Structure-activity relationships of 2-chloro-N6-substituted- 4′-thioadenosine-5′-uronamides as highly potent and selective agonists at the human A3 adenosine receptor

Article abstract of DOI:10.1021/jm050595e

We have established structure-activity relationships of novel 4′-thionucleoside analogues as the A3 adenosine receptor (AR) agonists. Binding affinity, selectivity toward other AR subtypes, and efficacy in inhibition of adenylate cyclase were studied. Fro

Full text of DOI:10.1021/jm050595e

J. Med. Chem. 2006, 49, 273-281
273
Structure-Activity Relationships of 2-Chloro-N6-substituted-4′-thioadenosine-5′-uronamides as
Highly Potent and Selective Agonists at the Human A3 Adenosine Receptor
Lak Shin Jeong,*,† Hyuk Woo Lee,‡ Kenneth A. Jacobson,§ Hea Ok Kim,† Dae Hong Shin,† Jeong A Lee,† Zhan-Guo Gao,§
Changrui Lu,§ Heng T. Duong,§ Prashantha Gunaga,† Sang Kook Lee,† Dong Zhe Jin,‡ Moon Woo Chun,‡ and
Hyung Ryong Moon|
Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha Womans UniVersity, Seoul 120-750, Korea, College of Pharmacy, Seoul
National UniVersity, Seoul 151-742, Korea, Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of
Diabetes, and DigestiVe and Kidney Disease, National Institute of Health, Bethesda, Maryland 20892, and College of Pharmacy, Pusan
National UniVersity, Pusan 609-753, Korea
ReceiVed June 23, 2005
We have established structure-activity relationships of novel 4′-thionucleoside analogues as the A3 adenosine
receptor (AR) agonists. Binding affinity, selectivity toward other AR subtypes, and efficacy in inhibition of
adenylate cyclase were studied. From this study, 2-chloro-N6-methyl-4′-thioadenosine-5′-methyluronamide
(36a) emerged as the most potent and selective agonist at the human A3 AR. We have also revealed that,
similar to 4′-oxoadenosine analogues, at least one hydrogen on the 5′-uronamide moiety was necessary for
high-affinity binding at the human A3 AR, presumably to allow this group to donate a H bond within the
binding site. Furthermore, bulky substituents on the 5′-uronamide reduced binding affinity, but in some
cases large 5′-uronamide substituents, such as substituted benzyl and 2-phenylethyl groups, maintained
moderate affinity with reduced efficacy, leading to A3 AR partial agonists or antagonists. In several cases
for which the corresponding 4′-oxonucleosides have been studied, the 4′-thionucleosides showed higher
binding affinity to the A3 AR.
Introduction
affinity to the A2A AR, compared to 1, while the opposite trend
was observed in the case of 2. This study indicated that the
binding site of the human A3 AR preferred the North sugar ring
puckering.
Adenosine is an endogenous signaling molecule regulating
many physiological functions through four specific subtypes (A1,
A2A, A2B, and A3) of adenosine receptors (ARs), at least one of
Recently, on the basis of a bioisosteric rationale and the high
binding affinity and selectivity of compounds 1-3, we have
reported the 2-chloro-N6-substituted-4′-thioadenosine-5′-methyl-
uronamides as highly potent and selective A3 AR agonists,
among which 4 (LJ-530) showed extremely high binding affinity
(Ki ) 0.28 nM) at the human A3 AR with high selectivity to
the human A1 and human A2A ARs.15 Compound 5 (LJ-529)
has also exhibited very high binding affinity (Ki ) 0.38 nM) at
the human A3 AR but less selectivity than 4 to the human A1
and A2A ARs.
Thus, on the basis of high binding affinity and selectivity of
4′-thionucleosides 4 and 5,15 we carried out an extensive
structure-activity relationship study of the N6-position and/or
methyluronamide of 4 and 5 to search for better A3 AR agonists
(Figure 1). We found that electronic as well as steric effects
play an important role in binding affinity and selectivity to the
ARs. In this paper, we report a full account of 4′-thioadenosine
derivatives as highly potent and selective A3 AR agonists.
which is expressed on the surface of most cells.1 Among these
four subtypes, the A3 subtype is the most recently identified
and is known to be involved in many diseases, such as cardiac2
and cerebral ischemia,3 cancer,4 asthma,5 glaucoma,6 and
inflammation.7,8 Thus, the A3 AR is regarded as a good
therapeutic target for the development of clinically useful agents.
A number of nucleoside and nonnucleoside derivatives have
been synthesized and tested for binding affinities at the human
A1, A2A, A2B, and A3 ARs.9-11 Among them, 1 (IB-MECA)11
showed potent binding affinity to the human A3 AR (Ki ) 1.0
nM) with 51- and 2900-fold selectivity for the human A1 and
A2A ARs, respectively, with binding affinity at the human A2B
AR >10 µM (Chart 1). The systematic structure-activity
relationship study of compound 1 led to 2 (Cl-IB-MECA),12
which showed increased selectivity in comparison to 1, while
exhibiting similar binding affinity (Ki ) 1.4 nM) to the human
A3 AR. Compound 1 was found to show potent in vivo
antitumor activity4 and is now undergoing phase II clinical trials,
and compound 2 is being used extensively as a pharmacological
tool for studying A3 AR.13
Results and Discussion
Chemistry. For an extensive structure-activity relationship
study of 4′-thionucleosides, we first synthesized the glycosyl
donors 4-thiosugars 14 or 15 on a preparative scale, which was
achieved, starting from D-gulonic γ-lactone (6) as shown in
Scheme 1.15 Compound 6 was converted to the diacetonide 7
by treating with acetone and anhydrous copper(II) sulfate under
acidic conditions. Treatment of 7 with LAH in ether followed
by treating the resulting diol 8 with mesyl chloride afforded
the dimesylate 9 in excellent yield. Cyclization to the 4-thio-
furanose 10 was achieved by heating 9 with anhydrous sodium
sulfide in DMF. Treatment of 10 with 30% acetic acid at room
On the other hand, a cyclopropyl-fused carbocyclic nucleoside
3 (MRS 3558),14 adopting a fixed conformation of the pseudo-
ribose ring, exhibited higher binding affinity (Ki ) 0.29 nM)
than compounds 1 and 2 at the human A3 AR. Compound 3
showed better selectivity for the A1 AR and similar binding
* To whom correspondence should be addressed: Tel: 82-2-3277-
3466. Fax: 82-2-3277-2851. E-mail: lakjeong@ewha.ac.kr.
† Ewha Womans University.
‡ Seoul National University.
§ National Institute of Health.
| Pusan National University.
10.1021/jm050595e CCC: $33.50 © 2006 American Chemical Society
Published on Web 11/19/2005

274 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Jeong et al.
Chart 1. Binding Affinity (A1, A2, or A3 AR) or Functional Activation of the Adenosine Derivatives at Human Adenosine Receptorsa
a Ki (nM) values were determined (radioligand) in membrances from mammalian cells expressing hA1 AR ([3H]R-PIA), hA2 AR ([3H]CGS21680), or hA3
AR (125I-AB-MECA). EC50 values at the hA2B AR in a cAMP assay were determined in stably transfected CHO cells.11,15,20,21
resulting diols with TBSOTf yielded disilyl ethers, which were
debenzoylated with sodium methoxide to give the 4′-hydroxy-
methyl analogues 23-28, respectively. Oxidation of 23-28 with
PDC in DMF gave the acid derivatives 29-34, respectively.
