Effect of Nitrogen Atom Substitution in A3 Adenosine Receptor Binding

Aug 9, 2017 - The discovery of 2,6-diaryl-2-acetamidopyridines represents a step forward in the search of chemically simple, potent, and selective ant...
5 downloads 13 Views 3MB Size
Article pubs.acs.org/jmc

Effect of Nitrogen Atom Substitution in A3 Adenosine Receptor Binding: N‑(4,6-Diarylpyridin-2-yl)acetamides as Potent and Selective Antagonists Jhonny Azuaje,†,‡,# Willem Jespers,§,# Vicente Yaziji,†,‡ Ana Mallo,†,‡ María Majellaro,†,‡ Olga Caamaño,‡ María I. Loza,∥ María I. Cadavid,∥ José Brea,∥ Johan Åqvist,§ Eddy Sotelo,*,†,‡ and Hugo Gutiérrez-de-Terán*,§ †

Centro Singular de Investigación en Química Biolóxica e Materiais Moleculares (CIQUS), ‡Departamento de Química Orgánica, Facultade de Farmacia, and ∥Centro Singular de Investigación en Medicina Molecular y Enfermedades Crónicas (CIMUS), Universidade de Santiago de Compostela, 15782, Santiago de Compostela, Spain § Department of Cell and Molecular Biology, Uppsala University, Uppsala SE-75124, Sweden S Supporting Information *

ABSTRACT: We report the first family of 2-acetamidopyridines as potent and selective A3 adenosine receptor (AR) antagonists. The computer-assisted design was focused on the bioisosteric replacement of the N1 atom by a CH group in a previous series of diarylpyrimidines. Some of the generated 2-acetamidopyridines elicit an antagonistic effect with excellent affinity (Ki < 10 nM) and outstanding selectivity profiles, providing an alternative and simpler chemical scaffold to the parent series of diarylpyrimidines. In addition, using molecular dynamics and free energy perturbation simulations, we elucidate the effect of the second nitrogen of the parent diarylpyrimidines, which is revealed as a stabilizer of a water network in the binding site. The discovery of 2,6-diaryl-2-acetamidopyridines represents a step forward in the search of chemically simple, potent, and selective antagonists for the hA3AR, and exemplifies the benefits of a joint theoretical−experimental approach to identify novel hA3AR antagonists through succinct and efficient synthetic methodologies.



INTRODUCTION The regulation of the adenosinergic signaling system has progressively emerged as a valuable strategy to address serious unmet medical needs in diverse therapeutic areas.1,2 Our improved understanding of the physiology, pharmacology, and structural biology of the G-protein-coupled receptors (GPCRs) that signal for adenosine, namely, A1, A2A, A2B, and A3 adenosine receptors (ARs)3 is complemented with a plethora of small molecules that more or less effectively modulate them.4,5 The release of crystal structures of the human A2AAR in complex with different antagonists and agonists,6,7 and more recently of the A1AR in complex with a covalently bound xanthine antagonist,8 have contributed to our improved understanding of receptor−ligand interactions important for high affinity and selectivity. As a consequence the structurebased rational design of AR ligands is particularly mature, as recently reviewed by us7 and the case-study of a pharmaceutical company.6 The most recently characterized member of the AR family, albeit not yet at the structural level, is the A3AR.9 It is involved in a variety of vital physiological processes, and © 2017 American Chemical Society

accordingly the ligands modulating this receptor have been proposed as drugs with the associated pathologies, including modulation of cerebral and cardiac ischemic damage,10 glaucoma,11 inflammation,12 asthma and COPD,13 and tumor cell growth and cancer.14 In addition, there is a growing demand for pharmacological tools to study the human A3AR, all of which has made the identification of novel A3AR modulators a topic of great interest.15−18 In our laboratory, we are deeply engaged in the development of potent, selective, and structurally simple AR antagonists. Our integrated approach combines succinct and efficient synthetic methodologies assessed by a rational computer-based design of the exploratory series.19 In the framework of the A3AR, we have recently focused on amidopyrimidines as structurally simple, monocyclic scaffolds, where we could vary up to three points of diversity to end up with low molecular weight, potent, and selective A3 antagonists.20 Starting from A1AR-selective or Received: June 13, 2017 Published: August 9, 2017 7502

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

Figure 1. Rationale of the design strategy of targeted structures (3).

Table 1. Structure and Affinity Binding Data for the N-(4,6-Diarylpyridin-2-yl)acetamides 3 at the Human Adenosine Receptors

Ki (nM) or % at 1 μM comp

Ar

hA1a

hA2Ab

hA2Bc

hA3d

hA3 fold changee

3a 3b 3c 3d 3e 3f 3g 3h 3i 3j 3k 3l 3m 3n 3o 3p

Ph 4-Me-Ph 4-MeS-Ph 4-MeCO-Ph 2-MeO-Ph 3-MeO-Ph 4-MeO-Ph 4-CF3O-Ph 2,3-MeO-Ph 2,4-MeO-Ph 2,5-MeO-Ph 2,6-MeO-Ph 3,4-MeO-Ph 3,4-OCH2O-Ph 3,5-MeO-Ph 3,4,5-MeO-Ph

21.1 ± 0.4 7% 3% 2% 8% 239 ± 3 25% 1% 4% 2% 3% 1% 3% 91.1 ± 4 1% 8%

49% 1% 2% 4% 1% 21% 15% 1% 353 ± 4 2% 3% 1% 1% 9% 2% 15%

51.2 ± 2 2% 2% 3% 19% 96.2 ± 2 19% 1% 579 ± 3 1% 1% 3% 1% 27% 2% 4%

98.5 ± 3 20.5 ± 0.9 69.4 ± 1 7.2 ± 0.4 781 ± 3 14.9 ± 0.3 107.1 ± 2 20% 668 ± 4 26% 35.2 ± 1.4 21% 7.9 ± 0.9 2.8 ± 0.2 109.9 ± 2 116.5 ± 2

8.1 4.7 1.0 0.3 32.4 5.3 29.8 15.8 >50 5.6 5.3 1.5 0.9 2.0 4.5

a Displacement of specific [3H]DPCPX binding in human CHO cells expressed as Ki ± SEM in nM (n = 3) or percentage displacement of specific binding at a concentration of 1 μM (n = 2). bDisplacement of specific [3H]4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5ylamino]ethyl)phenol binding in human HeLa cells expressed as Ki ± SEM in nM (n = 3) or percentage displacement of specific binding at a concentration of 1 μM (n = 2). cDisplacement of specific [3H]DPCPX binding in human HEK-293 cells expressed as Ki ± SEM in nM (n = 3) or percentage displacement of specific binding at a concentration of 1 μM (n = 2). dDisplacement of specific [3H]NECA binding in human HeLa cells expressed as Ki ± SEM in nM (n = 3) or percentage displacement of specific binding at a concentration of 1 μM (n = 2). eExperimental effect of the bioisosteric replacement of pyrimidine (2) by pyridine series 3 in the A3 affinity. The fold change in affinity is calculated as ΔKi = Ki(3) /Ki(2).

