Structure–Activity Relationships of N⁶-Benzyladenosine-5′-uronamides as A₃-Selective Adenosine Agonists^†

Ribose Modifications

The SAR for adenosine derivatives with modifications in the ribose moiety has been found to be similar for A₃ receptors to that for A₁ and A_2a receptors.⁸ There is little toleration of substitution or deletion of the 2′- and 3′-hydroxyl groups, but there is considerable freedom of substitution at the 5′-position. 5′-Deoxy, 5′-thio ether, and 5′-uronamide substitutions are tolerated at A₃ receptors.⁸

NECA, 3, was reported to be among the most potent known agonists at A₃ receptors,⁶ although not A₃ selective. NECA favors both A₁ and A_2a receptors vs A₃ receptors by 1 order of magnitude in affinity. The 5′-uronamido group of NECA was varied (Table 1). The primary carboxamide, 1, was weaker than NECA at A₃ receptors, with roughly the same selectivity ratios. The N-methylamide, 2, was the most favored 5′-uronamide in binding to A₃ receptors. It was found to be nonselective but moderately potent (K_i 75 nM) at A₃ receptors.

Table 1.

Affinities of 5′-Uronamide Derivatives in Radioligand Binding Assays at Rat Brain A₁, A₂, and A₃ Receptors^a-c

			Ki (nM) or % inhibition
compd	R1	R2	Ki(A1)a	Ki(A2)b	Ki(A3)c	A1/A3	A2/A3
1 d	H	NH2	72.6	120	1410 ± 60	0.051	0.085
2 d	Me	NH2	83.6	66.8	72 ± 16	1.2	0.93
3 d	Et	NH2	6.3	10.3	113	0.071	0.091
4 d	cyclopropyl	NH2	6.4	13.4	1600 ± 70	0.0040	0.0084
5	H2NEt	NH2	0% (10−5)	12.5% (10−5)	14700 ± 2540	>1	>1
6	t-BocNHEt	NH2	0% (10−5)	0% (10−5)	18000 ± 4270	>1	>1
7	Me	OH	0% (10−4)	6.2 ± 4.2% (10−4)	6220 ± 1220f	>10	>10
8 e	Et	OH	44% (10−4)	31% (10−4)	5000	>1	>1

Open in a new tab

^a

Displacement of specific [³H]PIA binding, unless noted, in rat brain membranes expressed as K_i ± SEM in nM (n = 3).

^b

Displacement of specific [³H]CGS 21680 binding, unless noted, in rat striatal membranes, expressed as K_i ± SEM in nM (n = 3).

^c

Displacement of specific binding of [¹²⁵I] APNEA¹⁰ or [¹²⁵I]-N⁶-(4-amino-3-iodobenzyl)adenosine-5′-N-methyluronamide²⁸ from membranes of CHO cells stably transfected with the rat A₃-CDNA, expressed as K_i ± SEM in nM (n = 3–5).

^d

Values at A₁ and A₂ receptors are taken from Bruns et al.²⁹ K_i values at A₁ receptors are vs specific binding of [³H]-N⁶-cyclohexyladenosine. K_i values at A₂ receptors are vs specific binding of [³H]NECA in the presence of 50 nM CPA in rat striatal membranes.

^e

Values from ref 8.

^f

Vs[¹²⁵I]-N⁶-(4-amino-3-iodobenzyl)adenosine-5′-N-methyluronamide.

Two 5′-uronamide derivatives of ethylenediamine, 5 and 6, were synthesized (Figure 1) as functionalized congeners.¹⁷ They were found to be much less potent than NECA in receptor affinity, despite a relatively small added substituent (an amino group) in the case of 5. Although weak in receptor affinity, the derivatives were found to be A₃ selective, indicating the possibility of achieving greater potency and selectivity by combining the 5′-N-(2-amino-ethyl)carboxamido modification with other modifications.

Inosine (9-β-d-ribofuranosylhypoxanthine) binds weakly to adenosine receptors, with K_i values at A₁, A_2a, and A₃ receptors in the 20–50 μM range.⁸ The 5′-N-ethyluronamide derivative of inosine, 8, was also found to be a weak ligand with >20-fold selectivity for A₃ receptors.⁸ The N-methyl analogue, 7, was synthesized (Figure 2) and provided an apparent improvement in selectivity, with approximately the same A₃ affinity as the N-ethyl analogue, 8.

Purine Modifications

The 8-position of the purine moiety is not amenable to substitution in A₃ receptor binding.⁸ The 2-position may be substituted without eliminating recognition at A₃ receptors; however, among the few analogues examined there was no indication of enhancement of A₃ selectivity.⁸

Modification of the N1-position (either oxidation to the N-oxide or 1-deaza analogues) is tolerated at A₃ receptors.⁸ The 5′-N-ethyluronamide modification was combined with various changes at the N1-position. NECA N1-oxide, 9, was nonselective (Table 2), but the loss of potency versus NECA was greater for A₁ and A_2a receptors than for A₃ receptors. To introduce larger groups, such as 2- and 4-nitrobenzyl, NECA was alkylated at the N1-position (general reaction shown in Figure 3) resulting in 10 and 11. The resulting potency at A₁ and A_2a receptors was not substantially different from NECA N1-oxide, but the potency at A₃ receptors was increased by 1 order of magnitude. Thus, 10 and 11 displayed intermediate potency and some selectivity for A₃ receptors.

Table 2.

Affinities of N¹-Derivatives of Adenosine and Other Purines in Radioligand Binding Assays at Rat Brain A₁, A₂, and A₃ Receptors^a–c

			Ki (nM) or % inhibition
compd	R1	R3	Ki(A1)a	Ki(A2)b	Ki(A3)c	A1/A3	A2/A3
9 d	Et	O−	154	101	468	0.33	0.22
10 e	Et	CH3C6H4-4-NO2	341 ± 52	190 ± 52	60 ±3	5.7	3.2
11 e	Et	CH2C6H4-2-NO2	196 ± 52	83.5 ± 9.5	27 ±5	7.3	3.1

Open in a new tab

^a

Displacement of specific [³H]PIA binding in rat brain membranes expressed as K_i ± SEM in nM (n = 3–4).

^b

Displacement of specific [³H]CGS 21680 binding in rat striatal membranes, expressed as K_i ± SEM in nM (n = 3).

^c

Displacement of specific binding of [¹²⁵I]APNEA¹⁰ from membranes of CHO cells stably transfected with the rat A₃-cDNA, expressed as K_i ● SEM in nM (n = 3).

^d

Values from ref 8.

^e

Elemental analysis consistent with neutral imino form.

The position of substitution causing the greatest selective enhancement of potency at A₃ receptors was found to be the N⁶-position. We previously reported that among N⁶-aralkyl derivatives of adenosine, the benzyl group favored A₃ receptor selectivity and that this modification was compatible with the 5′-N-ethylcarboxamide.⁸ Indeed, N⁶-benzylNECA, 29 (Table 3), was 14-fold selective for A₃ receptors versus either A₁ or A_2a receptors. In this study we have modified both 5′-and N⁶-positions of 29 to enhance selectivity and potency.

Table 3.

Affinities of N⁶-Benzyladenosine-5′-uronamide Derivatives in Radioligand Binding Assays at Rat Brain A₁, A₂, and A₃ Receptors^a–c

