIP dose-response curve, the best fit was ob-
tained considering the coexistence of three receptor classes. If, for most VIP analogs, a fitting of similar quality was obtained for two or three classes, we favored the latter hypothesis after our analysis of the VIP dose-effect curve. Table 2 gives the calculated Kact for each class and the corresponding maximum stimulation of adenylate cyclase (expressed in % of total activity obtained with I0/zM VIP).
DISCUSSION This study was initially undertaken to search for VIP and PHI analogs that could act as VIP superagonist or antagonist on human lung membranes. In this respect, our data were negative as VIP was the most potent and efficient peptide at
VIP RECEPTORS IN H U M A N L U N G
343
TABLE 2 ACTIVATIONOF ADENYLATECYCLASEBY VIP AND VIP ANALOGSIN HUMAN LUNG MEMBRANES
Peptide
Kactl (nM)
% of Max. Activity
VIP Secretin PHI Helodermin (D-Hisl)-VIP (D-Ser2)-VIP (D-Asp3)-VIP (D-AIa4)-VIP (D-Phe2)-VIP (D-ArgZ)-VIP (D-Phe4)-PHI
0.08 6 5 3 3 1 0.5 0.1 200 300 100
20 28 14 20 35 10 15 10 27 28 20
Kact2 (nM) 2.5 3000 100 200 40 30 20 10 . . . . 1000
% of % of Max. Kact3 Max. Activity (nM) Activity 50 45 86 55 65 70 55 55 . . 50
300 --2000 -3000 1000 300
30 --20 -15 15 25
--
--
. .
The best estimated values for Kact and maximum activation (in % of that observed with 10/zM VIP) were derived from data presented in Fig. 2.
hand and as analogs like (D-Phe2)-VIP and (D-Arg2)-VIP, that are devoid of activity or behave as VIP antagonist in rat tissues [27], were able to stimulate human lung adenylate cyclase. The results gave, nevertheless, valuable information on the structural requirements for lung VIP receptor occupancy and subsequent adenylate cyclase activation: they substantiated the importance of the N-terminal portion of VIP for receptor site recognition, due probably to the presence of a beta bend in the 1-4 region [19]. They also demonstrated the highly unfavorable role of increasing hydrophobicity in position 2 and 4 and of having a basic D-amino acid residue in position 2. They also illustrated the key role of histidine in position 1, confirming previous studies on rat tissues and human heart membranes [25]. Such conclusions cannot be extended to other members of the VIP family: for example, (D-Phe4)-GRF(1-29)-NH2, when interacting with VIP receptors in both rat pancreatic and liver membranes, is more potent than G R F itself [28] and (D-Phe4)-glucagon, when binding to glucagon receptors in rat liver, is a glucagon superagonist ([19] and unpublished results). Although not substantiated in the present work, it is also highly likely that the C-terminal part of the molecule is involved in receptor recognition as PHM is found by Schachter et al. to be threefold more potent than PHI in human lung membranes [31]. Human lung VIP binding sites were heterogeneous based on binding studies as well as on adenylate cyclase activation data. The detailed computerized study of this heterogeneity was made possible by the high feasibility of binding (high specific 125I-VIP binding, low non-specific binding, and low tracer degradation due to the short time period needed to reach steady state binding). VIP, VIP analogs (except (DArg2)-VIP), PHI and the PHI analog (D-Phe4)-PHI discriminated two classes of binding sites: a high affinity class (42% of labeled sites) and a low affinity class (58% of labeled sites). Their K d values were in line with those obtained on rat lung membranes [20] but high affinity VIP receptors were proportionately more numerous in man than in rat lung membranes. It is clear, however, that the percentage of sites given for the two classes of receptors in both man (the present data) and rat [20] is indicative only as it was not demon-
strated that (12~I)iodo-VlP (in all possible forms) binds unspecifically (i.e., with the same affinity) to all receptor subclasses. Both classes of receptors in our preparations were coupled to adenylate cyclase and the correlation between Kd values for binding and Kact values for enzyme stimulation was good. Structural requirements for receptor occupancy and subsequent adenylate cyclase activation were quite similar in both classes as the order of peptide potency was comparable: VIP > (D-Alaa)-VIP > (D-AspZ)-VIP = (D-Ser2)VIP > (D-Hisl)-VIP > PHI > (D-Phez)-vIP > (D-Phe4)-PHI. (D-Arg2)-VIP was a non-selective partial agonist whose capability to activate adenylate cyclase could not be correlated with a given receptor class. The correlation between binding data and adenylate cyclase activation and Hill slope factors smaller than one for all VIP agonists (except (D-ArgZ)-VIP) allow to conclude that the heterogeneity of binding sites reflected the coexistence of two distinct classes of receptors rather than the presence of two receptor states that could be discriminated by agonists but not by antagonists. It should be noted that covalent cross-linking experiments on lung, like those on liver and intestine, also offer evidence for VIP receptor heterogeneity [5, 10, 14]. VIP and the three most potent VIP analogs ((D-Ala4)-VIP, (D-Aspa)-VIP, and (D-Ser2)-VIP) further activated adenylate cyclase at concentrations higher than those required for occupying the two sites just described (Table 2). This contribution of a third component had to be considered to obtain adequate fitting of dose-response curves (Table 2) but could not be detected in binding studies (Table 1) due probably to Kd values for this component close to the concentration of unlabeled VIP (10/zM) used for non-specific (lzsI)iodo-VIP binding determination. This third component could represent either a very low affinity VIP receptor or the feable interaction of VIP and the three most potent VIP analogs with a class of receptors having high affinity for a parent peptide. This last hypothesis is favored when considering how secretin and helodermin interacted with VIP receptors and activated adenylate cyclase. Indeed, secretin and helodermin binding data could not fit in a model involving two VIP binding sites only with relative proportions of 42% and 58%. By
344
ROBBERECHT ET AL.
