Abstract
The behavioral effects of the nonpeptidic delta opioid agonist SNC80 and a series of related piperazinyl benzamides derived from the parent compound BW373U86 were evaluated in rhesus monkeys. SNC80 (0.1–10 mg/kg) decreased response rates maintained by food-reinforcement in a dose- and time-dependent manner, with maximal effects occuring within 10 min of intramuscular injection. The potency of SNC80 and five other piperazinyl benzamides in this assay of schedule-controlled responding correlated with their affinity at cloned human delta opioid receptors but not with their affinity for cloned human mu receptors. Moreover, the effects of SNC80 were selectively antagonized by thedelta-selective antagonist naltrindole (1.0 mg/kg), but not by the mu selective antagonist quadazocine (0.1 mg/kg) or the kappa-selective antagonist nor-binaltorphimine (3.2 mg/kg). These findings indicate that SNC80 functions as a systemically active, delta-selective agonist with a rapid onset of action in rhesus monkeys. The antinociceptive effects of SNC80 were examined in a warm-water tail-withdrawal assay of thermal nociception. SNC80 (0.1–10 mg/kg) produced weak but replicable antinociceptive effects that were antagonized by naltrindole (1.0 mg/kg). SNC80 antinociception was also dose-dependently antagonized by BW373U86 (0.56–1.0 mg/kg), which was inactive in this procedure. These findings suggest that SNC80 may have higher efficacy than BW373U86 at delta opioid receptors. Moreover, SNC80 at doses up to 32 mg/kg did not produce convulsions, which suggests that SNC80 may also be safer than BW373U86. The effects of SNC80 were also examined in monkeys trained to discriminate cocaine (0.4 mg/kg i.m.) or self-administer cocaine (0.032 mg/kg/injection,i.v.). In drug discrimination studies, SNC80 (0.1–10 mg/kg) produced a dose-dependent and naltrindole-reversible increase in cocaine-appropriate responding, and complete substitution for cocaine was observed in five of seven monkeys tested. However, SNC80 (1.0–100 μg/kg/injection) did not maintain responding in monkeys trained to self-administer cocaine. Thus, despite its ability to produce cocaine-like discriminative stimulus effects, SNC80 may have relatively low abuse potential.
Opioids are thought to produce their effects by acting at three main types of opioid receptors, the mu, kappa anddelta receptors (Martin et al., 1976; Lordet al., 1977; Wood et al., 1981; Kieffer, 1995). Until recently, the only agonists available for probing the function ofdelta opioid receptors were peptidic compounds such as the highly selective delta agonists DPDPE (Mosberg et al., 1983) and [d-Ala2]deltorphin II (Erspameret al., 1989). Central administration of these peptidicdelta agonists produces antinociception in rodents (Heymanet al., 1987; Kovacs et al., 1988; Calcagnetti and Holtzman, 1991; Stewart and Hammond, 1994). These antinociceptive effects appear to be mediated by delta opioid receptors because they are are selectively antagonized by delta opioid antagonists such as naltrindole (Portoghese et al., 1988). Centrally administered delta peptides also produce abuse-related effects in rodents such as conditioned place preferences (Shippenberg et al., 1987), decreases in thresholds for electrical brain stimulation (Duvauchelle et al., 1997), maintenance of drug self-administration (Devine and Wise, 1994) and substitution for the discriminative stimulus effects of known drugs of abuse such as mu agonists (Shearman and Herz, 1982; Ukai and Holtzman, 1988) and cocaine (Ukai et al., 1993; Suzukiet al., 1997). Moreover, like cocaine and many other abused drugs, peptidic delta agonists increase extracellular levels of dopamine in the nucleus accumbens (Spanagel et al., 1990;Longoni et al., 1991), a neuroanatomic site that has been implicated in the reinforcing effects of drugs (see Koob, 1992, for review). Although these behavioral and neurochemical effects suggest that stimulation of delta receptors may be associated with relatively high abuse potential, few other side effects have been attributed to peptidic delta agonists. For example, peptidicdelta agonists produce relatively mild effects on respiration (Kiritsy-Roy et al., 1989) and gastrointestinal motility (Porreca et al., 1984). Taken together, these studies suggest that delta agonists may produce clinically useful analgesic effects without producing many of the side effects associated with morphine-like opioid agonists. However, it is unlikely that these peptidic delta agonists could be used for the clinical treatment of pain because they do not readily cross the blood-brain barrier after systemic administration.
BW373U86 was recently described as the first systemically active, nonpeptidic agonist selective for delta opioid receptors (Chang et al., 1993). In vitro, BW373U86 binds with subnanomolar affinity and moderate (10–20-fold) selectivity todelta opioid receptors and potently inhibits electrically evoked contractions of the mouse vas deferens, a smooth muscle preparation sensitive to peptidic delta agonists (Changet al., 1993). In vivo, BW373U86 and the peptidicdelta agonists produce similar but not identical behavioral profiles (see Negus and Picker, 1996, for review). First, BW373U86 produces naltrindole-reversible antinociception in mice, but it is less effective than peptidic delta agonists. For example, although both BW373U86 and peptidic agonists produce antinociceptive effective in models of inflammatory pain that use chemical noxious stimuli (Wild et al., 1993; Takasuna et al., 1994), BW373U86 is not effective in producing antinociception when high intensity noxious thermal stimuli are used (Wild et al., 1993). This difference suggests that BW373U86 may have relatively lower efficacy at delta receptors than peptidic agonists. Second, BW373U86 produces some mu receptor-mediated effects in addition to its delta agonist effects in mice, indicating that BW373U86 is less selective than peptidic delta agonists (e.g., Wild et al., 1993). Third, BW373U86 and peptidic delta agonists do not produce identical discriminative stimulus effects in pigeons trained to discriminate either BW373U86 or DPDPE from saline (Comer et al., 1993b; Jewett et al., 1996). Finally, BW373U86 produces convulsant effects in mice (Comer et al., 1993a), whereas peptidic delta agonists do not consistently produce convulsions and may produce anticonvulsant effects under some conditions (Tortella et al., 1988).
The availability of BW373U86 has facilitated the evaluation ofdelta agonist effects in primates. For example, BW373U86 produced dose-dependent rate-decreasing effects in rhesus monkeys responding for food under an FR schedule of reinforcement (Neguset al., 1993, 1994). Antagonism studies with receptor-selective opioid antagonists suggested that these effects of BW373U86 were mediated by delta opioid receptors and provided the first unequivocal evidence for a deltareceptor-mediated behavioral effect in rhesus monkeys. BW373U86 also produced antinociceptive effects in a model of bradykinin-induced hyperalgesia and allodynia, suggesting that BW373U86 might be effective in treating pain associated with inflammation (Butelman et al., 1995). However, in primates as in rodents, BW373U86 did not produce antinociception when high intensity noxious thermal and electrical stimuli were used, although it did produce convulsant effects (Dykstra et al., 1993; Negus et al., 1994). Unlike peptidic delta agonists in rodents, BW373U86 did not produce reinforcing effects or generalization to the discriminative stimulus effects of mu opioid agonists or cocaine in rodents, but similar to peptidic delta agonists, BW373U86 produced minimal respiratory depressant effects (Dykstraet al., 1993; Negus et al., 1994, 1995).