			Ki (nM) or % inhibition
compd	R1	R5	Ki(A1)a	Ki(A2)b	Ki(A3)c	A1/A3	A2/A3
12	H	H	580 ± 99	423 ± 46	246 ± 35	2.6	1.7
13	Me	H	898 ± 124	597 ± 42	16 ± 1	56	37
14	Me	3-Cl	916 ± 121	559 ± 78	21.9 ± 0.3	42	25
15	Me	3-Br	65 ± 2	64 ± 9	1.9 ± 0.3d	34	34
16	Me	3-I	54 ± 5	56 ± 8	1.1 ± 0.3d	49	51
17	Me	3-NO2	735 ± 5	441 ± 45	19 ± 1d	39	23
18	Me	3-NH2	1000 ± 60	794 ± 118	28 ± 13d	36	28
19	Me	3-NHCOCH3	2970 ± 330	988 ± 185	41.4 ± 0.7d	72	24
20	Me	3-CH3	322 ± 23	415 ± 33	10 ± 3d	32	42
21	Me	3-CF3	496 ± 50	574 ± 49	31 ± 6d	16	19
22	Me	4-Cl	478 ± 38	2730 ± 210	11 ± 10d	28	160
23	Me	4-Br	516 ± 38	2460 ± 380	12 ± 5d	43	200
25	Me	4-NH2	431 ± 45	1590 ± 180	14 ± 3d	31	110
26	Me	4-NH2-3-I	18.0 ± 5.0	197 ± 84	1.27 ± 0.18d	14	160
27	Me	3-SO3H·Et3N	19600 ± 2000	9070 ± 1150	1310 ± 170d	15	6.9
28	Me	4-SO3H·Et3N	0% (10−4)	0% (10−4)	4500 ± 480d	>100	>100
29	Et	H	87	95	6.8	14	14
30a	Et	H(R,R4 = CH3)	3.2 ± 0.1	259 ± 15	18 ± 4d	0.18	14
30b	Et	H (S, R4 = CH3)	65 ± 4	1650 ± 120	494 ± 91d	0.13	3.3
31	Et	3-F	51 ± 7	32 ± 4	10.7 ± 1.2	4.8	3.0
32	Et	3-Cl	22 ± 3	19.0 ± 0.3	1.1 ± 0.3	20	17
33	Et	3-Br	10.9 ± 1.5	6.2 ± 0.7	2.8 ± 1.9d	3.9	2.2
34	Et	3-I	7.7 ± 0.9	7.2 ± 0.6	0.88 ± 0.21	8.8	8.2
35	Et	2-NO2	31 ± 4	24 ± 3	2.8 ± 0.5	11	8.6
36	Et	3-NO2	78 ± 10	35 ± 7	8.7 ± 1.2	11	4.0
37	Et	4-NO2	49 ± 9	574 ± 64	9.0 ± 1.3	5.4	64
38	Et	3-CH3	36.5 ± 1.2	17.7 ± 1.8	1.2 ± 0.1	30	15
39	Et	2-OMe	52 ± 5	21 ± 3	7.1 ± 0.3	7.3	3.0
40	Et	3-OMe	69 ± 8	38 ± 6	4.3 ± 0.6	16	8.8
41	Et	4-OMe	209 ± 30	609 ± 34	11 ± 3	19	55
42	cyclopropyl	H	112 ± 13	55 ± 6	103 ± 22	1.1	0.53

Open in a new tab

^a

Displacement of specific [³H]PIA binding, unless noted, in rat brain membranes expressed as K_i ± SEM in nM (n = 3–6).

^b

Displacement of specific [³H]CGS 21680 binding, unless noted, in rat striatal membranes, expressed as K_i ± SEM in nM (n = 3–6).

^c

Displacement of specific binding of [¹²⁵I]APNEA¹⁰ or [¹²⁵I]-N⁶-(4-amino-3-iodobenzyl)adenosine-5′-N-methyluronamide²⁸ from membranes of CHO cells stably transfected with the rat A₃-cDNA, expressed as K_i ± SEM in nM (n = 3–5).

^d

Vs[¹²⁵I]-N⁶-(4-amino-3-iodobenzyl)adenosine-5′-N-methyluronamide.

Disubstituted N⁶- and 5′-uronamide derivatives of adenosine were prepared by two main routes (Figures 2 and 3).^18,19The best overall yields were obtained by the route introduced by Olsson et al.¹⁸ (Figure 2). This route consisted of oxidation of 2′,3′-isopropylideneinosine, 48, followed by treatment with thionyl chloride to form at once the acid chloride and 6-chloro derivative. The acid chloride was displaced with an alkylamine, and the stable intermediate 51 was isolated, for further reaction with substituted benzyl bromides. An alternate method for preparing N⁶-benzyladenosines (Figure 3) consisted of the Dimroth rearrangement¹⁹ of N1-alkylated 5′-uronamidoadenosine derivatives, 55.

As with simple 5′-uronamides, the 5′-N-methylcarboxamide analogue of N⁶-benzyladenosine were most selective for A₃ receptors (compare 12, 13, 29, and 42). 13 was 37–56-fold more selective for A₃ receptors, whereas the 5′-N-cyclopropyl analogue, 42, was nonselective and much less potent. Although the 5′-N-methyl substitution was generally favored over the 5′-N-ethyl substitution, there were several exceptions, such as the 3-nitro and 3-methyl analogues, in which the potency was more favorable in the N-ethyl series (36 and 38 vs 17 and 20, respectively).

Comparisons of benzyl group substitutions were made for both the N-ethyl- and the N-methyluronamide series. The stereoselectivity of binding at A₃ receptors for enantiomers at the carbon α to N⁶ has been demonstrated for the R- and S-isomers of N⁶-(phenylisopropyl)adenosine.⁸ As for A₁ and A_2a receptors,^20,21 the R-isomer is favored. For the doubly modified N⁶-(1-phenylethyl)-5′-uronamidoadenosine analogues, 30a and 30b, the R-configuration is also favored by a factor of 27 at A₃ receptors. At A₁ and A_2a receptors the stereoselectivity factors for R-vs S-configurations were 20-fold and 6.4-fold, respectively. Thus, A₃ receptors resembles both A₁ and A_2a receptors in stereoselectivity at the position α to N⁶.

Substituents such as an electron-withdrawing group (e.g. nitro) or an electron-donating group (e.g. methoxy) were placed at each position of the benzene ring to probe electronic and positional effects of substitution. Among N⁶-benzyl-5′-N-ethyluronamides, selectivity for A₃ vs A₁ receptors ranged from 4- to 30-fold and was greatest for the 3-methyl derivative, 38. 3-Chloro-, 32, 3-methoxy-, 40, and 3-nitro-, 41, derivatives were also very selective for A₃ vs A₁ receptors. Selectivity for A₃ vs A_2a receptors ranged from 2- to 64-fold and was greatest for the 4-nitro derivative, 37, and the 4-methoxy derivative, 41. In general, positional and/or steric effects appeared to be more important than electronic effects. For example, 4-position substitution provided high selectivity for A₃ vs A_2a receptors in both the N-methyl (e.g. 22, 23, and 25) and N-ethyl (e.g. 37 and 41) series.

Substitution at the 3-position generally favored A₃ potency and selectivity. Halo substituents were varied at the 3- and 4-positions of the benzyl substituent. 3-Halo derivatives were particularly potent. The 3- and 4-bromo derivatives in the 5′-N-methyl series, 15 and 23, respectively, were both relatively selective, but the affinity at A₃ receptors was 6-fold greater with substitution at the 3-position. There was a major dependence of the affinity at all three adenosine receptors on the halo atom, and affinities of ca. 1 nM were achieved. In the 5′-N-ethyl series, the order of potency at A₁ and A_2a receptors was I ~ Br > Cl > F. At A₃ receptors, the order was I ~ Cl > Br > F. In the 5′-N-ethyl series, the 3-iodobenzyl analogue, 34, and to a lesser degree the 3-bromo analogue, 33, were very highly potent and selective for A₃ receptors. N⁶-(3-Iodobenzyl)adenosine-5′-N-methyluronamide, 16, displayed a K_i value of 1.0 nM at A₃ receptors and selectivity versus A₁ and A_2a receptors of 49- and 51-fold, respectively.

Since compound 16 was both potent and selective and as such is a candidate for further pharmacological characterization, the selectivity for adenosine receptors versus other receptors was examined in a battery of radioreceptor binding assays (NovaScreen, Adheron Corporation, Hanover, MD).²² At a concentration of 10⁻⁵ M, the displacement of radioligand from α-adrenergic, β-adrenergic, dopamine (D₁ and D₂), serotonin 5-HT₁, central benzodiazepine, amino acid (γ-aminobutyric acid; N-methyl-d-aspartate; kainate; quisqualate; and glycine, strychnine sensitive and insensitive, MK-801), peptide (angiotensin II, vasopressin, bombesin, neurokinin, central CCK, neuropeptide Y, neurotensin, somatostatin, vasoactive intestinal peptide, growth factors) receptors was insignificant (0 ± 20%). The displacement of binding of radioligand from ion channels (Ca²⁺, Cl⁻, and K⁺) and from second messenger sites (forskolin, phorbol ester, and inositol triphosphate) and inhibition of monoamine oxidase was also insignificant. Curiously, the displacement of radioligand at serotonin 5-HT₂, phencyclidene, peripepheral cholecystokinin receptors was 50–70% at 10⁻⁵ M. The observation that high affinity was not observed at any of these sites emphasizes the selectivity of compound 16 for A₃ receptors.

