Structure-activity relationships of truncated D- and L-4′- thioadenosine derivatives as species-independent A3 adenosine receptor antagonists

Article abstract of DOI:10.1021/jm8008647

Novel D- and L-4′-thioadenosine derivatives lacking the 4′-hydroxymethyl moiety were synthesized, starting from D-mannose and D-gulonic γ-lactone, respectively, as potent and selective species-independent A₃ adenosine receptor (AR) antagonists. Among the novel 4′-truncated 2-H nucleosides tested, a N⁶-(3- chlorobenzyl) derivative 7c was the most potent at the human A₃ AR (K_i = 1.5 nM), but a N⁶-(3-bromobenzyl) derivative 7d showed the optimal species-independent binding affinity.

Full text of DOI:10.1021/jm8008647

J. Med. Chem. 2008, 51, 6609–6613
6609
Structure-Activity Relationships of Truncated D- and L-4′-Thioadenosine Derivatives as
Species-Independent A3 Adenosine Receptor Antagonists1
Lak Shin Jeong,*,† Shantanu Pal,† Seung Ah Choe,† Won Jun Choi,† Kenneth A. Jacobson,‡ Zhan-Guo Gao,‡ Athena M. Klutz,‡
Xiyan Hou,† Hea Ok Kim,† Hyuk Woo Lee,† Sang Kook Lee,† Dilip K. Tosh,† and Hyung Ryong Moon§
Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha Womans UniVersity, Seoul 120-750, Korea, Molecular Recognition Section,
Laboratory of Bioorganic Chemistry, National Institute of Diabetes & DigestiVe & Kidney Diseases, National Institutes of Health, Bethesda,
Maryland 20892, College of Pharmacy, Pusan National UniVersity, Busan 609-753, Korea
ReceiVed July 14, 2008
Novel D- and L-4′-thioadenosine derivatives lacking the 4′-hydroxymethyl moiety were synthesized, starting
from D-mannose and D-gulonic γ-lactone, respectively, as potent and selective species-independent A3
adenosine receptor (AR) antagonists. Among the novel 4′-truncated 2-H nucleosides tested, a N6-(3-
chlorobenzyl) derivative 7c was the most potent at the human A3 AR (Ki ) 1.5 nM), but a N6-(3-bromobenzyl)
derivative 7d showed the optimal species-independent binding affinity.
Chart 1. Rationale for the Design of the Target Nucleosides 7
Introduction
On the basis of the structure of adenosine, an endogenous
cell signaling molecule that binds to four specific subtypes (A1,
A2A, A2B, and A3) of adenosine receptors (ARsa),2 a number of
nucleoside analogues have been synthesized and evaluated as
adenosine receptor ligands.3 Among these, IB-MECA4 1 and
Cl-IB-MECA5 2 were discovered as potent and selective A3
AR full agonists (Ki ) 1.0 and 1.4 nM, respectively, at the
human A3 AR) and are being developed as anti-inflammatory
and anticancer agents. On the basis of the bioisosteric rationale,
we reported the 4′-thionucleosides 3 and 4, derivatives of
compounds 1 and 2, to also be highly potent and selective A3
AR full agonists.6 Compound 4 exhibited potent in vitro and in
vivo antitumor activities,7 resulting from the inhibition of Wnt
signaling pathway (Chart 1).
However, because of the structural similarity to adenosine,
most of these adenosine analogues were found to be A3 AR
agonists. Only a few nucleoside derivatives8 have been reported
to be A3 AR antagonists, but these generally exhibit weaker
and less selective human A3 AR antagonism than nonpurine
heterocyclic A3 AR antagonists. Although these nonpurine
heterocyclic A3 AR antagonists9 bound with high affinity at the
human A3 AR, they were weak or ineffective at the rat A3 AR,
indicating that they were not ideal for evaluation in small animal
models and thus as drug candidates.10 Therefore, it is highly
desirable to develop A3 AR antagonists that are independent of
species. The fact10 that nucleoside analogues show minimal
species-dependence at the A3 AR prompted us to search for
novel potent and selective A3 AR antagonists, derived from
nucleoside templates.
A molecular modeling study of the A3 AR indicated that
hydrogen of the 5′-uronamides of compounds 1-4 serves as a
hydrogen-bonding donor in the binding site of the A3 AR, which
is essential for the induced-fit required for the activation of the
A3 AR.11 On the basis of these findings, we appended extra
alkyl groups on the 5′-uronamides of compounds 1-4 to remove
hydrogen-bonding ability at this site, thus precluding the
conformational change required for activation of the A3 AR.
As expected, these 5′-N,N-dialkyl amide derivatives12 displayed
potent and selective A3 AR antagonism in which steric factors
were crucial for affinity in binding to the A3 AR. Within this
class, 5′-N,N-dimethylamide derivative 5 was discovered to be
the most potent full A3 AR antagonist. Encouraged with these
results, we designed and synthesized another new template to
remove the 5′-uronamide group of compound 4 in order to
minimize the steric repulsion at the binding region of the 5′-
uronamide group and to abolish its hydrogen-bonding ability.