contrast, the same data fitted perfectly with a three sites model as proposed in Table 1. Altogether our data in Tables 1 and 2 can be interpreted as follows: (a) secretin and helodermin interacted with high affinity secretin/helodermin receptors that were (very) poorly labeled with ('2~I)iodo-VIP. This was followed by adenylate cyclase activation (Kd, and Kact,); (b) secretin and helodermin interacted with high affinity VIP receptors (Kd2) but only the interaction of helodermin promoted detectable adenylate cyclase activation (Kactz of helodermin); (c) secretin and helodermin bound also to low affinity VIP receptors (Kd3) and this interaction of both peptides led to increased cyclic A M P production (Kact2 of secretin and Kact3 of helodermin). This interpretation is in good agreement with previous data on secretin and VIP receptors: secretin binds to both secretin receptors and VIP receptors and with or without resulting adenylate cyclase activation in guinea pig and rat
pancreas [1, 3, 24]; helodermin, a parent peptide of VIP and secretin extracted from Gila monster venom [6,12], interacts efficiently with VIP receptors in tissues devoid of secretin receptors [26,32] and with near equal potency with secretin receptors and VIP receptors in rat pancreas [8,26]. Any efficient interaction of PHI with secretin receptors would be difficult to demonstrate due to the low efficacy of secretin itself (Fig. 2). In conclusion, human lung membranes possess multiple VIP and secretin receptors coupled to adenylate cyclase. VIP is likely to cross-react with secretin receptors and secretin with VIP receptors. Helodermin recognized all receptors. The N-terminal part of VIP was essential for receptor site recognition. We were unable to find a superagonist or an antagonist for VIP receptors among the seven analogs monosubstituted in the first 4 amino acids of VIP and PHI that were tested.
REFERENCES 1. Bissonnette, B. M., M. J. Collen, H. Adachi, R. T. Jensen and J. D. Gardner. Receptors for vasoactive intestinal peptide and secretin on rat pancreatic acini. A m J Physiol 246: G710-G717, 1984. 2. Christofides, N. D., Y. Yiangou, P. J. Piper, M. A. Ghatei, M. N. Sheppard, K. Tatemoto, J. M. Polak and S. R. Bloom. Distribution of peptide histidine isoleucine in the mammalian respiratory tract and some aspects of its pharmacology. Endocrinology 115: 1958-1963, 1984. 3. Christophe, J., J. P. Conlon and J. D. Gardner. Interaction of porcine vasoactive intestinal peptide with dispersed pancreatic acinar cells from the guinea pig: binding of radioiodinated peptide. J Biol Chem 251: 4629-4634, 1976. 4. Christophe, J., P. Chatelain, G. Taton, M. Delhaye, M. Waelbroeck and P. Robberecht. Comparison of VIP-secretin receptors in rat and human lung. Peptides 2: Suppl 2, 113-118, 1981. 5. Couvineau, A. and M. Laburthe. The rat liver vasoactive intestinal peptide binding site. Molecular characterization by covalent cross linking and evidence for differences from the intestinal receptor. Biochem J 225: 473-479, 1985. 6. Coy, D. H. and J. Gardner. Solid-phase synthesis of porcine vasoactive intestinal peptide. Int J Pept Prot Res 15: 73-78, 1980. 7. Day, R., I. Lemaire, P. Mercier, H. Beaudoin and S. Lemaire. Asbestos-related increase in pulmonary levels of vasoactive intestinal peptide (VIP). Life Sci 33: 1869-1876, 1983. 8. Dehaye, J. P., J. Winand, C. Damien, F. Gomez, P. Poloczek, P. Robberecht, A. Vandermeers, M. C. Vandermeers-Piret, M. Stirvenart and J. Christophe. Receptors involved in helodermin action on rat pancreatic acini. Am J Physiol 251: 599-602, 1986. 9. Dey, R. D., W. A. Shannon and S. I. Said. Localization of VIP-immunoreactive nerves in airways and pulmonary vessels of dogs, cats and human subjects. Cell Tissue Res 220: 231-238, 1981. 10. Dickinson, K. E. J., M. Schachter, C. M. M. Miles, D. H. Coy and P. S. Sever. Characterization of vasoactive intestinal peptide (VIP) receptors in mammalian lung. Peptides 7: 791-800, 1986. 11. Ghatei, M. A., M. N. Sheppard, D. J. O'Shaughnessy, T. E. Adrian, G. P. McGregor, J. M. Polak and S. R. Bloom. Regulatory peptides in the mammalian respiratory tract. Endocrinology 111: 1248--1254, 1982. 12. Hoshino, M., C. Yanaihara, Y. M. Hong, S. Kishida, Y. Katsumaru, A. Vandermeers, M. C. Vandermeers-Piret, P. Robberecht, J. Christophe and N. Yanaihara. Primary structure of helodermin, a VIP-secretin-like peptide isolated from Gila monster venom. FEBS Lett 178: 233-239, 1984.
