Introduction

Top Abstract Introduction Methods Results Discussion References

The discovery that buspirone is effective in the treatment of anxiety (Riblet et al., 1982) led to the development of novel, putative anxiolytics that are agonists at the 5-HT_1A receptor subtype (Barrett and Vanover, 1993; Deakin, 1993; De Vry, 1996). Seven classes of 5-HT receptors (5-HT_1-7) are recognized currently, and the 5-HT₁ receptor family now has five subtypes (Humphrey et al., 1993; Hoyer et al., 1994; Martin and Humphrey, 1994). In addition to their role in anxiety, 5-HT_1A, 5-HT_1B and 5-HT_1D receptor subtypes have been implicated in psychiatric disorders such as schizophrenia and depression, and in migraine (Saxena, 1995). Although the 5-HT_1B receptor was once thought to be specific to rats and some other rodents, the human 5-HT_1D and the rat 5-HT_1B are now recognized variants of the same receptor (Saxena, 1994). Yet, h5-HT_1B (i.e., 5-HT_1D) and r5-HT_1B receptors have different pharmacological properties that apparently result from a single amino acid mutation (Martin and Humphrey, 1994). Thus, because compounds acting at 5-HT_1D/1B and 5-HT_1A receptors are targets for drug discovery, the species in which preclinical tests are conducted may be important.

Species differences in the DS effects of mixed 5-HT_1A/1B ligands have been reported (Barrett and Gleeson, 1992; Ybema et al., 1993; Mos et al., 1997). Prototypical 5-HT_1A agonists such as 8-OH-DPAT and buspirone produce DS effects in rats and pigeons which are presumably mediated by interactions with central 5-HT_1A receptors (Barrett and Zhang, 1991; Barrett and Gleeson, 1992; Sanger and Schoemaker, 1992; Rabin and Winter, 1993; Schreiber et al., 1995a). Barrett and Gleeson (1992) demonstrated that compounds with mixed 5-HT_1A/1B affinities (e.g., RU24969 and eltoprazine) mimic the DS effects of 8-OH-DPAT in pigeons, whereas they do not have similar effects in rats (Cunningham et al., 1987; Tricklebank et al., 1987; Gardner, 1989; Ybema et al., 1992, 1993). Species differences were also suggested to be important in the pigeon conflict procedure (Barrett et al., 1994), a preclinical test that has been used widely to characterize novel anxiolytic compounds (Nanry et al., 1991; Colpaert et al., 1992; Barrett and Vanover, 1993; Foreman et al., 1993; Kleven and Koek, 1996). Robust, anxiolytic-like activity of 5-HT_1A agonists can be readily demonstrated in this species, in contrast to results obtained in rats (Brocco et al., 1990; Howard and Pollard, 1990), thus making the pigeon an especially useful species for preclinical studies involving 5-HT_1A ligands (Barrett et al., 1994). This is particularly evident for the low-efficacy 5-HT_1A agonist buspirone (Yocca, 1990; Rabin and Winter, 1993) which is an effective anxiolytic in man (Riblet et al., 1982) and has easily identifiable anxiolytic-like effects in the pigeon conflict procedure (e.g., Kleven and Koek, 1996). Inasmuch as the pigeon may closely model therapeutic effects of 5-HT_1A agonists in humans, further comparative pharmacological studies in rats and pigeons are needed.

In the present study we examined the DS effects of a variety of putative 5-HT_1A agonists in both rats and pigeons trained to discriminate 8-OH-DPAT from saline. We also compared the ability of putative 5-HT_1A antagonists to block the DS effects of 8-OH-DPAT in rats and pigeons. Although many of the same compounds have been examined previously in rats and/or pigeons, an important contribution of the present study is that all of the compounds were examined in the same laboratory by similar procedures. Altogether, the DS effects of 8-OH-DPAT appear to be pharmacologically similar in both species under the conditions that were used in this study, e.g., training dose and route of administration; yet, some compounds unexpectedly produced different results in rats and pigeons. Although the results support previous conclusions about the ability of compounds with 5-HT_1A agonist properties to engender 8-OH-DPAT-like DS effects, other factors such as apparent intrinsic activity or actions at other receptors may determine whether or not particular 5-HT_1A ligands produce similar results in rats and pigeons.

