Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Kappa opioid binding sites have been previously characterized in mammalian brain in rats, guinea pigs, rhesus monkeys and humans (Zukin et al., 1988; Clark et al., 1989; Nock et al., 1990; Kim et al., 1996; Emmerson et al., 1994). In rats, guinea pigs and humans, it has been suggested that at least two distinct populations of kappa receptors could be detected. These have been variously defined by different authors, but some characteristics are common to several naming systems. For example, the prototypic arylacetamide congeners (e.g., U50,488 and U69,593) display high affinity for a subset of kappa receptors, which tend to be termed kappa-1 (Zukin et al., 1988; Clark et al., 1989). Other kappa opioid ligands, in particular, the benzomorphans [e.g., EKC, bremazocine], also display high to moderate affinity for so-called kappa-2 sites (Zukin et al., 1988). A third subpopulation (kappa-3) has also been proposed (Clark et al., 1989; Pasternak, 1993). Kappa-1 and kappa-2 sites display a differential distribution within the nervous system of a single species, and differential proportions between different species (e.g., guinea pigs and rats; Zukin et al., 1988). Recent radioligand binding studies in human cortex membranes also support the notion of at least two kappa receptor subpopulations similar to kappa-1 and kappa-2 (Kim et al., 1996), whereas others (Rothman et al., 1992) have described further subdivisions.

Several in vivo studies in rodents have also supported the notion of kappa receptor subpopulations. For example, an arylacetamide isothiocyanate derivative known as UPHIT selectively antagonized the antinociceptive effects of U69,593 but not bremazocine in mice (Horan et al., 1991). Similarly, cross-tolerance studies with these two agonists supported the conclusion that they were stimulating two functional subpopulations of kappa receptors (Horan and Porreca, 1993). Kappa opioid receptors have been cloned for a variety of species, including rats and humans (Raynor et al., 1994; Simonin et al., 1995; Zhu et al., 1995). In functional studies of the cloned kappa receptors, only one receptor population is usually detected, with greatest similarity to the high-affinity kappa-1 population, as earlier described by radioligand binding and in vivo pharmacology studies (Lai et al., 1994; Raynor et al., 1994; Simonin et al., 1995; Zhu et al., 1995). Intriguingly, Mansour et al. (1995) reported that dynorphin A(1-13) and -neoendorphin displaced only 70% to 80% of [³H]EKC bound to cloned human and rat kappa receptors and suggested that this may have been due to a lack of access by opioid peptides to a particular region within the kappa receptors.

The pharmacology of several kappa opioid agonists has been evaluated in rhesus monkeys in a variety of assays, including antinociception, sedation, diuresis and drug discrimination (Dykstra et al., 1987a; France et al., 1994). The effects of arylacetamide and benzomorphan kappa agonists are not readily differentiated in any of the above procedures in terms of selectivity for a particular effect. For example, representative compounds from the arylacetamide (e.g., U50,488) and benzomorphan (e.g., EKC) classes exhibit qualitatively similar antinociceptive, discriminative and diuretic effects in rhesus monkeys (Dykstra et al., 1987a). Similarly, their sensitivity to the opioid antagonist quadazocine cannot be readily differentiated in antinociception and drug discrimination assays (Dykstra et al., 1987b; see also France et al., 1994). Arylacetamides and benzomorphans can be more readily differentiated in an assay of respiratory depression in rhesus monkeys, in which the arylacetamides have a limited effect, whereas the benzomorphans have a more pronounced effect (Howell et al., 1988; France et al., 1994). However, this difference is more likely due to agonist effects of the benzomorphans at mu receptors than at a particular kappa receptor site (e.g., McGilliard and Takemori, 1978; Howell et al., 1988; Butelman et al., 1993a).

Some in vivo support for the existence of kappa subpopulations in primates was obtained with the selective kappa antagonist nor-BNI (3.2 mg/kg s.c.), which antagonized the antinociceptive effects of some (e.g., U50,488 and U69,593) but not other [e.g., CI-977 (enadoline), EKC and bremazocine] kappa agonists in the warm water tail withdrawal assay in rhesus monkeys (Butelman et al., 1993b).

