Eur. J. Biochem. 91, 31 -38 (1978)

The Interaction of Caerulein with the Rat Pancreas 2. Specific Binding of [3H]Caerulein on Dispersed Acinar Cells

Jean CHRISTOPHE, Philippe DE NEEF, Monique DESCHODT-LANCKMAN, and Patrick ROBBERECHT

Department of Biochemistry and Nutrition, Medical School of the UniversitC Libre de Bruxelles

(Received March 28, 1978)

1. [3H]Caerulein was bound to dispersed acinar cells from rat pancreas in a rapid, reversible, specific, saturable, and temperature-dependent manner. Binding decreased above pH 6.5. Treatment of intact cells with 2,4-dinitrophenol and oligomycin, p-chloromercuribenzoate, diisopropylfluoro- phosphate and glutaraldehyde impaired [3H]caerulein binding whereas the addition of EGTA increased binding. The C-terminal octapeptide of pancreozymin, desulfated caerulein and pentagastrin inhibited binding of [3H]caerulein whereas vasoactive intestinal polypeptide, secretin, bombesin or carbamoylcholine were without effect. The good resistance of [3H]caerulein to inactivation by acinar cells at 37 "C was reflected in the high proportion of tracer remaining capable of binding to fresh acinar cells.

2. Scatchard plots of [3H]caerulein binding were curvilinear with an upward concavity. The addi- tion of an excess of unlabeled caerulein resulted in the release of as much as 65 % of bound [3H]caeru- lein within 1 min at 37 "C. The dissociation of remainder followed much slower kinetics.

3 . The results suggested that intact rat pancreatic acinar cells have one class of caerulein binding sites existing in two states: one with high affinity and another with low affinity, the proportion of sites in each state depending on the degree af site occupancy (negative cooperativity), and on the intracellular concentration of nucleotides.

Caerulein, a decapeptide isolated from the skin of the Australian amphibian Hyla cacrulea, is similar in chemical structure and in biologic activity to chole- cystokinin-pancreozymin [l -41. In general the po- tency of caerulein is approximately the same as that of the C-terminal octapeptide of cholecystokinin- pancreozymin.

The interaction of monotritiated [3H]caeruleiii with rat pancreatic plasma membranes has been reported in the preceding paper [5] . A study with dispersed intact acinar cells was undertaken to observe the effect of extramembrane events on caerulein binding, e.g. effect of endogenous nucleotides and calcium fluxes. The present communication deals with the kinetics, stoichiometry and general chemical specificity of [3H]caerulein interaction. A further report [6] focuses on the structural requirements for the binding of peptides of the cholecystokinin-pan- creozymin family and the way in which this correlates

Ahhrevintions. EGTA, ethyleneglycol-bis-(2-an1inoethylether)- N,N,N',N'-tetraacetic acid.

Enzymes. Adenylate cyclase or ATP pyrophosphatelyase (cyc- lizing) (EC 4.6.1 .I); phospholipase A or phosphatide acyl-hydrolase (EC 3.1.1.4); neuraminidase (EC 3.2.1.1X); collagenase (EC 3.4.24.3); hyaluronidase (EC 3.2.1.35).

with the ability to stimulate secretion, calcium outflux and adenylate cyclase activity. Some of the present data have been published in abstract form [7].

MATERIALS AND METHODS

Preparation of Isolated Rat Pancreatic Acinar Cells

Male Wistar albino rats (200- 250 g) fed ad libitum were sacrificed by decapitation. Isolated pancreatic acinar cells were prepared following the procedure of Amsterdam and Jamieson [8] using crude collagenase and hyaluronidase, EDTA and mild shearing forces. The digestion of the rat pancreas and the isolation of acinar cells offered no more difficulty than that of the guinea pig. The final three washings of the cells were conducted in media containing 0.5 mM calcium concentration. Light microscopy revealed that 96 - 98 % of the cells were zymogen-containing exocrine cells. Their viability was demonstrated by their ability to exclude trypan blue. Cell concentration was deter- mined by counting an appropriately diluted suspen- sion in a conventional Neubauer hemocytometer. Unless otherwise specified, the dispersed acinar cells

32 Caerulein Receptor of Pancreatic Acinar Cells

were suspended in Krebs/Ringer/bicarbonate buffer (pH 7.4, equilibrated with 95 % O2 and 5 % COZ) con- taining 0.5 mM calcium chloride, 1 % (w/v) bovine serum albumin, 14 mM glucose, and 0.1 mg/ml Trasy- lo1 (i.e. 500 kallikrein inhibitor units/ml). The average cell concentration was 15 x lo6 cellsiml.