This led to the discovery of compounds 6a-6e as highly potent
and selective human A3 AR antagonists, which were more potent
and selective than compound 5.1 Among these, compound 6e
* To whom correspondence should be addressed. Phone: 82-2-3277-3466.
Fax: 82-2-3277-2851. E-mail: lakjeong@ewha.ac.kr.
†
Laboratory of Medicinal Chemistry, College of Pharmacy, Ewha
Womans University.
‡
Molecular Recognition Section, Laboratory of Bioorganic Chemistry,
National Institute of Diabetes & Digestive & Kidney Diseases, National
Institutes of Health.
§
College of Pharmacy, Pusan National University.
a Abbreviations: AR, adenosine receptor; CCPA, 2-chloro-N6-cyclopen-
tyladenosine; CHO, Chinese hamster ovary; IB-MECA, N6-(3-iodobenzyl)-
5′-N-methylcarboxamidoadenosine; Cl-IB-MECA, 2-chloro-N6-(3-iodobenzyl)-
5′-N-methylcarboxamidoadenosine; I-AB-MECA, N6-(4-amino-3-iodobenzyl)-
5′-N-methylcarboxam idoadenosine; NECA, 5′-N-ethylcarboxamidoaden-
osine.
10.1021/jm8008647 CCC: $40.75
2008 American Chemical Society
Published on Web 09/24/2008

6610 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Brief Articles
Scheme 2. Synthesis of Truncated L-4′-Thioadenosine
Scheme 1. Synthesis of Truncated D-4′-Thioadenosine
Derivatives 7a-7ea
Derivatives 7f and 7g
To determine whether a stereochemical preference exists in
the binding to the A3 AR, the L-enantiomers 7f and 7g of D-4′-
thionucleosides were synthesized as illustrated in Scheme 2.
D-Gulonic γ-lactone was converted to the diol 11 according to
our previously published procedure.13 One-step conversion of
the diol 11 into the L-glycosyl donor 12 was achieved using
excess Pb(OAc)4, indicating that oxidative diol cleavage,
oxidation of the resulting aldehyde to the acid, and oxidative
decarboxylation occurred simultaneously.13 Using the same
synthetic strategy shown in Scheme 1, L-4′-thioadenosine
derivatives 7f and 7g were synthesized from L-glycosyl donor
12.
a Reagents and conditions: (a) 6-chloropurine, ammonium sulfate, HMDS,
170 °C, 15 h, then TMSOTf, DCE, rt to 80 °C, 3 h; (b) 2 N HCl, 45 °C,
15 h; (c) RNH2, Et3N, EtOH, rt, 1-3 d.
also showed species-independent binding affinity, as indicated
by its high affinity at the rat A3 AR.1 On the basis of these
results, it is of interest to systematically establish structure-activity
relationships by modifying the C2 and N6 positions of the purine
moiety of compounds 6a-6e in order to develop novel A3 AR
antagonists. In this article, we extend previous observations that
truncated D-4′-thioadenosine derivatives 6a-6e containing 2-Cl
substitution are selective A3 AR antagonists.1 A series of 2-H
analogues were prepared and characterized biologically. The
binding affinities at the human A3 AR were compared with those
at the rat A3 AR to develop species-independent A3 AR
antagonists. We also compared the binding affinities of D-4′-
thionucleosides with those of the corresponding L-4′-thionucleo-
sides to determine a stereochemical preference. Thus, here we
report a full account of truncated D- and L-4′-thioadenosine
derivatives 7 as highly potent and species-independent A3 AR
antagonists.
Initial binding experiments were performed using adherent
mammalian cells stably transfected with cDNA encoding the
appropriate human ARs (A1 AR and A3 AR in CHO cells and
A
2A AR in HEK-293 cells).14,15 Binding was carried out using
1 nM [3H]CCPA, 10 nM [3H]CGS-21680, or 0.5 nM [125I]I-
AB-MECA as radioligands for A1, A2A, and A3 ARs, respec-
tively. As shown in Table 1, most of the synthesized compounds
exhibited high binding affinity at the human A3 AR with low
binding affinities at the human A1 AR and human A2A AR.