Kappa-receptor subtype binding sites have not been extensively characterized in rhesus monkeys (Young et al., 1986; France et al., 1994; Emmerson et al., 1994). Our aim in the present study was therefore to confirm in a test of thermal antinociception in rhesus monkeys whether such subtypes could be detected in this species and whether these subtypes could explain the apparent selectivity of antagonism of nor-BNI.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Warm Water Tail Withdrawal

Animals. One male and two female adult rhesus monkeys (Macaca mulatta) were used; their body weights ranged between 9 and 13.2 kg. They were housed individually with free access to water and were fed ~30 biscuits (Purina monkey chow) daily, supplemented with fresh fruit twice weekly. Animals used in this study were maintained in accordance with the University Committee on the Use and Care of Animals, University of Michigan, and Guidelines of the Committee on the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Health Council (Department of Health, Education and Welfare, publication no. NIH 85-23, revised 1983).

Procedure. The procedure used in the present study was similar to that described by Dykstra and Woods (1986). The subjects were seated in primate restraint chairs, and the lower part of the shaved tail (~15 cm) was dipped into warm water maintained at temperatures of 40°, 50° and 55°C. Tail withdrawal latencies were timed manually by a push-button switch connected to a computer. If an animal failed to remove its tail from water, the maximum (cutoff) latency, 20 sec, was recorded. Each experimental session began with control determinations at each temperature. Agonist dosing then began, using a cumulative dosing procedure, with a 30-min interinjection interval. Agonist dosing occurred at the beginning of each cycle; starting 15 min after injection, the animals were tested at three temperatures in a varying order, with ~2-min intervals between tests in the same animal. Agonist doses increased by 0.25 or 0.5 log units in each consecutive cycle. Experimental sessions were conducted no more than twice weekly, with a minimum of 72 hr between sessions (other than in experiment 3; see table 1).

Design. Base-line dose-effect curves for different opioid agonists (U50,488, EKC, CI-977 and alfentanil) were determined. The present data were obtained from three separate experiments in which nor-BNI (10 mg/kg) was subcutaneously administered on day 0. Cumulative dose-effect curves were then redetermined for the above opioid agonists, according to the order shown in table 1. The interval between the end and the beginning of consecutive experiments was 2 weeks.

Data analysis. Individual tail withdrawal latencies were converted to % MPE by the following formula: % MPE = [(test latency  control latency)/(cutoff latency  control latency)] × 100%. Mean A₅₀ values were obtained from individual A₅₀ values, which were calculated by least-squares regression using the portion of the dose-effect curves spanning the 50% MPE. In addition, 95% confidence limits were calculated after log transformation of individual A₅₀ values. Mean A₅₀ values were considered to be significantly different from base line when their 95% confidence limit values did not overlap. Dose ratios were calculated by dividing mean A₅₀ values in the presence of nor-BNI by the base-line A₅₀ values.

Radioligand Binding

Membrane preparation. The brain tissue used for all experiments was obtained from one male adult rhesus monkey (weight ~11 kg). After euthanasia by pentobarbital (100 mg/kg i.v.), the entire brain was excised and immediately placed in ice-cold Tris-HCl buffer (50 mM Tris, pH 7.4 at 25°C, 0.1 mM phenylmethylsulfonyl fluoride); all further steps were performed at 4°C. The brain tissue was cut into small pieces and washed in ice-cold Tris several times. Membrane vesicles were prepared in a Dounce homogenizer (pestle clearance, 60-90 µm) and centrifuged at 40,000 × g for 10 min. The membrane pellets were resuspended (5 ml per g of wet weight) and stored at 70°C.