Various N6-substituted acid derivatives 29-34 were converted
to the various uronamides 35-40g, as shown in Scheme 3. In
the previous paper,15 acids were converted to the corresponding
Figure 1. Structure of the desired 4′-thionucleosides (R ) alkyl or
5′-methyluronamides upon treatment with potassium carbonate
and dimethyl sulfate in acetone followed by reacting the
resulting esters with 40% methylamine, but an improved method
was now employed for the synthesis of the uronamides. Direct
coupling of the N6-methyladenosine derivative 30 with various
amines such as methylamine, dimethylamine, cyclopropylamine,
cyclopropylmethylamine, isoamylamine, morpholine, 4-ben-
zylpiperidine, 4-(4-fluorophenyl)piperazine, 3-fluorobenzyl-
amine, 2-(3-fluorophenyl)ethylamine, and 3,3-diphenylpropyl-
amine using EDC and HOBt afforded various 2-chloro-N6-
methyladenosine-5′-uronamides 36a-k after desilylation. Simi-
larly, N6-cyclopropyl-, cyclopentyl-, 3-iodobenzyl-, and 2-me-
thylbenzyladenosine derivatives 31-34 were also converted to
the various uronamides 37a-e, 38a,b, 39a-m, and 40a-g,
respectively.
arylalkyl, R′ ) aminoalkyl or aminoarylalkyl).
temperature for 2 h gave 11 in 90% yield, on the basis of the
recovered starting material.
Oxidative cleavage of diol 11 with lead tetraacetate gave an
aldehyde, which was reduced with sodium borohydride to give
12. Protection of 12 with a benzoyl group afforded 13.
Treatment of 13 with mCPBA in methylene chloride gave the
sulfoxide 14, which served as a glycosyl donor for the
condensation with nucleobases. The sulfoxide 14 was converted
to another glycosyl donor 15 by refluxing with acetic anhydride
at 100 °C.
The glycosyl donor 14 was directly condensed with 2,6-
dichloropurine in the presence of TMSOTf and triethylamine
in a solution of acetonitrile and 1,2-dichloroethane to give the
â-anomer 16 (54%) with trace amounts of the R-anomer, while
condensation of another glycosyl donor 15 with silylated 2,6-
dichloropurine produced the same â-anomer 16 (60%) and its
R-anomer in a 9:1 ratio,15 but the two-step yield from sulfoxide
14 was only 37% (Scheme 2). Anomeric configurations were
Biological Activity. All AR experiments were performed
using adherent CHO (Chinese hamster ovary) cells stably
transfected with cDNA encoding the human ARs.11 Binding at
the human A3 AR in this study was carried out using [125I]I-
AB-MECA [N6-(4-amino-3-iodobenzyl)adenosine-5′-N-ethyl-
carboxamidoadenosine] as a radioligand. In cases of weak
binding, the percent inhibition of radioligand binding to the
human A3 AR was determined at 10 µM. Furthermore, the
percent activation of the human A3 AR (inhibition of adenylate
cyclase in comparison to the full agonist 2) was determined at
10 µM. Binding at human A1 [using [3H]R-PIA, N6-(2-phenyl-
isopropyl)adenosine, or [3H]NECA, 5′-N-ethylcarboxamido-
adenosine] or A2A AR [using [3H]CGS21680, (2-[p-(2-carboxy-
ethyl)phenylethylamino]-5′-N-ethylcarboxamidoadenosine] was
carried out as described in the Experimental Section.
1
easily assigned by H NOE experiments between 1′-H and 4′-
H. To examine the effects of N6-substituents and/or 5′-
uronamides in binding to the binding site of the A3 AR,
compound 16 was treated with various amines such as ammonia,
methylamine, cyclopropylamine, cyclopentylamine, 3-iodoben-
zylamine, and 2-methylbenzylamine to afford various N6-
substituted purine nucleosides 17-22. For the synthesis of the
various uronamides, N6-substituted nucleosides 17-22 were first
converted to the corresponding acids 29-34. Treatment of 17-
22 with 80% aqueous acetic acid followed by treating the

4′-Thionucleosides as A3 AR Agonists
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 275
Scheme 1. Synthesis of Thiosugar Intermediatesa
a Reagents: (a) CH3COCH3, H2SO4, CuSO4, rt; (b) LiAlH4, ether; (c) MsCl, Et3N, CH2Cl2; (d) Na2S, DMF, heat; (e) 30% AcOH; (f) Pb(OAc)4, EtOAc;
(g) NaBH4, MeOH; (h) BzCl, pyridine; (i) m-CPBA, CH2Cl2; (j) Ac2O.
Scheme 2. Synthesis of 4′-Thionucleoside 5′-Carboxylic Acid Intermediatesa
a Reagents and conditions: (a) 2,6-dichloropurine, TMSOTf, Et3N, CH3CN-ClCH2CH2Cl, rt to 83 °C; (b) silylated 2,6-dichloropurine, TMSOTf,
ClCH2CH2Cl, rt to 80 °C; (c) R1NH2, Et3N, EtOH, rt; (d) 80% AcOH, 70 °C; (e) TBSOTf, pyridine, 50 °C; (f) NaOMe, MeOH; (g) PDC, DMF.
As shown in Table 1, a variety of N6-alkyl, cycloalkyl, and
arylalkyl substituents on 4′-thioadenosine-5′-uronamide deriva-
tives have produced high affinity at the human A3 AR subtype,
among which 36a (R ) CH3, R′ ) NHCH3)15 showed the most
potent binding affinity (Ki ) 0.28 ( 0.09 nM) at the A3 AR.
Compound 36a was selective for the human A3 AR versus the
human A1 AR by 4800-fold and showed extremely low binding
affinity at the human A2A AR (20% inhibition at 10 µM). It
was also selective for the human A3 AR versus the rat A1 and
rat A2A ARs by 700- and 23 000-fold, respectively.15 However,
we have not determined the binding affinity of 36a at rat A3
AR, since it is expected to be weak due to the presence of the
N6-methyl group. The 4′-thionucleosides 36a and its N6-(3-
iodobenzyl) analogue 39a showed higher binding affinity to the
human A3 AR and higher selectivity in comparison to the human
A1 and A2A ARs than the corresponding 4′-oxonucleosides 1
and 2. However, due to the presence of the small N6-alkyl group,
the selectivity of the N6-methyl derivative 36a is not expected
to be general across species.
All compounds (36b,f-h, 37d,e, 39g-i, 40c,f,g) with dialkyl
substitution at the 5′-uronamides showed reduced binding
affinity at all the subtypes of ARs, indicating that the amide
hydrogen, important for hydrogen bonding in the binding site
of ARs, contributes to the observed affinity. Molecular modeling
of adenosine receptors based on the template of the high-
resolution structure of rhodopsin has already predicted that such
a H-bond occurs between the exocyclic NH and the side chain
of a conserved Asn residue in the sixth transmembrane helical
domain of several ARs.11a,17 In the case of monoalkyl 5′-
uronamide derivatives, bulky substituents generally reduced
binding affinity at the human A3 AR, regardless of the
substituent present at the N6 position. A variety of N6-alkyl,
cycloalkyl, and arylalkyl substituted 4′-thio-5′-uronamide ad-
enosine derivatives have produced moderate to high affinity at

276 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Jeong et al.
Scheme 3. Synthesis of 4′-Thionucleoside 5′-Uronamide Derivativesa
a Reagents: (a) NHR2R3, EDC, HOBT, DIPEA, CH2Cl2; (b) TBAF, THF.
the human A3 AR subtype. Among all the synthesized 5′-
uronamide derivatives, 5′-methyluronamide derivatives exhibited
the most potent binding affinity at the human A3 AR, in which
the binding affinity of the N6 substituted compounds was in
the following order: CH3 > 3-iodobenzyl > H > 2-methyl-
benzyl > cyclopropyl > cyclopentyl. 5′-Uronamide substituents
consisting of C1-C5 alkyl and cycloalkyl groups were capable
of maintaining namomolar affinity and in some cases selectivity
for the human A3 AR in comparison to A1 and A2A ARs. This
conclusion applied to N6-methyl, cyclopropyl, and 3-iodobenzyl
derivatives, but not to all N6-substitution. While the 5′-N-
cyclopropyl and cyclopropylmethyl uronamide derivatives
displayed Ki values of 1-3 nM at the human A3 AR in the
above N6 series, the corresponding N6-(2-methylbenzyl) ana-
logues were at least an order of magnitude less potent in binding.