Among the novel 2-H truncated adenosine derivatives tested,
compound 7c (R ) 3-chlorobenzyl) showed the highest binding
affinity (Ki ) 1.5 ( 0.4 nM) at the human A3 AR with high
selectivities versus the A1 AR (570-fold selective) and the A2A
AR (290-fold selective). Compound 7e (R ) 3-iodobenzyl) was
also very potent (Ki ) 2.5 ( 1.0 nM), with selectivities of 210-
and 92-fold versus the A1 and A2A AR, respectively. N6-
Substituted adenosine derivatives 7a-7e without a 2-chloro
substituent showed a very similar pattern to the corresponding
2-chloro derivatives 6a-6e in the binding affinity at the human
A3 AR but showed less selectivity versus the other subtypes of
ARs. In the 3-halobenzyl series, the order of binding affinity
for 2-H analogues was as follows: Cl > I > Br > F, indicating
that the size of halogen alone does not determine the binding
affinity at the human A3 AR. It is interesting to note that 2-H
derivatives are less lipophilic than the corresponding 2-Cl
derivatives, conferring more water solubility on the molecules
for further biological evaluation. For example, the cLogP values
of corresponding structures 6c and 7c are 1.84 and 1.12,
respectively. To determine a stereochemical preference, the
binding affinities of D-series were compared with those of
L-series. As shown in Table 1, L-type nucleosides 7f and 7g
Results and Discussion
The D-glycosyl donor 8 was subjected to the Lewis acid-
catalyzed condensation for the synthesis of the final D-4′-
thionucleosides lacking a 4′-hydroxymethyl group, as shown
in Scheme 1. The D-glycosyl donor 8 was condensed with
6-chloropurine in the presence of TMSOTf as a Lewis acid to
give ꢀ-6-chloropurine derivative 9 as a single diastereomer. The
anomeric configuration of compound 9 was easily confirmed
by 1H NOE experiment between 3′-H and H-8. Removal of the
isopropylidene group of 9 was achieved with 2 N HCl in THF
to give 10. The 2-H intermediate 10 was converted to the novel
N6-methyl derivative 7a and N6-3-halobenzyl derivatives 7b-7e
by treating with methylamine and 3-halobenzylamines, respec-
tively. This route parallels the synthesis of the 2-chloro-N6-
susbtituted-4′-thiopurine analogues 6a-6e that we reported
earlier.1

Brief Articles
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6611
Table 1. Binding Affinities of Known A3 AR Agonists, 1-4 and Antagonist 5, and Truncated 4′-Thioadenosine Derivatives 6a-6e and 7a-7g at Three
Subtypes of ARs
affinity, Ki, nM ( SEM (or % inhibition at 10-5 M)a,b
c
compd
hA1
hA2A
rA3
hA3
1(IB-MECA)
2 (Cl-IB-MECA)
3 (thio-IB-MECA)
4 (thio-Cl-IB-MECA)
51
222 ( 22
17.3
193 ( 46
6220 ( 640
55.4 ( 1.8
(20%)
(38%)
(34%)
2490 ( 940
1070 ( 180
1430 ( 420
860 ( 210
790 ( 190
530 ( 97
(6.1%)
2900
5360 ( 2470
ND
223 ( 36
>10,000
45.0 ( 1.4
(48%)
1.1
0.33
1.0
1.4 ( 0.3
0.25 ( 0.06
0.38 ( 0.07
15.5 ( 3.1
3.69 ( 0.25
7.4 ( 1.3
1.66 ( 0.90
8.99 ( 5.17
4.16 ( 0.50
4.8 ( 1.7
7.3 ( 0.6
1.5 ( 0.4
6.8 ( 3.4
2.5 ( 1.0
(12.6%)
1.86 ( 0.36
0.82 ( 0.27
321 ( 74
658 ( 160
36.2 ( 10.7
6.2 ( 1.8
6.1 ( 1.8
3.89 ( 1.15
(28 ( 10%)
98 ( 28
17 ( 5
6.3 ( 1.3
48 ( 7
ND
ND
5
6a (R1 ) Cl, R2 ) methyl)
6b (R1 ) Cl, R2 ) 3-fluorobenzyl)
6c (R1 ) Cl, R2 ) 3-chlorobenzyl)
6d (R1 ) Cl, R2 ) 3-bromobenzyl)
6ed (R1 ) Cl, R2 ) 3-iodobenzyl)
7a (R1 ) H, R2 ) methyl)
7b (R1 ) H, R2 ) 3-fluorobenzyl)
7c (R1 ) H, R2 ) 3-chlorobenzyl)
7d (R1 ) H, R2 ) 3-bromobenzyl)
7e (R1 ) H, R2 ) 3-iodobenzyl)
7f (R1 ) Cl, R2 ) 3-bromobenzyl)
7g (R1 ) Cl, R2 ) 3-iodobenzyl)
(18%)
(18%)
341 ( 75
(22 ( 5%)
1260 ( 330
440 ( 110
420 ( 32
230 ( 65
(45.7%)
(-8.0%)
(-0.95%)
(18.4%)
a ND: Not determined. All binding experiments were performed using adherent mammalian cells stably transfected with cDNA encoding the appropriate
human AR (A1 AR and A3 AR in CHO cells and A2A AR in HEK-293 cells) or the rat A3 AR (CHO cells). Binding was carried out using 1 nM [3H]CCPA,
10 nM [3H]CGS-21680, or 0.5 nM [125I]I-AB-MECA as radioligands for A1, A2A, and A3 ARs, respectively. Values are expressed as mean ( sem, n ) 3-4
(outliers eliminated), and normalized against a nonspecific binder, 5′-N-ethylcarboxamidoadenosine (NECA, 10 µM). Data for compounds 6a-6e at the
human ARs and compound 6e at the rat A3 AR were reported in ref 1. b A value expressed as a percentage refers to percent inhibition of specific radioligand
binding at 10 µM, with nonspecific binding defined using 10 µM NECA. c A functional assay was also carried out at this subtype: percent inhibition at 10
µM forskolin-stimulated cyclic AMP production in CHO cells expressing the human A3 AR, as a mean percentage of the response of the full agonist 3 (n
) 1 - 3). None of the analogues 5-7 activated the hA3 AR (>10% of full agonist effect) by this criterion. d Compound 6e at 10 µM displayed <10% of
the full stimulation of cyclic AMP production, in comparison to 10 µM NECA; no inhibition of the stimulatory effect of 150 nM NECA in CHO cells
expressing human A2B AR (ref 1).
were totally devoid of binding affinities at all subtypes of ARs,
indicating that the D-series induced optimal interaction with all
subtypes of ARs.