In view of BW373U86’s relatively low selectivity for delta vs. mu opioid receptors, modest antinociceptive activity and pronounced convulsant activity, there have been efforts to develop more selective, more efficacious and safer deltaagonists. One promising series of compounds has been synthesized byCalderon et al. (1994). These investigators resolved the isomers of BW373U86 and synthesized a series of BW373U86 derivatives, some of which display much greater selectivity for delta vs. mu opioid receptors than the parent compound. For example, SNC80 ((+)-4-[(aR)-a-((2S,5R)-4-allyl-2,5-dimethyl-1-piperazinyl)-3-methoxybenzyl]-N,N-diethylbenzamide), the O-methylated derivative of (+)-BW373U86 (fig.1), has an affinity of ∼1 nM fordelta opioid receptors. Moreover, SNC80 has ≥500-fold selectivity for delta vs. mu opioid receptors, which is much greater than the receptor selectivity of the parent compound BW373U86 and similar to that of the most selective peptidic delta agonists (Calderon et al., 1994;Bilsky et al., 1995; Knapp et al., 1996).
In vivo functional studies in mice also suggest that SNC80 may provide a significant advance over BW373U86. First, SNC80 is more effective than BW373U86 in producing antinociception. For example, i.p. SNC80 produced antinociception in the warm-water tail flick test, whereas i.p. BW373U86 was not effective (Wild et al., 1993;Bilsky et al., 1995). Second, the antinociceptive effects of SNC80 appeared to be mediated entirely by delta opioid receptors (Bilsky et al., 1995; Kamei et al., 1995), whereas the antinociceptive effects of BW373U86 appeared to be mediated by both delta and mu receptors (Wildet al., 1993). Finally, effective antinociceptive doses of SNC80 were reported to produce a lower incidence of convulsions than BW373U86. These findings suggest that novel delta agonists such as SNC80 may be more selective, more efficacious and safer than BW373U86 and may represent more promising compounds for development as analgesics. However, the effects of SNC80 and related compounds in species other than mice have not been reported. Moreover, it is not known whether the enhanced delta-receptor selectivity, analgesic activity and safety of SNC80 is associated with a higher potential for abuse. Finally, the behavioral effects of other piperazinyl benzamides structurally related to BW373U86 and SNC80 are not known, although some of these compounds have an affinity and selectivity for delta receptors similar to or greater than that of SNC80 (Knapp et al., 1996; Calderon et al., 1997).
Accordingly, the purpose of the present study was to evaluate the behavioral effects of SNC80 and related compounds in rhesus monkeys under conditions similar to those we used to examine BW373U86 (Neguset al., 1993, 1994, 1995). Initially, the time course, potency and delta receptor selectivity of SNC80 and other piperazinyl benzamides were evaluated in an assay of food-maintained, schedule-controlled behavior. Subsequently, the antinociceptive effects of selected compounds were examined in a warm-water tail-withdrawal assay of thermal nociception. In addition, the overt behavioral effects, and in particular the possible convulsant effects, of SNC80 were investigated in observational studies. Finally, the abuse-related effects of SNC80 were evaluated in two assays. First, because peptidicdelta agonists have been found to produce cocaine-like discriminative stimulus effects in rodents (Ukai et al., 1993; Suzuki et al., 1997), we examined the effects of SNC80 in rhesus monkeys trained to discriminate cocaine from saline. Second, the reinforcing effects of SNC80 were evaluated in an assay of drug self-administration, in which SNC80 was substituted for cocaine in monkeys trained to self-administer cocaine.
Methods
Subjects
Fourteen male and six female rhesus monkeys (Macaca mulatta) weighing 4.5 to 12 kg were used as subjects. All monkeys had prior exposure to drugs (primarily dopaminergic and opioid compounds) and behavioral procedures. The subjects were individually housed, and water was freely available. Their diet consisted of PMI Feeds Jumbo monkey diet (2–6 biscuits/day for monkeys in operant studies involving food-maintained responding, 4–6 biscuits/day for monkeys used in tail-withdrawal and observational studies) and was supplemented with fresh fruit twice daily. A 12-hr light/dark cycle was in effect (lights on from 7:00 a.m. to 7:00 p.m.). All housing and procedures were in compliance with NIH guidelines on care and use of animal subjects in research and were approved by the McLean Hospital Institutional Animal Care and Use Committee.
Assay of Schedule-Controlled Behavior
Apparatus and procedure.
The effects of deltaagonists on schedule-controlled behavior were studied in four monkeys, and experiments were conducted in each monkey’s home cage (dimensions: 60 × 65 × 75 cm). The home cages of all monkeys were modified to include an operant response panel (28 × 28 cm) mounted on the front wall. Three square translucent response keys (6.4 × 6.4 cm) were arranged 2.54 cm apart in a horizontal row 3.2 cm from the top of the operant panel. Each key could be transilluminated by red, green or yellow stimulus lights (Superbright LEDs). In addition, three circular translucent panels (1.9 cm in diameter) were located in a vertical column below the center response key and could be transilluminated by red, green or yellow stimulus lights (Superbright LEDs). The operant panel also supported an externally mounted pellet dispenser (Gerbrands, Model G5210) that delivered 1-g fruit-flavored food pellets (Precision Primate Pellets Formula L/I Banana Flavor or Precision Purified Pellets Formula L/P Grape Flavor, P. J. Noyes Co., Lancaster, NH) to a food receptacle mounted on the cage beneath the operant response panel. The panel was controlled by a MED-PC interface and an IBM-compatible computer programmed in MEDSTATE Notation (MED Associates, East Fairfield, VT).
Experiments incorporated a multiple-cycle procedure. For training, each of five cycles was 15 min long and consisted of two components: a 10-min pretreatment period followed by a 5-min response period. During the pretreatment period, no stimulus lights were illuminated and responding had no scheduled consequences. During the response period, the center key was transilluminated yellow, and the subjects could respond for up to 10 food pellets on an FR30 schedule of reinforcement. If all 10 food pellets were earned before 5 min had elapsed, the lights were turned off, and responding had no scheduled consequences for the remainder of that response period. All monkeys were trained until they responded at rates >1.0 responses/sec during all five cycles for 10 consecutive days. Sessions were conducted 5 days a week.
Test sessions were conducted no more than twice a week and only after a training session during which the monkeys responded at rates greater than 1.0 responses/sec for all five cycles. Test sessions were conducted using two different protocols, a time course protocol and a cumulative dosing protocol. In the time course protocol, a single dose of the test compound was injected at the beginning of the test session, and 5-min response cycles identical to those described above began at 10, 30, 100 and 300 min and 24 hr after the injection. All time points were examined during a single test session. In the cumulative dosing protocol, test sessions were identical to training sessions except that a dose of the test compound was administered at the beginning of each 15-min cycle, and each dose increased the total cumulative dose by ¼ or ½ log units. In experiments examining the effects of pretreatment with the delta-selective opioid antagonist naltrindole (1.0 mg/kg) and the mu-selective opioid antagonist quadazocine (0.1 mg/kg), a single dose of the antagonist was administered 15 min before the beginning of a cumulative dosing test session. In experiments examining the effects of pretreatment with the kappa-selective opioid antagonist nor-BNI (3.2 mg/kg), dose-effect curves for the kappaagonist U50,488 were determined 1 hr and 10 days after administration of nor-BNI, and a dose-effect curve for SNC80 was determined after 7 days. All antagonist doses and pretreatment times were based on prior studies (Butelman et al., 1993; Negus et al., 1993, 1994).
