Anilide Derivatives of an 8-Phenylxanthine Carboxylic Congener Are Highly Potent and Selective Antagonists at Human A_2B Adenosine Receptors

Abstract

No highly selective antagonists of the A_2B adenosine receptor (AR) have been reported; however such antagonists have therapeutic potential as antiasthmatic agents. Here we report the synthesis of potent and selective A_2B receptor antagonists. The structure–activity relationships (SAR) of 8-phenyl-1,3-di-(n-propyl)xanthine derivatives in binding to recombinant human A_2B ARs in HEK-293 cells (HEK-A_2B) and at other AR subtypes were explored. Various amide derivatives of 8-[4-[[carboxymethyl]oxy]phenyl]-1,3-di-(n-propyl)xanthine, 4a, were synthesized. A comparison of aryl, alkyl, and aralkyl amides demonstrated that simple anilides, particularly those substituted in the para-position with electron-withdrawing groups, such as nitro, cyano, and acetyl, bind selectively to human A_2B receptors in the range of 1–3 nM. The unsubstituted anilide 12 had a K_i value at A_2B receptors of 1.48 nM but was only moderately selective versus human A₁/A_2A receptors and nonselective versus rat A₁ receptors. Highly potent and selective A_2B antagonists were a p-aminoacetophenone derivative 20 (K_i value 1.39 nM) and a p-cyanoanilide 27 (K_i value 1.97 nM). Compound 27 was 400-, 245-, and 123-fold selective for human A_2B receptors versus human A₁/A_2A/A₃ receptors, respectively, and 8.5- and 310-fold selective versus rat A₁/A_2A receptors, respectively. Substitution of the 1,3-dipropyl groups with 1,3-diethyl offered no disadvantage for selectivity, and high affinities at A_2B receptors were maintained. Substitution of the p-carboxymethyloxy group of 4a and its amides with acrylic acid decreased affinity at A_2B receptors while increasing affinity at A₁ receptors. 1,3-Di-(cyclohexylmethyl) groups greatly reduced affinity at ARs, although the p-carboxymethyloxy derivative 9 was moderately selective for A_2B receptors. Several selective A_2B antagonists inhibited NECA-stimulated calcium mobilization in HEK-A_2B cells.

Introduction

The alkylxanthine theophylline, 1 (Figure 1), a weak nonselective adenosine antagonist,¹ is effective when used therapeutically for the treatment of asthma, but its use is associated with unpleasant side effects, such as insomnia and diuresis.² In recent years, use of theophylline as a bronchodilator for relief of asthma has been supplanted by drugs of other classes, i.e. selective β₂-adrenergic agonists, corticosteroids, and recently leukotriene antagonists.³ All of these compounds have limitations, so the development of a theophylline-like drug with reduced side effects is still desirable. The mechanism of action of theophylline in asthma has long been the subject of controversy. It is recognized that at therapeutically relevant doses, theophylline and its closely related analogue caffeine block endogenous adenosine acting as a local modulator in the brain and other organs. Adenosine activates four subtypes of G protein-coupled adenosine receptors (ARs): A₁/A_2A/A_2B/A₃.⁴ In comparison to the other known actions of theophylline, e.g. inhibition of phosphodiesterases, it is more potent in antagonism of ARs. Although lung tissue from asthmatic patients is hyperresponsive in adenosine-induced contraction,⁵ the adenosine antagonist properties of theophylline had been doubted as an explanation of the therapeutic properties,⁶ until relatively recently. One reason was that enprofylline, 3-propylxanthine, 3, formerly used clinically as an antiasthmatic in Europe, is much weaker than theophylline as an antagonist of two AR subtypes previously well-defined pharmacologically, i.e. A₁/A_2A ARs. With the recent focus on A_2B ARs,⁷ it has been noted that therapeutic concentrations of enprofylline block human A_2B receptors and proposed that antagonists selective for this subtype may have potential use as antiasthmatic agents.^8,9 Enprofylline has a K_i value of 7 μM and is somewhat selective in binding to human A_2B ARs.^9,10 A_2B ARs are expressed in some mast cells, such as canine BR mastocytoma cells, in which they appear to be responsible for triggering acute Ca²⁺ mobilization and degranulation.^11,12 A_2B ARs also trigger Ca²⁺ mobilization¹⁰ and participate in a delayed IL8 release from human HMC-1 mast cells.¹³ Other functions associated with the A_2B AR are the control of cell growth and gene expression,¹⁴ endothelial-dependent vasodilation,¹⁵ and fluid secretion from intestinal epithelia.¹⁶ Adenosine acting through A_2B ARs can stimulate chloride permeability in cells expressing the cystic fibrosis transport regulator.¹⁷

Figure 1.

Structures of various xanthines that act as antagonists at A_2B receptors.

Although AR subtype-selective probes are available for the A₁/A_2A/A₃ ARs,¹⁸ only few weakly selective antagonists^19,20 and no selective agonists^21,22 are known for the A_2B receptor. Recently radioligand binding assays^9,10,23 have been reported which will aid in the identification of selective antagonists. Although there are several classes of non-xanthine adenosine antagonists^24–26 that have been found to be potent and slightly selective A_2B receptor antagonists, we have selected xanthines as a suitable lead for the development of structure–activity relationships (SAR). Among xanthines, an 8-phenyl group is associated with increased affinity at A_2B receptors.^7,27 The 8-phenyl analogue of theophylline, 2, displayed a 22-fold enhancement of binding affinity at A_2B receptors.¹⁹ Leads for achieving moderate selectivity (at least 20-fold versus A₁/A_2A/A₃ ARs) have recently been identified among derivatives of 8-[4-[(carboxymethyl)oxy]phenyl]-1,3-dipropylxanthine,^19,20,27 4a. Compound 4b (MRS 1204, N-hydroxysuccinimide ester of 4a) displayed approximately 20-fold selectivity for human A_2B receptors versus A₁/A_2A/A₃ ARs.¹⁹ A 1,2-dimethylmaleimide derivative, 4c (MRS 1595), bound to human A_2B receptors with a K_i of 19 nM and was selective versus human A₁/A_2A/A₃ receptors by 160-, 100-, and 35-fold, respectively.²⁰ The A_2B receptor selectivity of enprofylline was lost in its 8-aryl-substituted analogues.²⁰ Aryl amide derivatives were previously reported to be highly potent antagonists at A_2A receptors in human platelets.²⁸ In the present study, these and additional amide derivatives of 4a were synthesized, to identify novel, selective A_2B receptor antagonists.