Thus, the effects of substitution at N6 and 5′-positions were
interdependent. To further illustrate this point, 5′-morpholin-
eamides were ∼10-fold less potent in other N6-series than was
the N6-(3-iodobenzyl) derivative 39i. A 5′-(4-benzylpiperidine)
derivative 40g was considerably weaker in A3 AR binding in
the N6-(2-methylbenzyl) series than in other N6-series. A 5′-N-
(3-methylbutyl) uronamide derivative 36e was also highly potent
and selective in binding to the A3 AR. 5′-Uronamide substituents
consisting of substituted benzyl and 2-phenylethyl groups
generally resulted in intermediate affinity at the A3 AR.
Compounds 35, 36a,d, 38a, 39a,b,f, and 40b,e were full
agonists in an assay of human A3 receptor-mediated inhibition
of cyclic AMP in transfected CHO cells.16 However, in some
cases large 5′-uronamide substituents reduced efficacy, leading
to A3 AR partial agonists or antagonists. For example, a 5′-N-
(3-fluorobenzyl) uronamide derivative 36i was a potent and
selective partial agonist of the A3 AR. The corresponding 5′-
N-(2-(3-fluorophenyl)ethyl) uronamide derivative 36j, although
5-fold less potent in binding, appeared to display less intrinsic
efficacy toward the A3 AR. The corresponding 5′-N-(3,3-
diphenylpropyl) uronamide derivative 36k displayed only
moderate binding affinity and appeared to be a putative
antagonist at the A3 AR due to its ability to bind but not activate
the receptor. In the series of N6-(3-iodobenzyl) derivatives, 39c
and 39l appeared to be either low efficacy agonists or
antagonists. The 5′-N-(3-methylbutyl) uronamide 39c was also
selective in binding to the A3 AR (Ki ) 42 nM).
The reduction of intrinsic efficacy has been demonstrated
previously for a variety of sterically bulky groups at N6- and
C2-positions11 and for smaller substituents, such as thioethers,
at the 5′-position.10 These results are novel in that the range of
5′-substitutents found to maintain moderate binding affinity at
the human the A3 AR has been greatly expanded, leading to an
initial structure-efficacy relationship.
Conclusion
We have established structure-activity relationships of novel
4′-thio analogues of compound 2. An improved synthetic route
to this series involved the direct condensation of the sulfoxide
14 with 2,6-dichloropurine, which was efficiently synthesized
on a preparative scale from D-gulonic γ-lactone (5). This route
was superior to the previous route based on condensation using
4-thioribosyl acetate 15 as an intermediate. From this study,
2-chloro-N6-methyladenosine 5′-methyluronamide (36a) emerged
as the most potent and selective agonist at the human A3 AR.
We have also revealed that, similar to 4′-oxoadenosine ana-
logues, at least one hydrogen on the 5′-uronamide moiety is
necessary for nanomolar binding affinity at the human A3 AR,
presumably to allow this group to donate a H bond within the

4′-Thionucleosides as A3 AR Agonists
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 277
Table 1. Potency of 2-Chloro-4′-thioadenosine-5′-uronamide Derivatives at Human A1, A2A, and A3 AR and Maximal Agonist Effects at Human A3 AR
Expressed in CHO Cellsa
Ki (nM) or % inhibition at 10 µM
hA3 AR
% activationb
compd
R
R′
hA1 AR
hA2A AR
Ki (nM)
at 10 µM
35
H
NHCH3
NHCH3
N(CH3)2
cyclopropyl-NH
cyclopropylmethyl-NH
isoamyl-NH
89.2 ( 11.7
1330 ( 240
10%
47.8 ( 5.7
638 ( 35
4070 ( 1220
4%
10%
28%
67%
15%
158 ( 29
0.40 ( 0.06
0.28 ( 0.09
1500 ( 1300
2.82 ( 1.03
1.10 ( 0.03
1.63 ( 0.17
3870 ( 580
3500 ( 340
2700 ( 880
17.4 ( 3.8
85.6 ( 35.6
415 ( 16
2.24 ( 1.21
0.67 ( 0.07
5.56 ( 1.77
4020 ( 740
4440 ( 160
4.27 ( 0.33
2.83 ( 0.63
0.38 ( 0.07
0.89 ( 0.18
41.9 ( 11.3
2.96 ( 1.03
3.64 ( 0.60
18.2 ( 13.4
878 ( 285
1440 ( 830
510 ( 69
100 ( 5
119 ( 12
8.3 ( 5.9
98 ( 8
36a
36b
36c
36d
36e
36f
36g
36h
36i
36j
36k
37a
37b
37c
37d
37e
38a
38b
39a
39b
39c
39d
39e
39f
39g
39h
39i
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopropyl
cyclopentyl
cyclopentyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
3-iodobenzyl
2-methylbenzyl
2-methylbenzyl
2-methylbenzyl
2-methylbenzyl
2-methylbenzyl
2-methylbenzyl
2-methylbenzyl
20%
12%
2770 ( 580
44%
14%
24%
6%
34%
41%
30%
15%
4890 ( 380
604 ( 110
325 ( 13
33%
109 ( 9
92 ( 4
morpholine
38 ( 6
4-Benzyl -piperidine
4-(4-F-Ph)-piperazine
3-F-benzyl-NH
2-(3-F-phenyl)ethyl-NH
3,3-diphenyl-propyl-NH
NHCH3
3.1 ( 1.3
2.9 ( 4.0
57 ( 2
11 ( 5
20%
0.1 ( 2.8
99 ( 3
37.3 ( 2.5
3.2 ( 0.3
4.6 ( 0.3
32%
NHCH2CH3
97 ( 3
cyclopropyl-NH
4-benzylpiperidine
morpholine
NHCH3
NHCH2CH3
NHCH3
NHCH2CH3
isoamyl-NH
cyclopropyl-NH
cyclopropylmethyl-NH
cyclobutyl-NH
4-benzylpiperidine
4-(4-F-Ph)-piperazine
morpholine
3-Cl-benzyl-NH
3-(trifluoromethyl)benzyl-NH
2-phenylethyl-NH
3,3-diphenylpropyl-NH
NHCH3
NHCH2CH3
N(CH3)2
cyclopropyl-NH
cyclopropylmethyl-NH
morpholine
85 ( 4
0
25%
11%
83 ( 5
105 ( 2
95 ( 3
114 ( 9
100 ( 6
12 ( 6
87 ( 4
96 ( 9
13.9 ( 1.8
2.0 ( 2
193 ( 46
144 ( 55
31%
92.0 ( 10.9
1720 ( 690
126 ( 2
38%
24%
18%
1710 ( 340
0%
921 ( 76
149 ( 29
223 ( 36
292 ( 120
31%
326 ( 23
2220 ( 420
549 ( 78
17%
12%
4%
28%
11%
100 ( 3
74 ( 19
96 ( 10
94 ( 6
39j
39k
39l
308 ( 148
354 ( 18
433 ( 141
343 ( 48
2.18 ( 0.46
2.50 ( 1.10
632 ( 70
33 ( 9
31 ( 11
7.1 ( 2.7
31 ( 14
99 ( 6
20%
5%
11%
3%
39m
40a
40b
40c
40d
40e
40f
40g
507 ( 66
75.8 ( 2.5
0%
76.7 ( 9.3
1980 ( 310
4%
1320 ( 270
429 ( 97
22.5%
202 ( 51
3890 ( 630
17%
112 ( 9
9.1 ( 0.4
91 ( 3
111 ( 11
61 ( 4
31 ( 4
27.8 ( 3.8
29.7 ( 11.3
7670 ( 800
4-benzylpiperidine
26%
15%
49200 ( 17500
a All AR experiments were performed using adherent CHO cells stably transfected with cDNA encoding the human or rat ARs. Percent activation of the
human A3 AR was determined at 10 µM. Binding was carried out as described in the Experimental Section using as radioligand [3H]R-PIA or [3H]NECA
at the human A1 AR and [3H]CGS 21680 at the human A2A AR. Values from the present study are expressed as mean ( SEM, n ) 3-5. b Percent activity
(inhibition of foskolin-stimulated cyclic AMP formation) at 10 µM, relative to 10 µM Cl-IB-MECA (A3).
chromatography was performed using silica gel 60 (230-400 mesh).
binding site. Furthermore, bulky substituents on the 5′-urona-
mide reduced binding affinity, although in some cases large 5′-
uronamide substituents, such as substituted benzyl and 2-phe-
nylethyl groups, maintained moderate affinity with reduced
efficacy, leading to A3 AR partial agonists or antagonists. In
several cases for which the corresponding 4′-oxonucleosides
have been studied, the 4′-thionucleosides showed higher binding
affinity to the A3 AR. These findings may facilitate the
identification of the modes of binding of nucleoside derivatives
at ARs.