The 3-methyl and 3-trifluoromethyl analogues, 20 and 21, respectively, were prepared in the N-methyl series. The trifluoromethyl analogue was weaker at the A₃ receptors. Thus an electron-withdrawing group at the 3-position is not favorable for potency, further supported by the relatively modest potency of the 3-nitro derivative, 17. In the N-methyl series, 3- and 4-amino derivatives, 18 and 25, respectively, were prepared by treatment of the 6-chloropurine intermediate R′ = CH₃), 51a, with 3- or 4-aminobenzylamine. The intermediate 3-aminobenzylamine, 57, was prepared via catalytic reduction of the 3-nitro derivative, 56 (Figure 4A). The nucleophilic attack of the purine ring occurred selectively at the arylamine. Since the arylamino derivative 25 was prepared as a precursor for radioiodination, the expected major product of direct iodination, (3-iodo-4-aminobenzyl)adenosine-5′-N-methyluronamide, 26, was prepared as a standard for purification and pharmacology. This compound proved to be less selective for A₃ versus A₁ receptors than the corresponding 3-iodo derivative, 16. The disubstituted benzylamine intermediate 61 (Figure 4B) was prepared via protection of the alkylamine as the tert-butyloxycarbonyl derivative. The 3-aminobenzyl derivative, 25, was also N-acetylated at the 2′,3′-isopropylidene-protected stage using acetic anhydride to yield after deprotection compound 19. This acetamido derivative demonstrated that there is an unfavorable interaction at this site on the receptor. The 3-acetamido group is unfavorable at A₁ and to a lesser extent at A_2a and A₃ receptors. Thus, compound 19 is 72-fold and 24-fold selective for A₃ vs A₁ and A_2a receptors, respectively.

Two sulfo analogues were prepared as a putative peripherally selective agonist, by analogy to our previous study of A₁-selective sulfoadenosine derivatives.²³ Synthesis of the 4-sulfo derivative, 28, required intermediate 63, which was prepared directly from benzylamine (Figure 4C). A 3-sulfo intermediate, 66, leading to adenosine derivative 27, was prepared via sulfonation of 4-bromobenzylamine followed by catalytic hydrogenation in basic medium (Figure 4D). Compound 28 weakly displaced radioligand from A₃ but not A₁ or A_2a receptors; thus, it is a highly selective ligand. It is evident that a negative charge at the 4-benzyl position is poorly tolerated at all three receptor subtypes. The 3-sulfo derivative, 27, was slightly more potent at A₃ receptors, but considerably less selective than the 4-isomer.

Chemistry

New compounds were characterized (and resonances assigned) by 300-MHz proton nuclear magnetic resonance mass spectroscopy using a Varian GEMINI-300 FT-NMR spectrometer. Unless noted, chemical shifts are expressed as ppm downfield from tetramethylsilane. Synthetic intermediates were characterized by chemical ionization mass spectrometry (NH₃) and adenosine derivatives by fast atom bombardment mass spectrometry (positive ions in a noba or m-bullet matrix) on a JEOL SX102 mass spectrometer. In the EI mode accurate mass was determined using a VG7070F mass spectrometer. C, H, and N analyses were carried out by Atlantic Microlabs (Norcross, GA), and ±0.4% was acceptable. All adenosine derivatives were judged to be homogeneous using thin-layer chromatography (silica, 0.25 mm, glass backed, Alltech Assoc., Deerfield, IL) following final purification. The following benzylamine derivatives were purchased from Aldrich (St. Louis, MO): 4-amino, 2-methoxy, 3-methyl, 3-trifluoromethyl, 3-fluoro, 3-bromo, 3-iodo, 4-chloro. 4-Nitrobenzyl bromide was purchased from Fluka (Ronkonoma, NY). 3-Nitrobenzyl bromide was purchased from Lancaster (Windham, NH). 2-Chloroadenosine and 5′-carboxamido derivatives of adenosine (compounds 3 and 4) were purchased from Research Biochemicals International (Natick, MA). Compound 1 was the gift of Dr. John W. Daly (NIH). Analytical TLC plates and silica gel (230–400 mesh) were purchased from VWR (Bridgeport, NJ).

Inosine-5′-N-methyluronamide (7)

2′,3′-Isopropylideneinosine-5′-carboxylic acid¹⁸ (51, 50 mg, 0.155 mmol), EDAC (59 mg, 0.31 mmol), and N-hydroxysuccinimide (36 mg, 0.31 mmol) were dissolved in DMF (1 mL). Methylamine (40% in water, 50 μL) was added, and the mixture was stirred for 90 min. Water was added, and the solid residue (compound 50a) was separated and dried in vacuo. Compound 50a was purified by column chromatography (silica, eluted with chloroform–methanol–25% ammonium hydroxide, 80:20:1) to yield 49.8 mg (96%) of the pure product. The isopropylidene group was removed using 0.1 N HCl, warming at 70°C for 1.5h, and neutralizing with NaHCO₃. The title compound was purified by column chromatography (silica gel, chloroform–methanol–ammonium hydroxide, 80:20:1) to obtain 26 mg of the pure compound (57% yield). ¹H NMR (DMSO-d₆): δ 3.29 (d, J = 7.9 Hz, 3H, CH₃), 4.13–4.20 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.49–4.58 (m, 1H, H-2′), 5.58 (d, J = 6.4 Hz, 1H, OH-2′), 5.68 (d, J = 4.7 Hz, 1H, OH-3′), 5.94 (d, J = 6.9 Hz, 1H, H-1′), 8.13 (s, 1H, H-2), 8.26–8.34 (m, 1H, NH-Me), 8.41 (s, 1H, H-8).

N⁶-Benzyladenosine-5′-N-methyluronamide (13)

2′,3′-O-Isopropylidene-6-chloropurine-5′-N-methyluronamide (51a, 30 mg, 85 μmol), benzylamine (9.7 μL, 89 μmol), and triethylamine (23.7 μL, 0.70 mmol) were dissolved in absolute ethanol (1 mL). The solution was stirred at 65 °C for 16 h in a sealed vessel. The solvent was evaporated under a stream of nitrogen, and water was added to remove the triethylammonium salt. The supernatant was removed and discarded, and the insoluble residue containing intermediate was used without further purification.

Hydrochloric acid (1 N, 1 mL) was added, and the resulting solution was heated to 60 °C for 40 min. After cooling in an ice bath, sodium bicarbonate solution was added to neutralize. The suspension was extracted with ethyl acetate, and the solvent was removed under vacuum to obtain 15 mg of the title compound (R_f = 0.69, silica TLC plate, chloroform–methanol–ammonium hydroxide 80:20:1, 46% yield overall). ¹H NMR (DMSO-d₆): δ 2.70 (d, J = 4.3 Hz, 3H, CH₃), 4.14 (m, 1H, H-3′), 4.31 (s, 1H, H-4′), 4.59 (dd, J = 4.6 Hz, J = 7.5 Hz, 1H, H-2′), 4.71 (br s, 2H, N⁶-CH₂Ph), 5.96 (d, J = 7.4 Hz, 1H, H-1′), 7.30 (m, 5H, phenyl), 8.29 (s, 1H, H-2), 8.43 (s, 1H, H-8), 8.56 (br s, 1H, N⁶H-CH₂Ph), 8.86 (m, 1H, NH-Me).

N⁶-(3-Iodobenzyl)adenosine-5′-N-methyluronamide (16)

2′,3′-O-Isopropylidene-N-methyl-6-chloropurine-5′-uronamide (51a, 35 mg, 99 μmol), 3-iodobenzylamine hydrochloride (28 mg, 104 μmol), and triethylamine (41 μL, 0.30 mmol) were dissolved in absolute ethanol (1 mL). The solution was stirred at 75 °C for 16 h in a sealed vessel. The solvent was evaporated under a stream of nitrogen, and water was added to remove the triethylammonium salt. The supernatant was removed and discarded, and the insoluble residue containing intermediate 52a (R = 3-iodobenzyl) was used without further purification.

Hydrochloric acid (1 N, 1 mL) was added, and the resulting solution was heated to 60 °C for 4 h. After cooling in an ice bath, sodium bicarbonate solution was added to neutralize. A white solid formed and was filtered, washed with water, and dried to give 28 mg of the title compound (R_f = 0.74, silica TLC plate, chloroform–methanol–ammonium hydroxide, 80:20:1, 55% yield overall). ¹H NMR (DMSO-d₆): δ 2.70 (d, J = 4.6 Hz, 3H, CH₃), 4.11–4.18 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.54–4.63 (m, 2H, H-2′), 4.67 (br s, 2H, N⁶-CH₂Ph), 4.96 (br s, 2H, NH₂), 5.53 (d, J = 6.4 Hz, 1H, OH-2′), 5.71 (d, J = 4.1 Hz, 1H, OH-3′), 5.97 (d, J = 7.6 Hz, 1H, H-1′), 7.10 (t, J = 7.7 Hz, 1H), 7.35 (d, J = 7.7 Hz, 1H), 7.58 (d, J = 7.8 Hz, 1H), 7.72 (s, 1H), 8.29 (s, 1H, H-2), 8.44 (s, 2H, H-8), 8.56 (br s, 1H, N⁶H-CH₂Ph), 8.80–8.89 (m, 1H, NH-Me).