Results

The structures of the xanthine derivatives 4a-40 tested for affinity in radioligand binding assays at ARs are shown in Tables 1–3. Most of the xanthines are derivatives of the carboxylic congener 4a²⁷ in which the carboxylic acid group has been condensed to various amines through amide coupling reactions shown in Scheme 1. This approach was taken based on the high potency in the A_2B receptor binding assay of an N-hydroxysuccinimide ester 4b (K_i value of 9.75 nM)¹⁹ and related acyl hydrazides,²⁰ such as 4c.

Table 1.

Affinities or Antagonistic Activities of Xanthine Carboxylic Acid Derivatives in Radioligand Binding Assays at A₁/A_2A/A_2B/A₃ Receptors

compd	R1	X	R2	Ki (nM)				hA1/hA2B	hA2A/hA2B	hA3/hA2B
				rA1a	rA2Ab	hA2Bc	hA3d
4a	n-Pr	OCH2	OH	58	2200	40 ± 4	3910 ± 2140	4.4	15	98
				175 ± 57 (h)	595 ± 128 (h)		75700 ± 6500 (r)
4b	n-Pr	OCH2	O-succinimide	153	127	9.75 ± 4.80	227
4c	n-Pr	OCH2	NHN-dimethylmaleyl	11.1 ± 2.4	126 ± 41	26.6 ± 4.0	670 ± 154	110	74	25
				3030 ± 1110 (h)	1970 ± 550 (h)
4d	n-Pr	OCH2	NH(CH2)2NH2	1.2	63	7.75 ± 0.14	25.6 ± 5.0	0.9	2.4	3.3
				6.82 ± 1.57 (h)	18.4 ± 0.03 (h)
5	allyl	OCH2	OH	756 ± 147	4290 ± 570	141 ± 29 (h)	816 ± 91	12	17	5.8
				1660 ± 580 (h)	2370 ± 290 (h)		173000 ± 18000 (r)
6	n-butyl	OCH2	OH	43.1 ± 9.9	874 ± 107	48.0 ± 16.9	90.3 ± 14.2	3.1	53	1.9
				149 ± 83 (h)	2540 ± 1250 (h)		27500 ± 2500 (r)
7	Bn	OCH2	OH	679 ± 190	25 ± 3% (10−4)e	1760 ± 110
8	n-Pr	CH=CH	OH	15	800	60 ± 2	30 ± 14	2.3	3.2	0.5
				140 ± 3 (h)	190 ± 71 (h)		15000 ± 1700 (r)
9	c-HexMe	CH=CH	OH	602 ± 24	<10% (10−5)e	199 ± 52	922 ± 399	25	7.6	4.6
				4890 ± 530 (h)	1518 ± 980 (h)
10	Bn	CH=CH	OH	201 ± 23	4450 ± 1230	469 ± 23

^aDisplacement of specific binding in rat brain membranes ([³H]R-PIA) or recombinant human A₁ receptors in HEK-293 cells ([¹²⁵I]IABA), expressed as K_i ±SEM (n = 3–5).

^bDisplacement of specific binding in rat striatal membranes ([³H]CGS21680) or recombinant human A_2A receptors in HEK-293 cells ([¹²⁵I]iodo-ZM241385), expressed as K_i ± SEM (n = 3–5).

^cDisplacement of specific [³H]ZM241385 or [¹²⁵I]IABOPX binding at human A_2B receptors expressed in HEK-293 cells, in membranes, expressed as K_i ± SEM (n = 3–4).

^dDisplacement of specific [¹²⁵I]IAB-MECA or [¹²⁵]IABA binding at human A₃ receptors expressed in HEK cells, in membranes, expressed as K_i ± SEM (n = 3–4).

^e% Displacement of specific binding at the designated molar concentration.

Table 3.

Affinities or Antagonistic Activities of Miscellaneous Xanthine Derivatives in Radioligand Binding Assays^a at A₁/A_2A/A_2B/A₃ Receptors

				Ki (nM)
compd	R1	X	R2	rA1	rA2A	hA2B	hA3
34	n-Pr	CH=CH	NHN-dimethylmaleyl	3.94 ± 1.20	406 ± 105	16.7 ± 3.0	31.0 ± 3.1
				105 ± 5 (h)	223 ± 55 (h)
35	n-Pr	CH=CH	NH-Ph(2-COCH3)	7.67 ± 2.20	143 ± 50	3.65 ± 0.98	121 ± 138
36	c-HexMe	CH=CH	NH-Ph(2-COCH3)	b (10−5)	b (10−5)	b (10−5)
37	Bn	CH=CH	NH-Ph(2-COCH3)	34300	b (10−5)	b (10−5)
38	Bn	OCH2	NH-Ph(2-COCH3)	b (10−5)	b (10−5)	b (10−5)
39	Et	OCH2	NH-Ph(4-CH3)	34.9 ± 0.3	71.1 ± 7.7	1.78 ± 0.43	b (10−6)
40	Et	OCH2	NH-Ph(4-CH2CONH(CH2)2NH2)	65.0 ± 15.4	1370 ± 490	15.2 ± 6.8

^aThe methods of each binding assay are shown in Table 1.

