The Pharmacological Profile of (R)-3,4-Dihydro-N-isopropyl-3-(N-isopropyl-N-propylamino)-2H-1-benzopyran-5-carboxamide, a Selective 5-Hydroxytryptamine_1A Receptor Agonist

Materials and Methods

Compounds.

The following radioactive compounds were used in the binding studies (the receptor and specific activity in parentheses): [³H]RX821002 (α₂-adrenoceptor; 53.0 Ci/mmol), [³H]strychnine (strychnine sensitive glycine site; 23.5 Ci/mmol), [³H]DAMGO (μ-opioid; 59 Ci/mmol), [³H]DPDPE (δ-opioid; 33 Ci/mmol), [¹²⁵I]neuropeptide Y (neuropeptide Y; 2118 Ci/mmol) and [¹²⁵I]somatostatin (somatostatin; 1867 Ci/mmol) from Amersham Pharmacia Biotech UK, Ltd. (Little Chalfont, Buckinghamshire, UK); [³H]8-OH-DPAT (5-HT_1A; 135.5 and 162.9 Ci/mmol), [³H]5-HT (5-HT_1B and 5-HT₆; 23.4 and 28.2 Ci/mmol), [³H]ketanserin (5-HT_2A; 64.6 and 60.1 Ci/mmol), [³H]prazosin (α₁-adrenoceptor; 25.0 and 72.2 Ci/mmol), [³H]rauwolscine (α₂-adrenoceptor; 82.3 Ci/mmol), [³H]dihydroalprenolol (β-adrenoceptor; 60.4 and 59.2 Ci/mmol), [³H]SCH23390 (dopamine D₁; 81.0 and 85.5 Ci/mmol), [³H]l-quinuclidinylbenzilate (muscarinic; 44.3 and 36.4 Ci/mmol), [³H]nicotine (nicotinic; 63 Ci/mmol), [³H]pyrilamine (histamine H₁; 31.2 Ci/mmol), [³H]tiotidine (histamine H₂; 83.7 Ci/mmol), [³⁵S]TBPS (GABA_A; 69.2 Ci/mmol), [³H]MK-801 (N-metyl-d-aspartate; 20.3 Ci/mmol), [³H]dl-α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid (dl-α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid; 53 Ci/mmol), [³H]flunitrazepam (benzodiazepine; 82.5 Ci/mmol), [³H]paroxetine (5-HT uptake site; 23.1 Ci/mmol), [¹²⁵I]galanin (galanin; 2200 Ci/mmol), [³H]mesulergine (5-HT_2C; 78 Ci/mmol), [³H]BRL 43694 (5-HT₃; 80 Ci/mmol), [³H]baclofen (GABA_B; 30.6 Ci/mmol), [³H]raclopride (dopamine D₂, D_2A, and D₃; 46.0 Ci/mmol), [³H]L-354,718 (CCK_A; 87 Ci/mmol) and [³H]L-365,260 (CCK_B; 73 Ci/mmol) were obtained from PerkinElmer Life Sciences (Boston, MA). NAE-086, [propyl-2,3-³H] (specific activity, 40.2 Ci/mmol) and NAE-086, [amido-¹⁴C] (specific activity, 52 mCi/mmol) were synthesized by Dr. Tom Werner, AstraZeneca R&D Södertälje (formerly Astra Arcus).