Data analysis.
Operant response rates from each cycle were converted to percent of control using the average rate from the previous training day as the control value. The log dose that produced a 50% decrease in the percent control rate of responding (ED50) was determined for each subject by linear regression from the linear portion of the dose-effect curve, and these values were averaged across subjects to yield a mean ED50 value (±95% confidence limits). ED50 values were considered to be significantly different if 95% confidence limits did not overlap. In addition, ED50 values (in mol/kg) for six piperazinyl benzamides were statistically correlated with their binding affinities at cloned human delta and mu opioid receptors (Knapp et al., 1996) using the StatView statistical analysis program (Abacus Concepts, Berkeley, CA). The criterion for significance was set a priori at P < .05.
Assay of Thermal Nociception
Apparatus and procedure.
The antinociceptive effects of selected opioids were studied in four monkeys, and each drug was tested in a group of three monkeys. Experiments were conducted no more than twice a week. During each experiment, the monkeys were seated in an acrylic restraint chair so that their tails hung down freely behind. An Apple IIe microcomputer was used to measure and record time intervals. The bottom 10 cm of each monkey’s shaved tail was immersed in a thermal container of warm water. If the subject did not withdraw its tail within 20 sec, the timer was stopped, and a tail-withdrawal latency of 20 sec was assigned to that measurement. Tail-withdrawal latencies were measured using four different water temperatures (42°, 46°, 50° and 54°C) to generate complete temperature-effect curves. When high doses of fentanyl and U50,488 were tested, 58°C was used and 42°C was omitted. Temperatures above 58°C were not tested to prevent the possibility of tissue damage. During any one cycle of measurements, subjects were exposed to water heated to four different temperatures as described above, and the order in which temperatures were presented was the same across subjects. The order in which the temperatures were presented was varied according to a modified latin square sequence across cycles.
Test sessions consisted of five cycles. Before the first test cycle, baseline tail-withdrawal latencies from 42°, 46°, 50° and 54°C water were determined. For cumulative dosing experiments with test compounds, a single drug dose was administered at the start of each 30 min cycle, and each injection increased the total dose by ¼ or ½ log increments. Fifteen minutes after each injection, tail-withdrawal latencies were redetermined as described above. In pretreatment experiments, naltrindole (1.0 mg/kg), quadazocine (0.1 mg/kg) or the delta opioid BW373U86 (0.56–1.0 mg/kg) was administered 15 min before the beginning of cumulative dosing test session.
Data analysis.
Tail-withdrawal data were analyzed as described previously (Negus et al., 1993; Gatch et al., 1995, 1996). For each animal at each set of latencies, the temperature that corresponded to a 10-sec latency was determined by fitting a line on the temperature-effect curve to the two points that fell immediately above and below 10 sec and interpolating the temperature that corresponded to a 10-sec latency. This measure was referred to as the T10 value. The T10 from each monkey’s baseline was subtracted from the T10 determined from each subsequent test cycle, which provided a measure of change relative to the baseline of each animal, or ΔT10. The ΔT10s were then averaged across monkeys and a standard error of the mean was calculated. The maximum effect of each drug (Emax) was determined for each monkey as the maximum ΔT10 produced by that drug. The log dose of each drug producing a ΔT10 of 3°C (ED3°) was interpolated from the linear portion of the dose-effect curve for each monkey and averaged to yield a mean ED3°value (±95% confidence limits). Emax and ED3° values were considered to be significantly different if 95% confidence limits did not overlap.
Observational Studies
Apparatus and procedure.
Observational studies were conducted in three monkeys (including two monkeys also used in studies of thermal nociception). Monkeys were seated in acrylic restraint chairs in a quiet room equipped with a video camera. A technician was present in the room throughout each test session, and the same technician conducted all observational studies. Test sessions were conducted no more the twice per week and began between 1:00 and 2:00 p.m., ∼20 hr after the most recent feeding of biscuits the previous afternoon and ∼4 hr after the most recent feeding of the morning ration of fruit. Water was not available during test sessions. Each test session consisted of five or six 20-min cycles. The first cycle was used to observe baseline behaviors and activity levels, and monkeys did not receive an injection. At the beginning of the second cycle, monkeys received an injection of saline. At the beginning of all subsequent cycles, monkeys received either repeated injections of saline or cumulative doses of a test compound (SNC80, fentanyl or U50,488). In cumulative dosing studies, each dose increased the total dose by ¼ or ½ log increments. At the end of each cycle (i.e., 20 min after each injection), the monkey was offered a 1-g banana-flavored food pellet identical to those used in the operant studies. Each monkey was allowed a maximum of 10 sec to take and consume the food pellet (i.e., a 10-sec cutoff), and the latency was measured between the time when the food pellet was offered and the time when it was placed in the monkey’s mouth. Throughout the test session, the technician present in the observation room observed and recorded overt changes in behavior, and especially the presence of convulsions.
In addition, the last 5 min of each cycle (including the period when a food pellet was offered) were videotaped and scored by two trained and independent observers blind to the treatment conditions. The observers assigned an Activity Score, which was determined in relation to baseline behavior displayed during the initial, no-injection cycle and which was based on the following scale: −2 = little or no movement, eyes closed for extended periods, low muscle tone; −1 = decreased speed or frequency of movements, eyes open and moving, relaxed; 0 = normal activity; +1 = increased speed or frequency of movements, brief bouts of repetitive movements; +2 = continuous rapid movements or continuous stereotypies. Interrater reliablity for the two observers in assignment of Activity Scores was 82%. Finally, the observer also noted the presence of unusual behaviors, and especially convulsions. In addition to the formalized videotaping and scoring of the last 5 min of each cycle as described above, monkeys were also videotaped at other times if warranted by the behavioral effects of the drug.
Data analysis.
The mean latency to consume the food pellet (±S.E.M.) and the mean Activity Score (±S.E.M.) were graphed as a function of drug dose. Other behaviors are described but were not formally quantified.
Assay of Drug Discrimination
Apparatus.
The discriminative stimulus effects of selecteddelta opioids were examined in seven monkeys. Each monkey was housed individually in a well-ventilated, stainless steel chamber (56 × 71 × 69 cm). The home cages of all monkeys were modified to include an operant panel identical to that described above for studies of schedule-controlled behavior. Operation of the operant panels and data collection were accomplished with Apple IIGS computers located in a separate room.
Discrimination training.
Experimental sessions were conducted 5 days per week, and each daily session was composed of multiple cycles. Each cycle consisted of a 15-min timeout period followed by a 5-min response period. During the timeout, all stimulus lights were off and responding had no scheduled consequences. During the response period the right and left response keys were transilluminated red or green, and monkeys could earn up to 10 food pellets by responding under an FR30 schedule of food presentation. The position of the red and green stimulus lights was counterbalanced across monkeys. The center key was not illuminated at any time, and responding on the center key had no scheduled consequences. If all 10 food pellets were delivered before the end of the 5-min response period, the stimulus lights transilluminating the response keys were turned off and responding had no scheduled consequences for the remainder of that response period.
