Eur. J. Biochem. 193, 149-154 (1990) FEBS 1990
c;
Secretin receptors in the neuroglioma hybrid cell line NGl08-15 Characterization and regulation of their expression Denis GOSSEN, Michele TASTENOY, Patrick ROBBERECHT and Jean CHRISTOPHE Department of Biochemistry and Nutrition, Medical School, Universite Libre de Bruxelles, Belgium (Received March 19, 1990) - EJB 90 0304
Secretin receptors in membranes from the neuroblastoma-glioma hybrid cell line NG108-15 were investigated by l 2 ’I-secretin binding and adenylate cyclase activation. On both parameters the corresponding relative potencies of parent peptides were, respectively: secretin > helodermin > peptide histidine isoleucinamide = vasoactive intestinal peptide. With secretin analogs and secretin fragments, the order of potency for binding was: secretin = [Val’lsecretin > [Ala2]secretin = [Ala’ ‘]secretin > [Ala4,Va15]secretin> [Ala4]secretin > [~-Phe~]secretin > [DPhe2]secretin = secretin (2-27) > secretin (3-27) > secretin (7-27). Also, on adenylate cyclase, [DPhe4]secretin, [~-Phe~]secretin, secretin (2 - 27) and secretin (3 - 27) were partial agonists while secretin (7 -27) was ineffective. The differentiating agent N6,2’-0-dibutyryladenosine 3’,S’-monophosphate (1 mM) increased the density of secretin receptors and secretin-stimulated adenylate cyclase activity after a lag period of 4 h. After incubation for 24 h, receptor number and enzyme activity were increased 4- and 3-fold, respectively. These effects were inhibited totally by 1 pg/ml cycloheximide and halved by 5 pg/ml actinomycin D. They were mimicked by 1 mM sodium butyrate but were not reproduced by either 8-bromoadenosine 3’,5‘-monophosphate or the phosphodiesterase inhibitor rac-4-(3-Butoxy-4-methoxybenzyl)-2-imidazolidinone. For a long time after its discovery, the role of secretin was considered as limited to the gastrointestinal tract and included the neutralization of duodenum acidification after gastric emptying (through a stimulation of pancreatic bicarbonate flux) [l], the inhibition of gastric acid secretion (by hindering the postprandial release of gastrin) [2] and the provocation of pepsinogen discharge from gastric chief cells [3]. Recent studies have demonstrated that secretin is present in the nervous system of several mammalian species [4 - 81. In line with this observation, secretin is able to affect neuronal parameters. It activates tyrosine 3-monooxygenase (EC 1.14.16.2)in the rat superior cervical ganglia [9] and in the rat PC12 cell line [lo]. In the hypothalamus, secretin increases dopamine turnover [l 11 and reduces prolactin and luteinizing hormone secretion [4, 121. Apart from this, secretin activates adenylate cyclase and elevates cAMP levels in cultured mouse brain neurons and glial cells [13, 141, mouse neuroblastoma N18TG2 cells [I51 and the neuroblastoma x glioma NG10815 hybrid [13,15,16]. These data strongly suggest that secretin
Correspondence to J. Christophe, Department of Biochemistry and Nutrition, Medical School, Universite Libre de Bruxelles, Boulevard de Waterloo 115 , B-1000 Brussels, Belgium Abbreviations. Bt2cAMP, N6,2‘-O-dibutyryladenosine 3‘,5‘monophosphate; Br’cAMP, 8-hromoadenosine 3’,5’-monophosphate; PG E1, prostaglandin El ; PHI, peptide histidine isoleucinamide; VIP, vasoactive intestinal peptide; p[NH]ppG: guanosine 5’-[/1’,y-imido]triphosphate; Ro 20-1724, rac-4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone;KD,concentration required for half-maxima1 occupancy of binding sites with tracer; ICSOrconcentration required for half-maximal inhibition of tracer binding; EC50, concentration exerting half-maximal stimulation of adenykate cyclase; Gs, guanosine-nucleotide-binding stimulatory protein.
