Fluorosulfonyl- and Bis-(β-chloroethyl)amino-phenylamino Functionalized Pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine Derivatives: Irreversible Antagonists at the Human A₃ Adenosine Receptor and Molecular Modeling Studies

Abstract

A series of pyrazolotriazolopyrimidines was previously reported to be highly potent and selective human A₃ adenosine receptor antagonists (Baraldi et al. J. Med. Chem. 2000, 43, 4768–4780). A derivative having a methyl group at the N⁸ pyrazole combined with a 4-methoxyphenylcarbamoyl moiety at N⁵ position, displayed a K_i value at the hA₃ receptor of 0.2 nM. We now describe chemically reactive derivatives which act as irreversible inhibitors of this receptor. Electrophilic groups, specifically sulfonyl fluoride and nitrogen mustard (bis-(β-chloroethyl)amino) moieties, have been incorporated at the 4-position of the aryl urea group. Membranes containing the recombinant hA₃ receptor were preincubated with the compounds and washed exhaustively. The loss of ability to bind radioligand following this treatment indicated irreversible binding. The most potent compound in irreversibly binding to the receptor was 14, which contained a sulfonyl fluoride moiety and a propyl group at the N⁸ pyrazole nitrogen. The bis-(β-chloroethyl)amino derivatives displayed a much smaller degree of irreversible binding than the sulfonyl fluoride derivatives. A computer-generated model of the human A₃ receptor was built and analyzed to help interpret these results. The model of the A₃ transmembrane region was derived using primary sequence comparison, secondary structure predictions, and three-dimensional homology building, using the recently published crystal structure of rhodopsin as a template. According to our model, sulfonyl fluoride derivatives could dock within the hypothetical TM binding domain, adopting two different energetically favorable conformations. We have identified two amino acids, Ser247 and Cys251, both in TM6, as potential nucleophilic partners of the irreversible binding to the receptor.

Introduction

In the last 20 years, much effort has been made toward the goal of identifying ligands that interact selectively with the four adenosine receptor subtypes, A₁, A_2A, A_2B, and A₃.^1,2 In particular, while A₁ and A_2A subtypes have been pharmacologically characterized through the use of chemical probes, A_2B and A₃ subtypes are still in need of additional probes.³

The A₃ subtype, recently cloned from various species (human, rat, and rabbit),^4-9 seems to be involved in a variety of physiological functions. The principal second messengers activated by adenosine A₃ receptors are inhibition of adenylate cyclase³ and stimulation of phospholipases C¹⁰ and D.¹¹ In addition, activation of the rat A₃ adenosine receptors causes the release of histamine from mast cells,¹² which leads to processes such as inflammation and hypotension. It has been suggested that the blockade of the A₃ adenosine receptor may be considered as a potential lead toward the treatment of asthma and inflammation.^13,14 For these reasons, and with the aim of obtaining efficient probes for studying this receptor subtype, there has been an active search for potent and selective antagonists. In the past few years, many classes of heterocyclic compounds have been synthesized as A₃ adenosine antagonists, with some showing significant selectivity for the A₃ adenosine receptor subtype.¹⁵ In particular, our group recently reported a class of 5-[[(4-substituted-phenyl)amino]carbonyl]amino-8-alkyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine derivatives as potent and selective human A₃ adenosine receptor antagonists. These were ineffective on a rat model, due to the well-known differences in sequence homologies (72%) (Table 1).^16,17

Table 1.

Structures and Binding Affinity at hA₁, hA_2A, hA_2B, and hA₃ Adenosine Receptors of Selected Compounds Previously Reported

compd	R	hA1 (Ki, nM)a	hA2A (Ki, nM)b	hA2B (Ki, nM)c	hA3 (Ki, nM)d	hA1/hA3	hA2A/hA3	hA2B/hA3
1	CH3	1097 (928–1297)	1390 (1220–1590)	261 (226–301)	0.20 (0.17–0.24)	5485	6950	1305
2	C2H5	1026 (785–1341)	1040 (830–1310)	245 (188–320)	0.60 (0.51–0.70)	1710	1733	408
3	n-C3H7	1197 (1027–1396)	140 (120–155)	2056 (1637–2582)	0.80 (0.63–0.1.00)	1496	175	2570
4	n-C4H9	296 (269–326)	80 (65–92)	303 (260–352)	0.32 (0.27–0.34)	925	250	946

^aDisplacement of specific [³H]-DPCPX binding at human A₁ receptors expressed in CHO cells (n = 3–6).

^bDisplacement of specific [³H]-SCH58261 binding at human A_2A receptors expressed in CHO cells (n = 3–6).

^cDisplacement of specific [³H]-DPCPX binding at human A_2B receptors expressed in HEK-293 cells (n = 3–6).

^dDisplacement of specific [³H]MRE-3008F20 binding at human A₃ receptors expressed in CHO cells (n = 3–6). Data are expressed as geometric means, with 95% confidence limits.

