Design and synthesis of 4″,6″-unsaturated cyclic ADP-carbocyclic-ribose, a Ca2+-mobilizing agent selectively active in T cells

Article abstract of DOI:10.1016/j.tet.2008.07.068

We previously developed cyclic ADP-carbocyclic-ribose (cADPcR, 3a) as a stable mimic of cyclic ADP-ribose (cADPR, 1), a Ca²⁺-mobilizing second messenger. The unsaturated carbocyclic-ribose analogs of cADPR, i.e., 4″,6″-didehydro-cADPcR (8a) and

Full text of DOI:10.1016/j.tet.2008.07.068

Tetrahedron 64 (2008) 9754–9765
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Design and synthesis of 400,600-unsaturated cyclic ADP-carbocyclic-ribose,
a Ca2þ-mobilizing agent selectively active in T cells
Takashi Kudoh a, Takashi Murayama b, Minako Hashii c, Haruhiro Higashida c, Takashi Sakurai b,
Clarisse Maechling d, Bernard Spiess d, Karin Weber e, Andreas H. Guse e, Barry V.L. Potter f,
,
Mitsuhiro Arisawa a, Akira Matsuda a, Satoshi Shuto a
*
a Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
b Department of Pharmacology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8421, Japan
c Department of Biophysical Genetics, Kanazawa University Graduate School of Medicine, Kanazawa 920-8640, Japan
d
´
´
Institut Gilbert Laustriat, Departement de Pharmacochimie de la Communication Cellulaire, UMR 7175, Faculte de Pharmacie, 74 route du Rhin, BP60024, F-67401 Illkirch, France
e The Calcium Signalling Group, University Medical Center Hamburg-Eppendorf, Center of Experimental Medicine, Institute of Biochemistry and Molecular Biology I: Cellular Signal
Transduction, Martinistrasse 52, 20246 Hamburg, Germany
f Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology, University of Bath, Claverton Down, Bath BA2 7AY, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 June 2008
Received in revised form 16 July 2008
Accepted 17 July 2008
Available online 23 July 2008
We previously developed cyclic ADP-carbocyclic-ribose (cADPcR, 3a) as a stable mimic of cyclic ADP-ribose
(cADPR,1), a Ca2þ-mobilizing second messenger. The unsaturated carbocyclic-ribose analogs of cADPR, i.e.,
400,600-didehydro-cADPcR (8a) and its inosine congener 400,600-didehydro-cIDPcR (8b) were newly designed
and successfully synthesized using the key intramolecular condensation reaction with S-phenyl
phosphorothioate-type substrates. The Ca2þ-mobilizing potency of the compounds was examined in sea
urchin egg homogenates, NG108-15 neuronal cells, and permeabilized Jurkat T-lymphocytes, which may
indicate that 400,600-didehydro-cADPcR is the first cADPR analog selectively active in T cells. Acid–base
behavior and conformation of 8a were also investigated and compared with those of cADPcR.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
difficult4a,b,6a and consequently, cADPR analogs have been synthe-
sized predominantly by enzymatic and chemo-enzymatic methods
Considerable attention has been focused on cyclic ADP-ribose
(cADPR, 1, Fig. 1), a naturally occurring metabolite of NADþ origi-
nally reported by Lee and co-workers.1 cADPR has been shown to
mobilize intracellular Ca2þ in various cells, such as sea urchin eggs,
using ADP-ribosyl cyclase-catalyzed cyclization under mild condi-
tions.5 For example, 8-NH2-cADPR (2), which is a potent antagonist
of cADPR, was enzymatically synthesized and effectively used in
biological studies.2,5a However, the analogs obtained by these
methods are limited due to the substrate-specificity of the ADP-
ribosyl cyclase.4a,b,5
Therefore, in recent years, methods for the chemical synthesis of
cADPR analogs have been extensively studied to develop useful
cADPR analogs, which could not be prepared by enzymatic and
chemo-enzymatic methods.6–8 In the synthesis of cADPR and its
analogs, construction of the large 18-membered ring structure is
the key step, and we have developed an efficient method for
forming the 18-membered ring employing S-phenyl phosphoro-
thioate-type substrates.6
cADPR is in a zwitterionic form with a positive charge around
the N(1)–C(6)–N6 moiety (pKa¼8.3), making the molecule unstable,
since the charged adenine moiety attached to the anomeric carbon
of the N1-linked ribose can be an efficient leaving group. Accord-
ingly, cADPR is readily hydrolyzed at the unstable N1-ribosyl link-
age of its adenine moiety to produce ADP-ribose (ADPR), even in
neutral aqueous solution.9 Under physiological conditions, cADPR
pancreatic b-cells, smooth muscle, cardiac muscle, T-lymphocytes,
and cerebellar neurons and therefore, in recent years, cADPR has
been recognized as a general mediator involved in Ca2þ signaling.2
A most recent property of cADPR shown using CD38 knockout mice
is that it is critical for social behavior by regulating oxytocin
secretion.3
Synthesis of cADPR analogs has been extensively investigated,4–8
since these can be used in investigating the mechanism of cADPR-
mediated Ca2þ signaling pathways and are also expected to be lead
structures for the development of potential drug candidates, due to
the important physiological roles of cADPR.2 Because of the unique
18-membered pyrophosphate structure and instability of cADPR,
chemical synthesis of cADPR and its analogs has proved to be rather
* Corresponding author. Fax: þ81 11 7063769.
E-mail address: shu@pharm.hokudai.ac.jp (S. Shuto).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.07.068

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9755
NH2
NH2
O
N
NH2
HO
HO
HO
HO
N
N
N
N
HO
N
N
HO
HO
N
N
N
N
N
N
1
X
X
X
O
N
O
N
O
N
O
O
O
O
O
P
O
O
O
O
P
P
O
P
O
P
O
P
O
P
O
OH
OH
OH
OH
O
O
O
O
O
O
O
O
O
O
O
O
P
OH
OH
OH
OH
O
O
O
O
5: X = H(3"-deoxy-cADPcR)
6: X = NH2
(8-NH2-3"-deoxy-cADPcR)
3b: cIDPcR
3a: X = H(cADPcR)
4: X = NH2(8-NH2-cADPcR)
1: X = H(cADPR)
2: X = NH2(8-NH2-cADPR)
O
N
NH2
HO
O
HO
HO
6
1"
2"
HO
N
N
HO
N
N
HO
N
N
N
N
2
N
1
3"
8
4"
O
P
O
6"
9
N
N
5"
O
O
P
O
O
O
1'
P
O
O
O
O
P
2'
O
O
(P1)
P
(P2)
4'
OH
OH
OH
O
O
O
O
O
O
O
O
O
3'
P
5'
OH
OH
OH
O
O
O
8b: 4",6"-didehydro-cIDPcR
8a: 4",6"-didehydro-cADPcR
7: cIDPR
Figure 1. cADPR (1) and its analogues.
is also hydrolyzed at the N1-ribosyl linkage by cADPR hydrolase to
give the inactive ADPR.9 Therefore, we designed cyclic ADP-car-
bocyclic-ribose (cADPcR, 3a) as a stable mimic of cADPR, in which
the oxygen atom in the N1-ribose ring of cADPR is replaced by
a methylene group. Using the chemical synthetic method described
above, we accomplished the total synthesis of cADPcR,6d to show
that it is actually resistant to both enzymatic and chemical hydro-
lysis due to its stable N1-alkyl linkage instead of the unstable
N1-glycosidic linkage of cADPR, and that it, like cADPR, effectively
mobilizes intracellular Ca2þ in sea urchin eggs.6d
Further structure–activity relationship (SAR) studies employing
cADPcR as the lead structure with evaluation in the sea urchin egg
system have shown that the N1-ribose moiety may be critically
important for Ca2þ-mobilizing potency and also agonist/antagonist
switching of cADPR activity.7 For example, deletion of the 300-hy-
droxyl of cADPcR resulted in providing the most potent Ca2þ-mo-
bilizing compound 300-deoxy-cADPcR (5) in the series;7c only
a subtle change, i.e., ‘–O–’ into ‘–CH2–’, brought about a dramatic
switch from a highly potent antagonist 8-NH2-cADPR (2) to an al-
most equipotent agonist 8-NH2-cADPcR (4);7a and also, when the
300-hydroxyl was removed from 8-NH2-cADPcR, it gave a borderline
compound 8-NH2-300-deoxy-cADPcR (6), a potent partial agonist
with a remarkably low EC50 value.7e
Our attention has been focused on developing cell-type selec-
tive cADPR analogs, which can be useful as biological tools and/or
potential drug leads. Thus, we found that although cADPcR is more
potent than cADPR in neuronal cells,7b it is almost inactive in Tcells,
while cADPR works effectively as a Ca2þ-mobilizing second mes-
senger in the same T cell system.6e,7c These are very important
findings indicating that the target proteins and/or the mechanism
of action of cADPR in T cells and neuronal cells are different.
As described above, synthetic studies of compounds employing
cADPcR as the lead structure have provided a series of biologically
and chemically stable mimics of cADPR and disclosed SAR of cADPR
analogs. In the course of these studies, we newly designed and
synthesized the unsaturated carbocyclic-ribose analog of cADPR,
i.e., 400,600-didehydro-cADPcR (8a) and its inosine congener 400,600-
didehydro-cIDPcR (8b). Biological evaluation in several systems
demonstrated that 400,600-didehydro-cADPcR is the first cADPR an-
alog selectively active in T cells. Acid–base behavior and confor-
mation of 8a were also investigated and compared with those of
cADPcR. In this paper, we describe these results in detail.10
2. Results and discussion
2.1. Design of the target compounds
Carbocyclic nucleosides have been known as effective mimics of
natural nucleosides.11 An antibiotic aristeromycin (Fig. 2) is a nat-
urally occurring mimic of adenosine, in which the ribose of aden-
osine is replaced by a carbocyclic-ribose, and has antiviral and
antitumor activities, due to its competitive inhibition to adenosine
against S-adenosylhomocysteine (AdoHcy) hydrolase.12,13 A cADPR
analog, the adenosine moiety of which was replaced with aristero-
mycin, has been reported as a hydrolysis resistant cADPR mimic.5h
Another nucleosidic antibiotic neplanocin A (Fig. 2), the struc-
ture of which corresponds to the 40,60-unsaturated derivative of
aristeromycin, is one of the most potent inhibitor of AdoHcy hy-
drolase known so far.12,14 Because of the remarkable inhibitory ef-
fect on the enzyme, neplanocin A has significant antiviral and also
antitumor effects, clearly surpassing aristeromycin. Therefore, an
unsaturated carbocyclic-ribose may mimic the ribose more pre-
cisely than the corresponding saturated congener.
Similar improvements in pharmacological potency of com-
pounds by replacing a saturated tetrahydrofuran or thiophene ring
with the corresponding unsaturated cyclopentene congener have
also been observed, for example, in studies of carba-analogs of
prostanoids15a or
compounds with a cyclopentene ring, the
b
-lactam antibiotic carbapenems.15b In these
-electrons of the un-
p
saturated bond might effectively mimic the unpaired electrons on
oxygen or a sulfur atom in the ring.
In nucleosides, conformation around the glycoside linkage is
one of the determinants of their biological activities, and they
generally prefer
a conformation that avoids steric repulsion
NH2
N
NH2
N
N
N
N
N
HO
N
N
HO
HO OH
HO OH
neplanocin A
aristeromycin
Figure 2. Naturally occurring carbocyclic adenine nucleosides.

