A Pseudopeptide That Is a Potent Cholecystokinin Agonist in the Peripheral System Is Able to Inhibit the Dopamine-like Effects of Cholecystokinin in the Striatum*

There are no known specific effective cholecystoki- nin (CCK) receptor antagonists of both peripheral and central nervous systems. Here, we describe experi- ments which demonstrate that a synthetic pseudopeptide analogue of CCK-7 is a potent agonist in the pe- ripheral system and behaves as a selective and highly potent inhibitor of the dopamine-like effects of CCK in the striatum. This compound, t-butyloxycarbonyl-Tyr (SO8H)-NleQ(COCH2)G1y-Trp-Nle-Asp-Phe-NH~, is able to stimulate enzyme secretion from rat pan- creatic acini, with high efficacy and potency. It is also very potent in inhibiting the binding of labeled CCK-8 to rat pancreatic acini (IC60 = 6 nM) and to guinea pig and mouse brain membranes (IC60 = 0.7 nM). However, this compound is able to antagonize the effects of intrastriatally injected t-butyloxycarbonyl-[Nle2s~S’] CCK-8 in mice, with high potency.

Cholecystokinin (CCK),' a gastrointestinal hormone of 33 amino acid residues, was originally isolated from the gastrointestinal tract (1). It is a hormonal regulator of pancreatic and gastric secretions, contraction of the gall bladder, and intestine motility. The complete range of peripheral biological activities has been found in the carboxyl-terminal octapeptide (CCK-8) of the entire molecule (2). CCK-like peptides have since been detected in brain, the sulfated octapeptide (CCK-8) being the most abundant form (3). They are believed to act as neurotransmitters or neuromodulators (4). Cholecystokinin appears to be involved in a wide variety of physiological functions, including satiety ( 5 ) , respiration and thermoregulation (6), sedation (7), and analgesia (8). In addition, cumulative evidence indicates close functional inter-relationships between dopaminergic systems and cholecystokinin in certain brain regions, suggesting that CCK-8 may have a role in the pathogenesis of schizophrenia (9, 10) and in the therapeutic actions of antipsychotic drugs such as neuroleptics (11). The dopaminergic brain areas appear to be particularly rich in CCK-8 (12) and CCK-receptors (13). CCK-8 has been shown to closely interact with dopamine-mediated neurotransmission. CCK-8 appears to antagonize dopamine-induced effects 5 To whom correspondence and reprint requests should be addressed.
The abbreviations used are: CCK, cholecystokinin; BOC, t-butyloxycarbonyl; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesu1fonic acid; BH-CCK-8, Bolton-Hunter-labeled CCK-8. in the mesolimbic system (14, 15), to modulate the sensitivity and the number of striatal and limbic dopamine receptors (16,17), and to decrease dopamine turnover in the striatum (18). It became evident that, particularly to clarify further the role of CCK and dopamine in the central nervous system in general, it was necessary to have selective and potent central CCK-antagonists. Several CCK-receptor antagonists have been described recently, including amino acid (19), peptide (20), and nucleotide derivatives (21), as well as nonpeptidic compounds (22). All of them exhibited a significant degree of selectivity for peripheral CCK-receptors.
McDermott et al. (23) showed that a degradative activity present in synaptosomes and purified lysed synaptosomal membranes cleaves CCK-8 at the Metz8,Glyz9 bond, in uitro. Delineation of the degradative pathway was further reported (24), clearly indicating that brain synaptosomal membranes contain peptidases which cleave CCK-8, the initial endopeptidase cleavage being between Metz8 and GlyZ9. Furthermore, our previous results concerning the importance of the peptide bonds in CCK-related peptides showed that only the bond between Mets and Gly29 could be replaced without affecting either the binding to pancreatic or central CCK-receptors, or the biological activity (32). In order to obtain enzyme-resistant and potent analogues of CCK, we have synthesized the pseudopeptide Boc-Tyr(SO3H)-Nle\k(COCHz)G1y-Trp-Nle-Asp-Phe-NHz (compound 1, Scheme 1) in which the peptide bond between Nle and Gly has been replaced by a ketomethylene bond (COCH,). We report here experiments which demonstrate that compound 1, an analogue of the carboxylterminal heptapeptide of CCK, is a potent CCK agonist in the peripheral system able to antagonize the action of CCK in the mouse striatum.

Methods
Tissue Preparations-Dispersed acini from rat pancreas were prepared according to the previously described modifications (29) of the method of Peikin et al. (21). Mouse and guinea pig brain membranes were prepared following the procedures described by Pelaprat et al.

