SYNTHESIS AND EVALUATION OF LIGANDS FOR D₂-LIKE RECEPTORS: THE ROLE OF COMMON PHARMACOPHORIC GROUPS

Abstract

Arylcycloalkylamines, such as phenyl piperidines and piperazines and their arylalkyl substituents, constitute pharmacophoric groups exemplified in several antipsychotic agents. A review of previous reports indicates that arylalkyl substituents can improve the potency and selectivity of the binding affinity at D₂-like receptors. In this paper, we explored the contributions of two key pharmacophoric groups, i.e., 4′-fluorobutyrophenones and 3-methyl-7-azaindole groups, to the potency and selectivity of synthesized agents at D₂-like receptors. Preliminary observation of binding affinities indicates that there is little predictability of specific effects of the arylalkyl moieties but the composite structure is responsible for selectivity and potency at these receptors.

It is now widely accepted based on gene cloning and recombinant DNA techniques that there are least five major dopamine (DA) receptor subtypes classified as D₁, D₂, D₃, D₄ and D₅. Originally, these receptors were classified into only two groups, D₁-like and D₂-like receptors with D₁ and D falling into the first and D₂, D₃ and D₄ making up the later group.¹ Of the two groups, the D₂- receptors have been the subject of great therapeutic interest because of their involvement in several psychiatric disorders.² The D₂ subtype receptor has been identified as the primary site of action antipsychotic agents.³ In addition, they are also implicated in the reinforcing and dependency producing drugs of abuse.⁴ The D₄ receptor subtype mediates functions that include motor activity, initiation and inhibition of behavior and working memory.^5–7 More recently, the D₄ receptor subtype has attracted attention because of its association with the induction of penile erection.^8–10 While D₂ and D₄ subtypes have become potential targets for drug development for several therapeutic indications, the functions of the D₃ subtype have remained largely uncertain.²

Thousands of DA ligands have appeared in the literature over the years.² However, a cursory evaluation of the common structural features in D₂ and D₄ receptor subtype ligands reveals consistent presence of arylcycloalkylamines in the form of alkylated arylpiperidines such haloperidol (Chart 1) and piperazines such as clozapine. The nature of the alkylated moieties varies and there is little evidence to suggest the role of these alkyl moieties in the selectivity of the ligands for each receptor subtype. In an attempt to understand the structural contributions of pharmacophoric elements at D₂-like receptors, we have compared the haloperidol analog, 1 with Merck compound, L745,870 (Chart 1).^11,12 In addition, several other publications have evaluated 3 methyl-7-azaindole and 3-methylindole moieties for D4 receptor selectivities.¹³

Chart 1.

The comparison of the binding affinity data at cloned human D₂-like receptors suggests that the presence of the butyrophenone and the 3-methyl-7-azaindole moieties significantly affects binding affinity and selectivity of these compounds at the D₂-like receptors.¹¹ In particular, a comparison of compound 1 and L745,870 suggests that the presence of the 3-methyl-7-azaindole moiety on 4- chlorophenyl piperazine confers ~40-fold D₄ potency on L745,870 while the butyrophenone confers less than a 4-fold D₂ potency. In addition, the 3-methyl-7-azaindole moiety appears to have increased D₄ selectively from 15-fold to over 2200-fold. On the other hand, the arylpiperidine and arylpiperazine groups common among CNS drugs appear to have preferences for the D₂ and D₄ subtype receptors, respectively. The aim of this study was to further explore the role of the two alkyl moieties and their impact on D₂/D₄ selectivity and potency (Charts 2 and 3).

Chart 2.

Chart 3.

CHEMISTRY

The binding affinities of compounds 1, 2a–c, 6, 8, 10, 12 and 14 were previously reported.^11,14 However, the full details of the synthetic procedures for several of them were not provided nor discussed. The key intermediate for the synthesis of compounds 2a–c and 3a–c, 3-(4-chlorophenyl)-8-azabicyclo[3.2.1]octan-3-ol, was obtained by treating commercially available carbamate protected tropinone with 4-chlorophenyl magnesium bromide under Grignard reaction conditions to form a carbamate protected aminoalcohol (17) which was decarbamylated to the aminoalcohol 18.

Compounds 2a–c were obtained by simple alkylation of compound 18 with the appropriate alkylating groups (Scheme 1). Compound 18 was also subjected to treatment with 7-azaindole and formaldehyde under Mannich reaction conditions, to give the desired product, 3a (Scheme 3). Compound 3b was similarly prepared using indole instead of 7-azaindole and the synthesis of compound 3c was accomplished by alkylating compound 18 with 3-(2-bromoethylindole (Scheme 4). Compound 24 (Scheme 2), which was previously reported by our laboratory,¹⁵ served as the key intermediate for the synthesis of compounds 4 and 5. The first step was to convert the commercially available benzyl-protected pyrrolidinol, 20 to the carbamate-protected pyrrolidinol (21) in order to avoid the anticipated dechlorination that often accompanied debenzylation under hydrogenolysis conditions. Oxidation of compound 21 to form ketone 22 and subsequent Grignard reaction with 4-chlorophenyl magnesium bromide produced the carbamate-protected pyrrolidinol 23. Deprotection with potassium hydroxide produced 24 and subsequent alkylation with the appropriate alkylating agents yielded the desired compounds 4 and 5 (Schemes 3 and 4).