Binding experiments. Kappa-1 subtype receptor binding was determined using [³H]U69,593 (0.8-1.1 nM in 1 ml assay volume). We used 1 µM unlabeled U69,593 as non-specific binding definition. [³H]Bremazocine binding (0.8-1.2 nM in 0.5 ml assay volume) to kappa receptors (kappa-all binding) was determined in the presence of 1 µM DAMGO and 1 µM DPDPE to prevent bremazocine binding to mu or delta receptors. The non-specific binding definition was 10 µM bremazocine. Mu receptor binding was determined using [³H]DAMGO (1.0-1.5 nM in 0.5 ml assay volume) and 1 µM unlabeled DAMGO as nonspecific binding definition. Delta receptor binding was assayed with [³H]p-Cl-DPDPE (1.0-1.5 nM in 0.5 ml assay volume) with 10 µM unlabeled DPDPE as nonspecific binding definition. Binding was linear up to 400 µg of protein/assay (protein was determined according to Bradford, 1976). After incubation for 2 to 2.5 hr at 25°C, assay volumes were filtered through Whatman GF/B glass-fiber filters using a Brandel cell harvester (Gaithersburg, MD). Before use, filters were soaked in 0.05% (v/v) polyethyleneimine to decrease nonspecific radioligand binding. The radioactivity was counted by liquid scintillation in the presence of 3.5 ml of Ultima Gold scintillation fluid (Packard, Meriden, CT).

Data analysis. IC₅₀ values were determined using the InPlot program (GraphPAD Software, San Diego, CA). K_i values were calculated according to Linden (1982). Statistical significance of a between-group difference was based on two-tailed t tests of pK_i values.

Drugs. In antinociception studies, alfentanil HCl (National Institute on Drug Abuse, Bethesda, MD), CI-977 (Warner Lambert/Parke-Davis, Ann Arbor, MI), EKC methanesulfonate (Sterling Winthrop, Rensselaer, NY), nor-BNI (Dr. H. I. Mosberg, Department of Medicinal Chemistry, University of Michigan, Ann Arbor, MI) and U50,488 (Upjohn Co., Kalamazoo, MI) were dissolved in sterile water and administered subcutaneously in the scapular region at a volume of 0.1 ml/kg. The above drugs were also used in the radioligand binding studies, in addition to: bremazocine methanesulfonate (Sandoz, Basel, Switzerland), butorphanol tartrate (Bristol Myers, Wallingford CT), BW373U86 (Burroughs Wellcome, Research Triangle Park, NC), DAMGO and DPDPE (Dr. H. I. Mosberg), etonitazene HCl, dynorphin A(1-13) and A(1-17) (National Institute of Drug Abuse), E-2078 ([N-methyl-Tyr¹,N-methyl-Arg⁷,D-Leu⁸]dynorphin A(1-8) ethylamide; Eisai Co., Tsukuba, Japan), morphine sulfate (Mallinckrodt, St. Louis, MO) and U69,593 (Upjohn).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Warm Water Tail Withdrawal

Control tail withdrawal latencies and base-line dose-effect curves. The monkeys used in this study showed a consistent profile in tail withdrawal responses. They kept their tails in 40°C water for 20 sec (cutoff latency) and removed their tails from 50°C and 55°C water very rapidly (typically within 1.5 sec). During the first 3 hr after injection of nor-BNI alone, there were no elevated tail withdrawal latencies in 50°C and 55°C water (data not shown). Furthermore, no antinociceptive effects of nor-BNI were observed from 24 hr and beyond.

Selectivity of antagonism study. Pretreatment with 10 mg/kg nor-BNI caused a different pattern of rightward shifts of dose-effect curves among the presently studied opioid agonists. The cumulative dose-effect curves of the four opioid agonists were studied between 1 and 35 days after nor-BNI administration (figs. 1 and 2).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Cumulative dose-effect curves in 50°C water for U50,488 (top left), EKC (top right), CI-977 (bottom left) and alfentanil (bottom right) after pretreatment with 10 mg/kg of nor-BNI. Abscissae: agonist dose in mg/kg; ordinates: % MPE (±S.E.M.). Dose-effect curves were determined either before nor-BNI (base line, ) or 1 (), 3 (), 7 () 10 (), 14 () or 35 () days after nor-BNI. The U50,488 dose-effect curves at 10, 17, 21 and 28 days are not shown for the sake of clarity.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   A50 values for U50,488, EKC, CI-977 and alfentanil after pretreatment with 10 mg/kg nor-BNI. Abscissae: pretreatment time (days) after nor-BNI. The open symbol above BL represents the base-line value. Ordinates: mean A50 values (mg/kg) calculated by linear regression from individual dose-effect curves.