The unlabeled compounds were obtained from the following sources: (−) and (±)alprenolol, AstraZeneca R&D Mölndal (formerly Astra Hässle) (Mölndal, Sweden); butaclamol, cimetidine, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane hydrochloride (DOI), 8-OH-DPAT, MK-801, and galanin (1–16) (rat), Sigma/RBI (Natick, MA); citalopram and flupenthixol, Lundbeck A/S (Copenhagen, Denmark); methysergide, Sandoz AG (Basel, Switzerland); 1-{2-[bis(4-fluorophenyl)methoxy]ethyl}-4-(phenylpropyl)piperazine hydrochloride, Gist-Brokades n.V. (Delft, The Netherlands); 1-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline (EEDQ), 3-hydroxybenzylhydrazine dihydrochloride (NSD 1015), oxotremorine, 5-HT, maprotiline hydrochloride, (−)sulpride, (±)nicotine, GABA, glutamic acid, pargyline hydrochloride, and diazepam, Sigma-Aldrich (St. Louis, MO); glycine and phentolamine, Ciba Geigy AG (Basel, Switzerland); levallorphan, Hoffmann-La Roche AG (Basel, Switzerland); L-364,718, Merck Sharp and Dohme (Hoddesdon, UK); CCK-8, neuropeptide Y, and Tyr-somatostatin-14, Bachem AG (Bubendorf, Switzerland).

NAE-086 and related compounds were synthesized at AstraZeneca R&D Södertälje. Other chemicals were obtained from commercial sources and were of analytical grade.

Cells.

GH₄ZD10 (rat pituitary tumor) cells expressing the 5-HT_1A receptor and Ltk⁻ (mouse fibroblast) cells expressing human dopamine D_2A (long isoform) and D_2B (short isoform) were obtained from Dr. Olivier Civelli (Vollum Institute for Advanced Biomedical Research, Oregon Health Sciences University, Portland, OR). Chinese hamster ovary cells expressing human D₃, rat 5-HT₆, and rat 5-HT₇receptors were purchased from INSERM Institute (Paris, France).

Animals.

Male Sprague-Dawley rats (ALAB or B&K strain) were supplied by B&K Universal (Sollentuna, Sweden) (formerly ALAB Laboratorietjänst). Dogs (beagles) for the in vivo metabolism studies were supplied by Vema Hund AB (Örbyhus, Sweden). The experiments were approved by Stockholm South ethical committee for experiments on laboratory animals.

In Vitro Receptor Binding Studies.

Receptor binding assays for 5-HT_1A, bovine 5-HT_1B, 5-HT_2A, 5-HT₆, 5-HT₇, D₁, D₂, D_2A, D_2B, D₃, α₁, α₂, β, muscarinic, nicotinic, histamine H₁ and H₂, GABA_A, glutamate and galanin receptors, and the serotonin uptake site were carried out under conditions described previously (Jackson et al., 1995; Johansson et al., 1997). Receptor binding assays for μ- and δ-opioid receptors were carried out as described by Magnan et al. (1982).

Receptor binding assays at one concentration in triplicate for the following receptors were carried out by Battelle (Geneva, Switzerland): 5-HT_2C, 5-HT₃, glycine, GABA_B, CCK_A, CCK_B, neuropeptide Y, and somatostatin.

Cells were grown and the cell membranes were prepared essentially as described by Malmberg et al. (1993). TheK_i values (inhibition constants) of the compounds were determined from inhibition curves using the iterative nonlinear curve-fitting program LIGAND (Munson and Rodbard, 1980). The K_d values (dissociation constants) of the various radioligands used to calculate theK_i values were determined by saturation analysis.

The kinetic binding studies of [³H]NAE-086 were performed using rat cerebral cortex including hippocampus. The tissue was homogenized with an Ultra-Turrax in ice-chilled 50 mM Tris-HCl buffer, pH 7.5, containing 4 mM CaCl₂ and 5.7 mM ascorbic acid. The homogenate was centrifuged at 39,000g for 10 min, and the pellet was rehomogenized in the same buffer and recentrifuged. This procedure was repeated, and the final pellet was homogenized and suspended in assay buffer to give a final concentration of 10 mg of original wet weight/2 ml of incubation volume. To remove endogenous serotonin, membranes were preincubated for 10 min at 37°C, after which pargyline was added to give a final concentration of 10 μM, and the incubation was continued for further 10 min. The kinetic binding studies of [³H]NAE-086 were performed at 22°C using 5.8 nM radioligand. In the association studies, total and nonspecific binding (10 μM 5-HT) were determined at different time points (0–60 min). The dissociation was initiated by addition of 10 μM 5-HT and the binding was measured during 0 to 60 min. The experiments were run in triplicate. The association and dissociation rate constants (k₋₁ andk₊₁) were calculated as described byBennett (1978) and the K_d value (k₋₁/k₊₁) was determined.