13. Laburthe, M., C. Boissard, G. Chevalier, A. Zweibaum and G. Rosselin. Peptide receptors in human lung tumor cells in culture: vasoactive intestinal peptide (VIP) and secretin interaction with the CALU-1 and SW-900 cell lines. Regul Pept 2: 219-230, 1981. 14. Laburthe, M., B. Breant and C. Rouyer-Fessard. Molecular 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. Fur J Biochem 139: 181-187, 1984. 15. Leroux, P., H. Vaudry, A. Fournier, S. St Pierre and G. Pelletier. Characterization and localization of vasoactive intestinal peptide receptors in the rat lung. Endocrinology 114: 1506-1512, 1984. 16. Leys, K., A. H. Morice, O. Madonna and P. S. Sever. Autoradiographic localisation of VIP receptors in human lung. FEBS Lett 199: 198--202, 1986. 17. Matsuzaki, Y., Y. Hamasaki and S. I. Said. Vasoactive intestinal peptide: a possible transmitter of noradrenergic relaxation of guinea pig airways. Science 210: 1252-1253, 1980. 18. Morice, A., R. J. Unwin and P. S. Sever. Vasoactive intestinal peptide causes bronchodilatation and protects against histamine-induced bronchoconstriction in asthmatic subjects. Lancet II: 1225-1227, 1983. 19. Murphy, W. A., D. H. Coy and V. A. Lance. Superactive amidated COOH-terminal glucagon analogues with no methionine or tryptophan. Peptides 7: Suppl 1, 69-74, 1986. 20. Provow, S. and G. Veliqelebi. Characterization and solubilization of Vasoactive Intestinal Peptide receptors from rat lung membranes. Endocrinology 120: 2442-2452, 1987. 21. Richardson, A. and A. Humrich. A microcomputer program for the analysis of radioligand binding curves and other doseresponse data. Trends Pharmacol Sci 5: 47-49, 1984. 22. Robberecht, P., P. Chatelain, P. De Neef, J. C. Camus, M. Waelbroeck and J. Christophe. Presence of vasoactive intestinal peptide receptors coupled to adenylate cyclase in rat lung membranes. Biochim Biophys Acta 678: 76-82, 1981. 23. Robberecht, P., K. Tatemoto, P. Chatelain, M. Waelbroeck, M. Delhaye, G. Taton, P. De Neef, J. C. Camus, D. Heuse and J. Christophe. Effects of PHI on vasoactive intestinal peptide receptors and adenylate cyclase activity in lung membranes. A comparison in man, rat, mouse and guinea pig. Regul Pept 4: 241-250, 1982. 24. Robberecht, P., M. Waeibroeck, M. Noyer, P. Chatelain, P. De Neef, W. K6nig and J. Christophe. Characterization of secretin and vasoactive intestinal peptide receptors in rat pancreatic plasma membranes using the native peptides, secretin-(7-27) and five secretin analogues. Digestion 23: 201-210, 1982.
VIP R E C E P T O R S I N H U M A N L U N G
25. Robberecht, P., M. Waelbroeck, J. C. Camus, P. De Neef, D. Coy and J. Christophe. Effects of His I modifications on the ability of vasoactive intestinal peptide to stimulate adenylate cyclase from rat and human tissues. Peptides 5: 877-881, 1984. 26. Robberecht, P., M. Waelbroeck, J. P. Dehaye, J. Winand, A. Vandermeers, M. C. Vandermeers-Piret and J. Christophe. Evidence that helodermin, a newly extracted peptide from Gila monster venom, is a member of the secretin/VIP/PHI/family of peptides with an original pattern of biological properties. FEBS Lett 166: 277-282, 1984. 27. Robberecht, P., D. H. Coy, P. De Neef, J. C. Camus, A. Cauvin, M. Waelbroeck and J. Christophe. Interaction of vasoactive intestinal peptide (VIP) and N-terminally modified VIP analogs with rat pancreatic, hepatic and pituitary membranes. Eur J Biochem 159: 45-49, 1986. 28. Robberecht, P., M. Waelbroeck, D. Coy, P. De Neef, J. C. Camus and J. Christophe. Comparative structural requirements of thirty GRF analogs for interaction with GRF- and VIP receptors and coupling to adenylate cyclase in rat adenopituitary, liver and pancreas. Peptides 7: Suppl 1, 53-59, 1986.
345
29. Robberecht, P., N. Yanaihara, C. Yanaihara, P. De Neef, J. C. Camus, A. Vandermeers, M. C. Vandermeers-Piret and J. Christophe. Interaction of synthetic N- and C-terminal fragments of helodermin with rat liver VIP receptors. Peptides 7: Suppl l, 79-82, 1986. 30. Salomon, Y., C. Londos and M. Rodbell. A highly sensitive adenylate cyclase assay. Anal Biochem 58: 541-548, 1974. 3 I. Schachter, M., K. E. J. Dickinson, C. M. Miles and P. S. Sever. Characterisation of a high-affinity VIP receptor in human lung parenchyma. FEBS Lett 199: 125-129, 1986. 32. Taton, G., M. Delhaye, J. C. Camus, P. De Neef, P. Chatelain, P. Robberecht and J. Christophe. Characterization of the VIPand secretin-stimulated adenylate cyclase system from human lung. Pflugers Arch 391: 178--182, 1981. 33. Uddman, R., J. Alumets, O. Densert, R. Hakanson and F. Sundler. Occurrence and distribution of VIP nerves in the nasal mucosa and tracheo-bronchial wall. Acta Otolaryngol (Stockh) 86: 443-448, 1978. 34. Vandermeers, A., M. C. Vandermeers-Piret, P. Robberecht, M. Waelbroeck, J. P. Dehaye, J. Winand and J. Christophe. Purification of a novel pancreatic secretory factor (PSF) and a novel peptide with VIP- and secretin-like properties (helodermin) from Gila Monster venom. FEBS Lett 166: 273-276, 1984.
0196-9781/88 $3.00 + .00
Pharmacological Characterization of VIP Receptors in Human Lung Membranes P. ROBBERECHT, M. WAELBROECK, P. DE N E E F , J. C. C A M U S , D. H. C O Y * A N D J. C H R I S T O P H E 2
Department o f Biochemistry and Nutrition, Medical School UniversitO Libre de Bruxelles, B-IO00 Brussels, Belgium and *Department of Medicine, Medical School Tulane University, New Orleans, LA 70112 R e c e i v e d 26 M a y 1987 ROBBERECHT, P., M. WAELBROECK, P. DE NEEF, J. C. CAMUS, D. H. COY AND J. CHRISTOPHE. Pharmacological characterization of V1P receptors in human lung membranes. PEPTIDES 9(2) 33%345, 1988.--The ability of VIP, PHI, secretin, helodermin, and seven N-terminally D-amino monosubstituted V1P and PHI analogs to occupy (125I)iodo-VIPlabeled receptors and to activate adenylate cyclase was tested on human lung membranes purified by the method of Schachter et al. Best fitted Kd, Kact and % of max. values suggested the coexistence, in near equal proportions, of two classes of VIP-preferring binding sites coupled to adenylate cyclase that showed similar decreasing affinity for: VIP > (D-AIa4)-VIP > (D-Asp3)-VIP = (D-Ser2)-VIP > (D-His')-VIP > PHI > (D-Phe~)-VIP > (D-Phe~)-VIP. (D-Arg2)-VIP was a non-selective agonist. A third receptor type, coupled to adenylate cyclase and showing high affinity for secretin and helodermin but not for VIP, was also detected. VIP receptors
Human lung membranes
preparations. To achieve this aim, we compared the ability of VIP, PHI, secretin, helodermin and N-terminally modified VIP and PHI analogs to occupy VIP receptors and to activate adenylate cyclase in purified human lung membranes.
