Synthesis of adenophostin A analogues conjugating an aromatic group at the 5′-position as potent IP3 receptor ligands

Article abstract of DOI:10.1021/jm060310d

Previous structure-activity relationship studies of adenophostin A, a potent IP₃ receptor agonist, led us to design the novel adenophostin A analogues 5a-c, conjugating an aromatic group at the 5′-position to develop useful IP₃ recep

Full text of DOI:10.1021/jm060310d

5750
J. Med. Chem. 2006, 49, 5750-5758
Synthesis of Adenophostin A Analogues Conjugating an Aromatic Group at the 5′-Position as
Potent IP3 Receptor Ligands1
Tetsuya Mochizuki,† Yoshihiko Kondo,† Hiroshi Abe,† Stephen C. Tovey,‡ Skarlatos G. Dedos,‡ Colin W. Taylor,‡
Michael Paul,§ Barry V. L. Potter,§ Akira Matsuda,† and Satoshi Shuto*,†
Graduate School of Pharmaceutical Sciences, Hokkaido UniVersity, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan, Department of
Pharmacology, Tennis Court Road, UniVersity of Cambridge, Cambridge, CB 1PD, United Kingdom, and Wolfson Laboratory for Medicinal
Chemistry, Department of Pharmacy and Pharmacology, UniVersity of Bath, Bath BA2 7AY, United Kingdom
ReceiVed March 17, 2006
Previous structure-activity relationship studies of adenophostin A, a potent IP3 receptor agonist, led us to
design the novel adenophostin A analogues 5a-c, conjugating an aromatic group at the 5′-position to develop
useful IP3 receptor ligands. The common key intermediate, a D-ribosyl R-D-glucoside 10R, was stereose-
lectively synthesized by a glycosidation with the 1-sulfinylglucoside donor 11, which was conformationally
restricted by a 3,4-O-cyclic diketal protecting group. After introduction of an aromatic group at the 5-position
of the ribose moiety, an adenine base was stereoselectively introduced at the anomeric â-position to form
7a-c, where the tetra-O-i-butyryl donors 9a-c were significantly more effective than the corresponding
O-acetyl donor. Thus, the target compounds 5a-c were synthesized via phosphorylation of the 2′, 3′′, and
4′′-hydroxyls. The potencies of compounds 5a-c for Ca2+ release were shown to be indistinguishable from
that of adenophostin A, indicating that bulky substitutions at the 5′-position of adenophostin A are well-
tolerated in the receptor binding. This biological activity of 5a-c can be rationalized by molecular modeling
using the ligand binding domain of the IP3 receptor.
Introduction
4 and demonstrated that both compounds bind to IP3 receptors
and stimulate Ca2+ mobilization with potencies comparable to
IP3 itself.5,6a,b These results suggest that the pentofuranosyl
structure of the D-ribose would not be essential for the activity
but that the tetrahydrofuran ring could effectively restrict three-
dimensional positioning of the D-glucose 3,4-bisphosphate,
adenine (aromatic ring), and the third phosphate moiety of the
ring. Accordingly, the 5′-hydroxymethyl moiety of adenophostin
A seems to be unimportant for the binding and Ca2+ mobiliza-
tion, suggesting that this moiety may be appropriate for further
modification to develop novel adenophostin analogues of
biological importance.
Bioactive ligands conjugated with an aromatic group, such
as a fluorescent or a photoreactive aromatic residue, are useful
in labeling target biomolecules and can often effectively clarify
the biological mechanism of action. Nagano and co-workers
recently developed an IP3 analogue conjugating malachite green,
an aromatic fluorescent residue, at the 1-phosphate moiety to
show that the analogue induced specific inactivation of the IP3
receptors in cells upon laser irradiation.8 However, the conjuga-
tion also reduced its affinity for IP3 receptors,8b suggesting that
the 1-phosphate moiety might not be as suitable a site for this
kind of conjugating modification. Therefore, identification of
another site for the modification of IP3 receptor ligands without
reducing the binding affinity is needed. Adenophostin A might
be more suitable as the lead for this kind of modification because
of its higher affinity for the receptor compared with IP3 itself.
Attention has been focused on D-myo-inositol 1,4,5-trisphos-
phate (IP3, 1), an intracellular Ca2+-mobilizing second mes-
senger, because of its biological importance.2 Therefore, ana-
logues of IP3 have been designed and extensively synthesized
to develop specific ligands for IP3 receptors, which have proven
useful in investigating the mechanism of IP3-mediated Ca2+
signaling pathways.2,3 However, no simple and monomeric
analogue has surpassed IP3 itself either in binding affinity for
the IP3 receptor or in Ca2+-mobilizing activity.2,3a In recent
studies, however, dimeric analogues of IP3 have proven to be
super-potent in releasing Ca2+ from intracellular stores of
permeabilized hepatocytes.3b,c For example, a short urea-linked
dimer of IP3 of length ∼0.8 nm has an EC50 value more than
12-fold lower than IP3.3c
In 1993, Takahashi and co-workers isolated adenophostin A
(2) from Penicillium breVicompactum. They found it to be a
very potent IP3 receptor ligand, 10-100 times more potent than
IP3 in stimulating Ca2+ release and binding, with likewise greater
affinity than IP3 to IP3 receptors.4 The intriguing structural and
biological features have prompted several groups including ours
to perform synthetic studies aimed at the design of novel IP3
receptor ligands.5-7 Biological evaluation of these analogues,
based on adenophostin A, showed that the R-D-glucopyranose
structure, the three phosphate groups, and the adenine or another
aromatic ring are essential for significant activity.5 These
findings promoted further synthetic studies of adenophostin A
analogues, with the aim of developing useful biological tools
and/or potential drug leads.
Hydrophobic interactions have been recognized to play a
crucial role in the binding of many ligands to their target
biomolecules.9 Thus, introduction of an aromatic group to lead
compounds sometimes improves binding affinity due to interac-
tions between the aromatic ring and the aromatic and/or aliphatic
hydrophobic residues of the target protein located near the
binding site. Furthermore, such derivatization sometimes realizes
functional inversion of an agonist into an antagonist, which is
well-known, for example, in the development of histaminergic10a
We previously synthesized an adenophostin A analogue 3
lacking the adenine moiety and its des-hydroxymethyl derivative
* To whom correspondence should be addressed. Phone and Fax: +81-
11-706-3769. E-mail: shu@pharm.hokudai.ac.jp.