	Methods

Top Abstract Introduction Methods Results Discussion References

Rats

Male Sprague Dawley rats (Ico: OFA SD (I.O.P.S. Caw) Iffa Credo, Lyon, France), weighing between 240 and 260 g at the beginning of the studies, were used. Animals were housed in individual cages (Iffa Credo, Lyon, France; 28 cm × 21 cm × 18 cm) with metal grid floors in air-conditioned rooms (temperature, 21 ± 1°C; hygrometric degree, 55 ± 5%) under a 12-hr light-dark cycle (lights on from 7:00 A.M. to 7:00 P.M.). Filtered (0.22 µ) water was freely available, but access to standard laboratory food (A04, 4AR, Epinay sur Orge, France) was limited to 10 g/day, except during weekends when food was freely available between 5:00 P.M. Friday and 2:00 P.M. Sunday. Experiments were conducted between 9:00 A.M. and 5:00 P.M., Monday through Friday.

Apparatus. Experiments were conducted in standard operant conditioning chambers (model E10-10, Coulbourn Instruments, Lehigh Valley, PA) housed in light- and sound-attenuating enclosures that were ventilated by a fan, which also produced a masking noise. Each chamber contained a house-light that was mounted above a food pellet receptacle located between two levers, which were situated 2.5 cm above the grid floor. Food pellets (45 mg dustless pellets, Bioserv, Frenchtown, NJ) were delivered by a pellet dispenser (model E14-12, Coulbourn Instruments, Lehigh Valley, PA). Scheduling of reinforcement contingencies, reinforcement delivery and data recording were controlled by a SKED-11 system (State Systems, Kalamazoo, MI) implemented on a PDP-11 computer (Digital Equipment Corporation, Maynard, MA).

Discrimination procedure. Rats (n = 47) were trained to discriminate 8-OH-DPAT (0.16 mg/kg i.p.) from saline in a two-lever, food-reinforced FR10 drug discrimination paradigm by methods, including the training dose of 8-OH-DPAT (0.16 mg/kg i.p.), that were identical with those described recently (Koek et al., 1995; Kleven et al., 1997). The training drug 8-OH-DPAT (0.16 mg/kg i.p.) or saline were administered 15 min before 15-min training sessions during which responding on one of two levers was reinforced, depending on administration of either saline or drug. Discrimination training was continued until less than three responses were made on the injection-inappropriate lever before the first food reinforcement (FRF < 13), during ten consecutive sessions.

Pigeons

Male White Carneau pigeons (n = 15; Palmetto Pigeon Plant, Sumter, SC), weighing 500 to 650 g, were housed individually with unlimited access to water, food (Purina pigeon diet) and crushed oyster shell grit in an air-conditioned room (temperature, 21 ± 1°C; hygrometric degree, 55 ± 5%) and lighting was on from 7:00 A.M. to 7:00 P.M. To provide containment of fine particles, pigeons were housed in a room that was maintained at a pressure lower than that of the adjoining laboratory where the experiments were conducted.

Mixed grain was freely available in the home cage until the body weight was stable (i.e., S.E. less than 10% of the mean) during 5 consecutive days, at which time the free-feeding weight was calculated. Thereafter, pigeons were fed 5 g of mixed grain per day, until the pigeon's body weight was reduced to 80% of its free-feeding value. From then on, the animals were maintained at about 80% of their free-feeding weight in the following manner: on weekdays, mixed grain was given only when the body weight was less than 80% of its free-feeding weight. The quantity of mixed grain given was equal to the difference between the actual body weight and the 80% value, in grams. During weekends, generally between 10 to 20 g of food was given per day.

Apparatus. Experiments were conducted in standard operant conditioning chambers (model E10-10, Coulbourn Instruments, Lehigh Valley, PA). The chambers were housed in light- and sound-attenuating enclosures that were fan-ventilated. Each chamber contained two response keys (model E21-17, Coulbourn Instruments, Lehigh Valley, PA) mounted behind a 2.5-cm-diameter aperture on the midline of the front wall, 23 cm above the grid floor. The response keys could be transilluminated by red lights. Mixed grain (Friskies Repas Complet, Friskies, Brussels, Belgium) was presented by a feeder (model E14-10, Coulbourn Instruments, Lehigh Valley, PA) mounted behind a 5 × 5.5 cm aperture on the midline of the front panel, 17 cm below the response key. This aperture was illuminated during the 4-sec grain presentation by a white light. Scheduling of reinforcement contingencies, reinforcement delivery and data recording were controlled by a SKED-11 system (State Systems, Kalamazoo, MI) implemented on a PDP-11 computer (Digital Equipment Corporation, Maynard, MA).