Binding of [ 3H/Caerulein to Dispersed Rat Pancreatic Acinar Cells

Incubations were routinely conducted with gentle agitation at 37 "C, the concentration of [3H]caerulein being 3.6 nM. At the indicated times duplicate 100 yl samples were filtered on EHWP Millipore filters (cellu- lose acetate, 0.5 pM pore size). Each filter was quickly washed three times with 1.5 ml of cold Krebs/Ringer/ bicarbonate (pH 7.4) containing 0.5 mM calcium chloride and 1 % (w/v) bovine serum albumin. The filters were transferred to liquid scintillation vials and dried in an oven at 60 "C for 45 min. 5 ml toluene- based scintillation fluid was then added to each vial and the radioactivity was determined in a Packard liquid scintillation counter (model 3320). Nonspecific binding was determined as that which occurred in the presence of 40 pM nonradioactive caerulein. Unless otherwise specified, all values for binding of [3H]- caerulein to isolated pancreatic acinar cells are ex- pressed as specific binding, that is, total binding minus nonspecific binding.

[ 3H]Caerulein Dissociation from Dispersed Rat Pancreatic Acinar Cells

Acinar cells were preincubated with [3H]caerulein as described for caerulein binding. After 10 min incu- bation at 37 "C, 100 yl aliquots of these cells were diluted in 1 to 5 ml of Krebs/Ringer/bicarbonate buffer (pH 7.4) containing 1 % bovine serum albumin. At appropriate times, the radioactivity remaining bound to the cells at 37 "C was determined by filtration through 0.5 pm cellulose acetate filters (type EHWP from Millipore).

Materials

The source of most of the materials including [3H]caerulein and several unlabeled peptides has been indicated in the preceding paper [5]. Desulfated caeru- lein, bombesin, the C-terminal nonapeptide of bom- besin (nonabombesin), and litorin were generous gifts from Dr R. de Castiglione (Montedison Farm- italia, Milano, Italy). Crude collagenase from Clostri- dium histolyticum and crude bovine-testis hyaluroni- dase were purchased from Sigma Chemical Co. (St Louis, Mo., U.S.A.). Phospholipase A was obtained from Boehringer (Mannheim, Germany), and neur-

aminidase (impure, number N-2876, specific activity 30 units/mg) was obtained from Sigma. Trasylol was a gift from Bayer-Pharma (Brussels, Belgium). Bovine plasma albumin, fraction V (fatty-acid-poor), was purchased from Miles laboratories, Inc. (Kanka- kee, Ill., U.S.A.). 2,4-Dinitrophenol was obtained from Merck (Darmstadt, Germany) and oligomycin (15% A and 85% B) was obtained from Sigma. A-23187 was a gift from Eli Lilly and Co. (Lndiana- polis, Ind., U.S.A.). N-Morpholino-3-propanesulfonic acid (Mops) and N-tris(hydroxymethyl)methyl-3-ami- no-propane-sulfonic acid (Taps) and maleic acid were obtained from Serva (Heidelberg, Germany). All other reagents were of the highest purity available.

RESULTS

The binding of [3H]caerulein to isolated rat pan- creatic acinar cells depended on time and temperature. At 37 "C specific binding of [3H]caerulein was rapid (approximately 50 "/, of maximum binding occurred within 30 s : Fig. 1 A), was maximal by 5 - 6 min and then remained constant for at least 20 more min (Fig. 1 B). The maximum amount of bound [3H]caeru- lein was increased by lowering the incubation temper- ature from 37 "C to 22 "C, 15 "C, or 2 'C: for instance there was 40 more [3H]caerulein bound after 24 min at 15 "C than at 37 "C. At the lower temperatures, however, the rate of the binding reaction decreased somewhat and more time was necessary to attain equilibrium binding (Fig. 1 A and B).

The functional alteration of free [3H]caerulein in the presence of acinar cells incubated at 37 "C was tested. The amount of free [3H]caerulein which could bind to fresh cells decreased by one-third within 6 min of the first cell exposure but did not change markedly over the course of the next 40 min. A portion of the decrease was due to [3H]caerulein binding to the first cells utilized (data not shown).

