9
Regulatory Peptides, 16 (1986) 207-215 207 Elsevier RPT 00536 A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1 Florence Reyl-Desmars a, Christian Laboisse b and Miguel J.M. Lewin a aUnitb de Recherches de Gastroentbrologie (INSERM U.IO), and bUnit~ de Recherches de Biologic et Physiologic des Cellules Digestives ( INSERM U.239), Hrpital Bichat, Paris, France (Received 26 June 1986; revised manuscript received 17 October 1986; accepted for publication 20 October 1986) Summary Somatostatin receptors are demonstrated in the human derived gastric cell line HGT-1. Using 125I-Tyr11-somatostatin as ligand, two classes of sites were charac- terized with apparent dissociation constants KDI = 0.9 x 10-lo M and KD2 = 4 x 10-9 M and maximum binding capacities of N1 = 20 and N2 = 556 fmol per mg protein, respectively. These values are close to those previously reported in freshly isolated parietal cells (Reyl, F., Silve, C. and Lewin, M.J.M., Somatostatin receptors on isolated gastric cells. In S. Bonfils et al. (Eds.), Hormone Receptors in Digestion and Nutrition, Elsevier/North-Holland, Amsterdam, 1979, pp. 391-400). Somato- statin binding to the high affinity sites was partially inhibited by the non-hydrolysable guanyl nucleotide analog Gpp(NH)p and by pretreating the cells with islet activating protein (IAP). Furthermore, IAP counteracted the inhibitory effect of somatostatin on histamine stimulation of adenylate cyclase. These findings are interpreted in terms of somatostatin interaction with the 41 000 Da adenylate cyclase GTP-dependent inhibitory subunit, Ni. stomach; somatostatin; histamine; pertussis toxin; inhibitory subunit Address for correspondence: F. Reyl-Desmars, INSERM U.10, H6pital Bichat, 170 Boulevard Ney, 75877 Paris Cedex 18, France. Tel.: (1) 46.27.63.25. 0167-0115/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)

A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

Embed Size (px)

Citation preview

Page 1: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

Regulatory Peptides, 16 (1986) 207-215 207 Elsevier

RPT 00536

A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line

HGT-1

Florence Reyl-Desmars a, Christian Laboisse b and Miguel J.M. Lewin a

aUnitb de Recherches de Gastroentbrologie (INSERM U.IO), and bUnit~ de Recherches de Biologic et Physiologic des Cellules Digestives ( INSERM U.239), Hrpital Bichat, Paris, France

(Received 26 June 1986; revised manuscript received 17 October 1986; accepted for publication 20 October 1986)

Summary

Somatostatin receptors are demonstrated in the human derived gastric cell line HGT-1. Using 125I-Tyr11-somatostatin as ligand, two classes of sites were charac- terized with apparent dissociation constants KDI = 0.9 x 10-lo M and KD2 = 4 x 10-9 M and maximum binding capacities of N1 = 20 and N2 = 556 fmol per mg protein, respectively. These values are close to those previously reported in freshly isolated parietal cells (Reyl, F., Silve, C. and Lewin, M.J.M., Somatostatin receptors on isolated gastric cells. In S. Bonfils et al. (Eds.), Hormone Receptors in Digestion and Nutrition, Elsevier/North-Holland, Amsterdam, 1979, pp. 391-400). Somato- statin binding to the high affinity sites was partially inhibited by the non-hydrolysable guanyl nucleotide analog Gpp(NH)p and by pretreating the cells with islet activating protein (IAP). Furthermore, IAP counteracted the inhibitory effect of somatostatin on histamine stimulation of adenylate cyclase. These findings are interpreted in terms of somatostatin interaction with the 41 000 Da adenylate cyclase GTP-dependent inhibitory subunit, Ni.

stomach; somatostatin; histamine; pertussis toxin; inhibitory subunit

Address for correspondence: F. Reyl-Desmars, INSERM U.10, H6pital Bichat, 170 Boulevard Ney, 75877 Paris Cedex 18, France. Tel.: (1) 46.27.63.25.