or a secretin-like peptide could operate as a neuroregulatory factor in the mammalian central and peripheral nervous systems through receptors positively coupled to adenylate cyclase. In agreement, the presence of secretin binding sites has been demonstrated in membranes from specific rat brain areas [I71 and in mouse neuroblastoma N108TG2 membranes ~151. The NG108-15 cell line is a hybrid, obtained by the fusion of mouse neuroblastoma (N18TG2) and rat glioma (C6BU1) cells, that expresses a variety of receptors including secretin [I 61, muscarinic, a,-adrenergic, opioid and prostaglandin receptors [18]. The secretin receptors were not demonstrable in parent rat glioma (C6BU1) cells, suggesting their origin in the mouse neuroblastoma component of the hybridoma [I 51. In NG108-IS cells, the presence of secretin receptors was documented only by membrane adenylate cyclase activation and cAMP elevation in intact cells. We report here (a) the pharmacological characterization of these secretin receptors based on 1251-secretinbinding and adenylate cyclase coupling, using several analogues and fragments of secretin, and (b) the regulation (amplification) of the expression of these receptors by N6,2’-0-dibutyryladenosine 3’,5’-monophosphate (Bt,cAMP) and butyrate. MATERIALS AND METHODS Culture
NG108-15 cells (passage number 26) were kindly donated by Dr J. M. Maloteaux (Louvain, Belgium). The experiments presented in this paper were performed employing cells of passage numbers 30-SO. These cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 10%
150 (by vol.) foetal calf serum. This medium was further enriched withO.l mM hypoxanthine, 1 pM aminopterin, 16 pM thymidine and antibiotics (100 U/m1 penicillin streptomycin) [18]. For propagation, cells were dislodged from the flasks by firm tapping in order to release them into the medium. Cell medium was changed every 2-3 days. In case of cell pretreatment, the culture medium of preconfluent cells was renewed with fresh medium supplemented with the drug to be tested.
37‘C by adding 0.5 ml 0.5% sodium dodecyl sulfate solution containing 0.5 mM ATP, 0.5 mM CAMP and 20000cpm [8-3H]cAMP. CAMPwas separated from ATP by t w o successive chromatographies on Dowex 50-WX8 and neutral alumina. Under all conditions tested, CAMPproduction was linear during the incubation period and proportional to the amount of protein added.
Memhrane pwparurioii
Protein determination was measured according to Lowry et al. [23] using bovine serum albumin as standard.
+
NGIOX-15 cells wese harvested by agitation, pelleted by centrifugation at 100 x g for 5 rnin at 20”C, washed in the standard culture medium (without supplement), lyzed in hypotonic 1 mM NaHCO, (pH 7.0) then quickly frozen in liquid Nz. After thawing, the lyzate was centrifuged at 400 x g for 10 min at 4’C. The resulting supernatant was centrifuged at 10000 x g for 15 min, and the pellet was washed twice in 1 mM NaHC03. Thc final pellet was resuspended in 1 mM NaHCO, and immediately used for binding assay and adenylate cyclase assay. Rrrdioioditiut ion 0f sccretin ’
Synthetic porcine secretin was radioiodinated at pH 8.5 according to Jensen et al. [19] with minor modifications [20]. 1251-secretinwas separated from free 251by cellulose adsorption and elution with 12% bovine serum albumin. This ligand, radiolabelled at the level of N-terminal histidine, corresponded to a single homogeneous peak after radio-HPLC on a yBondapak C1 column (Waters, Milford, USA). Its specific radioactivity was 1.5 mCiinmol.