Scheme 1.

Scheme 3.

Scheme 4.

Scheme 2.

We have previously reported the detailed synthetic procedures for compound 6 including the synthesis of 1-(4-chlorophenyl)-1,4-diazepane, using CuI-catalyzed coupling of 4-chlorophenyl iodide with 1,4-diazepane.¹⁴ Mannich reaction involving 1-(4-chlorophenyl)-1,4-diazepane, formaldehyde and 7-azaindole produced compound 7 in good yield (Scheme 3). The synthesis of compounds 8 and 9 required the previously synthesized 9-methyl-3,9-diazabicyclo[4.2.1]nonane (25) as a key intermediate.¹⁴ To obtain compound 8, intermediate 25 was carbamylated using ethyl chloroformate (26) and then demethylated to yield the secondary amine of the bridged nitrogen. N-arylation of 27 with 4-chlorophenylboronic acid produced compound 28 which underwent deprotection to now produce the secondary amine of the other nitrogen (29) (Scheme 5). Alkylation of compound 29 with 4-chloro-4′-fluorobutyrophenone delivered the target compound 8 as desired. Compound 9 was obtained by reacting 9-(4-chlorophenyl)-3,9-diazabicyclo[4.2.1]nonane (29) with 7-azindole under Mannich reaction condition (Method A) (Scheme 3). The synthesis of compound 10 was previously reported ¹⁴ while compound 11 was synthesized using the commercially available 5-(4-chlorophenyl)-2,5-diazabicyclo[2.2.1]heptane (30) as the starting material. Compound 30 was treated with 7-azaindole and formaldehyde under Mannich reaction condition as before to form compound 11 (Scheme 3). The synthesis of compounds 12 and 14 were also previously reported¹⁴ and compounds 13 and 15 were synthesized using the same procedure for the synthesis of compound 11 (Scheme 3). 3-(4-chlorophenyl)-3-hydroxypyrrolidine 24, previously reported,¹⁵ (Scheme 2) the commercially available 1-(2-pyrimidyl)piperazine 31 served as starting materials for Mannich reaction mediated conversion to target compounds 13 and 15, respectively (Scheme 3).

Scheme 5.

Results and Discussion

We have previously shown that replacing the piperidine ring in haloperidol with a tropane moiety enhanced binding affinity to dopamine D₂ receptors.^15–16 Later, we compared the effect of a 4′-fluorobutyrophenone and 3-methyl-7-azaindole moieties on 4-(4-chlorophenyl)piperazine and observed that 3-methyl-7-azaindole moiety conferred both potency and D₄ selectivity.¹¹ At the time, we were drawn to the possibility that the distance between the aromatic ring and the N atom in the piperazine ring might be important. That hypothesis was tested using the potent tropane analog of haloperidol (2a) and the shorter arylalkyl groups (2b and 2c) but we failed to achieve either increased potency or D₄ selectivity.¹¹ To further explore other physicochemical aspects of the azaindole ring, we synthesized compounds 3a–c and evaluated their binding affinities for the D₂-like receptors. Compounds 3a [Ki (nM), D₂ = 588; D₄ = 7873] and 3b [Ki (nM), D₂ = 160; D₄ = 3007] are analogs of L745,870 and L741,626 with the piperazine and piperidinol moieties replaced with the tropanol moiety respectively. Surprisingly, both compounds had significantly reduced affinity for the D₂ receptor. In addition, one would have expected 3b to at least retain its binding affinity at the D₂ receptor (Ki of L741,626 at D₂ = 11.2 nM) since replacement of the piperidine ring with tropane (2a) in haloperidol enhanced potency at the D₂ subtype. Addition of a methylene group to extend the chain of 3b to 3c [Ki (nM) at D₂ = 53.0; D₄ = 277.5] improved affinity somewhat at all DA subtypes. In particular, affinity was significantly improved at the D₃ subtype with moderate selectivity when compared to the D₂ and D₄. Comparison of compounds 3a and 3b also suggests that the presence of the pyridine nitrogen in the azaindole analog detracts somewhat from binding affinity to the D₂-like receptors. Further attempts to improve binding affinity by synthesizing pyrrolidinol analogs of 3b and 3c (compounds 4 and 5, respectively) were also unsuccessful. These surprising observations led us to synthesize and evaluate the 3-methyl-7-azaindole and 4′-fluorobutyrophenone moieties on several aryl cycloalkylamine structures shown in Chart 3.