Discrimination procedure. Drug discrimination training in pigeons was identical with that used in rats (e.g., pretreatment interval and training session duration were 15 min), with the following exceptions: 1) the drug- or saline-appropriate keys were illuminated during shaping sessions and both keys were illuminated during training/test sessions, whereas the rat operant chambers did not have lights above the levers; 2) the training dose of 8-OH-DPAT was 0.31 mg/kg i.m.; and 3) the FR (30), training criterion FRF (<40) and test validation FRF (45) were higher than the corresponding values in rats. Note that the FRF criteria were proportionately similar to those used in rats studies. The training dose of 8-OH-DPAT in pigeons (0.31 mg/kg) was chosen to correspond to that used by Barrett and colleagues (Zhang and Barrett, 1991; Barrett and Gleeson, 1992), whereas the dose used for rat studies (0.16 mg/kg i.p.) was chosen for historical reasons unrelated to the present study.

Data Analysis

Test sessions generated data on two variables: 1) the selected manipulandum, i.e., saline or drug key/lever, representing the measure of discriminative responding; and 2) the response rate, i.e., the total number of responses made on either manipulandum during the 15-min session, expressed as a percentage of the response rate during the most recently preceding saline training session. Selection data were used to calculate the percentage of animals at each treatment condition selecting the drug manipulandum. Drug effects on this variable were analyzed by the Litchfield and Wilcoxon procedure (Litchfield and Wilcoxon, 1949; Tallarida and Murray, 1987), implemented by use of the research programming language RS/1 (Bolt Beranek and Newman Inc., Cambridge, MA), to estimate ED₅₀ values and their 95% confidence limits. When fewer than two intermediate effects were observed, 0 and/or 100% effects were transformed by means of Berkson's adjustment (Hubert, 1984) to permit the use of the Litchfield and Wilcoxon procedure.

Apparent pA₂ values and their 95% confidence limits were calculated by Schild linear regression analysis (Arunlakshana and Schild, 1959). Apparent pA₂ values were calculated with the slopes constrained to 1, in cases where the 95% confidence limits of the Schild plot slopes included 1.

Drugs

Drugs in this study were purchased from Research Biochemicals Intl. (Natick, MA): 8-OH-DPAT HBr, BMY-7378 dihydrochloride, clozapine, NAN-190 HBr, TFMPP and ()-pindolol; Sigma Chemical (Fresnes, France): buspirone HCl, haloperidol, prazosin HCl and yohimbine HCl; or Janssen Pharmaceutica (Beerse, Belgium): fentanyl citrate injectable. Flesinoxan HCl, GR-127,935 dihydrochloride, idazoxan HCl, tertatolol, nemonapride, (S)-WAY-100135 HCl, WAY-100635 dihydrochloride, S14506 and WY 50,324 were synthesized by J.-L. Maurel, Centre de Recherche Pierre Fabre. The following drugs were gifts: BMS 110100 (also designated BMY 14802) and gepirone HCl (Bristol Myers Squibb Company, Wallingford, CT), eltoprazine HCl (Duphar, Weesp, Netherlands), FG 5974 HCl (Kabi Pharmacia AB, Malmö, Sweden), indorenate HCl (Department of Pharmacology, CINVESTAV; Mexico City, Mexico), ipsapirone HCl (also designated BAY Q 7821 or TVQX 7821; Bayer AG, Wuppertal-Elberfeld, Germany), LEK 8804 tartrate (LEK Pharmaceutical and Chemical Co., Ljubljana, Slovenia), lisuride maleate (Schering-Plough Corporation, Bloomfield, NJ), LY228729 (Lilly Research Laboratories, Indianapolis, IN), MDL-72832 HCl and MDL-73005 mesylate (Hoechst Marion Roussel, Inc. Cincinnati, OH), metanopirone citrate (also designated tandospirone or SM-3997, Sumitomo Pharmaceuticals Co., Ltd., Osaka, Japan), penbutolol sulfate (Hoechst, Frankfurt, Germany). Drugs were dissolved and administered in distilled water, with the exception of metanopirone, FG 5974, GR-127,935, LY228729, S-14506, lisuride, LEK-8804 and NAN-190, which were prepared as suspensions in aqueous Tween 80 (2 drops/10 ml distilled water), and BMY 14802 and ()-pindolol, which were dissolved in distilled water to which several drops of acetic acid were added and then the pH adjusted to 5 to 7. All drugs were injected in a volume of 10 ml/kg in rats and 1 ml/kg in pigeons. Doses are expressed as weight of the free base.

For tests of agonist activity, drugs were administered i.p. in rats and i.m. in pigeons, 15 min before sessions. For time-course studies, 8-OH-DPAT was administered i.p. (rats) or i.m. (pigeons) 30, 60, 120 or 240 min before sessions. For tests of antagonist activity, drugs, were injected s.c. or i.p. (in cases where Tween 80 was needed) in rats 60 min before the session, 45 min before administration of 8-OH-DPAT, and in pigeons, i.m., 30 min before the session, 15 min before administration of 8-OH-DPAT. The order of treatment with individual drugs and doses was unsystematic.