On training days, monkeys were given an i.m. injection of either saline or 0.40 mg/kg cocaine 5 min after the beginning of each timeout period (i.e., 10 min before the response period). After the administration of saline, responding on only the green key (the saline-appropriate key) produced food, whereas after administration of 0.40 mg/kg cocaine, only responding on the red key (the cocaine-appropriate key) produced food. Responses on the inappropriate key reset the FR requirement on the appropriate key. Sessions consisted of one to five cycles each day, and if the training dose of cocaine was administered, it was administered only during the last cycle.
During the response period of each cycle, three dependent variables were determined using the following equations: (1) percent injection-appropriate responding prior to delivery of the first reinforcer: (injection-appropriate responses emitted before first reinforcer/total responses emitted before delivery of first reinforcer) × 100; (2) percent injection-appropriate responding for the entire response period: (injection-appropriate responses emitted during response period/total responses emitted during response period) × 100; and (3) response rate: (total responses emitted during response period/total time stimulus lights were illuminated).
Monkeys were considered to have acquired cocaine discrimination when the following three criteria were met for seven of eight consecutive training sessions: (1) the percent injection-appropriate responding before delivery of the first reinforcer was ≥80% for all cycles; (2) the percent injection-appropriate responding for the entire cycle was ≥90% for all cycles; and (3) response rates during saline training cycles were >1.0 response/sec.
Discrimination testing.
Once monkeys met criterion levels of cocaine discrimination, testing began. Test sessions were identical to training sessions except that responding on either key produced food, and test compounds were administered using either a substitution protocol or a pretreatment protocol. In the substitution protocol, test compounds (cocaine or the delta agonists SNC80, SNC162 and SNC243A) were administered using a cumulative dosing procedure, in which a dose of the test compound was administered 5 min after the beginning of each cycle, instead of either saline or the training dose of cocaine, and each dose increased the total dose by ¼ or ½ log units. In the pretreatment protocol, monkeys were pretreated with naltrindole, the dopamine antagonist flupenthixol, or SNC80 prior to the determination of a cumulative dose-effect curve for either SNC80 or cocaine. Naltrindole and SNC80 were administered 15 min before the beginning of cumulative dosing test sessions, and flupenthixol was administered 3 hr before the test session. These pretreatment times were based on our previously published studies using this drug discrimination procedure (Mello et al., 1995;Negus et al., 1995, 1996).
Training sessions were conducted on Mondays, Wednesdays and Thursdays, and test sessions were conducted on Tuesdays and Fridays. Test sessions were conducted only if the three criteria listed above under “Criteria for Discrimination” were met during the training day immediately preceding the test day. If responding did not meet criterion levels of discrimination performance, then training was continued until criterion levels of performance were obtained for at least 2 consecutive days.
Data analysis.
Graphs of the percent cocaine-appropriate responding (for the entire response period) and the response rate were plotted as a function of the cumulative dose of the test drug (log scale). A test drug was considered to have substituted for cocaine in a monkey if some dose of the test drug produced ≥90% cocaine-appropriate responding.
Assay of Drug Self-Administration
Surgery.
Four monkeys were used in drug self-administration studies, and these monkeys were surgically implanted with double-lumen Silicone rubber catheters (inside diameter 0.7 mm; outside diameter 2.0 mm). Catheters were implanted in the jugular or femoral vein, and the catheters exited in the midscapular region. All surgical procedures were performed under aseptic conditions. Monkeys were initially sedated with ketamine (5 mg/kg s.c.), and anesthesia was induced with sodium thiopental (10 mg/kg i.v). In addition, monkeys were treated with 0.05 mg/kg atropine to reduce salivation. After insertion of a tracheal tube, anesthesia was maintained with halothane (1–1.5% in oxygen). After surgery, monkeys were administered aspirin or acetaminophen (80–160 mg/day p.o.) for 3 days. The antibiotic procaine penicillin G (300,000 units/day i.m.) was administered every day for 5 days. The i.v. catheter was protected by a tether system consisting of a custom-fitted nylon vest connected to a flexible stainless steel cable and fluid swivel (Lomir Biomedical, Montreal, Canada). This flexible tether system permited monkeys to move freely. Catheter patency was periodically evaluated by i.v. administration of the short-acting barbiturate methohexital (3 mg/kg i.v.) or ketamine (2–3 mg/kg i.v.). The catheter was considered patent if i.v. administration of methohexital or ketamine produced loss of muscle tone within 10 sec after its administration.
Apparatus.
Each monkey was housed individually in a well-ventilated stainless steel chamber (64 × 64 × 79 cm). The home cages of all monkeys were modified to include an operant panel (28 × 28 cm) identical to that described above for schedule-controlled behavior. Two syringe pumps (model B5P-lE; Braintree Scientific, Braintree, MA; or model 980210, Harvard Apparatus, South Natick, MA) were mounted above each cage for delivery of saline or drug solutions through the intravenous catheters. Operation of the operant panels and data collection were accomplished with an IBM-compatible computer programmed in MEDSTATE Notation (MED Associates, East Fairfield, VT).
Drug self-administration training.
Food and i.v. drug or saline injections were available during three alternating components: a 5-min food component, a 100-min drug component and a second 5-min food component. During initial shaping of the lever press response, both food and i.v. injections were available under an FR1 schedule of reinforcement. Subsequently, the fixed ratio value was gradually increased to FR30. During the two food components, the center response key was transilluminated red. During the drug component, the center response key was transilluminated green. After the delivery of each food pellet or drug injection, there was a 10-sec timeout period, during which the stimulus light illuminating the center response key was turned off and responding had no scheduled consequences. The food and drug components were separated by 5-min timeout periods when the response key was dark and responding had no scheduled consequences. The entire food/drug/food session lasted 120 min and was conducted daily from 3:00 to 5:00 p.m.
In addition to the food/drug/food session described above, monkeys were given the opportunity to respond for additional food pellets during supplementary food sessions conducted from 7:00 to 8:00 p.m. and 6:00 to 7:00 a.m. During these sessions, food was available under an FR30/timeout 10-sec schedule, and a maximum of 25 pellets per session could be earned. These additional food sessions provided additional enrichment opportunities for the monkeys and behavioral information relevant to the detection of possible health problems.
During initial training, the solution available for self-administration during the drug component was 32 μg/kg/injection cocaine, and this maintenance dose of cocaine was available until self-administration behavior was stable. Stability was defined as three consecutive days during which monkeys acquired ≥25 injections/day and the response rate during the drug component of each session differed by no more than 20% from the mean drug component response rate. Once monkeys met criterion levels for stable cocaine self-administration, saline was substituted for cocaine until rates of injection-maintained responding were below 0.1 response/sec during the drug component for three consecutive days. This sequence of cocaine availability followed by saline availability was then repeated at least once. Because the primary purpose of these studies was to study drug self-administration, no formal criteria were established for rates of food-maintainted responding during the food components. However, response rates during food components were consistently >1.0 response/sec.
Drug self-administration testing.
After training, testing was initiated using substitution procedures in which different unit doses of cocaine (0.32–10 μg/kg/injection) or SNC80 (1.0–100 μg/kg/injection) were substituted for the maintenance dose of cocaine. Each substitution remained in effect for one session, and test sessions were conducted on Tuesdays and Fridays. Either saline or the maintenance dose of cocaine was reinstated for the remaining weekly sessions, with the restriction that test sessions were always preceded by training sessions during which the maintenance dose of cocaine was available. In addition, test days were occasionally omitted to allow several days of saline substitution. Saline substitution sessions were conducted frequently to maintain rapid extinction of behavior during saline substitution conditions.