VIP-Stimulated Adenylyl Cyclase in GH₄ZD10 Cells.

The GH₄ZD10 cells expressing 5-HT_1A receptors were cultured as described previously (Albert et al., 1990, Fowler et al., 1992; Johansson et al., 1997). The adenylyl cyclase assay used was based on the method ofDorflinger and Schonbrunn (1983) with some minor modifications (Fowler et al., 1992; Johansson et al., 1997).

In Vivo Metabolism of NAE-086.

Urine was collected over the first 24 h from rats and dogs administered [¹⁴C]NAE-086 solution. The pH of the urine was adjusted to pH 12 with 1 M NaOH. The urine was extracted with 3 volumes of chloroform or a mixture of hexane/diethyl ether/butanol (70:30:5). The organic phase was collected and evaporated to dryness, after which the residue was dissolved in mobile phase before HPLC chromatography and detection. A YMC C18 ODS-A column (150 × 4.6 S-3 μm) was used for HPLC fractionation. The mobile phase consisted of 0.05 M phosphate buffer, pH 2.5, containing 0.6 mM octyl sulfate and 17% acetonitrile. The flow rate was 1 ml/min. Detection was by UV absorption at 205 nm using a Spectra 100 (Spectra-Physics, San Jose, CA) spectrophotometer. The radioactive peaks were detected by a FLO-ONE/β - detector (Radiometric; Canberra Instrument Co., Meriden, CT). The HPLC pump used was an LKB 2150. The eluted peaks were compared with those of synthetic reference compounds. A gas chromatography-mass spectrometry procedure was used for structural confirmation of the separated urinary metabolites.

5-HTP and DOPA Accumulation.

The rats were administered with the test compound at the time noted before the intraperitoneal injection of 100 mg/kg NSD 1015. In some experiments, 6 mg/kg i.p. EEDQ was injected 24 h beforehand. The rats were killed with a guillotine 30 min after the NSD 1010 injections, the brains were removed rapidly, and the regions dissected were frozen immediately on dry ice. The samples were stored at −70°C until assayed.

5-HT and Dopamine Turnover.

The test compound was administered at the time noted before the rats were decapitated. The brain regions to be analyzed were treated as described above.

Determination of 5-HTP, DOPA, 5-HT, Dopamine, and Their Metabolites.

The concentrations of 5-HTP, DOPA, 5-HT, 5-HIAA, dopamine, 3,4-dihydroxyphenylacetic acid, and homovanillic acid in various brain regions were determined by the use of high-performance liquid chromatography with electrochemical detection according to a modification of the method of Magnusson et al. (1980) as described previously (Johansson et al., 1997).

Serum Corticosterone.

Rat serum corticosterone was determined as described by Kelder and Ross (1992). The rats were treated with saline injections once daily for 5 days before the experiment to reduce stress-induced corticosterone increase.

Flat Body Posture and Forepaw Treading.

Flat body posture and forepaw treading were determined as described by Johansson et al. (1997). The intensity of the behavior was tested between 10 and 12 min after the administration of the test compound.

Temperature Measurements.

Rectal temperature was measured as described by Johansson et al. (1997). Rectal temperature was recorded before the administration of the test compound and 30, 60, 90, and 120 min thereafter.

Cage-Leaving Response.

The cage-leaving response in rats was determined as described by Rënyi et al. (1986). The number of rats leaving the cage and the time to leave the cage during the observation period 10 to 22 min after the test compound were recorded.

Wet-Dog Shake Response.

Wet-dog shake response in rats was measured as described by Ross et al. (1992). The number of rats showing wet-dog shakes and the number of wet-dog shakes during the observation period 5 to 65 min after the challenge dose were recorded.

Salivation and Male to Male Mounting with Penile Erection.

Salivation and male to male mounting with penile erection were observed during the cage-leaving test.