Anhydrous solvents were purified by the standard procedures.
Synthesis. Methanesulfonic Acid (S)-((4R)-2,2-Dimethyl-[1,3]-
dioxolan-4-yl)((4S,5S)-5-methanesulfonyloxymethyl-2,2-dimeth-
yl-[1,3]dioxolan-4-yl)methyl Ester (9). To a stirred slurry of 6
(20.2 g, 113.5 mmol) and anhydrous CuSO4 (27.0 g, 169.8 mmol)
in dry acetone (650 mL) was added concentrated H2SO4 (1.7 mL)
and the mixture was stirred at room temperature for 24 h. The pH
of the solution was adjusted to 7 with Ca(OH)2, and the resulting
slurry was filtered and evaporated to afford 7 (28.7 g, 98%) as a
light-yellow solid, which was used in the next step without further
purification. To a stirred solution of 7 (25.1 g, 97.2 mmol) in
anhydrous THF (300 mL) was cautiously added LAH (7.2 g, 190.4
mmol) in several portions at 0 °C and the reaction mixture was
stirred at room temperature for 5 h. The reaction mixture was
quenched with ice, dried over anhydrous MgSO4, filtered through
a Celite pad, and evaporated. The resulting residue was purified
Experimental Section
1
General. Melting points are uncorrected. H NMR (400 MHz)
and 13C NMR (100 MHz) spectra were measured in CDCl3 or CD3-
OD and chemical shifts are reported in parts per million (δ)
downfield from tetramethylsilane as internal standard. Column

278 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Jeong et al.
by silica gel column chromatography (hexane/ethyl acetate ) 1:1)
to give 8 (24.1 g, 95%) as a white semisolid: 1H NMR (CDCl3) δ
1.29 (s, 3 H, CH3), 1.29 (s, 3 H, CH3), 1.36 (s, 3 H, CH3), 1.43 (s,
3 H, CH3), 3.26 (s, 1 H, OH), 3.28 (s, 1 H, OH), 3.36-3.79 (m, 4
H, HOCH2, OCH2), 3.96-4.03 (m, 2 H, HOCH2CH(OR)R′,
OCHRR′), 4.12-4.22 (m, 2 H). Anal. (C12H22O6) C, H.
To a solution of 8 (4.1 g, 15.7 mmol) in dry pyridine (20 mL)
was added methanesulfonyl chloride (6.1 mL, 47.1 mmol) at 0 °C
and the mixture was stirred at the same temperature for 4 h. The
reaction mixture was quenched with ice and the reaction mixture
was partitioned between EtOAc and H2O. The organic layer was
washed with aqueous NaHCO3 solution and brine, dried over
anhydrous MgSO4, and evaporated. The residue was purified by
silica gel column chromatography (hexane/ethyl acetate ) 2:1) to
give 9 (6.4 g, 98%) as a colorless syrup: 1H NMR (CDCl3) δ 1.37
(s, 3 H, CH3), 1.39 (s, 3 H, CH3), 1.46 (s, 3 H, CH3), 1.52 (s, 3 H,
CH3) 3.10 (s, 3 H, SO2CH3), 3.18 (s, 3 H, SO2CH3), 3.92-4.08
(dd, 1 H, J ) 7.1, 11.7 Hz, ROCHH), 4.08-4.20 (m, 2 H,
ROCHHCH(OR′)), 4.39-4.49 (m, 4 H, MsOCH2CH(OR)CH(OR′)),
4.81-4.87 (dd, 1 H, J ) 4.8, 6.6 Hz, MsOCHRR′). Anal.
(C14H26O10S2) C, H, S.
organic layer was dried over anhydrous MgSO4, filtered, and
evaporated. The resulting residue was purified by silica gel column
chromatography (hexane/ethyl acetate ) 2:1) to give 12 (5.5 g,
98%) as a syrup: 1H NMR (CDCl3) δ 1.33 (s, 3 H, CH3), 1.53 (s,
3 H, CH3), 2.41 (br s, 1 H, OH), 2.89 (dd, 1 H, J ) 1.5, 12.9 Hz,
6-CHH), 3.09 (dd, 1 H, J ) 4.9, 12.7 Hz, 6-CHH), 3.44 (td, 1 H,
J ) 1.0, 6.6 Hz, 4-H), 3.59 (d, 2 H, J ) 5.4 Hz, HOCH2), 4.71
(dd, 1 H, J ) 1.2, 5.6 Hz, 6a-H), 4.91 (td, 1 H, J ) 1.5, 5.3 Hz,
3a-H); FAB-MS m/z 190 (M+). Anal. (C8H14O3S) C, H, S.
Benzoic Acid (3aS,4R,6aR)-2,2-Dimethyltetrahydrothieno[3,4-
d][1,3]dioxol-4-ylmethyl Ester (13). To a solution of 12 (2.1 g,
11.1 mmol) in pyridine (20 mL) was added benzoyl chloride (1.9
g, 13.4 mmol) at 0 °C and the reaction mixture was stirred at room
temperature for 6 h, quenched with methanol, and evaporated. The
residue was dissolved in ether (50 mL) and the insoluble pyridinium
salt was filtered and washed with ether. The combined organic layer
was dried over anhydrous MgSO4, filtered, and evaporated. The
resulting residue was purified by silica gel column chromatography
(hexane/ethyl acetate ) 4:1) to afford 13 (3.2 g, 99%) as a colorless
oil: 1H NMR (CDCl3) δ 1.26 (s, 3 H, CH3), 1.47 (s, 3 H, CH3),
2.87 (dd, 1 H, J ) 1.5, 13.2 Hz, 6-CHH), 3.11 (dd, 1 H, J ) 4.9,
13.2 Hz, 6-CHH), 3.57 (m, 1 H, 4-H), 4.25 (dd, 1 H, J ) 8.0, 11.4
Hz, BzOCHH), 4.35 (dd, 1 H, J ) 5.8, 11.4 Hz, BzOCHH), 4.72
(dd, 1 H, J ) 1.2, 5.6 Hz, 6a-H), 4.91 (td, 1 H, J ) 1.2, 4.4 Hz,
3a-H), 7.35-7.99 (m, 5 H, Ph); FAB-MS m/z 294 (M+). Anal.
(C15H18O4S) C, H, S.