^b<10% Displacement of specific binding at the designated molar concentration.

Scheme 1.

Synthesis of Amide Derivatives of Xanthine Carboxylic Acid Congeners as Potentially Selective A_2B AR Antagonists^a

^a Reagents: (1) EDAC, DMAP, amines in DMF/CH₂Cl₂; (2) BOPCl, triethylamine, amines in CH₂Cl₂; or (3) SOCl₂, amines in pyridine/CH₂Cl₂.

Besides 4a, various other analogues of xanthine carboxylic acid congeners, 7-10, were synthesized according to established synthetic procedures of xanthines,²⁹ and their corresponding amide or acyl hydrazide conjugates, 34-38, were also prepared and tested for comparison. The yields and chemical characterization of all new compounds are reported in Table 4.

Table 4.

Yields and Chemical Characterization of Xanthine Derivatives

compd	% yield	mp (°C)	MS	formula	anal.
7	13	>310	EI: 482	C27H22N4O5	HRMSa
10	40	>310	FAB: 479	C28H22N4O4	C,H,N
12	71	301–302	CI: 462	C25H27N5O4	C,H,N
13	41	268	CI: 476	C26H29N5O4	C,H,N
14	46	269–270	EI: 551	C32H33N5O4	C,H,N
15	55	230	EI: 565	C33H35N5O4	C,H,N
16	49	215	FAB: 476	C26H29N5O4	C,H,N
17	13	225	CI: 558	C27H35N5O8	C,H,N
18	68	294	EI: 503	C27H29N5O5	C,H,N
19	29	269–270	EI: 503	C27H29N5O5	HRMSa
20	29	309–310	EI: 503	C27H29N5O5•0.23H2O	C,H,N
21	56	>310	CI: 520	C27H29N5O6	C,H,N
22	19	>310	CI: 505	C26H28N6O5	C,H,N
23	45	>310	FAB: 519	C27H30N6O5•1.8CH2Cl2	C,H,N
24	51	>310	FAB: 506	C26H27N5O6•0.60CH2Cl2	C,H,N
27	44	>310	CI: 487	C26H26N6O4	C,H,N
28	31	307	CI: 507	C25H26N6O6•0.43CH3OH	C,H,N
29	44	>310	CI: 530	C26H26F3N5O4•0.26CH3OH	C,H,N
30	48	298	CI: 480	C25H26FN5O4	C,H,N
31	31	309	CI: 496	C25H26ClN5O4•0.26(CH3)2CO	C,H,N
32	36	>310	CI: 540	C25H26BrN5O4	C,H,N
33	13	>310	CI: 588	C25H26IN5O4•0.60CH3OH	C,H,N
4c	79	>310	FAB: 509	C25H28N6O6	C,H,N
34	49	302–303	EI: 504	C26H28N6O5	C,H,N
35	42	281–283	CI: 500	C28H29N5O4•0.27MeOH	C,H,N
36	33	305	FAB: 608	C36H41N5O4	HRMSa
37	76	308–309	CI: 596	C36H29N5O4•0.60H2O	C,H,N
38	18	284	EI: 599	C35H29N5O5	HRMSa

^aHigh-resolution mass in EI or FAB⁺ mode (m/z) determined to be within acceptable limits: 7: calcd, 482.1590; found, 482.1597; 19: calcd, 503.2169; found, 503.2169; 36: calcd, 608.3237; found, 608.3251; 38: calcd, 599.2169; found, 599.2171. HPLC demonstrated >95% purity, retention times (mobile phase, min): 7: A, 9.94; B, 10.14; 19: A, 14.76; B, 17.52; 36: A, 23.97; B, 24.79; 38: A, 17.58; B, 24.20. Mobile phases consisted of: (A) 0.1 M TEAA (pH = 5.0)/CH₃CN, 80:20 to 20:80, in 30 min; and (B) H₂O/MeOH, 80:20 to 20:80, in 30 min, both with a flow rate of 1 mL/min.

The potency of the xanthine derivatives at human A_2B receptors was evaluated using two binding assays (Tables 1–3) and a functional assay (Figure 2). K_i values of xanthine derivatives were determined in displacement of binding of two nonselective radioligands: [³H]ZM241385 (4-(2-[7-amino-2-furyl[1,2,4]triazolo[2,3-a]-[1,3,5]triazin-5-yl]aminoethyl)phenol)²³ and [¹²⁵I]IABOPX (3-(4-amino-3-iodobenzyl)-8-phenyloxyacetate-1-propylxanthine),¹⁰ at human A_2B receptors stably expressed in HEK-293 cell membranes.¹⁰ Results with these two radioligands were identical. To determine selectivity, the xanthines were evaluated using standard binding assays at A₁/A_2A/A₃ receptors. The initial screening utilized rat brain A₁/A_2A receptors (with radioligands [³H]R-PIA and [³H]CGS21680), and selected compounds were examined at the recombinant human subtypes, using [³H]CPX (8-cyclopentyl-1,3-dipropylxanthine) (A₁)³⁰ and [¹²⁵I]ZM241385 (A_2A).³¹ Affinity at cloned human A₃ receptors expressed in HEK-293 cells was determined using [¹²⁵I]IABA (N⁶-(4-amino-3-[¹²⁵I]iodobenzyl)-adenosine)³² or [¹²⁵I]IAB-MECA (N⁶-(4-amino-3-iodobenzyl)adenosine-5′-N-methyluronamide).³³

Figure 2.

Inhibition by several selective A_2B AR antagonists of NECA-stimulated calcium mobilization in HEK-A_2B cells. Cells were loaded with Indo-1 for 1 h: (A) calcium mobilization in response to 10 nM and 1 μM NECA added at the arrow; (B) calcium mobilization in response to 10 nM NECA added at the arrow in cells pretreated for 2 min with 1% DMSO (control) or with 100 nM of the indicated antagonists. The results are typical of replicate experiments.