In a functional assay, percent inhibition at 10 µM forskolin-
stimulated cyclic AMP production in CHO cells expressing the
human A3 AR was measured as a mean percentage of the
response of the full agonist 3 (n ) 1 - 3). None of the analogues
6 and 7 activated the human A3 AR (>10% of full agonist
effect) by this criterion.
To determine if all final nucleosides show species-independent
binding affinity at the A3 AR, their binding affinity at the rat
A3 AR expressed in CHO cells was also measured (Table 1).
As expected, most of the compounds exhibited species-
independent binding affinity, indicating that they are suitable
for evaluation in small animal models or for further drug
development. Among the 2-H nucleoside analogues tested, a
N6-(3-bromobenzyl) derivative 7d exhibited the most potent
binding affinity at the rat A3 AR (Ki ) 6.3 ( 1.3 nM) followed
by N6-(3-chlorobenzyl) derivative 7c, N6-(3-iodobenzyl) deriva-
tive 7e, and N6-(3-fluorobenzyl) derivative 7b. N6-Methyl
derivative 7a was totally devoid of A3 AR binding affinity in
this species. In the 2-Cl-N6-substituted adenosine series, the
binding affinity was in the following order: I > Br ≈ Cl > F
> Me. The 2-Cl derivatives generally showed more potent and
species-independent binding affinity than the corresponding 2-H
analogues. Compound 6e exhibited the highest binding affinity
at the rat A3 AR (Ki ) 3.89 ( 1.15 nM) among all compounds
tested and was inactive as agonist or antagonist in a cyclic AMP
functional assay16,17 at the hA2B AR. It is interesting to note
that N6-methyl derivatives 6a and 7a showing high binding
affinities (Ki ) 3.69 ( 0.25 nM and 4.8 ( 1.7 nM, respectively)
at the human A3 AR lost their binding affinities at the rat A3
AR, indicating that there must be a larger N6 substituent for
species-independent binding affinity at the A3 AR.18,19
Conclusion
We have established structure-activity relationships of novel
truncated D- and L-4′-thionucleoside analogues as potent species-
independent A3 AR antagonists. The glycosyl donors 8 and 12
were efficiently synthesized from D-mannose and D-gulonic
γ-lactone, respectively, using ring closure of dimesylate with
sodium sulfide and one step conversion of the diol into the
acetate with lead tetraacetate as key steps. Among the novel
4′-truncated 2-H nucleosides tested, D-N6-(3-halobenzyl) deriva-
tives 7b-7e exhibited high binding affinities at the human A3
AR as well as at the rat A3 AR with very low binding affinities
at the human A1 and A2A ARs and a N6-(3-chlorobenzyl)
derivative 7c was the most potent at the human A3 AR, but at
the rat A3 AR 3-bromobenzyl derivative, 7d was the most potent.
Among both 2-H and 2-Cl analogues tested, 2-chloro-N6-(3-
iodobenzyl) derivative 6e was found to exhibit the most potent
binding affinity at the rat A3 AR. Because this class of potent
nucleoside human A3 AR antagonists showed species-indepen-
dence in interaction at this AR subtype, they are regarded as

6612 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20
Brief Articles
1
°C; UV (MeOH) λmax 273.5 nm; H NMR (DMSO-d6) δ 8.45 (s,
1 H), 8.43 (br s, 1 H, D2O exchangeable), 8.21 (s, 1 H), 7.36-7.30
(m, 1 H), 7.18-7.11 (m, 2 H), 7.03 (dt, 1 H, J ) 2.4, 8.4 Hz), 5.90
(d, 1 H, J ) 7.2 Hz), 5.53 (d, 1 H, J ) 6.4 Hz, D2O exchangeable),
5.35 (d, 1 H, J ) 4.0 Hz, D2O exchangeable), 4.70-4.66 (m, 2
H), 4.36-4.33 (m, 1 H), 3.41 (dd, 1 H, J ) 4.0, 10.8 Hz), 3.17 (d,
1 H, J ) 5.2 Hz), 2.79 (dd, 1 H, J ) 2.8, 10.8 Hz). 13C NMR
(DMSO-d6) δ 163.4, 160.9, 152.4, 143.2, 140.0, 130.2, 130.1, 123.1,
123.1, 113.8, 113.6, 113.4, 113.2, 78.3, 72.2, 61.6, 48.6, 34.4. Anal.
(C16H16FN5O2S) C, H, N, S.
good candidates for efficacy evaluation in small animal models
and for further drug development.
Experimental Section
General Methods. Melting points are uncorrected. 1H NMR (400
MHz) and 13C NMR (100 MHz) spectra were measured in CDCl3,
CD3OD or DMSO-d6, and chemical shifts are reported in parts per
million (δ) downfield from tetramethylsilane as internal standard.
Column chromatography was performed using silica gel 60
(230-400 mesh). Anhydrous solvents were purified by the standard
procedures. cLogP values were calculated using ChemDrawUltra,
version 11.0 (CambridgeSoft).