’
25
I-srcrrtin hitiding .studies
lZ51-secretinbinding to NGlOX-15 membranes was carried out as described previously [21] in 50 mM Tris/HCl pH 7.4 containing 5 mM MgCI2,0.5 mg/ml bacitracin, 100 kallikrein inhibitor unitsjml Trasylol, 1O h bovine serum albumin, 25 60 pM ‘251-secretin,increasing concentrations of unlabelled peptide and 100 pg membrane protein in a total volume of 120 pl. Incubations were conducted at 37°C and terminated by adding 2 ml ice-cold 50 mM sodium phosphate pH 7.4 to each sample, followed by rapid filtration through glass-fiber filters (GF/C; Whatman, Maidstone, Kent, UK) presoaked for 24 h in 0,1% poly(ethy1eneimine) in order to reduce nonspecific binding. The filters were rinsed three times with 2 ml of the same buffer and their radioactivity was measured. Nonspecific binding was determined in the presence of 1 pM unlabelled secretin and accounted for approximately 25% of total binding. Specific binding was defined as total binding minus non-specific binding and always represented less than 25% of the total radioactivity offered. Adenyla t e cycluse ~issuy Adenylate cyclase activity was determined according to the procedure of Salonion et al. [22]. Membrane protein (20 30 pg) was incubated in a total volume of 60 p1 containing 0.5 mM [a-”PIATP. 10 pM GTP, 5 mM MgCI2, 0.5 mM EGTA, 1 mM CAMP,0.5 m M theophylline, 10 mM phosphoenolpyruvate, 30 pg/mI pyruvate kinase and 30 mM Tris/HCl at a final pH of 7.5 [XI. The reaction was initiated by addition of membranes and was terminated after a 15-min incubation at
Protein determination
Chemicals [~-Phe’]Secretin and [~-Phe~]secretin, synthetic porcine vasoactive intestinal peptide (VIP) and porcine peptide histidine isoleucinamide (PHI) were generous gifts from D r D. H. Coy (Tulane University, New Orleans, LA, USA). Synthetic porcine secretin fragments and other secretin analogues were kindly given by Dr W. Konig (Hoechst Aktiengesellschaft, Frankfurt/Main, FRG). Synthetic helodermin came from Novabiochem (Laufelingen, Switzerland). Carrier-free NalZ51 (IMS 300, 600-800 mCi/ml) and [8-3H]cAMP (20 - 30 Ci/mmol) were purchased from Amersham International (Amersham, Bucks, UK) and [ z - ~ ~ P I A T P (30 Ci/mmol) from New England Nuclear (Boston, MA, USA). Bovine serum albumin (fraction V), bacitracin, phosphoenolpyruvate, pyruvate kinase, CAMP, GTP, ATP, R-bromoadenosine 3’,5’-monophosphate (Br’cAMP), prostglandin El (PGE1), isobutyrate, butyrate, 3-hydroxybutyrate were purchased from Sigma Chemical Co (St Louis, MO, USA). Kallikrein inhibitor (Trasylol) was a gift from Bayer (Leverkusen, FRG), forskolin was obtained from Calbiochem (La Jolla, CA, USA), rac-4-(3-butoxy-4-methoxybenzyl)-2iinidazolidinone (Ro 20-1 724) from Koche S.A. (Brussels, Belgium) and guanosine 5’-[/l,y-imido]triphosphate (p[NH]ppG) from Boehringer-Mannheim (Brussels, Belgium). Foetal calf serum and medium for cell culturcs were from Gibco Europe (Gent, Belgium). All other reagents were of the highest grade available. RESULTS Characteristics of 251-secretinbinding sites Specific 12’I-secretin binding to NG108-15 membranes reached a steady state after 30 rnin at 37 C (data not shown). Tracer bound after 30 min dissociated with a tl,z of 9 rnin as studied by isotopic dilution. This t l 1 2was transiently reduced to 2 min by addition of 10 pM GTP (Fig. 1). Saturation curves were obtained by incubating increasing concentrations of 1251-secretin(60- 1200 pM). The Scatchdrd transformation (Fig. 2) was compatible with the existence of one high-affinity class of binding sites exhibiting a KD of 0.3 5 0.2 nM (mean f SEM of four determinations from different membrane preparations). Competition curves were performed by incubating membranes for 30 min in the simultaneous presence of a 60 pM tracer concentration and increasing concentrations of unlabelled peptide. Unlabelled porcine secretin inhibited tracer bindingwith an ICso of2 nM (Fig. 3). Helodermin was tenfold less potent than secretin in inhibiting tracer displacement; PHI and VIP were barely effective. The selectivity of these secretin receptors was further evaluated
151 1%)
100-
1000 3 0
z
75-
m
z
w
&
50-