N⁶-(3-Aminobenzyl)adenosine-5′-N-methyluronamide (18)

Compound 51a (100 mg, 0.28 mmol) was dissolved in ethanol (1.5 mL) and treated with 3-aminobenzylamine hydrochloride (57, 47 mg, 0.30 mmol) and triethylamine (117 μL, 0.84 mmol). The solution was heated at 80 °C for 12 h in an oil bath. The solvent was evaporated, leaving a solid residue. The residue was treated with 1 N HCl (1.0 mL) and the mixture heated at 80 °C for 45 min. Sodium bicarbonate solution was added until pH 7, and the mixture was extracted three times with ethyl acetate. The solvent was removed under vacuum, and the residue was recrystallized from methanol–water to provide 56 mg of the pure product (50% yield). ¹H NMR (DMSO-d₆): δ 2.70 (d, J = 4.5 Hz, 3H, CH₃), 4.11–4.14 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.54–4.64 (m, 3H, H-2′, N⁶-CH₂Ph), 4.96 (br s, 2H, NH₂), 5.53 (d, J = 6.7 Hz, 1H, OH-2′), 5.72 (d, J = 3.9 Hz, 1H, OH-3′), 5.96 (d, J = 7.7 Hz, 1H, H-1′), 6.38 (d, J = 7.9 Hz, 1H), 6.46 (d, J = 7.7 Hz, 1H), 6.50 (s, 1H), 6.91 (t, J = 7.8 Hz), 8.28 (s, 1H, H-2), 8.40 (s, 2H, H-8, N⁶H-CH₂Ph), 8.87–8.97 (m, 1H, NH-Me).

N⁶-(3-Acetamidobenzyl)adenosine-5′-N-methyluronamide (19)

2′,3′-O-Isopropylidene-6-chloropurine-5′-N-methyluronamide (51a, 25 mg, 71 μmol), 3-aminobenzylamine hydrochloride (57, 11.3 mg, 71 μmol), and triethylamine (29 μL, 0.21 mmol) were dissolved in absolute ethanol (1 mL). The solution was stirred at 85 °C overnight in a sealed vessel. The solvent was evaporated under a stream of nitrogen and dried in vacuo. DMF (0.5 mL) was added followed by triethylamine (25 μL, 0.34 mmol) and acetic anhydride (32 μL, 0.34 mmol). After 30 min the reaction was complete. The solvent was removed under nitrogen, and a preparative TLC was performed (silica, CHCl₃–MeOH, 90:10) to recover 22 mg (46 mmol, 64% yield for both steps) of the isopropylidene derivative. ¹H NMR (CDCl₃): δ 1.34 (s, 3H, CH₃), 1.62 (s, 3H, CH₃), 2.13 (s, 3H, COCH₃), 2.61 (d, J = 4.9 Hz, 3H, NHCH₃), 4.71 (s, 1H, H-4′), 4.83 (br s, 2H, N⁶-CH₂Ph), 5.33 (s, 2H, H-2′ H-3′), 6.00 (s, 1H, H-1′), 6.75 (br s, 1H, NH-COCH₃), 7.09 (d, J = 7.6 Hz, 1H), 7.25 (t, J = 7.8 Hz, 1H), 7.41 (d, J = 7.8 Hz, 1H), 7.55 (s, 1H), 7.70 (s, 1H, H-2), 7.75 (s, 1H, H-8), 8.32 (br s, 1H, NHCH₃).

A 13-mg (27-μmol) sample of the latter compound was dissolved in 1 N HCl (0.5 mL) and warmed for 1 h at 70 °C. After cooling in an ice bath, sodium bicarbonate was added to neutralize. A solid crystallized after standing overnight at 4 °C, filtered, washed with water, and dried to give 11 mg (93% yield) of the title compound. A sample was purified by preparative TLC (silica gel, chloroform–methanol–ammonium hydroxide, 80:20:1) for microanalysis. ¹H NMR (DMSO-d₆): δ (s, 3H, COCH₃), 2.71 (d, J = 4.3 Hz, 3H, NHCH₃), 4.13 (m, 1H, H-3′), 4.31 (s, 1H, H-4′), 4.58 (m, 1H, H-2′), 4.67 (m, 2H, N⁶-CH₂Ph), 5.53 (d, J = 6.2 Hz, 1H, OH-2′), 5.72 (d, J = 4.2 Hz, 1H, OH-3′), 5.97 (d, J = 7.4 Hz, 1H, H-1′), 7.00 (d, J = 7.3 Hz, 1H), 7.19 (t, J = 7.8 Hz, 1H), 7.45 (s, 1H), 7.49 (d, J = 7.9 Hz, 1H), 8.28 (s, 1H, H-2), 8.42 (s, 1H, H-8), 8.52 (m, 1H, N⁶H-CH₂Ph), 8.90 (m, 1H, NHCH₃), 9.85 (s, NHCOCH₃).

N⁶-(3-(Trifluoromethyl)benzyl)adenosine-5′-N-methyluronamide (21)

The compound was prepared as described above for 16 using 3-(trifluoromethyl)benzylamine in 60% yield, except that due to aqueous insolubility, the hydrolysis on the isopropylidene protecting group was carried out in a 1:1 mixture of 1 N HCl and methanol. ¹H NMR (DMSO-d₆): δ 2.70 (d, J = 4.5 Hz, 3H, CH₃), 4.12–4.17 (m, 1H, H-3′), 4.30 (d, J = 1.2 Hz, 1H, H-4′), 4.54–4.64 (m, 1H, H-2′), 4.78 (br s, 2H, N⁶-CH₂Ph), 5.53 (d, J = 6.4 Hz, 1H, OH-2′), 5.72 (d, J = 4.3 Hz, 1H, OH-3′), 5.97 (d, J = 7.5 Hz, 1H, H-1′), 7.49–7.61 (m, 2H), 6.65 (d, J = 7.2 Hz, 1H), 7.71 (s, 1H), 8.29 (s, 1H, H-2), 8.45 (s, 1H, H-8), 8.64 (br s, 1H, N⁶H-CH₂Ph), 8.80–8.88 (m, 1H, NH-Me).

N⁶-(4-Amino-3-iodobenzyl)adenosine-5′-N-methyluronamide (26)

2′,3′-O-Isopropylidene-N-methyl-6-chloropurine-5′-uronamide (51a, 22.2 mg, 63 μmol) and 4-amino-3-iodobenzylamine trifluoroacetate (61, 63 μmol) were dissolved in absolute ethanol (1 mL). Triethylamine was added (80 mL), and the solution was stirred at 70 °C overnight. Ethanol was removed under nitrogen. Water (1 mL) was added and then removed from the resulting solid using a Pasteur pipette. The amorphous solid was dissolved in 1 N HCl and stirred at 70 °C for 1 h. After cooling the solution was neutralized with NaHCO₃ and then extracted with ethyl acetate (7 × 2 mL). After removal of the solvent in vacuo, the product was purified by column chromatography (silica, CHCl₃–MeOH–NH₄OH, 85:10:0.5) to obtain 17 mg (32 μmol, 51% yield overall). ¹H NMR (DMSO-d₆): δ 2.71 (d, J = 4.6 Hz, 3H, NHCH₃), 4.14 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.51 (br s, 2H, N⁶-CH₂Ph), 4.58 (m, 1H, H-2′), 5.07 (s, 2H, NH₂), 5.52 (d, J = 6.4 Hz, 1H, OH-2′), 5.71 (d, J = 4.2 Hz, 1H, OH-3′), 5.95 (d, J = 7.6 Hz, 1H, H-1′), 6.67 (d, J = 8.1 Hz, 1H), 7.08 (dd, J = 8.3 Hz, J = 1.7 Hz, 1H), 7.56 (d, J = 1.5 Hz, 1H), 8.30 (s, 1H, H-2), 8.36–8.41 (m, 2H, H-8, N⁶H-CH₂Ph), 8.90 (m, 1H, NHCH₃).

