Synthesis and agonist activity of cyclic ADP-ribose analogues with substitution of the northern ribose by ether or alkane chains

Article abstract of DOI:10.1021/jm060320e

Novel analogues of cADPR with adenine as base and ether (10a) or different alkane chain (10b-d) substitutions of the northern ribose were synthesized from protected imidazole nucleoside 1 in good yields. The pharmacological activities of cyclic inosine di

Full text of DOI:10.1021/jm060320e

J. Med. Chem. 2006, 49, 5501-5512
5501
Synthesis and Agonist Activity of Cyclic ADP-Ribose Analogues with Substitution of the
Northern Ribose by Ether or Alkane Chains
Jianfeng Xu,# Zhenjun Yang,# Werner Dammermann,† Liangren Zhang,*,# Andreas H. Guse,† and Li-he Zhang*,#
State Key Laboratory of Natural and Biomimetic Drugs, School of Pharmaceutical Sciences, Peking UniVersity, Beijing 100083, China, and
Center of Experimental Medicine, Institute of Biochemistry and Molecular Biology I: Cellular Signal Transduction, UniVersity Medical Center
Hamburg-Eppendorf, 20246 Hamburg, Germany
ReceiVed March 20, 2006
Novel analogues of cADPR with adenine as base and ether (10a) or different alkane chain (10b-d)
substitutions of the northern ribose were synthesized from protected imidazole nucleoside 1 in good yields.
The pharmacological activities of cyclic inosine diphosphoribose ether (cIDPRE) and the compounds (10a-
d) were analyzed in intact human Jurkat T-lymphocytes. The results indicate that the analogues 10a-d
permeate the plasma membrane and are weak agonists of the cADPR/ryanodine receptor signaling system
in intact human Jurkat T cells. They are the first membrane-permeant and biologically active cADPR analogues
that contain ether or alkane bridges instead of the northern ribose and retain adenine as its base.
Introduction
Cyclic ADP-ribose (cADPRa) is a universal Ca2+ mobilizing
second messenger first identified in the sea urchin egg system.1,2
Since its discovery, numerous cell systems have been described
to make use of the cADPR/ryanodine receptor (RyR)/Ca2+
signaling system to control Ca2+-dependent cellular responses
such as fertilization, secretion, contraction, proliferation, and
many more. 3,4
Because of the importance of the cADPR/RyR/Ca2+ signaling
system in cell regulation, analogues of cADPR have been
synthesized (Figure 1) and their biological activity has been
tested. The first and still very important analogues were
derivatized in the 8-position of adenine, e.g., 8-NH2-cADPR
or 8-Br-cADPR.5 These compounds turned out to be antagonists
of cADPR.5 In the years following 1993 many more cADPR
analogues with modifications in the (i) southern ribose, e.g.,
2′-phospho-6,7 or 3′-OCH3-,8 or the (ii) northern ribose, e.g.,
2′′-NH2-,9 or the (iii) pyrophosphate bridge, e.g., a triphosphate
bridge in cyclic adenosine triphosphoribose10 have been syn-
thesized and evaluated for their biological activity. Furthermore,
either the southern or the northern ribose was replaced by the
corresponding carbocyclic derivatives, termed cyclic aristero-
mycin diphosphoribose11 and cyclic adenosine diphosphocar-
* To whom correspondence should be addressed. For L.Z.: phone, 86-
10-82802567; fax, 86-10-82802724; e-mail, liangren@bjmu.edu.cn. For L.-
h.Z.:
phone, 86-10-82801700; fax, 86-10-82802724; e-mail,
zdszlh@bjmu.edu.cn.
# Peking University.
† University Medical Center Hamburg-Eppendorf.
a Abbreviations: cADPR, cyclic adenosine 5′-diphosphoribose; cIDPR,
cyclic inosine 5′-diphosphoribose; cADPcR, cyclic adenosine 5′-diphos-
phocarboribose; cIDPRE, cyclic inosine diphosphoribose ether; cIDP-DE,
cyclic inosine diphosphodiether; DMF, dimethylformamide; DMSO, di-
methyl sulfoxide; EGTA, ethylene glycol bis(2-aminoethyl ether)tetraacetic
acid; ESI, electrospray ionization; FAB, fast atom bombardment; 1H NMR,
proton nuclear magnetic resonance; 13C NMR, carbon-13 nuclear magnetic
resonance; 31P NMR, phosphorus-31 nuclear magnetic resonance; HMBC,
heteronuclear multiple bond correlation; HR-FAB-MS, high-resolution fast
atom bombardment mass spectrometry; HR-ESI-MS, high-resolution elec-
trospray ionization mass spectrometry; HPLC, high-performance liquid
chromatography; mAb, monoclonal antibody; MMTrCl, monomethoxytrityl
chloride; PSS, cyclohexylammonium S,S-diphenylphosphorodithioate; TOF,
time of flight; THF, tetrahydrofuran; Py, pyridine; TPSCl, triisopropylben-
zenesulfonyl chloride; TBDMSCl, tert-butyldimethylsilyl chloride; TEAA,
triethylammonium acetate; TEAB, triethylammonium bicarbonate; TBAF,
tetrabutylammonium fluoride; TMS, tetramethyl silicane.
Figure 1. Structures of cADPR analogues.
bocyclic ribose (cADPcR).12 Interestingly, in higher eukaryotic
cell types, cADPcR and its analogues were relatively weak in
their Ca2+ mobilizing activity,9,13 while the southern ribose
carbocyclic substitution was almost identical to cADPR.9 These
results indicated that the polar oxygen in the hemiacetal moiety
of the northern ribose might be required for receptor binding
of cADPR.
10.1021/jm060320e CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/10/2006

5502 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
both the northern and southern ribose were substituted by
corresponding ether strands.18
In the present study, novel analogues of cADPR with adenine
as base and ether or different alkane chain substitutions of the
northern ribose were synthesized and characterized; further, the
compounds were tested for their biological activity in Jurkat
T-lymphocytes.
For sake of clarity, in this paper the structural positions of
compounds 10a-d are numbered as follows: a single prime
numbering scheme is used for the position of the N9-ribosyl
moiety, and a double prime numbering scheme is used for the
position of the N1-substitution (Figure 2).
Figure 2. Structures of 10a-d
To initially test this hypothesis, a cADPR analogue with
substitution of the northern ribose with an ether strand termed
cyclic inosine diphosphoribose ether (cIDPRE, Figure 1) was
synthesized; its biological activity was proven in permeabilized
and intact higher eukaryotic cells.14 The biological activity of
cIDPRE and some 8-substituted derivatives was fully character-
ized in permeabilized and intact Jurkat T cells.15 Importantly,
substitution of the base adenine by hypoxanthine in N1-cyclic
Results and Discussion
Chemistry. The syntheses of compounds 10a-d are sum-
marized in Scheme 1. We reported previously the synthesis of
N1-ether substituted N1-cIDPR, cIDPRE.15 An N1 substitution
was carried out regioselectively on the protected inosine.
However, the same strategy cannot be used in the case of
adenosine. The N1 substitution on the adenine moiety is very
difficult, and an N6-substituted adenosine is the main product
instead. Blackburn’s group provided an efficient method to
construct the N1 substituted adenosine.19,20 By use of the same
inosine diphosphoribose (N1-cIDPR) did not alter the Ca2+
-
releasing properties compared to cADPR,16 though N7-cIDPR
did not release Ca2+ in sea urchin egg homogenates.17 A further
simplified cADPR analogue that still retained biological activity
is cyclic inosine diphosphodiether (cIDP-DE, Figure 1) in which
Scheme 1. Syntheses of Compounds 10a-da
a Reagents and condictions: (a) K2CO3, MeOH, room temp; (b) MMTrCl, Py, room temp; (c) TBAF, THF, room temp; (d) PSS, TPSCl, tetrazole, room
temp; (e) 80% AcOH, room temp; (f) POCl3, PO(OMe)3, 0 °C; (g) H3PO2, Et3N, Py, room temp; (h) AgNO3, 3 Å molecular sieves, Py, room temp; (i) 60%
HCOOH, room temp.

Cyclic ADP-Ribose Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18 5503
Figure 3. Lack of antagonist effect of compounds 10a-d in intact Jurkat T cells. Jurkat T cells were loaded with Fura 2/AM and analyzed by
ratiometric fluorometry on a Hitachi F-2000 flurometer. The cells were preincubated with each compound (500 mM final concentration) for 20
min. OKT3 was added 200 s after the beginning of each measurement, and each experiment was calibrated by the addition of Triton X-100 to
obtain a maximal ratio at 900 s and by subsequent addition of EGTA/Tris-base to obtain the minimal ratio.
starting material, 5-[(methoxymethylene)amino]-1-[5-O-(tert-
butyldimethylsilyl)-2,3-O-(isopropylidene)-â-D-ribofuranosyl]-
imidazole-4-nitrile 1, the formation of N1-ether or alkane chain
substituted adenosines (3a-d) was achieved regiospecifically
by condensation reaction with different amino alcohols (2a-
d) in the presence of a catalytic amount of K2CO3 in high yield.
The resulting hydroxyl groups were phosphorylated with
cyclohexylammonium S,S-diphenylphosphorodithioate (PSS) in
the presence of triisopropylbenzenesulfonyl choride (TPSCl) and
tetrazole in pyridine to give 6a-d in 65% yield. The presence
of the bis(phenylthio)phosphoryl group in the molecules was
supported by 31P NMR (49.39 ppm, s). Removal of the MMTr
group of 6a-d with 80% aqueous AcOH provided 7a-d,
respectively. The resulting terminal hydroxyl groups in 7a-d
were phosphorylated with the POCl3/PO(OMe)3 system.21 The
phosphorylated products were treated with H3PO2 and Et3N22
in pyridine to afford directly 5′-phenylthiophosphates 8a-d,
the substrates for the intramolecular cyclization reaction. The
presence of two different phosphate groups in 8a-d was
supported by 31P NMR (see Experiment Section). The intramo-
lecular cyclization of compounds 8a-d was achieved as
described in the synthesis of cIDPRE15 to afford the corre-
1
The structures 3a-d were confirmed by H NMR, 13C NMR,
and HMBC. Correlations between H-2 of adenine and C-1′′ of
the alkane or ether chain on the N1 position of the intermediate
(3a-d) were observed in the NMR HMBC spectrum. We tried
to complete the phosphorylation at the free hydroxyl group of
3a-d; however, it was very difficult to isolate the phosphory-
lated products in accordance with the report by Shuto in 2001.12
Therefore, the hydroxyl groups of intermediates 3a-d were
protected with MMTrCl, and then the 5′-tert-butyldimethylsilyl
(TBDMS) groups in 4a-d were removed by TBAF solution.