† Hokkaido University.
‡ University of Cambridge.
§ University of Bath.
10.1021/jm060310d CCC: $33.50 © 2006 American Chemical Society
Published on Web 08/19/2006

5′-Modified Adenophostin A Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5751
designed the sulfoxide donor 11 bearing a 3,4-O-cyclic diketal
protecting group in this study to realize the desired R-selective
glycosidation. The 3,4-O-cyclic diketal protecting group also
was advantageous for selective phosphorylation of the hydroxyls
at a latter stage.7c
The R-Selective Glycosidation. The glycosyl donor 11 with
the 3,4-O-cyclic diketal protection was prepared as reported
recently.11 Glycosidation with the acceptor 12, prepared ac-
cording to the previously reported method,17 and the donor 11
was investigated, and the results are summarized in Table 1.
The reactions were carried out under the conditions with Tf2O
and 2,6-di-tert-butyl-4-methylpyridine (DTBMP) as the pro-
moter using CH2Cl2, toluene, or Et2O as the solvent at low
temperature.14 Although the reaction with CH2Cl2 as the solvent
gave nonstereoselectively a mixture of R/â-glucosidation prod-
ucts (entry 1), the desired R-glucoside 10R was selectively
obtained in the reaction with toluene (entry 2, 50%, R/â ) 10:
1) or Et2O (entry 3, 71%, R/â ) 5:1) as the solvent. We found
that the R-glucoside 10R was obtained as the sole glycosidation
product in good yield when the reaction temperature was kept
at -78 °C and using Et2O as the solvent (entry 4).
Synthesis of 5′-Deoxy-5′-phenyladenophostin A. The syn-
thesis of the target compound 5a, conjugating a phenyl group
at the 5′-position, from the R-glucoside 10R was investigated,
as shown in Scheme 2. After removal of the 5-O-TBS group of
10R, oxidation of the 5-hydroxymethyl moiety of the resulting
13 was examined by various methods, where the Moffatt
oxidation gave the best result. The resulting aldehyde, without
purification because of its instability, was immediately treated
with PhMgBr in THF to give the 5-phenyl product 14a in 52%
yield from 13. Radical deoxygenation at the 5-position was
performed by successive treatment of 14a with PhOCSCl/
DMAP in MeCN and with Bu3SnH/AIBN in heating benzene
to form 15a. Acidic removal of the ketal protecting groups of
15a followed by acylation produced the 1,2,3′,4′-tetra-O-acetate
8a or the 1,2,3′,4′-tetra-O-i-butyrylate 9a.
The Vorbru¨ggen glycosylation with the acetate 8a was
examined as summarized in Table 2. The reaction was first
carried out with N6-benzoyladenine and TMSOTf/Et3N in MeCN
or dichloroethane (entries 1 and 2), where silylated N6-
benzoyladenine should be formed under the reaction conditions.
Although these reactions selectively produced the expected
â-nucleoside, the yields were low. The reaction using silylated
N6-benzoyladenine, prepared from N6-benzoyladenine and hex-
amethyldisilazane/pyridine, as the acceptor and SnCl4 as the
promoter in MeCN improved the yield but was shown to
produce the R/â-mixture nonstereoselectively (entry 3).18
In the course of the large scale synthetic study of the
antitumor nucleoside, 3′-ethynyluridine, we found that an O-i-
butyrylated 3-ethynylribosyl donor was very effective in the
Vorbru¨ggen glycosylation reaction.19 In the glycosylation with
the donors 8a and 9a, the steric hindrance due to the 5-phenyl
group might disturb the access of the nucleobase to the anomeric
â-position, as in the case of the 3′-ethynyluridine synthesis.19
Therefore, we investigated reactions using the tetra-O-i-butyryl
donor 9a. Although the reaction using the 9a by the TMSOTf
method was unsuccessful (Table 2, entry 4), the desired
â-nucleoside 7a was selectively obtained in 60% yield when
9a and silylated N6-benzoyladenine were treated with SnCl4 at
room temperature in MeCN (entry 5). The regio- and stereo-
chemistry of 7a was confirmed by its NOE and HMBC spectra,
as shown in Figure 2.
Figure 1. IP3 (1) and adenophostin A (2) and its analogues (3-4).
and adrenergic10b receptor antagonists. From this viewpoint,
introduction of an aromatic group into an appropriate position
of adenophostin A may be interesting.
With these findings in mind, we designed the novel adeno-
phostin A analogues 5a-c11 having an aromatic group at the
5′-position. In these target compounds, the structures of which
are shown in Figure 1, the 5′-hydroxyl in adenophostin A is
replaced with a phenyl, benzyl, or phenethyl group. A biological
evaluation of these compounds would clarify the steric toler-
ability around the 5′-hydroxymethyl moiety for introducing a
considerably bulky aromatic group in the binding site of the
IP3 receptors. Des-adenine analogue 6, having a phenyl group
at the 5-position, was also designed to confirm the role of the
adenine for the receptor binding in this series of the 5′-modified
adenophostin A analogues.
Results and Discussion
Synthetic Plan. Our synthetic plan for the target compounds
5a-c and 6 is summarized in Scheme 1. Compounds 7a-c,
the effective precursors for the target compounds 5a-c, would
be prepared from the disaccharides 8a-c or 9a-c bearing an
aromatic residue at the 5-position of the ribose moiety via the
â-selective adenine base introduction by the Vorbru¨ggen gly-
cosylation.12 Compounds 8a-c and 9a-c could be obtained,
via introduction of an aromatic residue at the 5-position of the
ribose moiety, from 10R, which was planned to be synthesized
by R-selective glycosidation with the sulfoxide donor 11 and
the acceptor 12. The donor 11 and the acceptor 12 would be
prepared from D-glucose and D-xylose, respectively. Sulfoxide
glycosyl donors, reported by Kahne and co-workers,13 are known
to be quite stable but are effectively activated under mild Lewis
acidic conditions.13,14 We most recently developed highly
R-selective radical and also SN1 C-glycosidation reactions based
on the conformational restriction of the pyranosyl donor to the
unusual 4C1-form15 by the 3,4-O-cyclic diketal protection.16 In
this restricted conformation, the kinetic anomeric effect was
enhanced, resulting in the high R-selectivity in these C-
glycosidation reactions.15 On the basis of these results, we
The three O-i-butyryl groups of 7a were removed simulta-
neously with NaOMe/MeOH to give 16a. The phosphate units

5752 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Mochizuki et al.