Data analysis.
The response rates during each food and drug component were calculated as (total responses emitted during component ÷ total time stimulus lights were illuminated). In addition to calculation of response rates for the entire component, response rates during each 25-min quartile of the 100-min drug component were also determined to provide information regarding the pattern of responding during the drug component. For the purposes of statistical analysis, response rates during substitution of cocaine or SNC80 were compared with response rates during saline substitution using a one-factor ANOVA, with dose of cocaine or SNC80 as the single factor. A significant ANOVA was followed by individual mean comparisons using the Duncan posthoc test (Winer, 1971). The criterion for significance was set a priori at P < .05. A substitution drug was considered to maintain self-administration if at least one dose of the drug maintained response rates significantly greater than those maintained by saline.
Drugs
The structures of the piperazinyl benzamides examined in this study are shown in figure 1. These piperazinyl benzamides and naltrindole were synthesized by Drs. Zhang and Rice at NIDDK/NIH. SNC67, SNC80, SNC121, SNC162 and SNC243A were free bases and were dissolved in 2% to 3% lactic acid in distilled water. SNC86 [(−)-dibenzoyl-l-tartaric acid salt] was dissolved in 10% DMSO in distilled water, and SNC89 (HCl salt) and naltrindole HCl were dissolved in distilled water. BW373U86 (Burroughs Wellcome, Research Triangle Park, NC), quadazocine methanesulfonate (Sanofi Pharmaceuticals, Malvern, PA) and nor-binaltorphimine 2HCl (Dr. P. S. Portoghese, Minneapolis, MN) were dissolved in distilled water. Fentanyl citrate, U50,488 methanesulfonate, and flupenthixol 2HCl (all purchased from Research Biochemicals, Natick, MA) also were dissolved in distilled water. Cocaine HCl (National Institute on Drug Abuse, Rockville, MD) was dissolved in sterile saline. Doses were calculated in the forms described above and were administered i.v. in a volume of 0.1 ml in the drug self-administration studies or i.m. in a volume of 0.1 to 2.0 ml in all other studies. For i.v. administration, aseptic precautions were taken in every phase of drug solution preparation and dispensing. Drug solutions were filter-sterilized using a 0.22-μm Millipore filter and stored in sterile, pyrogen-free vials. Sterility of the entire fluid path for drug solutions was maintained throughout the study.
Results
Assay of schedule-controlled responding.
In an initial series of studies, the effects of SNC80 and related compounds on rates of food-maintained operant responding were evaluated in a group of four monkeys. The overall mean control response rate ± S.E.M. was 2.05 ± 0.13 responses per second, and control response rates for individual monkeys ranged from 1.77 ± 0.05 to 2.53 ± 0.06 responses per second. As described previously for this procedure (e.g., Gatch et al., 1996), response rates were similar both across response cycles within a session and across sessions.
Figure 2 shows the time course of effects of SNC80 (1.0 and 3.2 mg/kg) on rates of schedule-controlled responding. A dose of 1.0 mg/kg SNC80 decreased response rates to ∼30% of control levels after 10 min, and response rates gradually recovered after 300 min. A dose of 3.2 mg/kg SNC80 eliminated responding after 10 min, and response rates were still suppressed to ∼60% of control levels after 300 min. However, response rates recovered to control levels after 24 hr.
Figure 3 shows cumulative dose-effect curves for SNC67, SNC80, SNC86, SNC89, SNC121 and SNC162. All six compounds produced dose-dependent decreases in response rates, although relative potencies varied over more than two orders of magnitude. The most potent compounds were SNC80, SNC86 and SNC162. SNC121 was less potent, and the least potent compounds were SNC67 and SNC89, which are the (−)-enantiomers of SNC80 and SNC86 respectively. Figure 3 also shows that the in vivo potency of these compounds correlated significantly with their affinity at cloned human deltareceptors (Knapp et al., 1996) (R = .897, P = .015). However, there was not a significant correlation between the potency of these compounds and their affinities at cloned humanmu receptors (R = .623; P = .186; data not shown). In addition to decreasing response rates, both SNC67 (56 mg/kg) and SNC121 (32 mg/kg) produced tonic/clonic convulsions. A dose of 56 mg/kg SNC67 produced a convulsion in one monkey, and this high dose of SNC67 was not tested in the other three monkeys. The same monkey convulsed following administration of 32 mg/kg SNC121. The convulsions lasted ∼5 min and were symptomatically treated by administration of diazepam (1.0–2.0 mg/kg). Convulsions were not observed with any of the other compounds.
Pretreatment with the delta opioid antagonist naltrindole (1.0 mg/kg) shifted the SNC80 dose-effect curve to the right and produced a ∼13-fold increase in the ED50 value for SNC80; however, this dose of naltrindole had no effect on the dose-effect curves or ED50 values for themu agonist fentanyl or the kappa agonist U50,488 (fig. 4 and table1). In contrast, themu-selective antagonist quadazocine (0.1 mg/kg) and thekappa-selective antagonist nor-BNI (3.2 mg/kg) did not alter the SNC80 dose-effect curve, although these antagonists did produce rightward shifts in the dose-effect curves and significant increases in the ED50 values for fentanyl and U50,488, respectively (fig. 4 and table 1). Naltrindole (1.0 mg/kg) also produced rightward shifts in the dose-effect curves and significant increases in ED50 values for SNC162 and SNC243A (fig. 5 and table 1). Note that SNC243A was not included in the correlational analysis shown in figure 3because the affinity of this compound at cloned human deltareceptors was not evaluated by Knapp et al. (1996). In rat brain tissue, however, SNC243A has an affinity and selectivity fordelta receptors similar to that of SNC80 (Calderon et al., 1997).
Assay of thermal nociception.
The antinociceptive effects of SNC80 and related compounds were examined in a warm-water tail-withdrawal assay of thermal nociception (n = 3 for each test drug). During baseline tail-withdrawal measurements, monkeys never removed their tails from water heated to 42°C, indicating that immersion of the tail alone was not sufficient to elicit the tail-withdrawal response. Mean tail-withdrawal latencies (±S.E.M.) from water heated to 46°, 50° and 54°C were 7.90 ± 4.26, 1.09 ± 0.20 and 0.97 ± 0.24 sec, respectively. The mean T10 value (±S.E.M.), which is the temperature corresponding to a tail-withdrawal latency of 10 sec, was 45.50 ± 0.91°C.
The antinociceptive effects of SNC80 administered alone or after various pretreatments are shown in figure6 and table2. SNC80 produced a relatively small but dose-dependent increase in ΔT10. The mean ED3°C for SNC80 (i.e., the dose producing a ΔT10 value of 3°C) was 0.93 mg/kg, and the mean Emax for SNC80 (i.e., the maximum ΔT10) was 4.1°C. For comparison, the ED3°Cand Emax values were .010 mg/kg and 12.0°C for the mu agonist fentanyl and 0.20 mg/kg and 11.3°C for thekappa agonist U50,488. Pretreatment with 1.0 mg/kg naltrindole produced a rightward shift in the SNC80 dose-effect curve and a significant 16-fold increase in the SNC80 ED3°C, but pretreatment with 0.1 mg/kg quadazocine had no effect (fig. 6, left; table 2). We reported previously that BW373U86 does not produce antinociceptive effects in this procedure (Negus et al., 1994; Gatch et al., 1995), and administration of BW373U86 alone at doses up to 1.0 mg/kg in the present study also failed to produce antinociceptive effects (data not shown). However, pretreatment with BW373U86 (0.56–1.0 mg/kg) produced dose-dependent rightward shifts in the SNC80 dose-effect curve, and 1.0 mg/kg BW373U86 produced a statistically significant 13-fold increase in the SNC80 ED3°C (fig. 6, right; table 2).