5504 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
sponding cyclic pyrophosphate compounds (9a-d) in 60%
yield. The cyclic pyrophosphate structures of 9a-b were
1
confirmed by the data from HR-ESI-MS, H NMR, and 31P
NMR. Deprotection of the isopropylidene group of 9a-d was
carried out with aqueous HCOOH at room temperature for 8 h
to furnish target compounds 10a-d. After repeated HPLC
1
purification, 10a-d were characterized by HR-ESI-MS, H
NMR, and 31P NMR (see Experiment Section).
Pharmacology. The cADPR/RyR signaling pathway is criti-
cally important for T-lymphocyte Ca2+ signaling and T cell
activation.4,23,24 Thus, we aimed to pharmacologically character-
ize the novel analogues in this cell system. Ca2+ signaling can
be analyzed in living T cells using fluorescent Ca2+ indicators,
both in cell suspensions9 and on the single cell level.25,26 Since
cIDPRE has been shown to be membrane-permeant and the
novel compounds were expected to be membrane-permeant too,
all analyses were carried out with intact cells. To assay the
antagonist property of any compound, cells were preincubated
and a quasi-physiological stimulation via the T cell receptor/
CD3 complex was carried out. To assay the agonist property
of any compound, the compound was added to the cell
suspension or single cell and any effect on Ca2+ signaling was
recorded directly.
The pharmacological activities of compounds 10a-d were
analyzed in intact human Jurkat T-lymphocytes. Preincubation
of 10a-d did not alter Ca2+ mobilization evoked by anti-CD3
monoclonal antibody (mAb) OKT3; rather, a small additional
stimulatory effect on [Ca2+]i was observed (Figure 3). Thus,
all four compounds appear to have no antagonist effect on the
cADPR/RyR signaling system.
Next, the four compounds were analyzed for their potential
activity as agonists of the cADPR/RyR signaling pathway. All
four compounds 10a-d were added to intact Jurkat T-
lymphocytes. At an extracellular concentration of 500 µM, they
induced a rapid but weak elevation of [Ca2+]i (Figure 4). It
appears that all compounds are membrane-permeant analogues
of cADPR, similar to cIDPRE and cIDP-DE,15,18 since they act
on intact cells. All compounds showed a typical biphasic Ca2+
mobilizing kinetics with an initial immediate Ca2+ peak and a
subsequent plateau phase (Figure 4). The induction of the
immediate peak was strongest for compound 10a at 500 µM,
while for the compounds 10b-d a less pronounced initial Ca2+
peak was observed. A quantitative comparison with cIDPRE15
is depicted in Figure 5; while compounds 10b-d produced small
Ca2+ peak elevations between 10 and 500 µM, there was an
increased effect of compound 10a at 500 µM (Figure 5A). In
comparison to cIDPRE, however, even the effect of compound
10a was approximate 2-fold smaller (Figure 5A). Similar results
were also obtained when analyzing the sustained Ca2+ signal
(Figure 5B); while cIDPRE showed the highest mean activity,
all of the new compounds were weaker. Again, as shown in
Figure 5A for the Ca2+ peak, the effect of compound 10a was
slightly stronger compared to compounds 10b-d.
Figure 4. Ca2+ mobilization by compounds 10a-d in intact Jurkat T
cells. Jurkat T cells were loaded with Fura 2/AM and analyzed by
ratiometric fluorometry on a Hitachi F-2000 flurometer. Compounds
were added 200 s after the beginning of measurement. Each experiment
was calibrated by the addition of Triton X-100 to obtain a maximal
ratio at 900 s and by subsequent addition of EGTA/Tris-base to obtain
the minimal ratio.
Since compound 10a was most effective among the novel
compounds, its Ca2+ mobilizing effect was also analyzed on
the single T cell level by confocal Ca2+ imaging experiments.
Individual single Jurkat T cells challenged by compound 10a
(500 µM) reacted rapidly with high increases in [Ca2+]i, although
the reactivity of some cells was delayed and smaller in amplitude
(Figure 6A). The mean value (red curve in Figure 6A) is similar
to the kinetics seen in Figure 4 although the initial peak was
not as pronounced compared to the cell suspension measure-
ments (see Figure 4, upper panel). Comparing the mean increase
in [Ca2+]i evoked by compound 10a to a vehicle control (Figure
6B, blue curve) or to a positive control (induction by anti-CD3
mAb OKT3; black curve in Figure 6B) suggests that compound
10a can substitute for almost the full sustained part of the quasi-
physiological activation via the TCR/CD3 complex. However,
in the initial phase, activation of RyR by compound 10a
obviously can only induce part of the signal (Figure 6B). Using
confocal Ca2+ imaging at a faster acquisition rate (approximately
one ratio per 150 ms), upon extracellular addition of compound
10a we demonstrate the induction of localized small Ca2+
signals in the vicinity of the cell border but also in many regions

Cyclic ADP-Ribose Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18 5505
Figure 5. Concentration response relationship of Ca2+ mobilization by compounds 10a-d in intact Jurkat T cells. Jurkat T cells were loaded with
Fura 2/AM and analyzed by ratiometric fluorometry on a Hitachi F-2000 flurometer. “Ca2+ peak” (A) and “Ca2+ plateau” (B) relate to the initial
Ca2+ peak (approximately 150-300 s) and the sustained Ca2+ plateau (800 s), respectively. Data are the mean ( SE (n ) 2-6).
deep in the cytosol (Figure 6C, time point 39.38 s). The confocal
image taken at 39.38 s in Figure 6C (marked and magnified
region) also demonstrates the recruitment of neighboring smaller
signals. Finally, a global Ca2+ signal occurred (Figure 6C).
Taken together, our study demonstrates that the four novel
cADPR analogues 10a-d are weak agonists of the cADPR/
RyR signaling system. The results also indicate that these four
compounds do not display any antagonist activity.
Compared to native cADPR, substitutions on C8 of the
adenine ring confer antagonistic activity to the cADPR ana-
logues.5 Other possibilities to construct cADPR antagonists
include modifications at the C3 of the southern ribose.8 Although
the compounds introduced in the current report do not display
any modifications at C8 of the adenine ring, we suspected that
the replacement of the northern ribose by different types of ether
bridges might also result in, at least, some antagonist properties.
However, this assumption turned out to be wrong. Thus, the
data support the idea that C8 modifications of the adenine base
are important constituents of cADPR antagonists. Whether this
also applies to analogues in which the northern ribose is replaced
by ether bridges similar to the ones used in the current study
remains to be demonstrated.
The results presented here confirm that the agonist effects of
cADPR analogues 10a-d are retained, although at lower
magnitude, even when the northern ribose is replaced by an
ether or alkane bridge. Unlike cIDPRE and cIDP-DE, com-
pounds 10a-d are analogues of cADPR in the sense that the
base adenine is preserved. They are the first membrane-permeant
cADPR analogues that contain ether or alkane bridges instead
of the northern ribose and retain adenine as its base. Except for
the difference in the base, the ether strand of 10a contains one
CH2 more than cIDPRE.15 Even though compound 10a has one
extra carbon, it appears that the oxygen can still bind, possibly
because of the increased flexibility provided by the longer chain.
The conformation of 10a may allow a better shift to an ideal
binding position to match the receptor binding site. However,
when compared to cIDPRE,15 the Ca2+ peak induced by
compound 10a was much less pronounced while a similar effect
of 10a on the sustained Ca2+ signal was observed. This result
may indicate that there are two independent targets for the
cADPR mimics. The initial Ca2+ peak evoked by the first target
may require a northern ribose closer to the natural molecule,
like in cIDPRE,15 while for the sustained phase the dependency
on this part of the molecule is not so strong.

5506 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
According to the results in the intact T cell suspension
measurements (Figures 4 and 5), the oxygen of the bridge that
connects N1 of adenosine to the phosphate moiety plays an
important role in the binding of the analogue to its target receptor
because the biological activities of compounds 10b-d are
weaker than that of compound 10a. The oxygen of the bridge
may contribute as a proton acceptor for the formation of a
hydrogen bond with a neighboring amino acid residue of the
receptor. However, the molecules with the alkane chains still
appear to fit into the cADPR receptor binding site, although
the oxygen at the right position is required for a better response.
Recently, we have demonstrated that analogues of cADPR
containing the base hypoxanthine instead of adenine, e.g., N1-
cIDPR,16 8-Br-cIDPR,27 cIDPRE,15 cIDP-DE,18 are almost not
hydrolyzable when incubated with native or recombinant
CD38.28 This is likely due to the amide-like bond between N1
and C1′′ of the northern ribose or its substitute, e.g., ether chain.
In contrast, the corresponding N1-C1′′ bond in cADPR was
hydrolyzed by both CD38 and, though to a weaker extent, ADP-
ribosyl cyclase from Aplysia californica.28 Although metabolism
experiments have not yet been carried out with the novel
analogues, a similar behavior as for cADPR can be expected.
Taken together, the first cADPR analogues, in which the
northern ribose is replaced by an ether or alkane bridge and the
base adenine is retained, were synthesized. The analogues were
weak agonists of the cADPR/RyR signaling system in intact
human Jurkat T cells but did not show antagonist activity. This
series of analogues enlarges our knowledge of the structure-
activity relationship of cADPR in the sense that even a
replacement of the northern ribose that no longer contains polar
components is sufficient for (weak) biological activity.
Experiment Section
Chemistry. Mass spectra were obtained on either VG-ZAB-HS
or Bruker APEX. High-resolution FAB (fast atom bombardment)
MS and HR-ESI-MS (ESI ) electrospray ionization) were per-
formed with Bruker BIFLEX III. 1H NMR and 13C NMR data were
recorded with a JEOL AL300 or a Varian VXR-500 spectrometer
using DMSO-d6 or D2O as solvent. Chemical shifts are reported in
parts per million downfield from TMS (1H and 13C). 31P NMR
spectra were recorded at room temperature by use of Bruker Avance
300 spectrometer (121.42 MHz); orthophosphoric acid (85%) was
used as an external standard. Compounds 10a-d were purified
twice on an Alltech preparative C18 reversed-phase column (2.2
cm × 25 cm) using a Gilson HPLC buffer system: MeCN/TEAB
(pH 7.5) and MeCN/TEAA (pH 7.0).
N1-[(5′′-Hydroxyl)ethoxyethyl]-5′-O-TBDMS-2′,3′-O-isopro-
pylideneadenosine 3a. A mixture of 1 (486 mg, 1.1 mmol), 2a
(144 mg, 1.3 mmol), and K2CO3 (8 mg, 0.06 mmol) in MeOH (15
mL) was stirred at room temperature for 4 h. The mixture was
evaporated, and the residue was partitioned between H2O and
EtOAc. The organic layer was washed with brine, dried (Na2SO4),
and evaporated. The residue was purified by silica gel column
chromatography (1:30 MeOH /CH2Cl2) to give compound 3a (477
Figure 6. Ca2+ mobilization by compound 10a in single T cells. Jurkat
T cells were loaded with Fura-2/AM and analyzed by confocal
ratiometric Ca2+ imaging as described in “Experimetal Section”.
Extracellular Ca2+ was present at 1 mM throughout the experiments.
In panel A the black curves in the diagram represent the behavior of
11 individual Jurkat T cells after the addition of compound 10a (500
µM) at the time point of 40 s, while the red curve shows the mean
value. Panel B shows the mean value curve (red) compared to effects
of anti-CD3 mAb OKT3 (black) and vehicle control (blue). In panel C
characteristic confocal ratiometric pseudocolor images of a single Jurkat
cell and magnifications of a defined subcellular region are shown. Basal
phase (before addition of compound 10a (final concentration of 0.5
mM)), pacemaker phase, and global phase are indicated. The image
acquisition rate was approximately 1 ratio image per 150 ms.
1
mg, 85%). H NMR (500 MHz, CDCl3) δ 0.00 (s, 6H, (CH3)2Si),
0.85 (s, 9H, (CH3)3C-), 1.38, 1.61 (each s, each 3H, (CH3)2C),
3.56-3.57 (m, 2H, H-5′′), 3.69-3.71 (m, 2H, H-4′′), 3.71-3.84
(m, 4H, H-2′′, H-5′′), 4.18-4.30 (m, 2H, H-1′′), 4.35-4.40 (m,
1H, H-4′), 4.90 (dd, 1H, J H3 ,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′),
′
5.12 (dd, 1H, JH1′, H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.02 (d,
1H, JH2′,H1′ ) 2.5 Hz, H-1′), 7.76, 7.82 (each s, each 1H, H-2, H-8);
13C NMR (125 MHz, CDCl3) δ 154.6, 148.1, 141.2, 136.6, 123.7,
114.1, 91.2, 87.0, 85.3, 81.3, 72.7, 68.1, 63.5, 61.6, 47.4, 27.2, 25.9,
25.4, 18.3, -5.6, -5.5. ESI-TOF+-MS: calcd for C23H39N5O6Si
[(M + 1)+], 510.3; found, 510.3. Anal. (C23H39N5O6Si) C, H, N.
N1-(5′′-Monomethoxytrityloxyethoxyethyl)-5′-O-TBDMS-2′,3′-
O-isopropylideneadenosine 4a. A mixture of 3a (477 mg, 0.935