Statistics.

The mean and S.E.M. were used as measures of central tendency and variation, respectively. Analysis of variance (ANOVA) was performed using SYSTAT 7.0: New Statistics (SPSS Inc., Chicago, IL), followed by a post hoc test as stated for each experiment. In some behavioral experiments Mann-Whitney Utest was used. Medians ± quartiles were calculated according toSnedecor and Cohran (1967).

Results

In Vitro Experiments

Receptor Binding Profiles of NAE-086 and Its Metabolites.

Both NAE-086 and the metabolite NAE-111 showed high affinity and selectivity for rat hippocampal 5-HT_1A receptors withK_i values of 4.5 ± 0.7 nM (n = 8) and 0.26 ± 0.03 nM (n = 3), respectively (Table 1). TheS-enantiomer of NAE-086 (NAE-084) had more than 200 times less potency (K_i 1100 ± 200 nM). Other related compounds had low affinity for the 5-HT_1A receptor. Apart from an affinity of 240 nM at 5-HT₇ receptor the affinity of NAE-086 was less than 500 nM for all other 30 receptors studied (data not shown; see Materials and Methods for list of receptors tested).

Kinetics of [³H]NAE-086 Binding.

The time course of the binding of [³H]NAE-086 to membranes of rat cerebral cortex including hippocampus at 22°C is shown in Fig.2. The observed association rate constantk_obs obtained from the linear relationship was 0.0835 and the forward rate constant (k₊₁ =k_obs −k₋₁/[L]) for the [³H]NAE-086 binding at 22°C was 0.009 × 10⁹M⁻¹min⁻¹. The dissociation rate constant (k₋₁), determined by addition of 10 μM 5-HT, was 0.0315 min⁻¹. The dissociation was completed after 60 min with a half-time dissociation value (t_1/2) of 22 min (graphically determined). The calculated ratio of the rate constants (k₋₁/k₊₁) gave an estimate of the K_d value of 3.5 nM, which is in good agreement with theK_i value obtained using [³H]8-OH-DPAT, i.e., 4.5 ± 0.7 nM.

Effects on VIP-Stimulated cAMP Production.

5-HT (1 μM), 8-OH-DPAT (1 μM), and buspirone (3 μM) reduced the cAMP response to VIP by 55, 50, and 20%, respectively (Table2). These responses could be antagonized by 10 μM (−)alprenolol. (−)Alprenolol per se did not affect VIP-stimulated cAMP production. NAE-086 and its enantiomer, NAE-084, were tested concomitantly with the standard compounds. NAE-086 reduced the cAMP response concentration dependently, the maximum reduction (at 10 μM) being 79% of that found with 1 μM 5-HT (Fig.3). The combination of 10 μM NAE-086 + 1 μM 5-HT produced a slightly lower inhibition of cAMP than that seen with 1 μM 5-HT alone. This suggests that the inhibition seen at 10 μM NAE-086 is close to the maximum obtainable for this compound. The concentration of NAE-086 producing half the inhibition observed at 10 μM was determined graphically from the mean data to be 220 nM. The inhibitory effect of 1 μM NAE-086 was antagonized by 1 to 10 μM (−)alprenolol (Fig. 3, insert). NAE-084 (10 nM–10 μM concentration range) was without effect upon either VIP-stimulated cAMP production per se or upon the inhibitory effect of 1 μM 5-HT (data not shown). This indicates that over this concentration range, NAE-084 does not interact with the 5-HT_1A receptor-adenylyl cyclase complex.

In Vivo Experiments

Minimum effective doses of NAE-086 in various in vivo experiments are summarized in Table 3.

In Vivo Metabolism of NAE-086.