structure of A2AAR in complex with the reference triazolotriazine antagonist 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3a][1,3,5]triazin-5-ylamino]ethyl)phenol (ZM241385) suggested that the N1 in the 4-amidopyrimidine series (2) could interact with a potentially conserved structural water molecule linked to Ser2717.42, for which evidence in the hA3AR homology model was suggested by a GRID analysis with a water probe.21 The effect of a “necessary nitrogen” in the structure−activity (SAR) of heterocycles is recognized to be a complex phenomenon. 23 It can be involved in hydrogen-bond interactions with the receptor and/or solvent in the binding cavity, but can also be related to other effects such as altering the energy profile to achieve the required bioactive conformation or modulate different physicochemical properties of the scaffold, all of which can play a role in the ligand-binding profile. Accordingly, we have designed a new series of bioisosteres [e.g., N-(4,6-diarylpyridin-2-yl)acetamides (3)], derived from the structurally simple N-(4,6-diarylpyrimidin-2-

promiscuous AR ligands, we have documented highly potent and specific A3AR antagonists by the selective introduction of methoxyphenyl fragments into the privileged 4-amidopyrimidine scaffold.20,21 As part of this exploration, we characterized the key contribution of the nitrogen atom near the amide fragment, N3 for the two isomeric series shown in Figure 1 (2amidopyrimidines, series 1 and 4-amidopyrimidines, series 2, incoming blue arrows). Together with the exocyclic nitrogen of the amido group (outgoing blue arrows, Figure 1), this nitrogen is part of a double hydrogen bond with the AR conserved residue Asn2506.55 (note the Ballesteros−Weinstein topological nomenclature of GPCRs as superscripts),22 as observed in all A2AAR antagonists crystallized to date. The actual role of the other nitrogen atom of the pyrimidine ring (N1, red circles in Figure 1) was also investigated. The relative position of this nitrogen between the 2-amido (1) and the 4-amidopyrimidines (2) was related, by means of a quantitative structure−activity relationship (QSAR) analysis, with the systematic higher affinity of the later series. A comparison with the crystal 7503

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

yl)acetamide scaffold (2). According to the design, this new series of hA3AR antagonists would allow us to determine whether the N1 nitrogen of pyrimidines is needed for high affinity to this receptor. Building upon our computational model of the pyrimidine series, we herein make use of our flexible synthetic methodology to report the design, synthesis, pharmacological evaluation, and structural interpretation of this novel family of potent and highly selective A3AR antagonists. The new series represents a step forward in the chemical simplicity of the scaffold as compared to the previous series of diarylpyrimidines.20,21 The most salient features of the SAR and structure−selectivity (SSR) relationships in this series are analyzed and provide a detailed map of the structural requirements for specific and potent binding to the hA3AR. In particular, we identified the role of the N1 nitrogen atom in the pyrimidine series by means of all-atom molecular dynamics (MD) simulations and free energy perturbation (FEP) calculations. Altogether, this report complements our previous findings and contributes to a better comprehension of the binding mode of these monocyclic antagonists (Figure 1, Cpds 1−3), paving the way for further improvements and lead optimization phases of these already potent and selective hA3AR antagonists. Chemistry. An efficient two-step synthetic strategy was optimized (Table 1, header) aiming to accomplish an exhaustive exploration of the pyridine template with diverse aryl fragments. The preparative method, inspired by the synthetic pathway employed for the synthesis of the diarylpyrimidine congeners (1 and 2),20,21 relied on the availability of the key precursor 4 and a collection of 16 commercially available aryl boronic acids (6a−p). Briefly, acetylation of the amino group of the heterocyclic precursor 4, by treatment with acetyl chloride in pyridine, afforded the key intermediate 5 in near quantitative yield. The palladium-catalyzed bisarylation of 5, employing the standard conditions of the reliable Suzuki− Miyaura cross-coupling reaction,24 provided the targeted N(4,6-diarylpyridin-2-yl)acetamides (3a−p) in yields ranging from 78 to 95%. Biological Evaluation. The affinities of the obtained compounds for the four human adenosine receptor subtypes were determined in vitro using radioligand binding assays. The experimental protocols were in essence the same as described in the report of our N-(2,6-diarylpyrimidin-4-yl)acetamide series 2, the affinities of which are included in Supplementary Table S1 for the sake of comparison.20,21 In brief, human adenosine receptors were expressed in transfected CHO (A1AR), HeLa (A2AAR and A3AR), and HEK-293 (A2BAR) cells. (3H)-1,3-Dipropyl-8-cyclopentylxanthine ([3H]DPCPX) for A1AR and [3H]NECA for A3AR were employed as radioligands in binding assays. The resulting biological data of the N-(2,6-diarylpyridin-4-yl)acetamide series 3 (Table 1) are expressed as Ki ± SEM (nM, n = 3), or alternatively as a percentage of inhibition of specific binding at 1 μM (n = 2, average) for those compounds that did not fully displace radioligand binding. Functional Assay at Adenosine A3 Receptors. Three representative ligands with high affinity and selectivity for the hA3AR (3d, 3m, and 3n, Table 1) were also studied through cAMP experiments. The functional evaluation was carried out with intact cells expressing the hA3AR.20,21 The inhibition of forskolin-stimulated cAMP production by the receptor agonist was used as a read-out. Concentration−response curves of the selected compounds over 0.1 μM NECA-induced A3AR

Figure 2. Effect of 3d (○, dashed fitting), 3m (●, black fitting) and 3n (◊, blue fitting) on 0.1 μM NECA-induced cAMP decrease of 10 μM forskolin-stimulated hA3 ARs. Points represent the mean ± SEM (vertical bars) of two separate experiments.

activation are shown in Figure 2. cAMP formation was measured by enzyme immunoassay (GE Healthcare). Antagonistic potency, measured as KB, was calculated from the formula: KB = (IC50)/((2 + ([A]/[A50]n)1/n − 1), where IC50 is the concentration of the antagonist that inhibits the agonist stimulation by 50%, [A] is the concentration of the agonist in the assay, [A50] is the concentration of the agonist that elicits the half-maximum response, and n is the slope of the concentration response curve.25 The tested derivatives (3d, 3m, and 3n) fully reverted the A3AR-elicited inhibition of cAMP accumulation, unequivocally validating its antagonistic behavior at the hA3AR (Figure 2, KB = 57.2 nM, 36.8 nM, and 1.9 nM, respectively). Moreover, a comparative analysis with the Ki data reported in Table 1 (3d, Ki = 7.20 nM; 3m, Ki = 7.95 nM and 3n, Ki = 2.88 nM) revealed a reasonable correspondence between functional and affinity values data. Structure−Activity Relationships and Molecular Modeling. In the following section the SARs arising from the pharmacological data (Table 1, describing the 16 novel N-(4,6diarylpyridin-2-yl)acetamides 3a−p) are established and compared to the corresponding data reported for the pyrimidine congeners 2a−p (Table S1 and Figure S2 of the Supporting Information).20,21 The SAR is further interpreted on the basis of the computational 3D model of the hA3AR used in the design of the two series of ligands. A first inspection of the pharmacological data reveals that pyridine is a promising scaffold to develop hA3AR ligands. Five compounds (3b, 3d, 3k, 3m, and 3n) show high affinity for the receptor (Ki < 50 nM). They generally exhibit excellent selectivity profiles with negligible affinities for any other AR, with the only exception of the residual affinity of compound 3n for the hA1AR, which is still 30-fold selective for the A3 vs the A1 receptor. A closer examination of the data and comparison with the related pyrimidines (Table S1 and Figure S2 of the Supporting Information) reveals some prominent features of the SAR within this series. The new series contains potent and highly selective A3 antagonists, in a few cases equipotent (see the fold change between pairs 3c/2c and 3n/2n in Figure S2, bearing 4methyltiophenyl and piperonyl groups, all with Ki values at hA3AR in the low nanomolar range as stated in Table 1 and Table S1), and in one case even more potent than the corresponding pyrimidines (e.g., compound 3d with a 4acetylphenyl is almost 3-fold more potent than 2d). However, in many cases the bioisosteric replacement results in a slight loss of potency as compared to the corresponding pyrimidines (e.g., compounds 3a−b, 3f, 3h−i, 3k−m, 3o−p show a 3.8-fold 7504

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

Figure 3. (A) Binding mode of compound 3d (magenta) in hA3AR (helices TM1 → TM7 colored blue → red), with the residues interacting with the ligand shown and labeled. (B) Superposition of the final binding mode determined for the whole series in the binding cavity of A3AR (mesh surface representation).