A series of xanthine carboxylic acid derivatives 4a-10 allowed comparison of the effects of substitutions at the 1- and 3-positions and variation of the linkage between the carboxylate group and the 8-phenyl ring. Among 8-(4-carboxymethyloxyphenyl) derivatives differing only in the 1- and 3-substitutents, 4a-7, affinity at A_2B receptors decreased in the order: 1,3-dipropyl ⩾ 1,3-di-n-butyl > 1,3-diallyl > 1,3-dibenzyl. The diallyl derivative was more A_2B receptor-selective but less potent than the dipropyl derivative. Thus, 4a was selected for further derivatization as amides.

8-(4-Phenylacrylic) acid derivatives 8-10 tended to be more potent at A₁ receptors and less potent at A_2B receptors than the 8-(4-carboxymethyloxyphenyl) derivatives. The 1,3-dicyclohexylmethyl derivative 9 was 25-fold selective for A_2B receptors. A primary carbox-amide 11 was more potent than the carboxylic acid 4a at A₁ (3-fold) and A_2A (29-fold) receptors and equipotent at A_2B receptors.

The AR affinities of aryl, 12 and 18-33, alkyl, 17, and aralkyl, 13-16, amides of 4a were compared. A benzyl amide 13 and simple anilides had the highest affinity of binding, in the nanomolar range, to human A_2B receptors. Selectivities for the human A_2B versus rat A₁ receptors ranged from 1-fold (30) to 27-fold (20), while comparisons within the same species (human) generally led to greater selectivities. Anilides substituted in the para-position with groups such as nitro, cyano, and acetyl displayed the highest selectivity. An N-methylanilide of 4a, 16, was 40- and 92-fold selective for human A_2B receptors versus rat A₁/A_2A receptors; thus the N-methylation reduced affinity by 3.7-fold but increased selectivity. An ortho-substituted acetophenone 18 was 120-, 160-, and 23-fold selective for human A_2B receptors versus human A₁/A_2A/A₃ receptors and 10- and 160-fold selective versus rat A₁/A_2A receptors. The para-substituted acetophenone 20 was more potent at A_2B receptors than the corresponding ortho- and meta-isomers. Other highly potent and moderately selective A_2B antagonists were a p-trifluoromethyl derivative, 29 (K_i value 2.14 nM), and a p-cyanoanilide, 27 (K_i value 1.97 nM), which was highly selective versus the other human subtypes but only 8.5-fold selective versus rat A₁ receptors. A p-nitro derivative, 28, bound to human A_2B receptors with a K_i of 1.52 nM but was only 35-fold selective versus human A₁ receptors. A p-iodo derivative, 33 (K_i value 2.13 nM), was 140-, 2400-, and 600-fold selective for human A_2B receptors versus human A₁/A_2A/A₃ receptors. A p-toluide of 4a, 25, was reported previously²⁷ and displayed a K_i value at human A_2B receptors of 1.88 nM.

The introduction of the dimethylmaleimido group²⁰ in 34 and in the p-acryloyl derivative 35 resulted in moderate selectivity for A_2B receptors versus other human but not rat ARs.

Introduction of a p-acrylic acid group in an anilide derivative, e.g. 36 versus 18, decreased selectivity. Bulky 1,3-di(cyclohexylmethyl) groups or 1,3-dibenzyl groups in anilide derivatives 37 and 38, respectively, greatly diminished binding to ARs. Substitution of the 1,3-dipropyl groups with ethyl, as in 40 and 41, offered no disadvantage for selectivity, and high affinities were maintained.

The functional effects of several selective A_2B antagonists in inhibiting the effects of NECA in HEK-A_2B cells were examined (Figure 2). Several selective A_2B AR antagonists at 100 nM nearly completely inhibited NECA-stimulated calcium mobilization. In comparison, XAC(8-[4-[[[[(2-aminoethyl)amino]carbonyl]methyl]oxy]-phenyl]-1,3-dipropylxanthine), which has a K_i value of 12.3 nM in binding to human A_2B ARs,²¹ inhibited the effect by roughly one-half. Thus, the potency of the xanthines in the functional assay was parallel to results from the binding assay.

Discussion

We have identified amide derivatives of 4a, such as 20, as adenosine antagonists which are potent and selective for human A_2B receptors. High affinity for the receptor has been achieved through the formation of anilides and a benzyl amide of 4a. To increase selectivity, substitution of the aryl ring in the anilide derivatives was carried out. It appears that electron-withdrawing substituents at the 2- or 4-position are best suited for A_2B receptor selectivity. Further SAR studies are in progress to enhance the pharmacological profile of these xanthine derivatives as A_2B receptor antagonists. Although this study reveals xanthines having high selectivity for the human subtypes, there is still a need for improving A_2B receptor selectivity in the rat.

There is also a need to enhance water solubility in potent and selective antagonists such as 18, 27, and 33, which are highly hydrophobic. An attempt to introduce the charged p-carboxylate group, in 24, resulted in lower affinity, although selectivity was maintained.

High-affinity compounds, such as 18 and 27, may be the first selective pharmacological probes needed to investigate the physiological role of this AR subtype and in tritiated form may be suitable as selective radioligands for the A_2B receptor. There is evidence that both A_2B/A₃ ARs may play a role in asthma.¹² The A₃ AR mediates the degranulation of rat RBL mast-like cells³⁴ and is present in high density in human blood eosinophils.³⁵ The availability of antagonists selective for the A_2B receptor should provide an opportunity to explore the importance of these two receptor subtypes in asthma.

Materials and Methods

Materials.

Compounds 4a, 11, 25, and 26 were synthesized as reported.²⁷ Compounds 5-7 and 10 were synthesized as reported.²⁹ Compounds 39 and 40 were synthesized as reported.²⁸ Compounds 8 and 9 were obtained from Dr. Susan Daluge, Glaxo, Research Triangle, NC. R-PIA, and 2-chloroadenosine were purchased from Research Biochemicals International (Natick, MA). All other agents were purchased from Aldrich (St. Louis, MO).