Benzoic Acid (3aS,4R,6R,6aR)- and (3aS,4R,6S,6aR)-6-Ac-
etoxy-2,2-dimethyltetrahydrothieno[3,4-d][1,3]dioxol-4-ylmeth-
yl Ester (15). To a stirred solution of 13 (1.4 g, 4.6 mmol) in
CH2Cl2 (30 mL) was added a solution of m-CPBA (1.0 g, 4.6 mmol,
80%) in CH2Cl2 (15 mL) dropwise at -78 °C and the mixture was
stirred at the same temperature for 45 min. The reaction mixture
was quenched with aqueous saturated NaHCO3 solution and
extracted with CH2Cl2, and the organic layer was washed with brine,
dried over anhydrous MgSO4, filtered, and evaporated. The residue
was purified by silica gel column chromatography (hexane/ethyl
acetate ) 2:1) to give 14 (1.4 g, 95%) as a white solid: 1H NMR
(CDCl3) of one isomer δ 1.33 (s, 3 H, CH3), 1.49 (s, 3 H, CH3),
3.23 (dd, 1 H, J ) 4.1, 14.4 Hz, 6-CHH), 3.38 (dd, 1 H, J ) 6.3,
14.4 Hz, 6-CHH), 3.47 (m, 1 H, 4-H), 4.73 (dd, 1 H, J ) 9.0,
11.9, Hz, BzOCHH), 4.89 (dd, 1 H, J ) 4.9, 11.9 Hz, BzOCHH),
5.02 (t, 1 H, J ) 6.1 Hz, 3a-H), 5.24 (m, 1 H, 6a-H), 7.41-8.03
(m, 5 H, Ph).
A solution of 14 (3.5 g, 11.3 mmol) in Ac2O (90 mL) was heated
at 100 °C for 6 h and evaporated. The residue was partitioned
between ethyl acetate and H2O. The organic layer was washed with
saturated aqueous NaHCO3 solution and brine, dried over anhydrous
MgSO4, filtered, and evaporated. The residue was purified by silica
gel column chromatography (hexane/ethyl acetate ) 3:1) to give
15 (2.5 g, 62%) as a syrup: 1H NMR (CDCl3) of one isomer δ
1.30 (s, 3 H, CH3), 1.50 (s, 3 H, CH3), 2.03 (s, 3 H, COCH3), 3.77
(dd, 1 H, J ) 5.8, 9.5 Hz, 4-H), 4.37 (dd, 1 H, J ) 9.8, 11.4 Hz,
BzOCHH), 4.23 (dd, 1 H, J ) 6.1, 11.4 Hz, BzOCHH), 4.94 (d, 1
H, J ) 5.6 Hz, 3a-H), 4.98 (d, 1 H, J ) 5.6 Hz, 6a-H), 6.06 (s, 1
H, 6-H), 7.42-8.06 (m, 5 H, Ph); FAB-MS m/z 293 (M+ - OAc).
Anal. (C17H20O6S) C,H, S.
Benzoic Acid (3aS,4R,6R,6aR)-6-(2,6-Dichloropurin-9-yl)-2,2-
dimethyltetrahydrothieno[3,4-d][1,3]dioxol-4-ylmethyl Ester (16)
and Its r-Isomer. To a suspension of 2,6-dichloropurine (9.93 g,
64.2 mmol) in a mixture of dry CH3CN (100 mL) and 1,2-
dichloroethane (50 mL) were added triethylamine (9.0 mL, 64.2
mmol) and TMSOTf (23.2 mL, 128.5 mmol), and the mixture was
stirred at room temperature until the mixture was clear. The resulting
solution was added to a solution of 14 (10.0 g, 32.1 mmol) in dry
1,2-dichloroethane (50 mL) dropwise. An additional amount of
triethylamine (9.0 mL, 64.2 mmol) in dry 1,2-dichloroethane (25
mL) was added to the reaction mixture to initiate the Pummerer
reaction. After being stirred at room temperature for 5 min, the
reaction mixture was heated at 83 °C for 24 h. The reaction was
quenched with ice and the reaction mixture was partitioned between
EtOAc and H2O. The separated organic layer was washed with
(3aS,4R,6aR)-4-((4R)-2,2-Dimethyl-[1,3]dioxolan-4-yl)-2,2-
dimethyltetrahydrothieno[3,4-d][1,3]dioxole (10). To a stirred
solution of 9 (10.1 g, 24.2 mmol) in DMF (260 mL) was added
sodium sulfide nonahydrate (6.97 g, 29.0 mmol) and the mixture
was heated at 100 °C for 3 h. After being cooled to room
temperature, the mixture was diluted with ether and washed with
H2O (three times) and brine. The organic layer was dried over
anhydrous MgSO4, filtered, and evaporated. The resulting residue
was purified by silica gel column chromatography (hexane/ethyl
1
acetate ) 5:1) to give 10 (5.9 g, 94%) as a syrup: H NMR (CDCl3)
δ 1.33 (s, 3 H, CH3), 1.34 (s, 3 H, CH3), 1.44 (s, 3 H, CH3), 1.52
(s, 3 H, CH3), 2.88 (d, 1 H, J ) 12.7 Hz, 6-CHH), 3.09 (td, 1 H,
J ) 2.2, 13.1 Hz, 6-CHH), 3.24 (d, 1 H, J ) 8.8 Hz, 4-H), 3.76
(dd, 1 H, J ) 5.7, 8.6 Hz, 5′-CHH), 3.98 (td, 1 H, J ) 6.0, 9.0 Hz,
4′-H). 4.15 (dd, 1 H, J ) 6.3, 8.5 Hz, 5′-CHH), 4.93 (d, 2 H, J )
1.9 Hz, 3a-H, 6a-H); 13C NMR (CDCl3) δ 23.8, 24.6, 25.6, 26.1,
36.9, 56.5, 68.3, 75.6, 82.3, 84.5; FAB-MS m/z 260 (M+). Anal.
(C12H20O4S) C, H, S.
(1R)-1-((3aS,4R,6aR)-2,2-Dimethyltetrahydrothieno[3,4-d]-
[1,3]dioxol-4-yl)ethane-1,2-diol (11). A solution of 10 (5.4 g, 20.1
mmol) in 30% aqueous AcOH (150 mL) was stirred at room
temperature for 4 h and the reaction mixture was evaporated. The
resulting residue was purified by silica gel column chromatography
(hexane/ethyl acetate ) 1:1) to give 11 (1.89 g, 43%) and recovered
starting material 9 (2.82 g): 1H NMR (CDCl3) δ 1.34 (s, 3 H, CH3),
1.53 (s, 3 H, CH3), 2.90 (dd, 1 H, J ) 2.2, 12.9 Hz, 6′-CHH), 3.09
(dd, 1 H, J ) 4.8, 12.9 Hz, 6′-CHH), 3.27 (dd, 1 H, J ) 1.9, 7.8
Hz, 4′-H), 3.49 (br s, 2 H, 2′ × OH), 3.58-3.72 (m, 2 H,
HOCHHCH(OH)), 3.79 (dd, 1 H, J ) 2.9, 11.0 Hz, HOCHH), 4.93
(m, 2 H, 3a-H, 6a-H). Anal. (C9H16O4S) C, H, S.
(3aS,4R,6aR)-(2,2-Dimethyltetrahydrothieno[3,4-d][1,3]dioxol-
4-yl)methanol (12). To a stirred solution of 11 (2.5 g, 11.2 mmol)
in ethyl acetate (50 mL) was added Pb(OAc)4 (5.4 g, 12.3 mmol)
at 0 °C and the reaction mixture was stirred for 10 min, at which
time TLC indicated the absence of starting material. The reaction
mixture was filtered, the filtrate was diluted with EtOAc, and the
organic layer was washed with saturated aqueous NaHCO3 solution,
dried over anhydrous MgSO4, and evaporated. The residue was
purified by silica gel column chromatography (hexane/ethyl acetate
) 5:1) to give aldehyde (2.1 g, 98%) as a syrup: 1H NMR (CDCl3)
δ 1.33 (s, 3 H, CH3), 1.52 (s, 3 H, CH3), 2.61 (dd, 1 H, J ) 3.9,
13.2 Hz, 6-CHH), 2.87 (dd, 1 H, J ) 4.1, 12.7 Hz, 6-CHH), 3.93
(s, 1 H, 4-H), 4.92 (t, 1 H, J ) 4.1 Hz, 6a-H), 5.10 (d, 1 H, J ) 5.4
Hz, 3a-H), 9.43 (s, 1 H, CHO).