0167-0115/86/$03.50 © 1986 Elsevier Science Publishers B.V. (Biomedical Division)

Page 2: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

208

Introduction

Besides other possible mechanisms mediated by Ca 2 ÷ [1,2], or protein dephos- phorylation [3-5], somatostatin inhibitory effect is suggested to involve cyclic AMP. This involvement has recently received strong supports by the demonstration of so- matostatin inhibition of adenylate cyclase activation [6-11] or cAMP production in freshly isolated cells or cultured cells from the pituitary and other tissues [1,2,12-15]. That this inhibition can be counteracted by the guanyl nucleotide GTP, or by lAP, the Bordetella pertussis toxin known to ADP ribosylate the 41 000 Da GTP-depen- dent adenylate cyclase inhibitory subunit, suggested the involvement of this subunit in the mediation of somatostatin inhibitory effect in the pituitary [11,13,14,16,17].

Somatostatin is a strong inhibitor of gastric acid secretion [18] and the presence of specific receptor sites for this peptide has been documented on acid secreting par- ietal cells of the stomach, using enzymatically dissociated cell suspensions [19,20]. On this material, however, the study of surface receptors is not easy because of a possible occurrence of proteolytic degradation. In this respect, the gastric human-derived HGT-1 cell line represents a better model. Although they do not apparently secrete acid, these cells have been previously reported to express several surface receptors occurring on the normal gastric parietal cell and particularly adenylate cyclase coupled histamine H2 receptors [21-23]. So far, however, the characterization of somatostatin receptors on this cell line has not been documented.

In the present study we provide evidence for the presence of somatostatin receptor sites on the HGT-1 cells and further demonstrate a coupling of these receptor sites with a GTP-dependent inhibitory subunit interacting with the H2 receptor stimulated adenylate cyclase.

Material and Methods

Cell culture The HGT-1 cell line originating from a human gastric tumor [22] was grown in

Dulbecco's modified Eagle's medium (Gibco, France), supplemented with 10% fetal bovine serum. Five days after seeding, cells were preincubated or not (control cells) overnight with cholera toxin (Calbiochem) and islet activating protein (lAP, List Biologicals, Campbell, CA, U.S.A.) added in the culture medium.

Binding studies Tyrl 1-Somatostatin - 14 (Bachem, Switzerland) was iodinated by the chloramine-T

method (2200 Ci/mmol) [19,23] and purified by chromatography on a carboxymethyl cellulose (CM 52) column. Cells were incubated in 96-well plates with the iodinated analog (20 pM) in the presence (nonsaturable binding) or absence (total binding) of 10-6 M unlabeled somatostatin-14 (Sanofi, Montpellier, France). Before incubation, the cells were washed twice with a modified Krebs buffer containing 10 mM NazHPO4, 0.5 mM NaH2PO4, 11 mM glucose, 20 mM NaHCO3, 5 mM KCI, 80 mM NaCI, 1 mM CaC12, 1.5 mM MgC12, 50 mM Hepes adjusted to pH 7.4 with

Page 3: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

209

Tris, 0.1% bovine serum albumin, and 0.01% bacitracin. After 0-30 min incubation at 37°C, the cells were rapidly rinsed with saline buffer, dissolved in 0.05 N NaOH, and the radioactivity was counted in a Gamma spectrophotometer. Degradation of ~25I-Tyr 1 ~-somatostatin in the presence of the cells was measured by high pressure liquid chromatography. After 15 min and 30 min of incubation, 82 and 73% of the tracer, respectively, were still remaining in the medium.