(2R,3R,4S)-2-(6-(3-Chlorobenzylamino)-9H-purin-9-yl)-tet-
rahydrothiophene-3,4-diol (7c). Yield 85%; [R]23.9 -162.5 (c
D
0.096, DMSO); FAB-MS m/z 378 [M + H]+; mp 165.0-165.3
1
°C; UV (MeOH) λmax 274.5 nm. H NMR (DMSO-d6) δ 8.46 (s,
Synthesis. 6-Chloro-9-((3aR,4R,6aS)-2,2-dimethyltetrahy-
drothieno[3,4-d][1,3]dioxol-4-yl)-9H-purine (9). 6-Chloropurine
(3.91 g, 25.3 mmol), ammonium sulfate (84 mg, 0.63 mmol), and
HMDS (50 mL) were refluxed under inert and dry conditions
overnight. The solution was evaporated under high vacuum. The
resulting solid was redissolved in 1,2-dichloroethane (20 mL) cooled
in ice. The solution of 81 (2.76 g, 12.6 mmol) in 1,2-dichloroethane
(20 mL) was added to this mixture dropwise. TMSOTf (4.6 mL,
25.3 mmol) was added dropwise to the mixture. The mixture was
stirred at 0 °C for 30 min, at rt for 1 h, and then heated at 80 °C
for 2 h. The mixture was cooled, diluted with CH2Cl2, and washed
with saturated NaHCO3 solution. The organic layer was dried with
anhydrous MgSO4 and evaporated under reduced pressure. The
yellowish syrup was subjected to a flash silica gel column
chromatography (CH2Cl2:MeOH ) 50:1) to give 9 (3.59 g, 90%)
1 H), 8.44 (br s, 1 H, D2O exchangeable), 8.22 (s, 1 H), 7.39-7.24
(m, 4 H), 5.90 (d, 1 H, J ) 10.4 Hz), 5.53 (d, 1 H, J ) 6.4 Hz,
D2O exchangeable), 5.35 (d, 1 H, J ) 4.0 Hz, D2O exchangeable),
4.71-4.67 (m, 2 H), 4.38-4.33 (m, 1 H), 3.47-3.31 (m, 2 H),
2.80 (dd, 1 H, J ) 3.2, 10.8 Hz). 13C NMR (DMSO-d6) δ 154.3,
152.4, 142.8, 140.0, 132.8, 130.1, 126.9, 126.6, 125.8, 78.3, 72.2,
61.6, 56.0, 34.4. Anal. (C16H16ClN5O2S) C, H, N, S.
(2R,3R,4S)-2-(6-(3-Bromobenzylamino)-9H-purin-9-yl)-tet-
rahydrothiophene-3,4-diol (7d). Yield 71%; [R]23.7D -100.71 (c
0.139, DMSO); FAB-MS m/z 422 [M]+; mp 183.0-184.0 °C; UV
(MeOH) λmax 270.0 nm. 1H NMR (DMSO-d6) δ 8.46 (s, 1 H), 8.43
(br s, 1 H, D2O exchangeable), 8.21 (s, 1 H), 7.53 (s, 1 H)
7.42-7.24 (m, 3 H), 5.90 (d, 1 H, J ) 7.2 Hz), 5.53 (d, 1 H, J )
6.4 Hz, D2O exchangeable), 5.35 (d, 1 H, J ) 4.0 Hz, D2O
exchangeable), 4.71-4.66 (m, 2 H), 4.37-4.34 (m, 1 H), 3.41 (dd,
1 H, J ) 4.0, 10.8 Hz), 3.06 (q, 1 H, J ) 7.2 Hz). 2.79 (dd, 1 H,
J ) 2.8, 10.8 Hz). 13C NMR (DMSO-d6) δ 154.2, 152.4, 143.0,
140.0, 130.4, 129.8, 129.4, 126.2, 121.5, 78.3, 72.2, 61.6, 45.5,
34.5. Anal. (C16H16BrN5O2S) C, H, N, S.
(2R,3R,4S)-2-(6-(3-Iodobenzylamino)-9H-purin-9-yl)-tetrahy-
drothiophene-3,4-diol (7e). Yield 88%; [R]23.8D -97.08 (c 0.137,
DMSO); FAB-MS m/z 370 [M + H]+; mp 198.8-199.8 °C; UV
(MeOH) λmax 271.5 nm. 1H NMR (DMSO-d6) δ 8.45 (s, 1 H), 8.43
(br s, 1 H, D2O exchangeable), 8.21 (s, 1 H), 7.72 (s, 1 H), 7.56
(d, 1 H, J ) 7.2 Hz), 7.35 (d, 1 H, J ) 7.6 Hz), 7.10 (merged dd,
1 H, J ) 7.6 Hz), 5.90 (d, 1 H, J ) 7.2 Hz), 5.53 (d, 1 H, J ) 6.4
Hz, D2O exchangeable), 5.35 (d, 1 H, J ) 4.4 Hz, D2O exchange-
able), 4.71-4.66 (m, 2 H), 4.37-4.34 (m, 1 H), 3.41 (dd, 1 H, J
) 2.8, 10.8 Hz), 3.15 (d, 1 H, J ) 5.2 Hz), 2.79 (dd, 1 H, J ) 2.8,
10.8 Hz). 13C NMR (DMSO-d6) δ 154.2, 152.4, 149.2, 142.9, 140.0,
137.0, 135.7, 135.3, 130.4, 126.6, 94.7, 78.3, 72.2, 61.6, 42.2, 34.4.