It is evident from the binding data summarized in Table 1 that affinity increased when the 4-methoxyphenyl carbamoyl moiety at the 5 position and small alkyl chains (methyl (1), ethyl (2), propyl (3), butyl (4)) at the N⁸ pyrazole nitrogen were combined simultaneously. These derivatives are the most potent and selective human A₃ adenosine receptor antagonists ever reported, with affinity at the human A₃ adenosine receptor in the subnanomolar range (0.2–0.8 nM) and selectivities versus other receptor subtypes ranging from 200 to 7000.¹⁷

Beginning with these experimental observations, we proposed the design of new compounds in this series by replacing the 4-methoxy substituent on the phenyl ring with alkylating moieties such as fluorosulfonyl and bis(β-chloroethyl)amino groups (Figure 1), as potential irreversibly binding human A₃ adenosine antagonists (9–16).

Figure 1.

Rational design for the synthesis of irreversible human A₃ adenosine antagonists.

The use of irreversible antagonists represents a useful approach to the study of receptor structure and function, such as ligand binding site mapping,¹⁸ physiological receptors functioning,¹⁹ and determining the relationship between receptor occupancy and a tissue or cellular response.²⁰ A similar approach has been successful with A₁^21--24 and A_2A²⁵ adenosine receptor subtypes, as clearly reported in the literature.

Also for the A₃ receptor, this approach has been used both for agonists (adenosine derivatives)²⁶ and antagonists (dihydropyridine derivatives).²⁷ While the agonists showed a good binding profile in a rat model,²⁶ at the human A₃ adenosine receptors only racemic dihydropyridine derivatives have been tested, and the percentage of irreversibility reached only 56% at 100 nM concentration.²⁷ Thus, there is a need for improved irreversibly binding A₃ antagonists.

In addition, the introduction of two different alkylating moieties could help to identify the nature of nucleophiles on the receptor which may interact with these derivatives, and consequently this may help identify the nature of some amino acid residues present within the binding site. In fact, while a fluorosulfonyl group is highly reactive²⁸ and could interact indifferently with any kind of nucleophile (e.g., –NH₂, OH, SH), it is well-known that nitrogen mustard, which is less reactive than fluorosulfonyl moiety,²⁹ reacts only with nitrogen nucleophiles (e.g., $-{\text{NH}}_2,-\text{NH}-\text{C(NH)}={\text{NH}}_2$).

Chemistry

Potentially irreversible human A₃ receptor antagonists (9–16) were prepared following, as previously reported,¹⁷ the general synthetic pathway for the preparation of urea derivatives, as summarized in Schemes 1 and 2.

Scheme 1^a.

^a Reagents: (i) PhNCO (17), THF, reflux, 18 h; (ii) FSO₃H, −10 °C, 2 h.

Scheme 2^a.

^a Reagents: (i) ClCOOCCl₃, Et₃N, dioxane, reflux, 5 h; (ii) 5–8, THF, reflux, 18 h.

Fluorosulfonyl derivatives (10, 12, 14, 16) were prepared starting from the previously reported amino compounds, 5–8, which reacted with the commercially available phenyl isocyanate 17, to afford the phenylcarbamoyl derivatives, 18–21. After purification by flash chromatography, the resulting compounds, 18–21, reacted at −10 °C with fluorosulfonic acid leading to the final derivatives, 10, 12, 14, and 16.^25e The final products were purified by crystallization from EtOAc–light petroleum (Scheme 1).

The mustard derivatives (9, 11, 13, 15) were prepared in a similar manner by reaction of the amino compounds 5–8 with freshly prepared 4-bis-(β-chloroethyl)aminophenyl isocyanate, 22. The latter compound was prepared starting from the corresponding aniline derivative,³⁰ 23, by reaction with diphosgene (1.2 equiv) in the presence of triethylamine (1.2 equiv) in dry dioxane.³¹ After 5 h the mixture was concentrated at reduced pressure, and the presence of isocyanate was confirmed by infrared spectroscopy, which showed a characteristic peak at 2250 cm⁻¹. The crude residue was utilized for coupling reactions without further purification (Scheme 2).

Results and Discussion

The reactive pyrazolo-triazolo-pyrimidine derivatives (9–16) were tested for irreversible binding at the human A₃ adenosine receptors in CHO cells using the radioligand [¹²⁵I]AB-MECA (Table 2).^32,33

Table 2.

Irreversible Inhibition of Radioligand ([¹²⁵I]AB-MECA) Binding to Human A₃ Adenosine Receptors in CHO Cell Membranes, Following a 1 h Incubation with Compounds (9–16) at Concentrations of 1, 10, and 100 nM

	ligand		inhibition (% of control)a
	R	R1	1 nM	10 nM	100 nM
9	CH3	N(CH2CH2Cl)2	0	11 ± 6	22 ± 5
10	CH3	SO2F	NDb	NDb	NDb
11	C2H5	N(CH2CH2Cl)2	0	0	18 ± 17
12	C2H5	SO2F	4.1 ± 5.1	30 ± 5	65 ± 4
13	n-C3H7	N(CH2CH2Cl)2	7.8 ± 10	11 ± 9	16 ± 14
14	n-C3H7	SO2F	47 ± 7	70 ± 3	79 ± 8
15	n-C4H9	N(CH2CH2Cl)2	0	7.0 ± 4.9	9.0 ± 7.2
16	n-C4H9	SO2F	31 ± 6	40 ± 11	73 ± 7

^an = 3–7.