We recently reported that compound 6, a diazepane analog of haloperidol, has a favorable atypical antipsychotic profile.¹⁴ Replacement of the butyrophenone moiety with the 3-methyl-7-azaindole moiety led to the formation of compound 7. Evaluation of the binding affinities of the two compounds shows 7 has a 7- and 3-fold lower affinity for the D₂ and D₄ receptors, respectively when determined using the same assay procedure in the same laboratory (Table 2). The Ki values in parenthesis for compound 6 are obtained in a different laboratory. This observation demonstrates that the 3-methyl-7-azaindole moiety does not necessarily confer D₄ receptor potency on its own but is dependent on the amine to which it is attached. A similar observation was made when the bridged analog of compounds 6 and 7, i.e., 8 [Ki (nM); D₂ = 178.4; D₄ = 41.8) and 9 [Ki (nM); D₂ > 10,000; D₄ = 583.7] were evaluated. Compound 8 did meet our initial criterion for further evaluation. At 5-HT receptors, 8 has weak binding affinity for 5-HT_1A and 5-HT_2C receptors but a moderate affinity for 5-HT_2A receptors [Ki (nM); 5-HT_1A= 2332; 5-HT_2A = 194.8; 5-HT_2C = 3513)]. Compared to its un-bridged counterpart, compound 6, 8 has an 8-fold lower affinity for the 5-HT_2A receptor. These differences can be exploited to investigate the correlation of D₂/D₄ affinity ligands without substantial 5-HT binding affinity and the absence of extrapyramidal activity associated with typical antipsychotic agents. Compound 10 is the boat-constrained analog of 1 and 11 is an analog of 10 with the butyrophenone replaced with the 3-methyl-7-azaindole moiety. Compound 10 has moderate affinity for the D₂ and D₃ receptors and a weak affinity for the D₄ receptor. However, unlike the previous two pairs of compounds, 11 has a significantly higher affinity for all the D₂-like receptors. Indeed, compound 11 has the highest D₃ receptor affinity of all the compounds tested [Ki (nM); D₂ = 62.0; D₃ = 11.0; D₄ = 69.0] in this paper.

Table 2.

Binding Affinity Constants of Synthetic Compounds to D₂-like Receptors.

Compounds	a Binding Data of Synthetic Compounds, Ki ± SEM (n) in nM
	D2	D3	D4	D2/D4
Clozapineb	130	240	54	2.4
6b	43.3±13.3 (130)	158.8±35.1 (567)	6.6±0.6 (56)	6.6 (2.3)
7 c	970 (n = 2)	370 (219 – 631)	18.6 (14.5 – 24.0)	5.1
8b	178.4±29.2	548.1±246.0	41.8±9.0	4.3
9	>10,000	335.5±178.0	583.7±114.9	>17
10c	170.0 (123 – 234)	220.0 (148 – 339)	513.0 (447 – 589)	0.33
11 c	62.0 (38.0 – 100)	11.0 (7.94 – 15.1)	69.0 (56.2 – 85.1)	0.90
12 c	33.0 (21.9 – 50.1)	200.0 (144.5 – 275.4)	11.0 (8.9 – 12.3)	3.0
13	MPA	MPA	1213±260
14b	98.0±15.3	244.1±106.0	6.5±0.8	15
15c	1170 (n = 2)	1500 (912 – 2399)	56.0 (45.7 – 69.2)	21

^a= Data obtained from the NIMH-PDSP. Data for compounds 6 (parenthesis) 7, 10, 11, 12, and 15 were provided by A. W. Schmidt, at Pfizer laboratories as described in ref 15. Ki is the mean value obtained on triplicate or quadruplicate determinations unless otherwise indicated. MPA = Missed primary assay threshold of 50% inhibition.

^bBinding data were previously reported (ref ¹⁴).

^cThe data in brackets is the range of the Mean relative to the SEM.

We have also previously shown that replacement of the piperidine in haloperidol with the pyrrolidine ring (12) results in an analog with reduced binding affinity at the D₂ subtype.¹⁵ Separation and evaluation of the enantiomers indicated that the (+)-enantiomer is the eutomer and its behavioral profile was desirable.¹⁷ Thus, we opined that replacement of the 4′-fluorobutyrophenone moiety in 12 with the 3-methyl-7-azaindole moiety might be useful. The results indicate that there was little or no affinity for both D₂ and D₃ receptor subtypes and over 100-fold lower affinity for the D₄ receptor subtype. Synthesis and evaluation of compound 15 [Ki (nM); D₂ = 1170; D₃ = 1500; D₄ = 56.0), a 7-azaindole counterpart of the previously reported compound 14, also resulted in diminished binding affinity for the D₂-like receptors. It is important to note however that the binding affinities were determined in different laboratories and hence inter-laboratory differences (see the results for the determination of haloperidol, compound 2a and 6) may play a role as well.

Overall, these results suggest that the N-arylalkyl substituents on aryl cycloalkylamines can modify significantly the binding affinities of the resulting compounds at the dopamine receptor subtypes. There does not appear to be a specific and predictable effect of the nature of the arylalkyl moiety and the binding affinity appears to be due to a combination of effects involving the two component parts. These observations are consistent with the fact that the pharmacophoric elements of both typical and atypical antipsychotic drugs are found in both the cycloalkylamine and the arylalkyl moieties of these compounds.

Experimental

Melting points were determined on a Gallenkamp (UK) apparatus and are uncorrected. NMR spectra were obtained on a Varian 300 MHz Mercury Spectrometer. Elemental analyses were carried out by Atlantic Microlab, Inc., Norcross, GA and are within 0.4% of theory unless otherwise noted. Flash chromatography was performed with Davisil grade 634 silica gel. N,N-Dimethylformamide was distilled from CaSO₄ and stored over 4Ǻ molecular sieves. 4-Chloro-4′-fluorobutyrophenone was obtained from Sigma-Aldrich but was purified by distillation under reduced pressure to a colorless liquid prior to use. Other starting materials were used without further purification.