A total of seven metabolites were identified in the urine of the rat and dog (Fig. 1). (R)-3,4-Dihydro-3-(N-ethyl-N-isopropylamino)-N-isopropyl-2H-1-benzopyran-5-carboxamide was the most predominant metabolite in rat urine, accounting for as much as 20% of the dose. (R)-3,4-Dihydro-N-isopropyl-3-(N-isopropylamino)-2H-1-benzopyran-5-carboxamide and (R)-3-amino-3,4-dihydro-2H-1-benzopyran-5-carboxamide were the most predominant urinary metabolites in dogs accounting for levels as high as 17 and 9%, respectively. NAE-111 levels amounted to about 3% of dose in rat urine, whereas only trace amounts were present in the urine of dogs.

5-HTP Accumulation.

NAE-086 reduced dose dependently the accumulation of 5-HTP. The time course of this decrease in hypothalamus after s.c. (1 mg/kg) and p.o. (3 mg/kg) administration is shown in Fig.4. The time courses in the other three brain regions examined were similar (data not shown). The maximal effect after s.c. administration was seen when NAE-086 was given 30 min before NSD 1015 and the rats were killed 30 min thereafter, whereas that after p.o. administration was seen when NAE-086 was given 1 h before NSD 1015. At the doses of NAE-086 used here, the maximal s.c. effect was about 65% of the total 5-HTP formation in control rats whereas the maximal effect after p.o. administration was about 50%. The duration of the effect was quite long-lasting, with significant effects 4 h after administration. The dose-response relationship was determined when NAE-086 was given s.c. 30 min before NSD 1015 (Fig.5). This experiment also included rats pretreated with 6 mg/kg i.p. EEDQ 24 h beforehand. This compound irreversibly destroys many receptors, including 5-HT_1A receptors. Because of large receptor reserve of 5-HT_1A receptors in the raphe nuclei, by partially inactivating receptors, EEDQ shifted the dose-response curves to the right without changing the maximal response. The ratio between the ED₅₀ doses with and without EEDQ treatment (q) gives an estimation of the fraction receptors not inactivated by EEDQ and also gives an estimation of the intrinsic efficacy of the compound (Minneman and Abel, 1984; Cox et al., 1993). As shown in Table 4 the qvalues obtained for the different agonists suggest a reserve of 5-HT_1A receptors in the raphe nuclei of about 80% for NAE-086 and 8-OH-DPAT and 50 to 60% for buspirone.

5-HT Turnover (5-HIAA/5-HT).

After oral administration, NAE-086 reduced the 5-HIAA/5-HT ratio in the rat brain by about 40% in hypothalamus and frontal cortex but less in hippocampus and particularly in striatum (Fig. 6A). In this experiment the inhibition in frontal cortex was also significant at the lowest dose (0.1 mg/kg) examined (Fig. 6B). The low inhibition of the 5-HIAA/5-HT ratio in striatum was also observed in the rats treated with NSD 1015 after oral administration of NAE-086 (Fig. 6C). The corresponding effects of subcutaneous NAE-086, also in NSD 1015-treated rats, showed less pronounced inhibition in striatum than in the other brain regions, and the inhibition in frontal cortex was similar to that in hypothalamus (Fig. 6D). Subcutaneous administration of 8-OH-DPAT in NSD 1015-treated rats also produced less of an effect in striatum than other regions examined (Fig. 6E).

DOPA Accumulation.

A significant increase in DOPA accumulation in hypothalamus was found 30 min after 3 mg/kg p.o. NAE-086 (128 ± 8% of controls, p < 0.05; Dunnett's ttest). No significant increases were obtained in other regions.

Dopamine Turnover.

The ratio of 3,4-dihydroxyphenylacetic acid/dopamine in striatum was increased significantly (153 ± 10%; p < 0.05; Dunnett's t test) 30 min after 3 mg/kg p.o., but the effect had already disappeared 30 min later.

Corticosterone Secretion.

NAE-086 increased serum corticosterone in rats with a significant effect at 0.3 mg/kg s.c. and 1 mg/kg p.o. when determined 1 h after administration (Fig.7A). This effect had disappeared 2 h after administration (Fig. 7B). The increase in corticosterone after a challenge dose (1 mg/kg s.c.) of NAE-086 was attenuated significantly 24 h after single doses of either 3 mg/kg s.c. or 10 mg/kg p.o. NAE-086 (Fig. 7C).