Scheme 1
Table 1. Glycosidation with the Donor 11 and the Acceptor 12a
11 + 12 f 10R + 10â
respectively, via the Vorbru¨ggen glycosylation and subsequent
phosphorylation, as summarized in Scheme 2.
Synthesis of the 5-Phenyl Derivative Lacking the Adenine.
The 5-phenyl adenophostin A analogue 6 lacking the adenine
moiety was synthesized via hydride reduction at the anomeric
position of the furanose moiety of the tetra-O-acetyl-5-phenyl
derivative 8a (Scheme 4). When 8a was treated with Et3SiH
and TMSOTf in CH2Cl2 at 0 °C,23 the expected reduction
proceeded effectively and the 2-O-acetyl group was concomi-
tantly removed under these conditions to give 60% of the
diacetate 22 along with 8% of the corresponding triacetate 21.
After removal of the other acetyl groups of 22 with NaOMe/
MeOH, the resulting triol 23 was then phosphorylated and
debenzylated by the method described above to produce the
final target compound 6.
entry
solvent
temp (°C)
time (h)
yield (%)
R/âb
1:1
10:1
5:1
1
2
3
4
CH2Cl2
toluene
Et2O
-78 to -40
-78 to -20
-78 to -20
-78
2
2
2
8
73
50
71
78
Et2O
only R
a Reactions were carried out in Tf2O (2 equiv) and DTBMP (5 equiv) in
1
the presence of molecular sieves (4A). b Determined by H NMR.
were selectively introduced onto the three hydroxyls using the
phosphoramidite method with o-xylene N,N-diethylphosphor-
amidite (XEPA).20 Thus, 16a was treated with XEPA and
imidazolium triflate21 in CH2Cl2, followed by oxidation with
m-CPBA, to give the desired 2′,3′′,4′′-trisphosphate derivative
17a in 90% yield. Finally, the benzyl protecting groups were
all removed in one step by catalytic hydrogenation with Pd black
in aqueous MeOH/CHCl3 to furnish the target trisphosphate 5a
in 99% yield as a sodium salt, after treatment with ion-exchange
resin.
Synthesis of the Benzyl and Phenethyl Derivatives. The
other target compounds, the 5′-benzyl derivative 5b and the 5′-
phenethyl derivative 5c, were synthesized basically according
to the route described above for the synthesis of 5a (Scheme
2).
Grignard addition of PhCH2CH2MgCl to the aldehyde
prepared by the Moffatt oxidation of 13 and subsequent Barton
deoxygenation gave the 5-phenethyl product 15c, as in the above
case of the 5-phenyl derivative 15a. On the other hand, in a
similar procedure using BnMgCl, the yields of the Grignard
addition product and the subsequent deoxygenation product were
insufficient (40 and 18%, respectively, Scheme 2).
Compound 13 was converted into the corresponding 5-iodo
and 5-O-Tf derivatives 18 and 19, respectively, and their SN2-
type substitution using the benzyl organo-copper reagent was
examined (Scheme 3). Treatment of the 5-iodide 18 with
BnMgCl/CuI/HMPA in THF22 produced a mixture of the desired
5-benzyl product 15b and the 5-chloro product 20 in 77% yield
in a ratio of approximately 1:1. However, when the 5-O-triflate
19 was subjected to the same reaction conditions, the 5-benzyl
product 15b was isolated in good yield (66%) along with 20%
of 20.
Biological Activity. A low affinity Ca2+ indicator trapped
within the intracellular stores of permeabilized cells was used
to monitor Ca2+ release mediated by recombinant rat type 1
IP3 receptors in DT40 cells lacking endogenous IP3 receptors.24
Table 3 summarizes the results obtained with each of the
compounds, and Figure 3 compares the concentration-dependent
Ca2+ release by 5b, as a typical example of the newly
synthesized compounds, with those of IP3 and adenophostin A.
Maximally effective concentrations of each of the ligands (IP3,
adenophostin A, 5a-c, and 6) released a similar fraction of the
intracellular Ca2+ stores (∼80%), but they differed significantly
in their potencies. All the 5′-modified adenophostin analogues
5a-c showed significant Ca2+ release similar to the parent
compound adenophostin A. The half-maximally effective con-
centrations (EC50) for 5a (2.1 ( 0.4 nM), 5b (2.8 ( 0.6 nM),
5c (2.7 ( 0.4 nM), and also for adenophostin A (2.1 ( 0.2
nM) are about 13-fold lower than the EC50 for IP3 (24.8 ( 2.1
nM). Compound 6, the des-adenine analogue of 5a, had an EC50
∼2-fold higher than that of IP3
Therefore, 5a-c are full agonists with significant Ca2+
-
mobilizing potency comparable to adenophostin A. It is interest-
ing that all the three 5′-modified analogues 5a-c have very
similar activities, despite their difference in 5′-chain length (n
) 0, 1, or 2) and closely resemble the parent compound. Thus,
the 5′-aromatic moiety of 5a-c does not disturb receptor binding
and probably does not interact with the receptor, suggesting that
it may be located outside the binding pocket. The dramatic
decrease of the Ca2+-mobilizing activity in 6, the analogue
without an adenine, shows that the adenine ring enhances the
The 5-benzyl and 5-phenethyl products 15b and 15c were
successfully converted into the target compounds 5b and 5c,

5′-Modified Adenophostin A Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5753
Scheme 2
Conditions: (a) TBAF, THF, rt, 82%; (b) (1) Moffatt ox, (2) Grignard reagent, THF, rt, 52% (PhMgBr, 14a), 40% (BnMgCl, 14b), 58% (PhCH2CH2MgCl,
14c); (c) (1) PhOCSCl, DMAP, MeCN, rt, (2) Bu3SnH, AlBN, PhH, reflux, 93% (15a), 18% (15b), 64% (15c); (d) (1) 90% TFA, rt, (2) R1 O, Et3N, DMAP,
2
MeCN, rt, 66% (8a), 88% (9a), 69% (9b), 62% (9c); (e) silylated N6-benzoyladenine, SnCl4, MeCN, rt, 60% (7a), 67% (7b), 60% (7c); (f) NaOMe, MeOH,
rt, quant (16a), 96% (16b), quant (16c); (g) XEPA, imidazolium triflate, CH2Cl2, -40 °C-rt, then m-CPBA, -40 C, 90% (17a), 70% (17b), quant (17c);
(h) H2 (60 psi), Pd-black, aq MeOH-CHCl3, rt, 99% (5a), quant (5b), 43% (5c).