Preparation of 3-(4-chlorophenyl)-8-azabicyclo[3.2.1]octan-3-ol (18)

A Grignard reagent, p-chlorophenyl magnesium bromide,¹⁸ was generated in situ by reacting 4-bromochlorobenzene (5.82 g, 30.40 mmol), Mg (0.8 g, 32.90 mmol) and I₂ (ca 1 mg) in anhydrous Et₂O (20 ml), and refluxing the mixture for 4 h. A solution of N-carbethoxy tropinone (16) (2 g or 10.1 mmol) in anhydrous THF (10 mL) was slowly added to the reaction mixture and further refluxing continued for 18 hours. The resulting mixture was allowed to cool to room temperature, saturated NH₄Cl solution (50 mL) was added, and the mixture was extracted with EtOAc (3×50 mL). The combined organic phase was washed with H₂O (50 mL) followed by brine (50 mL), dried over Na₂SO₄, filtered, and the filtrate was concentrated in vacuo. Column chromatography (gradient solvent of 8:2 to 7:3 hexane/EtAOc) on silica gel afforded ethyl-3-(4-chlorophenyl)-3-hydroxy-8-azabicyclo[3.2.1]octane-8-carboxylate (17) as a yellowish oil which solidified on standing at room temperature for one day, 1.3 g, 41.5 %. ¹HNMR (300 MHz, CDCl₃): δ 1.26 (3H, t, J = 7.1 Hz), 1.55 (2H, d, J = 9.2 Hz), 1.81 (2H, d, J = 14.5 Hz), 1.96 (2H, m), 2.27 (4H, d, J = 6.8 Hz), 4.15 (2H, q, J = 7.1, J = 7.3 Hz), 7.28 (4H, d, J = 4.0 Hz). A mixture of KOH (3.2 g, 56.5 mmol) in ethylene glycol (20 mL) was added to a solution of 17 (2.5 g or 8.1 mmol) in MeOH (10 mL) and the resulting mixture was heated at 150 °C, with constant stirring, for 4 hours and then allowed to cool to room temperature. Water (200 mL) was added, and the mixture was extracted with EtOAc (2×100 mL) followed by CH₂Cl₂ (3×100 mL). The organic phases were combined, washed with H₂O (400 mL), brine (100 ml) and dried (Na₂SO₄). The organic phase was filtered, and the filtrate was concentrated in vacuo and the residue was column chromatographed on silica gel (4:2 CH₂Cl₂/MeOH) to give white yellowish crystals of 3-(4-chlorophenyl)-8-azabicyclo[3.2.1]octan-3-ol (18) (1.40 g, 73%). ¹HNMR (300 MHz, CDCl₃): δ 1.82 (4H, m), 2.17 (2H, dd, J = 3.7, J = 3.7 Hz), 2.34 (2H d, J = 7.5 Hz), 3.56 (2H, brs), 7.27 (2H, d, J = 8.7 Hz), 7.48 (2H, d, J = 8.7 Hz).

4-[3-(4-chlorophenyl)-3-hydroxy-8-azabicyclo[3.2.1]oct-8-yl]-1-(4-fluorophenyl)butan-1-one (2a)

A mixture of 18 (0.46 g, 1.94 mmol), 4-chloro-4′-fluorobutyrophenone (0.60 g, 3 mmol), KI (0.5 g, 3 mmol), K₂CO₃ (0.70 g, 5.1 mmol) in DME (10 mL) was refluxed under N₂ for 18 h. The reaction mixture was allowed to cool to room temperature, H₂O (20 mL) was added, and the mixture was extracted with EtOAc (4 × 50 mL). The combined organic phase was washed with brine, dried over Na₂SO₄ and concentrated in vacuo. Column chromatography was carried out using CH₂Cl₂ followed by 2:3 EtOAc/MeOH to afford a yellowish oil of 2a (0.18 g, 23%) which was then converted to its HCl salt, mp 237.2 –238.3 °C. ¹HNMR (300 MHz, CD₃OD): δ 2.24 (m, 6H), 2.57 (m, 2H,) 2.72 (d, J = 9.0 Hz, 2H), 3.17 (m, 2H), 3.27 (t, J = 6.6 Hz, 2H), 4.19 (brs, 2H), 7.26 (dd, J_{H- F} = 8.7 Hz, J_H-H = 8.8 Hz, 2H), 7.37 (d, J = 8.7 Hz, 2H), 7.56 (d, J = 8.7, 2H), 8.11 (dd, J_H-F = 5.4 Hz, J_H-H = 8.8 Hz, 2H). Anal. (C₂₃H₂₆Cl₂FN₂O.0.5H₂O): C 61.75, H 6.08, N 3.13. Found: C 61.93, H 6.03, N 3.20.