L2 is accommodated in the inner cavity of the receptor, and the analogous substituent in L3 is oriented toward the extracellular crevice. Notably, some of the compounds bearing these substitutions need to shift the position of the ring by 1.5−2.5 Å toward the extracellular cavity, losing the second direct Hbond between N1 and Asn2506.55 while maintaining the rest of the interactions described above (see Figure 3B). The effect of o-methoxylation is different in the pyridine (3) as compared to the pyrimidine series (2). The fact that the o-methoxyl is close to the variable atom in the ring (N or CH) results in a different torsional energetic profile between the two series (Figure S3 in Supporting Information), which can provide a rationale for the different effect of this substitution in the affinities for the two series. Substitution in the para position generally leads to improvements in affinity (compounds 3b−d and 3m−n). The binding mode of one of the most active compounds within this group (3d, Ki = 7.2 nM) is depicted in Figure 3, showing the optimal shape complementarity of the 4-COMePh substituents in L2/L3 and even potential H-bonds with Tyr151.35. However, 4-methoxyphenyl substitution leads to a moderately active compound (3g) or, in combination with the ortho methoxylation, to the completely inactive 3j. This comes as a surprise, since the most interesting 2-acetamidopyrimidines were precisely obtained by p-methoxylation (see affinities of compounds 2g and 2j in Table S1, and a comparison with the corresponding pyrimidines in Figure S2). Also conserved between the two series is the role of the p-trifluoromethoxyphenyl substitution, in this case detrimental for the affinity. The next step was to elucidate the role in receptor binding of the bioisosteric replacement of the N1 in the N-(2,6diarylpyrimidin-4-yl)acetamides (2) by CH in the current pyridine series (3). Our binding models show that this position is located deep in the binding pocket, with no residues in the vicinity that could act as a hydrogen bond donor to the N1 of pyrimidines (2). Instead, for this series we had initially modeled an interaction with a water molecule in the binding site.20 This hypothesis is here examined through comparative modeling of the pyridine (3) and pyrimidine series (2) based on all-atom molecular dynamics (MD) and free energy perturbation (FEP) simulations. We first analyzed the water occupancy of the binding site during the course of unbiased MD simulations of the A3AR in complex with representative compounds 2a and 3a, both bearing the simplest aromatic substituents (benzene). The results show a water network in the first hydration shell around the pyrimidine (Figure 4A, density corresponding to

lower affinity on average, with a standard deviation of 4.7-fold). This effect is more pronounced for derivatives bearing 2methoxyphenyl or 4-methoxyphenyl fragments (3e and 3g respectively), both with around a 30-fold loss in affinity as compared to 2e and 2g, respectively. In addition, the combination of substitutions on ortho and para positions in the phenyl rings results in the completely inactive N-[4,6-(2,4dimethoxyphenylpyridin-2-yl)]acetamide 3j, in stark contrast to the effect of these same substituents on the pyrimidine series, leading to one of the most active compound 2j (Ki = 5.4 nM, Table S1). The homology model of the hA3AR, used for the design of the current pyridine series (3),20 was employed here to gain further understanding of the SAR that we qualitatively discussed above. An automated docking exploration did not suggest any alternative binding mode which is conserved along the series, and instead the analogous binding mode to that previously reported for the pyrimidines was found for 15 out of 16 compounds in the present series. Figure 3 shows the detailed binding mode for the potent compound 3d, together with a superposition of the 16 compounds in the series as obtained after energetic refinement (see Experimental Section). The heterocyclic core is stabilized by a π-stacking interaction with the side chain of Phe168EL2, in such a way that a hydrogen bond between the exocyclic amido group (donating) is established with Asn2506.55 and in most cases accompanied by a second H-bond between the same residue (donating) and the nitrogen of the pyridine nucleus (accepting). The aromatic substituent in position 6 (L2) is sitting at the deep transmembrane cavity, defined by Leu903.32, Leu91 3.33, Met1775.38, Phe1825.43, Trp2436.48, and Leu2466.51, while the other aromatic ring (L3) faces the extracellular vestibule, in a region defined by Tyr151.35, Ala692.61, Val722.64, Leu2647.37, Tyr2657.36, and Ile2687.39. Finally, the acetamido group (L1) is pointing toward the extracellular loops, surrounded by Ile253,6.58 Val169EL2, and Met1745.35. On the light of this binding mode, a few conclusions can be extracted about the SAR of the series, which does not always follow the same trend as in the pyrimidine series 2. The detrimental effect on ligand binding of the methoxylation in the ortho position is only compensated by a simultaneous methoxylation at the relative position 5; e.g., compare the unsubstituted diphenyl derivative 3a with compound 3e bearing ortho-methoxyphenyls, and with compounds 3i−l, which combine ortho with meta or para methoxylations. In our models, the 2-methoxylphenyl group in 7505

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

Figure 4. Structure of the complex of pyrimidine 2a (A) and pyridine 3a (B) with the hA3AR. The average water occupancy calculated from unbiased MD trajectories is illustrated as a volume density map (red), with assigned structural waters labeled as referred to in the text.

>50 fold for 2j/3j), which was correctly captured in our model (see Table 2). The latter pair is particularly challenging, not only because of the extreme difference in binding affinity, but especially because the 2,4-dimethoxy substitution can be theoretically accommodated in a total of four possible binding orientations while respecting the binding mode of the pyridine/ pyrimidine series proposed (see Figure S5 in the Supporting Information). Consequently, we performed parallel FEP simulations in the four different binding modes. The results indicate that the best correlation with the experimental difference in affinity are obtained when the 2,4-dimethoxy at the L2 site interacts with N2506.55 (see Table 2, pose b and pose c, indicated in bold). These two poses differ in the position of the 2,4-dimethoxy at the L3 site oriented to the extracellular crevice (Figure S5B,C). The orientation of the substituents on L3, however, seem to have less influence in binding affinity, since the calculated values on the two orientations are both compatible with the experimental difference in affinity between the two compounds. Altogether, the remarkable agreement with the experimental data for the three representative pairs of compounds (Table 2) further reinforces the suggestion arising from unbiased MD simulations about the effect of the N1 in the pyrimidine series (2) in the stabilization of a water network in the binding site.

waters W5−W9), which is highly stable as compared to the pyridine complex (see lower density in this region in the corresponding map in Figure 4B). Interestingly, a comparison with the high-resolution crystal structure of the A2AAR in complex with a triazolotriazine reference antagonist26 allows the identification of some structural waters which are analogous to the positions here identified for the hA3AR (Figure S4 of the Supporting Information). In particular, we identified the analogy between W5(A3)/W2527(A2B), W6(A3)/W10(A3)/W2522(A2B), and W7(A3)/W2584(A2B), whereas W8−9(A3) seems to replace the NH group in the triazolotriazine. In the A2AAR, it has been recognized that antagonist binding can be strongly influenced by the perturbation or stabilization of water molecules in the binding site.27 Here, we suggest a similar effect in the binding of pyrimidine and pyridine scaffolds to the hA3AR. Thus, in a second stage we assessed the energetics of the water-mediated interactions in the same pair of compounds through FEP simulations. The corresponding transformation of compound 2a into 3a only involves converting the C−H to N in position 1. The moderate experimental loss in affinity of about 5 fold, corresponding to a difference of 1.2 kcal/mol in relative free energies of binding, is correctly reproduced by the FEP simulations in our model (Table 2). To further validate these results, we applied the same analysis for the pair of compounds 2g/3g, bearing the 4-methoxy substitution, and 2j/3j, bearing the 2,4-dimethoxy substitution pattern. The transformation of pyrimidine to pyridine for these compounds showed the most drastic effect in experimental affinities (30 fold for 2g/3g and



CONCLUSIONS In summary, we report the design, synthesis, and pharmacological and structural characterization of a novel family of structurally simple, highly potent, and totally selective antagonists of the hA3AR. The resulting series of N-(4,6diarylpyridin-2-yl)acetamides (3) were designed as bioisosteres of a parent series of pyrimidines (2), and indeed the most remarkable pyridines (3d and 3m) show affinity values in the low nanomolar range (Ki < 10 nM), comparable to the best pyrimidines of former series, while devoid of any affinity toward the remaining ARs (≤10% displacement of 1 μM concentrations). Their antagonistic behavior was corroborated through functional cAMP experiments, and a joint analysis of the current and previous series now allows for a comprehensive understanding of binding mode and SARs of the two scaffolds. Consequently, the comparative analysis of the SAR of the pyrimidine (2) and newly synthesized pyridine (3) series, based on the same homology model of the hA3AR used for their design, shows a binding mode that is essentially the same in the two cases. Using this as a starting point, we designed a MD/

Table 2. Experimental and Calculated Relative Binding Free Energies between Pairs of Pyrimidine/Pyridine Ligands transformation 2a → 3a 2g → 3g 2j → 3j

ΔΔGexp (kcal/mol)a

ΔΔGcalc (kcal/mol)

1.24 2.01 >3.09b

pose

a b c d

1.83 3.19 −0.52 7.44 5.17 −2.59

± 0.14 ± 0.10 ± 0.15 ± 0.22 ± 0.19 ± 0.06

a The relative binding free energies (ΔΔGexp) were calculated from experimentally determined Ki values using the relation ΔGobind,exp = RT ln Ki. bNo experimental value could be determined (no full displacement at 1 μM), and the calculated ΔΔGexp represents the detection threshold.