V
50-
0 W
w Ln
2
0 z 0 3
0-
m U
30-
+
U U
0 U Y
I
l o g [ P e p t i d e ] /M 5
0
15
10 T I M E bin)
Fig. 1. Kinetics of dissociation of prebound '251-secretinfrom control NG108-15 membranes. After a 30-min preincubation with tracer, 0.1 pM unlabelled secretin was added in the absence (0)or presence (a) of 10 pM GTP. The radioactivity remaining bound was calculated, after subtraction of non-specific binding, and estimated as a percentage of tracer bound before addition of unlabelled secretin. The results, presented on a semi-log scale, were the means of experiments performed in duplicate on three separate membrane preparations
Fig. 3. Doseleifect curves of inhibition of '251-secretin binding (upper panel) and adenylate cyclase activation (lower panel} in control NG108-15 membranes by secretin (a),helodermin ( O ) , PHI (A) and VIP (B). Binding data are expressed as the percentage of tracer specifically bound in the absence of unlabelled peptide and are the means of two experiments performed in duplicate. Data on adenylate cyclase activation, in the presence of 10 pM GTP, are expressed as a percentage of the rate of production of cAMP/mg protein in the presence of 1 pM secretin (after subtraction of the basal unstimulated value) and are the means of two experiments in duplicate
\
4001
[ALA4, VAL5] S e c r e t in
0 10
20
30 40 50 60 70 E ifmol.mg protein-11
80
90
Fig. 2. Scatchard representation of' a saturation curve obtained after incubating increasing '251-secretin concentrations with control (a) NG108-15 cell membranes and membranes f r o m cells pretreated with 1 mM B t 2 c A M P (0) or 1 mM butyrate ( A } , f o r 24 h. Tracer binding was conducted for 30 min at 37°C. Non-specific binding was determined in the presence of 0.1 pM secretin. The results were representative of three others performed in duplicate
with secretin analogues and secretin fragments. The results in Figs 4 and 5 show the following decreasing order of potency: porcine secretin = [Val'lsecretin > [Alaz]secretin = [Ala' ']secretin > [Ala4,Va15]secretin> [Ala4]secretin > [DPhe4]secretin > [~-Phe']secretin = secretin (2 - 27) > secretin (3 - 27) > secretin (7 - 27).
'-10
-9
-8
-7
-6
log [ P e p t i d e ] /M
Fig. 4. Dose/
c;
Secretin receptors in the neuroglioma hybrid cell line NGl08-15 Characterization and regulation of their expression Denis GOSSEN, Michele TASTENOY, Patrick ROBBERECHT and Jean CHRISTOPHE Department of Biochemistry and Nutrition, Medical School, Universite Libre de Bruxelles, Belgium (Received March 19, 1990) - EJB 90 0304
Secretin receptors in membranes from the neuroblastoma-glioma hybrid cell line NG108-15 were investigated by l 2 ’I-secretin binding and adenylate cyclase activation. On both parameters the corresponding relative potencies of parent peptides were, respectively: secretin > helodermin > peptide histidine isoleucinamide = vasoactive intestinal peptide. With secretin analogs and secretin fragments, the order of potency for binding was: secretin = [Val’lsecretin > [Ala2]secretin = [Ala’ ‘]secretin > [Ala4,Va15]secretin> [Ala4]secretin > [~-Phe~]secretin > [DPhe2]secretin = secretin (2-27) > secretin (3-27) > secretin (7-27). Also, on adenylate cyclase, [DPhe4]secretin, [~-Phe~]secretin, secretin (2 - 27) and secretin (3 - 27) were partial agonists while secretin (7 -27) was ineffective. The differentiating agent N6,2’-0-dibutyryladenosine 3’,S’-monophosphate (1 mM) increased the density of secretin receptors and secretin-stimulated adenylate cyclase activity after a lag period of 4 h. After incubation for 24 h, receptor number and enzyme activity were increased 4- and 3-fold, respectively. These effects were inhibited totally by 1 pg/ml cycloheximide and halved by 5 pg/ml actinomycin D. They were mimicked by 1 mM sodium butyrate but were not reproduced by either 8-bromoadenosine 3’,5‘-monophosphate or the phosphodiesterase inhibitor rac-4-(3-Butoxy-4-methoxybenzyl)-2-imidazolidinone. For a long time after its discovery, the role of secretin was considered as limited to the gastrointestinal tract and included the neutralization of duodenum acidification after gastric emptying (through a