T H E neurotransmitter VIP (Vasoactive Intestinal Peptide) is a physiological regulator of the respiratory tract that could serve as a helpful pharmacological agent in pulmonary disease when applied exogenously. Indeed: (1) nerve endings containing VIP are associated to airways smooth muscle, submucosal glands, and blood vessels in the trachea and stem bronchi [11,33]; (2) VIP nerves are closely related with bronchial smooth muscle and with the tunica adventitia of pulmonary arteries [9]; (3) specific VIP receptors are identiffed by their binding characteristics in pulmonary membranes from several mammalian species including man [4, 10, 15, 23, 31] and by their morphological properties in alveolar capillaries, pulmonary smooth muscle, alveolar walls and bronchial epithelium in rat [15] and man [16]; (4) pulmonary VIP receptors are efficiently coupled to adenylate cyclase [4, 22, 23, 32]; (5) VIP may mediate the nonadrenergic relaxation of guinea pig tracheal smooth muscle [17]; (6) VIP relaxes isolated guinea pig tracheal muscle dose-dependently [2]; (7) VIP causes bronchodilation and protects against histamine-induced bronchoconstriction in asthmatic patients [18]; (8) local levels of VIP increase in rat lung exposed to asbestos [7], and (9) VIP receptors are found in human lung tumor cells in culture [13]. It was thus felt of interest to examine whether VIP analogs (synthetic or natural) might act as superagonist or antagonist on (sub)classes of VIP receptors in human lung
METHOD
Human Lung Specimens Normal lung tissue from 6 male patients (35-55 years old) suffering from bronchopulmonary cancer was obtained during segmentectomy or lobectomy. As soon as possible after tissue devascularization, the tumor was dissected out and the remaining healthy tissue was frozen on dry-ice then stored in liquid nitrogen until use.
Preparation of Lung Membranes After thawing of the tissue, we used the method described by Schachter et al. [31] with only one modification consisting in the omission of leupeptin from the homogenizing medium. Aliquots of crude membranes were stored in liquid nitrogen.
Binding Studies on Lung Membranes Synthetic VIP was radioiodinated by the chloramine-T technique and purified as previously described [3,24]. (~zsI)iodo-VIP with specific radioactivity of 200/~Ci//~g and
'Aided by Grant 3.4571.85 from The Fonds de la Recherche Scientifique M6dicale (Belgium), Grants ROI-AM-17010-9 and AM-31670 from the National Institutes of Health (USA) and a Grant from the Queen Elisabeth Medical Foundation (Belgium). ZRequests for reprints should be addressed to J. Christophe, Department of Biochemistry and Nutrition, Medical School, Universit6 Libre de Bruxelles, Boulevard de Waterloo 115, B-1000 Brussels, Belgium.
339
340
ROBBERECHTETAL.
LUNG
® 100-
~I~. = ~ , ~ , f -
Helodermin
"'~,.~ X N~x , ~,
\\{Secret in
50O Z D 0 cn
u~cu ~--I
O"
r~
,
,
,
,
• ""-
a.i
--
a
--
!
!
®
is
~,~-.,
(D-ARG2) -rIP .~
(D-PHE21-VlP
\
\ \ /
co
-11
-10
-9
-8
-7
-6
-5
[PEPTIDE CONCENTRATION](10g M) FIG. 1. Inhibition of (l~al)iodo-VIP binding to human lung membranes by VIP (©), PHI (A), secretin (0) and helodermin (&) (upper panel A) and by VIP (©), (D-AIa4)-VIP (T), (D-Asp3)VIP (A), (D-Ser2).VIP (A), (D-His~)-VIP (0), {D-Phe2)-VIP ([~), (D-Arg~)-VlP (m) and (D-Phe4)-PHI (V) (lower panel B). The results were expressed as a percentage of tracer specifically bound in the absence of unlabeled peptide and were the means of four experiments performed in duplicate.
0.4 iodine atom/molecule on an average was stored at -20°C. B i n d i n g o f (12~l)iodo-VIP. Binding studies were carried out in a total volume of 0.120 ml in a 50 mM Tris/maleate buffer (pH 7.4) containing 5 mM MgC12, 0.5 mg/ml bacitracin, 100 kallikrein inhibitor unit/ml Trasylol, 1% bovine serum albumin, 20-30 pM (125I)iodo-VIP, increasing concentrations of unlabeled peptide and 10 to 15/zg tissue protein. Incubation was conducted at 37°C for 10 min, i.e., until binding equilibrium was attained. The incubation was terminated by rapid filtration under reduced pressure through Whatman GF/C filters (Maidstone, UK) pre-soaked overnight in 0.5% polyethyleneimine [31] in order to reduce non-specific tracer binding. The filters were washed 3 times with 3 ml ice-cold buffer containing 50 mM Tris, 0.5 mM EDTA, 0.2% bovine serum albumin (pH 7.4) and counted in a Gamma Spectrometer. Non-specific binding was defined as ('~5I)iodo-VIP bind-
ing in the presence of 1/zM VIP and never exceeded 20% of the total bound radioactivity. Adenylate
Cyclase Assay
Adenylate cyclase activity was determined according to the Salomon et al. procedure [30] as previously detailed [32]. T e n / z M GTP was included in all assays. Cyclic A M P production was always linear and proportional to the amount of tissue protein added. Peptides and Chemicals
VIP, PHI, VIP analogs and PHI analogs were assembled by solid-phase methods and purified by ion-exchange and partition chromatography as previously described [6]. Natural helodermin was prepared as detailed in [34]. Synthetic
VIP RECEPTORS IN H U M A N L U N G
K d
341
TABLE 1 VALUES AND PROPORTIONS OF ('2~I)IODO-VIPLABELED RECEPTORS FOR OCCUPANCY BY VIP AND VIP ANALOGSIN HUMAN LUNG MEMBRANES
Peptide
Kdl (nM)
% of Sites
VIP Secretin PHI Helodermin (D-Hisl)-VIP (D-Ser2)-VIP (D-Asp'~)-VIP (D-Ala4)-VIP (D-Phe2)-VIP (D-Arg2)-VIP (D-Phe4)-VIP
0.08 3 3 1 1 0.7 0.5 0.1 10 200 100
43 8 43 10 43 43 43 40 40 100 40
Kd2 (nM) 2.5 300 100 50 20 20 20 10 100 . . 2000
% of Sites 57 32 57 30 57 57 57 60 60 . 60
Kd3 (nM) - -
% of Sites - -
1000 -2000 ------
60 -60 ------
--
--
.