9756
T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
sea urchin eggs and in neuronal cells.6d,7b We also planned to
B
A
synthesize and evaluate the 400,600-unsaturated analog 8b of cIDPcR,
since cyclic inosine diphosphate ribose (7, cIDPR) and its derivatives
H
2 N
H
2 N
9
9
N
N
N
N
Hβ
H
1
1
6"
6"
4"
N
N
synthesized recently have been disclosed to have potent Ca2þ
mobilizing effects.5g,8b–e
-
4"
NH2
NH2
OH
OH
OH
4",6"-didehydro-cADPcR (8a)
OH
cADPcR (3a)
2.2. Synthetic plan
Figure 3. Possible steric effect between the carbocyclic and adenine moieties: A, in
cADPcR (3a); B, in 400,600-didehydro-cADPcR (8a).
We planned to synthesize the target 400,600-didehydro-cADPcR
(8a) and its inosine congener 8b by a route, as summarized in
Scheme 1, in which intramolecular condensation reaction with a
S-phenyl phosphorothioate-type substrate forming the 18-mem-
bered pyrophosphate structure was employed as the key step. This
method was developed by us6 and has been effectively used by
other groups8 in the synthesis of a variety of cADPR analogs.
Thus, treatment of S-phenyl phosphorothioate-type substrates
9a or 9b having the N1-400,600-unsaturated carbocyclic-ribose with
AgNO3/MS 3 Å as a promoter6b–d was expected to form the desired
cyclized products, subsequent acidic treatment of which for
deprotection would furnish the target 8a or 8b. The S-phenyl
phosphorothioate-type substrates 9a and 9b could be converted
from the N1-unsaturated carbocyclic-ribosyl adenosine derivative
10a and its inosine congener 10b, which would be constructed by
condensation between the optically active cyclopentenyl amine 11
with known nucleoside 136d,17a or 14,6c,17b readily prepared from
inosine. The amine 11 would be prepared from 12, which can be
between the nucleobase and sugar moieties.16 Therefore, in cADPR
and its analogs, the most stable conformation can be that in which
steric repulsion between the adenine moiety and the N1- and
N9-ribose moieties, particularly between the adenine H-8 and the
N9-ribose H-20 and also between the adenine H-2 and the N1-ri-
bose H-200, is minimal. It should be noted that, in cADPcR, the sp3-
hydrogens on the tetrahedral C600, especially the H-600
b, which is
absent in cADPR, can be sterically repulsive to the adenine H-2
(Fig. 3A), which might make its conformation different from that of
cADPR at least to some extent. When the carbocyclic-ribose of the
N1-moiety is replaced with the corresponding 4,6-unsaturated one,
a sterically repulsive sp3-hydrogen on the C600 is absent, as shown in
Figure 3B. Therefore, we thought that, in 400,600-didehydro-cADPcR
(8a), three-dimensional positioning of the N1-unsaturated carbo-
cyclic-ribose and the adenine moieties might be more analogous to
that in cADPR.
Another difference between cADPR and cADPcR is the pKa value
for protonation at the N6-position. The pKa value of cADPcR is
somewhat higher compared with that of cADPR,6d which might
affect features important for binding to target proteins. In-
troduction of an unsaturated bond into the carbocyclic-ribose of
cADPcR might lower the pKa and accordingly, the pKa of 400,600-
didihydro-cADPcR might be more similar to that of cADPR.
Taking these results and considerations into account, we de-
cided to synthesize the 400,600-unsaturated analog 8a (400,600-dide-
hydro-cADPcR) of cADPcR as a new target and investigate its
Ca2þ-mobilizing effect.
synthesized from D-ribose by the procedure recently developed by
Jeong and co-workers.18
2.3. Synthesis
The chiral cyclopentenyl amine 11 was prepared from a known
chiral cyclopentenone 1218as shown in Scheme 2.19 1,2-Reduction
of the enone system of 12 with NaBH4/CeCl3 stereoselectively gave
the allylic
a-alcohol 15, of which mesylation and subsequent
treatment with LiN3 in HMPA/DMSO afforded the
b-azide 16. Re-
We previously synthesized cyclic IDP-carbocyclic-ribose (3b),6a,c
the inosine congener of cADPcR, and showed that it was inactive in
moval of the TBDPS group of 16, followed by reduction of the azido
group with Ph3P in aqueous THF gave the desired amine 11.
O
O
O
O
X
X
N
N
N
N
N
N
O
HO P
O
O
PhS P
O
OH
O
O
N
TBSO
N
O
O
8a or 8b
O
O
9a,b
O
O
a series: X = NH
10a,b
b series: X = O
NH2
O
OH
O
TBDPSO
HO
HO
O
O
O
O
HO OH
D-ribose
O
11
12
NO2
O
N
+
CN
N
N
N
N
N
N
N
NO2
N
TBSO
TBSO
N
CH(OMe)
or
N
HO
O
O
O
O
O
O
O
HO
OH
inosine
14
13
Scheme 1. Synthetic plan for the target compounds 8a and 8b.

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9757
was obtained in 65% yield. Similarly, the other substrate 9b was
condensed to give the cyclization product 20b in 57% yield. Finally,
removal of the isopropylidene groups of 20a and 20b with aqueous
HCO2H, followed by treatment with aqueous ammonia furnished
400,600-didehydro-cADPcR (8a) and its inosine congener 400,600-
didehydro-cIDPcR (8b), respectively.
1. MsCl, Et3N
CH2Cl2, 0 °C
2. LiN3
OH
NaBH4, CeCl3 TBDPSO
MeOH, 0 °C
HMPA/DMSO
12
O
O
90% from 12
15
N3
2.4. Ca2D-mobilizing activity in sea urchin egg homogenate
1. TBAF, THF
2. Ph3P, aq. THF
TBDPSO
11
The Ca2þ-mobilizing ability of 400,600-didehydro-cADPcR (8a) and
its inosine congener 8b in sea urchin egg homogenate was de-
termined fluorometrically by monitoring Ca2þ with fura-2,24 and
the results were compared with those of cADPR (1) and cADPcR
(3a) (Fig. 4). 400,600-Didehydro-cADPcR (8a) showed slight Ca2þ re-
O
O
16
89%
Scheme 2. Synthesis of the chiral cyclopentenyl amine 11.
400,600-Didehydro-cADPcR (8a) and -cIDPcR (8b) were success-
fully synthesized from the chiral cyclopentenyl amine 11 as shown
in Scheme 3. The 400,600-didehydro-N1-carbocyclic-ribosyl adenosine
derivative 10a was obtained in 58% yield by the treatment of
a mixture of the amine 11 and the imidazole nucleoside 13 with
K2CO3 in MeOH at room temperature. When the amine 11 was
heated with inosine derivative 14 in DMF at 50 ꢀC and then the
resulting mixture was treated with K2CO3, it produced the corre-
sponding N1-substituted inosine derivative 10b in 35% yield. The
500-hydroxy group of 10a or 10b was protected with a dimethoxy-
trityl (DMTr) group and the 50-O-TBS group of the product was re-
moved with TBAF to give 17a or 17b, respectively. Treatment of 17a
or 17b with an S,S0-diphenylphosphorodithioate (PSS)/2,4,6-triiso-
propylbenzenesulfonyl chloride (TPSCl)/pyridine system20 gave the
50-bis(phenylthio)phosphate 18a or 18b, respectively. After removal
of 500-O-DMTr group of 18a or 18b with aqueous AcOH, a phosphoryl
group was introduced at the resulting 500-primary hydroxyl by
Yoshikawa’s method with POCl3/(EtO)3PO,21 followed by treatment
of the product with H3PO2 and Et3N22 in the presence of N-methyl-
maleimide (NMM) in pyridine,6d to afford the corresponding
S-phenyl phosphorothioate 9a or 9b, respectively. Although the
yield of this step was low, probably due to the unstable allylic
phosphate system in the reaction product, the substrates 9a and 9b
for the key intramolecular condensation reaction23 were obtained.
When a solution of 9a in pyridine was added slowly to a mixture
of a large excess of AgNO3 and Et3N in the presence of MS 3 Å in
pyridine at room temperature,6 the desired cyclization product 20a
lease above 10 mM (Fig. 4A). The EC50 value of 8a was w190 mM
(assuming that it is a full agonist), which is over 2000 times larger
than that of cADPcR (EC50¼79 nM). Ca2þ release by 1
m
M cADPcR
after treatment with 8a was not affected up to 3
mM and consis-
tently reduced at 10 and 30 mM, indicating that 8a does not have
antagonistic activity (Fig. 4B). Thus, 8a is only slightly active for
cADPR-induced Ca2þ mobilization in sea urchin egg homogenate.
The inosine congener 8b did not show any agonistic and antago-
nistic activity on cADPR-induced Ca2þ release.
2.5. Ca2D-mobilizing activity in neuronal cells
The effect of 400,600-didehydro-cADPcR (8a) and its inosine
congener 8b on the Ca2þ-mobilizing activity was also studied in
neuronal cells, using the combined patch voltage-clamp and Ca2þ
-
monitoring technique.7b The membrane of NG108-15 mouse neu-
roblastoma x rat glioma hybrid cell was sealed with a patch pipette,
disrupted to infuse the compound, and then voltage-clamped. The
two analogs 8a and 8b did not evoke any apparent changes in
[Ca2þ]i by themselves at the resting membrane potential of around
ꢁ40 mV, as previously observed for cADPcR.7b,c We next examined
the effect of 8a and 8b on depolarization-induced [Ca2þ]i elevation
using an identical protocol for testing cADPcR (Fig. 5). Membrane
depolarization from ꢁ40 to ꢁ20 mV in control cells evoked a slight
[Ca2þ
]
increase with an average peak value of 167ꢂ18%
i
O
O
O
O
X
N
N
X
N
N
N
N
N
N
O
ODMTr
e
OH
RO
11
O
O
(PhS)2P
O
c
a or b
+
10a,b
O
O
13 or 14
O
O
19a,b
a series: X = NH
b series: X = O
d
17a,b: R = H
18a,b: R = PO(SPh)2
OH
O
O
X
N
O
O
X
N
HO
N
X
N
N
N
N
O
N
N
N
O
O
PhS P
O
O
P
O
O
P
O
O
P
O
P
O
P
O
N
O
N
O
HO
O
O
O
O
g
h
O
f
OH
O
O
O
O
O
O
HO
O
O
9a,b
20a,b
8a,b
Scheme 3. Synthesis of the target compounds 8a and 8b. Reagents and conditions: (a) K2CO3, MeOH, rt, 58% (10a); (b) (1) DMF, 50 ꢀC, (2) K2CO3, DMF, 50 ꢀC, 35% (10b); (c)
(1) DMTrCl, pyridine, rt, (2) TBAF, THF, AcOH, rt, quant. (17a), 61% (17b); (d) PSS, TPSCl, pyridine, rt, 74% (18a), 71% (18b); (e) aq 60% AcOH, rt, 80% (19a), 79% (19b); (f) (1) POCl3,
(EtO)3PO, 0 ꢀC, (2) H3PO2, Et3N, NMM, pyridine, 0 ꢀC–rt, 28% (9a), 36% (9b); (g) AgNO3, MS 3 Å, Et3N, pyridine, rt, 65% (20a), 57% (20b); (h) aq HCO2H, rt, (2) aq NH3, 86% (8a),
82% (8b).