The pH dependence of t3H]caerulein binding to acinar cells was tested after a 10-min incubation period at 37 "C in media of a pH between 4.8 and 9.0. As shown in Fig. 2, specific binding of [3H]caerulein was low at alkaline pH. As the pH was decreased, binding increased and became maximal in the 4.8 - 6.5 range. Unspecific binding remained low at all pH values.

The effect of cell concentration in the range from 3 to 32 x lo6 cells/ml, on binding after 10 min at 37 ^C was tested. Binding was a linear function of cell con- centration in the range of 3-25 x lo6 cells/ml (data not shown). At a cell concentration of 25 x lo6 cells/ ml, 16% of the added tracer was bound. Further increases in the cell concentration increased binding only moderately. Most experiments were performed using 13 x lo6- 17 x lo6 cellsiml.

J . Christophe, P. De Neef, M. Deschodt-Lanckman, and P. Robberecht

20 1 1

33

A

0 1 2 3 4 5 0 5 10 15 20 25 Time (min) Time (rnin)

Fig. 1. Time course of specific binding o f ' / 3H/caeru/ein to rat pcmcrratir acinur cr1l.s as u function of temperature. The cells were incubated with 3.6 nM [3H]caerulein at the temperature indicated (37 "C: 0 ; 22 "C: A; 15 "C: 0; 2 "C: +), as described under Materials and Methods. For each experimental value, the nonspecific binding was subtracted. (A) Results are the means of five experiments performed in duplicate. (B) The time studies were recorded over 24 rnin at 37 "C (0-0) and 15 "C ((3-0)

*O 1

PH

Fig.2. Binding of [3H/caerulein to ral pancreatic acinar cells as a function o f p H (0). The cells were incubated for 10 min at 37 "C with 3.6 nM [3H]caerulein. The incubation solutions were prepared by mixing one portion of concentrated acinar cells in Krebs/kinger/ bicarbonate plus four portions of a buffer made of 0.154 M acetic acid, 0.154 M maleic acid, 0.154 M Mops, 0.154 M Taps and 0.308 M NaCl (1/1/1/1/2, by vol.). The pH was adjusted with HCl or NaOH and determined at 37 "C. Nonspecific binding was deter- mined in the presence of 0.4pM unlabeled caerulein (0). The results are the means of 4 experiments performed in duplicate

Magnesium and the calcium chelator EGTA in- creased the binding of [3H]caerulein (Fig. 3A). Maxi- mal binding was obtained in media containing 5 mM EGTA together with 6. 8 mM M2' , while 60-75 % of maximal binding was observed with 2.2 mM Ca2 ' combined with 1 or 6.8 mM Mg2+. When increasing the Mg2+ concentration from 0.9 mM to 7.1 mM, the stimulatory effects observed at steady state, i.e. after 10 min at 37 "C, were more marked in a calcium-free

medium supplemented with 5 mM EGTA than in media containing Ca2 '- (Fig. 3 B). The inhibitory effect of Ca2+ increased in the concentration range 0.5- 2.5 mM. Finally, the inhibitory effects noted when extracellular calcium exceeded 0.5 mM were not reversed by increasing Mg2 + concentration (Fig. 3 C).

The binding of [3H] caerulein was unaffected by pretreating acinar cells for 20 min at 37 "C with phos- pholipase A (0.06 to 1.60 pg/ml) or neuraminidase (at 350 pg/ml).

Three reagents capable of modifying surface pro- teins markedly reduced [3H]caerulein binding (data not shown). The inhibitory effect of 5 mM diisopropyl- fluorophosphate on [3H]caerulein binding was evident within 2 min of addition of the compound. Glutaral- dehyde markedly reduced the binding of [3H]caerulein when used at 0.45 or 0.90% concentration at 4 "C. The binding of [3H]caerulein was almost abolished in the presence of 125 pM p-chloromercuribenzoate. A pronounced inhibitory effect was also evident at a 60 pM concentration provided the reagent was present during a 10 minpreincubation period (data not shown).