^bNot determined due to the insolubility of the sample at 1 mM in DMSO.

Prior to radioligand binding, membranes containing hA₃ receptors were preincubated with the reactive antagonist (9–16), followed by a thorough washing, to remove free antagonist.

The data summarized in Table 2 showed that compounds containing a fluorosulfonyl moiety (10, 12, 14, 16) proved to be irreversible antagonists with different degrees of potency at the human A₃ adenosine receptor, while the corresponding nitrogen mustard derivatives (9, 11, 13, 15) were unable to bind covalently at this receptor subtype.

In the fluorosulfonyl series, derivative 10 could not be evaluated for irreversible receptor binding due to its low solubility in DMSO, which was less than 1 mM. Other fluorosulfonyl derivatives were sufficiently soluble in DMSO to prepare stock solutions for the binding studies. (The final concentration of DMSO in buffer must be maximum 1%.) The three fluorosulfonyl compounds (12, 14, 16) showed good irreversible properties at concentrations of 1–100 nM. In particular, high efficiency was observed at a concentration of 100 nM, with the percentage of inhibition being 65–79%. This effect was quite evident also at a concentration of 10 nM, with irreversibility in the range of 30–70%. The maximal irreversible inhibition was observed with the derivative having a propyl chain at the N⁸ position (14, 100 nM, 79%).

In addition, the presence of a small alkyl chain on the N⁸ position (Table 1) conferred at the compound an optimized structural feature required for the interaction with human A₃ adenosine receptor, as previously demonstrated.^16,17

In contrast, derivatives with a nitrogen mustard group on the phenyl ring (9, 11, 13, 15) showed much less significant irreversible activity at the human A₃ adenosine receptor. In fact, only at a concentration of 100 nM was slight irreversibility observed (9–22%).

With the aim to evaluate the selectivity of these derivatives vs other adenosine receptor subtypes, the most effective irreversible compound (14) was also tested in competitive binding assays at human A₁ and A_2A receptors expressed in CHO cells. At the human A₁ receptor, no significant displacement of binding (<5%) was observed for compound 14 at 100 nM, a concentration which effectively irreversibly inhibited most of the hA₃ receptors. However, at the human A_2A receptor, a $K_{\mathrm{i}}$ for binding of 14 was determined to be 50 ± 5 nM. The irreversibility of binding of 14 at the human A_2A receptor was not evaluated. This result potentially compromises use of these antagonists as selective hA₃ receptor probes; however, the lower affinity at the A_2A receptor would suggest that there likely exists a concentration range of selectivity for irreversible inhibition of the hA₃ adenosine receptor subtype.

This large difference in receptor interaction between the fluorosulfonyl and mustard series can be explained with two different hypotheses. The inactivity of the mustard derivatives can be attributed to the absence in the binding site of a nitrogen nucleophile (e.g., NH₂), which is the only one able to interact with the aziridinium salt generated by the bis-(β-dichloroethyl)amino function, or to the spatial orientation of this group within the binding site, which does not permit the nucleophiles to properly bind.

A molecular modeling study was performed to elucidate the hypothetical binding mode of the new A₃ irreversible antagonists and to determine which amino acid residues in the TM region might bond covalently with the ligand. Very little structural information was available concerning the details of ligand/GPCR interactions. However, a 2.8 Å resolution crystal structure of bovine rhodopsin was very recently published.³⁴ Using a homology approach and this structure as a template, we built an improved model of the transmembrane region of the human A₃ receptor. This model may be considered a further refinement of the hypothetical binding site model of the A₃ receptor antagonists previously proposed.³⁵ Details of the model building are given in the Experimental Section.

We selected derivative 14, which contained a sulfonyl fluoride moiety and a propyl group at the N⁸ pyrazole nitrogen, and the corresponding nitrogen mustard analogue 13 as reference ligands in our docking studies. The most stable ligand conformations of both irreversible antagonists are shown in Figure 2.

Figure 2.

Side view of the A₃-14 (right side) and A₃-13 (left side) complex models. The side chains of the two important residues in proximity to the docked ligand molecules are highlighted and labeled.

Interestingly, the receptor–ligand complex containing derivative 14 appeared more energetically stable by approximately 15 kcal/mol. In derivative 13, the presence of the (bis-(β-chloroethyl)amino) moiety (derivative 13), characterized by a high degree of conformational flexibility, increased the steric hindrance inside the binding cavity, decreasing the relative stability of the complex. This may be the reason for the low efficacy of the nitrogen mustard derivatives as irreversible antagonists.