9758
T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
Ca2þ from intracellular stores with similar potency.5g While cADPcR
(3a) as compared to cADPR (1) was only a weak agonist for Ca2þ
release in permeabilized Jurkat T cells,6e 400,600-didehydro-cADPcR
(8a) was much more potent (Fig. 6A). At 100
cADPcR induced a robust response while almost no effect of
cADPcR was obtained (Fig. 6A). Interestingly, at 500
M 400,600-
m
M 400,600-didehydro-
m
didehydro-cADPcR a significantly higher amplitude was observed
as compared to cADPR (Fig. 6A). Though cIDPR showed a concen-
tration–response curve very similar to cADPR,5g its carbocyclic
inosine congener cIDPcR (3b) was not active up to 500 mM (Fig. 6A).
In the next series of experiments, the potential antagonistic
effect of cADPcR, cIDPcR, 400,600-didehydro-cADPcR, and 400,600-
didehydro-cIDPcR on Ca2þ release induced by a submaximal
concentration of cADPR (30
significant differences as compared to controls were observed up to
500 M for all compounds (Fig. 6B).
mM) was analyzed. No statistically
m
2.7. NMR titration study
We next investigated the acid–base behavior of the ionizable
groups for 400,600-didehydro-cADPcR (8a) and also for cADPcR (3a),
since the ionizable groups, such as N6-imino or the pyrophosphate
group, may have a critical role in the binding of the compound with
target biomolecules.
The 1H and 31P NMR titration experiments were done in the pH
range 2.5–11.5 for cADPcR and 400,600-didehydro-cADPcR in 2H2O.
Provided that the observed chemical shift variation for the phos-
phorus and for the proton resonances mainly depends on the
electronic effects accompanying the protonation process, the pro-
tonation equilibrium constants could be derived from the chemical
shift variations of the nuclei as a function of pH.25 Both compounds
show only one protonation equilibrium in the studied pH range,
which corresponds to an imino base equilibrium on the adenine
ring (Fig. 7).4 The corresponding pKa values determined are, re-
spectively, 8.65 and 8.45 for cADPcR and 400,600-didehydro-cADPcR.
No acidic pKa value was evidenced showing that phosphate group
protonation takes place at a pH lower than 2.5.
Therefore, while the ionizable groups of cADPcR and of 400,600-
didehydro-cADPcR behave similarly in the pH range 2.5–11.5, in-
troduction of an unsaturated bond into the carbocyclic-ribose of
cADPcR (8.65) lowered the pKa, as we expected. The pKa of 400,600-
didehydro-cADPcR (8.45) was rather similar to that of cADPR
(8.3).26
Figure 4. Ca2þ-mobilizing activity of 400,600-didehydro-cADPcR (8a) and its inosine
congener 8b in sea urchin egg homogenate. The activity was measured fluorometri-
cally with fura-2. (A) Dose-dependent Ca2þ release by the compounds. The Ca2þ re-
lease by each compound was normalized by Ca2þ release with 10
m
M cADPR. (B)
Evaluation of antagonistic activity. Ca2þ release by 1
mM cADPcR was measured after
application of indicated concentrations of the compounds. Reduction in the cADPcR-
induced Ca2þ release by 8a corresponds to its agonistic activity, indicating that it has
no antagonistic activity. Data are meanꢂSEM (n¼3).
Difference in chemical shift values, Dd
¼
dp dd (ppm), between
ꢁ
fully protonated, dp, and fully deprotonated, dd, state is shown in
Table 1. The adenine ring protons H2 and H8, close to the imine
group, are good reporter nuclei for the protonation, and conse-
quently show relatively large chemical shift variation, in almost the
same extent for both compounds.
(meanꢂSEM; n¼4) of the pre-depolarization level, resulting from
activated voltage-activated Ca2þ channels (VACCs) and subsequent
Ca2þ release by Ca2þ-induced Ca2þ release (CICR) mechanism. In-
jection of 8a at 10 mM, which could induce a near maximum re-
sponse and discriminate the difference in efficiency, showed slight
and statistically non-significant potentiating effect on the de-
In the N9-linked ribose ring, only H10, the nearest from the pro-
tonable group, shows a significant variation in chemical shift values,
and the curves for both compounds are almost superimposed.
The phosphorus nuclei of P1, the phosphate group linked to the
N9-ribose moiety, and P2, the phosphate linked to the N1-car-
boxyribose moiety, are differently affected by protonation in
cADPcR and 400,600-didehydro-cADPcR (Fig. 8A). Both phosphate
groups of 400,600-didehydro-cADPcR are quite equally sensitive to the
protonation process, whereas only phosphate P2, linked to the
carbocyclic-ribose ring, shows significant chemical shift variation
during protonation (Fig. 8A).
Finally, in case of the N1-linked carbocyclic-ribose ring, one can
see in Table 1 clear differences in cADPcR and 400,600-didehydro-
cADPcR data. The most interesting nucleus is the H100 proton, which
shows a negative chemical shift variation during protonation in the
case of cADPcR and a positive chemical shift variation in case of the
didehydro analog 8a (Fig. 8B). Such a wrongway shift, as for
polarization-induced [Ca2þ
] ] level was
rise. The average [Ca2þ
i
i
207ꢂ22% (n¼9) at the initial peak. The inosine congener 8b showed
no effect on depolarization-induced [Ca2þ]i increases (Fig. 5A and
B). On the other hand, injection of 10 mM of cADPcR facilitated
depolarization-induced [Ca2þ]i increases: the peak [Ca2þ]i level was
295ꢂ30% (n¼4) of the pre-depolarization level, which was equal or
more potent than that of natural cADPR, as previously described.7b,c
Therefore, our results indicate that 8a does not possess apparent
Ca2þ-mobilizing activity in NG108-15 neuronal cells, in contrast to
cADPcR as the potent ligand.
2.6. Ca2D-mobilizing activity in T cells
The Ca2þ-mobilizing effect of 8a and 8b was evaluated also in
permeabilized Jurkat T cells. Both cADPR (1) and cIDPR (7) released

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9759
Figure 5. Effects of cADPR, cADPcR, 400,600-didehydro-cADPcR (8a), and its inosine congener 8b on membrane depolarization-evoked [Ca2þ]i increases in patch-clamped NG108-15
cells. (A) Traces show [Ca2þ]i changes in fura-2 loaded cells infused with 10 mM of cADPcR, cADPR, 8a, 8b or without compound (buffer). At about 100 s before the beginning of each
trace, cell membranes were ruptured with pipettes filled with these test compounds. Then, the membranes were voltage-clamped at ꢁ40 mV, followed by depolarization to
ꢁ20 mV at the indicated point. (B) [Ca2þ]i levels were plotted at the peak after depolarization to ꢁ20 mV and are given as percentages of that at ꢁ40 mV. Each bar is meanꢂSEM of
4–9 experiments. *Values in cells pretreated with cADPcR or cADPR were significantly higher than those in cells without compounds at p<0.05.
cADPcR, may be explained by a conformational change, already
observed, for instance, when a phosphate group approaches the
proton,27 or by a modification in ring currents during protonation.28
No unexpected effect is seen in H200 titration profile, this proton
is much more affected by the protonation in cADPcR than in 400,600-
didehydro-cADPcR. Almost the same profile, but in a lesser extent,
is seen for H600 nuclei.
2.8. Conformational analysis
The three-dimensional structure of biologically active com-
pounds in aqueous solution is very important from the viewpoint of
bioactive conformation. The conformations of 400,600-didehydro-
cADPcR (8a) were constructed by molecular dynamics calculations
with a simulated annealing method based on the NOE constraints
of the intramolecular proton pairs measured in D2O.7c
In Figure 9, the structure of 8a obtained is shown with those of
cADPR (1) and cADPcR (3a) by the same method.7c The calculated
structures of cADPR (Fig. 9A) and cADPcR (Fig. 9B) were analogous,
each adopting a 20-endo form in the N9-ribose moiety and a syn-
form around the N9-glycosyl linkage. However, their three-di-
mensional arrangement of the N1-ribose moiety and the adenine
ring is somewhat different. In cADPcR, the adenine H-2 is located
above the center of the N1-carbocyclic-ribose ring due probably to
its repulsion to the H-600
b, where the H-2 is located just above the
ring-oxygen of the N1-ribose in cADPR.
In 400,600-didehydro-cADPcR (8a), arrangement of the N1-ribose
moiety and the adenine ring is inverted compared with those of
cADPR and cADPcR, in which the rather bulky 6-NH2 group is lo-
cated just above the plane sp2-C6, and the H2 is near to the H100 of
the carbocyclic-ribose. This may be because a sterically repulsive
b
-hydrogen on the sp3-C6 in cADPcR is absent in 400,600-didehydro-
cADPcR. In addition, the flat C100–C600–C400–C500-structure might
distort the conformation of the large 18-membered ring. As a result,
its entire three-dimensional structure of the molecule is
Figure 6. Effects of cADPR and its analogues in permeabilized Jurkat T cells. Jurkat T
cells were permeabilized by saponin. Ca2þ stores were refilled by addition of ATP and
ATP-regenerating system. [Ca2þ] was determined fluorometrically using fura-2/free
acid. (A) Compounds were added at the concentrations indicated. Data are differences
between the Ca2þ concentration upon addition and before addition of compound.
MeanꢂSEM (n¼3–6). (B) Compounds were added at the concentrations indicated prior
to addition of cADPR (30 mM). Data represent the Ca2þ release by cADPR (30 mM) and
are differences between the Ca2þ concentration upon addition and before addition of
cADPR. MeanꢂSEM (n¼3–6).
NH2
NH
N
N
N
N
N
N
N
H
N
Figure 7. Imino base equilibrium on the adenine moiety.