SNC162 and SNC243A also produced dose-dependent increases in ΔT10 values (fig. 7). In general, however, the effects of these compounds were smaller and more variable than those of SNC80, and SNC243A was completely ineffective in one of the three monkeys tested. As a result, ED3°C and Emax values were not calculated for these compounds.
Observational effects.
Because doses of SNC80 up to 10 mg/kg did not produce convulsions in the assays of schedule-controlled behavior or thermal nociception, additional observational studies were conducted in three monkeys to determine if higher doses of SNC80 would produce convulsions or other overt behavioral effects. In these studies, repeated administration of saline for five consecutive cycles had little effect on the latency to take a food pellet [3.4 (±2.8) and 1.9 (±0.4) sec for first and last cycles, respectively] or the activity score [−0.33 (±0.21) and −0.17 (±0.3) for first and last cycles, respectively]. SNC80 at doses up to 32 mg/kg produced a dose-dependent increase in the latency to take a food pellet and small decreases in the Activity Score (fig. 8). In addition, one monkey vomited after administration of 32 mg/kg SNC80. However, SNC80 did not produce convulsions in any of the monkeys, and did not produce any other notable behavioral effects.
For comparison, the effects of the mu agonist fentanyl and the kappa agonist U50,488 were also examined in this procedure (fig. 8). Fentanyl (0.032–0.1 mg/kg) and U50,488 (0.1–1.8 mg/kg) also produced dose-dependent increases in the latency to take a food pellet, with fentanyl being ∼100-fold more potent than SNC80 and U50,488 ∼10-fold more potent. However, both fentanyl and U50,488 produced greater decreases than SNC80 in the Activity Score, and at the highest doses tested, both drugs produced profound sedation. In addition, fentanyl and U50,488 produced overt toxic effects at the highest doses tested. Fentanyl at a dose of 0.1 mg/kg dramatically decreased the regularity and frequency of breathing in one monkey, and the opioid antagonist naltrexone was administered to prevent respiratory arrest. This dose of fentanyl was not tested in the other two monkeys. U50,488 produced intermittent tremors of the arms and legs and rapidly alternating contraction and dilation of the pupils at a dose of 1.0 mg/kg in all three monkeys. A higher dose of 1.8 mg/kg U50,488 was tested in only two monkeys, and this dose produced polymyoclonus in one monkey.
Assay of cocaine discrimination.
The cocaine-like discriminative stimulus effects of SNC80 and related compounds were examined in seven monkeys trained to discriminate 0.4 mg/kg i.m. cocaine from saline. Cocaine produced a dose-dependent increase in cocaine-appropriate responding in all seven monkeys, with complete substitution occuring at the training dose of 0.4 mg/kg (fig.9). The 0.4 mg/kg dose of cocaine had little effect on response rates, but a higher dose of 1.3 mg/kg cocaine eliminated responding in three monkeys and decreased response rates in the other four monkeys.
SNC80 also produced dose-dependent increases in cocaine-appropriate responding and decreases in response rates (fig. 9). The highest dose of SNC80, 10 mg/kg, eliminated responding in two monkeys, but in the remaining five monkeys, this dose produced a mean (±S.E.M.) of 94% (±4.0%) cocaine-appropriate responding. SNC162 and SNC243A also increased cocaine-appropriate responding and decreased response rates, but these effects were smaller and more variable than those of SNC80. At the highest dose tested (10 mg/kg), SNC162 eliminated responding in three of six monkeys and produced 83% (±6.1) cocaine-appropriate responding in the remaining three monkeys, whereas SNC243A eliminated responding in two of six monkeys and produced 40% (±17%) cocaine-appropriate responding in the remaining monkeys.
The effects of pretreatment with either the delta antagonist naltrindole (1.0 mg/kg) or the dopamine antagonist flupenthixol (0.01 mg/kg) were examined in three monkeys in which SNC80 substituted for cocaine (fig. 10). Naltrindole pretreatment decreased the cocaine-like discriminative stimulus effects of SNC80 and produced a rightward shift in the SNC80 dose-effect curve, but flupenthixol pretreatment had no effect. We have shown previously that this dose of flupenthixol is effective in shifting the dose-effect curve for cocaine to the right (Negus et al., 1996).
Figure 11 shows the effects of SNC80 pretreatment (0.32–1.0 mg/kg) on the cocaine discrimination dose-effect curve in six monkeys. In these pretreatment studies, saline was administered during the first cycle of the cumulative dosing test session to assess the effects of SNC80 alone, and cumulative doses of cocaine (0.013–0.4 mg/kg) were administered during subsequent cycles. Pretreatment with 0.32 mg/kg SNC80 produced exclusively saline-appropriate responding during the initial saline-injection cycle, whereas 1.0 mg/kg SNC80 produced>60% cocaine-appropriate responding (points above “Sal” in fig. 11). Thus, as in the cumulative dosing studies described above, SNC80 produced a dose-dependent increase in cocaine-appropriate responding. However, SNC80 was slightly more potent in these pretreatment studies than in the cumulative dosing studies. The lower dose of 0.32 mg/kg SNC80 had little effect on the cocaine dose-effect curve. Pretreatment with the higher dose of 1.0 mg/kg SNC80 shifted the cocaine dose-effect curve upward and to the left. However, the effects of SNC80 in combination with cocaine never exceeded the effects of either drug alone.
Assay of drug self-administration.
The reinforcing effects of SNC80 were examined in four monkeys trained to respond for food and cocaine (32 μg/kg/injection). Substitution of saline and lower unit doses of cocaine (0.32–10 μg/kg/injection) for the maintenance dose of cocaine in these monkeys yielded a cocaine self-administration dose-effect curve with an inverted U shape (fig.12, top). Unit doses of 1.0, 3.2 and 10 μg/kg/injection maintained response rates significantly higher than those maintained by saline (P < .05), and peak response rates were maintained by 3.2 μg/kg/injection cocaine. The unit dose of cocaine available during the drug component of the experimental session had little effect on rates of food-maintained responding during the food components occuring either before or after the drug component (fig. 12, center and bottom, respectively).
The effects of substituting SNC80 (1.0–100 μg/kg/injection) for cocaine are also shown in figure 12. No dose of SNC80 maintained response rates significantly higher than those maintained by saline; however, the highest dose of SNC80 tested, 100 μg/kg/injection, maintained response rates significantly lower than those maintained by saline (P < .05). The unit dose of SNC80 available during the drug component did not significantly alter rates of food maintained responding during either of the two food components. However, there was a trend for response rates during the second food component to decrease after self-administration of 32 and 100 μg/kg/injection SNC80.