Adenylate cyclase assay Cells were permeabilized by replacing the medium in the flasks by 3 ml of distilled

water. After 5 min, cells were detached by scraping with a rubber policeman, centri- fuged in a minifuge and resuspended at 4°C in 4 mM Tris-HCl, pH 7.6. Adenylate cyclase activity was measured as already described [9,10] in 50 mM Tris-HC1 buffer, pH 7.6, containing 4 mM MgC12, 1 mM EGTA, 1 mM isobutylmethyl xanthine, 0.5 mM cyclic AMP, 5 mM phosphocreatin, 0.4 mg/ml creatin phosphokinase, 20 000 cpm cyclic [3-8-all]AMP (Amersham), 0.5 x 106 cpm alpha-[32p]ATP (Amersham) without (basal activity) or with the agents studied in a final volume of 60 #1. The reaction was started by the addition of cells (10 #g protein) and run for 10 min at 37°C. Radioactive samples containing the cAMP formed were counted in a Beckman liquid spectrophotometer.

Results

The radiolabeled analog 12SI-Tyr11-somatostatin specifically bound to the HGT- 1 cells according to a time-dependent process (Fig. 1). At 37"C a steady state was reached within 15 min incubation with a maximal bound radioactivity corresponding to 5.5 + 0.5 fmol somatostatin per mg protein. Nonsaturable (non-specific) binding

O.E

o o

v

.c_

o

0

total

1 non

saturable

10 20 30 t ime (min)

Fig. 1. Kinetics of 12SI-TyrXl-somatostatin binding to HGT-I cells. The trace (2 x 10 -11 M) was incu- bated at 37"C, pH 7.4, with cells (0.1 x 106 cells per well) in the presence (non-saturable) or the absence (total binding) of 10-6 M native somatostatin. Mean values + 1 S.E. o f three experiments with triplicate determinations.

Page 4: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

210

o•_. ° 0 . 4 ~

= 0.2.

~ 0.1. o

f o

A

I'0 15 Time (min)

Fig. 2. Kinetics of dissociation of 12 H.Tyr ~ l.somatostatin binding. At the times indicated by the arrows 10 -6 M unlabeled somatostatin were added to the medium.

represented less than 27-30% of total binding. Radiolabeled somatostatin specifically bound with an apparent association constant Ka of 3.7 x 109 M- 1. min-1 and was totally chased by the addition of 10-6 M unlabeled somatostatin in the medium (Fig. 2). Under the conditions of the study (low ligand concentration) bound-somatostatin release occurred with an apparently single dissociation constant, Kd, of 0.16 min-1.

Specific binding was displaced by 50 and 100% by 10- lo M and 10- T M unlabeled somatostatin, respectively, but no displacement occurred with up to 10-5 M rGRF (rat growth hormone-releasing factor), PHI (peptide histidine isoleucine), or VIP (vasoactive intestinal peptide) (Fig. 3). However, somatostatin binding was partially displaced by Gpp(NH)p (50% displacement with 10 -5 M). Non-linear regression analysis of saturation curves according to Scatchard was consistent with the presence

10(;

._=

E o

0 i i I i i _12 -11 _I0 _9 _8 _7 _6 _5 _4

T !

S 14~'*e'T~ Gp~) ( NH IP ~ t

concentration (IogM)

Fig. 3. Displacement of 12 S i_Tyrl l_somatostatin specific binding by Gpp(NH)p and various pepfidos (20 rain incubation, 37"C, pH 7.4). Mean values + l S.E. of three experiments with triplicate determinations.

Page 5: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

211

of two classes of binding sites: high affinity sites with an apparent dissociation con- stant (KD) of 0.9 x 10-10 4- 0.2 M and an estimated number (Bm~) of 0.44 x 10-11 M (i.e., 20 + 2 fmol per mg protein) and low affinity sites with a KD of 4.0 x 10 - 9

+ 0.5 M and a Bma~ of 12.25 x 10-11 M (i.e., 556 4- 50 fmol per mg protein) (Fig. 4). Treatment of the cells by 100 ng/ml IAP resulted in a marked increase in the apparent dissociation constant of the low affinity sites (8 -4- 0.2 x 10 -9 M vs. 4 + 0.5 × 10 - 9 M ) but no significant change in Bma, (556 4- 80 vs. 556 4- 50 fmol per mg protein). Similar results were obtained with the addition of 10 -4 M Gpp(NH)p in the incubation medium (not shown). However, the apparent dissociation constant of the high affinity sites was unchanged (0.9 4- 0.1 vs. 0.9 + 0.2 x 10 -1° M) while the estimated number of these sites was decreased by 57% (8.6 4- 0.4 vs. 20 4- 2 fmol per mg protein). In sharp contrast to IAP treatment, treatment with cholera toxin had no effect on somatostatin binding (Fig. 5).