Table 2. Vorbru¨ggen Glycosylation with Donors 8a and 9a
Scheme 3
entry
donor
Lewis acid
methoda
solvent
yieldb (R/â)
1
2
3
4
5
8a
8a
8a
9a
9a
TMSOTf
TMSOTf
SnCl4
TMSOTf
SnCl4
A
A
B
A
B
MeCN
(ClH2C)2
MeCN
MeCN
MeCN
18% (only â)
26% (only â)
66% (1:1)
12% (only â)
60% (only â)
a A: N6-Bz-adenine (4 equiv), TMSOTf (10 equiv), and Et3N (4 equiv)
at room temperature. B: silylated N6-Bz-adenine (5 equiv), prepared by
heating N6-Bz-adenine in hexamethyldisilazane/pyridine, and SnCl4 (6 equiv)
1
at room temperature. b The ratio was determined by H NMR.
Conditions: (a) I2, Ph3P, imidazole, PhMe, reflux, 93% (18); (b) Tf2O,
DTBMP, CH2Cl2, 90% (19); (c) 18, BnMgCl, CuI, HMPA, THF, -50-0
°C, 77% (15b/20 ) 1:1); (d) 19, BnMgCl, CuI, HMPA, THF, -50 °C,
66% (15b), 20% (20).
Figure 2. NOE (a) and HMBC (b) data of 7a.
activity for a 5′-phenyl derivative, which is consistent with
previous results.6,7
experiments of 5a, when the H-8 of the adenine ring was
irradiated, NOEs were observed both at H-1′ (2.9%) and at H-2′
(3.6%) of the ribose moiety, suggesting its syn-like conformation
around the ribosyl linkage, similar to adenophostin A.
Conformational Analysis and Molecular Modeling. The
conformation of the 5′-deoxy-5′-phenyladenophostin A (5a) was
investigated by 1H NMR experiments. The large coupling
constants (about 9 Hz) between the protons H-2′′, -3′′, -4′′, and
-5′′ of the glucose moiety show that the pyranose ring adopts a
normal 4C1-chair conformation with the vicinal equatorial trans-
phosphates, similar to adenophostin A. The coupling constants
of the ribose moiety, determined by the proton-decoupling
method, were J1′,2′ ) 3.1 Hz, J2′,3′ ) 5.6 Hz, and J3′,4′ ) 6.3
Hz, respectively. The furanose ring of nucleosides generally is
in equilibrium between C2′-endo and C3′-endo forms, and their
ratio is known to be calculated by the equation, [C2′-endo] (%)
) [J1′,2′ /(J1′,2′ + J3′,4′)] × 100.7e,25 Thus, the [C2′-endo] (%) of
5a was calculated to be 31%, which suggests that the C3′-endo
form is somewhat more stable than the C2′-endo one. However,
the energy difference would be rather small, therefore, both of
the two forms of 5a might exist in solution.26 In NOE
The binding mode of 5a-c to the IP3 receptor was investi-
gated by molecular modeling using the recently reported X-ray
crystal structure of the type 1 IP3 receptor binding core
(1N4K).27 The structures of the 5′-modified analogues 5a-c
were built in MOE (Molecular Operating Environment, v2003.02,
Chemical Computing Group, Canada) and minimized using the
MMFF94 force field. These structures were then placed in the
IP3 receptor (1N4K) by aligning the 3,4-trans-bisphosphate
group on the glucose ring with the 4,5-trans-bisphosphate of
IP3 from the crystal structure. Following removal of IP3 from
the pocket, the heavy atoms of the receptor and associated water
molecules were fixed, and the structure was allowed to
minimize. The result is shown in Figure 4, where the side chains
are colored magenta (5a), green (5b), or orange (5c). The
conformations show that none of the substituents at the

5754 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Mochizuki et al.
Table 3. Ca2+ Release via Type 1 IP3 Receptors Stimulated by IP3 and Adenophostin A and Its Analogues
relative potencya
compound
IP3 (1)
EC50, nM
Hill slope
Ca2+ release, %
n
IP3
adenophostin
24.8 ( 2.1
2.1 ( 0.2
2.1 ( 0.4
2.8 ( 0.6
2.7 ( 0.4
53.3 ( 10.6
1.21 ( 0.06
1.54 ( 0.13
1.62 ( 0.24
1.16 ( 0.08
1.42 ( 0.16
1.11 ( 0.09
78 ( 2
76 ( 1
77 ( 1
76 ( 2
79 ( 2
79 ( 3
11
12
5
5
6
1
0.09 ( 0.01
1
1.0 ( 0.23
0.87 ( 0.20
0.87 ( 0.15
0.03 ( 0.01
adenophostin A (2)
12.8 ( 4.5
13.4 ( 1.7
12.7 ( 3.8
9.1 ( 1.4
5a
5b
5c
6
4
0.45 ( 0.05
a Ca2+ release evoked by each of the ligands is shown with potencies relative to IP3 or adenophostin A, calculated by a comparison of EC50 values
determined in parallel experiments.
Scheme 4
Figure 3. Concentration-dependent release of Ca2+ from intracellular
stores by IP3 (b), Adenophostin A (O), and 5b (9). Results are means
( S.E.M., n g 5.
Conditions: (a) Et3SiH, TMSOTf, CH2Cl2, 0 °C-rt, 8% (21), 60% (22);
(b) NaOMe, MeOH, rt, 86%; (c) XEPA, imidazolium triflate, CH2Cl2, -40
°C-rt, then m-CPBA, -40 °C, 85%; (d) H2 (60 psi), Pd-black, aq MeOH-
CHCl3, rt, 87%.