Cyclic ADP-Ribose Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18 5507
mmol) and MMTrCl (577 mg, 1.87 mmol) in pyridine (10 mL)
was stirred at room temperature for 8 h. The mixture was
evaporated, and the residue was partitioned between H2O and
EtOAc. The organic layer was washed with brine, dried (Na2SO4),
and evaporated. The residue was purified by silica gel column
chromatography (1:60 MeOH/ CH2Cl2) to give compound 4a (658
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.45 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.28 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 7.20-7.50 (m, 10H, Ar-H), 7.80 (br s, 1H, NH), 8.25, 8.32
(each s, each 1H, H-2, H-8). 13C NMR (125 MHz, CDCl3) δ 154.6,
148.1, 141.2, 136.6, 132.5, 129.5, 128.9, 125.6, 123.7, 114.1, 91.2,
87.0, 85.3, 81.3, 72.7, 68.1, 63.5, 61.6, 47.4, 27.2, 25.4. 31P NMR
(CDCl3, 81 MHz, decoupled with 1H) δ 49.38 ppm (s). ESI-TOF+-
MS: calcd for C29H34N5O7PS2 [(M + 1)+], 660.2; found, 660.2.
Anal. (C29H34N5O7PS2) C, H, N.
1
mg, 90%). H NMR (500 MHz, CDCl3) δ 0.00 (s, 6H, (CH3)2Si),
0.86 (s, 9H, (CH3)3C-), 1.29, 1.58 (each s, each 3H, (CH3)2C),
3.17-3.21(m, 2H, H-5′′), 3.57-3.65 (m, 2H, H-4′′), 3.73-3.90
(m, 7H, H-1′′, H-2′′, OCH3 ), 4.13-4.22(m, 1H, H-5′a), 4.30-
4.40 (m, 2H, H-4′, H-5′b), 4.85 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′
) 6.0 Hz, H-3′), 5.00 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz,
H-2′), 6.00 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.81-7.44 (m, 14H,
Ar-H), 7.81, 7.82 (each s, each 1H, H-2, H-8). 13C NMR (75 MHz,
CDCl3) δ 158.5, 152.8, 148.5, 144.5, 135.7, 130.3, 129.2, 128.4,
127.8, 127.1, 126.8, 114.0, 113.2, 113.0, 91.0, 87.1, 86.2, 85.2,
81.2, 70.8, 68.1, 63.5, 63.0, 55.2, 47.7, 27.2, 25.9, 25.3, 18.3, -5.5,
-5.4. ESI-TOF+-MS: calcd for C43H55N5O7Si [(M + 1)+], 782.4;
found, 782.3. Anal. (C43H55N5O7Si) C, H, N.
N1-(5′′-Phosphonoxyethoxyethyl)-5′-O-[(phenylthio)phospho-
ryl]-2′,3′-O-isopropylideneadenosine 8a. POCl3 (282 µL, 3.02
mmol) was added to a solution of 7a (200 mg, 0.302 mmol) in
PO(OMe)3 (3 mL) at 0 °C, and the mixture was stirred at the same
temperature for 35 min, then quenched by aqueous saturated
NaHCO3 (6 mL). The resulting solution was stirred at 0 °C for 15
min, then concentrated in vacuo. The residue was dissolved in 1
mL of TEAB (0.1 M, pH 7.5).The solution was purified by a C18
reversed-phase column (2.2 cm × 25 cm) using a linear gradient
of 0-60% CH3CN in TEAB buffer (0.1 M, pH 7.5). The
appropriate fractions were collected and evaporated. The residue
was coevaporated with pyridine (5 mL × 3). The residue was mixed
with H3PO2 (120 µL, 2.4 mmol) and Et3N (156 µL, 1.1 mmol) in
pyridine (3 mL). The mixture was stirred at room temperature in
the dark for 12 h, then evaporated in vacuo. The residue was
partitioned between H2O and CHCl3, and the aqueous layer was
washed with CHCl3 (5 mL × 3) and evaporated in vacuo. The
residue was dissolved in 1 mL of TEAB buffer (0.1 M, pH 7.5),
then applied to a C18 reversed-phase column (2.2 cm × 25 cm)
developed by a linear gradient of 0-60% CH3CN in TEAB buffer
(0.1 M, pH 7.5) within 30 min to give 8a (110 mg, 56% for two
N1-(5′′-Monomethoxytrityloxyethoxyethyl)-2′,3′-O-isopropyl-
ideneadenosine 5a. A mixture of 4a (658 mg, 0.84 mmol), TBAF
(1 M in THF, 8.4 mL, 8.4 mmol), and AcOH (266 µL,4.2 mmol)
in THF (15 mL) was stirred at room temperature for 2 h. The
mixture was evaporated, and the residue was purified by silica gel
column chromatography (1:40 MeOH/CH2Cl2) to give compound
5a (533 mg, 95%). 1H NMR (500 MHz, CDCl3) δ 1.28, 1.60 (each
s, each 3H, (CH3)2C), 3.19-3.22(m, 2H, H-5′′), 3.59-3.61 (m, 2H,
H-4′′), 3.72-3.88 (m, 7H, H-5′ , H-2′′; OCH3 ), 4.11-4.35(m, 2H,
H-1′′), 4.43-4.44 (m, 1H, H-4′), 4.93 (dd, 1H, JH3′,H4′ ) 3.0 Hz,
JH2′,H3′ ) 6.0 Hz, H-3′), 4.99 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′
)
1
6.0 Hz, H-2′), 5.75 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.81-7.42 (m,
14H, Ar-H), 7.43 (s,1H, H-2), 7.78 (s,1H, H-8). 13C NMR (125
MHz, CDCl3) δ 158.4, 153.9, 148.5, 144.4, 144.3, 140.5, 138.1,
135.7, 130.2, 128.3, 127.7, 126.8, 114.1, 113.1, 93.6, 86.2, 86.0,
83.7, 81.3, 70.8, 67.8, 63.0, 63.0, 55.2, 48.2, 27.5, 25.1. ESI-TOF+-
MS: calcd for C37H41N5O7 [(M + 1)+], 668.3; found, 668.3. Anal.
(C37H41N5O7) C, H, N.
steps) as a triethylammonium salt. H NMR (500 MHz, D2O) δ
1.36, 1.58 (each s, each 3H, (CH3)2C), 3.50-3.79 (m, 8H, H-5′,
H-2′′, H-4′′, H-5′′), 4.40-4.46 (m, 2H, H-1′′), 4.59-4.60 (m, 1H,
H-4′), 5.12 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.43
(dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.18 (d, 1H,
JH2′,H1′ ) 2.5 Hz, H-1′), 7.30-7.60 (m, 5H, Ar-H), 8.28, 8.35 (each
s, each 1H, H-2, H-8). 31P NMR (D2O, 81 MHz, decoupled with
1H) δ 2.06 ppm (s), 17.79 ppm (s). HRMS (ESI-TOF-) calcd for
C23H31N5O11P2S [(M - 1)-], 646.1138; found, 646.1129.
N1-(5′′-Monomethoxytrityloxyethoxyethyl)-5′-O-[bis(phenylth-
io)phosphoryl]-2′,3′-O-isopropylideneadenosine 6a. To a solution
of 5a (533 mg, 0.798 mmol) in pyridine (10 mL) was added TPSCl
(483 mg, 1.586 mmol), PSS (909 mg, 2.394 mmol), and tetrazole
(168 mg, 2.394 mmol), and the mixture was stirred at room
temperature for 12 h. The mixture was evaporated, and the residue
was partitioned between H2O and CH2Cl2. The organic layer was
washed with brine, dried (Na2SO4), and evaporated. The residue
was purified by silica gel column chromatography (1:40 MeOH/
CH2Cl2) to give compound 6a (483 mg, 65%). 1H NMR (500 MHz,
CDCl3) δ 1.29, 1.61 (each s, each 3H, (CH3)2C), 3.20-3.23(m,
2H, H-5′′), 3.59-3.62 (m, 2H, H-4′′), 3.74-3.91 (m, 7H, H-5′ ,
H-2′′, OCH3 ), 4.12-4.35 (m, 2H, H-1′′), 4.43-4.45 (m, 1H, H-4′),
4.95 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.00 (dd,
1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.76 (d, 1H, JH2′,H1′
) 2.5 Hz, H-1′), 6.81-7.46 (m, 24H, Ar-H), 7.45 (s,1H, H-2),
7.79 (s,1H, H-8). 13C NMR (125 MHz, CDCl3) δ 158.4, 153.9,
148.5, 144.4, 144.3, 140.5, 138.1, 135.7, 132.5, 130.2, 129.4, 129.1,
128.3, 127.8, 126.8, 125.6, 114.1, 113.1, 93.6, 86.2, 86.0, 83.7,
81.3, 70.8, 67.8, 63.0, 62.9, 55.2, 48.1, 27.5, 25.1. 31P NMR (CDCl3,
N1-[(5′′-O-Phosphoryl)ethoxyethyl]-2′,3′-O-isopropylidene-5′-
O-phosphoryladenosine 5′,5′′-Cyclicpyrophosphate 9a. A solu-
tion of 8a (15 mg, 23 µmol) in pyridine (5 mL) was added slowly
over 20 h, using a syringe pump, to a mixture of AgNO3 (83 mg,
490 µmol) and 3 Å molecular sieves (2.0 g) in pyridine (50 mL) at
room temperature in the dark. The 3 Å molecular sieves were
filtered off with Celite and washed with H2O. The combined filtrate
was evaporated, and the residue was partitioned between CHCl3
and H2O. The aqueous layer was evaporated, and the residue was
dissolved in 0.1 M TEAB buffer (1.0 mL), which was applied to
a C18 reversed-phase column (2.2 cm × 25 cm). The column was
developed using a linear gradient of 0-60% CH3CN in TEAB
buffer (0.1 M, pH 7.5) within 30 min to give 9a (7.4 mg, 60%) as
1
a triethylammonium salt. H NMR (500 MHz, D2O) δ 1.38, 1.61
(each s, each 3H, (CH3)2C), 3.30-3.80 (m, 8H, H-5′, H-2′′, H-4′′,
H-5′′), 4.45-4.48 (m, 2H, H-1′′), 4.58-4.59 (m, 1H, H-4′), 5.10
(dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.45 (dd, 1H,
JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.15 (d, 1H, JH2′,H1′
)
1
81 MHz, decoupled with H) δ 49.39 ppm (s). ESI-TOF+-MS:
2.5 Hz, H-1′), 8.30, 8.38 (each s, each 1H, H-2, H-8). 31P NMR
calcd for C49H50N5O8PS2 [(M + 1)+], 932.3; found, 932.3. Anal.
(C49H50N5O8PS2) C, H, N.
(D2O, 81 MHz, decoupled with H) δ -9.69 ppm (d, JP,P ) 10.0
1
Hz), -10.61 ppm (d, JP,P ) 10.0 Hz). HRMS (ESI-TOF-) calcd
for C17H25N5O11P2 [(M - 1)-], 536.0948; found, 536.0940.
N1-[(5′′-O-Phosphoryl)ethoxyethyl]-5′-O-phosphoryladeno-
sine 5′,5′′-Cyclicpyrophosphate 10a. A solution of 9a (7.4 mg,
13.8 µmol) in 60% HCOOH (5 mL) was stirred for 8 h and then
evaporated under reduced pressure. The purification of the residue
was performed with the same procedure as for compound 9a by
HPLC on a C18 reversed-phase column, eluting with a linear
gradient of 0-60% CH3CN in TEAB buffer (0.1 M, pH 7.5) to
give the target molecule 10a (6.2 mg, 90%). 1H NMR (500 MHz,
D2O) δ 3.20-3.71(m, 8H, H-5′, H-2′′, H-4′′, H-5′′), 4.45-4.48 (m,
2H, H-1′′), 4.50-4.60 (m, 1H, H-4′), 5.05 (dd, 1H, JH3′,H4′ ) 3.0
N1-[(5′′-Hydroxyl)ethoxyethyl]-5′-O-[bis(phenylthio)phospho-
ryl]-2′,3′-O-isopropylideneadenosine 7a. A solution of 6a (483
mg, 0.519 mmol) in 80% aqueous AcOH (10 mL) was stirred at
room temperature for 8 h. The mixture was evaporated, and the
residue was partitioned between aqueous saturated NaHCO3 and
EtOAc. The organic layer was washed with brine, dried (Na2SO4),
and evaporated. The residue was purified by silica gel column
chromatography (1:30 MeOH/CH2Cl2) to give compound 7a (291
1
mg, 85%). H NMR (500 MHz, DMSO-d6) δ 1.34, 1.56 (each s,
each 3H, (CH3)2C), 3.42-3.75 (m, 8H, H-5′, H-2′′, H-4′′, H-5′′),
4.37-4.51 (m, 3H, H-1′′, OH), 4.59-4.62 (m, 1H, H-4′), 5.05 (dd,