As shown in Figure 2, we identified a hypothetical binding site of both 14 and 13 surrounded by TMs 3, 5, 6, and 7 following the Monte Carlo/annealing sampling. Both the sulfonyl fluoride and nitrogen mustard moieties were close to TM6, pointing toward the extracellular environment. Two residues, Ser247 and Cys251, were tentatively implicated in irreversible binding. The side chain of Ser247 (TM6) was 3.6 Å from the S atom of the sulfonyl fluoride and 4.9 Å from one of the two ${\mathrm{C}}_{\alpha }$ of the β-chloroethyl moiety. Moreover, the side chain of Cys251 (TM6) was 4.2 Å from the S atom of the sulfonyl fluoride, and 5.0 Å from one of the two ${\mathrm{C}}_{\alpha }$ of the β-chloroethyl moiety. As already anticipated, Cys251 (TM6) would be a more likely candidate for the irreversible binding, based solely on its specific chemical reactivity. However, it is useful to emphasize the proximity of the Ser247 to the putative ligand binding site. A major structural difference among the hypothetical binding sites of adenosine receptor subtypes is that the A₃ receptor does not contain the histidine residue in TM6 common to all A₁ (His251 in h_A₁) and A₂ (His250 in h_A_2A) receptors. This histidine has been shown to participate in both agonist and antagonist binding to the A_2A receptors. In the A₃ receptor, this histidine in TM6 is replaced with a serine residue, i.e., Ser247, a candidate for the site of irreversible binding, due to its proximity to the electrophilic group.

Nevertheless, it is important to note that these new irreversible antagonists share a common question with all other irreversibly binding compounds: is it necessary to find a nucleophile residue in proximity to the docked conformation or can they bind provisionally elsewhere on the receptor, on the way down to the TM site, and remain long enough to form a covalent bond? In this study we do not answer this question, but we can speculate that in the proximity of the TM binding domain there are at least two residues tentatively implicated in the irreversible binding.

Conclusions

In summary, a new series of pyrazolo-triazolo-pyrimidines, bearing chemoreactive moieties (4-fluorsulfonyl- and 4-bis-(β-chloroethyl)amino-) as irreversible antagonists for the human A₃ adenosine receptor, has been reported. The results from an irreversible binding assay demonstrate that one of the reported compounds, 5-[[(4-fluorosulfonyl-phenyl)amino]carbonyl]amino-8-n-propyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (14), irreversibly inhibited the binding of [¹²⁵I]AB-MECA by 79% at a 100 nM concentration. In contrast, all the derivatives bearing the 4-bis-(β-chloroethyl)amino function failed to irreversibly bind the hA₃ adenosine receptors. These observations were analyzed by molecular modeling studies, demonstrating that Cys251 or Ser247 are the most likely sites for covalent binding. These hypotheses may now be tested using a site directed mutagenesis approach.

In conclusion, these results provide new pharmacological probes for improving our understanding of the pathophysiological role of the human A₃ adenosine receptor, yielding new information about the structural requirements necessary for the receptor recognition.

Experimental Section

Chemistry. General.

Reactions were routinely monitored by thin-layer chromatography (TLC) on silica gel (precoated F₂₅₄ Merck plates). Products were visualized with iodine or potassium permanganate solution. Infrared spectra (IR) were measured on a Perkin-Elmer 257 spectrophotometer. ¹H NMR were determined in CDCl₃ or DMSO-d₆ solutions with a Bruker AC 200 spectrometer. Peaks positions were given in parts per million (δ), downfield from tetramethylsilane as internal standard, and J values were given in hertz. Light petroleum ether referred to the fractions boiling at 40–60 °C. Melting points were determined on a Buchi-Tottoli instrument and were uncorrected. Chromatography was performed using Merck 60–200 mesh silica gel. All products reported showed IR and ¹H NMR spectra in agreement with the assigned structures. Organic solutions were dried over anhydrous magnesium sulfate. Elemental analyses were performed by the microanalytical laboratory of Dipartimento di Chimica, University of Ferrara, and were within ±0.4% of the theoretical values for C, H, and N.

General Procedures for the Preparation of 5-[[(Substituted-phenyl)amino]carbonyl]amino-8-alkyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (9, 11, 13, 15, 18–21).

The amino precursor (5–8) (10 mmol) was dissolved in freshly distilled THF (15 mL), and the appropriate isocyanate (17, 22) (13 mmol) was added. The mixture was refluxed under argon for 18 h. Then the solvent was removed under reduced pressure, and the residue was purified by flash chromatography (EtOAc–light petroleum 4–6) to afford the desired compounds 9, 11, 13, 15, 18–21.

5-[[(Phenyl)amino]carbonyl]amino-8-methyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (18):

yield 90%, yellow solid; mp 165–170 °C (EtOAc–light petroleum); IR (KBr) 3255–2990, 1680, 1620, 1510, 1450 cm⁻¹; ¹H NMR (DMSO-d₆) δ 4.20 (s, 3H, N⁸-CH₃); 6.61 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.11–7.37 (m, 6H, C₆H₅ and 3-H Furane); 7.76 (d, 1H, J = 2, 5-H Furane); 8.15 (s, 1H, N⁸-CH); 8.67 (bs, 1H, CO-NH-Ph); 11.16 (bs, 1H, 5-NH-CO). Anal. (C₁₈H₁₄N₈O₂) C, H, N.