Radioligand binding. Table 3 summarizes radioligand binding affinities and binding site densities in whole rhesus monkey brain membranes as determined in equilibrium saturation binding experiments of [³H]DAMGO (mu), [³H]p-Cl-DPDPE (delta), [³H]U69,593 (kappa-1) and [³H]bremazocine (kappa-all). The K_d values for these radioligands, as determined in the equilibrium saturation binding experiments, were essentially identical to the K_i values of the unlabeled compounds as determined in equilibrium radioligand displacement experiments (3.2 nM for [³H]DAMGO vs. 3.3 nM for DAMGO, 2.3 nM for [³H]p-Cl-DPDPE vs. 2.4 nM for DPDPE, 1.2 nM for [³H]U69,593 vs. 1.3 nM for U69,593 and .39 nM for both labeled and unlabeled bremazocine). This indicates good internal consistency of the binding experiments. The B_max value for the kappa-1 receptor subtype was 29% of the B_max value determined with bremazocine, a ligand that presumably interacted with both kappa-1 and kappa-2 receptor subtypes.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The present study documents the presence of two kappa receptor binding sites in rhesus monkeys, a species in which kappa agonist pharmacology has been investigated in a variety of assays (e.g., antinociception, sedation, diuresis and drug discrimination; Dykstra et al., 1987a, 1987b; France et al., 1994). The presently characterized kappa receptor subpopulations have some similarity with those previously reported in rats and guinea pigs. For example, a smaller population was found, for which arylacetamides (such as U50,488 and U69,593) had high affinity, whereas a larger population was found for which benzomorphans, but not U50,488 or U69,593, had high affinity (e.g., Zukin et al., 1988; Nock et al., 1990). These findings further reinforce the notion that the occurrence of subpopulations of kappa opioid receptors is a robust finding in several species, including human (e.g., Rothman et al., 1992; Kim et al., 1996) and nonhuman primates. All of the binding data in this report are derived from the brain of a single male animal; however, the presently obtained K_d and B_max values for [³H]U69,593 and [³H]p-Cl-DPDPE are within a 2- to 3-fold range of values determined previously from a different monkey brain (Emmerson et al., 1994). A slightly greater difference was observed between the presently obtained [³H]DAMGO saturation binding data and previously reported values (Emmerson et al., 1994) but only with respect to K_d (and not B_max) values. Therefore, there seem not to be major differences between the binding characteristics of the brains from these two animals.

Binding evidence in other species suggested that the kappa antagonist nor-BNI had a relative selectivity (10-fold) for kappa-1 receptors (i.e., in rats and humans, see Nock et al., 1990; Kim et al., 1996). Consistent with this, a pretreatment dose of nor-BNI (3.2 mg/kg s.c.) antagonized the effects of two arylacetamides (and putative kappa-1 agonists) U50,488 and U69,593 but not those of the benzomorphans EKC or bremazocine in the warm water tail withdrawal assay in rhesus monkeys (Butelman et al., 1993b). Notably, the arylacetamide CI-977 was not antagonized in that study, indicating that arylacetamide structure of an agonist is not necessarily predictive of its sensitivity to nor-BNI in vivo. In the present radioligand binding studies, nor-BNI displayed only a very moderate binding selectivity for [³H]U69,593 sites, relative to [³H]bremazocine sites. This suggested that nor-BNI would have, at most, a very moderate antagonist selectivity against in vivo effects mediated by kappa-1 vs. non-kappa-1 receptors. This suggestion was presently tested by studying the antagonist effects of a larger dose of nor-BNI in the same assay (10 mg/kg s.c.) than had previously been studied (i.e., 3.2 mg/kg; Butelman et al., 1993b). In the present study, nor-BNI was again found to be a very long-lasting kappa-selective antagonist, as shown by the prolonged antagonism of U50,488 and the minimal or absent effects against the selective mu agonist alfentanil. A slightly smaller and relatively less long-lasting (i.e., 3-10 days), but significant, degree of antagonism was observed against the benzomorphan kappa agonist EKC. Thus, consistent with the present nor-BNI binding data, a sufficiently large nor-BNI dose can block the antinociceptive effect of an agonist that presumably acts on kappa-2 receptors. Recent evidence suggests that the antinociceptive effects of EKC in rhesus monkeys may also be partially mediated by mu receptors (Ko et al., in press). However, this is unlikely to be the reason for the presently reported antagonism of EKC by nor-BNI (10 mg/kg) because this nor-BNI dose was ineffective against the mu-selective agonist alfentanil. In previous antinociception studies in mice, nor-BNI exhibited some mu-antagonist effects but only up to several hours after administration, whereas its selective kappa antagonist effects were observable for several days after administration (e.g., Endoh et al., 1992; Broadbear et al., 1994).