9760
T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
significantly different from that of cADPR or cADPcR, which can be
the reason why 400,600-didehydro-cADPcR is almost inactive in sea
urchin egg and neuronal cell systems.
2.9. Discussion
As described above, the 400,600-unsaturated analogs 8a and 8b
have been successfully synthesized, which, concomitant with the
previous examples, clearly demonstrates that the strategy employ-
ing a S-phenyl phosphorothioate-type substrate in the key intra-
molecular condensation reaction forming the pyrophosphate linkage
is very effective in total syntheses of cADPR-related compounds.
Considering the highly potent biological activity of neplanocin
A, the 40,60-unsaturated carbocyclic adenosine analog, we designed
and synthesized 400,600-didehydro-cADPcR (8a). However, 400,600-
didehydro-cADPcR is shown to be almost inactive in sea urchin
homogenate and also in neuronal cell systems. The conformational
analysis study suggested that the entire three-dimensional struc-
ture of 400,600-didehydro-cADPcR is significantly different from that
of cADPR and cADPcR, which might be the reason why 400,600-
didehydro-cADPcR has only slight Ca2þ-mobilizing activity in the
sea urchin and neuronal cell systems.
The titration studies suggest that significant conformational
change of the N1-carbocyclic-ribose moiety may occur dependent
on the adenine ring protonation in cADPcR, but not in 400,600-dide-
hydro-cADPcR. Although similar protonation-dependent conforma-
tional changes have been observed in biologically active phosphate
compounds, such as inositol trisphosphate or inositol tetraki-
sphosphate,26 this is the first example disclosed in cADPR-related
A
-8.8
-9.0
P2 (3a)
-9.2
-9.4
-9.6
P2 (8a)
-9.8
-10.0
-10.2
-10.4
-10.6
P1 (3a)
P1 (8a)
2
4
6
8
10
12
pH
6.2
6.0
5,8
5.6
5.4
5.2
5.0
4.8
4.6
4.4
4.2
B
H1" (8a)
H1" (3a)
2
4
6
8
10
12
pH
31
Figure 8. (A) P NMR titration of cADPcR (3a) (circles) and 400,600-didehydro-cADPcR
(8a). Full symbols stands for P1 (phosphate group linked to the N9-ribose moiety) and
open symbols stands for P2 (phosphate group linked to the N1-carboxyribose moiety).
1
(B) H NMR titration of H100 nucleus of cADPcR (3a) (circles) and 400,600-didehydro-
cADPcR (8a) (triangles).

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9761
B
A
H2
H2
N6
N6
N1-carbocyclic
ribose
-
N1-ribose
C
N6
H2
N1-carbocyclic-
ribose
Figure 9. Calculated structures of cADPR (1, A), cADPcR (3a, B), and 400,600-didehydro-cADPcR (8a, C) by molecular dynamics calculations with a simulated annealing method based
on the NOE data in D2O.
compounds. This kind of conformational change might be impor-
tant in the interaction between the target proteins and cADPR or
its analogs, and the difference between cADPcR and 400,600-dide-
hydro-cADPcR in the biological activity might be related to the
protonation-dependent conformational behavior of the molecule.
We showed that 400,600-didehydro-cADPcR is active in T cells, but
is almost inactive in sea urchin egg homogenate and also in neu-
ronal cells, which may indicate some selectivity in T cells. cADPR
analogs having selectivity between mammalian cell types, i.e.,
compounds selectively active in some mammalian cells but inactive
in other mammalian cells would be useful tools to investigate the
target biomolecules as well as the Ca2þ-mobilizing mechanisms.
We recently identified cADPcR and its 300-deoxy derivative 5 as the
first kind of mammalian cell-type selective compound: these
compounds are significantly active in neuronal cells but almost
inactive in T cells. 400,600-Didehydro-cADPcR was identified as an-
other example having selectivity between mammalian cell types.
The cell-type selectivities of 400,600-didehydro-cADPcR and cADPcR
are complementary, i.e., T cell-selective and neuronal cell-selective,
although no other cell types have yet been studied. These results
may indicate that the recognition mechanism of cADPR by the
target proteins could be different in T cells and neuronal cells.
However, on the other hand, our results may simply reflect the
difference between the cell preparations used: while in the neu-
ronal model compounds were infused by the patch pipette, per-
meabilized T cells were prepared by several wash steps during
which cytosolic proteins are lost. Although cytosolic proteins also
diffuse into patch pipettes, the loss of potentially important factors
is slow and likely less complete.
It is interesting that some aspects of the SAR of compounds in
neuronal cells are likely to be related to that in sea urchin eggs. For
example, cADPcR and 300-deoxy-cADPcR are significantly active
both in sea urchin egg homogenate and neuronal cells, but inactive
in T cells.7c On the contrary, 400,600-didehydro-cADPcR and -cIDPcR
are inactive in sea urchin egg homogenate and neuronal cells, but
active in T cells. On the other hand, the EC50 of cADPR in sea urchin
egg homogenates is 79 nM, while active concentrations in mam-
malian cells, e.g., neuronal or T cells, are in the micromolar-range.
Clear-cut conclusions require molecular identification of cADPR
receptors. Recent data showing that cADPR, in addition to induction

9762
T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
of intracellular Ca2þ release via ryanodine receptors, may also be
involved in the fine tuning of the plasma membrane Ca2þ and Naþ
channel TRPM2.30 Thus, this second cADPR target complicates the
situation dramatically, especially when data are obtained in intact
cells.30
desilylated product (649 mg, 93%) as a colorless oil. A mixture of
the desilylated product (649 mg, 3.00 mmol) and Ph3P (1.58 g,
6.00 mmol) in THF/H2O (1:1, 30 mL) was stirred at room temper-
ature for 7 h. The reaction mixture was partitioned between EtOAc
and H2O, the organic layer was washed with H2O, and the com-
bined water layer was evaporated and lyophilized to give 11
3. Conclusion
(533 mg, 96%) as a white solid. 1H NMR (DMSO-d6, 500 MHz)
d 5.51
(m, 1H), 5.02 (d, 1H, J¼5.6 Hz), 4.25 (d, 1H, J¼5.6 Hz), 4.01 (d, 1H,
The unsaturated carbocyclic-ribose analogs of cADPR, i.e., 400,600-
didehydro-cADPcR (8a) and its inosine congener 400,600-didehydro-
cIDPcR (8b), designed as novel stable mimics of cADPR, were
successfully synthesized using the key intramolecular condensa-
tion reaction with the S-phenyl phosphorothioate-type substrates.