Anal. (C16H16IN5O2S) C, H, N, S.
as a foam: [R]23.6 -157.63 (c 0.144, DMSO); FAB-MS m/z 313
D
[M + H]+; UV (MeOH) λmax 265.0 nm. 1H NMR (CDCl3) δ 8.67
(s, 1 H), 8.23 (s, 1 H), 5.88 (s, 1 H), 5.25-5.19 (m, 1 H), 3.69 (dd,
1 H, J ) 4.0, 13.2 Hz), 3.18 (d, 1 H, J ) 12.8 Hz), 1.51 (s, 3 H),
1.28 (s, 3 H). 13C NMR (CDCl3) δ 152.0, 151.4, 151.1, 144.3,
132.6, 111.9, 89.6, 84.3, 70.3, 40.8, 26.4, 24.6. Anal.
(C12H13ClN4O2S) C, H, N, S.
(2R,3R,4S)-2-(6-Chloro-9H-purin-9-yl)-tetrahydrothiophene-
3,4-diol (10). Hydrochloric acid (2 N, 12 mL) was added to a
solution of 9 (2.59 g, 8.28 mmol) in THF (20 mL), and the mixture
was stirred at room temperature overnight. The mixture was
neutralized with 1 N NaOH solution, and then the volatiles were
carefully evaporated under reduced pressure. The mixture was
subjected to a flash silica gel column chromatography (CH2Cl2:
MeOH ) 20:1) to give 10 (1.79 g, 79%) as a white solid: [R]23.5
D
-109.14 (c 0.164, DMSO); FAB-MS m/z 273 [M + H]+; mp
1
192.3-192.8 °C; UV (MeOH) λmax 264.5 nm. H NMR (DMSO-
d6) δ 9.02 (s, 1 H), 8.81 (s, 1 H), 6.02 (d, 1 H, J ) 7.2 Hz), 5.62
(d, 1 H, J ) 6.0 Hz, D2O exchangeable), 5.43 (d, 1 H, J ) 4.1 Hz,
D2O exchangeable), 4.74-4.70 (m, 1 H), 4.40-4.36 (m, 1 H), 3.47
L-4-Thiosugar acetate 12 was synthesized from D-gulonic acid
γ-lactone according to a similar procedure1,13 used for the prepara-
tion of 8 (Scheme 1). Then L-4-thiosugar acetate 12 was converted
to 13 according to a similar procedure used for the preparation of
9. The final L-4′-thio nucleosides 7f and 7g were synthesized from
12 according to the described general procedure for the synthesis
of 7a-7e.
(dd, 1 H, J ) 4.0, 11.2 Hz), 2.83 (dd, 1 H, J ) 2.8, 11.2 Hz). 13
C
NMR (DMSO-d6) 152.1, 151.6, 149.2, 146.6, 131.3, 78.6, 72.1,
62.4, 34.7. Anal. (C9H9ClN4O2S) C, H, N, S.
General Procedure for the Synthesis of 7a-7e. To a solution
of 10 in EtOH (5 mL) was added appropriate amine (1.5 equiv) at
room temperature, and the mixture was stirred at rt for a time period
ranging from 2 h to 3 d and evaporated. The residue was purified
by a flash silica gel column chromatography (CH2Cl2:MeOH )
20:1) to give 7a-7e.
The 1H, 13C NMR, UV, and mp data of L-series compounds
were the same as for the D-series of compounds as described above,
except that the specific optical rotations were in the opposite
direction. Yields of the L-series compounds were comparable with
those of the D-series of compounds.
(2R,3R,4S)-Tetrahydro-2-(6-(methylamino)-9H-purin-9-yl)th-
iophene-3,4-diol (7a). Yield 83%; [R]22.8 -175.60 (c 0.123,
Binding Assays.1,6 Human A1 and A2A ARs. For binding to
human A1 AR, [3H]CCPA (1 nM) was incubated with membranes
(40 µg/tube) from CHO cells stably expressing human A1 ARs 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 NECA. For human A2A AR binding,
membranes (20 µg/tube) from HEK-293 cells stably expressing
human A2A ARs were incubated with 15 nM [3H]CGS21680 at 25
°C for 60 min in 200 µL 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.
D
DMSO); FAB-MS m/z 268 [M + H]+; mp 223.9-224.8 °C; UV
(MeOH) λmax 266.0 nm. 1H NMR (DMSO-d6) δ 8.40 (s, 1 H), 8.23
(s, 1 H), 7.72 (br s, 1 H, D2O exchangeable), 5.89 (d, 1 H, J ) 7.2
Hz), 5.51 (d, 1 H, J ) 6.4 Hz, D2O exchangeable), 5.32 (d, 1 H,
J ) 4.4 Hz, D2O exchangeable), 4.70-4.64 (m, 1 H), 4.37-4.33
(m, 1 H), 3.40 (dd, 1 H, J ) 4.0, 10.8 Hz), 2.95 (s, 3 H), 2.79 (dd,
1 H, J ) 3.2, 10.8 Hz). 13C NMR (DMSO-d6) δ 154.9, 152.5,
148.8, 139.5, 119.5, 78.3, 72.2, 61.5, 34.6, 27.0. Anal.