The values were deduced from the best fit of data presented in Fig. 1. porcine secretin was a generous gift from D. W. Krnig (Hoechst Aktiengesellschaft, Frankfurt, FRG). Carrier-free N a 1~5I (600-800 mCi/ml) and (8-all)cyclic AMP (20-30 Ci/mole) were purchased from Amersham International (Amersham, Bucks, GB). (c~-3ZP)ATP (30 Ci/mmole) was from New England Nuclear (Boston, MA). Bovine serum albumin (fraction V), bacitracin, ATP, cyclic AMP, GTP, and polyethyleneimine were obtained from Sigma Chemical Co. (St Louis, MO). Kallikrein inhibitor (Trasylol) was a gift from Bayer (Brussels, Belgium). All other reagents were of the highest purity degree available. All peptides were dissolved in 0.01 M sodium phosphate buffer (pH 7.5), enriched with 0.9% sodium chloride, 0.1% sodium azide and 0.05% Tween 20. This buffer reduced peptide binding to stock tubes and, when diluted in the media used for binding studies or adenylate cyclase assay, was uninfluential on the parameters investigated yet minimized peptide binding to the test tubes. The results of binding studies were analyzed assuming that the ligand recognized one, two or three subclasses of receptors, using the program of Richardson and Humrich [21]. RESULTS
General Characteristics of (125I)iodo-VIPBinding The present data fully confirmed previous results from Schachter et al. [31] showing that tracer binding was rapid at 37°C: steady state binding was attained after 5 min already then maintained for at least 15 more min. This binding was rapidly and totally reversible in the presence of GTP (data not shown). During the first 15 min of incubation, tracer degradation did not exceed 15% as measured by trichloroacetic acid precipitation (it is acknowledged that this relatively low percentage probably represented a minimal figure of tracer degradation). The cations magnesium (5 to 20 mM) and manganese (0.25 to 2.5 mM) but not calcium (0.5 to 5.0 mM) increased tracer binding to a maximum of 50% whereas EDTA and EGTA were without effect (data not shown).
Specificity of VIP Binding Sites and Structural Requirements for Receptor Occupancy The capability of peptides to occupy VIP binding sites
was evaluated by their ability to inhibit (125I)iodo-VIP binding: the data shown in Fig. 1A were expressed in % of tracer bound in the presence of increasing concentrations of unlabeled peptide. VIP binding sites were highly specific as PHI, secretin and helodermin were, respectively, 30-, 500- and 600-fold less potent than VIP based on the peptide concentration required for 50% tracer binding inhibition (Kd). The N-terminal part of VIP was critical for receptor recognition: a systematic substitution of the first four amino acids by their D-isomer led to compounds 2- to 8-fold less potent than VIP, the potency decreasing with D-substitutions closer to the N-terminus. Substitution of serine in position 2 by a large D-Phe or D-Arg residue reduced the peptide potency even more: 50- and 200-fold, respectively. (D-Phe4)-PHI was 60fold less potent than PHI itself.
Evidence for the Presence of Subclasses of VIP Binding Sites All inhibition c u r v e s - - e x c e p t the curve for (D-Arg2)V I P - - s h o w e d a Hill slope factor smaller than one, suggesting that more than one class of binding sites was occupied by tracer and/or unlabeled peptide. The use of a curve fitting program [21] allowed to determine the proportions of sites and corresponding K d values for the peptides capable to occupy these sites. The results in Table 1 indicate the best fit after considering, successively, the potential existence of one, two or three binding sites. The average proportions of high affinity and low affinity binding sites for VIP and PHI analogs were 42% and 58%, respectively. The selectivity, i.e., the capability of peptides to discriminate between both classes of binding sites was inferred from the ratio of Kd values for low and high affinity sites. VIP, (D-Hisl)-VIP, (D-Ser2)-VIP, (D-Asp3)-VIP, PHI and (D-Phe4)-PHI showed comparable selectivity, ranging from 20 to 40. (D-AIa4)-VIP recognized high affinity binding sites preferentially. (DPhe2)-VIP and (D-Arg~)-VIP had a lower selectivity and (DArgZ)-VIP could even be considered as a non-selective ligand. For secretin and helodermin the best fit, when considering two classes of binding sites, gave 25% high affinity sites and 75% low affinity sites, i.e., proportions significantly different from those obtained with VIP and PHI analogs.
ROBBERECHT ET AL.
342
LUNG
® I00-
PHI . ~ vIp
.o.--'" /
/
50 Z o
~, ,~-'"
~
,
He.lodermin
I--
---'" ~ I.u
0 dl,
Iii
®
J
Secretin ,*,
,.
.
.
.