Titration studies on the compounds suggest that significant con-
formational change might occur dependent on protonation, which
might be important for the biological activity of cADPR and its
analogs. Evaluation of Ca2þ-mobilizing potency using three differ-
ent biological systems, i.e., sea urchin egg homogenates, NG108-15
neuronal cells, and permeabilized Jurkat T-lymphocytes may
indicate that 400,600-didehydro-cADPcR is the first cADPR analog to
be selectively active in T cells. Thus, this study may be an entry to
developing T cell-selective cADPR analogs, which can be useful as
biological tools and/or drug leads.
J¼15.6 Hz), 3.96 (d, 1H, J¼15.6 Hz), 3.65 (m, 1H), 1.26, 1.23 (each s,
each 3H); 13C NMR (DMSO-d6, 125 MHz)
d 145.6, 129.3, 110.1, 87.5,
83.2, 61.6, 58.1, 27.5, 25.9; HRMS (FAB, positive) calcd for C9H16NO3:
186.1130 (MHþ), found: 186.1128.
4.1.3. N-1-{(1R,2S,3R)-4-Hydroxymethyl-2,3-
isopropylidenedioxycyclopent-4-en-1-yl}-50-O-(tert-
butyldimethylsilyl)-20,30-O-isopropylideneadenosine (10a)
A mixture of 13 (39 mg, 90
mmol), 11 (17 mg, 90 mmol), and
K2CO3 (1 mg, 7 mol) in MeOH (1 mL) was stirred at room tem-
m
perature for 28 h and then evaporated. The residue was partitioned
between EtOAc and H2O, and the organic layer was washed with
brine, dried (Na2SO4), and evaporated. The residue was purified by
column chromatography (SiO2, 5% MeOH in CHCl3) to give 10a
(31 mg, 58%) as an oil. 1H NMR (CDCl3, 500 MHz)
d 7.80 (s, 1H), 7.62
(s, 1H), 6.00 (d, 1H, J¼2.8 Hz), 5.91 (m, 1H), 5.73 (m, 1H), 5.22 (d, 1H,
J¼5.7 Hz), 5.04 (dd, 1H, J¼2.8, 6.1 Hz), 4.86 (dd, 1H, J¼6.1, 2.3 Hz),
4.59 (d,1H, J¼5.7 Hz), 4.46 (d, 1H, J¼15.1 Hz), 4.40 (d,1H, J¼15.1 Hz),
4.38 (ddd, 1H, J¼2.3, 3.7, 4.0 Hz), 3.82 (dd, 1H, J¼3.7, 11.3 Hz), 3.77
(dd, 1H, J¼4.0, 11.3 Hz), 1.59, 1.46, 1.36, 1.35 (each s, each 3H), 0.84 (s,
9H), 0.030, 0.024 (each s, each 3H); 13C NMR (CDCl3, 125 MHz)
4. Experimental
4.1. General
Chemical shifts are reported in parts per million downfield from
Me4Si (1H), MeCN (13C) or H3PO4
(
31P). All of the 1H NMR assign-
d 154.4, 151.9, 144.3, 140.8, 136.7, 123.7, 122.9, 114.1, 112.6, 91.3, 86.9,
ments described were in agreement with COSY spectra. Thin-layer
chromatography was done on Merck coated plate 60F254. Silica gel
chromatography was done on Merck silica gel 5715. Reactions were
carried out under an argon atmosphere.
85.4, 84.7, 83.6, 81.3, 65.9, 63.5, 59.9, 27.4, 27.2, 25.8, 25.3,18.0, ꢁ5.5,
ꢁ5.6; HRMS (FAB, positive) calcd for C28H44N5O7Si: 590.3010
(MHþ), found: 590.3019; UV (MeOH) lmax¼260, 295 (sh) nm.
4.1.4. N-1-{(1R,2S,3R)-4-Dimethoxytrityloxymethyl-2,3-
isopropylidenedioxycyclopent-4-en-1-yl}-20,30-O-
isopropylideneadenosine (17a)
4.1.1. (1R,2S,3R)-1-Azido-4-(tert-butyldiphenylsilyloxymethyl)-2,3-
O-isopropylidene-4-cyclopenten-2,3-diol (16)
A mixture of 12 (1.19 g, 2.82 mmol), NaBH4 (160 mg, 4.23 mmol),
and CeCl3$7H2O (1.05 g, 2.82 mmol) in MeOH (30 mL) was stirred at
0 ꢀC for 10 min, and then aqueous AcOH (1 M, 2 mL) was added. The
mixture was partitioned between EtOAc and H2O, and the organic
layer was washed with brine, dried (Na2SO4), and evaporated. A
A mixture of 10a (117 mg, 0.199 mmol) and DMTrCl (202 mg,
0.60 mmol) in pyridine (2 mL) was stirred at room temperature for
10 h. After addition of MeOH (2 mL), the resulting mixture was
evaporated. The residue was partitioned between EtOAc and H2O,
and the organic layer was washed with brine, dried (Na2SO4), and
mixture of the residue, MsCl (327
m
L, 4.23 mmol), and Et3N
evaporated. A mixture of the residue, TBAF (1.0 M in THF, 400
mL,
(1.18 mL, 8.46 mmol) in CH2Cl2 (30 mL) was stirred at 0 ꢀC for
15 min, and then ice cooled water was added. The mixture was
partitioned between EtOAc and ice cooled water, and the organic
layer was washed with brine, dried (Na2SO4), and evaporated. A
mixture of the residue, LiN3 (276 mg, 5.64 mmol), and HMPA
0.40 mmol), and AcOH (11 L, 0.2 mmol) in THF (2 mL) was stirred
m
at room temperature for 1 h and then evaporated. The residue was
purified by column chromatography (SiO2, 5% MeOH in EtOAc) to
give 17a (174 mg, quant.) as an oil. 1H NMR (CDCl3, 500 MHz)
d 7.66
(s, 1H), 7.63 (s, 1H), 6.84–7.47 (m, 13H), 6.00 (m, 1H), 5.97 (m, 1H),
5.78 (d, 1H, J¼4.3 Hz), 5.05 (m, 3H), 4.55 (d, 1H, J¼5.6 Hz), 4.48 (m,
1H), 4.01 (m, 1H), 3.92 (m, 1H), 3.82 (m, 1H), 3.79 (s, 6H), 3.75 (m,
1H), 1.64, 1.42, 1.37, 1.30 (each s, each 3H); 13C NMR (CDCl3,
(245 mL, 1.41 mmol) in DMSO (30 mL) was stirred at room tem-
perature for 36 h. The mixture was partitioned between EtOAc and
H2O, and the organic layer was washed with brine, dried (Na2SO4),
and evaporated. The residue was purified by column chromatog-
raphy (SiO2, 5% EtOAc in hexane) to give 16 (1.14 g, 90%) as a col-
125 MHz) d 167.7, 158.6, 154.0, 151.7, 144.6, 144.5, 139.8, 138.1, 136.0,
135.7, 132.4, 130.8, 129.9, 129.9, 128.8, 128.0, 127.9, 126.9, 125.3,
121.4, 114.2, 113.2, 112.6, 93.9, 86.7, 85.7, 84.6, 84.0, 83.6, 81.4, 68.1,
66.1, 63.1, 61.3, 55.2, 38.7, 30.3, 28.9, 27.5, 26.2, 25.2, 23.7, 22.9;
HRMS (FAB, positive) calcd for C43H48N5O9: 778.3452 (MHþ),
found: 778.3458; UV (MeOH) lmax¼260, 295 (sh) nm.
orless oil. 1H NMR (CDCl3, 500 MHz)
d 7.38–7.70 (m, 10H), 5.87 (m,
1H), 5.06 (d, 1H, J¼5.6 Hz), 4.61 (d, 1H, J¼5.6 Hz), 4.38 (m, 3H), 1.33,
1.32 (each s, each 3H), 1.10 (s, 9H); 13C NMR (CDCl3, 125 MHz)
d
150.9,135.5,134.8,133.2,133.2,129.8,127.8,122.2,112.1, 84.3, 83.3,
69.8, 61.1, 27.3, 26.8, 26.0, 19.3; HRMS (FAB, positive) calcd for
25H32N3O3Si: 450.2213 (MHþ), found: 450.2224.
C
4.1.5. N-1-{(1R,2S,3R)-4-Dimethoxytrityloxymethyl-2,3-O-
isopropylidenedioxycyclopent-4-en-1-yl}-50-O-{bis(phenylthio)-
phosphoryl}-20,30-O-isopropylideneadenosine (18a)
4.1.2. (1R,2S,3R)-1-Amino-4-hydroxymethyl-2,3-O-
isopropylidenedioxy-4-cyclopentene (11)
After stirring a mixture of PSS (883 mg, 2.3 mmol) and TPSCl
(631 mg, 2.1 mmol) in pyridine (8 mL) at room temperature for 1 h,
17a (600 mg, 0.771 mmol) was added, and the resulting mixture
was stirred at room temperature for further 2 h and then evapo-
rated. The residue was partitioned between EtOAc and H2O, and the
A mixture of 16 (1.49 g, 3.31 mmol) and TBAF (1.0 M in THF,
5.00 mL, 5.00 mmol) in THF (20 mL) was stirred at room temper-
ature for 3 h, and then evaporated. The residue was purified by
column chromatography (SiO2, 25% EtOAc in hexane) to give the