A series of incubations were conducted in the presence of reagents capable of reducing the intra- cellular ATP concentration. A 50% reduction in maximal binding was obtained when the cells were preincubated for 2 min with 1 mM 2,4-dinitrophenol. Inhibition was also evident with a 30-s preincubation. 0.75 mM 2,4-dinitrophenol was still inhibitory (33 % reduction of binding) but no effect was obtained at 0.50 mM (Fig.4A). Oligomycin (10 pM) and 10 mM NaF exerted almost no immediate effects (Fig.4B) but were strongly inhibitory after 10 min of preincu- bation (Fig. 4 C).

34 Caerulein Receptor of Pancreatic Acinar Cells

I I. , I I

3 1 2 3 4 5 Time (rnin)

Fig. 3. &ffects of Cu" , Mg2', and EGTA OIZ [3H]carrulein hindikg to vui pamwutic ac'inur cdls. All cells were preincubated under the final ionic conditions for 10 min at 37 "C before addition of 3.6 nM [3H]caerulein. (A) The 5-min time study of specific binding was conducted in the presence of either 5 mM EGTA and 6.8 mM Mgz+ (A), 5 mM EGTA and 1 mM Mg2' (O), 2.2 mM Ca" and 6.8 mM Mg2' (+), or 2.2 mM Ca" and 1 mM Mg" (0). (B) Specific binding was recorded after 10 min in the presence of increasing concentrations of Mg''., and four concentrations of calcium (5 mM EGTA and no Ca" added: 0 ; 0 mM Ca" added: 0; 0.5 mM Ca2'.: A; or 2.5 mM Ca" : A). (C) Specific binding was recorded after 10 min in the presence of increasing concentrations of Ca", and three concentrations of magnesium (3.6 mM Mg": A; 1.2 mM Mg2' : 0, or 0 mM Mg2' : A). Results are the means of four experiments performed in duplicate

5 10 15 2 0 25 30

0 5 10 0 5 10 Time (min) Time (min)

Fig. 4. Effects of2,4-dinitrophenol, oligomycin and NaFon the time course of specific hinding of" 3HJcaerulein on ratpancreatic acinar cells. Cells were incubated at 37 "C in the presence oi3.6 nM [3H]caerulein as described under Materials and Methods. The results are the means of three experiments performed in duplicate. (A) The effects of three concentrations of 2,4-dinItrophenol(I .O mM: 0 ; 0.75 mM: .; and 0.5 niM: 0) were compared with control data (0). (B) 1 mM 2,4-dinitrophenol (o), 10 pM oligomycin (0), and 10 mM NaF (A) were added 2 min before addition of the tracer. (C) The cells were preincubated for 10 min with 10 pM oligomycin (0) or 10 mM NaF (A) before addition of the tracer. The 0.6 final concentration of ethanol used in the presence of oligomycin exerted no effect on control data (0)

Binding Specificity and Characteristics using a 10-min incubation period at 37 T, i.e. under of Binding Sites steady-state conditions. Inhibition of binding of

3.6 nM [3H]caerulein by native caerulein could be detected after the addition of 2 nM caerulein, was half- maximal at 18 nM and was maximal at approx. 1 pM

The ability of unlabeled caerulein to inhibit binding of [3H]caerulein to pancreatic acinar cells was tested

J. Christophe, P. De Neef, M. Deschodt-Lanckman, and P. Robberecht 35

loo 1 - Y- -

0 4f I

-8 -7 -6 -5 -4 log Concentration (log M)

Fig. 5 . Inhibition of [ 3H]caerulein hinding to pancreatic minor crll.c by t h i w tiirulogs qf ilie cliolec:l;srokinin-pancreozymin-gastrin family and hy secrefiiz, vasoactive intestinal polypeptide, hornhesin, nonabomhesin, litorin, and carbamoylcholine. Pancreatic acinar cells were incubated for 10 min at 37 "C with 3.6 nM [3H]caerulein in the presence of increasing concentrations of unlabeled caerulein (A), C-terminal octapeptide of pancreozymin (O), pentagastrin (A), desulfated caerulein (O), secretin, vasoactive intestinal polypeptide, bombesin, nonabombesin, litorin, and carbamoylcholine (0). Specific binding was determined as described under Materials and Methods. The results, expressed as percentage of radioactivity bound in presence of tracer only, were the means of seven experiments performed in duplicate

0.15

c .- - 2 8 - ,I 0.1c I

a,

U

U 3

P . $0.05

0

\ \ \

1 2 3 4 Caerulein bound (nM)