Synthesis.

Proton nuclear magnetic resonance spectroscopy was performed on a Varian GEMINI-300 spectrometer and spectra were taken in DMSO-d₆ or CDCl₃. Unless noted, chemical shifts are expressed as ppm downfield from tetramethylsilane or relative ppm from DMSO (2.5 ppm). Chemical ionization (CI) mass spectrometry was performed with a Finnigan 4600 mass spectrometer and electron impact (EI) mass spectrometry with a VG7070F mass spectrometer at 6 kV for high-resolution mass. FAB (fast atom bombardment) mass spectrometry was performed with a JEOL SX102 spectrometer using 6-kV Xe atoms. Elemental analysis (±0.4% acceptable) was performed by Atlantic Microlab Inc. (Norcross, GA). All melting points were determined with a Unimelt capillary melting point apparatus (Arthur H. Thomas Co., PA) and were uncorrected. All xanthine derivatives were homogeneous as judged using TLC (MK6F silica, 0.25 mm, glass backed; Whatman Inc., Clifton, NJ). Where needed, evaluation of purity was done on a Hewlett-Packard 1090 HPLC system using an OD-5–60 C18 analytical column (150 mm × 4.6 mm; Separation Methods Technologies, Inc., Newark, DE) in two different linear gradient solvent systems, at a flow rate of 1 mL/min. One solvent system (A) was 0.1 M TEAA (pH = 5.0)/CH₃CN, 80:20 to 20:80, in 30 min, and the other (B) was H₂O/MeOH, 80:20 to 20:80, in 30 min. Peaks were detected by UV absorption using a diode array detector.

General Procedure for the Preparation of Amide Derivatives of 4a and 7-10. Method A (Carbodiimide).

A solution of a 4a analogue (0.0517 mmol), the desired amine compound (0.103 mmol), EDAC (20 mg, 0.103 mmol), and DMAP (4 mg, 0.032 mmol) in 2 mL of anhydrous DMF/CH₂Cl₂ (1:1 v/v) was stirred at room temperature for 24 h. The mixture was evaporated to dryness under reduced pressure, and the residue was purified by preparative silica gel TLC (CHCl₃: MeOH = 20:1) and crystallization in MeOH/ether or MeOH/CH₂Cl₂ to afford the desired compounds (12-14, 18, 36).

Method B (BOP-Cl).

A solution of a 4a analogue (0.0517 mmol), the desired amine compound (0.103 mmol), BOP-Cl (14 mg, 0.0517 mmol), and triethylamine (20 μL, 0.206 mmol) in 2 mL of anhydrous CH₂Cl₂ was stirred at room temperature for 24 h. The mixture was treated with the same procedure as method A for purification of the desired compounds (15, 17, 19, 20, 38).

Method C (Acid Chloride).

A solution of a 4a analogue (0.0517 mmol) in 1 mL of thionyl chloride was stirred at 70 °C for 4 h and the excess thionyl chloride was removed by nitrogen stream. To the residue was added a solution of the desired amine compound (0.103 mmol) in 1 mL of anhydrous pyridine and 1 mL of anhydrous CH₂Cl₂. The mixture was stirred at room temperature for 24 h, then subjected to the same procedure as method A for purification of the desired compounds (16, 21, 22, 27-35, 37).

8-[4-[(Carboxymethyl)oxy]phenyl]-1,3-dibenzylxanthine (7):

¹H NMR (DMSO-d₆) 4.23 (s, 2H, −OCH₂−), 5.10 (s, 2H, −NCH₂−), 5.23 (s, 2H, −NCH₂−), 6.88 (d, 2H, J = 8.8 Hz, Ar), 7.22–7.41 (m, 10H, 2×-Ph), 8.01 (d, 2H, J = 8.8 Hz, Ar).

8-(4-(2-Carboxy-trans-vinyl)phenyl)-1,3-dibenzylxanthine (10):

¹H NMR (DMSO-d₆) 5.12 (s, 2H, −NCH₂−), 5.26 (s, 2H, −NCH₂−), 6.63 (d, 1H, J = 15.6 Hz, −CH=), 7.227.43 (m, 10H, 2×-Ph), 7.63 (d, 1H, J = 15.6 Hz, −CH=), 7.84 (d, 2H, J = 8.8 Hz, Ar), 8.17 (d, 2H, J = 8.8 Hz, Ar).

8-[4-[(Phenylcarbamoylmethyl)oxy]phenyl]-1,3-di(n-propyl)xanthine (12):

¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.58 and 1.74 (2m, 4H, 2×-CH₂−), 3.87 and 4.02 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.80 (s, 2H, −OCH₂−), 7.06–7.12 (m, 1H, −Ph), 7.14 (d, 2H, J = 8.8 Hz, Ar), 7.33 (t, 2H, J = 7.8 Hz, −Ph), 7.64 (d, 2H, J = 7.8 Hz, −Ph), 8.09 (d, 2H, J = 8.8 Hz, Ar), 10.13 (s, 1H, −NH).

8-[4-[((4-Acetylphenyl)carbamoylmethyl)oxy]phenyl]-1,3-di(n-propyl)xanthine (20):

¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.58 and 1.74 (2m, 4H, 2×-CH₂−), 2.54 (s, 3H, −COCH₃), 3.87 and 4.02 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.85 (s, 2H, −OCH₂−), 7.15 (d, 2H, J = 8.8 Hz, Ar), 7.79 (d, 2H, J = 7.8 Hz, Ar), 7.96 (d, 2H, J = 7.8 Hz, Ar), 8.09 (d, 2H, J = 8.8 Hz, Ar), 10.48 (s, 1H, −NH−).