7506

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

(EtOH) to give 198 mg, 85%. Mp 156−158 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 9.12 (s, 1H), 8.45 (s, 1H), 7.87 (d, J = 8.2 Hz, 2H), 7.66 (d, J = 8.3 Hz, 2H), 7.29 (t, J = 7.3 Hz, 4H), 2.43 (s, 3H), 2.41 (s, 3H), 1.96 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 169.2, 156.3, 152.2, 151.8, 139.2, 136.2, 135.6, 129.7, 129.5, 127.1, 126.8, 114.4, 109.8, 24.4, 21.3, 21.2. HRMS (EI) m/z calcd. for C21H20N2O [M]+: 316.1574, found: 316.1577. N-(4,6-Bis(4-(methylthio)phenyl)pyridin-2-yl)acetamide (3c). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 199 mg, 71%. Mp 127−128 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 8.69 (s, 1H), 8.39 (s, 1H), 7.90 (d, J = 8.4 Hz, 2H), 7.68−7.59 (m, 3H), 7.40−7.27 (m, 4H), 2.52 (s, 3H), 2.51 (s, 3H), 2.07 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 168.9, 155.6, 152.0, 151.1, 140.3, 140.3, 135.4, 134.9, 127.5, 127.2, 126.5, 126.3, 113.9, 109.6, 24.6, 15.4. HRMS (EI) m/z calcd. for C21H20N2OS [M]+: 380.1017, found: 380.1019. N-(4,6-Bis(4-acetylphenyl)pyridin-2-yl)acetamide (3d). Purified by column chromatography (n-hexane−ethyl acetate 5:1) and then recrystallized (EtOH) to give 178 mg, 65%. Mp 144−145 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.50 (s, 1H), 8.16 (s, 1H), 8.13− 8.01 (m, 6H), 7.87−7.77 (m, 2H), 7.74 (d, J = 1.3 Hz, 1H), 2.66 (d, J = 2.0 Hz, 6H), 2.28 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 197.7, 197.5, 168.8, 155.0, 151.9, 150.8, 142.7, 137.4, 129.0, 128.8, 127.5, 127.0, 115.3, 111.0, 26.7, 24.9. HRMS (EI) m/z calcd. for C23H20N2O3 [M]+: 372.1471, found: 372.1475. N-(4,6-Bis(2-methoxyphenyl)pyridin-2-yl)acetamide (3e). Purified by column chromatography (n-hexane−ethyl acetate 6:1) and then recrystallized from EtOH to give 192 mg, 75%. Mp 138−140 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.61 (s, 1H), 8.32 (s, 1H), 7.70 (s, 1H), 7.67 (dd, J = 7.6, 1.5 Hz, 1H), 7.44 (dd, J = 7.5, 1.5 Hz, 1H), 7.41−7.32 (m, 2H), 7.13−6.93 (m, 4H), 3.86 (s, 3H), 3.81 (s, 3H), 2.06 (s, 3H).13C NMR (75 MHz, CDCl3) δ (ppm): 168.7, 157.0, 156.7, 153.9, 150.9, 148.6, 130.9, 130.8, 129.9, 128.9, 128.4, 122.1, 120.9, 112.8, 111.5, 111.4, 55.6, 55.6, 24.5. HRMS (EI) m/z calcd. for C21H20N2O3 [M]+: 348.1476, found: 348.1475. N-(4,6-Bis(3-methoxyphenyl)pyridin-2-yl)acetamide (3f). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 200 mg, 78%. Mp 134−135 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.99 (s, 1H), 8.45 (s, 1H), 7.67 (s, 1H), 7.56 (d, J = 7.2 Hz, 1H), 7.44−7.28 (m, 3H), 7.25 (s, 1H), 7.13− 6.82 (m, 2H), 3.88 (s, 3H), 3.82 (s, 3H), 2.01 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 169.1, 160.1, 156.0, 152.1, 151.9, 140.2, 139.9, 130.0, 129.8, 119.7, 119.3, 115.1, 115.1, 114.6, 112.9, 112.2, 110.6, 55.4, 55.3, 24.4. HRMS (EI) m/z calcd. for C21H20N2O3 [M]+: 348.1473, found: 348.1477. N-(4,6-Bis(4-methoxyphenyl)pyridin-2-yl)acetamide (3g). Purified by column chromatography (n-hexane−ethyl acetate 3:1) and then recrystallized from EtOH to give 195 mg, 76%. Mp 161−163 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.30 (s, 1H), 8.17 (s, 1H), 7.75 (d, J = 8.8 Hz, 2H), 7.53 (d, J = 8.8 Hz, 2H), 7.42 (d, J = 1.3 Hz, 1H), 6.84 (d, J = 5.9 Hz, 2H), 6.81 (d, J = 5.9 Hz, 2H), 3.70 (s, 3H), 3.69 (s, 3H), 1.95 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 168.6, 160.5, 155.8, 151.6, 151.2, 131.6, 130.9, 128.4, 128.1, 114.3, 114.1, 113.7, 108.8, 55.3, 24.8. HRMS (ESI) m/z calcd. for C21H21N2O3 [M + H]+: 349.1547, found: 349.1551. N-(4,6-Bis(4-(trifluoromethoxy)phenyl)pyridin-2-yl)acetamide (3h). Puri-fied by column chromatography (n-hexane−ethyl acetate 3:1) and then recrystallized from EtOH to give 201 mg, 60%. Mp 112−114 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.41 (s, 1H), 8.17 (s, 1H), 8.07−7.97 (m, 2H), 7.78−7.70 (m, 2H), 7.61 (d, J = 1.3 Hz, 1H), 7.32 (dt, J = 16.5, 6.2 Hz, 4H), 2.22 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 168.8, 155.0, 151.9, 150.7, 150.0, 137.2, 137.0, 128.8, 128.4, 122.1, 121.4, 121.1, 118.7, 114.6, 110.4, 24.8. HRMS (EI) m/z calcd. for C21H14F6N2O3 [M]+: 456.0912, found: 456.0910. N-(4,6-Bis(2,3-dimethoxyphenyl)pyridin-2-yl)acetamide (3i). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 207 mg, 69%. Mp 152−154 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 9.21 (s, 1H), 8.34 (s, 1H), 7.82 (s, 1H), 7.39−7.22 (m, 1H), 7.17−7.00 (m, 3H), 6.96 (dd, J = 7.8, 2.1 Hz, 2H), 3.87 (s, 6H), 3.79 (s, 3H), 3.73 (s, 3H), 1.94 (s, 3H).

FEP scheme to elucidate the role of the additional nitrogen atom (N1) in the pyrimidine series. The effect in ligand binding affinities, due to the bioisosteric N1 to CH replacement, was reproduced in well-converged FEP simulations for three pairs of compounds, providing a reasonable explanation about the role of the extra nitrogen in the pyrimidine series in the stabilization of a water network in the binding site. We also showed how this approach is useful to elucidate the most probable binding mode in cases where several docking alternatives were possible, as is the case of the pair 2j/3j. This study reports novel and simple scaffolds as potent and selective antagonists of the hA3AR, and illustrates the utility of our computer-assisted design of AR antagonists.