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9763
organic layer was washed with brine, dried (Na2SO4), and evapo-
rated. The residue was purified by column chromatography (SiO2,
1% MeOH in CHCl3) to give 18a (591 mg, 74%) as a foam. 1H NMR
6.34 (d, 1H, J¼2.0 Hz), 6.08 (m, 1H), 5.62 (m, 1H), 5.48 (d, 1H,
J¼5.8 Hz), 5.43 (dd, 1H, J¼2.0, 5.9 Hz), 4.99 (dd, 1H, J¼5.9, 1.3 Hz),
4.72 (m, 1H), 4.67 (m, 3H), 4.20 (m, 1H), 4.13 (m, 1H), 3.19 (q, 6H,
J¼7.3 Hz), 1.63, 1.55, 1.43, 1.40 (each s, each 3H), 1.27 (t, 9H,
(CDCl3, 500 MHz)
d 7.68 (s, 1H), 7.56 (s, 1H), 6.82–7.46 (m, 23H),
5.99 (m, 2H), 5.95 (m, 1H), 5.19 (dd, 1H, J¼2.1, 6.2 Hz), 4.97 (d, 1H,
J¼7.3 Hz); 13C NMR (D2O, 125 MHz)
d 151.5, 151.4, 150.9, 146.9,
0
0
J¼5.3 Hz), 4.94 (dd, 1H, J3 ,2 ¼6.2, 2.7 Hz), 4.43 (m, 2H), 4.41 (d, 1H,
J¼5.3 Hz), 4.37 (m, 1H), 3.99 (d, 1H, J¼14.5 Hz), 3.79 (s, 6H), 3.76 (d,
1H, J¼14.5 Hz), 1.60, 1.40, 1.37, 1.27 (each s, each 3H); 13C NMR
144.6, 144.2, 133.0, 133.0, 129.9, 129.5, 128.3, 122.3, 119.6, 115.1,
114.9, 92.0, 86.7, 86.6, 84.5, 84.4, 83.8, 81.9, 70.6, 66.3, 62.0, 47.1,
26.6, 26.4, 25.3, 24.7, 8.7; 31P NMR (D2O, 202 MHz)
d
16.7 (s), 1.7 (s);
(CDCl3, 125 MHz)
d
158.6, 154.3, 151.1, 144.7, 144.6, 140.7, 137.4,
HRMS (FAB, negative) calcd for C28H34N5O12P2S: 726.1405
136.0, 135.8, 135.3, 135.2, 135.1, 135.1, 129.9, 129.7, 129.4, 128.6,
128.0, 127.9, 126.9, 125.8, 124.2, 122.0, 114.6, 113.2, 112.4, 90.9, 86.7,
84.7, 84.5, 83.7, 81.1, 65.7, 61.3, 55.2, 27.5, 27.1, 26.1, 25.3; 31P NMR
[(MꢁH)ꢁ], found: 726.1392; UV (H2O) lmax 261 nm.
4.1.8. 400,600-Didehydro-cyclic ADP-carbocyclic-ribose
diacetonide (20a)
(CDCl3, 202 MHz)
C
d 50.7 (s); HRMS (FAB, positive) calcd for
55H57N5O10PS2: 1042.3284 (MHþ), found: 1042.3293; UV (MeOH)
To a mixture of AgNO3 (127 mg, 0.75 mmol), Et3N (105 mL,
lmax¼260, 295 (sh) nm.
0.75 mmol), and MS 3 Å (200 mg) in pyridine (35 mL), a solution of
9a (30 mg, 36 mol) in pyridine (35 mL) was added slowly over
m
4.1.6. N-1-{(1R,2S,3R)-4-Hydroxymethyl-2,3-isopropylidene-
dioxycyclopent-4-en-1-yl}-50-O-{bis(phenylthio)-
15 h, using a syringe-pump, at room temperature under shading.
The MS 3 Å was filtered off with Celite and washed with H2O. To the
combined filtrate and washings was added TEAA buffer (2.0 M, pH
7.0, 2 mL), and the resulting solution was evaporated. The residue
was partitioned between EtOAc and H2O, and the aqueous layer
was evaporated. A solution of the residue in H2O (5 mL) was applied
to a C18 reverse phase column (1.1ꢃ17 cm), and the column was
developed using a linear gradient of 0–30% MeCN in TEAA buffer
(0.1 M, pH 7.0, 400 mL). Appropriate fractions were evaporated and
excess TEAA was removed by C18 reverse phase column chroma-
tography (1.1ꢃ17 cm, eluted with 50% aqueous MeCN). Appropriate
fractions were evaporated and the residue was lyophilized to give
20a (17 mg, 65%) as a triethylammonium salt. 1H NMR (D2O,
phosphoryl}-20,30-O-isopropylideneadenosine (19a)
A solution of 18a (551 mg, 0.529 mmol) in aqueous 60% AcOH
(6 mL) was stirred at room temperature for 3 h and then evapo-
rated. The residue was partitioned between EtOAc and aqueous
saturated NaHCO3, and the organic layer was washed with H2O and
brine, dried (Na2SO4), and evaporated. The residue was purified by
column chromatography (SiO2, 10% MeOH in EtOAc) to give 19a
(314 mg, 80%) as an oil. 1H NMR (CDCl3, 500 MHz)
d 7.69 (s,1H), 7.63
(s, 1H), 7.28–7.49 (m, 10H), 5.99 (d, 1H, J¼2.0 Hz), 5.87 (m, 1H), 5.72
(m, 1H), 5.17 (dd, 1H, J¼2.0, 6.2 Hz), 5.12 (d, 1H, J¼5.7 Hz), 4.94 (dd,
1H, J¼6.2, 1.7 Hz), 4.46 (d, 1H, J¼5.7 Hz), 4.37–4.44 (m, 4H), 4.33 (d,
1H, J¼15.3 Hz), 1.59, 1.45, 1.35, 1.30 (each s, each 3H); 13C NMR
500 MHz) d 8.71 (s, 1H), 8.43 (s, 1H), 6.39 (m, 1H), 6.31 (m, 1H), 5.73
(CDCl3, 125 MHz)
d
154.1, 152.4, 144.6, 140.8, 137.7, 135.3, 135.2,
(m, 2H), 5.60 (d, 1H, J¼5.2 Hz), 5.54 (m, 1H), 4.96 (m, 1H), 4.81 (d,
1H, J¼5.2 Hz), 4.58 (m, 1H), 4.52 (m, 1H), 4.06 (m, 1H), 3.86 (m, 1H),
3.21 (q, 6H, J¼7.3 Hz), 1.65, 1.49, 1.46, 1.42 (each s, each 3H), 1.29 (t,
135.1, 135.0, 129.7, 129.6, 129.4, 125.7, 125.7, 125.6, 125.6, 124.0,
122.4, 114.5, 112.5, 90.9, 85.0, 84.9, 84.5, 84.4, 83.4, 81.1, 66.4, 65.9,
59.7, 27.3, 27.1, 25.8, 25.2; 31P NMR (CDCl3, 202 MHz)
d
50.7 (s);
9H, J¼7.3 Hz); 13C NMR (D2O, 125 MHz)
d 151.3, 150.9, 147.4, 145.3,
144.5, 126.8, 119.6, 115.2, 114.1, 91.7, 86.7, 84.7, 82.9, 82.5, 81.1, 68.7,
HRMS (FAB, positive) calcd for C34H39N5O8PS2: 740.1978 (MHþ),
found: 740.1987; UV (MeOH) lmax¼259, 295 (sh) nm.
64.2, 62.2, 47.1, 26.8, 26.4, 25.8, 24.7, 8.7; 31P NMR (D2O, 202 MHz)
d
ꢁ10.5 (d, J¼15.3 Hz), ꢁ10.6 (d, J¼15.3 Hz); HRMS (FAB, negative)
4.1.7. N-1-{(1R,2S,3R)-2,3-Isopropylidenedioxy-4-
(phosphonoxymethyl)cyclopent-4-en-1-yl}-50-O-
(phenylthiophosphoryl)-20,30-O-isopropylideneadenosine (9a)
calcd for C22H28N5O12P2: 616.1215 [(MꢁH)ꢁ], found: 616.1188; UV
(H2O) lmax 260 nm.
A mixture of POCl3 (93
mL, 1.0 mmol) and 19a (72 mg, 98
m
mol)
4.1.9. 400,600-Didehydro-cyclic ADP-carbocyclic-ribose (8a)
in PO(OEt)3 (1.0 mL) was stirred at 0 ꢀC for 5 h. After addition of
aqueous saturated NaHCO3 (3 mL), the resulting mixture was stir-
red at 0 ꢀC for 10 min. To the mixture were added triethylammo-
nium acetate (TEAA) buffer (2.0 M, pH 7.0, 1 mL) and H2O (5 mL),
and the resulting solution was applied to a C18 reversed phase
column (1.1ꢃ17 cm). The column was developed using a linear
gradient of 0–66% MeCN in TEAA buffer (0.1 M, pH 7.0, 400 mL).
Appropriate fractions were evaporated and excess TEAA was re-
moved by C18 reversed phase column chromatography (1.1ꢃ17 cm,
eluted with 80% aqueous MeCN). Appropriate fractions were
evaporated and the residue was co-evaporated with pyridine
(2.0 mLꢃ3). A mixture of the residue, N-methylmaleimide (43 mg,
A solution of 20a (9.0 mg, 13 mmol) in aqueous 60% HCO2H
(1 mL) was stirred at room temperature for 4 h and then evapo-
rated. After the residue had been co-evaporated with H2O
(2 mLꢃ3), aqueous 28% NH3 (1 mL) was added. The mixture was
stirred at room temperature for 4 h and then evaporated. After the
residue had been co-evaporated with H2O (2 mLꢃ3), the resulting
residue was dissolved in TEAB buffer (0.1 M, pH 7.0, 100 mL), and the
solution was lyophilized to give 8a (6.0 mg, 86%) as a triethy-
lammonium salt. 1H NMR (D2O, 500 MHz, Kþ salt)
d
8.63 (s, 1H),
8.49 (s, 1H), 6.31 (m, 1H), 6.07 (d, 1H, J¼5.7 Hz), 5.44 (m, 1H), 5.29
(m, 1H), 4.91 (m, 2H), 4.67 (m, 1H), 4.62 (m, 2H), 4.34 (m, 2H), 4.07
(m, 1H); 13C NMR (D2O, 125 MHz)
d
153.1, 151.6, 147.4, 146.0, 143.3,
124.4,121.0, 91.3, 85.0, 74.1, 73.6, 73.2, 71.0, 69.0, 64.7, 61.7; 31P NMR
(D2O, 202 MHz)
0.39 mmol), H3PO2 (40 mL, 0.78 mmol), and Et3N (54 mL, 0.39 mmol)
was stirred at 0 ꢀC for 5 h under shading. After addition of TEAA
buffer (1.0 M, pH 7.0, 2 mL), the resulting mixture was evaporated.
The residue was partitioned between EtOAc and H2O, and the
aqueous layer was evaporated. A solution of the residue in H2O
(5 mL) was applied to a C18 reversed phase column (1.1ꢃ17 cm), and
the column was developed using a linear gradient of 0–35% MeCN
in TEAA buffer (0.1 M, pH 7.0, 400 mL). Appropriate fractions were
evaporated and excess TEAA was removed by C18 reversed phase
column chromatography (1.1ꢃ17 cm, eluted with 60% aqueous
MeCN). Appropriate fractions were evaporated and the residue was
lyophilized to give 9a (23 mg, 28%) as a triethylammonium salt. 1H
d
ꢁ10.0 (d, J¼15.3 Hz), ꢁ10.6 (d, J¼15.3 Hz); NOE
correlations H-2–H-100, H-2–H-200, H-2–H-600, H-2–H-300, H-8–H-10,
H-100–H-600, H-100–H-500a, H-10–H-20, H-10–H-40, H-20–H-30, H-20–H-
40, H-600–H-300, H-50a–H-40, H-50a–H-50b, H-40–H-50b, H-500a–H-500b,
H-200–H-300; HRMS (FAB, negative) calcd for C16H20N5O12P2:
536.0589 [(MꢁH)ꢁ], found: 536.0577; UV (H2O) lmax 260 nm.
4.1.10. N-1-{(1R,2S,3R)-4-Hydroxymethyl-2,3-
isopropylidenedioxycyclopent-4-en-1-yl}-50-O-(tert-
butyldimethylsilyl)-20,30-O-isopropylideneinosine (10b)
A
mixture of 14 (511 mg, 0.868 mmol) and 11 (241 mg,
NMR (D2O, 500 MHz)
d
8.42 (s, 1H), 8.41 (s, 1H), 7.17–7.29 (m, 5H),
1.30 mmol) in DMF (400
m
L) was stirred at 50 ꢀC for 45 h. The