N⁶-(3-Sulfobenzyl)adenosine-5′-N-methyluronamide Triethylammonium Salt (27)

A suspension of 2′,3′-O-isopropylidene-6-chloropurine-5′-N-methyluronamide (51a, 40 mg, 113 μmol), 3-sulfobenzylamine (66, 21.2 mg, 113 μmol), and triethylamine (47 μL, 0.339 mmol) in absolute ethanol (1 mL) was stirred at 90 °C overnight in a sealed vessel. The solvent was removed under nitrogen, and the isopropylidene derivative was purified by column chromatography (silica, CHCl₃–MeOH-NH₄–OH, 80:20:1). The isolated homogeneous band was dissolved in 1 N HCl and warmed at 65 °C for 30 min. The solvent was removed under nitrogen. The title compound was purified using reverse phase cartridges (Alltech Maxiclean, 900 mg) to obtain 38 mg (59% overall yield) as the triethylammonium salt. ¹H NMR (DMSO-d₆): δ 1.16 (t, J = 7.1 Hz, 9H, (CH₃CH₂)₃NH⁺), 2.70 (d, J = 4.6 Hz, 3H, NHCH₃), 3.09 (m, 6H, (CH₃CH₂)₃NH), 4.13 (d, J = 4.3 Hz, 1H, H-3′), 4.29 (s, 1H, H-4′), 4.59 (dd, J = 4.6 Hz, J = 7.6 Hz, 1H, H-2′), 4.73 (m, 2H, N⁶-CH₂Ph), 5.62 (br s, 2H, OH), 5.95 (d, J = 7.6 Hz, 1H, H-1′), 7.25 (m, 2H), 7.43 (d, J = 7.6 Hz, 1H), 7.60 (s, 1H), 8.29 (s, 1H, H-2), 8.42 (s, 1H, H-8), 8.63 (br s, 1H, N⁶H-CH₂Ph), 8.93 (m, 2H, NHCH₃, (CH₃CH₂)₃NH⁺).

N⁶-(4-Sulfobenzyl)adenosine-5′-N-methyluronamide (28)

Compound 51a (30 mg, 85 μmol), p-sulfobenzylamine (63, 17 mg, 90 μmol), and triethylamine (34 μL, 0.27 mmol) were combined in absolute ethanol (1 mL) and heated to 90 °C for 3 days. The mixture was filtered, and the filtrate was reduced in volume by evaporation, leaving a viscous syrup. The isopropylidene-protected intermediate was purified (23.5 mg, recovered) on a TLC plate run in chloroform–methanol–acetic acid (85:10:5). Hydrochloric acid (1 mL, 1 N) was added, and the solution was warmed at 60 °C for 40 min. After cooling and evaporation of the solvent, methanol and ethyl acetate were added. The resulting preciptate was separated and dried to provide 14.3 mg (29% yield) of the title compound. A sample was purified by preparative TLC (chloroform–methanol–ammonium hydroxide, 70:30:1) for microanalysis. ¹H NMR (DMSO-d₆): δ 2.69 (d, J = 43.5 Hz, 3H, NHCH₃), 4.15 (d, J = 4.5 Hz, 1H, H-3′), 4.32 (s, 1H, H-4′), 4.58 (dd, J = 4.6 Hz, J = 7.2 Hz, 1H, H-2′), 4.71 (br s, 2H, N⁶-CH₂Ph), 5.98 (d, J = 7.4 Hz, 1H, H-1′), 7.27 (d, J = 7.8 Hz, 1H), 7.51 (d, J = 8.1 Hz, 1H), 8.33 (s, 1H, H-2), 8.50 (br s, 2H, H-8, N⁶H-CH₂Ph), 8.82 (br s, 1H, NHCH₃).

(R)-N⁶-(1-Phenylethyl)adenosine-5′-N-ethyluronamide (30a)

2′,3′-O-Isopropylidene-6-chloropurine-5′-N-ethyluronamide (51b, 30 mg, 81 μmol), (R)-(+)-α-methylbenzylamine (11.1 μL, 86 μmol), and triethylamine (22 μL, 162 μmol) were dissolved in absolute ethanol (1 mL). The solution was stirred at 80 °C for 24 h in a sealed vessel. The solvent was removed under nitrogen. HCl (1 N) (0.5 mL) was added and the solution stirred at 60 °C for 30 min. After cooling, NaHCO₃ was added until pH 7, and the solution was extracted with ethyl acetate (4 × 2 mL). The organic layers were combined, and the solvent was evaporated, leaving 20 mg (48 μmol, 60% yield overall) of the title compound. A sample was purified by preparative TLC for microanalysis, as described for compound 19. ¹H NMR (DMSO-d₆): δ 1.05 (t, J = 7.1 Hz, 3H, CH₃), 1.53 (d, J = 6.8 Hz, 3H, CHCH₃), 3.15–3.28 (m, 2H, CH₂), 4.08–4.17 (m, 1H, H-3′), 4.28 (s, 1H, H-4′), 4.54–4.63 (m, 1H, H-2′), 5.46–5.57 (m, 2H, OH-2′ N⁶-CHPh), 5.72 (d, J = 4.3 Hz, 1H, OH-3′), 5.94 (d, J = 7.3 Hz, 1H, H-1′), 7.13–7.22 (m, 1H), 7.28 (t, J = 7 Hz, 2H), 7.44 (d, J = 7.7 Hz, 2H), 8.21 (s, 1H, H-2), 8.42 (br s, 2H, H-8, N⁶HCH₂Ph), 8.83 (m, NH-Et).

(S)-N⁶-(1-Phenylethyl)adenosine-5′-N-ethyluronamide (30b)

Same procedure as above using (S)-(—)-α-methylbenzylamine. ¹H NMR (DMSO-d₆): δ 1.05 (t, J = 7.0 Hz, 3H, CH₃), 1.54 (d, J = 7.2 Hz, 3H, CHCH₃), 3.15–3.26 (m, 2H, CH₂), 4.08–4.15 (m, 1H, H-3′), 4.28 (s, 1H, H-4′), 4.54–4.63 (m, 1H, H-2′), 5.46–5.58 (m, 2H, OH-2′, N⁶-CHPh), 5.72 (d, J = 4.0 Hz, 1H, OH-3′), 5.94 (d, J = 7.8 Hz, 1H, H-1′), 7.13–7.22 (m, 1H), 7.28 (t, J = 7 Hz, 2H), 7.43 (d, J = 7.7 Hz, 2H), 8.21 (s, 1H, H-2), 8.41 (br s, 2H, H-8, N⁶H-CH₂Ph), 8.87 (m, NH-Et).

N⁶-(3-Fluorobenzyl)adenosine-5′-N-ethyluronamide (31)

2′,3′-O-Isopropylidene-6-chloropurine-5′-N-ethyluronamide (51b, 30 mg, 81 μmol) and 3-fluorobenzylamine (10.8 mg, 86 μmol) were dissolved in absolute ethanol (1 mL). Triethylamine (17 μL, 0.12 mmol) was added, and the solution was warmed at 80 °C for 16 h in a sealed vessel. No starting material (R_f = 0.41, silica TLC plates, chloroform–methanol, 95:5) remained in the mixture. The solvent was evaporated under a stream of nitrogen, and water was added to remove the triethylammonium salt. The supernatant was removed and discarded, and the insoluble residue containing intermediate 52b (R = 3-fluorobenzyl) was used without further purification.

For removal of the isopropylidene group, hydrochloric acid (1 N, 1 mL) was added, and the resulting solution was heated to 60 °C for 45 min or until complete as judged by TLC. The intermediate 52b and product displayed R_f values of 0.51 and 0.10, respectively (silica, chloroform–methanol, 95:5). After cooling, sodium bicarbonate solution was added to neutralize. A white solid precipitated and was filtered, washed with water, and dried to give 23 mg of the title compound (68% yield overall). ¹H NMR (DMSO-d₆): δ 1.06 (t, J = 7.2 Hz, 3H, CH₃), 3.20 (m, 2H, CH₂), 4.00–4.16 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.56–4.64 (m, 1H, H-2′), 4.71 (br s, 2H, N⁶-CH₂Ph), 5.53 (d, J = 6.4 Hz, 1H, OH-2′), 5.73 (d, J = 4.3 Hz, 1H, OH-3′), 5.97 (d, J = 7.4 Hz, 1H, H-1′), 7.03 (dt, J = 8.6 Hz, J = 2.0 Hz, H_benzyl-5), 7.14 (d, J = 12.7 Hz, H_benzyl-2), 7.17 (d, J = 8.3 Hz, H_benzyl-6), 7.33 (dd, J = 14.0 Hz, J = 7.9 Hz, H_benzyl-4), 8.26 (s, 1H, H-2), 8.44 (s, 1H, H-8), 8.9 (br s, 1H, N⁶H-CH₂Ph), 8.83 (t, J = 5.6 Hz, 1H, NH-Et).