Consistent with a previous report [24], histamine stimulated adenylate cyclase ac- tivity of HGT-1 cells with an apparent concentration for half maximal effect of 0.96 4- 0.14x 10 -5 M corresponding to 3.6 4- 0.36 pmol cAMP per 106 cells per min (i.e., 1.75-fold the basal value) (not shown). Histamine stimulation of adenylate cy- clase activity was concentration-dependently inhibited by somatostatin (Fig. 6). The inhibition was however only partial, 30--40% at the maximal somatostatin concen- tration tested (10 -a M). Somatostatin had no effect on the basal adenylate cyclase activity (not shown). Treatment of the cells with IAP prevented the inhibitory effect of somatostatin by 50% with a toxin concentration of 40 ng/ml and by 100% with a toxin concentration of 100 ng/ml.

501~ • control a * lAP

40

a 20 z o

o \ , ,\ , , , , , , 0 02 (14 0.6 0.8 2 6 10

BOUND (~10-"M)

Fig. 4. Scatchard analysis o f 125I-Tyrll-somatostatin specific binding (Fig. 3) to HGT-1 control cells as compared to HGT-I cells treated by 100 ng/ml IAP. Mean values + 1 S.E. of three experiments with triplicate determinations.

Page 6: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

212

~ 3 7 5

.L20250

.E

~0.125 o E

• _ ° contro l , _ . + l A P

. _ , + C T T

i ! I

10 2 0 3 0 t i m e ( ra in )

Fig. 5. Effect of 100 ng/ml lAP or 5 x 10 - t° M cholera toxin treatment of HGT-1 cells on ~2SI-Tyr 11- somatostatin specific binding time course. Mean values + 1 S.E. of three experiments with triplicate determinations.

100

== m

u

E 50.

.c:

0

_%,~ ~- T ~ - I A P 100 ng/ml

"~"-l.IAP *0 .vm,

control

-1'1 -10 -b -§ somatos ta t l n 14 (log M)

Fig. 6. Effect of lAP treatment on somatostatin inhibition of 10-* M histamine-stimulated adenylate cyclase activity in HGT-I cells. Mean values + 1 S.E. of three experiments with triplicate determinations.

Discussion

The present s tudy s t rongly suppor t s the presence o f a t least two classes o f soma-

tos ta t in recep tor sites on H G T - 1 cells. The co r r e spond ing Kv values (0.9 x 1 0 - l o

M and 4.5 x 10 -9 M) are close to those prev ious ly observed in freshly i sola ted

gastr ic cells, i.e. 0.8 x 10 -1° M and 4 x 10 -9 M, respect ively [19]. Fu r the rmore ,

several lines o f evidence are p rov ided which a rgue for the existence o f a negat ive

coupl ing o f the high affinity soma tos t a t i n recep tor sites to adeny la te cyclase.

Page 7: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

213

(1) The non hydrolysable guanyl nucleotide analog Gpp(NH)p displaced soma- tostatin binding. Such a finding is generally interpreted in terms of an altosteric interaction between the receptor and the GTP-binding adenylate cyclase regulatory protein Ni [16,17] albeit that it has recently been suggested that GTP-binding regu- latory protein might also exist in non-adenylate cyclase-dependent receptor systems [1,2]. Moreover, the effect described here of Gpp(NH)p on high affinity binding of somatostatin to HGT- 1 cells is consistent with the previous study of Enjalbert et al. [25] on rat adenohypophyseal membranes. In this study, a 68% decrease in Bm~x but no change in KD (0.97 x 10 -9 M) of the (unique) site was observed in the presence of 5 x 10 -5 M Gpp(NH)p. On the other hand, the effect of Gpp(NH)p on the low affinity sites observed in our study is consistent with the finding of Koch et al. [26] on GH4 C1 cell membranes suggesting an increase in the KD (1.1 x 10 -1° M to 4 x 10-lo M) but no change in the Bm~x for the unique site demonstrated.