stimulation of pancreatic bicarbonate flux) [l], the inhibition of gastric acid secretion (by hindering the postprandial release of gastrin) [2] and the provocation of pepsinogen discharge from gastric chief cells [3]. Recent studies have demonstrated that secretin is present in the nervous system of several mammalian species [4 - 81. In line with this observation, secretin is able to affect neuronal parameters. It activates tyrosine 3-monooxygenase (EC 1.14.16.2)in the rat superior cervical ganglia [9] and in the rat PC12 cell line [lo]. In the hypothalamus, secretin increases dopamine turnover [l 11 and reduces prolactin and luteinizing hormone secretion [4, 121. Apart from this, secretin activates adenylate cyclase and elevates cAMP levels in cultured mouse brain neurons and glial cells [13, 141, mouse neuroblastoma N18TG2 cells [I51 and the neuroblastoma x glioma NG10815 hybrid [13,15,16]. These data strongly suggest that secretin
Correspondence to J. Christophe, Department of Biochemistry and Nutrition, Medical School, Universite Libre de Bruxelles, Boulevard de Waterloo 115 , B-1000 Brussels, Belgium Abbreviations. Bt2cAMP, N6,2‘-O-dibutyryladenosine 3‘,5‘monophosphate; Br’cAMP, 8-hromoadenosine 3’,5’-monophosphate; PG E1, prostaglandin El ; PHI, peptide histidine isoleucinamide; VIP, vasoactive intestinal peptide; p[NH]ppG: guanosine 5’-[/1’,y-imido]triphosphate; Ro 20-1724, rac-4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone;KD,concentration required for half-maxima1 occupancy of binding sites with tracer; ICSOrconcentration required for half-maximal inhibition of tracer binding; EC50, concentration exerting half-maximal stimulation of adenykate cyclase; Gs, guanosine-nucleotide-binding stimulatory protein.
or a secretin-like peptide could operate as a neuroregulatory factor in the mammalian central and peripheral nervous systems through receptors positively coupled to adenylate cyclase. In agreement, the presence of secretin binding sites has been demonstrated in membranes from specific rat brain areas [I71 and in mouse neuroblastoma N108TG2 membranes ~151. The NG108-15 cell line is a hybrid, obtained by the fusion of mouse neuroblastoma (N18TG2) and rat glioma (C6BU1) cells, that expresses a variety of receptors including secretin [I 61, muscarinic, a,-adrenergic, opioid and prostaglandin receptors [18]. The secretin receptors were not demonstrable in parent rat glioma (C6BU1) cells, suggesting their origin in the mouse neuroblastoma component of the hybridoma [I 51. In NG108-IS cells, the presence of secretin receptors was documented only by membrane adenylate cyclase activation and cAMP elevation in intact cells. We report here (a) the pharmacological characterization of these secretin receptors based on 1251-secretinbinding and adenylate cyclase coupling, using several analogues and fragments of secretin, and (b) the regulation (amplification) of the expression of these receptors by N6,2’-0-dibutyryladenosine 3’,5’-monophosphate (Bt,cAMP) and butyrate. MATERIALS AND METHODS Culture
NG108-15 cells (passage number 26) were kindly donated by Dr J. M. Maloteaux (Louvain, Belgium). The experiments presented in this paper were performed employing cells of passage numbers 30-SO. These cells were grown in Dulbecco’s modified Eagle’s medium supplemented with 10%
150 (by vol.) foetal calf serum. This medium was further enriched withO.l mM hypoxanthine, 1 pM aminopterin, 16 pM thymidine and antibiotics (100 U/m1 penicillin streptomycin) [18]. For propagation, cells were dislodged from the flasks by firm tapping in order to release them into the medium. Cell medium was changed every 2-3 days. In case of cell pretreatment, the culture medium of preconfluent cells was renewed with fresh medium supplemented with the drug to be tested.
37‘C by adding 0.5 ml 0.5% sodium dodecyl sulfate solution containing 0.5 mM ATP, 0.5 mM CAMP and 20000cpm [8-3H]cAMP. CAMPwas separated from ATP by t w o successive chromatographies on Dowex 50-WX8 and neutral alumina. Under all conditions tested, CAMPproduction was linear during the incubation period and proportional to the amount of protein added.
