4-Amino-6-benzylamino-1,2-dihydro-2-phenyl-1,2,4-triazolo[4, 3-a]quinoxalin-1-one (1) has been found to be an A(2A) versus A(1) selective antagonist (Colotta et al. Arch. Pharm. Pharm. Med. Chem. 1999, 332, 39-41). In this paper some novel triazoloquinoxalin-1-ones 4-25 bearing different substituents on the 2-phenyl and/or 4-amino moiety of the parent 4-amino-1, 2-dihydro-2-phenyl-1,2,4-triazolo[4,3-a]quinoxalin-1-one (3) have been synthesized and tested in radioligand binding assays at bovine A(1) and A(2A) and cloned human A(3) adenosine receptors (AR). Moreover, the binding activities at the above-mentioned AR subtypes of the 1,4-dione parent compounds 26-31 and their 5-N-alkyl derivatives 33-37 were also evaluated. The substituent on the 2-phenyl ring exerted a different effect on AR subtypes, while replacement of a hydrogen atom of the 4-amino group with suitable substituents yielded selective A(1) or A(3) antagonists. Replacement of a hydrogen atom of the 4-NH(2) with an acyl group, or replacement of the whole 4-NH(2) with a 4-oxo moiety, shifted the binding activity toward the A(3) AR. The binding results allowed elucidation of the structural requirements for the binding of these novel tricyclic derivatives at each receptor subtype. In particular, A(1) and A(2A) binding required the presence of a proton donor group at position-4, while for A(3) affinity the presence of a proton acceptor in this same region was of paramount importance.

A molecular simplification approach of previously reported 2-arylpyrazolo[3,4-c]quinolin-4-ones was applied to design 2-arylpyrazolo[4,3-d]pyrimidin-7-one derivatives as new human A(3) adenosine receptor antagonists. Substituents with different lipophilicity and steric hindrance were introduced at the 5-position of the bicyclic scaffold (R(5) = H, Me, Et, Ph, CH(2)Ph) and on the 2-phenyl ring (OMe, Me). Most of the synthesized derivatives were highly potent hA(3) adenosine receptor antagonists, the best being the 2-(4-methoxyphenyl)pyrazolo[4,3-d]pyrimidin-7-one (K(i) = 1.2 nM). The new compounds were also highly selective, being completely devoid of affinity toward hA(1), hA(2A), and hA(2B) adenosine receptors. On the basis of the recently published human A(2A) receptor crystallographic information, we propose a novel receptor-driven hypothesis to explain both A(3) AR affinity and A(3) versus A(2A) selectivity profiles of these new antagonists.

In the past few years much effort in our laboratory has been directed toward the study of adenosine receptor antagonists, and recently we focused our attention on 2-aryl-1,2,4-triazolo[4,3-a]quinoxaline-1,4-diones and 2-aryl-1,2,4-triazolo[4,3-a]quinoxalin-4-amino-1-ones, some of which were potent and/or selective A(3) receptor antagonists. In the present paper, a new series of triazoloquinoxaline derivatives is described. Most of the new compounds, biologically evaluated in radioligand binding assays at bovine (b) A(1) and A(2A) and at human (h) A(1) and A(3) adenosine receptors, showed high hA(3) adenosine receptor affinity and selectivity. In particular, 2-(4-nitrophenyl)-1,2,4,5-tetrahydro-1,2,4-triazolo[4,3-a]quinoxaline-1,4-dione (1), also tested at the hA(2A) ARs, shows the best binding profile with a high hA(3) affinity (K(i) = 0.60 nM) and strong selectivity vs hA(1) and vs hA(2A) receptors (both selectivity ratios greater than 16 600). To interpret our experimental results, we decided to theoretically depict the putative transmembrane binding motif of our triazoloquinoxaline analogues on hA(3) receptor. Structure-activity relationships have been explained analyzing the three-dimensional structure of the antagonist-receptor models obtained by molecular docking simulation.