(D-HIS I) -VIP (D-ALA4) -VIP ~ , . ~ ~
100-
iii I-J >.. z iii i-i
tpPO
50
" ~ o-'"
.,"
-I0
//7/
• -""
/r/q-
-9
-8
~ ~'~'-+,7
sp31-/
-7
/
Co- HE21-vI
-6
-5
[PEPTIDE CONCENTRATION] (log M) FIG. 2. Dose-effect curves of activation of human lung membrane adenylate cyclase by VIP (O), PHI (A), secretin (O), helodermin (&) (upper panel A) and by VIP (O), (D-Ala~)-VIP(T), (D-Asp3)-VIP (A), (D-Ser2)-VIP (A), (D-Hisl)-VIP (0), (D-Phe2)-VIP (rq), (D-Arg2)-VIP (11) and (D-Phe4)-PHI (V) (lower panel B). The results were expressed as a percentage of adenylate cyclase activation observed in the presence of 10/xM VIP after subtraction of the basal value and were the means of four experiments performed in duplicate. The mean_+SEM basal activity was 58.3---4.0 pmol cyclic AMP.min l.mg protein -t and the mean±SEM activity in the presence of 10/zM VIP was 138.2±8.0 pmol cyclic AMP.min-Umg protein 1.
This fitting improved, however, when a three classes model was considered (see Table 1 and Discussion).
Adenylate Cyclase Activation All peptides tested were able to stimulate adenylate cyclase but with variable efficacy (Figs. 2A and 2B). (D-Arg2)VIP and (D-Phe2)-VIP, although active, were partial agonists only with an intrinsic activity of 2 7 - 2 ~ as compared to that of VIP. Most curves were compatible with an activation process mediated by more than one class of binding sites. By analogy with the data obtained in binding studies, the results were fitted according to models with one, two or three classes of functional receptors. For the VIP dose-response curve, the best fit was ob-
tained considering the coexistence of three receptor classes. If, for most VIP analogs, a fitting of similar quality was obtained for two or three classes, we favored the latter hypothesis after our analysis of the VIP dose-effect curve. Table 2 gives the calculated Kact for each class and the corresponding maximum stimulation of adenylate cyclase (expressed in % of total activity obtained with I0/zM VIP).
DISCUSSION This study was initially undertaken to search for VIP and PHI analogs that could act as VIP superagonist or antagonist on human lung membranes. In this respect, our data were negative as VIP was the most potent and efficient peptide at
VIP RECEPTORS IN H U M A N L U N G
343
TABLE 2 ACTIVATIONOF ADENYLATECYCLASEBY VIP AND VIP ANALOGSIN HUMAN LUNG MEMBRANES
Peptide
Kactl (nM)
% of Max. Activity
VIP Secretin PHI Helodermin (D-Hisl)-VIP (D-Ser2)-VIP (D-Asp3)-VIP (D-AIa4)-VIP (D-Phe2)-VIP (D-ArgZ)-VIP (D-Phe4)-PHI
0.08 6 5 3 3 1 0.5 0.1 200 300 100
20 28 14 20 35 10 15 10 27 28 20
Kact2 (nM) 2.5 3000 100 200 40 30 20 10 . . . . 1000
% of % of Max. Kact3 Max. Activity (nM) Activity 50 45 86 55 65 70 55 55 . . 50
300 --2000 -3000 1000 300
30 --20 -15 15 25
--
--
. .
The best estimated values for Kact and maximum activation (in % of that observed with 10/zM VIP) were derived from data presented in Fig. 2.
hand and as analogs like (D-Phe2)-VIP and (D-Arg2)-VIP, that are devoid of activity or behave as VIP antagonist in rat tissues [27], were able to stimulate human lung adenylate cyclase. The results gave, nevertheless, valuable information on the structural requirements for lung VIP receptor occupancy and subsequent adenylate cyclase activation: they substantiated the importance of the N-terminal portion of VIP for receptor site recognition, due probably to the presence of a beta bend in the 1-4 region [19]. They also demonstrated the highly unfavorable role of increasing hydrophobicity in position 2 and 4 and of having a basic D-amino acid residue in position 2. They also illustrated the key role of histidine in position 1, confirming previous studies on rat tissues and human heart membranes [25]. Such conclusions cannot be extended to other members of the VIP family: for example, (D-Phe4)-GRF(1-29)-NH2, when interacting with VIP receptors in both rat pancreatic and liver membranes, is more potent than G R F itself [28] and (D-Phe4)-glucagon, when binding to glucagon receptors in rat liver, is a glucagon superagonist ([19] and unpublished results). Although not substantiated in the present work, it is also highly likely that the C-terminal part of the molecule is involved in receptor recognition as PHM is found by Schachter et al. to be threefold more potent than PHI in human lung membranes [31]. Human lung VIP binding sites were heterogeneous based on binding studies as well as on adenylate cyclase activation data. The detailed computerized study of this heterogeneity was made possible by the high feasibility of binding (high specific 125I-VIP binding, low non-specific binding, and low tracer degradation due to the short time period needed to reach steady state binding). VIP, VIP analogs (except (DArg2)-VIP), PHI and the PHI analog (D-Phe4)-PHI discriminated two classes of binding sites: a high affinity class (42% of labeled sites) and a low affinity class (58% of labeled sites). Their K d values were in line with those obtained on rat lung membranes [20] but high affinity VIP receptors were proportionately more numerous in man than in rat lung membranes. It is clear, however, that the percentage of sites given for the two classes of receptors in both man (the present data) and rat [20] is indicative only as it was not demon-
strated that (12~I)iodo-VlP (in all possible forms) binds unspecifically (i.e., with the same affinity) to all receptor subclasses. Both classes of receptors in our preparations were coupled to adenylate cyclase and the correlation between Kd values for binding and Kact values for enzyme stimulation was good. Structural requirements for receptor occupancy and subsequent adenylate cyclase activation were quite similar in both classes as the order of peptide potency was comparable: VIP > (D-Alaa)-VIP > (D-AspZ)-VIP = (D-Ser2)VIP > (D-Hisl)-VIP > PHI > (D-Phez)-vIP > (D-Phe4)-PHI. (D-Arg2)-VIP was a non-selective partial agonist whose capability to activate adenylate cyclase could not be correlated with a given receptor class. The correlation between binding data and adenylate cyclase activation and Hill slope factors smaller than one for all VIP agonists (except (D-ArgZ)-VIP) allow to conclude that the heterogeneity of binding sites reflected the coexistence of two distinct classes of receptors rather than the presence of two receptor states that could be discriminated by agonists but not by antagonists. It should be noted that covalent cross-linking experiments on lung, like those on liver and intestine, also offer evidence for VIP receptor heterogeneity [5, 10, 14]. VIP and the three most potent VIP analogs ((D-Ala4)-VIP, (D-Aspa)-VIP, and (D-Ser2)-VIP) further activated adenylate cyclase at concentrations higher than those required for occupying the two sites just described (Table 2). This contribution of a third component had to be considered to obtain adequate fitting of dose-response curves (Table 2) but could not be detected in binding studies (Table 1) due probably to Kd values for this component close to the concentration of unlabeled VIP (10/zM) used for non-specific (lzsI)iodo-VIP binding determination. This third component could represent either a very low affinity VIP receptor or the feable interaction of VIP and the three most potent VIP analogs with a class of receptors having high affinity for a parent peptide. This last hypothesis is favored when considering how secretin and helodermin interacted with VIP receptors and activated adenylate cyclase. Indeed, secretin and helodermin binding data could not fit in a model involving two VIP binding sites only with relative proportions of 42% and 58%. By
344
ROBBERECHT ET AL.