(2) Somatostatin potently, although partially, inhibited histamine-induced adenyl- ate cyclase activation, in concentration ranging from 10-11 M to 10 -9 M (IC5o = 10-lo M). This range suggests that only the high affinity sites would be involved in the inhibitory mechanism. Such a partial inhibitory effect of somatostatin has been already observed by others on VIP- or GRF-induced cAMP accumulation in GH4 C1 [26,27] and in normal pituitary cells [8], respectively. It is also consistent with our own observations on somatostatin inhibition of GRF-induced adenylate cyclase ac- tivation in the normal pituitary [10,11] as well as in the GH3 cell line [9].

(3) Somatostatin inhibition of adenylate cyclase activation is prevented by IAP, a toxin protein believed to specifically interact with the GTP-binding regulatory in- hibitory subunit Ni of adenylate cyclase [16,17]. The decrease of the number of high affinity sites observed after lAP treatment and the above reported observations on Gpp(NH)p effect on somatostatin binding and somatostatin inhibition of histamine stimulated adenylate cyclase, are consistent with the following model:

R * N i C ~ R N i + C

where R stands for somatostatin receptor, Ni for the GTP binding inhibitory subunit and C for the catalytic subunit, with Gpp(NH)p or IAP inducing a shift from a high affinity (R*) to a low affinity (R) receptor state. The high affinity receptor sites appear kinetically similar to those previously described on membrane preparations [25]. Low affinity sites may represent biologically silent or deactivated receptors which could correspond either to membranous or internalized receptors [3,4,20,28]. Since IAP treatment was found to totally counteract somatostatin effect on adenylate cyclase while it reduced by only 57% the number of high affinity binding sites one may suggest that only a part of those sites are actively coupled with adenylate cyclase. Indeed, other mechanisms not involving adenylate cyclase inhibition, have been sug- gested for somatostatin receptor coupling to cellular function, e.g. inhibition of Ca 2 + mobilization via a GTP-dependent regulatory protein [29-31] or protein dephos- phorylation via the activation of phosphoprotein phosphatases [4,5]. So far, the pres- ence of such alternative pathways has not been investigated in the HGT-1 cells. Particularly, in the light of the observations previously reported on freshly isolated

Page 8: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

214

par ie ta l cells [3-5] the s tudy o f soma tos t a t i n recep tor in te rna l iza t ion and the char-

ac te r iza t ion o f a soma tos t a t in s t imula ted p h o s p h a t a s e act ivi ty in cytosol ic f ract ions

should deserve fur ther exper iments on H G T - 1 cells. W i t h this respect it is to be

po in ted ou t tha t the present s tudy was l imited to the initial , reversible phase o f so-

ma tos t a t i n b ind ing which very l ikely co r re sponds to the occupancy o f surface recep-

tors.

A l t h o u g h the poss ib i l i ty canno t be ruled out tha t a s t ruc tura l difference exists for

soma tos t a t i n and h is tamine receptors between H G T - 1 and n o r m a l par ie ta l cell, we

suggest tha t the molecu la r mode l d o c u m e n t e d in this s tudy on the H G T - I cell could

at least pa r t i a l ly account for soma tos t a t i n inh ib i t ion o f h is tamine s t imula ted gastr ic

acid secret ion, in vivo.

References

1 Koch, B.D., Dorflinger, L.J. and Schonbrunn, A., Pertussis toxin blocks both cyclic AMP-mediated and cyclic AMP-independent actions of somatostatin, J. Biol. Chem., 260 (1985) 13138-13145.

2 Reisine, T. and Guild, S., Pertussis toxin blocks somatostatin inhibition of calcium mobilization and reduces the affanity of somatostatin receptors for agonists, J. Pharm. Exp. Ther., 235 (1985) 551-557.