8-[4-[((4-Methylcarbamoyl)phenylcarbamoylmethyl)-oxy]phenyl]-1,3-di(n-propyl)xanthine (23).

A solution of 20 mg of 21 (0.0358 mmol) in 1 mL of 40% aqueous methylamine was stirred at room temperature for 1 h. The mixture was evaporated to dryness under reduced pressure, and the residue was purified by preparative silica gel TLC (CHCl₃: MeOH = 20:1) and crystallization in MeOH/CH₂Cl₂ to give 9 mg of 23: ¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.58 and 1.73 (2m, 4H, 2×-CH₂−), 2.76 (s, 3H, −NHCH₃), 3.86 and 4.01 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.82 (s, 2H, −OCH₂−), 7.14 (d, 2H, J = 8.8 Hz, Ar), 7.71 (d, 2H, J = 7.8 Hz, Ar), 7.81 (d, 2H, J = 7.8 Hz, Ar), 8.09 (d, 2H, J = 8.8 Hz, Ar), 8.33 (m, 1H, −NHCH₃), 10.34 (s, 1H, −NH−).

8-[4-[((4-Carboxyphenyl)carbamoylmethyl)oxy]phenyl]-1,3-di(n-propyl)xanthine (24).

A suspension of 20 mg of 21 (0.0385 mmol) in 1 mL of 1 N NaOH solution was stirred for 2 h to turn to a clear solution. The mixture was neutralized by adding 1 mL of 1 N HCl. The precipitate was collected by filtration and purified by low-pressure C18 column chromatography using linear gradient elution of 1 M triethylammo-niumacetate buffer (pH = 7.0) and CH₃CN (90/10 to 40/60) to give 10 mg of 24: ¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.58 and 1.74 (2m, 4H, 2×-CH₂−), 3.87 and 4.01 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.84 (s, 2H, −OCH₂−), 7.14 (d, 2H, J = 8.8 Hz, Ar), 7.77 (d, 2H, J = 8.8 Hz, Ar), 7.92 (d, 2H, J = 8.8 Hz, Ar), 8.09 (d, 2H, J = 8.8 Hz, Ar), 10.45 (s, 1H, −NH−).

8-[4-[((4-Cyanophenyl)carbamoylmethyl)oxy]phenyl]-1,3-di(n-propyl)xanthine (27):

¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.58 and 1.74 (2m, 4H, 2×-CH₂−), 3.86 and 4.01 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.85 (s, 2H, −OCH₂−), 7.13 (d, 2H, J = 8.8 Hz, Ar), 7.80 (d, 2H, J = 7.8 Hz, Ar), 7.84 (d, 2H, J = 7.8 Hz, Ar), 8.09 (d, 2H, J = 8.8 Hz, Ar), 10.58 (s, 1H, −NH−).

8-(4-(2-Carboxy-trans-vinyl)phenyl)-1,3-di(n-propyl)xanthine N′,N′-(1,2-dimethylmaleyl)hydrizide (34):

¹H NMR (CDCl₃) 1.01 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.72 and 1.89 (2m, 4H, 2×-CH₂−), 2.05 (s, 6H, 2×-CH₃), 4.02 and 4.17 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 6.67 (d, 1H, J = 15.6 Hz, −CH=), 7.63 (d, 2H, J = 8.8 Hz, Ar), 7.74 (d, 1H, J = 15.6 Hz, −CH=), 8.09 (d, 2H, J = 8.8 Hz, Ar), 9.43 (s, 1H, −NH−).

8-[4-[2-(2-Acetylphenyl)carbamoyl-trans-vinyl]phenyl]-1,3-di(n-propyl)xanthine (35):

¹H NMR (DMSO-d₆) 0.89 (2t, 6H, J = 7.8 Hz, 2×-CH₃), 1.59 and 1.76 (2m, 4H, 2×-CH₂−), 3.88 and 4.04 (2t, 4H, J = 6.8 Hz, 2×-NCH₂−), 4.79 (d, 3H, J = 4.9 Hz, −COCH₃), 6.93 (d, 1H, J = 15.6 Hz, −CH=), 7.51 (d, 1H, J = 15.6 Hz, −CH=), 7.57 (t, 1H, J = 7.8 Hz, Ar), 7.69 (t, 1H, J = 7.8 Hz, Ar), 7.75 (d, 2H, J = 7.8 Hz, Ar), 8.03 (d, 2H, J = 7.8 Hz, Ar), 8.18 (d, 2H, J = 7.8 Hz, Ar), 8.53 (t, 1H, J = 5.8 Hz, −NH−).

8-[4-[2-(2-Acetylphenyl)carbamoyl-trans-vinyl]phenyl]-1,3-dibenzylxanthine (37):

¹H NMR (DMSO-d₆) 4.79 (d, 3H, J = 5.8 Hz, −COCH₃), 5.13 (s, 2H, −NCH₂−), 5.27 (s, 2H, −NCH₂−), 6.93 (d, 1H, J = 15.6 Hz, −CH=), 7.24–7.41 (m, 10H, 2×-Ph), 7.46 (d, 1H, J = 15.6 Hz, −CH=), 7.57 (t, 1H, J = 8.8 Hz, Ar), 7.69 (t, 1H, J = 7.8 Hz, Ar), 7.75 (d, 2H, J = 8.8 Hz, Ar), 8.03 (d, 2H, J = 7.8 Hz, Ar), 8.20 (d, 2H, J = 7.8 Hz, Ar), 8.54 (t, 1H, J = 5.8 Hz, −NH−).

