N-(4-Arylpiperazinoalkyl)acetamide derivatives of 1,3- and 3,7-dimethyl-1H-purine-2,6(3H,7H)-diones and their 5-HT₆, 5-HT₇, and D₂ receptors affinity

Introduction

Compounds with affinity for serotoninergic (5-HT₆, 5-HT₇) and dopaminergic type 2 receptors (D₂Rs) exhibit potential for the treatment of diverse central nervous system disorders, e.g., depression [1, 2], schizophrenia [3], as well as dementia [4] and Alzheimer disease [5]. Importantly, derivatives with mixed receptor binding profile, often called serotonin/dopamine modulators, may attenuate both negative, positive symptoms of schizophrenia, as well as improve cognitive deficits [6, 7].

A vast group of compounds with mixed serotonin/dopamine receptor binding profile belongs to a class of long-chain arylpiperazines (LCAPs). Typically, in such derivatives, the arylpiperazine moiety is connected via an alkyl linker to the terminal fragment. For several years, we have been interested in developing of LCAPs with different xanthine moieties playing a role of a cyclic amide core in the terminal fragment. In our previous paper, we have described a series of derivatives possessing 8-alkoxy-1,3-dimethyl-1H-purine-2,6(3H,7H)-dione core, which behaved as 5-HT_1A receptors agonists [8].

The aim of our study was to investigate structure-affinity relationships for the 5-HT₆, 5-HT₇, and D₂ receptors in a group of LCAPs with xanthine moieties in the terminal fragment and an additional amide bond in the linker. For this purpose, we have designed a series of new N-(arylpiperazinylalkyl)acetamide derivatives of 1,3- and 3,7-dimethyl-1H-purine-2,6(3H,7H)-dione with a different length polymethylene spacer between the amide moiety and the basic nitrogen atom. Structural modifications in the amine fragment comprised introduction of 1-(2,3-dichlorophenyl)piperazine and 1-(benzo[d]isothiazol-3-yl)piperazine, as substructures related to aripiprazole and ziprasidone (Figure 1), respectively. To further examine an impact of replacement of sulfur atom with oxygen atom on affinity, the analogues with 1-(benzo[d]isoxazol-3-yl)piperazine moiety were also synthesized.

Figure 1

Aripiprazole, ziprasidone, and a general structure of 8-alkoxy-1,3-dimethyl-1H-purine-2,6(3H,7H)-dione derivatives [8].

Results and discussion

The designed compounds were synthesized in reaction of appropriate xanthinylacetic acids, obtained according to the methods previously reported by Maślankiewicz et al. [9], and arylpiperazinoalkylamines, using 1,1′-carbonyldiimidazole (CDI) as activating agent (Schemes 1 and 2).

Scheme 1

Scheme 2

In the first step, 1,3- or 3,7-dimethyl-1H-purine-2,6(3H,7H)-dione was alkylated by treatment with ethyl 2-chloroacetate in the presence of anhydrous K₂CO₃ and TEBA (N,N,N-triethylbenzylammonium chloride) in acetone, yielding ethyl 2-(1,3-dimethyl-2,6-dioxo-2,3-dihydro-1H-purin-7(6H)-yl)acetate [9] and 2-(3,7-dimethyl-2,6-dioxo-2,3,6,7-tetrahydro-1H-purin-1-yl)acetate [9], respectively. The esters were subsequently hydrolyzed with 10% solution of NaOH in a water-acetone mixture (2:1), and the resulting acids were isolated after acidification of the reaction mixture. The final products were obtained by acylation of the appropriate arylpiperazino-(4-(benzo[d]isothiazol-3-yl)piperazino-, benzo[d]isoxazol-3-yl)piperazinyl-, or 4-(2,3-dichlorophenyl)piperazino-) derivative of alkylamine with 2-(1,3-dimethyl-2,6-dioxo-2,3-dihydro-1H-purin-7(6H)-yl)acetic acid (I) [9] or 2-(3,7-dimethyl-2,6-dioxo-2,3,6,7-tetrahydro-1H-purin-1-yl)acetic acid (II) [9] at room temperature in DMF, using CDI as activating agent. The structures of the newly synthesized compounds 1–15 were confirmed by analysis of ¹H NMR and ¹³C NMR spectra, LC/MS, and elemental analysis. The investigated compounds were pharmacologically tested as free bases.

The newly synthesized N-(arylpiperazinylalkyl)acetamide derivatives of 1,3- and 3,7-dimethyl-1H-purine-2,6(3H,7H)-dione were tested in competition binding experiments for 5-HT₆, 5-HT₇, and D₂ receptors. Compounds displayed high to low affinity for the tested receptors, ranging from 14 nm to 100 μm for 5-HT₆Rs, from 100 nm to 4789 nm for 5-HT₇Rs, and 1–2090 nm for D₂Rs (Tables 1 and 2).