(C10H13N5O2S) C, H, N, S.
(2R,3R,4S)-2-(6-(3-Fluorobenzylamino)-9H-purin-9-yl)tet-
rahydrothiophene-3,4-diol (7b). Yield 82%; [R]23.7D -141.22 (c
0.114, DMSO); FAB-MS m/z 362 [M + H]+; mp 180.5-180.7
Human and Rat A3 ARs. For competitive binding assay, each
tube contained 100 µL of membrane suspension (from CHO cells

Brief Articles
Journal of Medicinal Chemistry, 2008, Vol. 51, No. 20 6613
(7) (a) Lee, E. J.; Min, H. Y.; Chung, H. J.; Park, E. J.; Shin, D. H.;
Jeong, L. S.; Lee, S. K. A novel adenosine analog, thio-Cl-IB-MECA,
induces G0/G1 cell cycle arrest and apoptosis in human promyelocytic
leukemia HL-60 cells. Biochem. Pharmacol. 2005, 70, 918–924. (b)
Chung, H.; Jung, J.-Y.; Cho, S.-D.; Hong, K.-A; Kim, H.-J.; Shin,
D.-H.; Kim, H.; Kim, H. O.; Lee, H. W.; Jeong, L. S.; Gong, K. The
antitumor effect of LJ-529, a novel agonist to A3 adenosine receptor,
in both estrogen receptor-positive and estrogen-negative human breast
cancers. Mol. Cancer Ther. 2006, 5, 685–692.
(8) (a) Jacobson, K. A.; Siddiqi, S. M.; Olah, M. E.; Ji, X. d.; Melman,
N.; Bellamkonda, K.; Meshulam, Y.; Stiles, G. L.; Kim, H. O.
Structure-activity relationships of 9-alkyladenine and ribose-modified
adenosine derivatives at rat A3 adenosine receptors. J. Med. Chem.
1995, 38, 1720–1735. (b) Volpini, R.; Costanzi, S.; Lambertucci, C.;
Vittori, S.; Kliotz, K.-N.; Lorenzen, A.; Cristalli, G. Introduction of
alkynyl chains on C-8 of adenosine led to very selective antagonists
of the A3 adenosine receptor. Bioorg. Med. Chem. Lett. 2001, 11,
1931–1934. (c) Gao, Z.-G.; Kim, S.-K.; Biadatti, T.; Chen, W.; Lee,
K.; Barak, D.; Kim, S. G.; Johnson, C. R.; Jacobson, K. A. Structural
determinants of A3 adenosine receptor activation: nucleoside ligands
at the agonist/antagonist boundary. J. Med. Chem. 2002, 45, 4471–
4484.
stably expressing the human or rat A3 AR, 20 µg protein), 50 µL
of [125I]I-AB-MECA (0.5 nM), and 50 µL of increasing concentra-
tions of the nucleoside derivative in Tris ·HCl buffer (50 mM, pH
7.4) containing 10 mM MgCl2. Nonspecific binding was determined
using 10 µM of NECA 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, Gaithersburg, MD). Filters were washed
three times with 9 mL of ice-cold buffer. Radioactivity was
determined in a Beckman 5500B γ-counter.
For binding at all three subtypes, Ki values are expressed as mean
( sem, n ) 3-4 (outliers eliminated), and normalized against a
nonspecific binder, 5′-N-ethylcarboxamidoadenosine (NECA, 10
µM). Alternately, for weak binding, a percent inhibition of specific
radioligand binding at 10 µM, relative to inhibition by 10 µM
NECA assigned as 100%, is given.
Acknowledgment. This work was supported by the Korea
Research Foundation grant (KRF-2008-E00304) and the Intra-
mural Research Program of NIDDK, NIH, Bethesda, MD.
(9) Baraldi, P. G.; Tabrizi, M. A.; Romagnoli, R.; Fruttarolo, F.; Merighi,
S.; Varani, K.; Gessi, S.; Borea, P. A. Pyrazolo[4,3-e]1,2,4-triazolo[1,5-
c]pyrimidine ligands, new tools to characterize A3 adenosine receptors
in human tumor cell lines. Curr. Med. Chem. 2005, 12, 1319–1329.
(10) Gao, Z. G.; Blaustein, J.; 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.
Supporting Information Available: Elemental analyses data
for all unknown compounds and pharmacological methods. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(11) Kim, S.-K.; Gao, Z.-G.; Jeong, L. S.; Jacobson, K. A. Docking studies
of agonists and antagonists suggest an activation pathway of the A3
adenosine receptor. J. Mol. Graphics Modell. 2006, 25, 562–577.