EXPERIMENTAL SECTION

Chemistry. Commercially available starting materials, reagents, and solvents were purchased (Sigma-Aldrich) and used without further purification. After extraction from aqueous phases, the organic solvents were dried over anhydrous sodium sulfate. The reactions were monitored by thin-layer chromatography (TLC) with 2.5 mm Merck silica gel GF 254 strips, and the purified compounds each showed a single spot; unless stated otherwise, UV light and/or iodine vapor were used for detection of compounds. The Suzuki cross-coupling reactions were performed in coated Kimble vials on a PLS (6 × 4) organic synthesizer with orbital stirring. Purity and identity of all tested compounds were established by a combination of HPLC, mass spectrometry, and NMR spectra. Purification of isolated products was carried out by column chromatography (Kieselgel 0.040−0.063 mm, E. Merck) or medium pressure liquid chromatography (MPLC) on a CombiFlash Companion (Teledyne ISCO) with RediSep prepacked normal-phase silica gel (35−60 μm) columns followed by recrystallization. Melting points were determined on a Gallenkamp melting point apparatus and are uncorrected. The purity and identity of all tested compounds were established by a combination of HPLC, mass spectrometry, and NMR spectroscopy as described below. The NMR spectra were recorded on Bruker AM300 and XM500 spectrometers. Chemical shifts are given as δ values against tetramethylsilane as internal standard and J values are given in Hz. Mass spectra were obtained on a Varian MAT-711 instrument. High resolution mass spectra were obtained on an Autospec Micromass spectrometer. Analytical HPLC was performed on an Agilent 1100 system using an Agilent Zorbax SB-Phenyl, 2.1 mm × 150 mm, 5 μm column with gradient elution using the mobile phases (A) H2O containing 0.1% CF3COOH and (B) MeCN and a flow rate of 1 mL/min. The purity of all tested compounds was determined to be >95%. Acetylation of 4amino-2,6-dichloropyridine (4) (95%) was performed by following a previously reported general procedure.20,21 General Procedure for the Synthesis of Compounds 3a−p. A mixture of the acetamide 5 (0.43 mmol), aryl boronic acid 6 (1.3 mmol), Pd(PPh3)4 (0.043 mmol), and Na2CO3 (2.1 mmol) in 5 mL of a mixture of DME/H20 (3:1) in coated Kimble vials was stirred with orbital stirring at 110 °C for 12 h. The resulting mixture was concentrated in vacuo, and then ethyl acetate was added. This solution was washed with water and 1 N NaOH to remove the remaining boronic acid. The organic layer was collected, dried over Na2SO4, filtered, and concentrated in vacuo. The residue was purified by column or preparative chromatography on silica gel. N-(4,6-Diphenylpyridin-2-yl)acetamide (3a). Purified by column chromatography (n-hexane−ethyl acetate 7:1) and then recrystallized from EtOH to give 174 mg, 82%. Mp 143−144 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.50 (s, 1H), 8.28 (s, 1H), 7.82 (d, J = 7.6 Hz, 2H), 7.58 (d, J = 7.6 Hz, 2H), 7.53 (s, 1H), 7.29−7.33 (m, 6H), 1.91 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 169.0, 156.3, 152.1, 152.0, 138.9, 138.5, 129.2, 129.1, 129.0, 128.8, 127.3, 126.9. 115.4, 110.3, 24.2. HRMS (ESI) m/z calcd. for C19H17N2O [M + H]+: 289.1335, found: 289.1333. N-(4,6-Di-p-tolylpyridin-2-yl)acetamide (3b). Purified by column chroma-tography (n-hexane−ethyl acetate 6:1) and then recrystallized 7507

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

C NMR (75 MHz, CDCl3) δ (ppm): 169.1, 153.7, 153.1, 151.4, 148.8, 147.3, 146.8, 133.7, 133.5, 124.2, 124.1, 122.5, 122.3, 121.7, 112.9, 112.8, 61.1, 60.9, 55.9, 55.9, 24.3. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1682, found: 408.1686. N-(4,6-Bis(2,4-dimethoxyphenyl)pyridin-2-yl)acetamide (3j). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 219 mg, 73%. Mp 168−169 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 9.45 (s, 1H), 8.28 (s, 1H), 7.69 (s, 1H), 7.65 (d, J = 8.3 Hz, 1H), 7.39 (d, J = 8.1 Hz, 1H), 6.75− 6.37 (m, 4H), 3.85 (s, 6H), 3.83 (s, 3H), 3.76 (s, 3H), 1.87 (s, 3H). 13 C NMR (75 MHz, CDCl3) δ (ppm): 169.1, 161.3, 158.2, 157.9, 153.7, 151.3, 148.4, 131.8, 131.4, 121.8, 121.5, 121.2, 112.3, 104.9, 99.0, 98.9, 55.6, 55.5, 55.4, 24.1. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1684, found: 408.1684. N-(4,6-Bis(2,5-dimethoxyphenyl)pyridin-2-yl)acetamide (3k). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 225 mg, 75%. Mp 171−172 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 9.00 (s, 1H), 8.33 (s, 1H), 7.74 (s, 1H), 7.30 (d, J = 2.0 Hz, 1H), 7.00 (d, J = 2.4 Hz, 1H), 6.97− 6.83 (m, 4H), 3.81 (s, 3H), 3.80 (s, 3H), 3.75 (s, 3H), 3.73 (s, 3H), 1.99 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 168.9, 153.8, 153.8, 153.6, 151.4, 151.2, 150.9, 148.7, 129.3, 129.2, 121.9, 116.3, 115.8, 115.6, 114.7, 113.2, 112.9, 56.4, 56.3, 55.8, 55.7, 24.3. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1685, found: 408.1686. N-(4,6-Bis(2,6-dimethoxyphenyl)pyridin-2-yl)acetamide (3l). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 201 mg, 67%. Mp 222−224 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 10.46 (s, 1H), 8.03 (s, 1H), 7.66 (ddd, J = 19.3, 10.3, 9.0 Hz, 1H), 7.59−7.38 (m, 1H), 7.28 (td, J = 8.4, 2.5 Hz, 1H), 7.11 (s, 1H), 6.60 (dd, J = 13.3, 8.4 Hz, 1H), 3.74 (s, 2H), 3.57 (s, 2H), 1.71 (s, 1H). 13C NMR (75 MHz, CDCl3) δ (ppm): 169.17, 158.32, 157.53, 150.89, 150.84, 144.62, 132.12, 131.99, 129.85, 129.47, 128.57, 128.41, 125.18, 118.13, 117.61, 115.76, 104.24, 103.92, 77.55, 77.13, 76.71, 55.88, 55.30, 23.46. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1685, found: 408.1684. N-(4,6-Bis(3,4-dimethoxyphenyl)pyridin-2-yl)acetamide (3m). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 216 mg, 72%. Mp 156−157 °C. 1 H NMR (300 MHz, CDCl3) δ (ppm): 8.57 (s, 1H), 8.35 (s, 1H), 7.60 (dd, J = 5.4, 1.5 Hz, 1H), 7.54 (dd, J = 8.4, 1.9 Hz, 1H), 7.32 (dd, J = 8.3, 2.0 Hz, 1H), 7.22 (d, J = 1.9 Hz, 1H), 6.96 (dd, J = 8.4, 5.5 Hz, 2H), 3.97 (s, 3H), 3.94 (s, 3H), 3.93 (s, 3H), 3.92 (s, 3H), 2.11 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 168.9, 155.8, 151.7, 151.7, 150.1, 150.1, 149.3, 149.2, 131.8, 131.4, 119.9, 119.4, 113.9, 111.4, 111.1, 110.2, 109.9, 109.2, 56.1, 56.0, 56.0, 55.9, 24.6. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1685, found: 408.1687. N-(4,6-Bis(benzo[d][1,3]dioxol-5-yl)pyridin-2-yl)acetamide (3n). Purified by column chromatography (n-hexane−ethyl acetate 3:1) and then recrystallized from EtOH to give 221 mg, 80%. Mp 192−194 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.52 (s, 1H), 8.30 (s, 1H), 7.53−7.40 (m, 3H), 7.32−7.13 (m, 2H), 7.00−6.73 (m, 2H), 6.01 (s, 2H), 6.00 (s, 2H), 2.11 (s, 3H).13C NMR (75 MHz, CDCl3) δ (ppm): 168.9, 155.6, 151.7, 151.4, 148.6, 148.4, 148.2, 133.2, 132.7, 121.2, 120.9, 113.9, 109.4, 108.7, 108.4, 107.5, 107.2, 101.4, 101.4, 24.6. HRMS (EI) m/z calcd. for C21H16N2O5 [M]+: 376.1057, found: 376.1058. N-(4,6-Bis(3,5-dimethoxyphenyl)pyridin-2-yl)acetamide (3o). Purified by column chromatography (n-hexane−ethyl acetate 4:1) and then recrystallized from EtOH to give 219 mg, 73%. Mp 153−1155 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 8.84 (s, 1H), 8.41 (s, 1H), 7.62 (s, 1H), 7.15 (d, J = 2.1 Hz, 2H), 6.83 (d, J = 1.9 Hz, 2H), 6.60− 6.47 (m, 2H), 3.86 (s, 6H), 3.81 (s, 6H), 1.99 (s, 3H). 13C NMR (75 MHz, CDCl3) δ (ppm): 169.0, 161.2, 161.2, 155.9, 152.1, 151.9, 140.8, 140.7, 115.1, 110.6, 105.5, 104.9, 101.5, 101.0, 55.5, 55.4, 24.5. HRMS (EI) m/z calcd. for C23H24N2O5 [M]+: 408.1685, found: 408.1684. N-(4,6-Bis(3,4,5-trimethoxyphenyl)pyridin-2-yl)acetamide (3p). Purified by column chromatography (n-hexane−ethyl acetate 2:1) and then recrystallized from EtOH to give 231 mg, 67%. Mp 167−168 °C. 1H NMR (300 MHz, CDCl3) δ (ppm): 9.28 (s, 1H), 8.34 (s, 1H), 13