3 Reyl, F. and Lewin, M.J.M., Somatostatin is a potent activator of phosphoprotein phosphatases in the digestive tract, Biochim. Biophys. Acta, 675 (1981) 297-300.

4 Reyl, F.J. and Lewin, M.J.M., lntracellular receptor for somatostatin in gastric mucosal cells: decom- position and reconstitution of somatostatin-stimulated phosphoprotein phosphatases, Proc. Natl. Acad. Sci. USA, 79 (1982) 978-982.

5 Hierowski, M.T., Liebow, C., du Sapin, K. and Schally, A.V., Stimulation by somatostatin of de- phosphorylation of membrane proteins in pancreatic cancer MIA PaCa-2 cell line, FEBS Lett., 179 (1985) 252-256.

6 Jakobs, K.H., Aktories, K. and Schultz, G., A nucleotide regulatory site for somatostatin inhibition of adenylate cyclase in $49 lymphoma cells, Nature, 303 (1983) 177-178.

7 Aktories, K., Schultz, G. and Jakobs, K.H., Adenylate cyclase inhibition and GTPase stimulation by somatostatin in $49 lymphoma cyc- variants are prevented by islet-activating protein, FEBS Lett., 158 (1983) 169-173.

8 Spada, A., Vallar, L. and Giannattasio, G., Presence of an adenylate cyclase dually regulated by somatostatin and human pancreatic growth hormone releasing factor in GH-secreting cells, Endocri- nology, 115 (1984) 1203-1209.

9 Reyl-Desmars, F. and Zeytin, F., Somatostatin inhibits growth hormone-releasing factor-stimulated adenylate cyclase activity in GH3 cells, Biochem. Biophys. Res. Commun., 127 (1985) 986--991.

10 Reyl-Desmars, F., Baird, A. and Zeytin, F.N, GRF is a highly potent activator of adenylate cyclase in the normal human, bovine and rat pituitary: interaction with somatostatin, Biochem. Biophys. Res. Commun., 127 (1985) 977-985.

11 Reyl-Desmars, F., Baird, A. and Zeytin, F.N., Mechanism of action of growth hormone releasing factor in normal pituitary cells and in the GH3 cell-line: interaction with somatostatin. In M.J.M. Lewin and S. Bonfils (Eds.), Regulatory Peptides in Digestive, Nervous and Endocrine Systems, IN- SERM Symposium No. 25, Elsevier Science Publishers B.V., Amsterdam, 1985, pp. 35-38.

12 Bilezikjian, L.M. and Vale, W.W., Stimulation of adenosine 3',5'-monophosphate production by growth hormone-releasing factor and its inhibition by somatostatin in anterior pituitary cells in vitro, Endocrinology, 113 (1983) 1726-1731.

13 Cronin, M.J., Hewlett, E.L., Evans, W.S., Thorner, M.O. and Rogol, A.D., Human pancreatic tumor growth hormone (GH)-releasing factor and cyclic adenosine 3',5'-monophosphate evoke GH release from anterior pituitary cells: the effects of pertussis toxin, cholera toxin, forskolin, and cycloheximide, Endocrinology, 114 (1984) 904--913.

Page 9: A somatostatin receptor negatively coupled to adenylate cyclase in the human gastric cell line HGT-1

215

14 Yajima, Y., Akita, Y. and Saito, T., Pertussis toxin blocks the inhibitory effects of somatostatin on cAMP-dependent vasoactive intestinal peptide and cAMP-independent thyrotropin releasing hor- mone-stimulating prolactin secretion of GH3 cells, J. Biol. Chem., 261 (1986) 2684-2689.

15 Guillemin, R., Zeytin, F., Ling, N., B6hlen, P., Esch, F., Brazeau, P., Bloch, B. and Wehrenberg, W.B., Growth hormone-releasing factor: chemistry and physiology (41813), Proc. Soc. Exp. Biol. Med., 175 (1984) 407--413.

16 Hsia, J.A., Moss, J., Hewlett, E.L. and Vaughan, M., ADP-ribosylation of adenylate cyclase by per- tussis toxin, J. Biol. Chem., 259 (1984) 1086-1090.