To a stirred solution of aldehyde (5.6 g, 30.0 mmol) in MeOH
(70 mL) was carefully added sodium borohydride (1.3 g, 33.6
mmol) in several portions at 0 °C, and the reaction mixture was
stirred for 30 min at the same temperature and neutralized with
glacial AcOH. After the removal of the solvent, the mixture was
partitioned between EtOAc (150 mL) and brine (100 mL). The

4′-Thionucleosides as A3 AR Agonists
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 279
saturated NaHCO3 solution (three times) and brine. The organic
layer was dried over anhydrous MgSO4 and evaporated. The residue
was purified by a silica gel column chromatography (CH2Cl2/EtOAc
) 20:1-10:1) to give 16 (8.34 g, 54%) as a white foam and its
R-anomer (trace).
General Procedure C for the Synthesis of 29-34. To a stirred
solution of 4′-hydroxymethyl analogues (23-28) in dry DMF (10
mL) was added pyridinium dichromate (10.0 equiv) and the reaction
mixture was stirred at room temperature for 20 h. After being poured
into water (50 mL per mmol), the reaction mixture was stirred at
room temperature for 1 h. The precipitate was filtered, and the filter
cake was washed with excess water and dried under high vacuum
to give a brownish solid, which was used in the next step without
further purification.
â-anomer 16: UV (MeOH) λmax 269 nm (pH 7); 1H NMR
(CDCl3) δ 1.39 (s, 3 H, CH3), 1.64 (s, 3 H, CH3), 4.11 (td, 1 H, J
) 2.7, 6.8 Hz, 4-H), 4.57 (dd, 1 H, J ) 6.6, 11.4 Hz, BzOCHH),
4.75 (dd, 1 H, J ) 7.1, 11.4 Hz, BzOCHH), 5.19 (dd, 1 H, J )
2.7, 5.4 Hz, 3a-H), 5.40 (dd, 1 H, J ) 1.9, 5.4 Hz, 6a-H), 6.10 (d,
1 H, J ) 1.9 Hz, 6-H), 7.37-7.97 (m, 5 H, Ph), 8.38 (s, 1 H, H-8);
FAB-MS m/z 482 (M+ + 1). Anal. (C20H18Cl2N4O4S) C, H, N, S.
R-anomer: UV (MeOH) λmax 269 nm (pH 7); 1H NMR (CDCl3)
δ 1.37 (s, 3 H, CH3), 1.64 (s, 3 H, CH3), 4.08 (m, 1 H, 4-H), 4.51
(dd, 1 H, J ) 6.3, 11.7 Hz, BzOCHH), 4.62 (dd, 1 H, J ) 7.1,
11.7 Hz, BzOCHH), 5.01 (m, 1 H, 3a-H), 5.11 (m, 1 H, 6a-H),
6.50 (d, 1 H, J ) 2.2 Hz, 6-H), 7.37-7.99 (m, 5 H, Ph), 8.87 (s,
1 H, H-8); FAB-MS m/z 482 (M+ + 1). Anal. (C20H18Cl2N4O4S)
C, H, N, S.
2-Chloropurin-9-yl)-3,4-dihydroxytetrahydrothiophene-2-car-
boxylic Acid Methyl Amide (35). Compound 23 (150 mg, 0.27
mmol) was oxidized with pyridinium dichromate (2.6 g, 6.91 mmol)
to give the crude 29, which was methylated with dimethyl sulfate
(1 mL, 10.57 mmol) and K2CO3 (100 mg, 0.72 mmol) in acetone
(6 mL) at room temperature for 2 h to give methyl ester (200 mg)
after the usual workup. A stirred solution of crude methyl ester
(200 mg) was treated with methylamine (20 mL, 40.0 mmol, 2 N
THF solution) at room temperature for 24 h in a sealed tube. The
volatiles were removed under reduced pressure, and the resulting
residue was purified by silica gel column chromatography (hexane/
ethyl acetate ) 1:2) to give silyl amide (80 mg, 51%) as a white
foam: UV (MeOH) λmax 264 nm (pH 7); 1H NMR (CDCl3) δ 0.01
(m, 12 H, 4′ × Si-CH3), 0.79 (s, 9 H, C(CH3)3), 0.85 (s, 9 H,
C(CH3)3), 2.67 (d, 3 H, J ) 4.1 Hz, N-CH3), 3.78 (d, 1 H, J )
4.8 Hz, 2-H), 4.35 (m, 1 H, 3-H), 4.57 (m, 1 H, 4-H), 5.70 (d, 1 H,
J ) 5.6 Hz, 5-H), 7.72 (br s, 2 H, exchangeable with D2O, NH2),
7.72 (br s, 1 H, exchangeable with D2O, NH), 8.10 (s, 1 H, H-8).
To a stirred solution of silyl amide (140 mg, 0.24 mmol) in THF
(5 mL) was added TBAF (0.25 mL, 0.25 mmol, 1 M THF solution)
and the reaction mixture was stirred at room temperature for 1 h.
The solvent was evaporated and the resulting residue was purified
by silica gel column chromatography (CH2Cl2/MeOH ) 10:1) to
Benzoic Acid (3aS,4R,6R,6aR)-6-(6-Amino-2-chloropurin-9-
yl)-2,2-dimethyltetrahydrothieno[3,4-d][1,3]dioxol-4-ylmethyl Es-
ter (17). A solution of 16 (454 mg, 0.94 mmol) in saturated
ethanolic ammonia (25 mL) was heated at 50 °C in a sealed tube
for 24 h. The reaction mixture was evaporated to dryness under
reduced pressure. The resulting residue was purified by silica gel
column chromatography (hexane/ethyl acetate ) 1:1) to afford 17
(148 mg, 32%) and debenzoylated compound (235 mg, 65%). A
solution of debenzoylated compound (235 mg) in pyridine (10 mL)
was treated with benzoyl chloride (0.12 mL, 0.758 mmol) at room
temperature for 12 h. The reaction mixture was partitioned between
EtOAc and water and the organic layer was dried, filtered, and
evaporated. The resulting residue was purified by silica gel column
chromatography (hexane/ethyl acetate ) 2:1) to afford compound
give 35 (58 mg, 69%) as a white solid: mp 233-235 °C; [R]26
D
-20.2 (c 0.1, MeOH); UV (MeOH) λmax 264 nm (pH 7); 1H NMR
(DMSO-d6) δ 2.70 (d, 3 H, J ) 4.1 Hz, N-CH3), 3.82 (d, 1 H, J
) 4.6 Hz, 2-H), 4.36 (dd, 1 H, J ) 4.6, 8.7 Hz, 3-H), 4.53 (m, 1
H, 4-H), 5.60 (d, 1 H, J ) 5.4 Hz, exchangeable with D2O, OH),
5.78 (d, 1 H, J ) 5.1 Hz, exchangeable with D2O, OH), 5.82 (d, 1
H, J ) 5.5 Hz, 5-H), 7.87 (br s, 2 H, exchangeable with D2O,
NH2), 8.33 (br q, 1 H, exchangeable with D2O, NH), 8.55 (s, 1 H,
H-8); 13C NMR (DMSO-d6) δ 51.8, 57.5, 62.5, 75.4, 78.1, 118.4,
140.1, 150.5, 153.0, 156.7, 170.7; FAB-MS m/z 345 (M+). Anal.
(C11H13ClN6O3S) C, H, N, S.
1
17 (240 mg, 80%): UV (MeOH) λmax 264 nm (pH 7); H NMR
(CDCl3) δ 1.37 (s, 3 H, CH3), 1.63 (s, 3 H, CH3), 4.01 (td, 1 H, J
) 2.7, 7.3 Hz, 4-H), 4.57 (dd, 1 H, J ) 6.8, 11.5 Hz, BzOCHH),
4.77 (dd, 1 H, J ) 7.6, 11.5 Hz, BzOCHH), 5.23 (dd, 1 H, J )
2.7, 5.6 Hz, 3a-H), 5.38 (dd, 1 H, J ) 1.9, 5.6 Hz, 6a-H), 6.02 (d,
1 H, J ) 1.9 Hz, 6-H), 6.29 (br s, 2 H, NH2), 7.40-8.02 (m, 6 H,
Ph, H-8); FAB-MS m/z 463 (M+ + 1). Anal. (C20H20ClN5O4S) C,
H, N, S.