Fig. 6 . Typical Scatchard plot of'[ 3HH]cne,ulein specijically hound to dispersed rat pancreatic acinar cells. Cells were incubated for 10 min at 37 "C with 3.6 nM [3H]caerulein and varying concentrations of native caerulein. The bound/free ratio of [3H]caerulein was calcu- lated by subtracting nonspecifically bound radioactivity. The graphic resolution of the curves into two linear components is shown by the broken lines

(Fig. 5) . The three analogs of the cholecystokinin- pancreozymin gastrin family, the C-terminal octa- peptide of pancreozymin, desulfated caerulein, and pentagastrin, inhibited the binding of [3H]caerulein with variable efficiency. On the other hand, caerulein binding sites were not affected by secretin, vasoactive intestinal polypeptide, bombesin, nonabombesin, lito- rin, and carbamoylcholine even when these agents were used at concentrations as high as 0.1 mM (Fig. 5).

Scatchard plots [9] of the results such as those in Fig.5 curved sharply with an upward concavity at high binding site occupancy (Fig. 6). The curve could be described by two straight lines. One possible expla- nation for these results is that there were two functio- nally independent classes of binding sites. Based on

100 - - m

c .- - ._ ._ c 75

." D 5 50 a c a, 3

.- - & 25 s - I I

0

Unlabeled caerulein added

Unspecific --------0- -----." ------- 0 __-- - - - - -____- 0

0 5 10 15 20 25 Time (min)

Fig. 7. Time course of' dissociation of bound [3H]cac.i-ulein from pancreatic acinar cellsat 37 "CandatpH 7.4. Cells were preincubated with 3.6 nM [3H]caerulein at 37 "C for 10 min. The dissociation reaction was initiated by adding 40 pM nonradioactive caerulein (0). No addition was made in the control incubation (A). Very low unspecific binding (0) was observed in a simultaneous incubation conducted in the presence of 40 pM unlabeled caerulein added at time zero of the preincubation. Results are the means of five experi- ments performed in duplicate and are expressed as the percentage of the [3H]caerulein bound to cells at the beginning of the incubation

this assumption the apparent affinity and capacity for each class of binding sites was calculated using results from six separate experiments. These calculations (means 2 S.E.M.) suggested the presence of 67000 k 7000 high-affinity sites per cell with an apparent Kd of 12.1 k 0.8 nM, and 101 000 k 21 000 low affinity sites per cell with an apparent Kd of 72 k 8 nM.

The Dissociation of [3H]Cuerulein

The addition of non-radioactive caerulein to a final concentration of 40 pM caused a marked disso- ciation of bound [3H]caerulein. This dissociation could not be described by a single exponential (Fig. 7 and 8A): during the initial 30 s, 51 % of the bound

Caerulein Receptor of Pancreatic Acinar Cells

I I I 1 I

0 1 2 3 4 5 Tlme (min)

t T c i

1 I I I

0 1 2 3 4 5 Time (min)

Fig.8. Effects of temperature, p H , Ca2' , M g Z 4 , and EGTA on the time course of dissociation of' specifically bound [3H]caerulein ,from pancreatic acinar cells initiated by the addition of 3 p M nonradioactive caerulein. (A) Cells were first preincubated for 5 min at the indicated temperatures (37 "C: 0 ; 22 "C: 0; 15 "C: A, and 3 "C: A) then labeled with 3.6 nM [3H]caerulein at the same temperatures. These tempera- tures were maintained during the dissociation reaction. (B) The specific conditions tested during preincubation (5 min), binding (5 min) in the presence of 3.6 nM [3H]caerulein, and dissociation (5 min) were one of the following pH values: pH 6.0: (a), pH 6.7 (O), pH 7.5 (A), and pH 8.6 (A) adjusted at 37 "C as indicated in Fig.3. (C) The incubation medium at 37 "C contained 5 mM EGTA and 6.8 mM Mg2+ (0) or 2.2 mM Ca2+ and 1.0 mM Mg". (0). All results are the means of four experiments performed in duplicate

[3H]caerulein dissociated and 13 % more after 1 min. The remaining portion of the dissociation curve was a complex exponential process having much lower rate constants. The fraction of the radioactivity remaining bound after 25 min (26 %) was significantly higher than nonspecific binding (Fig. 7). Unlabeled pentagastrin (0.62 mM) was as efficient as 12 pM unlabeled caerulein in inducing the fast dissociation of [3H]caerulein (data not shown).