Memhrane pwparurioii
Protein determination was measured according to Lowry et al. [23] using bovine serum albumin as standard.
+
NGIOX-15 cells wese harvested by agitation, pelleted by centrifugation at 100 x g for 5 rnin at 20”C, washed in the standard culture medium (without supplement), lyzed in hypotonic 1 mM NaHCO, (pH 7.0) then quickly frozen in liquid Nz. After thawing, the lyzate was centrifuged at 400 x g for 10 min at 4’C. The resulting supernatant was centrifuged at 10000 x g for 15 min, and the pellet was washed twice in 1 mM NaHC03. Thc final pellet was resuspended in 1 mM NaHCO, and immediately used for binding assay and adenylate cyclase assay. Rrrdioioditiut ion 0f sccretin ’
Synthetic porcine secretin was radioiodinated at pH 8.5 according to Jensen et al. [19] with minor modifications [20]. 1251-secretinwas separated from free 251by cellulose adsorption and elution with 12% bovine serum albumin. This ligand, radiolabelled at the level of N-terminal histidine, corresponded to a single homogeneous peak after radio-HPLC on a yBondapak C1 column (Waters, Milford, USA). Its specific radioactivity was 1.5 mCiinmol.
’
25
I-srcrrtin hitiding .studies
lZ51-secretinbinding to NGlOX-15 membranes was carried out as described previously [21] in 50 mM Tris/HCl pH 7.4 containing 5 mM MgCI2,0.5 mg/ml bacitracin, 100 kallikrein inhibitor unitsjml Trasylol, 1O h bovine serum albumin, 25 60 pM ‘251-secretin,increasing concentrations of unlabelled peptide and 100 pg membrane protein in a total volume of 120 pl. Incubations were conducted at 37°C and terminated by adding 2 ml ice-cold 50 mM sodium phosphate pH 7.4 to each sample, followed by rapid filtration through glass-fiber filters (GF/C; Whatman, Maidstone, Kent, UK) presoaked for 24 h in 0,1% poly(ethy1eneimine) in order to reduce nonspecific binding. The filters were rinsed three times with 2 ml of the same buffer and their radioactivity was measured. Nonspecific binding was determined in the presence of 1 pM unlabelled secretin and accounted for approximately 25% of total binding. Specific binding was defined as total binding minus non-specific binding and always represented less than 25% of the total radioactivity offered. Adenyla t e cycluse ~issuy Adenylate cyclase activity was determined according to the procedure of Salonion et al. [22]. Membrane protein (20 30 pg) was incubated in a total volume of 60 p1 containing 0.5 mM [a-”PIATP. 10 pM GTP, 5 mM MgCI2, 0.5 mM EGTA, 1 mM CAMP,0.5 m M theophylline, 10 mM phosphoenolpyruvate, 30 pg/mI pyruvate kinase and 30 mM Tris/HCl at a final pH of 7.5 [XI. The reaction was initiated by addition of membranes and was terminated after a 15-min incubation at
Protein determination
Chemicals [~-Phe’]Secretin and [~-Phe~]secretin, synthetic porcine vasoactive intestinal peptide (VIP) and porcine peptide histidine isoleucinamide (PHI) were generous gifts from D r D. H. Coy (Tulane University, New Orleans, LA, USA). Synthetic porcine secretin fragments and other secretin analogues were kindly given by Dr W. Konig (Hoechst Aktiengesellschaft, Frankfurt/Main, FRG). Synthetic helodermin came from Novabiochem (Laufelingen, Switzerland). Carrier-free NalZ51 (IMS 300, 600-800 mCi/ml) and [8-3H]cAMP (20 - 30 Ci/mmol) were purchased from Amersham International (Amersham, Bucks, UK) and [ z - ~ ~ P I A T P (30 Ci/mmol) from New England Nuclear (Boston, MA, USA). Bovine serum albumin (fraction V), bacitracin, phosphoenolpyruvate, pyruvate kinase, CAMP, GTP, ATP, R-bromoadenosine 3’,5’-monophosphate (Br’cAMP), prostglandin El (PGE1), isobutyrate, butyrate, 3-hydroxybutyrate were purchased from Sigma Chemical Co (St Louis, MO, USA). Kallikrein inhibitor (Trasylol) was a gift from Bayer (Leverkusen, FRG), forskolin was