(30).
Arnyhe Release-Amylase release was measured using the procedure already described (29). Briefly, acini were resuspended in the standard incubation solution complemented with 1% bovine serum albumin, 1 mM calcium, and 5 mM theophylline containing about 1 mg of protein/ml, and samples (0.5 ml) were incubated at 37 'C for 30 min. Amylase activity was determined by the method of Ceska et al. (31) using the Phadebas@ reagent. Amylase release was measured as the difference of amylase activity at the end of incubation that was released into the extracellular medium, with and without secretagogue, and expressed as the percentage of maximal stimulation obtained with B o c -[ N~~~.~~] C C K -~ (40 & 5% of the total amylase contained in the acini) minus the basal amylase secretion (10 & 2% of the total amylase contained in the acini) obtained without secretagogue.
Binding of '261-BH-CCK-8 to Pancreatic Acini-Binding of lZ6I-CCK-8 to rat pancreatic acini was performed as previously described (29). Briefly, samples (0.5 ml containing 4 mg/ml protein) were incubated with the appropriate peptide concentrations for 30 min at 37 "C in the presence of 10 p~ Iz6I-BH-CCK-8 plus various concen- or compound 1. After centrifugation at 10,000 X g for 10 min and washings, the radioactivity associated percentage of the value obtained with labeled CCK-8 alone. The with the acinar pellet was measured. Values are expressed as the specific activities of the various preparations used in our experiments were 2000 Ci/mmol. Acini from three rat pancreata were suspended in 100 mI of standard incubation solution. Specific binding in the absence of any unlabeled CCK-peptide was 13 k 3% of the total radioactivity present in the sample. Nonspecific binding was determined in the presence of 1 p~ B O C -[ N~~*~~~' ] C C K -~ and was always less than 15% of the total binding.
Binding of lZ6I-BH-CCK-8 to Mouse or Guinea Pig Brain Membranes-Binding of '"I-BH-CCK-8 to guinea pig or mouse brain membranes was performed according to Pelaprat et al. (30). The buffer used was 50 mM Tris -HCl, 5 mM MgCl,, 0.1 mg/ml bacitracin (pH 7.4) (Tris-MgClz-bacitracin buffer). Briefly, displacement experiments were performed by incubation of 0.5 ml of brain membranes (approximately 0.5 mg of protein) in the presence of 20 pM lZ5I-BH-CCK-8 for 60 min at 25 "C plus various concentrations of Boc- [NleZ8s31]CCK-8 or compound 1 in a total volume of 0.5 ml. Nonspecific binding was determined in the presence of 1 p~ Bo~-\Nle~*''] CCK-8 and was always less than 25% of the total binding. Total binding was about 10% of the total radioactivity contained in the sample.
Intrastriatal Injection Techniques-The drugs to be tested were dissolved in a pH 6 phosphate buffer. The injections were made freehand in a volume of 1 pl, directly into the right striatum of conscious, nonrestrained mice, by means of a 5-pl Hamilton microsyringe and a 10-mm calibrated needle (gauge, 26s; final length below the skin, 3.5 mm; duration of injection, 2-3 s) (25, 26). The injection point was slightly internal (1 mm) and caudal (1 mm) to the right orbitus. The exact location of injection was verified macroscopically after the experiment (33), the rate of success being approximately 90%. Control mice always received an injection of buffer (1 pl). After injection of the drug or vehicle, mice were placed individually in small square plexiglass cages (10 X 10 X 15 cm). The number of complete turns was counted for 2 min at times 2-4, 4-6, and 6-8 min postinjection. Rotations toward the injection side were named ipsilateral (and noted +), those away from the injection side were named contralateral (and noted -).
Statistics-Statistical analyses were performed using analysis of variance (ANOVAs) followed by Tukey's test for comparison of the means, Homogeneity of variances was assessed using Bartlett's test.

RESULTS
Compound 1 was as potent and efficacious as B~c -[ N l e~"~~] CCK-8, in increasing amylase release from dispersed rat pancreatic acini (Fig. I), maximal stimulation being obtained at a concentration of 0.3 nM. Compound 1 inhibited the binding of 1251-BH-CCK-8 to rat isolated pancreatic acini with almost the same potency as B O C -[ N~~~~*~~] C C K -~ (Fig. 2, ICW = 5 nM). To directly investigate the ability of compound 1 to interact with central CCK-receptors, we measured its ability to inhibit the binding of '251-BH-CCK-8 to guinea pig and mouse brain membranes. Compound 1 was found almost as potent as B O C -[ N~~~~,~~] C C K -~ in inhibiting the binding of labeled CCK-8 to guinea pig and mouse brain membranes (Fig. 3, A and 23, = 0.7 nM). Results are summarized in Table I.
Given our previous observation that the natural CCK-8 exhibited dopaminomimetic properties following intrastriatal injection (25), we have investigated the behavioral effect of B O C -[ N~~*~,~~] C C K -~ and of compound 1 after their direct unilateral injection within the mouse striatum, which is a simple model designated for the screening of drugs, which gave the same effects as dopamine (26). As shown in Fig. 4, B O~-[ N~~~~,~~] C C K -~ injected directly within the mouse striatum induced a strong contralateral turning behavior (termed  binding was 10 f 2% of the total radioactivity present in the sample, and nonspecific binding was always less than 2% of total binding. In each experiment, each value was determined in duplicate, and the results given are the means from five separate experiments.   [ANOVA, F (4,181) = 32.5; p < O.OOl] (Fig. 6), a significant inhibition being already observed at a concentration as little as 0.01 pM. Compound 1 at different concentrations (10 nM to 1 p~) was able to produce a parallel rightward shift in the dose-response curve of the action of intrastriatally injected B O C -[ N~~~~,~~] C C K -~ with no change in the maximal response ( Fig. 7). With increasing concentrations of compound 1, the magnitude of the rightward shift was proportional to the concentration of the pseudopeptide derivative (Fig. 7).