5-[[(Phenyl)amino]carbonyl]amino-8-ethyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (19):

yield 94%, yellow solid; mp 180 °C (EtOAc–light petroleum); IR (KBr) 3250–2985, 1675, 1610, 1510, 1460 cm⁻¹; ¹H NMR (DMSO-d₆) δ 1.67 (t, 3H, J = 7, CH₃-CH₂-N⁸); 4.47 (q, 2H, J = 7, CH₃-CH₂-N⁸); 6.62 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.16–7.42 (m, 6H, C₆H₅ and 3-H Furane); 7.72 (d, 1H, J = 2, 5-H Furane); 8.26 (s, 1H, N⁸-CH); 8.61 (bs, 1H, CO-NH-Ph); 11.16 (bs, 1H, 5-NH-CO). Anal. (C₁₉H₁₆N₈O₂) C, H, N.

5-[[(Phenyl)amino]carbonyl]amino-8-n-propyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (20):

yield 98%, yellow solid; mp 155 °C (EtOAc–light petroleum); IR (KBr) 3240–2985, 1674, 1615, 1510, 1470 cm⁻¹; ¹H NMR (DMSO-d₆) δ 1.00 (t, 3H, J = 7, CH₃-CH₂-CH₂-N⁸); 2.02–2.13 (m, 2H, CH₃-CH₂-CH₂-N⁸); 4.36 (t, 2H, J = 7, CH₃-CH₂-CH₂-N⁸); 6.62 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.16–7.49 (m, 6H, C₆H₅ and 3-H Furane); 7.69 (d, 1H, J = 2, 5-H Furane); 8.24 (s, 1H, N⁸-CH); 8.61 (bs, 1H, CO-NH-Ph); 11.16 (bs, 1H, 5-NH-CO). Anal. (C₂₀H₁₈N₈O₂) C, H, N.

5-[[(Phenyl)amino]carbonyl]amino-8-n-butyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (21):

yield 92%, white solid; mp 145 °C (EtOAc–light petroleum); IR (KBr) 3240–2980, 1675, 1610, 1500, 1470 cm⁻¹; ¹H NMR (DMSO-d₆) δ 0.96 (t, 3H, J = 7, CH₃-CH₂-CH₂-CH₂-N⁸); 1.30–1.41 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 1.98–2.06 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 4.39 (t, 2H, J = 7, CH₃-CH₂-CH₂-CH₂-N⁸); 6.62 (dd, 1H, J = 2, J = 4, 4-H Furane); 7.16–7.42 (m, 6H, C₆H₅ and 3-H Furane); 7.67 (d, 1H, J = 2, 5-H Furane); 8.23 (s, 1H, N⁸-CH); 8.62 (bs, 1H, CO-NH-Ph); 11.16 (bs, 1H, 5-NH-CO). Anal. (C₂₁H₂₀N₈O₂) C, H, N.

5-[[(4-Bis-β-cloroethylamino-phenyl)amino]carbonyl]amino-8-methyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (9):

yield 82%, yellow solid; mp 165 °C (EtOAc–light petroleum); IR (KBr) 3230–2990, 1670, 1615, 1510, 1460 cm⁻¹; ¹H NMR (DMSO-d₆) δ 3.60–3.73 (m, 8H, N(CH₂CH₂Cl)₂); 4.21 (s, 3H, CH₃-N⁸); 6.62 (dd, 1H, J = 2, J = 4, 4-H Furane); 6.66 (d, 2H, J = 9, CO-NH-Ph); 7.24 (d, 1H, J = 4, 3-H Furane); 7.51 (d, 2H, J = 9, 4-Ph- N(CH₂CH₂Cl)₂); 7.66 (d, 1H, J = 2, 5-H Furane); 8.22 (s, 1H, N⁸-CH); 8.60 (bs, 1H, CO-NH-Ph); 10.86 (bs, 1H, 5-NH-CO). Anal. (C₂₂H₂₁N₉O₂Cl₂) C, H, N.

5-[[(4-Bis-β-cloroethylamino-phenyl)amino]carbonyl]amino-8-ethyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (11):

yield 85%, white solid; mp 185 °C (EtOAc–light petroleum); IR (KBr) 3235–2985, 1675, 1620, 1511, 1450 cm⁻¹; ¹H NMR (DMSO-d₆) δ: 1.67 (t, 3H, J = 6.8, CH₃-CH₂-N⁸); 3.62–3.74 (m, 8H, N(CH₂CH₂Cl)₂); 4.46 (q, 2H, J = 6.8, CH₃-CH₂-N⁸); 6.62 (dd, 1H, J = 2, J = 4, 4-H Furane); 6.71 (d, 2H, J = 9, CO-NH-Ph); 7.24 (d, 1H, J = 4, 3-H Furane); 7.51 (d, 2H, J = 9, 4-Ph-N(CH₂CH₂Cl)₂); 7.66 (d, 1H, J = 2, 5-H Furane); 8.25 (s, 1H, N⁸-CH); 8.59 (bs, 1H, CO-NH-Ph); 11.04 (bs, 1H, 5-NH-CO). Anal. (C₂₃H₂₃N₉O₂Cl₂) C, H, N.