2-[3-(4-chlorophenyl)-3-hydroxy-8-azabicyclo[3.2.1]oct-8-yl]-1-(4-fluorophenyl)ethanone (2b)

A mixture of 18 (0.700 g, 2.94 mmol), 2-chloro-4′-fluoroacetophenone (0.83 g, 3.83 mmol), KI (0.800 g, 2.94 mmol), and K₂CO₃ (2.44 g, 17.6 mmol) in DME (10 mL) was refluxed for 18 hours. The reaction was cooled to rt., H₂O (20 mL) was added, and the mixture was extracted with EtAOc (3×50 mL). The combined organic phase was washed with brine (50 mL), dried over Na₂SO_4, filtered, and the filtrate was concentrated in vacuo. Column chromatography (7:3 hexane/EtAOc) resulted in a yellowish oil of 2b (0.480 g, 44 %) which was converted into an HCl salt; mp 256.8– 257.5 °C. ¹HNMR (300 MHz, CD₃OD): δ 2.28 (7H, m), 2.71 (4H, m), 4.19 (2H, brs), 7.34 (2H, t, J_{H- F} = 8.8, J_H-H = 8.8 Hz), 7.38 (2H, d, J = 8.8, J_H-H = 8.8 Hz) 7.59 (2H, d, J = 8.8), 8.18 (2H, dd, J_H-F = 5.5, J_H-H = 8.8 Hz). Calcd for C₂₁H₂₂Cl₂FNO₂·0.2H₂O: C 60.94, H 5.45, N 3.38. Found: C 60.96, H 5.34, N 3.40.

3-(4-chlorophenyl)-8-(4-fluorobenzyl)-8-azabicyclo[3.2.1]octan-3-ol (2c)

A mixture of18 (0.50 g, 2.10 mmol), 4-fluorobenzyl bromide (0.517 g, 2.70 mmol), KI (0.350 g, 2.10 mmol), and K₂CO₃ (1.74 g, 12.6 mmol) was refluxed in DME (6 ml) for 18 h. After cooling to room temperature, H₂O (20 mL) was added and the mixture was extracted with EtAOc (4×50 ml). Organic phases were combined, washed with brine (30 mL), dried over Na₂SO_4, filtered, and the filtrate was concentrated in vacuo. Column chromatography on silica gel (starting with 100 % hexane and then 7:3 hexane/EtAOc) yielded a yellowish oil of 2c (0.425 g, 59 %), which was then converted into an HCl salt, mp 110.4–111.1 °C. ¹HNMR (300 MHz, CDCl₃): δ 2.09 (2H, d, J = 14 Hz), 2.18 (2H, m), 2.80 (2H, m), 3.10 (2H, d, J = 14 Hz), 3.70 (2H, brs), 4.07 (2H, d, J = 6.0 Hz), 7.12 (2H, t, J_H-F = 8.5, J_H-H = 8.5 Hz), 7.26 (2H, d, J = 8.5), 7.78 (2H, d, J = 8.6 Hz), 7.87 (2H, dd, J_H-F = 5.3, J_H-H = 8.5 Hz). Calcd for C₂₀H₂₂Cl₂FNO·0.25H₂O: C 62.10, H 5.86, N 3.62. Found: C 62.05, H 6.01, N 3.60.

3-(4-chlorophenyl)-8-(1H-pyrrolo[2,3-b]pyridin-3-ylmethyl)-8-azabicyclo[3.2.1]octan-3-ol (3a)

Method A: A mixture of 18 (0.60 g, 2.50 mmol), 7-azaindole (0.400 g, 3.40 mmol), AcOH (6 drops, 17 M), and CH₂O (0.203 g, 2.50 mmol) in CH₂Cl₂ (5 mL) was stirred at room temperature for 18 hrs. The reaction mixture was basified with NaOH (10 % aqueous solution) and extracted with CH₂Cl₂ (4×25 mL). The combined organic phase was washed with brine (20 mL), dried over Na₂SO₄, filtered, and the filtrate was concentrated in vacuo. Purification by preparatory TLC (4:2 CH₂Cl₂/MeOH) yielded flaky white crystals of 3a (0.46 g, 50 %), mp 94.2–94.5 °C. 1HNMR (300 MHz, CD₃OD): δ 1.84 (2H, d, J = 14 Hz), 2.30 (6H, m), 3.45 (2H, brs), 3.84 (2H, brs), 7.14 (1H, dd, J = 5.0, J = 7.8 Hz), 7.27 (2H, d, J = 8.6 Hz), 7.46 (1H, s) 7.50 (2H, d, J = 8.6 Hz), 8.20 (2H, m). Calcd for C₂₁H₂₂ClN₃O·0.75H₂O: C 66.13, H 6.21, N 11.02; Found: C 66.12, H 6.04, N 10.87.

3-(4-chlorophenyl)-8-(1H-indol-3-ylmethyl)-8-azabicyclo[3.2.1]octan-3-ol (3b)

A mixture of 18 (0.50 g, 2.10 mmol), indole (0.250 g, 2.1 mmol), AcOH (6 drops, 17 M), and CH₂O (0.065 g, 2.10 mmol) in CH₂Cl₂ (5 mL) was stirred at room temperature for 18 h. The reaction mixture was basified (10 % aq. NaOH solution), extracted with CH₂Cl₂ (4×25 mL), the pooled organic layers were combined, washed with brine (20 mL), dried over Na₂SO₄ and the filtrate was concentrated in vacuo. Purification by preparatory TLC (4.5:0.5 CH₂Cl₂/2 M NH₃ in MeOH) yielded yellowish crystals of 3b (0.39 g, 51 %). Mp 72.4–73.1 C, ¹HNMR (300 MHz, CD₃OD): δ 1.83 (2H, d, J = 14 Hz), 2.26 (5H, m), 2.37 (2H, d, J = 6.1 Hz), 2.40 (1H, d, J = 7.2 Hz), 3.35 (1H, s), 3.84 (2H d, J = 11.0 Hz), 7.04 (1H, m), 7.11 (1H, m), 7.26 (2H, d, J = 8.6 Hz), 7.36 (1H, m), 7.52 (2H, d, J = 8.6 Hz), 7.70 (1H, m). Calcd for C₂₂H₂₃ClN₂O.1.0H₂O: C 68.65, H 6.55, N 7.28. Found: C 68.43, H 6.29, N 7.04.