8-[4-[[(2-Acetylphenyl)carbamoylmethyl]oxy]phenyl]-1,3-dibenzylxanthine (38):

¹H NMR (DMSO-d₆) 4.68 (s, 3H, −COCH₃), 4.71 (s, 2H, −OCH₂−), 5.12 (s, 2H, −NCH₂−), 5.26 (s, 2H, −NCH₂−), 7.15 (d, 2H, J = 8.8 Hz, Ar), 7.23–7.42 (m, 10H, 2×-Ph), 7.55 (t, 1H, J = 7.8 Hz, Ar), 7.68 (t, 1H, J = 7.8 Hz, Ar), 8.02 (d, 2H, J = 7.8 Hz, Ar), 8.11 (d, 2H, J = 8.8 Hz, Ar), 8.48 (t, 1H, J = 4.8 Hz, −NH−).

Pharmacology.

The human A_2B receptor cDNA was subcloned into the expression plasmid pDoubleTrouble.³⁶ The plasmid was amplified in competent JM109 cells and plasmid DNA isolated using Wizard Megaprep columns (Promega Corp., Madison, WI). A_2B ARs were introduced into HEK-293 cells by means of Lipofectin.³⁷

Cell Culture.

Transfected HEK cells were grown under 5% CO₂/95% O₂ humidified atmosphere at a temperature of 37 °C. Colonies were selected by growth of cells in 0.6 mg/mL G418. Transfected cells were maintained in DMEM supplemented with Hams F12 nutrient mixture (1/1), 10% newborn calf serum, 2 mM glutamine, and containing 50 IU/mL penicillin, 50 μg/mL streptomycin, and 0.2 mg/mL Geneticin (G418, Boehringer Mannheim). Cells were cultured in 10- cm diameter round plates and subcultured when grown confluent (approximately after 72 h).

Radioligand Binding Studies. At A_2B receptors:

Confluent monolayers of HEK-A_2B cells were washed with PBS followed by ice-cold buffer A (10 mM HEPES, 10 mM EDTA, pH 7.4) with protease inhibitors (10 mg/mL benzamidine, 100 mM phenylmethanesulfonyl fluoride, and 2 mg/mL of each aprotinin, pepstatin, and leupeptin). The cells were homogenized in a polytron (Brinkmann) for 20 s and centrifuged at 30000g and the pellets washed twice with buffer HE (10 mM HEPES, 1 mM EDTA, pH 7.4 with protease inhibitors). The final pellet was resuspended in buffer HE, supplemented with 10% sucrose, and frozen in aliquots at −80 °C. For binding assays membranes were thawed and diluted 5–10-fold with HE to a final protein concentration of approximately 1 mg/mL. To determine protein concentrations, membranes and bovine serum albumin standards were dissolved in 0.2% NaOH/0.01% SDS and protein was determined using fluorescamine fluorescence.³⁸

To prepare [¹²⁵I]IABOPX, 10 mL of 1 mM ABOPX in methanol/1 M NaOH (20:1) was added to 50 mL of 100 mM phosphate buffer, pH 7.3. One or 2 mCi of Na¹²⁵I was added, followed by 10 mL of 1 mg/mL chloramine-T in water. After a 20-min incubation at room temperature, 50 mL of 10 mg/mL Na-metabisulfite in water was added to quench the reaction. The reaction mixture was applied to a C18 HPLC column, eluting with a mixture of methanol and 4 mM phosphate, pH 6.0. After 5 min at 35% methanol, the methanol concentration was ramped to 100% over 15 min. Unreacted ABOPX eluted in 11–12 min; [¹²⁵I]IABOPX eluted at 18–19 min in a yield of 50–60% with respect to the initial ¹²⁵I.

Saturation binding assays for human A_2B ARs were performed with [³H]ZM214385 (17 Ci/mmol; Tocris Cookson, Bristol, U.K.)²³ or [¹²⁵I]IABOPX (2200 Ci/mmol).

In equilibrium binding assays the ratio of [¹²⁷I/¹²⁵I]ABOPX was 10–20/1. Radioligand binding experiments were performed in triplicate with 20–25 μg of membrane protein in a total volume of 0.1 mL of HE buffer supplemented with 1 U/mL adenosine deaminase and 5 mM MgCl₂. The incubation time was 3 h at 21 °C. Nonspecific binding was measured in the presence of 100 μM NECA. Competition experiments were carried out using 0.6 nM [¹²⁵I]IABOPX. Membranes were filtered on Whatman GF/C filters using a Brandel cell harvester (Gaithersburg, MD) and washed three times during 15–20 s with ice-cold buffer (10 mM Tris, 1 mM MgCl₂, pH 7.4). B_max and K_D values were calculated by Marquardt’s nonlinear least-squares interpolation for single-site binding models.³⁹ K_i values for different compounds were derived from IC₅₀ values as described, assuming a K_D value for [¹²⁵I]IABOPX of 36 nM.⁴⁰ Data from replicate experiments are tabulated as means ± SEM.

At other ARs:

[³H]CPX,³⁰ [¹²⁵I]iodo-ZM241385, and [¹²⁵I]IABA were utilized in radioligand binding assays to membranes derived from HEK-293 cells expressing recombinant human A₁/A_2A/A₃ ARs, respectively. Binding of [³H]R-N⁶-phenylisopropyladenosine⁴¹ ([³H]R-PIA; Amersham, Chicago, IL) to A₁ receptors from rat cerebral cortical membranes and of [³H]CGS21680⁴² (NEN, Boston, MA) to A_2A receptors from rat striatal membranes was performed as described. Adenosine deaminase (3 units/mL) was present during the preparation of the brain membranes, in a preincubation of 30 min at 30 °C, and during the incubation with the radioligands. All nonradioactive compounds were initially dissolved in DMSO and diluted with buffer to the final concentration, where the amount of DMSO never exceeded 2%. 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 with 3 mL of buffer each.

At least six different concentrations of competitor, spanning 3 orders of magnitude adjusted appropriately for the IC₅₀ of each compound, were used. IC₅₀ values, calculated with the nonlinear regression method implemented in Graph-Pad (Prism, San Diego, CA), were converted to apparent K_i values as described.⁴⁰ Hill coefficients of the tested compounds were in the range of 0.8–1.1.