The affinity for 5-HT₆ and D₂ receptors depended on the type of the xanthine core: 1,3-dimethyl-1H-purine-2,6(3H,7H)-dione derivatives displayed higher affinity for 5-HT₆ and D₂ receptors than their 3,7-dimethyl-1H-purine-2,6(3H,7H)-dione counterparts, e.g., compound 8 vs. compound 15 (Ki(D2) = 85 and 1 nm, respectively). Interestingly, a type of the xanthine core did not significantly influence affinity for 5-HT₇Rs.

The impact of the length of the polymethylene spacer between the amide moiety and the basic nitrogen atom was evident: compounds with trimethylene spacer displayed the highest affinity for 5-HT₆ and D₂ receptors. Compounds 4 and 5 displayed over 9- and 59-fold higher affinity for 5-HT₆ receptors than their lower homologues, compounds 1 and 3, respectively, and over 5- and 18-fold higher affinity than their higher homologues, compounds 6 and 8. Although a dimethylene group spacer was the least preferential for 5-HT₇Rs, trimethylene, and tetramethylene analogues display comparable affinity for these sites.

Further analysis of the affinity data suggested that replacement of carbostyryl fragment in aripiprazole with xanthinylacetamide moiety increased affinity of 2,3-dichlorophenylpiperazines for 5-HT₆Rs (compounds 5 and 12 vs. aripiprazole). At the same time, this modification significantly decreased affinity for 5-HT₇Rs. As a consequence, compounds 5 and 12 were classified as potent 5-HT₆/D₂ receptor ligands. The same modification of ziprasidone structure significantly increased the affinity for 5-HT₆ and 5-HT₇ receptors, while this effect was much smaller in case of D₂ receptors (4 vs. ziprasidone).

Structure-activity relationship studies in the arylpiperazine function suggested that highly lipophilic 1-(2,3-dichlorophenyl)piperazine moiety guaranteed high affinity for 5-HT₆ and D₂Rs. Compounds with 1-(benzo[d]isothiazol-3-yl)piperazine moiety displayed lower affinity for these receptors. Further substitution of the sulfur atom in 1-(benzo[d]isothiazol-3-yl)piperazine with oxygen atom, which decreased the lipophilicity of the fragment, also caused further loss of affinity. This can be easily observed with direct structural analogues 15 vs. 13 vs. 14, containing 4-(2,3-dichlorophenyl)piperazino, 4-(benzo[d]isothiazol-3-yl)piperazino, and 4-(benzo[d]isoxazol-3-yl)piperazino moieties, respectively. Simultaneously, an influence of arylpiperazine moiety on the affinity for 5-HT₇ receptors suggested that generally less lipophilic substituents were preferred, but this influence was much weaker than for 5-HT₆ and D₂ receptors, e.g., compound 15 had comparable affinity for 5-HT₇Rs to those of compounds 13 and 14.

Conclusions

We have designed and synthesized a series of 15 new N-(arylpiperazinylalkyl)acetamide derivatives of 1,3- and 3,7-dimethyl-1H-purine-2,6(3H,7H)-diones. The study allowed for identification of potent 5-HT₆/D₂ receptor ligands (compounds 5 and 12) and D₂ receptor agents (compounds 4, 8, and 15). Structure-affinity relationships studies showed that the 3,7-dimethyl-1H-purine-2,6(3H,7H)-dione core, a polymethylene spacer containing three methylene groups, and 2,3-dichlorophenypiperazine were the optimal elements for high affinity for 5-HT₆ and D₂ receptors.

Experimental

Melting points (mp) were determined with a Büchi Melting Point B-545 apparatus and are uncorrected. ¹H NMR and ¹³C NMR spectra were taken with a Varian Mercury-VX (300 MHz) spectrometer in DMSO-d₆ solutions, using signals of solvent’s residual ¹H and ¹³C atoms as internal standards (δ = 2.49 and 39.7 ppm, respectively). LC/MS analyses were performed on Waters Acquity TQD apparatus with eλ DAD detector. For mass spectrometry, ESI+ (electrospray positive) ionization mode was used. The progress of the reactions and the purity of compounds were routinely checked by TLC using Merck Kieselgel 60 F₂₅₄ sheets and eluting with dichloromethane/methanol, 90:10. Spots were detected by UV irradiation. All final compounds had purity over 95%. Elemental analyses were taken with Elementar Vario EL III apparatus. All in vitro radioligand binding assays were carried out using methods published by Zajdel et al. [10].