	Results

Top Abstract Introduction Methods Results Discussion References

Effects of 8-OH-DPAT. In both species, the 8-OH-DPAT versus saline discrimination was acquired within 100 sessions by more than 88% of the animals trained (table 1); however, the median sessions to criterion was significantly (P < .002) lower in pigeons than in rats. Stimulus control was maintained by 8-OH-DPAT in both species, as illustrated by the high percent correct-lever selections during drug and saline training sessions. Both rats and pigeons correctly chose the saline-appropriate manipulandum during 93 to 95% of saline training sessions; however, in contrast, rats made significantly (P < .02, paired t test) more errors during drug training sessions, i.e., correct lever selections were observed in 88% of drug training sessions versus 93% of saline training sessions.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Effects of 8-OH-DPAT in rats and pigeons trained to discriminate 8-OH-DPAT from saline. Top panels: symbols represent the percentage of animals selecting the drug-appropriate manipulandum during 15-min sessions; arrows indicate the training doses (i.e., 0.16 mg/kg i.p. and 0.31 mg/kg i.m., rats and pigeons, respectively). Middle panels: symbols represent rate of responding (Rsp Rate; mean ± S.E.M.) expressed as a percentage of rate immediately preceding control training sessions. Bottom panels: symbols represent the percentage of animals in which responding during test sessions exceeded the limits more than two S.D. from their own average control rates of responding. Errors are contained by the symbols with which they do not appear.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Time course of effects of 8-OH-DPAT in rats and pigeons trained to discriminate 8-OH-DPAT from saline. The details are as in figure 1. Discriminative stimulus and response-rate effects were examined 15, 30, 60, 120 and 240 min after administration of the training dose of 8-OH-DPAT in rats (0.16 mg/kg i.p.) and pigeons (0.31 mg/kg i.m.). (n = 7 animals/time point, with the exception of data obtained from the 15-min time point which are redrawn from figure 1).

Effects of other 5-HT_1A agonists. The results for 5-HT_1A agonists that produced high levels of drug-appropriate selection (i.e., 67%) in both rats and pigeons are summarized in table 2. These 5-HT_1A agonists engendered similar levels of drug-appropriate selection in both species, over a similar rank order of potencies. S 14506 was the most potent compound, with the ED₅₀ values differing from the least potent compound BMY 14802 by more than a factor of 50.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Relationship between potencies of 5-HT1A ligands in rat and pigeon. Numbers refer to data shown in table 2.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Drugs that do not have similar discriminative stimulus effects in pigeons (top panels) and rats (lower panels) trained to discriminate 8-OH-DPAT from saline. Values represent the percentage of animals selecting the drug-appropriate manipulandum during 15-min sessions. Open symbols: results of tests when drugs were administered alone. Closed symbols represent results when drugs were administered before the training drug, with the exception of the results shown in the lower left-hand panel, which were obtained when eltoprazine (2.5 mg/kg i.p.) was given together with GR 127,935 (0.63 mg/kg s.c.).

Effects of 5-HT_1A antagonists. In both rats and pigeons, the 5-HT_1A antagonists NAN-190, penbutolol, ()-pindolol, tertatolol and WAY-100635 decreased drug-appropriate selection from the high levels engendered by the training dose to less than 25% (table 5). The 5-HT_1A antagonist (S)-WAY-100135 did not block drug-appropriate selection in pigeons, whereas it was effective in rats. In contrast, as shown in figure 4, BMY-7378 antagonized 8-OH-DPAT-appropriate selection more completely in pigeons than in rats (67 vs. 14%, rats vs. pigeons, respectively).

Effects of other compounds. A variety of compounds that were examined as antagonists of the DS effects of 8-OH-DPAT in rats and pigeons failed to decrease drug-appropriate selection in more than 50% of animals tested (table 5). All of these compounds were tested up to doses that produced effects on response rates in both species (data not shown). The antipsychotics haloperidol and clozapine and the alpha-1 adrenergic antagonist prazosin engendered saline-appropriate selection in only 1 of 5 rats tested; fentanyl was ineffective at the highest dose that did not disrupt responding in most of the rats treated. Clozapine, prazosin and the mu opioid fentanyl did not antagonize drug-appropriate selection; and, as shown in figure 4, idazoxan, LEK8804 and yohimbine similarly failed to alter markedly drug-appropriate selection in pigeons or in rats.