3-(4-chlorophenyl)-8-(1H-pyrrolo[2,3-b]pyridin-3-ylethyl)-8-azabicyclo[3.2.1]octan-3-ol (3c)

A mixture of 18 (0.200 g, 0.84 mmol), 3-(2-bromoethyl)indole (0.094 g, 0.42 mmol), and NaHCO₃ (0.14 g, 1.68 mmol) in anhydrous CH₃CN (5 mL) was refluxed for 4 h under N₂. The reaction mixture was allowed to cool to rt and H₂O (15 mL) was added. The mixture was extracted with CH₂Cl₂ (4×25 ml) and the combined organic layers was washed with brine (20 mL), dried over Na₂SO₄ and the filtrate was concentrated in vacuo. White crystals of 3c (0.18 g, quantitative) were obtained after preparatory TLC purification (4:1 CH₂Cl₂/MeOH), mp 186.5–187.1 °C. 1HNMR (300 MHz, CD₃OD): δ 1.83 (2H, d, J = 14 Hz), 2.00 (2H, m), 2.32 (5H, m), 2.82 (2H, m), 3.03 (2H, m), 3.50 (2H, brs), 7.08 (1H, s), 7.10 (1H, dt, J = 1.2, J = 8.0 Hz), 7.28 (2H, d, J = 8.7 Hz), 7.32 (2H, m), 7.54 (2H, d, J = 8.7Hz), 7.56 (2H, m). Calcd for C₂₃H₂₅ClN₂O·0.75H₂O: C 70.04, H 6.77, N 7.10. Found: C 69.86, H 6.56, N 7.00.

3-(4-chlorophenyl)-1-(1H-indol-3-ylmethyl)pyrrolidin-3-ol (4)

A solution of 3-(4- chlorophenyl)pyrrolidin-3-ol (0.60 g, 3.04 mmol), indole (0.46 g, 3.95 mmol), AcOH (6 drops, 17 M), and CH₂O (0.08 ml, 3.04 mmol) in CH₂Cl₂ (5 mL) was stirred at room temperature for 18 h. The mixture was basified (10 % NaOH solution), extracted with CH₂Cl₂ (4×25 mL), the combined organic layers was washed with brine (20 mL), dried over Na₂SO₄ and the filtrate was concentrated in vacuo. Purification using preparatory TLC (4.5:0.5: CH₂Cl₂/2M NH₃ in MeOH) gave yellowish crystals of 4 (0.49 g, 49 %), mp 56.9–57.2 °C, ¹HNMR (300 MHz, CD₃OD): δ 2.13 (1H, m), 2.25 (1H, m), 2.94 (1H, m), 2.98 (2H, m), 3.08 (1H, m), 3.96 (2H, s), 7.03 (1H, m), 7.10 (1H, m), 7.24 (1H, s), 7.28 (2 H, d, J = 8.7 Hz), 7.35 (1H, m), 7.45 (2H, d, J = 8.7 Hz), 7.66 (1H, m). Calcd for C₁₉H₁₉ClN₂O·0.4H₂O: C 68.32, H 5.97, N 8.39. Found: C 68.40, H 5.86, N 8.28.

3-(4-Chlorophenyl)-1-[2-(1H-indol-3-yl)ethyl]pyrrolidin-3-ol (5)

A mixture of 3-(4- chlorophenyl)pyrrolidin-3-ol (0.40 g, 2.02 mmol), 3-(2-bromoethyl)indole (0.23 g, 1.01 mmol), and NaHCO₃ (0.68 g, 8.08 mmol) in anhydrous CH₃CN (5 mL) was refluxed for 4 h under N₂. The reaction mixture was allowed to cool to room temperature and H₂O (15 mL) was added. The mixture was extracted with CH₂Cl₂ (4×25 mL), the pooled organic layers was washed with brine (20 mL), dried over Na₂SO₄ and the filtrate was concentrated in vacuo. Purification by column chromatography (4:1 CH₂Cl₂/MeOH) afforded white crystals of 5 (0.15 g, 22 %), mp 54.1–55.7 °C. ¹HNMR (300 MHz, CDCl₃): δ 2.15 (1H, m), 2.28 (1H, m), 2.60 (2H, m), 2.92 (4H, m), 3.06 (1H, d, J = 9.7 Hz), 3.25 (1H, m), 6.98 (1H, s), 7.09 (2H, m), 7.26 (2H, m), 7.38 (2H, d, J = 8.6 Hz), 7.55 (1H, d, J = 7.8 Hz), 7.93 (1H, brs). Calcd for C₂₀H₂₁ClN₂O·0.5MeOH: C 68.99, H 6.50, N 7.85. Found: C 68.80, H 6.27, N 7.49.