At 22 "C and 14 "C, the initial component of the dissociation curve of [3H]caerulein was unaltered but the final extent of the dissociation was greater than at 37 "C. At 3 "C there was some reduction in the initial rate of the dissociation but again the final extent of the dissociation was greater than at 37 "C (Fig. 8A).

The first phase of dissociation of [3H]caerulein was unaffected by pH in the 6.0 to 8.6 range. How- ever, the fraction of [3H]caerulein remaining bound after 5 min of dissociation at pH 6.0 and 6.7 was about half of that seen at pH 7.5 and 8.6 (Fig. 8B). Varying the Ca2 '- and Mg2 + concentrations exerted no obvious effects (Fig. 8 C).

A 51-fold cell dilution was as efficient as a large excess (8 pM) of unlabeled caerulein or the combina- tion of these treatments in provoking the release of 71 % of the prebound [3H]caerulein after 7 rnin at 4 "C and at pH 7.4. An ll-fold cell dilution was already capable of provoking the dissociation of 50 % of pre- bound tracer (data not shown).

Neuraminidase (364 or 182 pg/ml), EGTA, the divalent cation ionophore A-23187, 1 mM 2,4-dinitro- phenol, 9 pM oligomycin, or 9 mM NaF did not alter the dissociation curve of [3H]caerulein induced by unlabeled caerulein (data not shown).

DISCUSSION

Specific Binding of [ 3H]Cuerulein

The cell suspensions used in the present investiga- tion contained at least 96 % acinar cells and contami- nation by red blood cells, centroacinar cells, etc. was minimal. It appears therefore justified to consider that our results could be attributed mostly to binding of [3H]caerulein to these cells. We have no direct evidence on the distribution of [3H]caerulein-binding sites in intact cells. A number of similarities between the present data and those obtained with plasma mem- branes [5] strongly suggest that most of the binding in dispersed acinar cells occurred at their plasma membranes. The partial dissociation of bound rH]- caerulein however might suggest that a fraction of caerulein also entered acinar cells (vide infva).

Nonspecific binding of [3H]caerulein did not change as a function of time or temperature, and in the 1 to 10 nM range of [3H]caerulein concentration, it was less than 1-2 % of total binding. Bombesin, nonabombesin, litorin and the muscarinic agent car- bamoylcholine, which also activates calcium outflux [lo, 111, did not affect caerulein binding.

Our results indicate that binding of [3H]caerulein to pancreatic acinar cells was rapid, specific, reversible and dependent on temperature and pH in the incuba- tion medium. Increased binding was obtained by lowering the temperature (Fig. l), in sharp contrast with reduced binding observed with pancreatic plasma membranes exposed to similar changes in temperature [5]. Another characteristic of binding which was distinct from that of plasma membranes [5] was the lack of inhibitory effect of phospholipase A sug-

J. Christophe, P. De Neef, M. Deschodt-Lanckman, and P. Robberecht 37

gesting that this effect when present is exerted on the exposed internal face of the plasma membrane.

Mg” ions and a calciuni-free medium favored the binding of [3H]caerulein probably as a result of an effect on the rate constant of binding (Fig.3) rather than decreased dissociation rate (Fig. 8C).

The effects of pH on the binding of [3H]caerulein although open to criticism due to the unphysiological nature of treatment suggest the contribution of a group having a pK, around 7.5 (Fig.2). p-Chloro- mercuribenzoate, diisopropylfluorophosphate and glutaraldehyde, three protein-modifying reagents cap- able of reacting with thiol groups, serine residues, and NH2 groups, respectively, strongly inhibited the bind- ing of [3H]caerulein to the outer surface of pancreatic acinar cells. These effects were comparable to those obtained on pancreatic plasma membranes [5] and occurred despite the presence of 1 ”/, bovine serum albumin, which added extra amino acid residues to the incubation medium.

The lower steady-state value of [3H]caerulein binding observed with 2,4-dinitrophenol (Fig. 4) might be attributed to an increase in the rate of dissociation of the tracer, since 2,4-dinitrophenol did not exert a significant effect on the rate of binding (Fig.4) and decreased the steady-state value previously obtained in the absence of this compound to a lower level (data not shown).