5508 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.40 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′
) 6.0 Hz, H-2′), 6.35 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 8.25, 8.58
(each s, each 1H, H-2, H-8). 31P NMR (D2O, 81 MHz, decoupled
δ 158.4, 153.7, 147.4, 147.3, 144.6, 140.3, 138.3, 138.2, 135.9,
132.7, 130.2, 129.4, 129.1, 128.3, 127.7, 126.8, 125.6, 125.0, 114.1,
113.0, 93.8, 86.1, 85.8, 83.6, 81.4, 63.1, 62.6, 55.2, 48.2, 27.5, 27.0,
25.5, 25.2. 31P NMR (CDCl3, 81 MHz, decoupled with 1H) δ 49.41
ppm (s). ESI-TOF+-MS: calcd for C49H50N5O7PS2 [(M + 1)+],
916.3; found, 916.3. Anal. (C49H50N5O7PS2) C, H, N.
1
with H) δ -9.69 ppm (d, JP,P ) 13.4 Hz), -10.38 ppm (d, JP,P
)13.4 Hz). HRMS (ESI-TOF-) calcd for C14H21N5O11P2 [(M -
1)-], 496.0635; found, 496.0629.
N1-[(4′′-Hydroxyl)butyl]-5′-O-TBDMS-2′,3′-O-isopropylidene-
adenosine 3b. A mixture of 1 (486 mg, 1.1 mmol), 2b (120 mg,
1.3 mmol), and K2CO3 (8 mg, 0.06 mmol) in MeOH (15 mL) was
stirred at room temperature for 4 h. The procedure was the same
N1-[(4′′-Hydroxyl)butyl]-5′-O-[bis(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 7b. A solution of 6b (435 mg,
0.475 mmol) in 80% aqueous AcOH (10 mL) was stirred at room
temperature for 8 h. The procedure was the same as for the synthesis
1
1
as for the synthesis of 3a to give 3b in 82% yield. H NMR (500
of 7a to give 7b in 83% yield. H NMR (300 MHz, CDCl3) δ
MHz, CDCl3) δ 0.00 (s, 6H, (CH3)2Si), 0.86 (s, 9H, (CH3)3C-),
1.39, 1.62 (each s, each 3H, (CH3)2C), 1.64-1.68 (m, 2H, H-3′′),
1.85-1.98 (m, 2H, H-2′′), 3.77-3.89 (m, 4H, H-4′′ , H-5′), 4.05-
4.20 (m, 2H, H-1′′), 4.38-4.42 (m, 1H, H-4′), 4.90 (dd, 1H, JH3′,H4′
) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.10 (dd, 1H, JH1′,H2′ ) 2.5 Hz,
JH2′,H3′ ) 6.0 Hz, H-2′), 6.02 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 7.72,
7.85 (each s, each 1H, H-2, H-8). 13C NMR (125 MHz, CDCl3) δ
154.6, 147.1, 141.5, 136.9, 123.7, 114.1, 91.2, 87.0, 85.2, 81.3,
63.5, 61.7, 47.1, 28.1, 27.2, 25.8, 25.7, 25.4, 18.3, -5.5, -5.6.
ESI-TOF+-MS: calcd for C23H39N5O5Si [(M + 1)+], 494.3; found,
494.3. Anal. (C23H39N5O5Si) C, H, N.
1.35, 1.63 (each s, each 3H, (CH3)2C), 1.76-1.79 (m, 2H, H-3′′),
1.89-1.93 (m, 2H, H-2′′), 3.05-3.08 (m, 2H, H-4′′), 3.82-3.98
(m, 2H, H-5′), 4.04-4.08 (m, 2H, H-1′′), 4.49-4.50 (m, 1H, H-4′),
4.89 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.26 (dd,
1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.98 (d, 1H, JH2′,H1′
) 2.5 Hz, H-1′), 6.83-7.43 (m, 10H, Ar-H), 7.63 (s, 1H, H-2),
7.70 (s, 1H, H-8). 13C NMR (125 MHz, CDCl3) δ 154.6, 147.1,
141.5, 136.9, 132.4, 129.4, 129.1, 125.6, 123.7, 114.1, 91.2, 87.0,
85.2, 81.3, 63.5, 61.7, 47.1, 28.1, 27.2, 25.7, 25.4. 31P NMR (CDCl3,
1
81 MHz, decoupled with H) δ 49.37 ppm (s). ESI-TOF+-MS:
calcd for C29H34N5O6PS2 [(M + 1)+], 644.2; found, 644.2. Anal.
(C29H34N5O6PS2) C, H, N.
N1-(4′′-Monomethoxytrityloxybutyl)-5′-O-TBDMS-2′,3′-O-iso-
propylideneadenosine 4b. A mixture of 3b (445 mg, 0.9 mmol)
and MMTrCl (557 mg, 1.8 mmol) in pyridine (10 mL) was stirred
at room temperature for 8 h. The procedure was the same as for
N1-(4′′-Phosphonoxybutyl)-5′-O-[(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 8b. POCl3 (290 µL, 3.11 mmol)
was added to a solution of 7b (200 mg, 0.311 mmol) in PO(OMe)3
(3 mL) at 0 °C, and the mixture was stirred at the same temperature
for 35 min. The following procedure was the same as for the
synthesis of 8a to give 8b in 52% yield. 1H NMR (500 MHz, D2O)
δ 1.35, 1.63 (each s, each 3H, (CH3)2C), 1.77-1.79 (m, 2H, H-3′′),
1.88-1.92 (m, 2H, H-2′′), 3.07-3.09 (m, 2H, H-4′′), 3.86-3.99
(m, 2H, H-5′), 4.06-4.10 (m, 2H, H-1′′), 4.51-4.52 (m, 1H, H-4′),
4.99 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.45 (dd,
1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.02 (d, 1H, JH2′,H1′
) 2.5 Hz, H-1′), 6.88-7.42 (m, 5H, Ar-H), 8.25 (s, 1H, H-2),
8.38 (s, 1H, H-8). 31P NMR (D2O, 81 MHz, decoupled with 1H) δ
2.84 ppm (s), 17.69 ppm (s). HRMS (ESI-TOF-) calcd for
C23H31N5O10P2S [(M - 1)-], 630.1189; found, 630.1182.
1
the synthesis of 4a to give 4b in 90% yield. H NMR (300 MHz,
CDCl3) δ 0.00 (s, 6H, (CH3)2Si), 0.82 (s, 9H, (CH3)3C-), 1.36,
1.59 (each s, each 3H, (CH3)2C),1.62-1.67 (m, 2H, H-3′′), 1.85-
1.88(m, 2H, H-2′′), 3.05-3.10 (m, 2H, H-4′′), 3.73-3.90 (m, 5H,
H-5′, OCH3 ), 3.92-4.05(m, 2H, H-1′′), 4.35-4.37 (m, 1H, H-4′),
4.86 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.07 (dd,
1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.00 (d, 1H, JH2′,H1′
) 2.5 Hz, H-1′), 6.78-7.41 (m, 14H, Ar-H), 7.60,7.78 (each s,
each 1H, H-2, H-8). 13C NMR (75 MHz, CDCl3) δ 158.2, 153.5,
147.4, 147.3, 144.5, 139.9, 138.3, 138.3, 135.9, 130.2, 128.9, 127.8,
126.8, 125.1, 114.2, 113.1, 93.9, 86.0, 85.6, 83.6, 81.4, 63.2, 62.7,
55.5, 48.7, 27.6, 27.0, 25.7, 25.4, 25.3, 18.3, -5.5, -5.6. ESI-
TOF+-MS: calcd for C43H55N5O6Si [(M + 1)+], 766.4; found,
766.3. Anal. (C43H55N5O6Si) C, H, N.
N1-[(4′′-O-Phosphoryl)butyl]-2′,3′-O-isopropylidene-5′-O-phos-
phoryladenosine 5′,5′′-Cyclicpyrophosphate 9b. A solution of 8b
(15 mg, 23.8 µmol) in pyridine (5 mL) was added slowly over 20
h, using a syringe pump, to a mixture of AgNO3 (86 mg, 507 µmol)
and 3 Å molecular sieves (2.07 g) in pyridine (50 mL) at room
temperature in the dark. The procedure was the same as for the
synthesis of 9a to give 9b in 60% yield. 1H NMR (500 MHz, D2O)
δ 1.39, 1.65 (each s, each 3H, (CH3)2C), 1.79-1.80 (m, 2H, H-3′′),
1.86-1.89 (m, 2H, H-2′′), 3.07-3.09 (m, 2H, H-4′′), 3.89-3.98
(m, 2H, H-5′ ), 4.08-4.11 (m, 2H, H-1′′), 4.49-4.50 (m, 1H, H-4′),
5.02 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.65 (dd,
1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.10 (d, 1H, JH2′,H1′
) 2.5 Hz, H-1′), 8.27 (s, 1H, H-2), 8.39 (s, 1H, H-8). 31P NMR
N1-(4′′-Monomethoxytrityloxybutyl)-2′,3′-O-isopropylidene-
adenosine 5b. A mixture of 4b (620 mg, 0.81 mmol), TBAF (1 M
in THF, 8.1 mL, 8.1 mmol), and AcOH (260 µL, 4.1 mmol) in
THF (20 mL) was stirred at room temperature for 2 h. The
procedure was the same as for the synthesis of 5a to give 5b in
90% yield. 1H NMR (300 MHz, CDCl3) δ 1.37, 1.63 (each s, each
3H, (CH3)2C), 1.70-1.80 (m, 2H, H-3′′), 1.85-1.91 (m, 2H, H-2′′
), 3.09-3.13 (m, 2H, H-4′′), 3.74-3.96 (m, 5H, H-5′ , OCH3 ),
3.98-4.03 (m, 2H, H-1′′), 4.49-4.50 (m, 1H, H-4′), 4.92 (dd, 1H,
JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.10 (dd, 1H, JH1′,H2′
)
2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.95 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 6.81-7.44 (m, 14H, Ar-H), 7.58 (s, 1H, H-2), 7.63 (s, 1H,
H-8). 13C NMR (75 MHz, CDCl3) δ 158.4, 153.7, 147.4, 147.3,
144.6, 140.3, 138.3, 138.2, 135.9, 130.2, 128.3, 127.8, 126.8, 125.0,
114.1, 113.0, 93.8, 86.1, 85.8, 83.6, 81.4, 63.1, 62.6, 55.2, 48.2,
27.5, 27.0, 25.5, 25.2. ESI-TOF+-MS: calcd for C37H41N5O6 [(M
+ 1)+], 652.3; found, 652.2. Anal. (C37H41N5O6) C, H, N.
1
(D2O, 81 MHz, decoupled with H) δ -9.65 ppm (d, JP,P ) 10.0
Hz), -10.35 ppm (d, JP,P ) 10.0 Hz). HRMS (ESI-TOF-) calcd
for C17H25N5O10P2 [(M - 1)-], 520.0999; found, 520.1004.
N1-[(4′′-O-Phosphoryl)butyl]-5′-O-phosphoryladenosine 5′,5′′-
Cyclicpyrophosphate 10b. A solution of 9b (7.4 mg, 14.3 µmol)
in 60% HCOOH (5 mL) was stirred for 8 h. The procedure was
1
N1-(4′′-Monomethoxytrityloxybutyl)-5′-O-[bis(phenylthio)-
phosphoryl]-2′,3′-O-isopropylideneadenosine 6b. To a solution
of 5b (475 mg, 0.73 mmol) in pyridine (10 mL) were added TPSCl
(444 mg, 1.458 mmol), PSS (831 mg, 2.19 mmol), and tetrazole
(154 mg, 2.19 mmol), and the mixture was stirred at room
temperature for 12 h. The procedure was the same as for the
the same as for the synthesis of 10a to give 10b in 90% yield. H
NMR (500 MHz, D2O) δ 1.77-1.79 (m, 2H, H-3′′), 1.87-1.89
(m, 2H, H-2′′), 3.05-3.07 (m, 2H, H-4′′), 3.90-3.96 (m, 2H, H-5′),
4.12-4.17 (m, 2H, H-1′′), 4.52-4.53 (m, 1H, H-4′), 4.95 (dd, 1H,
JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.40 (dd, 1H, JH1′,H2′
)
2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.08 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 8.30 (s, 1H, H-2), 8.38 (s, 1H, H-8). 31P NMR (D2O, 81
MHz, decoupled with 1H) δ -9.70 ppm (d, JP,P ) 10.0 Hz), -10.35
ppm (d, JP,P ) 10.0 Hz). HRMS (ESI-TOF-) calcd for C14H21N5O10P2
[(M - 1)-], 480.0686; found, 480.0678.
1
synthesis of 6a to give 6b in 65% yield. H NMR (300 MHz,
CDCl3) δ 1.39, 1.65 (each s, each 3H, (CH3)2C), 1.73-1.79 (m,
2H, H-3′′), 1.87-1.92 (m, 2H, H-2′′ ), 3.08-3.11 (m, 2H, H-4′′),
3.72-3.98 (m, 5H, H-5′ , OCH3 ), 4.00-4.03 (m, 2H, H-1′′), 4.48-
4.49 (m, 1H, H-4′), 4.97 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0
Hz, H-3′), 5.30 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′),
5.94 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.81-7.46 (m, 24H, Ar-H),
7.60 (s, 1H, H-2), 7.66 (s, 1H, H-8). 13C NMR (75 MHz, CDCl3)
N1-[(5′′-Hydroxyl)pentyl]-5′-O-TBDMS-2′,3′-O-isopropylide-
neadenosine 3c. A mixture of 1 (486 mg, 1.1 mmol), 2c (135 mg,
1.3 mmol), and K2CO3 (8 mg, 0.06 mmol) in MeOH (15 mL) was
stirred at room temperature for 4 h. The procedure was the same