Flat Body Posture and Forepaw Treading.

NAE-086 induced the 5-HT_1A receptor-mediated behavior syndrome with minimum effective dose of 0.3 mg/kg s.c. and 10 mg/kg p.o. (Table 3). The corresponding effect of 8-OH-DPAT was obtained at 0.3 mg/kg s.c. and >5 mg/kg p.o.

Hypothermia.

NAE-086 significantly and dose dependently decreased rectal temperature in rats at 0.3 and 1 mg/kg s.c. and 1 to 10 mg/kg p.o (Fig. 8, A and B). NAE-111 also potently reduced rectal temperature also at the lowest dose tested, 0.03 mg/kg s.c., and showed a minimum effective oral dose of 1 mg/kg (Fig. 8, C and D).

Cage-Leaving Response.

Acute administration of NAE-086 inhibited the cage-leaving response at a minimum effective dose of 1 mg/kg s.c. and 3 mg/kg p.o. (Table 3).

Attenuation of 5-HT_1A Receptor Responses.

A single administration of NAE-086 to ALAB Sprague-Dawley rats attenuated the effect of a challenge dose of 8-OH-DPAT or NAE-086 given 24 h later in the following tests: corticosterone secretion (Fig. 6C), hypothermia (Table 5), and the cage-leaving response (Table 6). B&K Sprague-Dawley rats were less sensitive in developing tolerance to the 8-OH-DPAT-induced 5-HT_1A syndrome (flat body posture and forepaw treading) and the inhibition of cage-leaving response, but these parameters were attenuated markedly 24 h after the last administration of 3 weeks treatment with 1 mg/kg s.c. or 3 mg/kg p.o. NAE-086 twice daily (Fig. 9; Table 7).

Wet-Dog Shake Response.

5-HT_1A receptor agonists do not induce wet-dog shakes because these are mediated by 5-HT₂ receptors. However, as reported previously (Ross et al., 1992) the number of spontaneous wet-dog shakes was increased significantly 24 h after a single administration of 8-OH-DPAT to ALAB Sprague-Dawley rats. As shown in Table 6, NAE-086 at 10 mg/kg s.c. or p.o., like buspirone and other 5-HT_1A receptor agonists, significantly increased spontaneous wet-dog shakes in the ALAB rat strain when observed 24 h later. In the B&K rat strain neither 0.1 mg/kg s.c. 8-OH-DPAT nor 1 mg/kg s.c. NAE-086 increased the number of spontaneous wet-dog shakes or those induced by 0.3 mg/kg s.c. DOI when tested 24 h later (Table 8). Repeated treatment of rats twice daily for 3 weeks with 8-OH-DPAT and NAE-086 did, however, double the number of DOI-induced wet-dog shakes when challenged 24 h after the last administration in this strain (Table 8).

Salivation and Male-to-Male Mounting with Penile Erection.

Acute treatment with doses of up to 30 mg/kg s.c. 8-OH-DPAT did not induce salivation or male-to-male mounting with penile erection in B&K Sprague-Dawley rats (Table 7). However, 24 h after the last administration of 3 weeks, twice daily, treatment with 0.1 mg/kg s.c. 8-OH-DPAT-treatment, a challenge dose of 1 mg/kg s.c. 8-OH-DPAT caused pronounced salivation, and male-to-male mounting with penile erection was observed during a period 3 to 12 min after injection. Similar effects were induced after 3 weeks treatment twice daily with 1 mg/kg s.c. or 3 mg/kg p.o. NAE-086.

Discussion

In Vitro Effects.

The in vitro receptor binding profile of NAE-086 indicates that this ligand is one of the most selective 5-HT_1A receptor agonists so far reported. The only other receptor of the 32 examined for which NAE-086 had some affinity was the 5-HT₇ receptor, and this was 50 times less than for the 5-HT_1A receptor. TheS-enantiomer of NAE-086 (NAE-084) has a much lower affinity for the 5-HT_1A receptor, demonstrating a greater degree of stereoselectivity for this receptor than previously reported for 8-OH-DPAT.