Dissociution of Bound [ 3H]Cuerulein

The initial rate of dissociation of specifically bound t3H]caerulein (Fig. 7) was significantly greater than that observed from pancreatic plasma membranes [5]. In addition, most of the radioactivity remaining bound after 1 min of dissociation was not released even after prolonged incubation. The latter fraction however could be reduced partially by lowering the tempera- ture and/or the pH in the incubation medium (Fig. 8). Such a phenomenon might correspond to the internali- zation of a fraction of the radioactive material escaping from the two-dimensional space of the membrane surface. The entry of 3H-labeled thyrotropin-releasing factor in a prolactin secreting clonal cell line [12], of ‘251-labeled insulin in human cultured lymphocytes [13], and of ‘251-labeled low density lipoproteins in fibroblasts [14] is well documented.

An ll-fold cell dilution induced the release of a large (70 %) fraction of the amount of prebound tracer which could be released either by a 51-fold cell dilution or a relative 24000-fold excess of unlabeled caerulein (8 pM). The tracer concentration of 0.33 nM obtained with an ll-fold cell dilution can be compared to the apparent Kd of 12 nM tentatively estimated for ‘high- affinity caerulein binding sites’ (vide infru). Dissocia- tion of bound [3H]caerulein did not depend on the concentration of nonradioactive caerulein in the incu-

bation medium suggesting that most of the binding sites in cells loaded at a 3.6 nM concentration were in a relatively low-affinity state. Unfortunately decreas- ing the [3H]caerulein concentration below the concen- tration presumed to provoke accelerated dissociation proved unfeasible since it did not provide sufficient radioactivity for accurate kinetic data to be obtained.

The washing of acinar cells on EHWP Millipore filters required on an average 30 s. Results in Fig.7 indicate that as much as 51 of the bound radioactiv- ity would have been lost by dissociation when routinely using EHWP Millipore filters. Such a figure appears justified on the basis of other experiments using GFB or GFC Whatman Ltd. filters (England). In these experiments [3H]caerulein binding to isolated pan- creatic acinar cells appeared to be 34% higher, on an average, than that estimated when cells incubated in parallel were filtered with EHWP Millipore filters, and this correlated well with a 33 % reduction in the dura- tion ofthe washing procedure (from 30 3 to 20 2 s). The data on binding and dissociation of [3H]caerulein as determined in a limited number of experiments with GFB or GFC Whatman filters were qualitatively similar to those observed with Millipore filters (not shown).

Number and Affinity of [ 3H]Cuerulein Binding Sites

Scatchard plots were curvilinear with an upward concavity (Fig. 8). Several different phenomena could give rise to this relation between [3H]caerulein binding and caerulein concentration [5]. According to one hypothesis, 40 ”/, of [3H]caerulein binding was due to high affinity binding sites whereas the second portion of the curve reflected mostly the contribution of some- what more abundant binding sites having a slightly lower affinity. The number of binding sites estimated for these functionally distinct classes were expressed as sites per cell and the calculation assumed that both types of sites were present on all cells.

It is also possible to interpret our results as not reflecting the presence of two functionally distinct classes of caerulein binding sites but rather the existence of only one class of sites which could exist in a high or a low affinity state. The most compelling observations supporting this conclusion are the dose- response curves of inhibition of [3H]caerulein binding, amylase secretion, and calcium outflux, given in the accompanying paper [6]. The concentrations at which these caerulein analogs produced half-maximal stimu- lation of secretion and calcium outflux were respec- tively, two orders and one order of magnitude lower than that observed for binding. However, the relative potencies of these various analogs in affecting these parameters were the same. To consider pre-existing heterogeneity of caerulein receptors as a mechanism, one would have to postulate the existence of several

38 J. Christophc, P. De Neef, M. Deschodt-Lanckman, and P. Robberecht: Caerulein Receptor of Pancreatic Acinar Cells

classes of hormone receptors possessing the same relative affinities for a number of analogs.

Additional indirect support for negative coopera- tivity is that the dissociation of bound [3H]caerulein from dispersed pancreatic acinar cells could not be described by a simple exponential time curve (Fig. 7).