contrast, the same data fitted perfectly with a three sites model as proposed in Table 1. Altogether our data in Tables 1 and 2 can be interpreted as follows: (a) secretin and helodermin interacted with high affinity secretin/helodermin receptors that were (very) poorly labeled with ('2~I)iodo-VIP. This was followed by adenylate cyclase activation (Kd, and Kact,); (b) secretin and helodermin interacted with high affinity VIP receptors (Kd2) but only the interaction of helodermin promoted detectable adenylate cyclase activation (Kactz of helodermin); (c) secretin and helodermin bound also to low affinity VIP receptors (Kd3) and this interaction of both peptides led to increased cyclic A M P production (Kact2 of secretin and Kact3 of helodermin). This interpretation is in good agreement with previous data on secretin and VIP receptors: secretin binds to both secretin receptors and VIP receptors and with or without resulting adenylate cyclase activation in guinea pig and rat
pancreas [1, 3, 24]; helodermin, a parent peptide of VIP and secretin extracted from Gila monster venom [6,12], interacts efficiently with VIP receptors in tissues devoid of secretin receptors [26,32] and with near equal potency with secretin receptors and VIP receptors in rat pancreas [8,26]. Any efficient interaction of PHI with secretin receptors would be difficult to demonstrate due to the low efficacy of secretin itself (Fig. 2). In conclusion, human lung membranes possess multiple VIP and secretin receptors coupled to adenylate cyclase. VIP is likely to cross-react with secretin receptors and secretin with VIP receptors. Helodermin recognized all receptors. The N-terminal part of VIP was essential for receptor site recognition. We were unable to find a superagonist or an antagonist for VIP receptors among the seven analogs monosubstituted in the first 4 amino acids of VIP and PHI that were tested.
REFERENCES 1. Bissonnette, B. M., M. J. Collen, H. Adachi, R. T. Jensen and J. D. Gardner. Receptors for vasoactive intestinal peptide and secretin on rat pancreatic acini. A m J Physiol 246: G710-G717, 1984. 2. Christofides, N. D., Y. Yiangou, P. J. Piper, M. A. Ghatei, M. N. Sheppard, K. Tatemoto, J. M. Polak and S. R. Bloom. Distribution of peptide histidine isoleucine in the mammalian respiratory tract and some aspects of its pharmacology. Endocrinology 115: 1958-1963, 1984. 3. Christophe, J., J. P. Conlon and J. D. Gardner. Interaction of porcine vasoactive intestinal peptide with dispersed pancreatic acinar cells from the guinea pig: binding of radioiodinated peptide. J Biol Chem 251: 4629-4634, 1976. 4. Christophe, J., P. Chatelain, G. Taton, M. Delhaye, M. Waelbroeck and P. Robberecht. Comparison of VIP-secretin receptors in rat and human lung. Peptides 2: Suppl 2, 113-118, 1981. 5. Couvineau, A. and M. Laburthe. The rat liver vasoactive intestinal peptide binding site. Molecular characterization by covalent cross linking and evidence for differences from the intestinal receptor. Biochem J 225: 473-479, 1985. 6. Coy, D. H. and J. Gardner. Solid-phase synthesis of porcine vasoactive intestinal peptide. Int J Pept Prot Res 15: 73-78, 1980. 7. Day, R., I. Lemaire, P. Mercier, H. Beaudoin and S. Lemaire. Asbestos-related increase in pulmonary levels of vasoactive intestinal peptide (VIP). Life Sci 33: 1869-1876, 1983. 8. Dehaye, J. P., J. Winand, C. Damien, F. Gomez, P. Poloczek, P. Robberecht, A. Vandermeers, M. C. Vandermeers-Piret, M. Stirvenart and J. Christophe. Receptors involved in helodermin action on rat pancreatic acini. Am J Physiol 251: 599-602, 1986. 9. Dey, R. D., W. A. Shannon and S. I. Said. Localization of VIP-immunoreactive nerves in airways and pulmonary vessels of dogs, cats and human subjects. Cell Tissue Res 220: 231-238, 1981. 10. Dickinson, K. E. J., M. Schachter, C. M. M. Miles, D. H. Coy and P. S. Sever. Characterization of vasoactive intestinal peptide (VIP) receptors in mammalian lung. Peptides 7: 791-800, 1986. 11. Ghatei, M. A., M. N. Sheppard, D. J. O'Shaughnessy, T. E. Adrian, G. P. McGregor, J. M. Polak and S. R. Bloom. Regulatory peptides in the mammalian respiratory tract. Endocrinology 111: 1248--1254, 1982. 12. Hoshino, M., C. Yanaihara, Y. M. Hong, S. Kishida, Y. Katsumaru, A. Vandermeers, M. C. Vandermeers-Piret, P. Robberecht, J. Christophe and N. Yanaihara. Primary structure of helodermin, a VIP-secretin-like peptide isolated from Gila monster venom. FEBS Lett 178: 233-239, 1984.