5′-position appear to show any favorable interactions with the
binding pocket of the receptor as they protrude into areas that
probably contain solvent. However, the limitations of this study
should be stressed because 1N4K.pdb represents only a truncated
form (residues 224-604) of the full length (2700) receptor
representing the core binding domain, and it is not known
whether the region in question could potentially be accessed
by other motifs of the full-length protein with which potential
ligand interactions may be possible. Regarding the conformation
of the analogues modeled on the receptor, while the preference
of a C2′-endo conformation in adenophostin A has been
previously reported, 5a seems to prefer a C3′-endo solution
conformation by NMR, as seen above. In a recent model
proposed by some of us for the binding of adenophostin A to
its receptor an extended 2′-endo conformation with a potential
cation-π interaction between adenine and protein was proposed
for optimal interaction in the ligand binding domain.7a The
energy difference between the two C2′-endo and C3′-endo forms
in solution is assumed to be relatively small, and the very high
activity of the analogues 5a-c, which is comparable to that of
adenophostin A, as reported here, suggests that 5a-c bind to
the receptor in a mode similar to that of adenophostin A. Thus,
we can propose a 2′-endo conformation on the protein, in line
with the established model for the parent ligand.
Figure 4. Modeling of analogues 5a-c into the ligand binding domain
of the IP3 receptor. Possible binding modes for 5a (magenta), 5b (green),
and 5c (orange) are shown. The adenine moiety of adenophostin A
shows a π-stacking interaction with the side-chain of Arg504 as
suggested for adenophostin A.7a
residue, at the 5′-position. This may become more relevant as
the structure of the full length receptor is clarified.
Conclusion. The novel adenophostin A analogues 5a-c
conjugating an aromatic group at the 5′-position, designed as
novel IP3 receptor ligands, were synthesized via two key
stereoselective glycosidation steps, in which the 1-sulfinylglu-
coside donor 11 conformationally restricted by a 3,4-O-cyclic
diketal protecting group and the tetra-O-i-butyryl donors 9a-
c, respectively, were effectively used to realize the desired
stereoselectivity. The potencies of 5a-c for the Ca2+ mobiliza-
tion proved to be indistinguishable from that of adenophostin
A. This activity can be rationalized by molecular modeling using
the ligand binding domain of the IP3 receptor where, subject to
certain caveats, it can be seen that the bulky substitutions have
no formal interaction with the protein. We therefore conclude
that the 5′-hydroxylmethyl moiety of adenophostin A is not
essential for biological activity and that, indeed, this part of
From these results, it may be conjectured that the 5′-aromatic
moiety does indeed exist outside the binding pocket during
interaction of these analogues 5a-c with the ligand binding
domain of the receptor. Therefore, adenophostin A analogues,
which are useful as biological tools for labeling target biomol-
ecules potentially could be developed by attaching a functional
aromatic residue, such as a fluorescent or photoactive aromatic

5′-Modified Adenophostin A Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5755
5-Deoxy-3-O-[2,6-di-O-benzyl-3,4-O-[(2S,3S)-2,3-dimethoxy-
butan-2,3-diyl]-R-D-glucopyranosyl]-1,2-O-isopropylidene-5-
phenethyl-R-D-ribofuranose (15c). Compound 15c (47 mg, 64%)
was prepared from 14c (75 mg, 0.10 mmol) by the procedure
described for the synthesis of 15a:11 1H NMR (CDCl3, 400 MHz)
δ 7.42-7.13 (m, 15H), 5.73 (d, 1H, J ) 3.8), 5.20 (d, 1H, J )
3.9), 4.79, 4.75 (each d, each 1H, J ) 12.4), 4.67 (dd, 1H, J ) 3.9,
3.9), 4.54 (s, 2H), 4.16-4.09 (m, 2H), 3.83-3.75 (m, 2H), 3.70-
3.60 (m, 4H), 3.30, 3.20 (each s, each 3H), 2.64 (m, 2H), 1.88-
1.73 (m, 2H), 1.52, 1.40 (each s, each 3H), 1.25, 0.92 (each m,
each 1H), 1.32 (s, 6H); FAB-HRMS calcd for C42H54O11Na,
757.3564; found, 757.3560 (MNa+).
the molecule will tolerate bulky substitutions, which can be
effective for conjugating a functional residue.
Experimental Section
General Methods. Chemical shifts are reported in ppm down-
field from tetramethylsilane (1H and 13C) or H3PO4 (31P), and J
values are given in hertz. The 1H NMR assignments described were
in agreement with COSY spectra. Thin-layer chromatography was
done on Merck coated plates, 60F254. Silica gel chromatography
was carried out using Merck silica gel 7734 or 9385. Reactions
were carried out under an argon atmosphere. Compounds 5a, 11,
and 13 were synthesized as described previously.11
3-O-[2,6-Di-O-benzyl-3,4-O-[(2S,3S)-2,3-dimethoxybutan-2,3-
diyl]-R-D-glucopyranosyl]-5-deoxy-5-iodo-1,2-O-isopropylidene-
R-D-ribofuranose (18). A mixture of 13 (64 mg, 0.10 mmol),
imidazole (27 mg, 0.40 mmol), Ph3P (52 mg, 0.20 mmol), and I2
(25 mg, 0.20 mmol) in toluene (2 mL) was heated under reflux for
15 min. After addition of aqueous saturated NaHCO3, I2 was added
to the resulting mixture to turn the aqueous layer brown, and then
aqueous saturated Na2S2O3 and toluene were added and partitioned.
The organic layer was washed with H2O and brine, dried (MgSO4),
and evaporated. The resulting residue was purified by column
chromatography (SiO2, hexane/AcOEt, 9:1) to give 18 (70 mg, 93%
as a form): 1H NMR (CDCl3, 400 MHz) δ 7.42-7.23 (m, 10H),
5.79 (d, 1H, J ) 3.8), 5.16 (d, 1H, J ) 4.1), 4.80 (d, 1H, J )
12.2), 4.76-4.71 (m, 2H), 4.60, 4.54 (each d, each 1H, J ) 12.0),
4.11 (dd, 1H, J ) 9.9, 9.9), 3.91-3.85 (m, 4H), 3.82 (m, 2H),
3.65 (dd, 1H, J ) 4.0, 9.9), 3.57 (dd, 1H, J ) 2.8, 11.3), 3.41 (dd,
1H, J ) 4.0, 11.3), 3.32, 3.19 (each s, each 3H), 1.55, 1.30 (each
s, each 3H,), 1.35 (s, 6H); FAB-HRMS calcd for C34H45O11INa
779.1905, found 779.1894 (MNa+).