5-[[(4-Bis-β-cloroethylamino-phenyl)amino]carbonyl]amino-8-n-propyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (13):

yield 78%, white solid; mp 195 °C (EtOAc–light petroleum); IR (KBr) 3230–2985, 1674, 1625, 1520, 1445 cm⁻¹; ¹H NMR (DMSO-d₆) δ 1.01 (t, 3H, J = 6.8, CH₃-CH₂-CH₂-N⁸); 1.97–2.08 (m, 2H, CH₃-CH₂-CH₂-N⁸); 3.62–3.76 (m, 8H, N(CH₂CH₂Cl)₂); 4.35 (t, 2H, J = 6.8, CH₃-CH₂-CH₂-N⁸); 6.62 (dd, 1H, J = 2, J = 4, 4-H Furane); 6.71 (d, 2H, J = 9, CO–NH-Ph); 7.24 (d, 1H, J = 4, 3-H Furane); 7.51 (d, 2H, J = 9, 4-Ph- N(CH₂CH₂Cl)₂); 7.66 (d, 1H, J = 2, 5-H Furane); 8.23 (s, 1H, N⁸-CH); 8.59 (bs, 1H, CO-NH-Ph); 10.85 (bs, 1H, 5-NH-CO). Anal. (C₂₄H₂₅N₉O₂Cl₂) C, H, N.

5-[[(4-Bis-β-cloroethylamino-phenyl)amino]carbonyl]amino-8-n-butyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (15):

yield 80%, pale yellow solid; mp 190 °C (EtOAc–light petroleum); IR (KBr) 3235–2980, 1672, 1620, 1510, 1450 cm⁻¹; ¹H NMR (DMSO-d₆) δ 0.99 (t, 3H, J = 6.8, CH₃-CH₂-CH₂-CH₂-N⁸); 1.23–1.47 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 1.89–2.04 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 3.64–3.74 (m, 8H, N(CH₂CH₂Cl)₂); 4.39 (t, 2H, J = 6.8, CH₃-CH₂-CH₂-CH₂-N⁸); 6.61 (dd, 2H, J = 2, J = 4, 4-H Furane); 6.71 (d, 2H, J = 9, CO-NH-Ph); 7.24 (d, 1H, J = 4, 3-H Furane); 7.51 (d, 2H, J = 9, 4-Ph-N(CH₂CH₂Cl)₂); 7.66 (d, 1H, J = 2, 5-H Furane); 8.23 (s, 1H, N⁸-CH); 8.59 (bs, 1H, CO-NH-Ph); 10.84 (bs, 1H, 5-NH-CO). Anal. (C₂₅H₂₇N₉O₂Cl₂) C, H, N.

General Procedures for the Preparation of 5-[[(4-Fluorosulfonyl-phenyl)amino]carbonyl]amino-8-alkyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (10, 12, 14, 16).

The appropriate phenyl-carbamoyl derivative (18–21) (0.086 mmol) were reacted with fluorosulfonic acid (0.1 mL) at −10 °C for 2 h. Then the mixture was quenched with ice water (1 mL), and the precipitate was filtered off to afford the desired compounds (10, 12, 14, 16) as a solid in a good yield.

5-[[(4-Fluorosulfonyl-phenyl)amino]carbonyl]amino-8-methyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (10):

yield 78%, yellow solid; mp 170 °C (EtOAc–light petroleum); IR (KBr) 3230–2995, 1670, 1615, 1500, 1320, 1470, 1160 cm⁻¹; ¹H NMR (DMSO-d₆) δ 4.14 (s, 3H, CH₃-N⁸); 6.76 (dd, 1H, J = 2, J = 4, 4-H Furane); 7.29 (d, 1H, J = 4, 3-H Furane); 7.89 (d, 2H, J = 9, CO-NH-Ph); 7.99 (d, 1H, J = 2, 5-H Furane); 8.14 (d, 2H, J = 9, 4-Ph-SO₂F); 8.80 (s, 1H, N⁸-CH); 10.15 (bs, 1H, CO-NH-Ph); 11.06 (bs, 1H, 5-NH-CO). Anal. (C₁₈H₁₃N₈SO₄F) C, H, N.