In a recent paper (Colotta et al. J. Med. Chem. 2000, 43, 1158-1164 we reported the synthesis and adenosine receptor binding activity of two sets of 2-aryl-1,2,4-triazolo[4,3-a]quinoxalines (A and B) some of which were potent and selective A 1 or A 3 antagonists. In this paper the synthesis of a set of 2-arylpyrazolo[3,4-c]quinolin-4-ones 1-10, 4-amines 11-18, and 4-aminosubstituted derivatives 19-35 are reported. The binding activity at bovine A 1 and A 2A and human cloned A 3 adenosine receptors showed that (i) the substituent on the appended 2-phenyl ring could be used to modulate A 1 and A 3 affinity, (ii) the 4-amino group was necessary for A 1 and A 2A binding activity, and (iii) a nuclear or extranuclear CdO proton acceptor at position 4 yielded potent and selective A 3 antagonists. These results are in agreement with those of the previously reported series A and B suggesting a similar adenosine receptor binding mode. In particular, the A 3 nanomolar affinity of 1-8, 31-33, and 35 confirms the hypothesis of the presence in the N-6 region of the adenosine A 3 subtype of a proton donor able to bind to a CdO proton acceptor at position 4.

A structural investigation on some 4-amido-2-phenyl-1,2-dihydro-1,2,4-triazolo[4,3-a]quinoxalin-1-one derivatives, designed as human A3 adenosine receptor (hA3 AR) antagonists, is described. In the new derivatives, some acyl residues with different steric bulk were introduced on the 4-amino group, and their combination with the 4-methoxy group on the 2-phenyl moiety, and/or the 6-nitro/6-amino substituent on the fused benzo ring, was also evaluated. Most of the new derivatives were potent and selective hA3 AR antagonists. SAR analysis showed that hindering and lipophilic acyl moieties not only are well tolerated but even ameliorate the hA3 affinity. Interestingly, the 4-methoxy substituent on the appended 2-phenyl moiety, as well as the 6-amino group, always exerted a positive effect, shifting the affinity toward the hA3 receptor subtype. In contrast, the 6-nitro substituent exerted a variable effect. An intensive molecular modeling investigation was performed to rationalize the experimental SAR findings.

Introduction. Glutamate (Glu) is probably the major excitatory transmitter in the CNS, but it is also likely to be involved in many pathological processes. The excitotoxicity of Glu is mainly mediated by the overstimulation of the ionotropic Glu receptors (iGluRs): i.e., N-methyl-D-aspartate (NMDA), R-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), and kainate (KA) receptors. 1,2 Accordingly, there has been a growing interest in the therapeutic potential of antagonists acting at these iGluRs. [3][4][5][6] As part of a program aimed at finding novel iGluR antagonists, we have recently reported the synthesis of a set of 4,5-dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-carboxylates (TQXs). 7 The combined glycine/ NMDA and AMPA receptor affinity of TQXs 7 suggested that this tricyclic system may represent a structure which, suitably modified, could lead to selective glycine/ NMDA or AMPA receptor antagonists. We have hypothesized that combining the TQX framework with a 8-(imidazol-1-yl) substituent might produce selective AMPA antagonists. In fact, a pharmacophore model 6,8 of the AMPA receptor for the binding of quinoxalinedione and heterocyclic-fused quinoxalinone antagonists claimed that the imidazol-1-yl substituent was essential for high AMPA receptor activity. To verify this hypothesis, we have replaced the 8-nitro substituent of our mixed glycine/NMDA and AMPA antagonist 7-chloro-8-nitro-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-carboxylic acid (1) 7 with the bioisoster imidazol-1-yl moiety (compounds 2 and 3, Chart 1). Moreover, although several nitrogen-containing heterocycles have been introduced as substituents on the benzo-fused moiety of AMPA antagonists, 9,10 to our knowledge the 1,2,4-triazol-4-yl substituent has never been considered in the structure-activity relationship (SAR) studies on quinoxalinone 9 and heterocyclic-fused quinoxalinone antagonists. 8 Thus, we synthesized and tested at iGluRs the 7-chloro-4,5-dihydro-8-(1,2,4-triazol-4-yl)-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-carboxylates 4 and 5 (Chart 1) to evaluate the influence of the 8-(1,2,4-triazol-4-yl) substituent at this crucial position in AMPA receptor recognition.

A series of 4,5-dihydro-1,2,4-triazolo[1,5-a]quinoxalin-4-ones bearing different substituents on the benzo-fused ring and at position 2 were synthesized and biologically evaluated for their binding at glycine/NMDA and AMPA receptors. Most of the reported compounds show combined glycine/NMDA and AMPA receptor binding activity providing further evidences of the structural similarities of the binding pockets of both receptor recognition sites. Moreover, this study has pointed out some differences for the binding at each receptor type. In particular, for the glycine/NMDA receptor-ligand interaction, the presence of a free acidic function at position 2 and an electron-withdrawing substituent(s) nonbulkier than chlorine atom(s) on the benzo-fused moiety is required. Functional antagonism at the NMDA receptor-ion channel complex was also performed on some selected compounds.

No abstract