3-O-[2,6-Di-O-benzyl-3,4-O-[(2S,3S)-2,3-dimethoxybutan-2,3-
diyl]-R-D-glucopyranosyl]-5-deoxy-1,2-O-isopropylidene-5-O-tri-
fluromethanesulfonyl-R-D-ribofuranose (19). To a mixture of 13
(100 mg, 0.15 mmol) and DTBMP (80 mg, 0.38 mmol) in CH2Cl2
(1.5 mL) was slowly added a solution of Tf2O (28 µL, 0.17 mmol)
in CH2Cl2 (1 mL) at -78 °C, and the mixture was stirred at the
same temperature for 1 h. The mixture was partitioned between
Et2O and aqueous saturated NaHCO3, and the organic layer was
dried (MgSO4) and evaporated. The resulting residue was purified
by column chromatography (SiO2, hexane/Et2O, 3:1) to give 19
(105 mg, 90% as a solid): 1H NMR (CDCl3, 270 MHz) δ 7.41-
7.24 (m, 10H), 5.75 (d, 1H, J ) 3.6), 5.08 (d, 1H, J ) 4.0), 4.82
(d, 1H, J ) 12.2), 4.77-4.63 (m, 4H), 4.57, 4.50 (each d, each
1H, J ) 12.2), 4.36 (m, 1H), 4.11 (dd, 1H, J ) 9.6, 9.6), 4.00 (dd,
1H, J ) 4.2, 9.2), 3.86-3.58 (m, 5H), 3.32, 3.18 (each s, each
3H), 1.54, 1.29 (each s, each 3H), 1.34 (s, 6H); FAB-HRMS calcd
for C35H45F3O14SNa, 801.2380; found, 801.2393 (MNa+). Anal.
(C35H45F3O14S): C, H, S.
General Procedure for the Glycosidation with Donor 11 and
Acceptor 12 (Table 1). A mixture of 11 (58 mg, 0.10 mmol), 12
(34 mg, 0.10 mmol), and DTBMP (103 mg, 0.50 mmol) was
azeotroped with toluene (3×). To the mixture of the resulting
residue and MS4A (powder, 100 mg) in CH2Cl2 was slowly added
Tf2O (34 µL, 0.20 mmol) at -78 °C, and the mixture was stirred
under the conditions indicated in Table 1. The mixture was poured
into aqueous saturated NaHCO3 and filtered through Celite, and
the filtrate was partitioned between AcOEt and aqueous saturated
NaHCO3. The organic layer was washed with brine, dried (Na2-
SO4), and evaporated. The resulting residue was purified by flash
column chromatography (SiO2, hexane/AcOEt, 50:1) to give an R/â-
mixture of glycosidation products 10R and 10â (entries 1-3) or
the R-anomer 10R (entry 4): 1H NMR (R/â-mixture, CDCl3, 270
MHz) for R-anomer δ 7.36-7.20 (m, 10H, Ph × 2), 5.64 (d, 1H,
H-1, J ) 3.3), 5.16 (d, 1H, H-1′, J ) 3.9), 4.79-4.46 (m, 5H,
OCH2Ph × 2, H-2), 4.20-4.04 (m, 3H, H-3, 4, 3′), 3.89-3.54 (m,
7H, H-5a, 5b, 2′, 4′, 5′, 6′a, 6′b), 3.28, 3.15 (each s, each 3H, OCH3),
1.49, 1.25 (each s, each 3H, Me), 1.31 (s, 6H, isopropylidene), 0.82
(s, 9H, t-Bu), 0.013, 0.00 (each s, each 3H, Si-CH3); for â-anomer
δ 7.38-7.17 (m, 10H, Ph × 2), 5.69 (d, 1H, H-1, J ) 3.6), 4.80
(m, 2H, OCH2Ph × 2), 4.67-4.62 (m, 2H, H-2, OCH2Ph), 4.55
(d, 1H, OCH2Ph, J ) 12.2), 4.46 (d, 1H, H-1′, J ) 7.3), 4.06 (m,
1H, H-3), 3.85 (m, 1H, H-4′), 3.80-3.64 (m, 7H, H-4, 5a, 5b, 3′,
5′, 6′a, 6′b), 3.53 (dd, 1H, H-2′, J ) 7.3, 8.9), 3.29, 3.17 (each s,
each 3H, OCH3), 1.33, 1.26 (each s, each 3H, Me), 1.32, 1.28 (each
s, each 3H, isopropylidene), 0.85 (s, 9H, t-Bu), 0.00 (s, 6H, Si-
CH3). 10R: [R]22.9D +142.4 (c 1.15, CHCl3); FAB-LRMS m/z 783
(MNa+). Anal. (C40H60O12Si): C, H.
5-Benzyl-5-deoxy-3-O-[2,6-di-O-benzyl-3,4-O-[(2S,3S)-2,3-
dimethoxybutan-2,3-diyl]-R-D-glucopyranosyl]-1,2-O-isopropyl-
idene-R-D-ribofuranose (15b) via the Barton Deoxygenation.
Compound 14b (145 mg, 40% as a solid), prepared from 13 (350
mg, 0.54 mmol) by the procedure described for the synthesis of
14a11 using BnMgCl (1.0 M in THF) instead of PhMgBr, was used
immediately to the next step because of the instability. Thus,
treatment of the obtained 14b (145 mg) by the procedure for the
synthesis of 15a11 gave 15b (13 mg, 18% as a solid): 1H NMR
(CDCl3, 400 MHz) δ 7.41-7.16 (m, 15H), 5.77 (d, 1H, J ) 3.8),
5.22 (d, 1H, J ) 3.8), 4.77 (s, 2H), 4.70 (dd, 1H, J ) 3.8, 4.3),
4.54, 4.50 (each d, each 1H, J ) 12.2), 4.17 (ddd, 1H, H-4, J )
3.6, 3.6, 9.2), 4.14 (dd, 1H, H-3′, J ) 9.2, 9.2), 3.84-3.74 (m,
2H), 3.70 (dd, 1H, J ) 4.3, 9.2), 3.66-3.32 (m, 3H), 3.28, 3.17
(each s, each 3H), 2.88, 2.72 (each m, each 1H), 2.06, 1.76 (each
m, each 1H), 1.52, 1.29 (each s, each 3H), 1.33 (s, 6H); FAB-
HRMS calcd for C41H52O11Na, 743.3408; found, 743.3431 (MNa+).