Cyclic ADP-Ribose Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18 5509
1
as for the synthesis of 3a to give 3c in 84% yield. H NMR (500
MHz, CDCl3) δ 0.00 (s, 6H, (CH3)2Si), 0.86 (s, 9H, (CH3)3C-),
1.39 (s, 3H, isopropyl CH3), 1.46-1.52 (m, 2H, H-3′′), 1.62 (s,
3H, isopropyl CH3), 1.63-1.67 (m, 2H, H-4′′), 1.83-1.89 (m, 2H,
H-2′′), 3.65-3.67 (m, 2H, H-5′′), 3.76-3.86 (m, 2H, H-5′), 4.00-
4.10 (m, 2H, H-1′′), 4.38-4.40 (m, 1H, H-4′), 4.89 (dd, 1H, JH3′,H4′
) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.11 (dd, 1H, JH1′,H2′ ) 2.5 Hz,
JH2′,H3′ ) 6.0 Hz, H-2′), 6.03 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 7.68,
7.82 (each s, each 1H, H-2, H-8). 13C NMR (125 MHz, CDCl3) δ
152.7, 145.5, 139.4, 134.4, 122.1, 112.3, 89.4, 85.3, 83.5, 79.5,
61.7, 60.3, 46.1, 30.3, 27.9, 26.5, 25.4, 24.1, 23.6, 21.1, 16.5, -7.2,
-7.3. ESI-TOF+-MS: calcd for C24H41N5O5Si [(M + 1)+], 508.3;
found, 508.3. Anal. (C24H41N5O5Si) C, H, N.
(s, 3H, isopropyl CH3), 1.48-1.70(m, 9H, H-2′′, H-3′′, H-4′′,
isopropyl CH3 ), 3.05 (m, 2H, H-5′′), 3.85-3.97 (m, 2H, H-5′),
4.05-4.09 (m, 2H, H-1′′), 4.52-4.53 (m, 1H, H-4′), 4.90 (dd, 1H,
JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.05 (dd, 1H, JH1′,H2′
)
2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.95 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 6.85-7.42 (m, 10H, Ar-H), 7.65, 7.70 (each s, each 1H,
H-2, H-8). 13C NMR (125 MHz, CDCl3) δ 152.7, 145.5, 139.4,
134.4, 132.7, 129.5, 129.2, 125.7, 122.1, 112.3, 89.4, 85.3, 83.5,
79.5, 61.7, 60.3, 46.1, 30.3, 27.9, 26.5, 25.4, 24.1, 23.6. 31P NMR
(CDCl3, 81 MHz, decoupled with 1H) δ 49.50 ppm (s). ESI-TOF+-
MS: calcd for C30H36N5O6PS2 [(M + 1)+], 658.2; found, 658.2.
Anal. (C30H36N5O6PS2) C, H, N.
N1-(5′′-Phosphonoxypentyl)-5′-O-[(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 8c. POCl3 (283 µL, 3.04 mmol)
was added to a solution of 7c (200 mg, 0.304 mmol) in PO(OMe)3
(3 mL) at 0 °C, and the mixture was stirred at the same temperature
for 35 min. The following procedure was the same as for the
synthesis of 8a to give 8c in 51% yield. 1H NMR (300 MHz, D2O)
δ 1.41 (s, 3H, isopropyl CH3), 1.52-1.75 (m, 9H, H-2′′, H-3′′,
H-4′′, isopropyl CH3), 3.20 (m, 2H, H-5′′), 3.85-4.00 (m, 2H, H-5′
), 4.10-4.21 (m, 2H, H-1′′), 4.52-4.53 (m, 1H, H-4′), 5.02 (dd,
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.38 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.05 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 6.90-7.35 (m, 5H, Ar-H), 8.15, 8.32 (each s, each 1H, H-2,
N1-(5′′-Monomethoxytrityloxypentyl)-5′-O-TBDMS-2′,3′-O-
isopropylideneadenosine 4c. A mixture of 3c (469 mg, 0.924
mmol) and MMTrCl (572 mg, 1.848 mmol) in pyridine (10 mL)
was stirred at room temperature for 8 h. The procedure was the
1
same as for the synthesis of 4a to give 4c in 91% yield. H NMR
(300 MHz, CDCl3) δ 0.00 (s, 6H, (CH3)2Si), 0.86 (s, 9H, (CH3)3C-
), 1.39 (s, 3H, isopropyl CH3 ), 1.45-1.78 (m, 9H, H-2′′, H-3′′,
H-4′′, isopropyl CH3), 3.06 (m, 2H, H-5′′), 3.78-3.83 (m, 5H, H-5′,
OCH3 ), 3.99-4.04 (m, 2H, H-1′′), 4.39-4.40 (m, 1H, H-4′), 4.89
(dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.09 (dd, 1H,
JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.03 (d, 1H, JH2′,H1′
)
1
2.5 Hz, H-1′), 6.81-7.44 (m, 14H, Ar-H), 7.64, 7.82 (each s, each
1H, H-2, H-8). 13C NMR (75 MHz, CDCl3) δ 158.3, 153.7, 147.4,
144.6, 141.3, 139.5, 136.3, 130.0, 128.1, 127.7, 124.3, 114.1, 112.8,
93.2, 86.4, 84.2, 81.9, 63.1, 55.5, 49.1, 29.4, 28.3, 27.5, 26.3, 25.2,
23.5, 17.2, -7.1, -7.2. ESI-TOF+-MS: calcd for C44H57N5O6Si
[(M + 1)+], 780.4; found, 780.3. Anal. (C44H57N5O6Si) C, H, N.
N1-(5′′-Monomethoxytrityloxypentyl)-2′,3′-O-isopropylidene-
adenosine 5c. A mixture of 4c (656 mg, 0.84 mmol), TBAF (1 M
in THF, 8.4 mL, 8.4 mmol), and AcOH (273 µL, 4.3 mmol) in
THF (20 mL) was stirred at room temperature for 2 h. The
procedure was the same as for the synthesis of 5a to give 5c in
H-8). 31P NMR (D2O, 81 MHz, decoupled with H) δ 3.01 ppm
(s), 17.65 ppm (s). HRMS (ESI-TOF-) calcd for C24H33N5O10P2S
[(M - 1)-], 644.1345; found, 644.1338.
N1-[(5′′-O-Phosphoryl)pentyl]-2′,3′-O-isopropylidene-5′-O-
phosphoryladenosine 5′,5′′-Cyclicpyrophosphate 9c. A solution
of 8c (15 mg, 23.2 µmol) in pyridine (5 mL) was added slowly
over 20 h, using a syringe pump, to a mixture of AgNO3 (84 mg,
494 µmol) and 3 Å molecular sieves (2.02 g) in pyridine (50 mL)
at room temperature in the dark. The procedure was the same as
1
for the synthesis of 9a to give 9c in 59% yield. H NMR (300
MHz, D2O) δ 1.38 (s, 3H, isopropyl CH3), 1.49-1.72(m, 9H, H-2′′,
H-3′′, H-4′′, isopropyl CH3), 3.25 (m, 2H, H-5′′), 3.95-4.15 (m,