5-[[(4-Fluorosulfonyl-phenyl)amino]carbonyl]amino-8-ethyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (12):

yield 72%, pale yellow solid; mp 165 °C (EtOAc–light petroleum); IR (KBr) 3235–2990, 1675, 1620, 1510, 1330, 1470, 1160 cm⁻¹; ¹H NMR (DMSO-d₆) δ 1.5 (t, 3H, J = 6.8, CH₃-CH₂-N⁸); 4.42 (q, 2H, J = 6.8, CH₃-CH₂-N⁸); 6.76 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.29 (d, 1H, J = 4, 3-H Furane); 7.96 (d, 2H, J = 9, CO-NH-Ph); 7.99 (d, 1H, J = 2, 5-H Furane); 8.14 (d, 2H, J = 9, 4-Ph-SO₂F); 8.86 (s, 1H, N⁸-CH); 10.19 (bs, 1H, CO-NH-Ph); 11.04 (bs, 1H, 5-NH-CO). Anal. (C₁₉H₁₅N₈SO₄F) C, H, N.

5-[[(4-Fluorosulfonyl-phenyl)amino]carbonyl]amino-8-n-propyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (14):

yield 75%, white solid; mp 155 °C (EtOAc–light petroleum); IR (KBr) 3235–2990, 1673, 1615, 1515, 1325, 1465, 1150 cm⁻¹; ¹H NMR (DMSO-d₆) δ 0.87 (t, 3H, J = 6.8, CH₃-CH₂-CH₂-N⁸); 1.90–1.95 (m, 2H, CH₃-CH₂-CH₂-N⁸); 4.35 (t, 2H, J = 6.8, CH₃-CH₂-CH₂-N⁸); 6.76 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.29 (d, 1H, J = 4, 3-H Furane); 7.89 (d, 2H, J = 9, CO-NH-Ph); 8.01 (d, 1H, J = 2, 5-H Furane); 8.13 (d, 2H, J = 9, 4-Ph-SO₂F); 8.85 (s, 1H, N₈-CH); 10.12 (bs, 1H, CO-NH-Ph); 11.04 (bs, 1H, 5-NH-CO). Anal. (C₂₀H₁₇N₈SO₄F) C, H,N.

5-[[(4-Fluorosulfonyl-phenyl)amino]carbonyl]amino-8-n-butyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine (16):

yield 77%, white solid; mp 145 °C (EtOAc–light petroleum); IR (KBr) 3225–2980, 1670, 1610, 1525, 1335, 1465, 1155 cm⁻¹; ¹H NMR (DMSO-d₆) δ 0.97 (t, 3H, J = 7, CH₃-CH₂-CH₂-CH₂-N⁸); 1.29–1.34 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 1.98–2.06 (m, 2H, CH₃-CH₂-CH₂-CH₂-N⁸); 4.43 (t, 2H, J = 7, CH₃-CH₂-CH₂-CH₂-N⁸); 6.65 (dd, 2H, J = 2, J = 4, 4-H Furane); 7.29 (d, 1H, J = 4, 3-H Furane); 7.85 (d, 2H, J = 9, CO–NH–Ph); 8.04 (d, 1H, J = 2, 5-H Furane); 8.10 (d, 2H, J = 9, 4-Ph-SO₂F); 8.78 (s, 1H, N⁸-CH); 10.11 (bs, 1H, CO-NH-Ph); 11.88 (bs, 1H, 5-NH-CO). Anal. (C₂₁H₁₉N₈SO₄F) C, H, N.

Biology. hA₁ and hA_2A Binding Assays.

Binding of [³H]-DPCPX to CHO cells transfected with the human recombinant A₁ adenosine receptor was performed as previously described.³⁶

Displacement experiments were performed for 120 min at 25 °C in 0.20 mL of buffer containing 1 nM [³H]-DPCPX, 20 μL of diluted membranes (50 μg of protein/assay), and at least six to eight different concentrations of examined compounds. Nonspecific binding was determined in the presence of 10 μM of CHA, and this is always ≤10% of the total binding.

Binding of [³H]-SCH58261 to CHO cells transfected with the human recombinant A_2A adenosine receptors (50 μg of protein/assay) was performed according to Varani et al.³⁷ In competition studies, at least six to eight different concentrations of compounds were used, and nonspecific binding was determined in the presence of 50 μM NECA for an incubation time of 60 min at 25 °C. Bound and free radioactivity were separated by filtering the assay mixture through Whatman GF/B glass-fiber filters using a Micro-Mate 196 cell harvester (Packard Instrument Company). The filter bound radioactivity was counted on Top Count (efficiency 57%) with Micro-Scint 20. The protein concentration was determined according to the Bio-Rad method³⁸ with bovine albumin as reference standard.

hA₃ Irreversible Binding Assay.

Binding of [¹²⁵I]AB-MECA to membranes prepared from CHO cells stably expressing the human A₃ receptor was performed as described.^32,33 The assay medium consisted of a buffer containing 10 mM Mg²⁺, 50 mM Tris, and 1 mM EDTA, at pH 7.4. The glass incubation tubes were filled with 100 μL of the membrane suspension (1.0 mg of protein/mL, stored at −80 °C in the same buffer), 20 μL of [¹²⁵I]AB-MECA (final concentration 0.6 nM), and 50 μL of a solution of the proposed antagonist. Nonspecific binding was determined in the presence of 100 μM NECA. Adenosine deaminase (3 units/mL) was present during the preparation of the membranes, in a preincubation of 30 min at 30 °C and during the incubation with the radioligands.