Apparent in vivo pA₂ analysis. Pretreatment with ascending doses of WAY-100635 increased the doses of 8-OH-DPAT needed to engender drug-appropriate selection in both pigeons and rats (fig. 5, table 6). WAY-100635 pretreatment produced dose-dependent antagonism of the DS effects of 8-OH-DPAT that appeared to be more readily surmountable in rats than in pigeons. That is, maximal drug-appropriate selection reached 100% at all pretreatment doses of WAY-100635 in rats, whereas apparently lower levels (67-80%) were achieved in pigeons treated with the 0.04 and 0.16 mg/kg doses; higher doses of 8-OH-DPAT (5.0 and 10 mg/kg) completely suppressed responding in most of the pigeons tested. Estimated ED₅₀ values for 8-OH-DPAT after treatment with saline or WAY-100635 ranged from 0.059 to 1.1 mg/kg and from 0.038 to 1.7 mg/kg, rat and pigeon, respectively (table 6).

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Ability of WAY-100635 to antagonize the discriminative stimulus and response rate-decreasing effects of 8-OH-DPAT in rats and pigeons. Top panels: values represent the percentage of animals (n = 7/dose) selecting the drug-appropriate manipulandum. Middle panels: values represent the mean ± S.E. rates of responding expressed as a percentage of responses obtained during the most recent saline control session. Bottom panels: values represent the percentage of animals in which responding during test sessions exceeded the limits more than two S.D. from their own average control rates of responding. The details are as in figure 1.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Schild regression plots for WAY-100635 antagonism of the DS and response-rate decreasing effects of 8-OH-DPAT in rats (top panel) and pigeons (lower panel) trained to discriminate 8-OH-DPAT from saline. Dose ratios are the ED50 values of 8-OH-DPAT in the presence of WAY-100635 (0.01, 0.04 or 0.16 mg/kg) divided by the ED50 values of 8-OH-DPAT in the presence of saline (shown in table 6). Note that the slopes of regression lines, where shown, are not constrained to 1. Open symbols, response-rate effects; closed symbols, 8-OH-DPAT-appropriate selection effects.

	Discussion

Top Abstract Introduction Methods Results Discussion References

In this study the DS effects of the 5-HT_1A agonist 8-OH-DPAT were shown to be mediated by similar mechanisms in rats and pigeons, although not all 5-HT_1A ligands produced identical effects. Most of the 5-HT_1A agonists produced 8-OH-DPAT-like DS effects in both species, and 5-HT_1A antagonists were generally equieffective. Idazoxan, yohimbine, LEK 8804 and BMY 7378 produced lower levels of drug-appropriate selection in pigeons, however. Whereas BMY 7378 exhibited lower intrinsic activity in pigeons, the remaining compounds did not act as antagonists in either species. These findings may be related to either effect at other receptors (e.g., the case of eltoprazine) or species differences in apparent intrinsic activity (e.g., BMY 7378). Despite evidence for similar mechanisms, compounds that exhibit 5-HT_1A agonist properties in other assays may not produce similar DS effects in both species.

Several findings suggested that there were species differences in the discriminability of 8-OH-DPAT, although it is not evident that this altered the primary results. In contrast to rats, pigeons 1) acquired the discrimination faster; 2) made fewer errors during drug training sessions than during saline training sessions; and 3) showed a higher separation between the training dose and the ED₅₀. These results indicate that the discriminability of 8-OH-DPAT was higher in pigeons than in rats (Overton, 1974, 1982; Colpaert et al., 1980; Colpaert and Janssen, 1982). Although the DS effects could be mediated via different substrates, our results generally agree with previous reports in rats (Sanger and Schoemaker, 1992; Rabin and Winter, 1993; Ybema et al., 1993; Schreiber et al., 1995a; Sánchez et al., 1996) and pigeons (Barrett and Gleeson, 1992; Barrett et al., 1994) and systematic differences were not observed.

Our results largely agree with most previous reports that intermediate to high intrinsic activity 5-HT_1A agonists produce high levels of drug-appropriate selection in 8-OH-DPAT-trained rats. For example, the relative potencies for several of the agonists tested in the present study were highly correlated (r = 0.88, P < .002) with results from a similar drug discrimination study also using 8-OH-DPAT as the training drug in rats (Sanger and Schoemaker, 1992). In pigeons, 5-HT_1A agonists such as WY 50,324, flesinoxan, buspirone and ipsapirone engendered 8-OH-DPAT-like DS effects, consistent with prior studies of these compounds (Barrett and Zhang, 1991; Zhang and Barrett, 1991; Barrett and Gleeson, 1992). Further, 8-OH-DPAT-like DS effects were engendered by the 5-HT_1A ligands S14506 (Colpaert et al., 1992), MDL-72832 (Mir et al., 1988), MDL-73005 (Moser et al., 1990), LY228729 (Foreman et al., 1993), metanopirone (Schreiber et al., 1995b) and the mixed 5-HT_1A agonist/5-HT_2A/2C antagonist FG5974 (Albinsson et al., 1994).To our knowledge, the DS effects of these compounds have not been examined previously in pigeons, but the results are nonetheless consistent with their 5-HT_1A agonist properties. Thus, most of the 5-HT_1A agonists that we examined engendered 8-OH-DPAT-like DS effects in both species.