N⁶-(3-Chlorobenzyl)adenosine-5′-N-ethyluronamide (32)

To a solution of NECA (50 mg, 0.162 mmol) in DMF (1 mL) was added 3-chlorobenzyl bromide (61 μL, 0.47 mmol), and the solution was stirred in a closed vessel for 2 days at 40 °C. DMF was evaporated under a stream of nitrogen. The residue was treated with acetone (1 mL), and ether (2 mL) was added. The solvent was removed using a Pasteur pipette, and the residue, an amorphous solid, was again extracted with chloroform to remove traces of the benzyl bromide. The dried residue was treated with methanol (2.0 mL), and concentrated NH₄OH (4.0 mL) was added. The mixture was warmed in a closed tube at 90 °C for 2 h with stirring. The mixture was reduced in volume by evaporation and cooled in an ice bath, resulting in precipitation of the chromatographically pure product. The white solid was isolated by filtration to give 43 mg (62% yield) of product which melted at 199–200 °C. ¹H NMR (DMSO-d₆): δ 1.06 (t, J = 7.1 Hz, 3H, CH₃), 3.14–3.25 (m, 2H, CH₂), 4.11–4.16 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.56–4.64 (m, 1H, H-2′), 4.71 (br s, 2H, N⁶-CH₂Ph), 5.53 (d, J = 6.4 Hz, 1H, OH-2′), 5.72 (d, J = 4.2 Hz, 1H, OH-3′), 5.97 (d, J = 7.6 Hz, 1H, H-1′), 7.2–7.4 (m, 4H, phenyl), 8.26 (s, 1H, H-2), 8.44 (s, 1H, H-8), 8.60 (br s, 1H, N⁶H-CH₂Ph), 8.82 (t, J = 5.8 Hz, 1H, NH-Et). Mass spectrum (CI-NH₃): m/e 433 (MH⁺, base).

N⁶-(2-Nitrobenzyl)adenosine-5′-N-ethyluronamide (35)

To a solution of NECA (51 mg, 0.165 mmol) in DMF (1 mL) was added 2-nitrobenzyl bromide (107 mg, 0.496 mmol), and the solution was stirred for 2 days at 40 °C. DMF was removed under vacuum giving a syrup that crystallized when acetone and ether were added. The solvent was removed using a Pasteur pipette, and the solid was dried in vacuo and purified by preparative TLC (silica gel, ethyl acetate—2-propanol—water, 4:1:2, upper phase, R_f = 0.25) to give compound 11 in 45% yield. ¹H NMR (DMSO-d₆): δ 1.04 (t, J = 6.7 Hz, 3H, CH₃), 3.11–3.27 (m, 2H, CH₂), 4.12–4.18 (m, 1H, H-3′), 4.30 (d, J = 2.0 Hz, 1H, H-4′), 4.49–4.57 (m, 1H, H-2′), 5.55 (s, 2H, N¹-CH₂Ph), 5.61 (br s, 1H, OH-2′), 5.74 (br s, 1H, OH-4′), 5.88 (d, J = 7.1 Hz, 1H, H-1′), 7.06 (s, 1H, N⁶H), 7.15 (d, J = 7.8 Hz, 1H, arom), 7.54 (t, J = 7.8 Hz, 1H, arom), 7.68 (t, J = 8.0 Hz, 1H, arom), 8.10 (d, J = 7.5 Hz, 1H, arom), 8.26 (s, 1H, H-2), 8.29 (s, 1H, H-8), 8.43 (t, J = 5.6 Hz, 1H, NH-Et).

Methanol (2.0 mL) and concentrated NH₄OH (2.0 mL) were added to compound 11, and the mixture was warmed in a closed vessel at 90 °C for 45 min. The solvent was evaporated, and the product was purified by preparative TLC (silica gel, ethyl acetate–2-propanol–water, 4:1:2, upper phase, R_f = 0.77) to give 25 mg of the title compound (34% yield overall). Mp: 181 °C dec. A sample was recrystallized from 2-propanol–ethyl ether for microanalysis. ¹H NMR (DMSO-d₆): δ 1.05 (t, J = 7.4 Hz, 3H, CH₃), 3.13–3.24 (m, 2H, CH₂), 4.09–4.16 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.56–4.63 (m, 1H, H-2′), 4.94–5.13 (m, 2H, N⁶-CH₂Ph), 5.54 (d, J = 7.0 Hz, 1H, OH-2′), 5.73 (d, J = 4.5 Hz, 1H, OH-4′), 5.97 (d, J = 7.7 Hz, 1H, H-1′), 7.46–7.59 (m, 2H, arom), 7.67 (t, J = 7.1 Hz, 1H, arom), 8.04 (d, J = 7.6 Hz, 1H, arom), 8.21 (s, 1H, H-2), 8.47 (s, 1H, H-8), 8.62 (m, 1H, N⁶H-CH₂Ph), 8.80 (t, J = 5.5 Hz, 1H, NH-Et).

N⁶-(3-Nitrobenzyl)adenosine-5′-N-ethyluronamide (36)

The title compound was synthesized in 55% yield, according the procedure for compound 35, using instead 3-nitrobenzyl bromide as starting material. ¹H NMR (DMSO-d₆): δ 1.06 (t, J = 7.1 Hz, 3H, CH₃), 3.11–3.24 (m, 2H, CH₂), 4.11–4.14 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.56–4.62 (m, 1H, H-2′), 4.82 (br s, 2H, N⁶-CH₂Ph), 5.54 (d, J = 6.5 Hz, 1H, OH-2′), 5.73 (d, J = 4.3 Hz, 1H, OH-3′), 5.97 (d, J = 7.3 Hz, 1H, H-1′), 7.61 (t, J = 8.0 Hz, 1H), 7.72 (d, J = 7.4 Hz, 1H), 7.95 (d, J = 7.6 Hz, 1H), 8.22 (br s, 1H), 8.26 (s, 1H, H-2), 8.46 (s, 1H, H-8), 8.68–8.76 (m, 1H, N⁶H-CH₂Ph), 8.82 (t, J = 5.5 Hz, 1H, NH-Et).

N⁶-(4-Nitrobenzyl)adenosine-5′-N-ethyluronamide (37)

To a solution of NECA (50 mg, 0.162 mmol) in DMF (0.5 mL) was added 4-nitrobenzyl bromide (53 mg, 0.245 mmol), and the solution was stirred for 2 days at 40 °C. DMF was removed in vacuo giving a syrup that crystallized when acetone and ether were added. The solvent was removed using a Pasteur pipette, and the solid was dried in vacuo and purified by preparative TLC (silica gel, ethyl acetate—2-propanol—water, 4:1:2, upper phase) to give compound 10 in 60% yield. ¹H NMR (DMSO-d₆): δ 1.06 (t, J = 7 Hz, 3H, CH₃), 3.20 (q, J = 7.2 Hz, 2H, CH₂), 4.12–4.18 (br s, 1H, H-3′), 4.29 (s, 1H, H-4′), 4.47–4.56 (m, 1H, H-2′), 5.39 (s, 2H, N¹-CH₂Ph), 5.57 (d, J = 6.3 Hz, 1H, OH-2′), 5.69 (d, J = 4.5 Hz, 1H, OH-3′), 5.87 (d, J = 7.3 Hz, 1H, H-1′), 7.58 (d, 2H, J = 8.7 Hz, arom), 8.19 (d, 2H, J = 8.8 Hz, arom), 8.27 (s, 1H, H-2), 8.37 (s, 1H, H-8). FAB⁺ (m-bullit): m/e 444 (MH⁺).

Methanol (1.0 mL) and concentrated NH₄OH (2.0 mL) were added to compound 10, and the mixture was warmed in a closed vessel at 90 °C for 45 min. The solvent was evaporated, and the product was purified by preparative TLC (silica gel, ethyl acetate—2-propanol—water, 4:1:2, upper phase) to give 21 mg of the pure product (R_f = 0.77, 49% yield). Mp: 196 °C dec. ¹H NMR (DMSO-d₆): δ 1.06 (t, J = 7.3 Hz, 3H, CH₃), 3.14–3.24 (m, 2H, CH₂), 4.11–4.17 (m, 1H, H-3′), 4.30 (s, 1H, H-4′), 4.56–4.65 (m, 1H, H-2′), 4.82 (br s, 2H, N⁶-CH₂Ph), 5.55 (d, J = 6.3 Hz, 1H, OH-2′), 5.73 (d, J = 4.2 Hz, 1H, OH-3′), 5.97 (d, J = 7.5 Hz, 1H, H-1′), 7.58 (d, 2H, J = 8.6 Hz, arom), 8.17 (d, 2H, J = 8.7 Hz, arom), 8.25 (s, 1H, H-2), 8.46 (s, 1H, H-8), 8.70 (br s, 1H, N⁶H-CH₂Ph), 8.81 (t, J = 5.6 Hz, 1H, NH-Et). Mass spectrum (CI-NH₃): m/e 444 (MH⁺, base).