DISCUSSION
In the present study, we investigated a synthetic pseudopeptide analogue of the carboxyl-terminal heptapeptide of CCK in which Met2' and Met31 were replaced by Nle (Scheme 11, a substitution that had proved in many cases, particularly in CCK-derived peptides, t o affect neither the biological activities nor the binding to CCK-receptors (27). The amino terminus of the pseudopeptide was acylated by a Boc group, because it had been demonstrated that CCK-analogues with an Nu-Boc protecting group were more potent than those bearing a free a-amino group (28) injected at very low doses directly into the mouse striatum, induced contra-lateral turning (maximum of activity at approximately 0.01 ng/pl). This compound was approximately 50 times more potent than the natural agonist CCK-8 (maximum of activity at 0.5 ng/rl) when injected directly within mice striatum (25), indicating that acylation of the amino terminus by a Boc group and the replacement of both methionines by norleucines is of a beneficial effect. A similar pattern was obtained with compound Boc-Tyr(S03H)-NleO(COCH2)Gly-Trp-Nle-Asp-Phe-NH2 (1) which exhibited the same efficacy and potency as B O C -[ N~~*~*~~] C C K -~ in stimulating amylase release from isolated rat pancreatic acini. This compound has the same potency as B O C -[ N~~~~*~~] C C K -~ in inhibiting the binding of labeled CCK-8 to rat pancreatic acini and interestingly to mouse or guinea pig brain membranes. In contrast, compound 1 injected directly into the mouse striatum did not induce any clear rotational behavior, but was able to dose dependently antagonize the effects of co-injected B O C -[ N~~~~~~* ] C C K -~. Compound 1 produced a parallel rightward shift in the doseresponse curve for the action 0fBoc-[Nle*~*~~]CCK-8 on turning, indicating that this antagonism is competitive in nature (Fig. 7). Compound 1 is approximately 10,000 times more potent than the well known CCK-antagonist, proglumide, and about 10 to 100 times more potent than 2-CCK-27-32-NHz in inhibiting the dopamine-like effects induced by CCK in the striatum (25). However, neither proglumide nor Z-CCK-27-32-NHz were full agonists in the peripheral system. The present results indicate that compound 1 is able to competitively antagonize CCK in the mouse striatum. That is, compound 1 is devoid of agonist activity, inhibits the interaction of CCK with its cell membrane receptors in the mouse brain, and causes a parallel rightward shift in the dose-response curve for the induced turning caused by CCK. The present findings also indicate that the peptide bond between Nle and Gly is not an essential requirement for the binding to peripheral and central CCK-receptors, but is essential for intrinsic CCK-like biological activity in the central nervous system, at least in the striatum. Although the peptide bond between Nle and Gly is not essential for binding to CCK-receptors (32), it is interesting to notice that its replacement by a ketomethylene bond (COCHz) does not influence at all the apparent affinity with which the pseudopeptide binds to its receptors, either in the peripheral system or in the central nervous system. The present data also indicate that it is possible to obtain very potent peripheral CCK-agonists by acylating the amino terminus in the carboxyl-terminal heptapeptide of CCK, by replacing the methionines by norleucines, and by modifying the peptide bond between norleucine and glycine by a ketomethylene bond in the heptapeptide; this modification led us to obtain a very potent selective central CCKantagonist. To our knowledge, compound 1 is the only described potent selective central CCK-antagonist. It provides an interesting insight in the knowledge of structure-activity relationships of CCK-related peptides and a very useful tool for the understanding of CCK actions and the study of its pharmacology particularly in the central nervous system.
Considerable controversy exists in the literature as to whether CCK-peptides facilitate or decrease dopamine transmission in this brain area (9, 10,15,25). The results obtained in the present study are in accordance with a mediation of CCK-induced turning behavior by CCK-receptors which are present within the striatum and suggest that, under acute conditions, CCK exerts a facilitatory influence on dopamine transmission within the striatum.