The significant correlation between agonist potencies in the two species indicates that the 5-HT_1A agonists produced their 8-OH-DPAT-like DS effects via similar mechanisms. Nonetheless, more compounds exhibited higher potency in pigeons. Although this could be related to the use of different training doses, it may be explained by pharmacodynamic factors such as differences in bioavailability or pharmacokinetics. Because different routes of administration were used, these factors may not be the same. Reports that 8-OH-DPAT is about 5 times more potent when given by the subcutaneous route than by the intraperitoneal route (Fuller and Snoddy, 1987; Sanger and Schoemaker, 1992) suggest a first-pass effect. Similar studies have not, to our knowledge, been conducted in pigeons; however, it would be reasonable to assume that when given by the intramuscular route, 8-OH-DPAT is rapidly absorbed and is not subject to first-pass metabolism. Nonetheless, potency estimates in the antagonism studies (and in particular the in vivo apparent pA₂ values) did not differ systematically. Thus, despite overall differences in agonist potencies, it is safe to conclude that the 8-OH-DPAT-like DS effects are pharmacologically similar in both species.

A variety of compounds (shown in tables 3 and 4) engendered intermediate levels of drug-appropriate selection in rats and/or pigeons and there were some apparent species differences. Species differences could result from 1) interactions with different receptor subtypes, 2) inherent differences in receptor-effector coupling or receptor reserve in the system(s) mediating the DS effects in rats and pigeons or 3) differences in the effective training dose (Colpaert, 1988; Koek and Woods, 1988). The involvement of different receptor subtypes is unlikely because of the high correlations between agonist and antagonist potencies. Moreover, the partial 5-HT_1A agonists BMY 14802 and MDL-73005 (Bristow et al., 1991; Rabin and Winter, 1993) engendered relatively high levels of drug-appropriate selection in pigeons (67% drug-appropriate selection), which indicated that low apparent efficacy did not uniformly yield intermediate levels of drug-appropriate selection in this species. Further, in contrast, idazoxan, which reportedly lacks efficacy in vitro at inhibiting forskolin-stimulated adenylate cyclase activity (Rabin and Winter, 1993), did not block 8-OH-DPAT in either pigeons or rats. Moreover, in contrast to BMY7378, yohimbine and the reportedly mixed 5-HT_1A agonist/5-HT_2A/2C antagonist LEK-8804 (Krisch and Bole, 1994) substituted partially, but did not antagonize the DS effects of 8-OH-DPAT in pigeons. Thus, factors other than intrinsic activity may explain partial 8-OH-DPAT-like DS effects.

Different training doses may affect the magnitude of drug-appropriate responding produced by low-efficacy agonists (Young et al., 1992), although the only compelling evidence for this is that the partial 5-HT_1A agonist BMY 7378 engendered low levels of substitution and exhibited antagonist properties in pigeons. In contrast, the opposite, weak antagonism and higher levels of drug-lever selection, was observed in rats. Nonetheless, the low-efficacy mixed 5-HT_1A agonists/₂-adrenoceptor antagonists yohimbine and idazoxan (Sanger and Schoemaker, 1992; Winter and Rabin, 1992; Rabin and Winter, 1993) produced lower levels of substitution in pigeons than in rats, but were not very effective in blocking the DS effects of 8-OH-DPAT in either species. Moreover, the partial agonists BMY14802 and MDL-73005 produced high levels of drug-appropriate selection in both rats (Sanger and Schoemaker, 1992; Rabin and Winter, 1993) and pigeons (present study). Thus, intermediate levels of drug-appropriate selection probably reflect other effects (e.g., rate-decreasing) that prevent the use of higher doses.