N⁶-(4-Methoxybenzyl)adenosine-5′-N-ethyluronamide (41)

To a solution of NECA (50 mg, 0.162 mmol) in DMF (0.5 mL) was added 4-methoxybenzyl chloride (33 μL, 0.24 mmol), and the solution was stirred for 3 days at 40 °C. DMF was removed in vacuo, giving a syrup that crystallized when acetone and ether were added. The solvent was removed using a Pasteur pipette. Methanol (1.0 mL) and concentrated NH₄OH (2.0 mL) were added, and the mixture was warmed in a closed vessel at 90 °C for 45 min. The solvent was evaporated, and the product was purified by preparative TLC (silica gel, ethyl acetate—2-propanol—water, 4:1:2) to give 4.9 mg of the pure product which crystallized from methanol (R_f = 0.82, 14% yield). ¹H NMR (DMSO-d₆): δ 1.07 (t, J = 7.5 Hz, 3H, CH₃), 3.2 (m, 2H, CH₂), 3.69 (s, OCH₃), 4.12 (d, J = 5.1 Hz, 1H, H-3′), 4.29 (s, 1H, H-4′), 4.59 (dd, J = 7.6 Hz, J = 4.7 Hz, 1H, H-2′), 4.63 (br s, 2H, N⁶-CH₂Ph), 5.96 (d, J = 7.7 Hz, 1H, H-1′), 6.84 (d, J = 8.5 Hz, 2H, arom), 7.27 (d, J = 8.7 Hz, 2H, arom), 8.25 (s, 1H, H-2), 8.40 (s, 1H, H-8), 8.45 (br s, 1H, N⁶H-CH₂Ph), 8.87 (t, J = 4.9 Hz, 1H, NH-Et). Mass spectrum (CI-NH₃): m/e 429 (MH⁺, base).

N⁶-Benzyladenosine-5′-N-cyclopropyluronamide (42)

To a solution of adenosine-5′-N-cyclopropyluronamide (20 mg, 0.062 mmol) in anhydrous DMF (1 mL) was added benzyl bromide (22 μL, 0.19 mmol), and the solution was stirred for 60 h at 40 °C. DMF was removed in vacuo, giving a syrup that crystallized when acetone and ether were added. The solvent was removed by decantation. Methanol (0.5 mL) and concentrated NH₄OH (2.0 mL) were added, and the mixture was warmed in a closed vessel at 90 °C for 2 h. The mixture was reduced in volume by evaporation and cooled in an ice bath, resulting in precipitation of the chromatographically pure product. The white solid was isolated by filtration, washed with water, and dried to give 15 mg (59% yield) of product which melted at 178–180 °C. ¹H NMR (DMSO-d₆): δ 0.46 (m, 2H, CH₂), 0.69 (m, 2H, CH₂), 2.70 (m, 1H, CH₃), 4.13 (m, 1H, H-3′), 4.27 (s, 1H, H-4′), 4.57 (m, 1H, H-2′), 4.71 (br s, 2H, N⁶-CH₂Ph), 5.53 (d, J = 6.5 Hz, 1H, OH-2′), 5.73 (d, J = 4.1 Hz, 1H, OH-3′), 5.95 (d, J = 7.5 Hz, 1H, H-1′), 7.15–7.43 (m, 5H, phenyl), 8.22 (s, 1H, H-2), 8.42 (s, 1H, H-8), 8.56 (br s, 1H, N⁶H-CH₂Ph), 8.88 (d, J = 3.8 Hz, 1H, NH-Me).

2′,3′-Isopropylideneadenosine-5′-carboxylic Acid (44)

2′,3′-Isopropylideneadenosine (43, Aldrich Chemical Co., St. Louis, MO, 0.5 g, 1.6 mmol) was dissolved in glacial acetic acid (11 mL), and chromium trioxide (0.222 g, 2.22 mmol) was added. A brown suspension formed, and the color changed gradually to dark green. The suspension was stirred at room temperature for 4 days. The dark solid formed was filtered, washed with water, and crystallized from MeOH to afford a white solid (0.28 g, 54 % yield): mp 257 °C; ¹H NMR (DMSO-d₆): δ 8.24 (s, 1H, H-8), 8.08 (s, 1H, H-2), 7.28 (br s, 2H, NH₂), 6.33 (s, 1H, H-1′), 5.50 (ABq, 2H, H-2′, H-3′), 4.68 (s, 1H, H-4′), 1.52, 1.35 (s, 3H, Me) ppm. High-resolution MS calcd for C₁₃H₁₅N₅O₅ 321.1058, found 321.1073. IR (KBr): ν 3000, 1706 cm⁻¹.

Methyl 2′,3′-Isopropylideneadenosine-5′-carboxylate (45)

An excess of diazomethane in ether (0.75 M) was added dropwise over 0.5 h to a suspension of compound 44 (0.22 g, 0.68 mmol) in dioxane-MeOH (1:1, 50 mL) until a clear yellowish solution was obtained. The reaction mixture was stirred at room temperature for 1.5 h. Nitrogen was bubbled through the solution until the yellow color disappeared. The solvent was removed, and the product was dried under high vacuum to produce a white solid (0.211 g, 92% yield): mp 221–222 °C; ¹H NMR (DMSO d₆) δ 8.25 (s, 1H, H-8), 8.05 (s, 1H, H-2), 7.32 (s, 2H, NH₂), 6.38 (s, 1H, H-1′), 5.61 (d, J = 6 Hz, 1H, H-2′), 5.43 (d, J = 6 Hz, 1H, H-3′), 4.85 (s, 1H, H-4′), 3.29 (s, 3H, CO₂Me), 1.51, 1.34 (s, 3H, Me) ppm. High-resolution MS caled for C₁₄H₁₇N₅O₅ 335.1238, found 335.1230.

2′,3′-Isopropylidene-6-chloropurine-5′-methyluronamide (51a)

2′,3′-Isopropylideneinosine-5′-carboxylic acid¹⁸ (1.1 g, 3.4 mmol) was added to a solution of thionyl chloride (0.51 mL, 0.68 mmol) and dimethylformamide (0.26 mL) in anhydrous chloroform (43 mL, dried over Al₂O₃). The mixture was heated to reflux with the exclusion of moisture for 6 h. After the mixture was cooled, the solvent was removed in vacuo leaving a syrup that was dissolved in chloroform (12 mL). The solution was cooled to 0 °C, and 2 mL of methylamine dissolved in 20 mL of chloroform was added. After being stirred for 15 min at <10 °C, the solution was extracted successively with HCl (0.1 N, 3 × 60 mL), sodium bicarbonate (0.5 M, 100 mL), and water (2 × 50 mL). The organic layer was dried (MgSO₄) and the sovent evaporated leaving 632 mg of the homogeneous (R_f = 0.75, chloroform–methanol–ammonium hydroxide, 80:20:1) title compound (53% yield). ¹H NMR (DMSO-d₆): δ 8.80 (s, 1H, purine), 8.73 (s, 1H, purine), 6.50 (s, 1H, H-1′), 5.4-5.5 (m, 2H, H-2′ and H-3′), 4.63 (s, 1H, H-4′), 2.16 (d, J = 4.7 Hz, 3H, NCH₃), 1.53 (s, 3H, i-Pr), 1.34 (s, 3H, i-Pr).

3-Aminobenzylamine Hydrochloride (57)

A mixture of 3-nitrobenzylamine hydrochloride (1.0g, 5.3 mmol) and 5% Pd/C (0.25 g) in methanol (30 mL) was shaken under hydrogen gas (50 psi) for 45 min. After filtration through Celite, the solvent was removed under vacuum to obtain the product in quantitative yield. ¹H NMR (MeOD): δ 3.96 (s, 2H, CH₂), 6.67–6.76 (m, 3H, H-2, H-4, H-6), 7.10–7.18 (m, 1H, H-5).

4-Amino-3-iodobenzylamine (61)

Compound 59 was prepared by treatment of 4-aminobenzylamine (0.30 g, 2.45 mmol) dissolved in methanol (10 mL) with di-tert-butyl dicarbonate (0.36 g, 1.65 mmol), added in portions with stirring over 0.5 h. The solvent was evaporated, and the residue was suspended in ethyl acetate and filtered. The solution was extracted with 0.5 M NaH₂PO₄ (3×) and dried (Na₂SO₄). Removal of solvent in vacuo left a clear oil, 4-((tert-butyloxycarbonyl)amino)benzylamine (59, 79% yield), which solidified upon standing to form a solid that melted at 72–75 °C.