9764
T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
reaction mixture was diluted with EtOAc and washed with H2O. The
water layer was evaporated and co-evaporated with toluene
(3 mLꢃ3) to recover 11. The organic layer was washed with brine,
dried (Na2SO4), and evaporated. A mixture of the residue and K2CO3
(240 mg, 1.74 mmol) in DMF (9 mL) was stirred at 50 ꢀC for 30 min.
The reaction mixture was diluted with EtOAc and washed with H2O
and brine, dried (Na2SO4), and evaporated. The residue was purified
by column chromatography (SiO2, 75% EtOAc in hexane) to give 10b
125.6, 125.3, 122.2, 114.7, 112.4, 90.8, 84.8, 84.7, 84.5, 84.4, 84.1, 81.0,
66.8, 66.2, 60.3, 59.9, 27.3, 27.1, 25.7, 25.3; 31P NMR (CDCl3,
202 MHz) d 50.8 (s); HRMS (FAB, positive) calcd for C34H38N4O19S2:
741.1818 (MHþ), found: 741.1831.
4.1.14. N-1-{(1R,2S,3R)-2,3-Isopropylidenedioxy-4-
(phosphonoxymethyl)cyclopent-4-en-1-yl}-50-O-
{(phenylthio)phospholyl}-20,30-O-isopropylideneinosine (9b)
Compound 9b (oil, 26 mg, 36%) was obtained from 19b (84 mg,
0.11 mmol) as described for the synthesis of 9a. 1H NMR (D2O,
(179 mg, 35%) as a brown foam. 1H NMR (CDCl3, 500 MHz)
d 8.01 (s,
1H), 7.87 (s, 1H), 6.09 (d, 1H, J¼2.6 Hz), 5.64 (m, 2H), 5.33 (d, 1H,
J¼5.7 Hz), 5.03 (dd, 1H, J¼2.6, 6.0 Hz), 4.88 (dd, 1H, J¼6.0, 2.3 Hz),
4.69 (d, 1H, J¼5.7 Hz), 4.51 (m, 2H), 4.41 (m, 3H), 1.61, 1.45, 1.38, 1.33
(each s, each 3H), 0.85 (s, 9H), 0.062, 0.036 (each s, each 3H); 13C
500 MHz) d 8.22 (s, 1H), 8.14 (s, 1H), 7.08–7.22 (m, 5H), 6.30 (m, 1H),
5.91 (m, 1H), 5.59 (m, 1H), 5.49 (d, 1H, J¼5.9 Hz), 5.36 (d, 1H,
J¼5.7 Hz), 5.10 (d, 1H, J¼5.9 Hz), 4.67 (m, 1H), 4.64 (m, 2H), 4.38 (d,
1H, J¼5.7 Hz), 4.29 (m,1H), 4.10 (m,1H), 3.18 (q,12H, J¼7.3 Hz),1.64,
1.51, 1.42, 1.39 (each s, each 3H), 1.26 (t, 18H, J¼7.3 Hz); 13C NMR
NMR (CDCl3, 125 MHz)
d 156.3, 151.9, 146.8, 145.4, 138.5, 123.7,
122.9, 114.2, 112.5, 91.4, 87.0, 85.5, 84.4, 84.3, 81.3, 63.5, 62.8, 59.9,
27.3, 27.2, 25.9, 25.3,18.3, ꢁ5.5, ꢁ5.6; HRMS (FAB, positive) calcd for
(D2O, 125 MHz) d 157.9, 149.2, 149.1, 147.7, 146.6, 141.4, 132.8, 132.8,
C
28H43N4O8Si: 591.2850 (MHþ), found: 591.2862; UV (MeOH) lmax
129.9, 129.4, 128.2, 124.2, 115.2, 113.6, 91.2, 86.5, 86.4, 84.9, 84.2,
245 nm.
83.8, 81.8, 67.0, 66.3, 62.2, 47.1, 26.6, 26.4, 25.3, 24.7, 8.7; 31P NMR
(D2O, 202 MHz)
C
lmax 244 nm.
d 17.4 (s), 1.0 (s); HRMS (FAB, negative) calcd for
4.1.11. N-1-{(1R,2S,3R)-4-Dimethoxytrityloxymethyl-2,3-
isopropylidenedioxycyclopent-4-en-1-yl}-20,30-O-
isopropylideneinosine (17b)
28H33N4O13P2S: 727.1246 [(MꢁH)ꢁ], found: 727.1230; UV (H2O)
Compound 17b (oil, 158 mg, 61%) was obtained from 10b
4.1.15. 400,600-Didehydro-cyclic IDP-carbocyclic-ribose
diacetonide (20b)
(195 mg, 0.330 mmol) as described for the synthesis of 17a. 1H NMR
(CDCl3, 500 MHz)
d
8.00 (s, 1H), 7.86 (s, 1H), 6.83–7.48 (m, 13H),
Compound 20b (13 mg, 57%) was obtained from 9b (amorphous
5.93 (d, 1H, J¼4.1 Hz), 5.90 (m, 1H), 5.83 (m, 1H), 5.17 (d, 1H,
J¼5.6 Hz), 5.09 (m, 1H), 5.07 (m, 1H), 4.61 (d, 1H, J¼5.6 Hz), 4.50 (m,
1H), 4.00 (m, 1H), 3.96 (m, 1H), 3.82 (m, 1H), 3.78 (s, 6H), 3.75 (m,
1H), 1.64, 1.39, 1.36, 1.27 (each s, each 3H); 13C NMR (CDCl3,
solid, 26 mg, 28
(D2O, 500 MHz)
m
d
mol) as described for the synthesis of 20a. 1H NMR
8.45 (s,1H), 8.19 (s,1H), 6.30 (m,1H), 6.22 (m,1H),
5.72 (d, 1H, J¼6.1 Hz), 5.61 (m, 1H), 5.55 (dd, 1H, J¼6.1, 3.0 Hz), 5.48
(d, 1H, J¼5.2 Hz), 4.93 (m, 1H), 4.60 (d, 1H, J¼5.2 Hz), 4.53 (m, 1H),
4.46 (m, 1H), 3.98 (m, 1H), 3.81 (m, 1H), 3.19 (q, 12H, J¼7.3 Hz), 1.62,
1.46, 1.45, 1.38 (each s, each 3H), 1.27 (t, 18H, J¼7.3 Hz); 13C NMR
125 MHz) d 171.0, 158.5, 156.0, 151.4, 146.2, 145.1, 144.6, 139.7, 135.9,
135.7, 132.8, 132.0, 132.0, 131.9, 131.9, 129.9, 129.8, 128.5, 128.4,
127.9, 127.8, 126.8, 125.7, 121.4, 114.2, 113.2, 112.5, 93.2, 86.6, 86.2,
84.6, 84.1, 81.5, 66.3, 62.9, 61.2, 60.3, 55.1, 27.4, 26.0, 25.2, 21.0, 14.1;
HRMS (FAB, positive) calcd for C43H47N4O10: 779.3292 (MHþ),
found: 779.3284; UV (MeOH) lmax 272, 265 nm.
(D2O, 125 MHz) d 158.8, 149.5, 148.1, 146.1, 142.6, 128.4, 124.6, 115.0,
113.3, 91.5, 86.7, 84.7, 83.8, 82.4, 81.5, 65.3, 64.3, 62.4, 47.1, 26.8,
26.4, 25.7, 24.7, 8.7; 31P NMR (D2O, 202 MHz)
d
ꢁ10.5 (d, J¼15.3 Hz),
ꢁ10.7 (d, J¼15.3 Hz); HRMS (FAB, negative) calcd for
C
22H27N4O13P2: 617.1055 [(MꢁH)ꢁ], found: 617.1049; UV (H2O) lmax
4.1.12. N-1-{(1R,2S,3R)-4-Dimethoxytrityloxymethyl-2,3-
isopropylidenedioxycyclopent-4-en-1-yl}-50-O-{bis(phenylthio)-
phosphoryl}-20,30-O-isopropylideneinosine (18b)
251 nm.
4.1.16. 400,600-Didehydro-cyclic IDP-carbocyclic-ribose (8b)
Compound 18b (oil, 149 mg, 71%) was obtained from 17b
Compound 8b (amorphous solid,10 mg, 82%) was obtained from
20b (13 mg, 16
(D2O, 500 MHz) d 8.35 (s, 1H), 8.19 (s, 1H), 6.22 (m, 1H), 5.99 (d, 1H,
(158 mg, 0.203 mmol) as described for the synthesis of 18a. 1H NMR
m
mol) as described for the synthesis of 8a. 1H NMR
(CDCl3, 500 MHz)
d
7.84 (s, 1H), 7.79 (s, 1H), 6.82–7.68 (m, 23H),
6.06 (d, 1H, J¼2.5 Hz), 5.88 (m, 1H), 5.82 (m, 1H), 5.15 (dd, 1H, J¼2.5,
6.3 Hz), 5.10 (d, 1H, J¼5.6 Hz), 4.96 (dd, 1H, J¼6.3, 3.1 Hz), 4.52 (d,
1H, J¼5.6 Hz), 4.46 (m, 1H), 4.39 (m, 2H), 3.99 (d, 1H, J¼15.5 Hz),
3.78 (s, 6H), 3.77 (d, 1H, J¼15.5 Hz), 1.62, 1.40, 1.37, 1.30 (each s, each
J¼6.2 Hz), 5.41 (m, 1H), 5.32 (dd, 1H, J¼6.2, 4.8 Hz), 4.88 (m, 1H),
4.81 (d, 1H, J¼4.0 Hz), 4.59–4.71 (m, 3H), 4.32 (m, 1H), 4.14 (d, 1H,
J¼4.0 Hz), 3.99 (m, 1H), 3.19 (m, 6H), 1.27 (m, 9H); 13C NMR (D2O,
125 MHz)
d 159.1, 151.5, 148.5, 144.9, 143.4, 126.3, 125.1, 91.1, 85.1,
3H); 13C NMR (CDCl3, 125 MHz)
d
171.1, 158.6, 158.5, 156.1, 150.9,
75.4, 73.3, 73.1, 71.2, 66.0, 64.9, 61.9, 47.1, 8.7; 31P NMR (D2O,
146.6, 145.4, 144.7, 139.0, 136.0, 135.8, 135.3, 135.2, 135.1, 135.0,
132.1,132.1,132.0,131.9,131.9,129.9,129.9,129.7,129.5,129.4,128.5,
128.4, 128.0, 127.9, 126.9, 125.7, 125.3, 121.9, 114.8, 113.2, 112.3, 90.8,
86.7, 84.8, 84.6, 84.4, 84.1, 80.9, 66.0, 61.2, 60.3, 55.2, 27.4, 27.1, 26.0,
202 MHz)
d
ꢁ9.9 (d, J¼15.3 Hz), ꢁ10.5 (d, J¼15.3 Hz); HRMS (FAB,
negative) calcd for C16H19N4O13P2: 537.0429 [(MꢁH)ꢁ], found:
537.0403; UV (H2O) lmax 251 nm.
25.3, 21.0, 14.2; 31P NMR (CDCl3, 202 MHz)
d
50.7 (s); HRMS (FAB,
4.2. Biological evaluations with sea urchin egg homogenate,
neuronal cells, and T cells
positive) calcd for C55H56N4O11PS2: 1043.3125 (MHþ), found:
1043.3132.
These bioassays were carried out as reported previously.7c
4.1.13. N-1-{(1R,2S,3R)-4-Hydroxymethyl-2,3-isopropylidene-
dioxycyclopent-4-en-1-yl}-50-O-{bis(phenylthio)-
4.2.1. NMR titrations
phospholyl}-20,30-O-isopropylideneinosine (19b)
NMR titrations were carried out as previously reported.29 The
experiments were performed in two steps in which 0.50 cm3 of the
same initial solution of studied compounds of 3.0ꢃ10ꢁ3 mol dmꢁ3
in 2H2O was successively subjected to potentiometric and NMR
titrations. It should be noted that the glass electrode was calibrated
in a concentration scale and the measurements done in 2H2O, so
that here pH means the cologarithm of the concentration of 2Hþ.
The processing of the pH measurements allowed the total con-
centration of the ligand and the acid as well as the macroscopic
Compound 19b (oil, 84 mg, 79%) was obtained from 18b
(149 mg, 0.143 mmol) as described for the synthesis of 19a. 1H NMR
(CDCl3, 500 MHz) d 7.86 (s, 1H), 7.83 (s, 1H), 7.27–7.67 (m, 10H), 6.05
(d, 1H, J¼2.5 Hz), 5.65 (m, 1H), 5.62 (m, 1H), 5.26 (d, 1H, J¼5.7 Hz),
5.13 (dd, 1H, J¼2.5, 6.3 Hz), 4.95 (dd, 1H, J¼6.3, 3.1 Hz), 4.59 (d, 1H,
J¼5.7 Hz), 4.45 (m, 1H), 4.40 (m, 4H), 1.61, 1.43, 1.36, 1.30 (each s,
each 3H); 13C NMR (CDCl3, 125 MHz)
d 156.2, 152.2, 146.7, 145.6,
139.2, 135.3, 135.3, 135.1, 135.1, 129.8, 129.5, 128.5, 128.4, 125.7,