The finding that the mixed 5-HT_1A/1B/1D agonist eltoprazine produced higher levels of drug-appropriate selection pigeons than in rats (Barrett and Gleeson, 1992; present study) provides the clearest demonstration of the influence of multiple receptor subtypes on differential DS effects in rats and pigeons. This finding is also consistent with recent results obtained in flesinoxan-trained pigeons (Mos et al., 1997). In the present study, eltoprazine was shown to have 5-HT_1B agonist activity that interfered with 8-OH-DPAT-appropriate selection in rats. Pretreatment with the 5-HT_1D/1B antagonist GR-127,935 (O'Neill et al., 1996; Pauwels and Colpaert, 1996) significantly antagonized the rate-decreasing effects of eltoprazine (2.5 mg/kg), and perhaps as a consequence, a high percentage of rats selected the drug lever. These results agree with the hypothesis that 5-HT_1B agonist actions interfere with the ability of eltoprazine to engender drug-appropriate selection in 8-OH-DPAT-trained rats (Barrett and Gleeson, 1992).

As noted previously (Schreiber et al., 1995a), 8-OH-DPAT interacts with sites other than 5-HT_1A receptors: alpha-2 adrenoceptors, D₂ receptors and 5-HT reuptake sites (Schoemaker and Langer, 1986), although these other effects do not seem to play a role in its DS effects. In this study the antipsychotics clozapine and nemonapride, which have been identified as 5-HT_1A agonists in in vitro functional studies (Newman-Tancredi et al., 1996; Assié et al., 1997) or, in the case of nemonapride, in vivo (Assié et al., 1997), engendered partial substitution in pigeons, but were relatively inactive in rats, probably because of response-rate effects. The ability of these antipsychotics to produce higher levels of drug-appropriate selection in pigeons may be related to the general observation that pigeons were relatively insensitive to rate effects, but clozapine did not act as a partial agonist insofar as it did not block the DS effects of the training dose of 8-OH-DPAT, consistent with the reported absence of partial agonist effects in vivo (Assié et al., 1997).

A variety of 5-HT_1A ligands were examined for their ability to antagonize the DS effects of 8-OH-DPAT. Compounds such as NAN-190 and BMY7378 were initially described as full antagonists, yet are partial agonists in some 5-HT_1A receptor models (e.g., Greuel and Glaser, 1992). BMY 7378 and NAN-190 have been reported to antagonize the DS effects of 8-OH-DPAT in pigeons (Barrett and Gleeson, 1992), and both were effective antagonists in the present study. BMY 7378 produced higher levels of drug-lever selection in rats than others have reported (Winter and Rabin, 1992; Rabin and Winter, 1993), although the inability of BMY 7378 to block the DS effects of 8-OH-DPAT has been mentioned previously (Winter and Rabin, 1992). With the exception of the BMY 7378 and (S)-WAY-100135 results, the 5-HT_1A antagonists WAY-100635 (Forster et al., 1995), ()-pindolol (Hjorth and Carlsson, 1986), penbutolol (Hjorth and Sharp, 1993), tertatolol (Jolas et al., 1993) and NAN-190 (Glennon et al., 1988) exhibited similar effects in rats and pigeons. The differential effects of BMY7378 might be related to its partial agonist properties, but it is not clear why (S)-WAY-100135 was ineffective in pigeons. (S)-WAY-100135 reportedly has partial 5-HT_1A agonist properties in vivo (Löscher and Hönack, 1993; Assié and Koek, 1996), but it produced only intermediate levels of substitution. Although overall, 5-HT_1A antagonists exhibited similar 8-OH-DPAT-blocking effects in rats and pigeons, predicting the outcome for compounds that have partial agonist properties appeared to be less straightforward.

The nearly identical in vivo apparent pA₂ values further indicate that the potency of WAY-100635 was very similar in both species despite the use of different training doses and routes of administration. In both species the 8-OH-DPAT dose-effect functions for drug-appropriate selection were shifted to the right by increasing doses of WAY-100635. However, the analysis assumes that 1) both the agonist and antagonist are tested at the time of peak activity and 2) agonists and antagonists interact competitively at a single receptor. The time-course studies in rats and pigeons indicated that maximal drug-appropriate selection occurs at least 30 to 45 min after 8-OH-DPAT is given; duration of action and peak effects of WAY-100635 were not investigated, although in rats, peak effects are reportedly observed within 1 hr after administration (Hjorth et al., 1996; Romero et al., 1996). With respect to drug-receptor interactions, the finding that the slopes of the Schild plots did not differ significantly from the theoretical value of 1 suggests that the interaction was competitive; however, the statistical reliability of this finding may be affected by the small number of doses that were examined. Overall, the results suggest that WAY-100635 is a competitive and reversible antagonist in both species.