NAE-111, a minor metabolite of NAE-086, is a potent 5-HT_1A receptor agonist with 17 times greater affinity than NAE-086 itself and appears to have the same binding profile as NAE-086. Although a minor metabolite in rats, NAE-111 may contribute significantly to the effects of NAE-086 due to its high potency, especially after oral administration (V. Hillegaart, J. Gabrielsson, D. Lake-Bakaar, and S. Ahlenius, manuscript in preparation). The inactive metabolite (R)-3,4-dihydro-3-(N-ethyl-N-isopropylamino)-N-isopropyl-2H-1-benzopyran-5-carboxamide accounted for 20% of the dose in the rat urine.

[³H]NAE-086 bound reversibly to 5-HT sensitive sites in the rat cerebral cortex/hippocampus with rapid association and dissociation rates at 22°C. Taken together with the findings that NAE-086 has a high selectivity for 5-HT_1Areceptors, there are good grounds to suggest that the 5-HT sensitive binding of [³H]NAE-086 occurs to this subtype of 5-HT receptors. In accordance with this suggestion is the observation that the calculated K_dvalue of 3.5 nM is close to the K_ivalue of 4.5 nM reported for the inhibition of the binding of [³H]8-OH-DPAT to hippocampal membranes.

In the VIP-stimulated adenylyl cyclase assay, NAE-086 was shown to inhibit the cyclase activity with an efficacy of 79% of that of 5-HT. Accordingly, in vitro NAE-086 is a 5-HT_1Areceptor agonist with an intrinsic efficacy greater than that of buspirone and similar to that of 8-OH-DPAT, albeit less potent. Similar results were obtained in experiments measuring forskolin-stimulated cAMP in rat hippocampal membranes (data not shown).

NAE-086 had no significant inhibitory effect on the neuronal 5-HT, dopamine, or noradrenaline transporters in vitro, nor did it inhibit monoamine oxidase A or B forms in vitro (S. B. Ross, unpublished results).

In Vivo Biochemical Effects.

The results obtained with the 5-HTP accumulation technique confirm that NAE-086 is a potent 5-HT_1A receptor agonist under in vivo conditions after both subcutaneous and oral administration. They also show that the efficacy of NAE-086 is similar to that of racemic 8-OH-DPAT. These results complement the data from hippocampal membranes, to show that the agonist effects of NAE-086 are also evident at the presynaptic, somatodendritic 5-HT_1A autoreceptors. The rapid onset of the effect after s.c. injection with maximal effect within 30 min indicates that it is mainly NAE-086 itself and not a metabolite that exerts the effect at the receptor under these conditions. After oral administration it is more likely that the compound is subjected to first-pass metabolism and that the active metabolite NAE-111 contributes to the overall pharmacological effect. The rather flat dose-response curves after oral administration shown in Fig. 5 may reflect this.

The short-lasting increase in dopamine turnover and DOPA accumulation in hypothalamus after NAE-086 administration is probably an indirect effect mediated by postsynaptic 5-HT_1A receptors, because previous experiments with 8-OH-DPAT have shown a similar effect that is resistant to 5-HT depletion withp-chlorophenylalanine (L.-G. Larsson and S. B. Ross, unpublished observation).

Behavioral Effects.