7.54 (s, 1H), 7.21 (s, 2H), 6.84 (s, 2H), 3.88 (s, 6H), 3.85 (s, 3H), 3.83 (s, 3H), 3.80 (s, 6H), 1.94 (s, 3H).13C NMR (75 MHz, CDCl3) δ (ppm): 169.2, 155.7, 153.6, 153.5, 152.4, 152.0, 139.3, 139.1, 134.4, 134.2, 114.5, 110.3, 104.6, 104.2, 60.9, 56.3, 56.1, 24.3. HRMS (EI) m/ z calcd. for C25H28N2O7 [M]+: 468.1899, found: 468.1898. Pharmacology. Radioligand binding competition assays were performed in vitro as previously described,20,21 using A1, A2A, A2B, and A3 human adenosine receptors expressed in transfected CHO (A1AR), HeLa (A2AAR and A3AR), and HEK-293 (A2BAR) cells. A brief description is given below. Adenosine A1 receptor competition binding experiments were carried out in membranes from CHO-A1 cells labeled with 2 nM [3H]DPCPX (Kd = 0.7 nM). Nonspecific binding was determined in the presence of 10 μM R-PIA. The reaction mixture was incubated at 25 °C for 60 min. Adenosine A2A receptor competition binding experiments were carried out in membranes from HeLa-A2A cells labeled with 3 nM [3H]ZM241385 (Kd = 1.5 nM). Nonspecific binding was determined in the presence of 50 μM NECA. The reaction mixture was incubated at 25 °C for 30 min. Adenosine A2B receptor competition binding experiments were carried out in membranes from HEK-293-A2B cells (Euroscreen, Gosselies, Belgium) labeled with 35 nM [3H]DPCPX (Kd = 21 nM). Nonspecific binding was determined in the presence of 400 μM NECA. The reaction mixture was incubated at 25 °C for 30 min. Adenosine A3 receptor competition binding experiments were carried out in membranes from HeLa-A3 cells labeled with 30 nM [3H]NECA (Kd = 8.7 nM). Nonspecific binding was determined in the presence of 100 μM R-PIA. The reaction mixture was incubated at 25 °C for 180 min. Functional Experiments. cAMP Assays. These assays were performed following previously described protocols.20,21 CHO-A3 cells were seeded (20 000 cells/well) with Dulbecco’s modified Eagle’s medium nutrient mixture F-12 (DMEM F-12), containing 10% fetal calf serum (FCS) and 1% L-glutamine in 96-well culture plates and incubated at 37 °C in an atmosphere with 5% CO2 for 24 h. Cells were washed 3 times with 200 μL of assay medium (DMEM-F12 and 25 mM HEPES pH = 7.4) and preincubated with assay medium containing 30 μM rolipram and test compounds at 37 °C for 15 min. Then, 0.1 μM NECA was added to each well and incubated for 10 min at 37 °C, after this time 10 μM forskolin was incubated for 5 min at 37 °C (total incubation time 30 min). cAMP levels were determined by employing cAMP enzyme immunoassay kit (PerkinElmer) by absorbance measurements at 450 nm in an Ultra Evolution detector (Tecan). Data were fitted by nonlinear regression using GraphPad Prism v2.01 (GraphPad Software). Computational Modeling. Our computational strategy for the structure-based design of adenosine receptor ligands involves a combination of homology modeling, ligand−receptor docking and free energy calculations, as recently reviewed.19 Homology Modeling of the hA3AR. A 3D structure of the inactive form of the receptor was generated at the beginning of this project.21 Briefly, the process consisted of the following sequential steps: (i) Manual curation of the sequence alignment with the template A2AAR (PDB code 3EML),28 (ii) Generation and selection of homology models and loop refinement procedures with Modeler,29 (iii) assessment of Asn/Gln/His rotamers and side chain protonation states with the Molprobity web server (http://molprobity.biochem. duke.edu/) and (iv) use of tools from the Schrödinger Suite for energetic structural refinements.30 Ligand Docking. A first exploration was done by means of the GOLD software,31 using the same protocol as previously described for the pyrimidine series.20 Each ligand was docked 20 times with default (high accuracy) genetic algorithm (GA) search parameters, using the scoring function Chemscore and allowing full flexibility for the ligand, including flipping of amide bonds. The search sphere was centered on the side chain (CD1) of Ile2667.39 and expanded with a radius of 15 Å, thus ensuring a generous enough search space comprising the antagonist binding site experimentally determined for adenosine receptors. The criterion for the selection of docking poses was based on a combination of the Chemscore ranking and the population (convergence) of the solutions according to a clustering criteria of 1 Å. 7508

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

site in complex with ligands 1a and 3a. The resulting MD trajectories were analyzed using the VolMap plugin implemented in VMD,42 to locate the relevant binding pocket hydration sites.