General Procedure A for the Synthesis of 18-22. To a solution
of 16 in anhydrous EtOH (20 mL/mmol) were added triethylamine
(3.0 equiv) and the appropriate amine (1.2 equiv). After being stirred
at room temperature for 24 h, the reaction mixture was evaporated.
The residue was purified by silica gel column chromatography
(hexane/ethyl acetate ) 1:1) to give various N6-substituted nucleo-
sides 18-22. Spectral data for compounds 18-22 are provided in
the Supporting Information.
General Procedure B for the Synthesis of 23-28. A solution
of N6-substituted nucleosides (17-22) in 80% aqueous AcOH
solution (30 mL) was stirred at 70 °C for 12 h. The solvent was
removed under reduced pressure and the mixture was neutralized
with methanolic ammonia. After evaporation, the residue was
purified by silica gel column chromatography (CH2Cl2/MeOH )
15:1) to give diol as a white foam.
To a stirred solution of diol in dry pyridine (20 mL) was added
a solution of TBDMSOTf (5.0 equiv) dropwise and the reaction
mixture was stirred at 50 °C for 5 h. The mixture was partitioned
between CH2Cl2 and H2O, and the organic layer was washed with
water, aqueous NaHCO3 solution, water, and brine; dried over
anhydrous MgSO4; and evaporated. The crude disilyl ether was
used in the next step without further purification.
To a stirred solution of disilyl ether in anhydrous methanol (30
mL) was added sodium methoxide (1.5 equiv) and the mixture was
stirred at room temperature for 4 h. After being neutralized with
glacial acetic acid, the mixture was evaporated. The resulting residue
was purified by silica gel column chromatography (hexane/ethyl
acetate ) 2:1) to give 4′-hydroxymethyl analogues 23-28. Spectral
data for compounds 23-28 are provided in the Supporting
Information.
General Procedure D for the Synthesis of 36a-40g. To a
solution of acid derivatives (30-34), EDC (1.5 equiv), HOBt (1.5
equiv), and appropriate amine (1.5 equiv) in CH2Cl2 (20 mL) was
added DIPEA (3.0 equiv), and the mixture was stirred at room
temperature for 12 h. The reaction mixture was evaporated and
the residue was purified by a silica gel column chromatography
(hexane/EtOAc ) 10:1-5:1) to give silyl amides as a white foam.
To a stirred solution of silyl amides in THF (5 mL) was added
TBAF (2.5 equiv) and the reaction mixture was stirred at room
temperature for 1 h. The solvent was evaporated and the resulting
residue was purified by silica gel column chromatography (CH2-
Cl2/MeOH ) 10:1) to give 36a-40g. Spectral data for compounds
36a-40g are provided in the Supporting Information.
Pharmacological Methods. [125I]-N6-(4-amino-3-iodobenzyl)-
adenosine-5′-N-methyluronamide (I-AB-MECA; 2000 Ci/mmol)
and [3H]cyclic AMP (40 Ci/mmol) were from Amersham Pharmacia
Biotech (Buckinghamshire, UK).
Cell Culture and Membrane Preparation. CHO (Chinese
hamster ovary) cells expressing the recombinant human A3 AR were
cultured in DMEM supplemented with 10% fetal bovine serum,
100 units/mL penicillin, 100 µg/mL streptomycin, 2 µmol/mL
glutamine, and 800 µg/mL geneticin. CHO cells expressing rat A3
ARs were cultured in DMEM and F12 (1:1). Cells were harvested
by trypsinization. After homogenization and suspension, cells were
centrifuged at 500g for 10 min, and the pellet was resuspended in
50 mM Tris‚HCl buffer (pH 8.0) containing 10 mM MgCl2, 1 mM
EDTA, and 0.1 mg/mL CHAPS. The suspension was homogenized
with an electric homogenizer for 10 s and was then recentrifuged

280 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1
Jeong et al.
(4) Fishman, P.; Madi, L.; Bar-Yehuda, S.; Barer, F.; Del Valle, L.;
Khalili, K. Evidence for involvement of Wnt signaling pathway in
IB-MECA mediated suppression of melanoma cells. Oncogene 2002,
21, 4060-4064.
(5) Meade, C. J.; Mierau, J.; Leon, I.; Ensinger, H. A. In vivo role of
the adenosine A3 receptor: N6-2-(4-Aminophenyl)ethyladenosine
induces bronchospasm in BDE rats by a neurally mediated mechanism
involving cells resembling mast cells. J. Pharmacol. Exp. Ther. 1996,
279, 1148-1156.
(6) Avila, M. Y.; Stone, R. A.; Civan, M. M. Knockout of A3 adenosine
receptors reduces mouse intraocular pressure. InVest. Opthalmol.
Visual Sci. 2002, 43, 3021-3026.
(7) Sajjadi, F. G.; Takabayashi, K.; Foster, A. C.; Domingo, R. C.;
Firestein, G. S. Inhibition of TNF-alpha expression by adenosine:
Role of A3 adenosine receptors. J. Immunol. 1996, 156, 3435-3442.
(8) Ramkumar, V.; Stiles, G. L.; Beaven, M. A.; Ali, H. The A3 adenosine
receptor is the unique adenosine receptor which facilitates release
of allergic mediators in mast cells. J. Biol. Chem. 1993, 268, 16887-
16890.
at 20 000 g for 20 min at 4 °C. The resultant pellets were
resuspended in buffer in the presence of 3 units/mL adenosine
deaminase, and the suspensions were stored at -80 °C until the
binding experiments. The protein concentration was measured using
the Bradford assay.18
Binding Assays to the Human A1 and A2A Receptors. For
binding to human A1 receptors, [3H]R-PIA (2 nM) was incubated
with membranes (40 µg/tube) from CHO cells stably expressing
human A1 receptors at 25 °C for 60 min in 50 mM Tris‚HCl buffer
(pH 7.4; MgCl2, 10 mM) in a total assay volume of 200 µL.
Nonspecific binding was determined using 10 µM of CPA. For
human A2A receptor binding, membranes (20 µg/tube) from HEK-
293 cells stably expressing human A2A receptors were incubated
with 15 nM [3H]CGS21680 at 25 °C for 60 min in 200 µL of 50
mM Tris‚HCl, pH 7.4, containing 10 mM MgCl2. NECA (10 µM)
was used to define nonspecific binding. Reaction was terminated
by filtration with GF/B filters.
Binding Assay to the Human A3 Receptors Using [125I]-4-
Amino-3-iodobenzyladenosine-5′-N-methyluronamide ([125I]I-
AB-MECA). For competitive binding assay, each tube contained
50 µL of membrane suspension (20 µg of protein), 25 µL of [125I]I-
AB-MECA (1.0 nM), and 25 µL of increasing concentrations of
the test ligands in Tris‚HCl buffer (50 mM, pH 7.4) containing 10
mM MgCl2, 1 mM EDTA. Nonspecific binding was determined
using 10 µM of Cl-IB-MECA in the buffer. The mixtures were
incubated at 25 °C for 60 min. Binding reactions were terminated
by filtration through Whatman GF/B filters under reduced pressure
using a MT-24 cell harvester (Brandell, Gaithersburgh, MD). Filters
were washed three times with 9 mL of ice-cold buffer. Radioactivity
was determined in a Beckman 5500B γ-counter.
Cyclic AMP Accumulation Assay. Intracellular cyclic AMP
levels were measured with a competitive protein binding method.19
CHO cells expressing recombinant human and rat A3 ARs were
harvested by trypsinization. After centrifugation and resuspension
in medium, cells were planted in 24-well plates in 1.0 mL medium.
After 24 h, the medium was removed, and cells were washed three
times with 1 mL of DMEM, containing 50 mM HEPES, pH 7.4.