2H, H-5′), 4.20-4.32 (m, 2H, H-1′′), 4.55 (m, 1H, H-4′), 5.05 (dd,
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.55 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.10 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 8.25, 8.35 (each s, each 1H, H-2, H-8). 31P NMR (D2O, 81
MHz, decoupled with1H) δ -9.99 ppm (d, JP,P ) 10.0 Hz), -10.72
ppm (d, JP,P ) 10.0 Hz). HRMS (ESI-TOF-) calcd for C18H27N5O10P2
[(M - 1)-], 534.1155; found, 534.1139.
1
90% yield. H NMR (300 MHz, CDCl3) δ 1.37 (s, 3H, isopropyl
CH3), 1.45-1.78 (m, 9H, H-2′′, H-3′′, H-4′′, isopropyl CH3), 3.06
(m, 2H, H-5′′), 3.73-3.91 (m, 5H, H-5′, OCH3 ), 3.95-4.03(m,
2H, H-1′′), 4.49-4.50 (m, 1H, H-4′), 4.95 (dd, 1H, JH3′,H4′ ) 3.0
Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.05 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′
) 6.0 Hz, H-2′), 5.78 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.81-7.44
(m, 14H, Ar-H), 7.64, 7.66 (each s, each 1H, H-2, H-8). 13C NMR
(75 MHz, CDCl3) δ 158.3, 153.3, 147.2, 144.7, 141.0, 139.0, 136.0,
130.2, 128.3, 127.7, 124.1, 114.1, 112.9, 93.4, 86.0, 83.9, 81.4,
62.9, 55.1, 48.7, 29.6, 28.4, 27.5, 25.2, 23.3. ESI-TOF+-MS: calcd
for C38H43N5O6 [(M + 1)+], 666.3; found, 666.2. Anal. (C38H43N5O6)
C, H, N.
N1-[(5′′-O-Phosphoryl)pentyl]-5′-O-phosphoryladenosine 5′,5′′-
Cyclicpyrophosphate 10c. A solution of 9c (7.3 mg, 13.7 µmol)
in 60% HCOOH (5 mL) was stirred for 8 h. The procedure was
1
the same as for the synthesis of 10a to give 10c in 89% yield. H
N1-(5′′-Monomethoxytrityloxypentyl)-5′-O-[bis(phenylthio)-
phosphoryl]-2′,3′-O-isopropylideneadenosine 6c. To a solution
of 5c (503 mg, 0.756 mmol) in pyridine (10 mL) were added TPSCl
(460 mg, 1.51 mmol), PSS (861 mg, 2.27 mmol), and tetrazole
(159 mg, 2.27 mmol). The mixture was stirred at room temperature
for 12 h. The procedure was the same as for the synthesis of 6a to
NMR (300 MHz, D2O) δ 1.55-1.69 (m, 6H, H-2′′, H-3′′, H-4′′),
3.32 (m, 2H, H-5′′), 3.86-4.05 (m, 2H, H-5′ ), 4.15-4.28 (m, 2H,
H-1′′), 4.38 (m, 1H, H-4′), 4.98 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′
) 6.0 Hz, H-3′), 5.45 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz,
H-2′), 6.15 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 8.15, 8.55 (each s, each
1
1H, H-2, H-8). 31P NMR (D2O, 81 MHz, decoupled with H) δ
1
give 6c in 65% yield. H NMR (300 MHz, CDCl3) δ 1.39 (s, 3H,
-9.72 ppm (d, JP,P ) 10.1 Hz), -10.10 ppm (d, JP,P ) 10.1 Hz).
HRMS (ESI-TOF-) calcd for C15H23N5O10P2 [(M - 1)-], 494.0842;
found, 494.0836.
isopropyl CH3), 1.41-1.72 (m, 9H, H2′′, H-3′′, H-4′′, isopropyl
CH3), 3.11 (m, 2H, H-5′′), 3.75-3.95 (m, 5H, H-5′, OCH3 ), 4.00-
4.10(m, 2H, H-1′′), 4.51-4.52 (m, 1H, H-4′), 4.89 (dd, 1H, JH3′,H4′
) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.10 (dd, 1H, JH1′,H2′ ) 2.5 Hz,
JH2′,H3′ ) 6.0 Hz, H-2′), 5.78 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.80-
N1-[(6′′-Hydroxyl)hexyl]-5′-O-TBDMS-2′,3′-O-isopropylide-
neadenosine 3d. A mixture of 1 (486 mg, 1.1 mmol), 2d (157 mg,
1.3 mmol), and K2CO3 (8 mg, 0.06 mmol) in MeOH (15 mL) was
stirred at room temperature for 4 h. The procedure was the same
7.56 (m, 24H, Ar-H), 7.56, 7.68 (each s, each 1H, H-2, H-8). 13
C
1
NMR (75 MHz, CDCl3) δ 158.3, 153.5, 147.2, 144.8, 140.9, 138.9,
136.1, 132.5, 130.2, 129.4, 129.1, 128.3, 127.7, 126.3, 124.1, 114.0,
112.9, 93.5, 86.1, 84.0, 81.6, 63.1, 55.1, 48.6, 29.5, 28.5, 27.6, 25.3,
23.4. 31P NMR (CDCl3, 81 MHz, decoupled with 1H) δ 49.45 ppm
(s). ESI-TOF+-MS: calcd for C50H52N5O7PS2 [(M + 1)+], 930.3;
found, 930.3. Anal. (C50H52N5O7PS2) C, H, N.
as for the synthesis of 3a to give 3d in 83% yield. H NMR (500
MHz, DMSO-d6) δ 0.00 (s, 6H, (CH3)2Si), 0.80 (s, 9H, (CH3)3C-
), 1.31 (s, 3H, isopropyl CH3 ), 1.32-1.69 (m, 11H, H-2′′, H-3′′,
H-4′′, H-5′′, isopropyl CH3), 3.34-3.38 (m, 2H, H-6′′ ), 3.64-
3.73 (m, 2H, H-5′), 3.92-4.08 (m, 2H, H-1′′), 4.19-4.20 (m, 1H,
H-4′ ), 4.32 (t, 1H, JH6′′,OH ) 4.5 Hz, OH), 4.88 (dd, 1H, JH3′,H4′
)
N1-[(5′′-Hydroxyl)pentyl]-5′-O-[bis(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 7c. A solution of 6c (457 mg,
0.491 mmol) in 80% aqueous AcOH (10 mL) was stirred at room
temperature for 8 h. The procedure was the same as for the synthesis
of 7a to give 7c in 82% yield. 1H NMR (300 MHz, CDCl3) δ 1.36
3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.28 (dd, 1H, JH1′,H2′ ) 2.5 Hz,
JH2′,H3′ ) 6.0 Hz, H-2′), 6.04 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 7.07
(s, br, NH), 8.05,8.06 (each s, each 1H, H-2, H-8). 13C NMR (125
MHz, CDCl3) δ 154.5, 147.3, 141.2, 136.7, 123.9, 114.1, 91.2, 87.0,
85.3, 81.3, 63.5, 62.5, 47.7, 32.5, 29.7, 28.6, 27.2, 26.3, 25.9, 25.2,