A refined binding mode was obtained by modification of the published docking pose of the reference 4-acetamidopyrimidine 2g to obtain each of the 16 compounds in the new series (3a−p). Each complex generated this way was geometrically optimized by partial energy minimization of the binding site with the Schrödinger suite (Macromodel tool, “ligand-receptor minimization” with the “embrace” option, selecting residues within 5 Å of any ligand atom as free to move; OPLS3 force field).30 MD and FEP Calculations. The hA3AR-3a complex obtained in the previous stage was inserted in the membrane and equilibrated under periodic boundary conditions (PBC) using the PyMemDyn protocol described elsewhere.32 Shortly, the starting structure is automatically embedded in a pre-equilibrated membrane consisting of 1-palmitoyl-2oleoylphosphatidylcholine (POPC) lipids, with the TM bundle aligned to its vertical axis. This hexagonal-prism shaped box is then soaked with bulk water and energy minimized with GROMACS 4.6.33 using the OPLS-AA force field34 for protein and ligands, combined with the Berger parameters for the lipids.35 The same setup is used for a 2.5 ns MD equilibration, where initial restraints on protein and ligand atoms are gradually released as described in detail in reference.32 The equilibrated binding site is then transferred to the MD software Q36 for free energy perturbation (FEP) calculations under spherical boundary conditions. A 25 Å sphere centered on the center of geometry of the ligand is considered for these MD simulations. Protein atoms in the boundary of the sphere (22−25 Å outer shell) had a positional restraint of 20 kcal/mol/Å2, while solvent atoms were subject to polarization and radial restrains using the surface constrained all-atom solvent (SCAAS)36,37 model to mimic the properties of bulk water at the sphere surface. Atoms lying outside the simulation sphere are tightly constrained (200 kcal/mol/Å2 force constant) and excluded from the calculation of nonbonded interactions. Long range electrostatics interactions beyond a 10 Å cut off were treated with the local reaction field method,38 except for the atoms undergoing the FEP transformation where no cutoff was applied. Solvent bond and angles were constrained using the SHAKE algorithm.39 All titratable residues outside the sphere were neutralized, and histidine residues were assigned a hydrogen atom on the δ nitrogen. Residue parameters as well as ligand partial charges were translated from the latest version of the OPLS-AA/M force field,40 whereas ligand partial charges were generated with the version of OPLS3 implemented in Macromodel,30 and the rest of the parameters for the ligand and lipids were inherited from the previous MD stage. The simulation sphere was warmed up from 0.1 to 298 K, during a first equilibration period of 0.61 ns, where an initial restraint of 25 kcal/ mol/Å2 imposed on all heavy atoms was slowly released. Additionally, a force constant of 15.0 kcal/mol/Å2 was applied during this equilibration period to maintain the distance between the N1 of Asn2506.55 and the exocyclic NH of the ligands between the reference threshold for hydrogen bond (1.8−2.2 Å). This restraint was gradually removed during additional 2 ns equilibration MD, where the rest of the system was unrestrained (except for the sphere boundary region as explained above). Thereafter the system was subject to seven parallel replicates of unrestrained MD, where the FEP protocol is applied for each ligand transformation. Each of these MD replicates starts with a 0.25 ns unbiased equilibration period, with different initial velocities. Thereafter the FEP protocol follows, consisting on 51, evenly distributed, FEP λ-windows of 30 ps each, along which the bioisosteric replacement of the ligand (CH → N1) is simulated. In order to fulfill a thermodynamic cycle and calculate relative binding free energies, parallel FEP transformations are run in a sphere of water for each ligand pair. In these water simulations, the same parameters apply (i.e., sphere size, simulation time, etc.), and the relative binding free energy difference was estimated by solving the thermodynamic cycle utilizing the standard Zwanzig formula.41 The same protocol was applied to the 2g/3g and 2j/3j series, where the four possible binding conformers of 2j/3j were manually introduced. Unbiased MD simulations for water-density calculations were carried out using the same spherical system setup and MD parameters as indicated for the FEP data collection period above. A total of 10 ns, unrestricted and unbiased MD was run for the hA3AR solvated binding



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jmedchem.7b00860. Experimental affinities of the pyrimidine series (2), and a comparison with the experimental affinities of the pyridines (3). Torsional scan of a representative pair of derivatives of each series. Comparative analysis of the water network binding site between hA2A and hA3 ARs. Binding modes and estimated affinity change for the pair 2j/3j (PDF) Molecular formula strings (CSV)



AUTHOR INFORMATION

Corresponding Authors

*(H.G.D.T.) Phone: +46 18 471 5056. Fax: +46 18 536971. Email: [email protected]. *(E.S.) Phone: +34 881815732. Fax: +34-881815704. E-mail: [email protected]. ORCID

Willem Jespers: 0000-0002-4951-9220 Johan Åqvist: 0000-0003-2091-0610 Eddy Sotelo: 0000-0001-5571-2812 Hugo Gutiérrez-de-Terán: 0000-0003-0459-3491 Author Contributions #

J.A. and W.J. contributed equally.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was financially supported by the Conselleriá de Cultura, Educación e Ordenación Universitaria of the Galician Government: (Grant: GPC2014/03), Centro Singular de Investigación de Galicia accreditation 2016-2019 (ED431G/ 09) and the European Regional Development Fund (ERDF), the Swedish Research Council (Grant 521-2014-2118) and the Swedish Strategic Research Program eSSENCE. The computations were performed on resources provided by the Swedish National Infrastructure for Computing (SNIC). Our laboratories are part of the European COST Actions CM1207 (GLISTEN) and CM15135 (MuTaLig).



ABBREVIATIONS USED AR, adenosine receptors; CD, circular dichroism; CHO cells, Chinese hamster ovary cells; c-AMP, cyclic adenosine monophosphate; GPCRs, G protein-coupled receptors; hA1, hA2A, hA2B, hA3 ARs, human A1, A2A, A2B, A3 adenosine receptors; HPLC, high performance liquid chromatography; MCR, multicomponent reaction; MD, molecular dynamics; NOE, nuclear overhauser effect; PDB, Protein Data Bank; RMSD, root mean square deviation; SAR, structure−activity relationships; SEM, standard error of the mean



REFERENCES

(1) Jacobson, K. A.; Gao, Z.-G. Adenosine Receptors as Therapeutic Targets. Nat. Rev. Drug Discovery 2006, 5, 247−264.