Functional assay:

HEK-A_2B cells from one confluent T75 flask were rinsed with Ca²⁺- and Mg²⁺-free Dulbecco’s phosphate-buffered saline (PBS) and then incubated in Ca²⁺- and Mg²⁺-free HBSS with 0.05% trypsin and 0.53 mM EDTA until the cells detached. The cells were rinsed twice by centrifugation at 250g in PBS and resuspended in 10 mL of HBSS composed of 137 mM NaCl, 5 mM KCl, 0.9 mM MgSO₄, 1.4 mM CaCl₂, 3 mM NaHCO₃, 0.6 mM Na₂HPO₄, 0.4 mM KH₃PO₄, 5.6 mM glucose, and 10 mM HEPES, pH 7.4, and the Ca²⁺-sensitive fluorescent dye Indo-1-AM (5 μM) 37 °C for 60 min. The cells were rinsed once and resuspended in 25 mL dye-free HBSS supplemented with 1 U/mL adenosine deaminase and held at room temperature. Adenosine receptor antagonists prepared as 100× stocks in DMSO or vehicle was added and the cells were transferred to a 37 °C bath for 2 min. Then the cells (1 million in 2 mL) were transferred to a stirred cuvette maintained at 37 °C within an Aminco SLM 8000 spectrofluorometer (SML Instruments, Urbana IL). The ratios of Indo-1 fluorescence obtained at 400 and 485 nm (excitation, 332 nm) was recorded using a slit width of 4 nm. NECA was added after a 100-s equilibration period.

Table 2.

Affinities or Antagonistic Activities of Xanthine Amide Derivatives in Radioligand Binding Assays^a at A₁/A_2A/A_2B/A₃ Receptors

		Ki (nM)
compd	R	rA1	rA2A	hA1	hA2A	hA2B	hA3	hA1/hA2B	hA2A/hA2B	hA3/hA2B
11	NH2	20.0 ± 3.8	76.3 ± 14.0			16.3 ± 4.2
12	NH-Ph	4.22 ± 0.88	45.6 ± 1.4	40.1 ± 5.1	25.8 ± 4.5	1.48 ± 0.63	137 ± 54	27	17	93
13	NH-CH2Ph	5.02 ± 0.55	25.9 ± 7.6	54.7 ± 21.2	23.8 ± 5.71	2.04 ± 0.17	79.2 ± 17.8	27	12	39
14	NH-CH(Ph)2	120 ± 21	20 ± 8% (10−6)			33.7 ± 17.0
15	N(CH2Ph)2	167 ± 49	2750 ± 800	690 ± 98	642 ± 198	9.88 ± 1.05	284 ± 14	70	65	29
16	N(CH3)Ph	218 ± 80	497 ± 250			5.42 ± 1.71
17	N(CH2COOEt)2	26.8 ± 2.4	999 ±144			43.4 ± 8.4
18 b	NH-Ph(2-COCH3)	27.9 ± 1.6	434 ± 129	335 ± 64	431 ± 176	2.74 ± 1.01	61.9 ± 3.4	120	160	23
19	NH-Ph(3-COCH3)	439 ± 111	949 ± 394	234 ± 28	58.9 ± 7.1	4.92 ± 0.55	352 ± 69	48	12	72
20 b	NH-Ph(4-COCH3)	37.6 ± 4.0	548 ± 183	157±8	112 ± 37	1.39 ± 0.30	230 ± 23	110	81	170
21	NH-Ph(4-COOCH3)	38.4 ± 3.9	541 ± 128	225 ± 9	3100 ± 1540	3.93 ± 1.35	363 ± 148	57	790	92
22	NH-Ph(4-CONH2)	10.2 ± 2.5	683 ± 167			7.75 ± 1.11
23	NH-Ph(4-CONHCH3)	24.8 ± 1.8	98.1 ± 49.6			3.34 ± 0.51
24	NH-Ph(4-COOH)	145 ± 28	220 ± 79	161 ± 53	39.0 ± 21.5	16.1 ± 4.7	>5000	10	2.4	>300
25	NH-Ph(4-CH3)	17.5 ± 5.0	126 ± 38			1.88 ± 0.76
26	NH-Ph(4-OH)	5.88 ± 1.06	63.3 ± 20.4			3.71 ± 0.76
27 b	NH-Ph(4-CN)	16.8 ± 3.6	612 ± 287	403 ± 194	503 ± 10.8	1.97 ± 0.31	570 ± 184	210	260	290
28	NH-Ph(4-NO2)	13.1 ± 3.9	1180 ± 360	57.0 ± 3.1	70.0 ± 10.7	1.52 ± 0.24	138 ± 17.1	38	46	91
29	NH-Ph(4-CF3)	44.6 ± 6.5	917 ± 258	61.2 ± 8.2	238 ± 28	2.14 ± 0.47	213 ± 94	29	110	100
30	NH-Ph(4-F)	2.72 ± 0.51	988 ± 518	17.9 ± 4.5	16.6 ± 3.6	2.22 ± 0.19	391± 147	8.1	7.5	176
31	NH-Ph(4-Cl)	6.35 ± 1.47	995 ± 550	49.7 ± 14.2	187 ± 38	2.47 ± 0.71	1870 ± 370	20	400	760
32	NH-Ph(4-Br)	7.46 ± 2.66	221 ± 36	73.5 ± 23.3	1640 ± 660	2.35 ± 0.01	2300 ± 420	31	700	980
33	NH-Ph(4-I)	15.7 ± 4.2	152 ± 47	293 ± 67	5140 ± 540	2.13 ± 0.12	1270 ± 130	140	2400	600

^aThe methods of each binding assay are shown in Table 1.

^b18, MRS 1668; 20, MRS 1706; 27, MRS 1754.