Compound 59 (29.6 mg, 0.133 mmol) was treated with I₂ (33.8 mg, 0.133 mmol), CaCO₃ (16.6 mg), and MeOH (0.4 mL), and the dark mixture was stirred at 70 °C for 6 h. After the mixture was cooled, the solvent was removed under vacuum. Saturated aqueous sodium bisulfite was added and extracted with ethyl acetate (3 × 3 mL). The organic layer was dried under vacuum, and the product was purified by preparative TLC (silica, CHCl₃-MeOH, 75:1) to give 60, 22 mg of 4-((tert-butyloxycarbonyl)-amino)-3-iodobenzylamine (47% yield). ¹H NMR MeOD-d₃): δ 1.43 (s, 9H, t-Bu), 4.03 (s, 2H CH₂Ph), 6.74 (d, 1H, J = 8.3 Hz), 7.02 (dd, 1H, J = 8.3, 1.9 Hz), 7.5 (br s, 1H).

Compound 60 was deprotected treating with trifluoroacetic acid (0.4 mL) for 10 min at room temperature. The excess of TFA was removed under nitrogen, and the residue, 4-amino-3-iodobenzylamine, was dried under vacuum and used without further purification.

4-Sulfobenzylamine (63)

Benzylamine (10.2 mL, 93 mmol) was added dropwise at 0 °C to 11% fuming H₂SO₄ (30 mL). After being stirred at room temperature for 30 min, the solution was warmed to 75 °C for 1 h. After cooling, the solution was poured into cold dioxane (400 mL). The solid obtained was filtered through glass and washed several times with dioxane. The compound was purified by dissolving in a minimum volume of aqueous NH₄OH and then precipitating upon addition of HCl until pH 1 while cooling, to obtain 7.0 g of the homogeneous product (40% yield). ¹H NMR (DMSO-d₆): δ 4.02 (br s, 2H, CH₂), 7.38 (d, J = 8.1 Hz, 2H), 7.62 (d, J = 8.2 Hz, 2H), 8.11 (br s, 3H, NH₃⁺). FAB⁺ (glycerol): 188 (MH⁺).

4-Bromo-3-sulfobenzylamine (65)

4-Bromobenzylamine hydrochloride (2.0 g, 9.0 mmol) was added portionwise to 18–24% fuming H₂SO₄ (4 mL) while cooling at 0 °C. After being stirred at room temperature for 1 h, the wine-colored solution was warmed to 100 °C for another hour. After cooling, the solution was poured into dioxane (40 mL), and the resulting precipitate was filtered and washed several times with dioxane. The amorphous solid was dissolved in a minimum volume of 4 N NaOH and then reprecipitated following acidification with 6 N HCl. The solid was washed with water and dried to obtain 1.9 g of product (79% yield). ¹H NMR (DMSO-d₆): δ 3.96–4.93 (m, 2H, CH₂), 7.26 (dd, J = 8.1 Hz, J = 2.3 Hz, 1H), 7.61 (d, J = 8.1 Hz, 1H), 8.05 (d, J = 2.2 Hz, 1H), 8.13 (br s, 3H, NH₃⁺). FAB⁺ (thioglycerol): 266/268 (MH⁺).

3-Sulfobenzylamine (66)

4-Bromo-3-sulfobenzylamine (1.0 g, 3.76 mmol) was suspended in 20 mL of water, and 4 N NaOH was added until dissolution (pH = 12.5). Next, 5% Pd (C) was added, and the mixture was hydrogenated (50 psi) overnight. It was filtered through glass wool, and 12 N HCl was added until pH = 1, but no precipitate formed. Inorganic salts precipitated following addition of ethanol. The remaining solution was partially lyophilized, and the solid was filtered and dried leaving 440 mg of the pure product (63% yield). ¹H NMR (DMSO-d₆): δ 4.03 (q, J = 5.8 Hz, 2H), CH₂), 7.33–7.38 (m, 2H), 7.57–7.63 (m, 1H), 7.78 (s, 1H), 8.13 (br s, 3H, NH₃⁺).

Compounds 15, 17, 20, 22, 23, 25, 33, 34, and 38 were synthesized according to the procedure for compound 16, using as starting materials the correspondening benzylamines. Compounds 12, 14, 39, and 40 were synthesized according with the procedure for compound 32.

Biological Methods: Receptor Binding. Materials

F-12 (Ham’s) medium, fetal bovine serum (FBS) and penicillin/streptomycin were from Gibco BRL (Gaithersburg, MD). [¹²⁵I]APNEA was prepared as descibed previously.²⁶ [³H]R-PIA was from Amersham (Arlington Heights, IL), and [³H]CGS 21680 was from DuPont NEN (Boston, MA). Adenosine deaminase (ADA) was from Boehringer Mannheim (Indianapolis, IN). Composition of lysis buffer: 10 mM Tris/5 mM EDTA, pH 7.4 at 5 °C. 50/10/1 buffer: 50 mM Tris; 10 mM MgCl₂; 1 mM EDTA, pH 8.26 at 5 °C. All other materials were from standard local sources and of the highest grade commercially available.

Cell Culture and Membrane Preparation

CHO cells stably expressing the A₃ receptor⁶ were grown in F-12 medium containing 10% FBS and penicillin/streptomycin (100 U/mL and 100 μg/mL, respectively) at 37 °C in a 5% CO₂ atmosphere. When cells had reached confluency, they were washed twice with Dulbecco’s phosphate buffer solution before dislodging after addition of 3 mL of trypsin-EDTA. For the final passage cells were grown in 150- × 50-mm tissue culture dishes. Cells were washed twice with 10 mL of lysis buffer. After addition of 5 mL of lysis buffer, cells were mechanically scraped and homogenized in an ice-cold Dounce homogenizer (20 strokes by hand). The suspension was centrifuged at 43000g for 10 min. The pellet was resuspended in the minimum volume of ice-cold 50/10/1 buffer required for the binding assay and homogenized in a Dounce homogenizer. Typically, six to eight 175-cm² flasks were used for a 48-tube assay. Adenosine deaminase was added to a final concentration of 3 units/mL, and the suspension was incubated at 37 °C for 15 min; the membrane suspension was subsequently kept on ice until use. When large batches (ca. 100 flasks) were processed, homogenization was performed with a Polytron (Brinkman, Luzern, Switzerland) and further workup was as described above. The preparation was stored at −70 °C and retained its [¹²⁵I]APNEA binding properties for at least 1 month.

Radioligand Binding

Binding of [¹²⁵I]APNEA to CHO cells stably transfected with the A₃ receptor clone was performed essentially as described.²⁶ Assays were performed in 50/10/1 buffer in glass tubes and contained 100 μL of the membrane suspension, 50 μL of [¹²⁵I]APNEA (final concentration 0.5 nM), and 50 μL of inhibitor. Inhibitors were routinely dissolved in DMSO and were then diluted with buffer; final DMSO concentrations never exceeded 1%. This concentration did not influence [¹²⁵I]APNEA binding. Incubations were carried out in duplicate for 1 h at 37 °C and were terminated by rapid filtration over Whatman GF/B filters, using a Brandell cell harvester (Brandell, Gaithersburg, MD). Tubes were washed three times with 3 mL of buffer. Radioactivity was determined in a Beckman gamma 5500B γ-counter. Nonspecific binding was determined in the presence of 40 μM R-PIA. K_i values were calculated according to Cheng-Prusoff,²⁷ assuming a K_d for [¹²⁵I]APNEA of 17 nM.⁶ Alternately, [¹²⁵I]-4-amino-3-iodobenzyladenosine-5′-N-methyluronamide²⁸ was used as the radioligand (final concentration 0.1 nM, K_d = 1.6 nM) with no significant difference in K_i values compared to [¹²⁵I]APNEA.

Binding of [³H]PIA to A₁ receptors from rat brain membranes and of [³H]CGS 21680 to A₂ receptors from rat striatal membranes was performed as described previously.^8,23

Rat cerebral cortical membranes and striatal membranes were prepared¹⁸ and treated with adenosine deaminase (2 units/mL) for 30 min at 37 °C prior to storage at −70 °C. Solid samples of the adenosine derivatives were dissolved in DMSO and stored in the dark at −20 °C. The stock solutions were diluted with DMSO to a concentration of ≤0.1 mM prior to adding to the aqueous medium. The final concentration of DMSO in the assay medium was generally 2%.

At least six different concentrations spanning 3 orders of magnitude, adjusted appropriately for the IC₅₀ of each compound, were used. IC₅₀ values, computer-generated using a nonlinear regression formula on the GraphPAD program (Institute for Scientific Information), were converted to apparent K_i values using K_d values²³ of 1.0 and 14 nM for [³H]PIA and [³H]CGS 21680 binding, respectively, and the Cheng-Prusoff equation.²⁷