3-{[4-(4-chlorophenyl)-1,4-diazepan-1-yl]methyl}-1H-pyrrolo[2,3-b]pyridine, HCl (7)

Using Method A and 1-(4-chlorophenyl)-1,4-diazepane and 7-azaindole as starting materials, compound 7 was obtained as the HCl salt, mp 158.0 –159.0 °C. ¹H NMR (300 MHz, CDCl₃): δ 9.30 (1H, brs), 8.28 (1H, d, J = 4.5 Hz), 7.96 (1H, dd, J = 1.5, 7.5 Hz), 7.19 (1H, s), 7.12 (2H, d, J = 9.0 Hz), 7.03(1H, dd, J = 4.8, 8.1 Hz), 6.58 (2H, d, J = 9.0 Hz), 3.79 (2H, s), 3.51 (2H, t, J = 4.8 Hz), 3.47 (2H, t, J = 6.3 Hz), 2.78 (2H, t, J = 4.8 Hz), 2.65 (2H, t, J = 4.8 Hz), 1.90 (2H, m). Calcd for C₁₉H₂₁ClN₄·HCl·0.35 H₂O: C 59.49, H 5.78, N 14.60; Found: C 59.40, H 5.54, N 14.59.

3-[(1H-pyrrolo[2,3-b]pyridin-3-yl)methyl]-9-(4-chlorophenyl)-3,9-diazabicyclo[4.2.1]nonane (9)

The method of Michaels and Zaugg was followed.¹⁹ Ethyl chloroformate (4 mL) was added in a drop-wise manner to compound 25¹⁴ (2.78 g, 19.8 mmol) dissolved in dry CH₂Cl₂ (50 mL) in the presence of molecular sieves (1 g). The reaction mixture was stirred for 16 h at rt, the solvent was evaporated and the resulting crude product subjected to column chromatography on silica gel (4:2 CH₂Cl₂:MeOH) to give 3,9-diazabicyclo[4.2.1]nonane-3-carboxylic acid ethyl ester, compound 26 as an oil; ¹H NMR (300MHz, CDCl₃): δ 1.23 (t, 3H, J = 6.9 Hz), 1.78–2.00 (m, 3H), 2.15–2.29 (m, 2H), 2.3–2.58 (m, 2H), 2.87 (s, 3H), 3.65–4.24 (m, 7H). Compound 26 (1.9 g, 8.95 mmol) was dissolved in ClCH₂CH₂Cl (10 mL) and stirred well at 0 °C. At 0 °C, α-chloroethyl chloroformate (1.9 g, 13.5 mmol) was added in a drop wise manner and the reaction was allowed to cool to room temperature. The reaction mixture was then refluxed for 3 h, MeOH (10 mL) was added and refluxing continued for an additional 1 h, solvent was evaporated and the crude reaction mixture was subjected to column chromatography (silica gel, 9:1 CH₂Cl₂:MeOH) to yield 3,9-diazabicyclo[4.2.1]nonane-3-carboxylic acid ethyl ester, 27 (0.55 g, 31 %). Without characterization, a mixture of compound 27 (0.5g, 2.5 mmol), 4-chlorophenylboronic acid (0.78g, 5 mmol), copper acetate (0.82 g, 4.56 mmol), Et₃N (0.5 mL) in CH₂Cl₂ (50 mL) with molecular sieves (0.8 g) was stirred in open air for 48 h. The reaction mixture was monitored by TLC until the formation of the product was observed before quenching with methanolic NH₃ solution. The resulting mixture was filtered over celite, extracted with CH₂Cl₂ (3 × 35 ml) dried and solvent evaporated under vacuum. The crude reaction mixture was subjected to column chromatography (silica gel, 4:1 CH₂Cl₂:MeOH ) to yield a yellowish syrup of 9-(4-chlorophenyl)-3,9-diazabicyclo[4.2.1]nonane-3-carboxylic acid ethyl ester, 28 (0.21g, 27 %). Compound 28 (0.2 g, 0.65 mmol), KOH (0.2 g in 0.2 ml H₂O) in ethylene glycol (0.2 mL) was heated at 90 °C overnight. The reaction mixture was extracted with CH₂Cl₂ (2 × 80 mL), dried over Na₂SO₄ and evaporated under vacuum. The crude product was subjected to column chromatography (silica gel, 4:2 CH₂Cl₂:MeOH) to yield compound 29 (0.09 g, 58 %). A mixture of 9-(4-chlorophenyl)-3,9-diazabicyclo[4.2.1]nonane, 29 (0.130g, 0.55 mmol), 7- azaindole (0.130g, 1.1mmol), AcOH (6 drops, 17M) CH₂O (0.016g, 0.53mmol) in butanol (8ml) was refluxed overnight under N₂. The excess butanol was removed in vacuo and residue was extracted with methylene chloride (3×60ml). The organic phase was dried (Na₂SO₄) and removed. The crude product was purified on silica gel column with 1:1 CH₂Cl₂ and EtOAc to yield compound 9 (80 mg, 62.2%) as a white hygroscopic crystalline solid. MP: 194–196 °C. ¹H NMR (300MHz, CDCl₃): δ 8.30 (m, 1H), 8.05 (m, 1H), 7.04 (m, 4H), 6.56 (d, J=6.5Hz, 2H), 3.58 (d, J=6.5Hz, 2H), 3.50 (m, 6H), 3.12 (m, 1H), 2.20-1.10 (m, 6H). Anal. Calcd. for C₂₁H₂₃N₄Cl·0.125H₂O : C, 68.33; H, 6.35; N, 15.18 Found C, 68.17; H, 6.30, N, 14.92.