3-O-[2,6-Di-O-benzyl-3,4-O-[(2S,3S)-2,3-dimethoxybutan-2,3-
diyl]-R-D-glucopyranosyl]-1,2-O-isopropylidene-5-phenethyl-R-
D-ribofuranose (14c). Compound 14c (118 mg, 58% as a solid)
was prepared from 13 (175 mg, 0.27 mmol) by the procedure
described for the synthesis of 14a,11 using PhCH2CH2MgCl (1.0
M in THF) instead of BnMgCl: 1H NMR (CDCl3, 400 MHz) δ
7.52-7.14 (m, 15H), 5.76 (d, 1H, J ) 3.8), 5.13 (d, 1H, J ) 4.1),
4.80, 4.72 (each d, each 1H, J ) 12.2), 4.68 (dd, 1H, J ) 3.9, 4.2),
4.56 (m, 2H), 4.19 (dd, 1H, J ) 4.2, 9.2), 4.15-4.09 (m, 3H),
3.83 (m, 1H), 3.75-3.58 (m, 4H), 3.30, 3.19 (each s, each 3H),
2.85, 2.69 (each m, each 1H), 2.36 (m, 1H), 2.05-1.82 (m, 2H),
1.59, 1.53 (each s, each 3H), 1.33, 1.28 (each s, each 3H); FAB-
LRMS m/z 773 (MNa). Anal. (C42H54O12‚1.5H2O): C, H.
Compound 15b from 19. After stirring a mixture of CuI (662
mg, 3.47 mmol) and BnMgCl (1.0 M in THF, 7.0 mL, 7.0 mmol)
in HMPA (8 mL) and THF (17 mL) at -50 °C for 1 h, a solution
of 19 (846 mg, 1.08 mmol) in THF (10 mL) was slowly added.
The resulting mixture was stirred at the same temperature for 2 h,
and then aqueous saturated NH4Cl and Et2O were added. The
mixture was partitioned, and the organic layer was washed with
brine, dried (Na2SO4), and evaporated. The resulting residue was
purified by column chromatography (SiO2, hexane/AcOEt, 10:1)
to give 15b (515 mg, 66%) and 5-chloro-5-deoxy-3-O-[2,6-di-O-
benzyl-3,4-O-[(2S,3S)-2,3-dimethoxbutan-2,3-diyl]-R-D-glucopy-
ranosyl]-1,2-O-isopropylidene-R-D-ribofuranose (20): [R]20.7
D
1
+154.1 (c 1.01, CHCl3); H NMR (CDCl3, 400 MHz) δ 7.32-
7.23 (m, 10H), 5.79 (d, 1H, J ) 3.6), 5.16 (d, 1H, J ) 4.1), 4.80,
4.74 (each d, each 1H, J ) 12.5), 4.71 (m, 1H), 4.58, 4.52 (each
d, each 1H, J ) 12.1), 4.42 (ddd, 1H, J ) 3.2, 3.2, 11.6), 4.15-
4.09 (m, 2H), 3.88 (ddd, 1H, J ) 2.0, 12.6, 12.6), 3.84-3.70 (m,
5H), 3.65 (dd, 1H, J ) 3.9, 9.9), 3.32, 3.18 (each s, each 3H),
1.55, 1.30 (each s, each 3H), 1.35 (s, 6H); FAB-HRMS calcd for
C34H45ClO11Na, 687.2548; found, 687.2568 (MNa+).
5-Benzyl-5-deoxy-3-O-(2,6-di-O-benzyl-3,4-di-O-isobutyryl-R-
D-glucopyranosyl)-1,2-di-O-isobutyryl-D-ribofuranose (9b). Com-

5756 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Mochizuki et al.
pound 9b (40 mg, 69% as a solid) was prepared from 15b (50 mg,
0.068 mmol) by the procedure described for the synthesis of 9a:11
1H NMR (CDCl3, 400 MHz) δ 7.34-7.12 (m, 15H), 6.47 (d, 0.15H,
J1, 2 ) 4.6), 6.13 (s, 0.85H), 5.42 (m, 1H), 5.32 (d, 1H, J ) 4.3),
5.04 (m, 1H), 4.91 (d, 0.85H, J ) 3.6), 4.74 (d, 0.15H, J ) 3.2),
4.57-4.40 (m, 4H), 4.30 (m, 1H), 4.21(m, 1H), 3.89 (m, 1H), 3.55
(dd, 1H, J ) 3.4, 9.9), 3.30 (m, 2H), 2.78, 2.64 (each m, each 1H),
2.55-2.34 (m, 4H), 2.07, 1.81 (each m, each 1H), 1.28-0.97 (m,
24H); FAB-HRMS calcd for C48H62O13Na, 869.4088; found,
869.4083 (MNa+).
12.6), 4.30 (m, 1H), 4.03 (m, 1H), 3.69-3.63 (m, 2H), 3.51 (m,
1H), 3.46-3.29 (m, 2H), 3.25 (dd, 1H, J ) 3.2, 9.8), 2.57 (m,
2H), 1.98 (m, 2H); UV (MeOH) λmax 261 nm; FAB-HRMS calcd
for C37H42N5O8, 684.3034; found, 684.3044 (MH+).
9-[5-Deoxy-3-O-(2,6-di-O-benzyl-R-D-glucopyranosyl)-5-phen-
ethyl-â-D-ribofuranosyl]adenine (16c). Compound 16c (13 mg,
quant. as a solid) was prepared from 7c (19 mg, 0.019 mmol) by
the procedure described for the synthesis of 16a:11 1H NMR
(DMSO-d6, 400 MHz) δ 8.31 (s, 1H), 8.12 (s, 1H),7.56-7.07 (m,
17H), 5.89 (d, 1H, J ) 4.4), 5.28 (m, 1H), 5.23 (br s, 1H), 4.94
(m, 1H), 4.78, 4.61 (each d, each 1H, J ) 11.4), 4.44 (m, 2H),
4.23 (m, 1H), 4.03 (m, 1H), 3.68-3.64 (m, 2H), 3.52-3.24 (m,
5H), 3.14 (dd, 1H, J ) 8.5, 8.5), 2.48 (m, 2H), 1.67-1.22 (m,
4H); UV (MeOH) λmax 259 nm; FAB-HRMS calcd for C38H43N5O8-
Na, 720.3009; found, 720.2998 (MNa+).