7509

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

(2) Chen, J.-F.; Eltzschig, H. K.; Fredholm, B. B. Adenosine Receptors as Drug Targets–What Are the Challenges? Nat. Rev. Drug Discovery 2013, 12, 265−286. (3) Fredholm, B. B.; IJzerman, A. P.; Jacobson, K. A.; Klotz, K.-N.; Linden, J. International Union of Pharmacology. XXV. Nomenclature and Classification of Adenosine Receptors. Pharmacol. Rev. 2001, 53, 527−552. (4) Baraldi, P. G.; Tabrizi, M. A.; Gessi, S.; Borea, P. A. Adenosine Receptor Antagonists: Translating Medicinal Chemistry and Pharmacology into Clinical Utility. Chem. Rev. 2008, 108, 238−263. (5) Müller, C. E.; Jacobson, K. A. Recent Developments in Adenosine Receptor Ligands and Their Potential as Novel Drugs. Biochim. Biophys. Acta, Biomembr. 2011, 1808, 1290−1308. (6) Jazayeri, A.; Andrews, S. P.; Marshall, F. H. Structurally Enabled Discovery of Adenosine A2A Receptor Antagonists. Chem. Rev. 2017, 117, 21−37. (7) Gutiérrez-de-Terán, H.; Sallander, J.; Sotelo, E. Structure-Based Rational Design of Adenosine Receptor Ligands. Curr. Top. Med. Chem. 2017, 17, 40−58. (8) Glukhova, A.; Thal, D. M.; Nguyen, A. T.; Vecchio, E. A.; Jörg, M.; Scammells, P. J.; May, L. T.; Sexton, P. M.; Christopoulos, A. Structure of the Adenosine A1 Receptor Reveals the Basis for Subtype Selectivity. Cell 2017, 168, 867−877. (9) Zhou, Q. Y.; Li, C.; Olah, M. E.; Johnson, R. A.; Stiles, G. L.; Civelli, O. Molecular Cloning and Characterization of an Adenosine Receptor: The A3 Adenosine Receptor. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 7432−7436. (10) Liang, B. T.; Jacobson, K. A. A Physiological Role of the Adenosine A3 Receptor: Sustained Cardioprotection. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 6995−6999. (11) Jacobson, K. A.; Civan, M. M. Ocular Purine Receptors as Drug Targets in the Eye. J. Ocul. Pharmacol. Ther. 2016, 32, 534−547. (12) Akkari, R.; Burbiel, J. C.; Hockemeyer, J.; Muller, C. E. Recent Progress in the Development of Adenosine Receptor Ligands as Antiinflammatory Drugs. Curr. Top. Med. Chem. 2006, 6, 1375−1399. (13) Polosa, R.; Blackburn, M. Adenosine Receptors as Targets for Therapeutic Intervention in Asthma and Chronic Obstructive Pulmonary Disease. Trends Pharmacol. Sci. 2009, 30, 528−535. (14) Fishman, P.; Bar-Yehuda, S.; Ardon, E. Targeting the A3 Adenosine Receptor for Cancer Therapy: Inhibition of Prostate Carcinoma Cell Growth by A3AR Agonist. Anticancer Res. 2003, 23, 2077−2083. (15) Baraldi, P. G.; Preti, D.; Borea, P. A.; Varani, K. Medicinal Chemistry of A3 Adenosine Receptor Modulators: Pharmacological Activities and Therapeutic Implications. J. Med. Chem. 2012, 55, 5676−5703. (16) Cheong, S. L.; Federico, S.; Venkatesan, G.; Mandel, A. L.; Shao, Y.-M.; Moro, S.; Spalluto, G.; Pastorin, G. The A 3 Adenosine Receptor as Multifaceted Therapeutic Target: Pharmacology, Medicinal Chemistry, and in Silico Approaches. Med. Res. Rev. 2013, 33, 235−335. (17) Squarcialupi, L.; Catarzi, D.; Varano, F.; Betti, M.; Falsini, M.; Vincenzi, F.; Ravani, A.; Ciancetta, A.; Varani, K.; Moro, S.; Colotta, V. Structural Refinement of pyrazolo[4,3- D ]Pyrimidine Derivatives to Obtain Highly Potent and Selective Antagonists for the Human A 3 Adenosine Receptor. Eur. J. Med. Chem. 2016, 108, 117−133. (18) Ciancetta, A.; Jacobson, K. Structural Probing and Molecular Modeling of the A3 Adenosine Receptor: A Focus on Agonist Binding. Molecules 2017, 22, 449. (19) Gutiérrez-de-Terán, H.; Keränen, H.; Azuaje, J.; Rodríguez, D.; Åqvist, J.; Sotelo, E. Computer-Aided Design of GPCR Ligands. Methods Mol. Biol. 2015, 1272, 271−291. (20) Yaziji, V.; Rodríguez, D.; Gutiérrez-de-Terán, H.; Coelho, A.; Caamaño, O.; García-Mera, X.; Brea, J.; Loza, M. I.; Cadavid, M. I.; Sotelo, E. Pyrimidine Derivatives as Potent and Selective A 3 Adenosine Receptor Antagonists. J. Med. Chem. 2011, 54, 457−471. (21) Yaziji, V.; Rodríguez, D.; Coelho, A.; García-Mera, X.; El Maatougui, A.; Brea, J.; Loza, M. I.; Cadavid, M. I.; Gutiérrez-deTerán, H.; Sotelo, E. Selective and Potent Adenosine A3 Receptor

Antagonists by Methoxyaryl Substitution on the N-(2,6-Diarylpyrimidin-4-Yl)acetamide Scaffold. Eur. J. Med. Chem. 2013, 59, 235−242. (22) Ballesteros, J. A.; Weinstein, H. Integrated Methods for the Construction of Three-Dimensional Models and Computational Probing of Structure-Function Relations in G Protein-Coupled Receptors. Methods Neurosci. 1995, 25, 366−428. (23) Pennington, L. D.; Moustakas, D. T. The Necessary Nitrogen Atom: A Versatile High-Impact Design Element for Multiparameter Optimization. J. Med. Chem. 2017, 60, 3552−3579. (24) Miyaura, N.; Suzuki, A. Palladium-Catalyzed Cross-Coupling Reactions of Organoboron Compounds. Chem. Rev. 1995, 95, 2457− 2483. (25) Leff, P.; Dougall, I. G. Further Concerns over Cheng-Prusoff Analysis. Trends Pharmacol. Sci. 1993, 14, 110−112. (26) Liu, W.; Chun, E.; Thompson, A. A.; Chubukov, P.; Xu, F.; Katritch, V.; Han, G. W.; Roth, C. B.; Heitman, L. H.; IJzerman, A. P.; Cherezov, V.; Stevens, R. C. Structural Basis for Allosteric Regulation of GPCRs by Sodium Ions. Science 2012, 337, 232−236. (27) Bortolato, A.; Tehan, B. G.; Bodnarchuk, M. S.; Essex, J. W.; Mason, J. S. Water Network Perturbation in Ligand Binding: Adenosine A2A Antagonists as a Case Study. J. Chem. Inf. Model. 2013, 53, 1700−1713. (28) Jaakola, V.-P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y. T.; Lane, J. R.; Ijzerman, A. P.; Stevens, R. C. The 2.6 Angstrom Crystal Structure of a Human A2A Adenosine Receptor Bound to an Antagonist. Science 2008, 322, 1211−1217. (29) Šali, A.; Blundell, T. Comparative Protein Modelling by Satisfaction of Spatial Restraints. J. Mol. Biol. 1993, 234, 779−815. (30) MacroModel, Version 9.7; Schrödinger, LLC: New York, NY, 2009. (31) Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Improved Protein-Ligand Docking Using GOLD. Proteins: Struct., Funct., Genet. 2003, 52, 609−623. (32) Gutiérrez-de-Terán, H.; Bello, X.; Rodríguez, D. Characterization of the Dynamic Events of GPCRs by Automated Computational Simulations. Biochem. Soc. Trans. 2013, 41, 205−212. (33) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435−447. (34) Kaminski, G. A.; Friesner, R. A.; Tirado-Rives, J.; Jorgensen, W. L. Evaluation and Reparametrization of the OPLS-AA Force Field for Proteins via Comparison with Accurate Quantum Chemical Calculations on Peptides. J. Phys. Chem. B 2001, 105, 6474−6487. (35) Berger, O.; Edholm, O.; Jähnig, F. Molecular Dynamics Simulations of a Fluid Bilayer of Dipalmitoylphosphatidylcholine at Full Hydration, Constant Pressure, and Constant Temperature. Biophys. J. 1997, 72, 2002−2013. (36) Marelius, J.; Kolmodin, K.; Feierberg, I.; Aqvist, J. Q: A Molecular Dynamics Program for Free Energy Calculations and Empirical Valence Bond Simulations in Biomolecular Systems. J. Mol. Graphics Modell. 199816, 213, 225.10.1016/S1093-3263(98)80006-5 (37) King, G.; Warshel, A. A Surface Constrained All-atom Solvent Model for Effective Simulations of Polar Solutions. J. Chem. Phys. 1989, 91, 3647−3661. (38) Lee, F. S.; Warshel, A. A Local Reaction Field Method for Fast Evaluation of Long-Range Electrostatic Interactions in Molecular Simulations. J. Chem. Phys. 1992, 97, 3100. (39) Ryckaert, J.-P.; Ciccotti, G.; Berendsen, H. J. Numerical Integration of the Cartesian Equations of Motion of a System with Constraints: Molecular Dynamics of N-Alkanes. J. Comput. Phys. 1977, 23, 327−341. (40) Robertson, M. J.; Tirado-Rives, J.; Jorgensen, W. L. Improved Peptide and Protein Torsional Energetics with the OPLS-AA Force Field. J. Chem. Theory Comput. 2015, 11, 3499−3509. (41) Zwanzig, R. W. High-Temperature Equation of State by a Perturbation Method. I. Nonpolar Gases. J. Chem. Phys. 1954, 22, 1420. 7510

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511

Journal of Medicinal Chemistry

Article

(42) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual Molecular Dynamics. J. Mol. Graphics 1996, 14, 33−38.

7511

DOI: 10.1021/acs.jmedchem.7b00860 J. Med. Chem. 2017, 60, 7502−7511