Response rates in sequential quartiles of the drug component are shown in figure 13 to illustrate patterns of self-administration behavior during different substitution conditions. The left of figure 13 compares the effects of substituting either saline or a unit dose of cocaine from the peak of the cocaine dose-effect curve (3.2 μg/kg/injection). During saline substitution, monkeys responded at high rates during the first quartile of the 100-min drug component, but response rates decreased to low levels during the last three quartiles of the component. In contrast, when 3.2 μg/kg/injection cocaine was available for self-administration, monkeys usually responded at relatively high rates throughout the 100-min drug component, although the highest rates of responding were still usually observed during the first quartile. Response rates were significantly higher during cocaine availability than during saline availability during the second, third and fourth quartiles (P < .05). The right of figure 13 compares the effects of substituting saline with the effects of substituting SNC80 (3.2–100 μg/kg/injection). Patterns of responding during SNC80 substitution were similar to those observed during saline substitution. Response rates were high during the first quartile and then decreased to low levels during the remaining three quartiles. Response rates during SNC80 availability were not higher than rates during saline availability for any of the quartiles of the drug component. However, response rates for the first quartile were significantly lower during availability of 100 μg/kg/injection SNC80 than during availability of saline.
Discussion
The findings of the present study indicate that SNC80, like the parent compound BW373U86, functions as a systemically activedelta-selective opioid agonist in rhesus monkeys. However, SNC80 differs from BW373U86 in several respects. Unlike BW373U86, SNC80 produces thermal antinociceptive effects and does not produce convulsions in rhesus monkeys even at relatively high doses. Also, although SNC80 is more effective than BW373U86 in producing cocaine-like discriminative stimulus effects, it does not maintain drug self-administration under conditions in which cocaine is reinforcing, which suggests that SNC80 has relatively low abuse potential. Taken together, these findings suggest that SNC80 and other relateddelta opioid agonists may have promise as analgesic drugs with relatively few side-effects.
Effects of SNC80 and related compounds in the assay of schedule-controlled responding maintained by food.
The time course, potency and receptor selectivity of SNC80 and related piperazinyl benzamides were initially examined in an assay of food-maintained, schedule-controlled behavior. This procedure has been used previously to assess the effects of the parent compound BW373U86 (Negus et al., 1993, 1994) as well as the effects of bothmu and kappa opioids (Negus et al., 1993; Butelman et al., 1996; Gatch et al., 1996). As with many other opioid agonists that have been examined in this procedure, SNC80 produced a dose-dependent decrease in response rates. In the present study, for example, SNC80, the mu agonist fentanyl and the kappa agonist U50,488 all decreased response rates. The principal difference between these compounds was their relative potency, with SNC80 being ∼100-fold less potent than fentanyl and 10-fold less potent that U50,488. Maximal effects of SNC80 were observed 10 min after i.m. administration and these effects diminished over several hours, indicating that relative to other opioid agonists, SNC80 has a rapid onset of action and moderately long duraton of action. For example, the time course of SNC80 observed in this study is similar to that of the parent compound BW373U86 (Negus et al., 1994). In relation to prototypical mu agonists, the time course of SNC80 is similar to that of fentanyl, whereas morphine has a slower rate of onset and longer duration of action (Gatch et al., 1996; Negus et al., 1998).
The pharmacological mechanisms underlying the rate-decreasing effects of SNC80 cannot be inferred by examination of the effects of SNC80 alone, because as noted above, agonists selective for mu,kappa and delta opioid receptors produce qualitatively similar effects. However, two findings suggested that the rate-decreasing effects of SNC80 were mediated by deltaopioid receptors. First, the rate-decreasing effects of SNC80 were antagonized by a delta receptor-selective dose of the opioid antagonist naltrindole, but not by a mu-selective dose of quadazocine or a kappa-selective dose of nor-BNI. Naltrindole also antagonized the rate-decreasing effects of SNC162 and SNC243A, two piperazinyl benzamides that have an affinity and selectivity for delta receptors similar to or higher than that of SNC80 (Knapp et al., 1996; Calderon et al., 1997).
Second, the relative potency of SNC80 and five other piperazinyl benzamides correlated with their relative affinities at cloned humandelta opioid receptors but not with their affinities at cloned human mu receptors (cf. Knapp et al., 1996). These potency comparisons also provided evidence for stereoselectivity. Two pairs of stereoenantiomers were investigated (SNC80 and SNC67; SNC86 and SNC89), and in both cases, the (+)-enantiomers (SNC80 and SNC86) were more potent than the (−)-enantiomers (SNC67 and SNC89). Moreover, our behavioral findings in primates correlate with other functional studies that have compared the relative potencies of these compounds. For example, BW373U86 is more potent than SNC80 in stimulating GTPγS binding in C6 glioma cells (Clark et al., 1997), inhibiting electrically induced contractions of the mouse vas deferens (Calderon et al., 1994) and producing antinociceptive and convulsant effects in mice (Wild et al., 1993; Bilsky et al., 1995). Taken together, these findings suggest that, like BW373U86 (Negus et al., 1994), SNC80 and some other related piperazinyl benzamides act as systemically active delta opioid agonists in rhesus monkeys.
Effects of SNC80 on thermal nociception.
The potential ofdelta opioid agonists to produce clinically relevant analgesic effects is one important reason for their development and evaluation, and this study provides the first evidence that activation of delta opioid receptors can produce thermal antinociception in primates. The maximal antinociceptive effects of SNC80 were relatively small in comparison with the effects of the high efficacy mu and kappa opioid agonists fentanyl and U50,488. However, the effects of SNC80 were similar to those obtained previously in this assay with the low efficacy muagonist nalbuphine (Gatch et al., 1995), which is used clinically as an analgesic. Pretreatment with adelta-selective dose of naltrindole, but not amu-selective dose of quadazocine, produced a rightward shift in the SNC80 dose-effect curve, which suggests that the antinociceptive effects of SNC80 were mediated by delta but notmu opioid receptors. In agreement with the present study, SNC80 was also found to produce delta receptor-mediated thermal antinociception in mice (Bilsky et al., 1995).
Studies using the assay of thermal nociception also provided an opportunity to compare the relative efficacy of SNC80 with the efficacies of other delta opioid agonists. Whereas SNC80 produced weak but consistent antinociception, BW373U86 did not produce antinociception under these conditions (see also Negus et al., 1994; Gatch et al., 1995). Moreover, pretreatments with BW373U86 reversibly antagonized the antinociceptive effects of SNC80. These findings suggest that SNC80 and BW373U86 act at a commondelta opioid receptor type, and that SNC80 has higher efficacy than BW373U86 at these receptors. SNC162 and SNC243A produced antinociceptive effects intermediate between the effects of SNC80 and BW373U86, suggesting that they may have an intermediate efficacy atdelta receptors. However, the weak and inconsistent effects of these compounds precluded a rigorous evaluation of their pharmacological mechanisms of action.