2-{(1H-pyrrolo[2,3-b]pyridin-3-yl)methyl}-5-(4-chlorophenyl)-2,5-diazabicyclo[2.2.1]heptane (11)

Using method A and 5-(4-chlorophenyl)-2,5-diazabicyclo[2.2.1]heptane and 7-azaindole as starting materials, compound 11 was obtained as the free base, mp 176–178 °C. ¹H NMR (CDCl₃): δ 10.03 (1H, brs), 8.29 (1H, d, J = 4.5 Hz), 8.01 (1H, d, J = 8.1 Hz), 7.22 (1H, s), 7.13 (2H, d, J = 9.0 Hz), 7.05 (1H, dd, J = 4.8, 7.8 Hz), 6.47 (2H, d, J = 9.0 Hz), 4.18 (1H, s), 3.85 (2H, s), 3.60 (1H, s), 3.36 (1H, d, J = 9.0 Hz), 3.29 (1H, d, J = 9.0 Hz), 2.93 (1H, d, J = 9.6 Hz), 2.73 (1H, d, J = 9.6 Hz), 2.00 (1H, d, J = 9.3 Hz), 1.86 (1H, d, J = 9.3 Hz). Calcd for C₁₉H₁₉ClN₄: C 67.35, H 5.65, N 16.54; Found: C 67.10, H 5.74, N 16.25.

3-(4-chlorophenyl)-1-(1H-pyrrolo[2,3-b]pyridin-3-ylmethyl)pyrrolidin-3-ol (13)

Using Method A and 3-(4-chlorophenyl)pyrrolidin-3-ol, 24 and 7-azaindole as starting materials, compound 13 was obtained in 37 % yield, mp 62.9–63.2 °C. ¹HNMR (300 MHz, CD₃OD): δ 2.15 (1H, m), 2.90 (1H, m), 2.95 (2H, m), 3.05 (1H, m), 3.93 (2H, s), 7.12 (1H, dd, J = 5.0, J = 7.8 Hz), 7.29 (2H, d, J = 8.6 Hz), 7.39 (1H, s), 7.47 (2H, d, J = 8.6 Hz), 8.16 (2H, m). Calcd for C₁₈H₁₈ClN₃O·0.4H₂O: C 64.53, H 5.66, N 12.54. Found: C 64.48, H 5.57, N 12.39.

3-{(4-(pyrimidin-2-yl)piperazin-1-yl)methyl}-1H-pyrrolo[2,3-b]pyridine (15)

Using method A and 1-(2-pyrimidyl)piperazine and 7-azaindole, produced compound 15 as a solid, mp 185–187 °C. ¹H NMR (CDCl₃): 9.00 (1H, brs), 8.31 (1H, dd, J = 1.5, 5.1 Hz), 8.28 (2H, d, J = 4.8 Hz), 8.09 (1H, dd, J = 1.5, 7.8 Hz), 7.24 (1H, s), 7.08 (1H, dd, J = 4.8, 8.4 Hz), 6.45 (1H, t, J = 4.8 Hz), 3.83 (4H, t, J = 5.4Hz), 3.73 (2H, s), 2.54 (4H, t, J = 5.4 Hz). Calcd for C₁₆H₁₈N₆: C 65.29, H 6.16, N 28.55; Found: C 65.02, H 6.20, N 28.26.

Biology

Receptor Binding Studies

Binding affinities reported in Tables 1 – 3 were conducted by the National Institute of Mental Health Psychoactive Drug Screening Program (NIMH-PDSP) unless otherwise stated. Details of the methods and radioligands used for the binding assays were previously reported.²⁰

Table 1.

Binding Affinity Constants of Synthetic Compounds to D₂-like Receptors.

Compounds	a Binding Data of Compounds, Ki ± SEM (nM)
	D2	D3	D4	D2/D4
Haloperidol	1.1±0.07 [ 0.89]	5.5±3.0[ 4.6]	12.7±7.2 [10.0]	0.10 [0.09]
2a	1.6 ±0.14 [0.31]	5.1±3.0 [0.71]	5.3± 0.99 [12.1]	0.30 [0.03]
2b	1231±145	>10,000	789±363	1.60
2c	1050±209	172±33	1015±179	1.03
3a	588±57.6	128±13	7873±1437	0.07
3b	160.3±11.8	25.0±1.7	3007±561	0.05
3c	53.0±6.4	18.0±1.6	277.5±26.5	0.19
4	MPA	2874±584	816.5±194.4	---
5	MPA	3074±553	MPA	---

^a= Data obtained from the NIMH-PDSP and those in square brackets are from ref 15. Ki is the mean value obtained on triplicate or quadruplicate determinations unless otherwise indicated. MPA = Missed primary assay threshold of 50% inhibition.