13. Laburthe, M., C. Boissard, G. Chevalier, A. Zweibaum and G. Rosselin. Peptide receptors in human lung tumor cells in culture: vasoactive intestinal peptide (VIP) and secretin interaction with the CALU-1 and SW-900 cell lines. Regul Pept 2: 219-230, 1981. 14. Laburthe, M., B. Breant and C. Rouyer-Fessard. Molecular 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. Fur J Biochem 139: 181-187, 1984. 15. Leroux, P., H. Vaudry, A. Fournier, S. St Pierre and G. Pelletier. Characterization and localization of vasoactive intestinal peptide receptors in the rat lung. Endocrinology 114: 1506-1512, 1984. 16. Leys, K., A. H. Morice, O. Madonna and P. S. Sever. Autoradiographic localisation of VIP receptors in human lung. FEBS Lett 199: 198--202, 1986. 17. Matsuzaki, Y., Y. Hamasaki and S. I. Said. Vasoactive intestinal peptide: a possible transmitter of noradrenergic relaxation of guinea pig airways. Science 210: 1252-1253, 1980. 18. Morice, A., R. J. Unwin and P. S. Sever. Vasoactive intestinal peptide causes bronchodilatation and protects against histamine-induced bronchoconstriction in asthmatic subjects. Lancet II: 1225-1227, 1983. 19. Murphy, W. A., D. H. Coy and V. A. Lance. Superactive amidated COOH-terminal glucagon analogues with no methionine or tryptophan. Peptides 7: Suppl 1, 69-74, 1986. 20. Provow, S. and G. Veliqelebi. Characterization and solubilization of Vasoactive Intestinal Peptide receptors from rat lung membranes. Endocrinology 120: 2442-2452, 1987. 21. Richardson, A. and A. Humrich. A microcomputer program for the analysis of radioligand binding curves and other doseresponse data. Trends Pharmacol Sci 5: 47-49, 1984. 22. Robberecht, P., P. Chatelain, P. De Neef, J. C. Camus, M. Waelbroeck and J. Christophe. Presence of vasoactive intestinal peptide receptors coupled to adenylate cyclase in rat lung membranes. Biochim Biophys Acta 678: 76-82, 1981. 23. Robberecht, P., K. Tatemoto, P. Chatelain, M. Waelbroeck, M. Delhaye, G. Taton, P. De Neef, J. C. Camus, D. Heuse and J. Christophe. Effects of PHI on vasoactive intestinal peptide receptors and adenylate cyclase activity in lung membranes. A comparison in man, rat, mouse and guinea pig. Regul Pept 4: 241-250, 1982. 24. Robberecht, P., M. Waeibroeck, M. Noyer, P. Chatelain, P. De Neef, W. K6nig and J. Christophe. Characterization of secretin and vasoactive intestinal peptide receptors in rat pancreatic plasma membranes using the native peptides, secretin-(7-27) and five secretin analogues. Digestion 23: 201-210, 1982.
VIP R E C E P T O R S I N H U M A N L U N G
25. Robberecht, P., M. Waelbroeck, J. C. Camus, P. De Neef, D. Coy and J. Christophe. Effects of His I modifications on the ability of vasoactive intestinal peptide to stimulate adenylate cyclase from rat and human tissues. Peptides 5: 877-881, 1984. 26. Robberecht, P., M. Waelbroeck, J. P. Dehaye, J. Winand, A. Vandermeers, M. C. Vandermeers-Piret and J. Christophe. Evidence that helodermin, a newly extracted peptide from Gila monster venom, is a member of the secretin/VIP/PHI/family of peptides with an original pattern of biological properties. FEBS Lett 166: 277-282, 1984. 27. Robberecht, P., D. H. Coy, P. De Neef, J. C. Camus, A. Cauvin, M. Waelbroeck and J. Christophe. Interaction of vasoactive intestinal peptide (VIP) and N-terminally modified VIP analogs with rat pancreatic, hepatic and pituitary membranes. Eur J Biochem 159: 45-49, 1986. 28. Robberecht, P., M. Waelbroeck, D. Coy, P. De Neef, J. C. Camus and J. Christophe. Comparative structural requirements of thirty GRF analogs for interaction with GRF- and VIP receptors and coupling to adenylate cyclase in rat adenopituitary, liver and pancreas. Peptides 7: Suppl 1, 53-59, 1986.
345
29. Robberecht, P., N. Yanaihara, C. Yanaihara, P. De Neef, J. C. Camus, A. Vandermeers, M. C. Vandermeers-Piret and J. Christophe. Interaction of synthetic N- and C-terminal fragments of helodermin with rat liver VIP receptors. Peptides 7: Suppl l, 79-82, 1986. 30. Salomon, Y., C. Londos and M. Rodbell. A highly sensitive adenylate cyclase assay. Anal Biochem 58: 541-548, 1974. 3 I. Schachter, M., K. E. J. Dickinson, C. M. Miles and P. S. Sever. Characterisation of a high-affinity VIP receptor in human lung parenchyma. FEBS Lett 199: 125-129, 1986. 32. Taton, G., M. Delhaye, J. C. Camus, P. De Neef, P. Chatelain, P. Robberecht and J. Christophe. Characterization of the VIPand secretin-stimulated adenylate cyclase system from human lung. Pflugers Arch 391: 178--182, 1981. 33. Uddman, R., J. Alumets, O. Densert, R. Hakanson and F. Sundler. Occurrence and distribution of VIP nerves in the nasal mucosa and tracheo-bronchial wall. Acta Otolaryngol (Stockh) 86: 443-448, 1978. 34. Vandermeers, A., M. C. Vandermeers-Piret, P. Robberecht, M. Waelbroeck, J. P. Dehaye, J. Winand and J. Christophe. Purification of a novel pancreatic secretory factor (PSF) and a novel peptide with VIP- and secretin-like properties (helodermin) from Gila Monster venom. FEBS Lett 166: 273-276, 1984.