5510 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
18.3, -5.4, -5.6. ESI-TOF+-MS: calcd for C25H43N5O5Si [(M +
1)+], 522.3; found, 522.3. Anal. (C25H43N5O5Si) C, H, N.
129.3, 125.8, 123.9, 114.1, 91.3, 87.0, 85.3, 81.4, 63.5, 62.5, 47.7,
32.5, 29.7, 28.7, 27.3, 25.9, 25.3. 31P NMR (D2O, 81 MHz,
decoupled with1H) δ 49.50 ppm (s) ESI-TOF+-MS: calcd for
C31H38N5O6PS2 [(M + 1)+], 672.2; found, 672.2. Anal. (C31H38N5O6-
PS2) C, H, N.
N1-(6′′-Monomethoxytrityloxyhexyl)-5′-O-TBDMS-2′,3′-O-iso-
propylideneadenosine 4d. A mixture of 3d (476 mg, 0.9 mmol)
and MMTrCl (557 mg, 1.8 mmol) in pyridine (10 mL) was stirred
at room temperature for 8 h. The procedure was the same as for
N1-(6′′-Phosphonoxyhexyl)-5′-O-[(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 8d. POCl3 (278 µL, 2.98 mmol)
was added to a solution of 7d (200 mg, 0.298 mmol) in PO(OMe)3
(3 mL) at 0 °C, and the mixture was stirred at the same temperature
for 35 min. The following procedure was the same as for the
synthesis of 8a to give 8d in 54% yield. 1H NMR (500 MHz, D2O)
δ 1.32 (s, 3H, isopropyl CH3 ), 1.48-1.80(m, 11H, H-2′′, H-3′′,
H-4′′, H-5′′, isopropyl CH3), 3.35 (m, 2H, H-6′′), 3.80-3.90 (m,
2H, H-5′), 4.03-4.08(m, 2H, H-1′′), 4.55 (m, 1H, H-4′), 4.99 (dd,
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.35 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.01 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 6.90-7.48 (m, 5H, Ar-H), 8.25, 8.35 (each s, each 1H, H-2,
H-8). 31P NMR (D2O, 81 MHz, decoupled with1H) δ 2.06 ppm
(s), 17.79 ppm (s). HRMS (ESI-TOF-) calcd for C25H35N5O10P2S
[(M - 1)-], 658.1502; found, 658.1496.
1
the synthesis of 4a to give 4d in 89% yield. H NMR (300 MHz,
CDCl3) δ 0.00 (s, 6H, (CH3)2Si), 0.86 (s, 9H, (CH3)3C-), 1.39 (s,
3H, isopropyl CH3), 1.45-1.78(m, 11H, H-2′′, H-3′′, H-4′′, H-5′′,
isopropyl CH3), 3.05 (m, 2H, H-6′′), 3.78-3.83 (m, 5H, H-5′, OCH3
), 3.99-4.04 (m, 2H, H-1′′), 4.39-4.40 (m, 1H, H-4′), 4.89 (dd,
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.10 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.92 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 6.85-7.45 (m, 14H, Ar-H), 7.64, 7.82 (each s, each 1H,
H-2, H-8). 13C NMR (125 MHz, CDCl3) δ 158.6, 154.1, 148.2,
145.3, 138.0, 136.3, 129.2, 128.3, 127.5, 126.6, 114.1, 112.9, 94.3,
86.1, 85.8, 83.6, 82.3, 63.5, 63.0, 54.9, 48.3, 28.8, 28.6, 28.0, 27.0,
26.7, 26.1, 25.6, 18.3, -5.4, -5.6. ESI-TOF+-MS: calcd for
C45H59N5O6Si [(M + 1)+], 795.4; found, 795.5. Anal. (C45H59N5O6-
Si) C, H, N.
N1-[(6′′-O-Phosphoryl)hexyl]-2′,3′-O-isopropylidene-5′-O-phos-
phoryladenosine 5′,5′′-Cyclicpyrophosphate 9d. A solution of 8d
(15 mg, 22.8 µmol) in pyridine (5 mL) was added slowly over 20
h, using a syringe pump, to a mixture of AgNO3 (82 mg, 486 µmol)
and 3 Å molecular sieves (1.98 g) in pyridine (50 mL) at room
temperature in the dark. The procedure was the same as for the
synthesis of 9a to give 9d in 60% yield. 1H NMR (500 MHz, D2O)
δ 1.39 (s, 3H, isopropyl CH3 ), 1.50-1.79(m, 11H, H-2′′, H-3′′,
H-4′′, H-5′′, isopropyl CH3), 3.43 (m, 2H, H-6′′), 3.85-3.94 (m,
2H, H-5′), 4.10-4.15(m, 2H, H-1′′), 4.58 (m, 1H, H-4′), 5.08 (dd,
1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.50 (dd, 1H, JH1′,H2′
) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 6.02 (d, 1H, JH2′,H1′ ) 2.5 Hz,
H-1′), 8.25, 8.38 (each s, each 1H, H-2, H-8). 31P NMR (D2O, 81
MHz, decoupled with 1H) δ -9.99 ppm (d, JP,P ) 10.0 Hz), -10.72
ppm (d, JP,P ) 10.0 Hz). HRMS (ESI-TOF-) calcd for C19H29N5O10P2
[(M - 1)-], 548.1312; found, 548.1295.
N1-(6′′-Monomethoxytrityloxyhexyl)-2′,3′-O-isopropylidene-
adenosine 5d. A mixture of 4d (645 mg, 0.80 mmol), TBAF (1 M
in THF, 8.0 mL, 8.0 mmol), and AcOH (254 µL, 4.0 mmol) in
THF (20 mL) was stirred at room temperature for 2 h. The
procedure was the same as for the synthesis of 5a to give 5d in
1
90% yield. H NMR (500 MHz, CDCl3) δ 1.35 (s, 3H, isopropyl
CH3 ), 1.39-1.78(m, 11H, H-2′′, H-3′′, H-4′′, H-5′′, isopropyl CH3),
3.02 (m, 2H, H-6′′), 3.74-3.77 (dd, 1H, JH4′,H5′a ) 1.5 Hz, JH5′b,H5′a
) 12.5 Hz, H-5′a), 3.78 (s, 3H, OCH3), 3.92-3.95 (dd, 1H, JH4′,H5′b
) 1.5 Hz, JH5′b,H5′b ) 12.5 Hz, H-5′b ), 3.99-4.02(m, 2H, H-1′′),
4.48 (m, 1H, H-4′), 4.98 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0
Hz, H-3′), 5.12 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′),
5.79 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.81-7.44 (m, 14H, Ar-H),
7.67, 7.71 (each s, each 1H, H-2, H-8). 13C NMR (125 MHz,
CDCl3) δ 158.4, 153.8, 147.3, 144.9, 138.3, 136.2, 130.2, 128.4,
127.7, 126.7, 114.1, 112.9, 93.8, 86.0, 85.8, 83.6, 81.4, 63.2, 63.1,
55.2, 48.3, 28.9, 28.6, 27.6, 26.5, 26.0, 25.2. ESI-TOF+-MS: calcd
for C39H45N5O6 [(M + 1)+], 680.3; found, 680.4. Anal. (C39H45N5O6)
C, H, N.
N1-[(6′′-O-Phosphoryl)hexyl]-5′-O-phosphoryladenosine 5′,5′′-
Cyclicpyrophosphate 10d. A solution of 9d (7.5 mg, 13.7 µmol)
in 60% HCOOH (5 mL) was stirred for 8 h. The procedure was
1
N1-(6′′-Monomethoxytrityloxyhexyl)-5′-O-[bis(phenylthio)-
phosphoryl]-2′,3′-O-isopropylideneadenosine 6d. To a solution
of 5d (489 mg, 0.72 mmol) in pyridine (10 mL) were added TPSCl
(437 mg, 1.438 mmol), PSS (820 mg, 2.16 mmol), and tetrazole
(152 mg, 2.16 mmol), and the mixture was stirred at room
temperature for 12 h. The procedure was the same as for the
the same as for the synthesis of 10a to give 10d in 90% yield. H
NMR (500 MHz, D2O) δ 1.48-1.76 (m, 8H, H-2′′, H-3′′, H-4′′,
H-5′′), 3.48 (m, 2H, H-6′′), 3.86-3.96 (m, 2H, H-5′), 4.15-4.18-
(m, 2H, H-1′′), 4.62 (m, 1H, H-4′), 5.05 (dd, 1H, JH3′,H4′ ) 3.0 Hz,
JH2′,H3′ ) 6.0 Hz, H-3′), 5.45 (dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′
)
6.0 Hz, H-2′), 6.05 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 8.21, 8.35 (each
s, each 1H, H-2, H-8). 31P NMR (D2O, 81 MHz, decoupled with
1H) δ -9.75 ppm (d, JP,P ) 10.5 Hz), -9.90 ppm (d, JP,P ) 10.5
Hz). HRMS (ESI-TOF-) calcd for C16H25N5O10P2 [(M - 1)-],
508.0999; found, 508.0988.
1
synthesis of 6a to give 6d in 64% yield. H NMR (500 MHz,
CDCl3) δ 1.32 (s, 3H, isopropyl CH3 ), 1.42-1.79(m, 11H, H-2′′,
H-3′′, H-4′′, H-5′′, isopropyl CH3), 3.15 (m, 2H, H-6′′), 3.75-3.85
(m, 5H, H-5′, OCH3 ), 4.05-4.10 (m, 2H, H-1′′), 4.45 (m, 1H,
H-4′), 4.86 (dd, 1H, JH3′,H4′ ) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.35
(dd, 1H, JH1′,H2′ ) 2.5 Hz, JH2′,H3′ ) 6.0 Hz, H-2′), 5.96 (d, 1H,
JH2′,H1′ ) 2.5 Hz, H-1′), 6.85-7.47 (m, 24H, Ar-H), 7.69, 7.75
(each s, each 1H, H-2, H-8). 13C NMR (125 MHz, CDCl3) δ 158.3,
154.1, 147.3, 145.2, 138.1, 136.0, 132.2, 129.9, 129.4, 129.1, 128.5,
128.2, 127.0, 125.6, 114.0, 112.6, 93.9, 86.5, 85.8, 83.3, 82.2, 63.6,
63.3, 56.1, 48.2, 29.0, 28.6, 28.2, 26.5, 26.0, 25.4. 31P NMR (D2O,
81 MHz, decoupled with1H) δ 49.48 ppm (s). ESI-TOF+-MS: calcd
for C51H54N5O7PS2 [(M + 1)+], 944.3; found, 944.3. Anal.
(C51H54N5O7PS2) C, H, N.
Pharmacology. Cell Culture. Jurkat T-lymphocytes (clone JMP)
were cultured in RPMI 1640 medium supplemented with Glutamax
I (2.06 mM), 25 mM HEPES, 110 units/mL penicillin, 110 µg/mL
streptomycin, and 7.5% (v/v) newborn calf serum (termed complete
medium).
Loading of Cells with Fura2/AM. The cells (1 × 107 cells)
were centrifuged (room temperature, 1500 rpm, 5 min), and then
the pellet was resuspended in 1 mL of fresh and warm (37 °C)
complete medium (see above). The cells were incubated for 5 min
at 37 °C. Then Fura 2-AM (4 µL of a 1 mg/mL stock solution)
was added, and the cells were incubated for 15 min in the dark.
Subsequently another 4 mL of complete medium was added to the
cell suspension and the cells were incubated for another 15 min at
37 °C in the dark. Then the cells were washed with Ca2+
measurement buffer (140 mM NaCl, 5 mM KCl, 1 mM MgSO4, 1
mM CaCl2, 20 mM HEPES, 1 mM NaH2PO4, 5.5 mM glucose,
pH 7.4) twice, resuspended in 5 mL of Ca2+ measurement buffer,
and kept in the dark at room temperature until use.
N1-[(6′′-Hydroxyl)hexyl]-5′-O-[bis(phenylthio)phosphoryl]-
2′,3′-O-isopropylideneadenosine 7d. A solution of 6d (435 mg,
0.461 mmol) in 80% aqueous AcOH (10 mL) was stirred at room
temperature for 8 h. The procedure was the same as for the synthesis
of 7a to give 7d in 82% yield. 1H NMR (500 MHz, CDCl3) δ 1.36
(s, 3H, isopropyl CH3 ), 1.38-1.75(m, 11H, H-2′′, H-3′′, H-4′′,
H-5′′, isopropyl CH3), 3.05 (m, 2H, H-6′′), 3.78-3.86 (m, 2H, H-5′),
3.98-4.06 (m, 2H, H-1′′), 4.35 (m, 1H, H-4′), 4.78 (dd, 1H, JH3′,H4′
) 3.0 Hz, JH2′,H3′ ) 6.0 Hz, H-3′), 5.25 (dd, 1H, JH1′,H2′ ) 2.5 Hz,
JH2′,H3′ ) 6.0 Hz, H-2′), 5.86 (d, 1H, JH2′,H1′ ) 2.5 Hz, H-1′), 6.89-
Determination of [Ca2+]i in Intact Cells. [Ca2+]i was analyzed
in Fura2-loaded Jurkat T-lymphocytes. Changes in Fura2 fluores-
cence were measured using a Hitachi F-2000 spectrofluorometer
operating in ratio mode (alternating excitation at 340 and 380 nm
7.43 (m, 10H, Ar-H), 7.70, 7.72 (each s, each 1H, H-2, H-8). 13
NMR (125 MHz, CDCl3) δ 154.5, 147.4, 141.2, 136.7, 132.4, 129.5,
C