All nonradioactive compounds (9–16) were initially dissolved in DMSO and diluted with buffer to the desired final concentartion. The amount of DMSO never exceeded 1%.

Incubations were terminated by rapid filtration over Whatman GF/B filters, using a Brandell cell harvester (Brandell, Gaithersburg, MD). The tubes were rinsed three times, each with 3 mL of buffer.

For studies of irreversible binding, membranes were incubated with the indicated (Table 1) concentration of ligand (9–16) for 1 h at room temperature. Following the incubation, the membrane suspension was centrifuged at 28 000 rpm for 30 min, and the pellet was resuspended in fresh buffer and kept at 4 °C for 30 min.

Membranes were washed a total of 9 times by sequential resuspension and centrifugation with buffer (50 mM Tris, 10 mM Mg²⁺, and 1 mM EDTA) at pH 7.4 containing 0.02% CHAPS. Adenosine deaminase (3 units/mL) was present during all steps. The final pellet was suspended in a fixed volume of the above buffer solution for the A₃ receptor binding assay as described.^32,33

Computational Approach.

All calculations were performed on a Silicon Graphics O2 R10000 workstation.

The human A₃ receptor model was built and optimized using MOE (2000.02) modeling package³⁹ based on the approach described by Moro et al.⁴⁰ Briefly, transmembrane domains were identified with the aid of Kyte-Doolittle hydrophobicity and E_min surface probability parameters.⁴¹ Transmembrane helices were built from the sequences and minimized individually. The minimized helices were then grouped together to form a helical bundle that matched the overall characteristics of the recently published crystal structure of bovine rhodopsin (PDB ID: 1F88). The helical bundle was minimized using the Amber94 all-atoms force field,⁴² until the rms value of the conjugate gradient (CG) was <0.01 kcal/mol/Å. A fixed dielectric constant of 4.0 was used throughout these calculations.

All pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine derivatives were fully optimized without geometric constraints using RHF/AM1 semiempirical calculations.⁴³ Vibrational frequency analyses were used to characterize the minima stationary points (zero imaginary frequencies). The software package Gaussian98 was utilized for all quantum mechanical calculations.⁴⁴

The ligands were docked into the hypothetical TMs binding site using the DOCK docking program, part of the MOE suite.³⁹ This program incorporates the use of manual and automatic docking procedures in combination with molecular mechanics within the Simulations module of MOE. The docking method employed enables nonbonded van der Waals and electrostatic interactions to be simultaneously monitored during the docking, and several possible conformations for the ligand were evaluated interactively. “Flexible” ligand docking was then used to define the lowest energy position of each ligand using a Monte Carlo/annealing based automated docking protocol. This uses a random iterative algorithm to sample changes in torsion angles and atomic positions while simultaneously recalculating internal and interaction energies. The automated docking procedure then selected the best structures and applied the totality of the binding site. During the docking, all torsion angles of the side chains on the ligand were allowed to vary. This docking procedure was followed by another sequence of CG energy minimization to a gradient threshold of < 0.1 kcal/mol/Å. Energy minimization of the complexes was performed using AMBER94 all-atom force field.

The interaction energy values were calculated as follows: $\Delta E_{(\text{complex})}=E_{(\text{complex})}-(E_{(\mathrm{L})}+{\mathrm{E}}_{(\text{receptor})})$. These energies are not rigorous thermodynamic quantities, but can only be used to compare the relative stabilities of the complexes. Consequently, these interaction energy values cannot be used to calculate binding affinities, since changes in entropy and solvation effects are not taken into account.

Abbreviations:

DMSO

dimethyl sulfoxide

DPCPX

1,3-dipropyl-8-cytclopentylxanthine

THF

tetrahydrofurane

CHO cells

Chinese hamster ovary cells

EDTA

ethylenediaminotetraacetate

NECA

5′-(N-ethylcarbamoyl)adenosine

TLC

thin-layer chromatography

mp

melting point

EtOAc

ethyl acetate

IR

infrared spectra

NMR

nuclear magnetic resonance

CDCl₃

deuterated chloroform

CHAPS

3-[(3-chloroamidopropyl)-dimethylammino]-1-propanesulfonate

[¹²⁵I]AB-MECA

[¹²⁵I]1-[6-[[(4-amino-3-iodophenyl)methyl]amino]-9H-purin-9-yl]-1-deoxy-N-methyl-β-d-ribofuranuronamide

MRE3008-F20

5-[[(4-methoxyphenyl)amino] carbonyl]amino-8-propyl-2-(2-furyl)-pyrazolo[4,3-e]1,2,4-triazolo[1,5-c]pyrimidine

SCH58261

5-amino-2-(2-furyl)-7-(2-phenylethyl)pyrazolo[4,3-e][1,2,4]-triazolo[1,5-c]pyrimidine