(12) (a) Gao, Z.-G.; Joshi, B. V.; Klutz, A.; Kim, S.-K.; Lee, H. W.; Kim,
H. O.; Jeong, L. S.; Jacobson, K. A. Conversion of A3 adenosine
receptor agonists into selective antagonists by modification of the 5′-
ribofuran-uronamide moiety. Bioorg. Med. Chem. Lett. 2006, 16, 596–
601. (b) Jeong, L. S.; Lee, H. W.; Kim, H. O.; Tosh, D. K.; Pal, P.;
Choi, W. J.; Gao, Z.-G.; Patel, A. R.; Williams, W.; Jacobson, K. A.;
Kim, H.-D. Structure-activity relationships of 2-chloro-N6-substituted-
4′-thioadenosine-5′-N,N-dialkyluronamides as the human A3 adenosine
receptor antagonists. Bioorg. Med. Chem. Lett. 2008, 18, 1612–1616.
(13) Gunaga, P.; Kim, H. O.; Lee, H. W.; Toshi, D. K.; Ryu, J.-S.; Choi,
S.; Jeong, L. S. Stereoselective functionalization of the 1′-position of
4′-thionucleosides. Org. Lett. 2006, 8, 4267–4270.
(14) (a) Klotz, K.-N.; Hessling, J.; Hegler, J.; Owman, C.; Kull, B.;
Fredholm, B. B.; Lohse, M. J. Comparative pharmacology of human
adenosine receptor subtypes: characterization of stably transfected
receptors in CHO cells. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1998, 357, 1–9. (b) Gao, Z. G.; Mamedova, L.; Chen, P.; Jacobson,
K. A. 2-Substituted adenosine derivatives: affinity and efficacy at four
subtypes of human adenosine receptors. Biochem. Pharmacol. 2004,
68, 1985–1993.
(15) Kim, Y. C.; Ji, X.-d.; Melman, N.; Linden, J.; Jacobson, K. A. Anilide
derivatives of an 8-phenylxanthine carboxylic congener are highly
potent and selective antagonists at human A2B adenosine receptors.
J. Med. Chem. 2000, 43, 1165–1172.
(16) 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.
(17) Post, S. R.; Ostrom, R. S.; Insel, P. A. Biochemical methods for
detection and measurement of cyclic AMP and adenylyl cyclase
activity. Methods Mol. Biol. 2000, 126, 363–374.
(18) Ohno, M.; Gao, Z.-G.; Van Rompaey, P.; Tchilibon, S.; Kim, S.-K.;
Harris, B. A.; Gross, A. S.; Duong, H. T.; Van Calenbergh, S.;
Jacobson, K. A. Modulation of adenosine receptor affinity and intrinsic
efficacy in adenine nucleosides substituted at the 2-position. Bioorg.
Med. Chem. 2004, 12, 2995–3007.
References
(1) The preliminary accounts of this work have been published in Jeong,
L. S.; Choe, S. A.; Gunaga, P.; Kim, H. O.; Lee, H. W.; Lee, S. K.;
Tosh, D. K.; Patel, A.; Palaniappan, K. K.; Gao, Z.-G.; Jacobson,
K. A.; Moon, H. R. Discovery of a new nucleoside template for human
A3 adenosine receptor ligands: D-4′-thioadenosine derivatives without
4′-hydroxymethyl group as highly potent and selective antagonists.
J. Med. Chem. 2007, 50, 3159–3162.
(2) Olah, M. E.; Stiles, G. L. The role of receptor structure in determining
adenosine receptor activity. Pharmacol. Ther. 2000, 85, 55–75.
(3) (a) Jacobson, K. A.; Gao, Z.-G. Adenosine receptors as therapeutic
targets. Nat. ReV. Drug DiscoVery 2006, 5, 247–264. (b) Klotz, K.-N.
Adenosine receptors and their ligands. Naunyn-Schmiedeberg’s Arch.
Pharmacol. 2000, 362, 382–391. (c) 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.
(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) 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.
(6) (a) 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. (b) Jeong, L. S.; Lee,
H. W.; Jacobson, K. A.; Kim, H. O.; Shin, D. H.; Lee, J. A.; Gao,
Z.-G.; Lu, C.; Duong, H. T.; Gunaga, P.; Lee, S. K.; Jin, D. Z.; Chun,
M. W.; Moon, H. R. Structure-activity relationships of 2-chloro-N6-
substituted-4′-thioadenosine-5′-uronamides as highly potent and selec-
tive agonists at the human A3 adenosine receptor. J. Med. Chem. 2006,
49, 273–281. (c) Jeong, L. S.; Lee, H. W.; Kim, H. O.; Jung, J. Y.;
Gao, Z.-G.; Duong, H. T.; Rao, S.; Jacobson, K. A.; Shin, D. H.; Lee,
J. A.; Gunaga, P.; Lee, S. K.; Jin, D. Z.; Chun, M. W. Design,
synthesis, and biological activity of N6-substituted-4′-thioadenosines
at the human A3 adenosine receptor. Bioorg. Med. Chem. 2006, 14,
4718–4730.
(19) Melman, A.; Gao, Z.-G.; Kumar, D.; Wan, T. C.; Gizewski, E.;
Auchamapach, J. A.; Jacobson, K. A. Design of (N)-methanocarba
adenosine 5′-uronamides as species-independent A3 receptor-selective
agonists. Bioorg. Med. Chem. Lett. 2008, 18, 2813–2819.
JM8008647