Many of the 5-HT_1Areceptor-mediated responses are originated postsynaptically, e.g., the behavioral syndrome, corticosterone secretion, and probably also the hypothermia (Lucki, 1992). We have observed that the same rat strain from different breeders differs in sensitivity for 5-HT_1A receptor-mediated effects, both in the potency to activate the receptors and to desensitize the receptors upon repeated treatment (Kelder and Ross, 2001). Thus, the ALAB Sprague-Dawley strain from B&K Universal and the Sprague-Dawley strain from Møllegaard (Vejle, Denmark) had high sensitivity for 8-OH-DPAT and the responses were attenuated strongly after a single dose (Larsson et al., 1990; Kelder and Ross, 1992; Rënyi et al., 1992; Ross et al., 1992). Other Sprague-Dawley rat strains, e.g., from B&K and from Charles River (Uppsala, Sweden), were less sensitive for this class of compounds and require repeated treatment to show attenuation of the receptor responses (Kelder and Ross, 2001). These discrepancies make comparisons between laboratories difficult and even comparisons between data obtained at different times in the same laboratory. Regardless of these differences, it is well established that 5-HT_1A receptor agonists cause a development of tolerance upon repeated treatment (for review, see De Vry, 1995). As shown in the present study NAE-086, like other 5-HT_1A receptor agonists, desensitized many 5-HT_1A receptor responses. Interestingly, however, it was found that, simultaneously with the development of the tolerance to the 5-HT_1A receptors, sensitization occurred to the 5-HT₂ receptor agonist DOI. Thus, in one experiment a single administration of NAE-086, like 8-OH-DPAT, induced an increase in spontaneous wet-dog shakes, a 5-HT_2A-mediated behavior (Schreiber et al., 1995), 24 h later. In another experiment, with less sensitive rats, 3 weeks of treatment with NAE-086 or 8-OH-DPAT increased the wet-dog shakes induced by a challenge dose of DOI when tested 24 h after the last administration of the 5-HT_1Aagonist. This sensitization of the 5-HT_2Areceptor response is probably a direct consequence of the tolerance toward the 5-HT_1A receptor responses, reflecting the balance between 5-HT_1A and 5-HT₂ receptor systems (Rënyi, 1991;Schreiber et al., 1995). The observations of salivation and male-to-male mounting with penile erection after a high challenge dose (1 mg/kg s.c.) of 8-OH-DPAT in rats that had been treated repeatedly with NAE-086 or 8-OH-DPAT are interesting and do not seem to have been reported previously. Although not further analyzed in the present study, these behavior changes indicate a strong sensitization of serotoninergic systems and may also involve other neuron systems, e.g., dopaminergic and cholinergic systems.

The finding in the present study that prolonged treatment with NAE-086 appears to sensitize 5-HT₂ receptor responses 1 day after the last administration may explain the observation of some side effects in the clinical phase I study of NAE-086 (E. Widerlöv, K.-G. Jostell, and F. Undén, manuscript in preparation). In this trial, three healthy volunteers had hallucinations and “nightmares” after a short period of repeated administration of NAE-086. Because the hallucinogenic action of certain psychotomimetic drugs seems to occur via stimulation of 5-HT₂ receptors, it seems plausible that the effect seen in the volunteers was triggered by hypersensitive 5-HT₂ receptors when the serotoninergic neurotransmission returned to normal after the NAE-086-induced blockade of the cell-firing via the somatodendritic autoreceptors had ended. The only 5-HT_1A receptor agonist that has reached therapeutic use during the almost two decades that this receptor subtype has been known is buspirone. Besides being a partial 5-HT_1A receptor agonist, this drug is also a dopamine receptor antagonist, which may explain why it is possible to use buspirone without side-effects of the type observed for NAE-086.

In summary, NAE-086 was shown in rats to be a very selective 5-HT_1A receptor agonist with an efficacy in vitro and in vivo similar to that of 8-OH-DPAT. It has all the effects typical for a 5-HT_1A receptor agonist on serotoninergic neurotransmission, endocrinology, and behavioral responses after subcutaneous and oral administration. In the rat a minor metabolite of NAE-086, NAE-111, with 17 times higher affinity for 5-HT_1A receptors but with similar binding profile, may contribute to the in vivo effects of NAE-086, especially after oral administration. One or more NAE-086 injections resulted in a diminution of the following 5-HT_1A-mediated responses: corticosterone secretion and behavioral effects including inhibition of cage-leaving and hypothermia. Simultaneously with the desensitization of the 5-HT_1A receptor responses, the 5-HT_2A receptor-mediated wet-dog shake response, spontaneous or DOI-induced, was hypersensitized.