T. Kudoh et al. / Tetrahedron 64 (2008) 9754–9765
9765
protonation constants to be determined. The HypNMR software25
was used to check protonation constants. The 1H NMR titration was
performed on a Bruker DPX-500 Fourier transform spectrometer
operating at 500.13 MHz. Spectra were acquired with water pre-
saturation over a spectral width of 10 ppm using a 3 s relaxation
Chem. 2005, 70, 4810–4819; (h) Bailey, V. C.; Fortt, S. M.; Summerhill, R. J.;
Galione, A.; Potter, B. V. L. FEBS Lett. 1996, 379, 227–230.
6. (a) Shuto, S.; Shirato, M.; Sumita, Y.; Ueno, Y.; Matsuda, A. J. Org. Chem. 1998, 63,
1986–1994; (b) Shuto, S.; Shirato, M.; Sumita, Y.; Ueno, Y.; Matsuda, A. Tetra-
hedron Lett. 1998, 39, 7341–7344; (c) Fukuoka, M.; Shuto, S.; Minakawa, N.;
Ueno, Y.; Matsuda, A. J. Org. Chem. 2000, 65, 5238–5248; (d) Shuto, S.; Fukuoka,
M.; Manikowsky, M.; Ueno, Y.; Nakano, T.; Kuroda, R.; Kuroda, H.; Matsuda, A.
J. Am. Chem. Soc. 2001, 123, 8750–8759; (e) Guse, A. H.; Cakir-Kiefer, C.;
Fukuoka, M.; Shuto, S.; Weber, K.; Matsuda, A.; Mayer, G. W.; Oppenheimer, N.;
Schuber, F.; Potter, B. V. L. Biochemistry 2002, 41, 6744–6751.
7. (a) Shuto, S.; Fukuoka, M.; Kudoh, T.; Garnham, C.; Galione, A.; Potter, B. V. L.;
Matsuda, A. J. Med. Chem. 2003, 46, 4741–4749; (b) Hashii, M.; Shuto, S.;
Fukuoka, M.; Kudoh, T.; Matsuda, A.; Higashida, H. J. Neurochem. 2005, 94, 316–
323; (c) Kudoh, T.; Fukuoka, M.; Ichikawa, S.; Murayama, T.; Ogawa, Y.; Hashii,
M.; Higashida, H.; Kunerth, S.; Weber, K.; Guse, A. H.; Potter, B. V. L.; Matsuda,
A.; Shuto, S. J. Am. Chem. Soc. 2005, 127, 8846–8855; (d) Kudoh, T.; Murayama,
T.; Ogawa, Y.; Matsuda, A.; Shuto, S. Nucleosides Nucleotides Nucleic Acids 2006,
25, 583–599; (e) Kudoh, T.; Murayama, T.; Matsuda, A.; Shuto, S. Bioorg. Med.
Chem. 2007, 15, 3032–3040.
8. For example: (a) Galeone, A.; Mayol, L.; Oliviero, G.; Piccialli, G.; Varra, M.
Tetrahedron 2002, 58, 363–368; (b) Huang, L.-J.; Zhao, Y.-Y.; Yuan, L.; Min, J.-M.;
Zhang, L.-H. J. Med. Chem. 2002, 45, 5340–5352; (c) Gu, X.; Yang, Z.; Zhang, L.;
Kunerth, S.; Fliegert, R.; Weber, K.; Guse, A. H.; Zhang, L. J. Med. Chem. 2004, 47,
5674–5682; (d) Xu, J.; Yang, Z.; Dammermann, W.; Zhang, L.; Guse, A. H.; Zhang,
L.-H. J. Med. Chem. 2006, 49, 5501–5512; (e) Moreau, C.; Wagner, G. K.; Weber,
K.; Guse, A. H.; Potter, B. V. L. J. Med. Chem. 2006, 49, 5162–5176.
delay and a
p/2 pulse. 16 K data points were sampled with a cor-
responding 1.14 s acquisition time. The spectra had a digital reso-
lution of 0.31 Hz per point. The temperature was controlled at
310.0ꢂ0.5 K. The proton resonances were assigned by performing
proton–proton (double quantum filtered gradient accelerated
COSY) and phosphorus–proton 2D correlation experiments at
acidic and basic pHs (2 and 11.5, respectively), thus allowing the
titration curves to be unambiguously characterized. 1H, 13C HSQC
were used for the full attribution of the two phosphorus nuclei.
Knowing non ambiguously 1H chemical shift, it was easy to sort 31
P
shift with HMQC 1H, 31P experiences using two different delay of
evolution, one corresponding to 40 Hz and the other 5 Hz coupling.
4.2.2. Calculations
Calculations were carried out using the molecular-modeling
software package SYBYL ver. 6.5 (Tripos, Inc.) on a Silicon Graphics
O2 computer as reported previously.7c
9. Lee, H. C.; Aarhus, R. Biochim. Biophys. Acta 1993, 1164, 68–74.
10. A preliminary account of this study: Kudoh, T.; Weber, K.; Guse, A. H.; Potter, B.
V. L.; Hashii, M.; Higashida, H.; Arisawa, M.; Matsuda, A.; Shuto, S. Tetrahedron
Lett. 2008, 49, 3976–3979.
11. Agrofoglio, L. A.; Challand, S. R. Acyclic, Carbocyclic and L-Nucleosides; Kluwer
Acknowledgements
Academic: Dordrecht, 1998.
12. Wolfe, M. S.; Borchardt, R. T. J. Med. Chem. 1991, 34, 1521–1530.
13. Kusaka, T.; Yamamoto, H.; Shibata, M.; Muroi, M.; Kishi, T. J. Antibiot. 1968, 21,
255–263.
14. (a) Yaginuma, S.; Muto, N.; Tsujino, M.; Sudate, Y.; Hayashi, M.; Otani, M.
J. Antibiot. 1981, 34, 359–366; (b) Shuto, S.; Minakawa, N.; Niizuma, S.; Kim,
S.-H.; Wataya, Y.; Matsuda, A. J. Med. Chem. 2002, 45, 748–751 and references
therein.
15. (a) Norstrom, A.; Bryman, I.; Lindblom, B.; Christensen, N. J. Prostaglandins
1985, 29, 337–346; (b) Andreotti, D.; Biondi, S.; Modugno, E. D. In Burger’s
Medicinal Chemistry and Drug Discovery, 6th ed.; Abraham, D. J., Ed.; John Wiley
& Sons: New York, NY, 2003; Vol. 5, pp 607–735.
16. Saenger, W. Principles of Nucleic Acid Structure; Springer: New York, NY, 1983.
17. (a) Hutchinson, E. J.; Taylor, B. F.; Blackburn, G. M. Chem. Commun. 1997, 1859–
1860; (b) De Napoli, L.; Messere, A.; Montesarchio, D.; Piccialli, G.; Varra, M. J.
J. Chem. Soc., Perkin Trans. 1 1997, 2079–2082.
We wish to acknowledge financial support from Ministry of
Education, Culture, Sports, Science, and Technology, Japan for Sci-
entific Research 17659027 and 17390027. We also acknowledge the
Wellcome Trust for a Biomedical Research Collaboration Grant
(068065 to B.V.L.P. and A.G.) and the Deutsche For-
schungsgemeinschaft for grant nos. GU 360/9-1 and 9-2 (to A.H.G.).
We are grateful to C. Antheaume and L. Allouche from Services de
´
´
RMN, Facultes de Pharmacie et de Chimie, Universite Louis Pasteur,
Strasbourg, France for technical assistance.
Supplementary data
18. Choi, W. J.; Moon, H. R.; Kim, H. O.; Too, B. N.; Lee, J. A.; Shin, D. H.; Jeong, L. S.
J. Org. Chem. 2004, 69, 2634–2636.
19. A similar transformation of a chiral cyclopentenone to the corresponding cy-
clopentenylamine has been previously reported: Marquez, V. E.; Lim, M.-I.;
Tseng, C. K.-H.; Markovac, A.; Priest, M. A.; Khan, M. S.; Kaskar, B. J. Org. Chem.
1988, 53, 5709–5714.
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.tet.2008.07.068.
20. Sekine, M.; Nishiyama, S.; Kamimura, T.; Osaki, Y.; Hata, T. Bull. Chem. Soc. Jpn.
1985, 58, 850–860; (b) Sekine, M.; Hata, T. Curr. Org. Chem. 1993, 3, 25–66.
21. Yoshikawa, M.; Kato, T.; Takenishi, T. Bull. Chem. Soc. Jpn. 1969, 42, 3505–3508.
22. Hata, T.; Kamimura, T.; Urakami, K.; Kohno, K.; Sekine, M.; Kumagai, I.; Shi-
nozaki, K.; Miura, K. Chem. Lett. 1987, 117–120.
References and notes
1. Clapper, D. L.; Walseth, T. F.; Dargie, P. J.; Lee, H. C. J. Biol. Chem. 1987, 262, 9561–
9568.
23. Intermolecular condensation between
a S-phenyl phosphorothioate and
2. (a) Galione, A. Science 1993, 259, 325–326; (b) Guse, A. H. Cell. Signalling 1999,
11, 309–316; (c) Cyclic ADP-ribose and NAADP: Structures, Metabolism and
Functions; Lee, H. C., Ed.; Kluwer Academic: Dordrecht, 2002.
3. Jin, D.; Liu, H.-X.; Hirai, H.; Torashima, T.; Nagao, T.; Lopatina, O.; Shnayder, N.
A.; Yamada, K.; Noda, M.; Seika, T.; Fujita, K.; Takasawa, S.; Yokoyama, S.;
Koizumi, K.; Shiraishi, Y.; Tanaka, S.; Hashii, M.; Yoshihara, T.; Higashida, K.;
Islam, M. S.; Yamada, N.; Hayashi, K.; Noguchi, N.; Kato, I.; Okamoto, H.; Mat-
sushima, A.; Salmina, A.; Munesue, T.; Shimizu, N.; Mochida, S.; Asana, M.;
Higashida, H. Nature 2007, 446, 41–45.
4. (a) Zhang, F.-J.; Gu, Q.-M.; Sih, C. J. Bioorg. Med. Chem. 1999, 7, 653–664; (b)
Shuto, S.; Matsuda, A. Curr. Med. Chem. 2004, 11, 827–845; (c) Guse, A. H. Curr.
Med. Chem. 2004, 11, 847–855; (d) Potter, B. V. L.; Walseth, T. F. Curr. Mol. Med.
2004, 4, 303–311.
a phosphomonoester promoted by I2 or AgNO3: see Ref. 20.
24. Shiwa, S.; Murayama, T.; Ogawa, Y. Am. J. Physiol. Regul. Integr. Comp. Physiol.
2002, 282, R727–R737.
25. Frassineti, C.; Ghelli, S.; Gans, P.; Sabatini, A.; Moruzzi, M. S.; Vacca, A. Anal.
Biochem. 1995, 231, 374–382.
26. Kim, H.; Jacobson, E. L.; Jacobson, M. K. Biochem. Biophys. Res. Commun. 1993,
194, 1143–1147.
27. (a) Martin, R. B. Acc. Chem. Res. 1985, 18, 32–38; (b) Blindauer, C. A.; Holy, A.;
Dvorakova, H.; Sigel, H. J. Chem. Soc., Perkin Trans. 2 1997, 2353–2363; (c)
Felemez, M.; Bernard, P.; Schlewer, G.; Spiess, B. J. Am. Chem. Soc. 2000, 122,
3156–3165; (d) Blum, H. C.; Bernard, P.; Spiess, B. J. Am. Chem. Soc. 2001, 123,
3399–3400.
28. (a) Rudiger, V.; Schneider, H. J. Chem.dEur. J. 2000, 6, 3771–3776; (b) Rudiger,
V.; Schneider, H. J. Magn. Reson. Chem. 2000, 38, 85–89.
29. Mernissi-Arifi, K.; Schmitt, L.; Schlewer, G.; Spiess, B. Anal. Chem. 1995, 67,
2567–2574.
30. (a) Togashi, K.; Hara, Y.; Tominaga, T.; Higashi, T.; Konishi, Y.; Mori, Y.; Tomi-
naga, M. EMBO J. 2006, 25, 1804–1815; (b) Kolisek, M.; Beck, A.; Fleig, A.;
Penner, R. Mol. Cell 2005, 18, 61–69.
5. For example: (a) Walseth, T. F.; Lee, H. C. Biochem. Biophys. Acta 1993, 1178, 235–
242; (b) Lee, H. C.; Aarhus, R.; Walseth, T. F. Science 1993, 261, 352–355; (c)
Zhang, F.-J.; Sih, C. J. Bioorg. Med. Chem. Lett. 1995, 5, 1701–1706; (d) Ashamu, G.
A.; Galione, A.; Potter, B. V. L. J. Chem. Soc., Chem. Commun. 1995, 1359–1360; (e)
Wagner, G. K.; Black, S.; Guse, A. H.; Potter, B. V. L. Chem. Commun. 2003, 1944–
1945; (f) Mort, C. J. W.; Migaud, M. E.; Galione, A.; Potter, B. V. L. Bioorg. Med.
Chem. 2004, 12, 475–488; (g) Wagner, G. K.; Guse, A. H.; Potter, B. V. L. J. Org.