obtained from Calbiochem (La Jolla, CA, USA), rac-4-(3-butoxy-4-methoxybenzyl)-2iinidazolidinone (Ro 20-1 724) from Koche S.A. (Brussels, Belgium) and guanosine 5’-[/l,y-imido]triphosphate (p[NH]ppG) from Boehringer-Mannheim (Brussels, Belgium). Foetal calf serum and medium for cell culturcs were from Gibco Europe (Gent, Belgium). All other reagents were of the highest grade available. RESULTS Characteristics of 251-secretinbinding sites Specific 12’I-secretin binding to NG108-15 membranes reached a steady state after 30 rnin at 37 C (data not shown). Tracer bound after 30 min dissociated with a tl,z of 9 rnin as studied by isotopic dilution. This t l 1 2was transiently reduced to 2 min by addition of 10 pM GTP (Fig. 1). Saturation curves were obtained by incubating increasing concentrations of 1251-secretin(60- 1200 pM). The Scatchdrd transformation (Fig. 2) was compatible with the existence of one high-affinity class of binding sites exhibiting a KD of 0.3 5 0.2 nM (mean f SEM of four determinations from different membrane preparations). Competition curves were performed by incubating membranes for 30 min in the simultaneous presence of a 60 pM tracer concentration and increasing concentrations of unlabelled peptide. Unlabelled porcine secretin inhibited tracer bindingwith an ICso of2 nM (Fig. 3). Helodermin was tenfold less potent than secretin in inhibiting tracer displacement; PHI and VIP were barely effective. The selectivity of these secretin receptors was further evaluated
151 1%)
100-
1000 3 0
z
75-
m
z
w
&
50-
V
50-
0 W
w Ln
2
0 z 0 3
0-
m U
30-
+
U U
0 U Y
I
l o g [ P e p t i d e ] /M 5
0
15
10 T I M E bin)
Fig. 1. Kinetics of dissociation of prebound '251-secretinfrom control NG108-15 membranes. After a 30-min preincubation with tracer, 0.1 pM unlabelled secretin was added in the absence (0)or presence (a) of 10 pM GTP. The radioactivity remaining bound was calculated, after subtraction of non-specific binding, and estimated as a percentage of tracer bound before addition of unlabelled secretin. The results, presented on a semi-log scale, were the means of experiments performed in duplicate on three separate membrane preparations
Fig. 3. Doseleifect curves of inhibition of '251-secretin binding (upper panel) and adenylate cyclase activation (lower panel} in control NG108-15 membranes by secretin (a),helodermin ( O ) , PHI (A) and VIP (B). Binding data are expressed as the percentage of tracer specifically bound in the absence of unlabelled peptide and are the means of two experiments performed in duplicate. Data on adenylate cyclase activation, in the presence of 10 pM GTP, are expressed as a percentage of the rate of production of cAMP/mg protein in the presence of 1 pM secretin (after subtraction of the basal unstimulated value) and are the means of two experiments in duplicate
\
4001
[ALA4, VAL5] S e c r e t in
0 10
20
30 40 50 60 70 E ifmol.mg protein-11
80
90
Fig. 2. Scatchard representation of' a saturation curve obtained after incubating increasing '251-secretin concentrations with control (a) NG108-15 cell membranes and membranes f r o m cells pretreated with 1 mM B t 2 c A M P (0) or 1 mM butyrate ( A } , f o r 24 h. Tracer binding was conducted for 30 min at 37°C. Non-specific binding was determined in the presence of 0.1 pM secretin. The results were representative of three others performed in duplicate
with secretin analogues and secretin fragments. The results in Figs 4 and 5 show the following decreasing order of potency: porcine secretin = [Val'lsecretin > [Alaz]secretin = [Ala' ']secretin > [Ala4,Va15]secretin> [Ala4]secretin > [DPhe4]secretin > [~-Phe']secretin = secretin (2 - 27) > secretin (3 - 27) > secretin (7 - 27).
'-10
-9
-8
-7
-6
log [ P e p t i d e ] /M
Fig. 4. Dose/