Previous studies have reported other differences in the effects of novel delta agonists that may be related to differential efficacy at delta receptors. For example, SNC80 was also more effective than BW373U86 in assays of thermal nociception in mice (Wild et al., 1993; Bilsky et al., 1995). Moreover, in receptor binding studies, the affinity of SNC80 fordelta opioid receptors was negatively modulated by sodium, an effect observed with most opioid agonists but not with opioid antagonists (Knapp et al., 1996). In contrast, BW373U86 displayed an antagonist-like profile under these conditions, in that sodium did not alter the affinity of BW373U86 for deltareceptors (Childers et al., 1993). However, other in vitro functional studies have not observed differences in the maximal effects of SNC80 and BW373U86. For example, SNC80 and BW373U86 produce similar maximal effects on GTPγS binding in C6 glioma cells (Clark et al., 1997). In addition, BW373U86 may be more effective than SNC80 in producing some behavioral effects (e.g. convulsions, see below). Additional research will be required to evaluate the degree to which differences in the effects of SNC80, BW37U86 and other novel deltaagonists reflect differences in efficacy at a common population ofdelta opioid receptors, or differences in some other pharmacological parameter, such as different profiles of activity at putative delta receptor subtypes or at nondeltareceptors.
Absence of convulsant effects with SNC80.
The possible convulsant effects of SNC80 were examined because the parent compound BW373U86 produced convulsions in mice, squirrel monkeys and rhesus monkeys at doses similar to or only slightly higher than those that produced other behavioral effects (Comer et al., 1993a;Dykstra et al., 1993; Negus et al., 1994). However, in contrast to these previous findings with BW373U86, SNC80 did not produce convulsions or other overt toxic effects in rhesus monkeys at doses up to 32 mg/kg: a dose >30 times greater than the ED50 for SNC80 in the assay of schedule-controlled behavior or the ED3°C for SNC80 in the assay of thermal nociception. For comparison, BW373U86 produced convulsions at doses <20 times greater than its ED50 in the assay of schedule-controlled responding (Negus et al., 1994). Similarly, themu and kappa agonists fentanyl and U50,488 also produced toxic effects characteristic of mu andkappa opioid agonists (e.g., Dykstraet al., 1987; Nussmeier et al., 1991) at doses <20 times greater than their ED50 values in the assay of schedule-controlled behavior or the ED3°C values in the assay of thermal nociception (present study). These results suggest that SNC80 may be relatively safe over a broader range of behaviorally active doses than either the parent compound BW373U86 or representative mu andkappa opioid agonists. SNC80 was also reported to have less toxicity than BW373U86 in mice (Bilsky et al., 1995).
Although SNC80 did not produce convulsions in this study, two other piperazinyl benzamides, SNC121 and SNC67, did produce tonic-clonic convulsions at doses only 3–4 times greater than their ED50 values in the assay of schedule- controlled responding. The receptor mechanisms underlying these effects were not investigated; however, these findings provide further evidence that there are important differences in the relative abilities of these piperazinyl benzamides to produce convulsant effects in rhesus monkeys. Moreover, it is important to note that within this series of compounds, enhanced antinociceptive activity such as that seen with SNC80 is not necessarily associated with enhanced convulsant activity. This suggests that the antincociceptive effects and convulsant effects of piperazinyl benzamides may be dissociable and may be mediated by different pharmacological mechanisms of action.
Abuse-related effects of SNC80.
SNC80 produced high levels of cocaine-appropriate responding in monkeys trained to discriminate cocaine from saline. These findings agree with previous reports that peptidic delta opioid agonists produce high levels of cocaine-appropriate responding in rats trained to discriminate cocaine (Ukai et al., 1993; Suzuki et al., 1997). In the present study, the cocaine-like discriminative stimulus effects of SNC80 were blocked by naltridole, suggesting that these effects were mediated by delta opioid receptors. However, the cocaine-like discriminative stimulus effects of SNC80 were not affected by a dose of the dopamine antagonist flupenthixol (0.01 mg/kg) that produed a rightward shift in the cocaine dose-effect curve (Neguset al., 1996). In addition, we have reported previously that naltrindole did not antagonize the discriminative stimulus effects of cocaine in these monkeys (Negus et al., 1995). Taken together, these findings suggest that SNC80 and cocaine produce similar discriminative stimulus effects in monkeys, but these effects appear to be mediated by different pharmacological mechanisms of action.
Like SNC80, the two other delta agonists SNC162 and SNC243A also produced dose-dependent increases in cocaine-appropriate responding; however, the effects of SNC162 and SNC243A were smaller and less consistent than those observed with SNC80. Furthermore, we have reported previously that BW373U86 produces even lower levels of cocaine-appropriate responding (Negus et al., 1995). Thus, the relative ability of these compounds to produce cocaine-like discriminative stimulus effects correlated well with their relative ability to produce thermal antinociception as described above (SNC80 > SNC162 ≥ SNC243A > BW373U86).
Pretreatments with SNC80 produced leftward and upward shifts in the cocaine-discrimination dose-effect curve. However, the effects of SNC80 and cocaine appeared to be additive at best, because levels of cocaine-appropriate responding produced by combinations of SNC80 and cocaine were similar to the effects of either SNC80 or cocaine alone. These findings suggest that SNC80 pretreatment did not enhance the discriminative stimulus effects of cocaine under these conditions.
Despite its ability to produce cocaine-like discriminative stimulus effects, SNC80 did not maintain self-administration under conditions in which cocaine functioned as a reinforcer. Rather, substition of SNC80 at doses up to 32 μg/kg/injection produced rates and patterns of self-administration similar to those produced by saline substitution, and a higher dose of 100 μg/kg/injection SNC80 was self-administered at rates lower than saline. BW373U86 also failed to maintain drug self-administration in rhesus monkeys maintained either on cocaine or on the mu opioid agonist alfentanil (Negus et al., 1994, 1995). However, these findings contrast with reports that in rodents, centrally administered peptidic deltaagonists produce conditioned place preferences (Shippenberg et al., 1987), decreases in thresholds for electrical brain stimulation (Duvauchelle et al., 1997), and maintenance of drug self-administration (Devine and Wise, 1994). The reason for this discrepancy is not clear and could be related to differences in routes of administration, species of the experimental subjects and/or procedures used to evaluate drug reinforcement. However, our findings suggest that SNC80 does not produce cocaine-like reinforcing effects and may have relatively low abuse potential.
Summary.
The results of the present study suggest that noveldelta agonists, and especially SNC80, may have significant advantages over other available delta agonists as potential analgesics. Unlike peptidic delta agonists such as DPDPE, SNC80 is systemically active. Moreover, in comparison with the systemically active parent compound BW373U86, SNC80 produces greater antinociceptive activity in assays of thermal nociception with a reduced incidence of convulsant activity. SNC80 produces cocaine-like discriminative stimulus effects, but it does not produce cocaine-like reinforcing effects in an assay of drug self-administration, suggesting that SNC80 may have relatively low abuse potential. These findings suggest that SNC80 and related nonpeptidic delta agonists may be promising candidates for development as safe and effective clinical analgesics.
Acknowledgments
The authors would like to thank Amy Calvert, Peter Fivel and Lenore Jensen for expert technical assistance and Elizabeth Hall, D.V.M., for veterinary assistance.
Footnotes
-
Send reprint requests to: Dr. S. Stevens Negus, Alcohol and Drug Abuse Research Center, Harvard Medical School, McLean Hospital, 115 Mill Street, Belmont, MA 02478-9106.
-
↵1 This work was supported by Grants RO1-DA02519, P50-DA04059, T32-DA07252 and K05-DA00101.
- Abbreviations:
- nor-BNI
- nor-binaltorphimine
- DPDPE
- (d-Pen2,d-Pen5)-enkephalin
- FR
- fixed ratio
- Received December 29, 1997.
- Accepted March 27, 1998.
- The American Society for Pharmacology and Experimental Therapeutics