Cyclic ADP-Ribose Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18 5511
(6) Vu, C. Q.; Lu, P. J.; Chen, C. S.; Jacobson, M. K. 2′-Phospho-cyclic
ADP-ribose, a calcium-mobilizing agent derived from NADP. J. Biol.
Chem. 1996, 271, 4747-4754.
(7) Guse, A. H. 1-(5-Phospho-beta-D-ribosyl) 2′-phosphoadenosine 5′-
phosphate cyclic anhydride induced Ca2+ release in human T-cell
lines. Eur. J. Biochem. 1997, 245 (2), 411-417.
(8) Ashamu, G. A.; Sethi, J. K.; Galione, A.; Potter, B. V. L. Roles for
adenosine ribose hydroxyl groups in cyclic adenosine 5′-diphosphate
ribose-mediated Ca2+ release. Biochemistry 1997, 36, 9509-
9517.
(9) Guse, A. H.; CakirKiefer, C.; Fukuoka, M.; Shuto, S.; Weber, K.;
Bailey, V.C.; Matsuda, A.; Mayr, G. W.; Oppenheimer, N.; Schuber,
F.; Potter, B. V. L. Novel hydrolysis-resistant analogues of cyclic
ADP-ribose: modification of the “northern” ribose and calcium
release activity. Biochemistry 2002, 41 (21), 6744-6751.
(10) Zhang, F. J.; Yamada, S.; Gu, Q. M. Synthesis and characterization
of cyclic ATP-ribose: A potent mediator of calcium release. Bioorg.
Med. Chem. Lett. 1996, 6 (11), 1203-1208.
(11) Bailey, V. C.; Fortt, S. M.; Summerhill, R. J.; Galione, A.; Potter,
B. V. L. Cyclic aristeromycin diphosphate ribose: a potent and poorly
hydrolysable Ca2+-mobilizing mimic of cyclic adenosine diphosphate
ribose. FEBS Lett. 1996, 379, 227-230.
(12) Shuto, S.; Fukuoka, M.; Manikowasky, A.; Ueno, Y.; Nakano, T.;
Kuroda, R.; Kuroda, H.; Matsuda, A. Total synthesis of cyclic ADP-
carbocyclic-ribose, a stable mimic of Ca2+-mobilizing second mes-
senger cyclic ADP-ribose. J. Am. Chem. Soc. 2001, 123 (36), 8750-
8759.
(13) 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. Synthesis of stable and cell-
type selective analogues of cyclic ADP-ribose, a Ca2+-mobilizing
second messenger. Structure-activity relationship of the N1-ribose
moiety. J. Am. Chem. Soc. 2005, 127 (24), 8846-8855
(14) Huang, L. J.; Zhao, Y. Y.; Yuan, L.; Min, J. M.; Zhang, L. H.
Syntheses and calcium mobilizing evaluations of N1-glycosyl
substituted stable mimics of Cyclic ADP-ribose. J. Med. Chem. 2002,
45, 5340-5352.
(15) Gu, X. F.; Yang, Z. J.; Zhang, L. R.; Zhang, L. H. Synthesis and
biological evaluation of novel membrane-permeant cyclic ADP-ribose
mimics: N1-[(5′′-O-phosphoryl-ethoxy)-methyl]-5′-O-phosphoryl-
inosine 5′,5′′-cyclicpyrophosphate (cIDPRE) and 8-substituted deriva-
tives. J. Med. Chem. 2004, 47, 5674-5682.
and emission at 495 nm). To test any antagonist activity of the
compounds, the cells in the cuvette (1 × 106 cells) were preincu-
bated with the compound (500 mM) for 20 min before the start of
the Ca2+ measurement. Anti-CD3 mAb OKT3 (10 µg/mL final
concentration) was added to the cell suspension at a time point
200 s after the start of each measurement. To test any agonist effect
of the compounds, each compound was added to the cell suspension
at the 200 s time point and data were recorded for the next 700 s.
To calibrate the Ca2+ determinations, Triton X-100 (0.1% (v/v))
and EGTA/Tris (12 mM/90 mM) were added sequentially at the
end of each experiment.
Ratiometric Ca2+ Imaging. Jurkat T-lymphocytes were loaded
with Fura2/AM as described above. Imaging experiments were
carried out on thin glass coverslips (0.1 mm) coated with bovine
serum albumin (5 mg/mL) and poly-L-lysine (1 mg/mL). Silicon
grease was used to seal a small chamber that consisted of a rubber
O-ring on the glass coverslip. Then 70 µL of Ca2+ measurement
buffer (composition see above) and 30 µL of cell suspension were
added into the small chamber. The coverslip with cells attached to
the bovine serum albumin/poly-L-lysine coating was mounted on
the stage of a fluorescence microscope (Leica DM IRB2). An
Improvision imaging system (Tu¨bingen, Germany) consisting of a
monochromator system (Polychromator IV, TILL Photonics, Graefelf-
ing, Germany) and a gray-scale CCD camera (type C4742-95-12ER;
Hamamatsu, Enfield, U.K.; operating in 8-bit mode) built around
the Leica microscope at 100-fold magnification was used. Illumina-
tion was carried out at 340 and 380 nm alternatively. The acquisition
rate was approximately one ratio per second for slow measurements
and one ratio per 150 ms for fast measurements. Raw data were
stored on a compact disk. Confocal images were obtained after
deconvolution using the volume deconvolution module of Openlab
software (Improvision). The removal of stray light was set to 0.7.
The deconvolved images were used to construct ratio images (340/
380). Finally the ratio values were converted into Ca2+ concentra-
tions by external calibration. Ratio images were subjected to a
median filter (3 × 3). Data processing was performed using Openlab
software, versions 4.0.2, 3.0.8, and 1.7.8 (Improvision).
(16) Wagner, G. K.; Guse, A. H.; Potter, B. V. L. Rapid synthetic route
toward structurally modified derivatives of cyclic adenosine 5′-
diphosphate ribose. J. Org. Chem. 2005, 70 (12), 4810-4819.
(17) Graeff, R. M.; Walseth, T. F.; Hill, H. K.; Lee, H. C. Fluorescent
analogs of cyclic ADP-ribose: Synthesis, spectral characterization,
and use. Biochemistry 1996, 35, (2), 379-386.
(18) Guse, A. H.; Gu, X. F.; Zhang, L. R.; Weber, K.; Zhang, L. H. A
minimal structural analogue of cyclic ADP-ribose. J. Biol. Chem.
2005, 280, 15952-15959.
Acknowledgment. This study was supported by the National
Natural Sciences Foundation of China (Grant No. SFCBIC
20320130046 to L.-h.Z. and Grant No. 20472007 to L.Z.), the
Ministry of Science and Technology of China (Grant Nos.
2004CB518904 and 2005BA711A04 to L.-h.Z.), the Research
Fund for the Doctoral Program of Higher Education (Grant No.
20040001140 to L.Z.), the Deutsche Foschungsgemeinschaft
(Grant Nos. GU 360/9-1 and 9-2 to A.H.G.), the Deutsche
Akademische Austauschdienst (Grant No. 423/PPP-vrc-sr to
A.H.G. and L.-h.Z.), and the Gemeinnu¨tzige Hertie-Stiftung
(Grant No. 1.01.1/04/010 to A.H.G.).
(19) Hutchinson, E. J.; Taylor, B. F.; Blackburn, G. M. Stereospecific
synthesis of 1,9-bis(beta-D-glycosyl)adenines: a chemical route to
stable analogues of cyclic-ADP ribose (cADPR). J. Chem. Soc.,
Chem. Commun. 1997, 1859-1860.
(20) Shuto, S.; Fukuoka, M.; Kudoh, T.; Garnham, C.; Galione, A.; Potter,
B. V. L.; Matsuda, A. Convergent synthesis and unexpected Ca2+
-
mobilizing activity of 8-substituted analogues of cyclic ADP-
carbocyclic-ribose. A stable mimic of the Ca2+-mobilizing second
messenger cyclic ADP-ribose. J. Med. Chem. 2003, 46, 4741-
4749.
Supporting Information Available: Elemental analysis data
for 3-7 and charts of HPLC data for 10a-d. This material is
available free of charge via the Internet at http://pubs.acs.org.
(21) Yoshikawa, M.; Kato, T.; Takenishi, T. Study of phosphorylation.
III. Selective phosphorylation of unprotected nucleotides. Bull. Chem.
Soc. Jpn. 1969, 42, 3505-3508.
References
(22) Hata, T.; Kamimura, T. A new method for the synthesis of
oligodeoxyribonucleotides bearing a 5′-terminal phosphate group.
Chem. Lett. 1987, 117-120.
(1) Clapper, D. L.; Walseth, T. F.; Dargie, P. J.; Lee, H. C. Pyridine
nucleotide metabolites stimulate calcium release from sea urchin egg
microsomes desensitized to inositol trisphosphate. J. Biol. Chem.
1987, 262, 9561-9568.
(23) Guse, A. H. Ca2+ signaling in T-lymphocytes. Crit. ReV. Immunol.
1998, 18, 419-448.
(24) Guse, A. H. Biochemistry, biology, and pharmacology of cyclic
adenosine diphosphoribose (cADPR). Curr. Med. Chem. 2004, 11,
847-855.
(2) Rusinko, N.; Lee, H. C. Widespread occurrence in animal tissues of
an enzyme catalyzing the conversion of NAD+ into a cyclic
metabolite with intracellular Ca2+-mobilizing activity. J. Biol. Chem.
1989, 264, 11725-11731.
(25) Kunerth, S.; Mayr, G. W.; Koch-Nolte, F.; Guse, A. H. Analysis of
subcellular calcium signals in T-lymphocytes. Cell. Signalling 2003,
15, 783-792.
(3) Lee, H. C. Multiplicity of Ca2+ messengers and Ca2+ stores: A
perspective from cyclic ADP-ribose and NAADP. Curr. Mol. Med.
2004, 4, 227-237.
(26) Kunerth, S.; Langhorst, M. F.; Schwarzmann, N.; Gu, X. F.; Huang,
L. J.; Yang, Z. J.; Zhang, L. R.; Mills, S. J.; Zhang, L. H.; Potter, B.
V. L.; Guse, A. H. Amplification and propagation of pacemaker Ca2+
signals by cyclic ADP-ribose and the type 3 ryanodine receptor in T
cells. J. Cell Sci. 2004, 117, 2141-2149.
(4) Guse, A. H. Regulation of calcium signaling by the second messenger
cyclic adenosine diphosphoribose (cADPR). Curr. Mol. Med. 2004,
4, 239-248.
(5) Walseth, T. F.; Lee, H. C. Synthesis and characterization of
antagonists of cADPR-induced Ca2+ release. Biochim. Biophys. Acta
1993, 1178 (3), 235-242.
(27) Wagner, G. K.; Black, S.; Guse, A. H.; Potter, B. V. L. First
enzymatic synthesis of an N1-cyclised cADPR (cyclic-ADP ribose)

5512 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 18
Xu et al.
analogue with a hypoxanthine partial structure: discovery of a
membrane permeant cADPR agonist. Chem. Commun. 2003, 15,
1944-1945.
L. H.; Potter, B. V. L.; Guse, A. H. Cellular effects and metabolic
stability of N1-cyclic inosine diphosphoribose and its devivatives.
Br. J. Pharmacol., in press.
(28) Kirchberger, T.; Wagner, G.; Xu, J. F.; Cordiglieri, C.; Wang, P.;
Gasser, A.; Fliegert, R.; Bruhn, S.; Flu¨gel, A.; Lund, F. E. Zhang,
JM060320E