Because WAY-100635 does not have high affinity for other serotonergic or nonserotonergic receptors (Forster et al., 1995; Sánchez et al., 1996; Kleven et al., 1997; Mos et al., 1997), it is likely that the DS effects of 8-OH-DPAT are mediated by the 5-HT_1A receptor. Further, the in vivo pA₂ analysis suggested that the response-rate effects of 8-OH-DPAT in rats are mediated by the same receptor as its DS effects, in contrast to results obtained in pigeons. Parallel shifts in the 8-OH-DPAT dose-effect function were observed after pretreatment with ascending doses of WAY-100635 and the slope of the Schild plot did not differ significantly from the theoretical value of 1. In contrast, the effects of 8-OH-DPAT on response rate in pigeons were not reversed by WAY-100635 in a dose-related manner. Thus, because the effects of WAY-100635 were readily surmounted by high doses of 8-OH-DPAT, response-rate effects of 8-OH-DPAT are probably not mediated exclusively by 5-HT_1A receptors in pigeons. But 5-HT_1A receptors may play a small role, whereas the relatively high affinity of 8-OH-DPAT for DA receptors could explain the inability of WAY-100635 to block the response-rate decreasing effects of higher doses in pigeons.

We examined the ability of the antipsychotics haloperidol and clozapine, the alpha-1 adrenergic antagonist prazosin and the mu opioid agonist fentanyl to either substitute for or block the DS of 8-OH-DPAT. It has been reported that 8-OH-DPAT has partial DA agonist effects (Ahlenius et al., 1991); however, findings that neither haloperidol nor clozapine reduced drug-appropriate selection indicate that DA does not play a role in its DS effects. Similarly, because prazosin was not effective as an antagonist, in agreement with previous work in either rats (Tricklebank et al., 1987; Arnt, 1989) or pigeons (Zhang and Barrett, 1991; Barrett and Gleeson, 1992), alpha-1 adrenoreceptors are probably not involved. Although the finding that fentanyl was inactive is consistent with the pharmacological selectivity of this discrimination, our results do not agree with those from recent studies conducted (Morgan and Picker, 1995) wherein mu (morphine and fentanyl) but not kappa opioids (U50,488 and bremazocine) attenuated the DS in 8-OH-DPAT-trained rats. Perhaps procedural factors such as differences in route and time of administration, dose range or schedule of reinforcement obscured the effects. Training dose is probably not a factor, because Morgan and Picker (1995) showed that fentanyl attenuated drug-lever selection engendered by both low (0.1 mg/kg i.p.) and high training doses (0.3 mg/kg i.p.). Further studies are clearly needed; nonetheless, because fentanyl did not alter the DS effects in pigeons, the generality of opioid/5-HT_1A interactions may be relatively limited.

One interesting species difference that was apparent in the present study is that pigeons were relatively insensitive to response-rate effects, not only those produced by 5-HT_1A agonists, but also those produced by other compounds such as antipsychotics. Although this could be related to schedule requirement, response topography or neurobiological differences, it may nonetheless offer an explanation for the remarkable sensitivity of the pigeon to the anticonflict effects of 5-HT_1A ligands. That is, 8-OH-DPAT-like DS effects occurred at doses close to those that decreased the rate of responding in rats, whereas pigeons clearly tolerated higher doses of 8-OH-DPAT and other 5-HT_1A agonists. Thus, anticonflict effects may be inhibited in rats by mechanisms, possibly, but not necessarily mediated also by 5-HT_1A receptors. Other mechanisms cannot be excluded, such as the fact that the pigeon does not express the rat 5-HT_1B receptor (i.e., human 5-HT_1D) but does express a receptor which is pharmacologically similar to the human 5-HT_1D receptor (Waeber et al., 1989a,b). It is possible that lower response-rate effects of 5-HT_1A agonists in pigeons are a consequence of such neurobiological differences.

General conclusions. Previous pharmacological characterizations in rats and pigeons have indicated that the DS effects of 8-OH-DPAT are mediated by interactions at 5-HT_1A receptors. Although most of these studies have been conducted in rats, results suggest that an identical relationship exists in pigeons, a species that has proven useful in preclinical assays of the anxiolytic activity of 5-HT_1A receptor agonists. In all, the present results confirm that 5-HT_1A agonists engender 8-OH-DPAT-like DS effects in both rats and pigeons; nevertheless, it is evident that 5-HT_1A agonist properties are necessary but not sufficient to engender 8-OH-DPAT-appropriate selection. Other factors such as apparent intrinsic activity or actions at other 5-HT receptor subtypes may play a larger role in determining whether or not some 5-HT_1A ligands produce similar results in rats and pigeons. Actions at other receptors may be particularly important in some cases because these effects can limit the ability to detect 5-HT_1A agonist properties in vivo by use of drug discrimination methods.