Cells were then treated with agonists and/or test compounds in the
presence of rolipram (10 µM) and adenosine deaminase (3 units/
mL). After 45 min forskolin (10 µM) was added to the medium,
and incubation was continued for an additional 15 min. The reaction
was terminated by removing the supernatant, and cells were lysed
upon the addition of 200 µL of 0.1 M ice-cold HCl. The cell lysate
was resuspended and stored at -20 °C. For determination of cyclic
AMP production, protein kinase A (PKA) was incubated with
[3H]cyclic AMP (2 nM) in K2HPO4/EDTA buffer (K2HPO4, 150
mM; EDTA, 10 mM), 20 µL of the cell lysate, and 30 µL of 0.1
M HCl or 50 µL of cyclic AMP solution (0-16 pmol/200 µL for
standard curve). Bound radioactivity was separated by rapid
filtration through Whatman GF/C filters and washed once with cold
buffer. Bound radioactivity was measured by liquid scintillation
spectrometry.
(9) (a) Klotz, K.-N. Adenosine receptors and their ligands. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 2000, 362, 382-391. (b) Baraldi,
P. G.; Cacciari, B.; Romagnoli, R.; Merighi, S.; Varani, K.; Borea,
P. A.; Spalluto, G. A3 adenosine receptor ligands: History and
perspectives. Med. Res. ReV. 2000, 20, 103-128.
(10) (a) van Tilburg, E. W.; von Frijtag Drabbe Ku¨nzel, J.; de Groote,
M.; Vollinga, R. C.; Lorenzen, A.; IJzerman, A. P. N6,5′-Disubstituted
adenosine derivatives as partial agonists for the human adenosine
A3 receptor. J. Med. Chem. 1999, 42, 1393-1400. (b) van Tilburg,
E. W.; van der Klein, P. A. M.; von Frijtag Drabbe Ku¨nzel, J.; de
Groote, M.; Stannek, C.; Lorenzen, A.; IJzerman, A. P. 5′-O-alkyl
ethers of N6,2-substituted adenosine derivatives: Partial agonists for
the adenosine A1 and A3 receptors. J. Med. Chem. 2001, 44, 2966-
2975.
(11) (a) Gao, Z.-G.; Kim, S. K.; Biadatti, T.; Chen, W.; Lee, K.; Barak,
D.; Kim, S. G.; Johnson, C. R.; Jacobson, K. A. Structural determinats
of A3 adenosine receptor activation: Nucleoside ligands at the
agonist/antagonist boundary. J. Med. Chem. 2002, 45, 4471-4484.
(b) Gao, Z.-G.; Blaustein, J. B.; Gross, A. S.; Melman, N.; Jacobson,
K. A. N6-Substituted adenosine derivatives: Selectivity, efficacy, and
species differences at A3 adenosine receptors. Biochem. Pharmacol.
2003, 65, 1675-1684. (c) Gao, Z.-G.; Jeong, L. S.; Moon, H. R.;
Kim, H. O.; Choi, W. J.; Shin, D. H.; Elhalem, E.; Comin, M. J.;
Melaman, N.; Mamedova, L.; Gross, A. S.; Rodriguez, J. B.;
Jacobson, K. A. Structural determinants of efficacy at A3 adenosine
receptors: Modification of the ribose moiety. Biochem. Pharmacol.
2004, 67, 893-901.
(12) Kim, H. O.; Ji, X.-d.; Siddiqi, S. M.; Olah, M. E.; Stiles, G. L.;
Jacobson, K. A. 2-Substitution of N6-benzyladenosine-5′-uronamides
enhances selectivity for A3-adenosine receptors. J. Med. Chem. 1994,
37, 3614-3621.
(13) Jacobson, K. A.; Knutsen, L. J. S. P1 and P2 purine and pyrimidine
receptors. In Purinergic and Pyrimidinergic Signalling I; Handbook
of Experimental Pharmacology; Abbracchio, M. P., Williams, M.,
Eds.; Springer-Verlag: Berlin, Germany, 2001; Vol. 151/I, pp 129-
175.
(14) Jacobson, K. A.; Ji, X-d.; Li, A.-H.; Melman, N.; Siddiqui, M. A.;
Shin, K.-J.; Marquez, V. E.; Ravi, R. G. Methanocarba analogues of
purine nucleosides as potent and selective adenosine receptor agonists.
J. Med. Chem. 2000, 43, 2196-2203.
(15) Jeong, L. S.; Jin, D. Z.; Kim, H. O.; Shin, D. H.; Moon, H. R.;
Gunaga, P.; Chun, M. W.; Kim, Y.-C.; Melman, N.; Gao, Z.-G.;
Jacobson. K. A. N6-Substituted D-4′-thioadenosine-5′-methylurona-
mides: Potent and selective agonists at the human A3 adenosine
receptor. J. Med. Chem. 2003, 46, 3775-3777.
Acknowledgment. This work was supported by a grant from
the Korea Health R & D Project, Ministry of Health & Welfare,
Korea (03-PJ2-P.G.4-BD02-0001).
(16) Klotz, K.-N.; Hessling, J.; Hegler, J.; Owman, B.; Kull, B.; Fredholm,
B. B.; Lohse M. J. Comparative pharmacology of human adenosine
subtypes-characterization of stably transfected receptors in CHO cells.
Naunyn-Schmiedeberg’s Arch. Pharmacol. 1998, 357, 1-9.
(17) Gao, Z. G.; Chen, A.; Barak, D.; Kim, S. K.; Mu¨ller, C. E.; Jacobson,
K. A. Identification by site-directed mutagenesis of residues involved
in ligand recognition and activation of the human A3 adenosine
receptor. J. Biol. Chem. 2002, 277, 19056-19063.
(18) Bradford, M. M. A rapid and sensitive method for the quantitation
of microgram quantities of protein utilizing the principle of protein-
dye binding. Anal. Biochem. 1976, 72, 248-254.
(19) (a) Nordstedt, C.; Fredholm, B. B. A modification of a protein-binding
method for rapid quantification of cAMP in cell-culture supernatants
and body fluid. Anal. Biochem. 1990, 189, 231-234. (b) Post S. R.;
Ostrom R. S.; Insel P. A. Biochemical methods for detection and
Supporting Information Available: Spectral data for com-
pounds 18-28 and 36-40 and results from elemental analyses data.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Olah, M. E.; Stiles, G. L. The role of receptor structure in determining
adenosine receptor activity. Pharmacol. Ther. 2000, 85, 55-75.
(2) Liang, B. T.; Jacobson, K. A. A Physiological role of the adenosine
A3 receptor: Sustained cardioprotection. Proc. Natl. Acad. Sci. U.S.A.
1998, 95, 6995-6999.
(3) von Lubitz, D. K. J. E.; Ye, W.; McClellan, J.; Lin, R. C.-S.
Stimulation of adenosine A3 receptors in cerebral ischemia. Neuronal
death, recovery, or both? Ann. N.Y. Acad. Sci. 1999, 890, 93-106.

4′-Thionucleosides as A3 AR Agonists
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 1 281
measurement of cyclic AMP and adenylyl cyclase activity. Methods
Mol. Biol. 2000, 126, 363-374.
(20) Tchilibon, S.; Kim, S.-K.; Gao, Z.-G.; Harris, B. A.; Blaustein, J.;
Gross, A. S.; Melman, N.; Jacobson, K. A. Exploring distal regions
of the A3 adenosine receptor binding site: Sterically-constrained N6-
(2-phenylethyl)adenosine derivatives as potent ligands. Bioorg. Med.
Chem. 2004, 12, 2021-2034.
(21) Tchilibon, S.; Joshi, B. V.; Kim, S.-K.; Duong H. T.; Gao, Z.-G.;
Jacobson, K. A. (N)-Methanocarba 2,N6-disubstituted adenine nucleo-
sides as highly potent and selective A3 adenosine receptor agonists.
J. Med. Chem. 2005, 48, 1745-1758.
JM050595E