In the present study, CI-977 (enadoline) was not significantly antagonized by nor-BNI (10 mg/kg). Other studies in rhesus monkeys are consistent with a kappa receptor mediation of the effects of CI-977. For instance, CI-977 was generalized by monkeys trained to discriminate EKC from saline (France et al., 1994), and its antinociceptive effects exhibited a similar sensitivity to naltrexone antagonism as bremazocine (Ko et al., in press). This suggests that the antinociceptive effects of CI-977 occur through a kappa receptor population that is not highly sensitive to nor-BNI. It should be noted that nor-BNI has previously antagonized the discriminative stimulus effects of CI-977 in squirrel monkeys (Bergman and Carey, 1997). This suggests that the present lack of nor-BNI sensitivity by CI-977 may not be observed across species or assays. CI-977 displayed an 11-fold binding selectivity for kappa-1 vs. kappa-all sites, in the same order as the arylacetamide congeners U69,593 and U50,488. Thus, the binding profile of CI-977 alone does not explain its lack of nor-BNI sensitivity because U50,488 and U69,593 are sensitive to nor-BNI antagonism (present study; Butelman et al., 1993b).

It is notable that both butorphanol and nalbuphine, compounds that are in clinical use (e.g., Gear et al., 1996), have some kappa-1 vs. kappa-all selectivity. Furthermore, the affinity of butorphanol for mu and kappa-1 sites appears to be almost equivalent, whereas nalbuphine has a 6.1-fold mu vs. kappa-1 selectivity. These affinity measures are consistent with the antagonist profile of these two compounds in a test of antinociception in squirrel monkeys (Dykstra, 1990). In that study, the apparent pA₂ value of butorphanol against a mu agonist (methadone) and putative kappa-1 agonist (U50,488) were equivalent, whereas nalbuphine was ~8-fold more potent in antagonizing methadone than U50,488 (Dysktra, 1990). Butorphanol and nalbuphine have both been evaluated in rhesus monkeys, and in this species they exhibit mu partial agonist effects when administered alone (e.g., Young et al., 1984; Walker et al., 1993; Gerak et al., 1994, Gerak and France, 1995; Butelman et al., 1995). The putative kappa receptor-mediated effects of butorphanol and nalbuphine are therefore more clearly investigated when their mu receptor-mediated effects are blocked by the pseudoirreversible, mu-selective antagonist clocinnamox (Vivian et al., 1997). For instance, under these conditions, a diuretic effect of butorphanol was detected (Vivian et al., 1997).

Overall, the present results document the presence of kappa-opioid binding subtypes in rhesus monkey brain. The characteristics of the two presently described populations have some similarity to those previously reported in rat, guinea pig and human brain, and as predicted by binding data, the kappa-selective antagonist nor-BNI displays a slight degree of selectivity for effects mediated by kappa-1 vs. non-kappa-1 receptors.