The first phase of dissociation was so rapid that negative cooperativity could not be demonstrated by the de Meyts procedure [15]. If negative cooperativity was indeed at the origin of this phenomenon, the dif- ficulty in using [3H]caerulein having a relatively low specific activity (13 Ci/mmol) was that the tracer con- centration utilized (3.6 nM) was of the same order as the apparent Kd of high-affinity binding sites (12.1 nM). It might well be that the real value of the Kd of sites in the highest affinity state was even lower than the reported value, were the tracer concentration utilized already able to induce negative cooperativity.

A second phenomenon could also give rise to variable affinity for only one class of caerulein binding sites in intact acinar cells. GTP and guanosine 5’-(p,y- imid0)triphosphate have been found to accelerate the rate of dissociation of [3H]caerulein from pancreatic plasma membranes (unpublished results). It is tempt- ing but speculative to consider that the inhibitory effects of 2,4-dinitrophenol and oligomycin on the binding of [3H]caerulein to dispersed acinar cells (Fig. 4) reflected variations in the concentration of nucleotides at the internal face of plasma membranes. Thus, alterations in the cell concentration of nucleo- tides might also decrease the affinity of caerulein receptors.

Aided by Grant 20.403 from the Fonds de la Recherche Scienti- fique Midicule (Belgium) and grant RO-IAM-17010 from the Natio- nal Institutes of Health (U.S.A.). We thank Dr J. L. Morgat for

supplying [3H]caerulein, Dr de Castiglione for supplying caerulein, bombesin and related peptides, Dr J. Morley for pentagastrin, Dr M. Ondetti for the C-terminal octapeptide of pancreozymin, Dr G. Wald (Bayer-Pharma, Brussels, Belgium) for Trasylol, and Dr J. Hutton and Mrs J. Ballinckx for their assistance in preparing the manuscript.

REFERENCES

1. Deschodt-Lanckman, M., Robberecht, P., de Neef, P., Labrie, F. & Christophe, J . (1975) Gastroenterology, 68, 318-325.

2. Lambert, M., Camus, J. & Christophe, J. (1973) Biochem. Biophys. Res. Cornrnun. 52,935 - 942.

3. Anastasi, N., Erspamer, V. & Endean, R. (1967) Experientiu (Busel) 23, 699 - 704.

4. Christophe, J. P., Frandsen, E. K., Conlon, T. P., Krishna, G. & Gardner, J. D. (1976) J . Biol. Chem. 251, 4640-4645.

5. Deschodt-Lanckman, M., Robberecht, P., Camus, J. & Christophe, J. (1978) Eur. J . Biochem. 91, 21 -29.

6 . Robberecht, P., Deschodt-Lanckman, M., Morgat, J . L. & Christophe, J . (1978) Eur. J . Biochem. 91, 39-48.

7. Christophe, J., Svoboda, M., Deschodt-Lanckman, M., Robbe- recht, P., Morgat, J. L. & Girma, J. P. (1976) in Peptides I976 (Loffet, A,, ed.) pp, 415-426, Editions de 1’Universite de Bruxelles, Brussels.

8. Amsterdam, A. & Jamieson, J. D. (1974) J. Cell Biol. 63,

9. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-675. 1037- 1056,1057- 1073.

10. Deschodt-Lanckman, M., Robberecht, P., de Neef, P., Lam- mens, M. &Christophe, J. (1976) J. Clin. Invest. 58,891 -898.

11. Gardner, J. D., Conlon, T. P., Klaeveman, H. L., Adams, T. D. & Ondetti, M. A. (1975) J . Clin. Invest. 56, 366-375.

12. Gourdji, D., Tixier-Vidal, A., Morin, A,, Pradelles, P., Mor- gat, J.-L., Fromageot, P. & Kerdelhue, B. (1973) Exp. Cell Res. 82, 39 -46.

13. Goldfine, I. D., Smith, G. J., Wong, K. Y. & Jones, A. L. (1977) Proc. Nut1 Acud. Sci. 74, 1368- 1372.

14. Goldstein, J. L. & Brown, M. S. (1977) Annu. Rev. Biochem.46,

15. de Meyts, P., Bianco, A. R. & Roth, J. (1976) J . B id . Chem. 251, 897 - 930.

1877-2888.

J. Christophe, P. De Neef, M. Deschodt-Lanckman, and P. Robberecht, Laboratoire de Chimie Biologique et de la Nutrition, Faculte de Mtdecine et de Pharmacie, Universite Libre de Bruxelles, Boulevard de Waterloo 115, B-1000 Bruxelles, Belgium