17 Katada, T., Bokoch, G.M., Northup, J.K., Ui, M. and Gilman, A.G., The inhibitory guanine nucleo- tide-binding regulatory component of adenylate cyclase, J. Biol. Chem., 259 (1984) 3568-3577.

18 Barros D'Sa, A.J.J., Bloom, S.R. and Baron, J.H., Inhibition by somatostatin of gastric acid and pepsin and G-cell release of gastrin, Gut, 19 (1978) 315-320.

19 Reyl, F., Silve, C. and Lewin, M.J.M., Somatostatin receptors on isolated gastric cells. In S. Bonfils, G. Rosselin and M.J.M. Lewin (Eds.), Hormone Receptors in Digestion and Nutrition, Elsevier/ North-Holland, Amsterdam, 1979, pp. 391-400.

20 Reyl-Desmars, F. and Lewin, M.J.M., Evidence for an intracellular somatostatin receptor in pancreas: a comparative study with reference to gastric mucosa, Biochem. Biophys. Res. Commun., 109 (1982) 1324-1331.

21 Gespach, C., Emami, S. and Rosselin, G., Gastric inhibitory peptide (GIP), pancreatic glucagon and vasoactive intestinal peptide (VIP) are cAMP-inducing hormones in the human gastric cancer cell line HGT-1. Homologous desensitization of VIP receptor activity, Biochem. Biophys. Res. Commun., 120 (1984) 641-649.

22 Cheret, A.M., Laboisse, C.L., Roumagnac, I., Augeron, C. and Lewin, M.J.M., Highly histamine- responsive clones from the human gastric adenocarcinoma cell line HGT-1, Agents Actions, 17 (1985) 436-440.

23 Greenwood, G.C., Hunter, W.M. and Glover, J.S., The preparation of 131I-labelled human growth hormone of high specific radioactivity, J. Biochem., 89 (1963) 114.

24 Laboisse, C.L., Augeron, C., Couturier-Turpin, M.H., Gespach, C., Cheret, A.M. and Potet, F., Char- acterization of a newly established human gastric cancer cell line HGT-1 bearing histamine H2-recep- tots, Cancer Res., 42 (1982) 1541-1548.

25 Enjalbert, A., Rasolonjanahary, R., Moyse, E., Kordon, C. and Epelbaum, J., Guanine nucleotide sensitivity of 125-I-iodo NTyr-somatostatin binding in rat adenohypophysis and cerebral cortex, En- docrinology, 113 (1983) 822-824.

26 Koch, B.D. and Schonbrunn, A., The somatostatin receptor is directly coupled to adenylate cyclase in GH4CI pituitary cell membrane, Endocrinology, 114 (1984) 1784-1790.

27 Dorflinger, L.J. and Schonbrunn, A., Somatostatin inhibits vasoactive intestinal peptide-stimulated cyclic adenosine monophosphate accumulation in GH pituitary cells, Endocrinology, 113 (1983) 1541-1550.

28 Lewin, M.J.M. and Reyl-Desmars, F., Internalization of somatostatin by digestive cells. In M.J.M. Lewin and S. Bonfils (Eds.), Regulatory Peptides in Digestive, Nervous and Endocrine Systems, IN- SERM Symposium No. 25, Elsevier Science Publishers B.V., Amsterdam, 1985, pp. 51-54.

29 Zeggari, M., Susini, C., Viguerie, N., Esteve, J.P., Vaysse, N. and Ribet, A., Tumor promoter inhibition of cellular binding of somatostatin, Biochem. Biophys. Res. Commun., 128 (1985) 850-857.

30 Matozaki, T., Sakamoto, C., Nagao, M. and Baba, S., Phorbol ester or diacylglycerol modulates somatostatin binding to its receptors on rat pancreatic acinar cell membranes, J. Biol. Chem., 261 (1986) 1414-1420.

31 Michell, B. and Kirk, C., G-protein control of inositol phosphate hydrolysis, Nature, 323 (1986) 112-113.