5′-Deoxy-5′-benzyladenophostin A Trisphosphate Derivative
17b. Compound 17b (75 mg, 70% as a solid) was prepared from
16b (60 mg, 0.088 mmol) by the procedure described for the
synthesis of 17a:11 1H NMR (CDCl3, 400 MHz) δ 8.21 (s, 1H),
7.76 (s, 1H), 7.46-6.98 (m, 29H), 6.17 (d, 1H, J ) 5.8), 5.96 (m,
1H), 5.67 (s, 2H), 5.41 (m, 2H), 5.32 (d, 1H, J ) 3.6), 5.19-4.76
(m, 12H), 4.64, 4.52 (each d, each 1H, J ) 12.0), 4.57 (dd, 1H, J
) 4.0, 4.0), 4.23 (m, 1H), 3.95 (m, 1H), 3.71-3.67 (m, 2H), 3.64
(dd, 1H, J ) 3.6, 9.8), 2.72, 2.57 (each m, each 1H), 2.18, 2.02
(each m, each 1H); FAB-HRMS calcd for C61H63N5O17P3, 1230.3432;
found, 1230.3445 (MH+).
5′-Deoxy-5′-phenethyladenophostin A Trisphosphate Deriva-
tive 17c. Compound 17c (31 mg, quant. as a solid) was prepared
from 16c (17 mg, 0.024 mmol) by the procedure described for the
synthesis of 17a:11 1H NMR (CDCl3, 400 MHz) δ 8.09 (s, 1H),
7.67 (s, 1H), 7.38-6.99 (m, 29H), 6.07 (d, 1H, J ) 5.8), 5.82 (dd,
1H, J ) 5.8, 12.6), 5.42-5.27 (m, 2H), 5.23 (d, 1H, J ) 3.6),
5.11-4.67 (m, 16H), 4.45 (m, 1H), 4.22 (m, 1H), 3.92 (m, 1H),
3.69-3.61 (m, 2H), 3.60 (dd, 1H, J ) 3.6, 9.8), 2.47 (m, 2H),
1.81-1.51 (m, 4H); FAB-HRMS calcd for C62H65N5O17P3,
1244.3589; found, 1244.3560 (MH+).
5′-Deoxy-5′-benzyladenophostin A 2′,3′′,4′′-Trisphosphate
(5b). Compound 5b (35 mg, quant. as a solid) was prepared from
17b (54 mg, 0.044 mmol) by the procedure described for the
synthesis of 5a:11 1H NMR (D2O, 400 MHz) δ 8.34 (s, 1H, H-2),
8.31 (s, 1H, H-8), 7.38-7.24 (m, 5H, Ph), 6.31 (m, 1H, H-1′), 5.38
(m, 1H, H-2′), 5.29 (m, 1H, H-1′′), 4.67 (m, 1H, H-4′), 4.51 (m,
1H, H-3′), 4.38 (m, 1H, H-3′′), 4.09-4.05 (m, 2H, H-4′′, H-5′′),
3.77 (m, 1H, H-2′′), 3.67 (m, 1H, H-6′′a), 3.57 (m, 1H, H-6′′b),
3.20 (m, 2H, H-6′a,b), 2.71 (m, 2H, H-5′a,b); 31P NMR (CDCl3,
202 MHz, H-decoupled) δ -3.38, -3.60, -3.97; UV (H2O) λmax
259 nm; FAB-HRMS calcd for C23H31N5O17P3, 742.0928; found,
742.0905 [(M - H)-].
5′-Deoxy-5′-phenethyladenophostin A 2′,3′′,4′′-Trisphosphate
(5c). Compound 5c (8 mg, 43% as a solid) was prepared from 20c
(31 mg, 0.025 mmol) by the procedure described for the synthesis
of 5a:11 1H NMR (D2O, 400 MHz) δ 8.34 (s, 1H, H-2), 8.31 (s,
1H, H-8), 7.38-7.24 (m, 5H, Ph), 6.31 (m, 1H, H-1′), 5.38 (m,
1H, H-2′), 5.29 (m, 1H, H-1′′), 4.67 (m, 1H, H-4′), 4.51 (m, 1H,
H-3′), 4.38 (m, 1H, H-3′′), 4.09-4.05 (m, 2H, H-4′′, H-5′′), 3.77
(m, 1H, H-2′′), 3.67 (m, 1H, H-6′′a), 3.57 (m, 1H, H-6′′b), 3.20
(m, 2H, H-6′a,b), 2.71 (m, 2H, H-5′a,b); 31P NMR (D2O, 202 MHz,
H-decoupled) δ -2.00, -3.29, -3.88; UV (H2O) λmax 259 nm;
FAB-HRMS calcd for C24H33N5O17P3, 756.1083; found, 756.1084
[(M - H)-].
(2R,3R,4S)-4-Acetoxy-2-benzyl-tetrahydrofuran-3-yl 3,4-di-
O-acetyl-2,6-di-O-benzyl-R-D-glucopyranoside (21) and (2R,3R,4S)-
2-Benzyl-4-hydroxytetrahydrofuran-3-yl 3,4-di-O-acetyl-2,6-di-
O-benzyl-R-D-glucopyranoside (22). To a mixture of 8a (20 mg,
0.028 mmol), Et3SiH (13 µL, 0.056 mmol) in CH2Cl2 (0.5 mL)
was slowly added TMSOTf (10 µL, 0.056 mmol) at 0 °C, and the
mixture was stirred at the same temperature for 20 min. The mixture
was partitioned between CHCl3 and aqueous saturated NaHCO3,
and the organic layer was washed with H2O and brine, dried
(MgSO4), and evaporated. The resulting residue was purified by
column chromatography (SiO2, hexane/AcOEt, 5:1-3:1) to give
5-Deoxy-3-O-(2,6-di-O-benzyl-3,4-di-O-isobutyryl-R-D-glu-
copyranosyl)-1,2-di-O-isobutyryl-5-phenethyl-D-ribofuranose (9c).
Compound 9c (41 mg, 62% as a solid) was prepared from 15c (57
mg, 0.077 mmol) by the procedure described for the synthesis of
9a:11 1H NMR (CDCl3, 400 MHz) δ 7.34-7.11 (m, 15H), 6.42 (d,
0.2H, J ) 3.6), 6.02 (s, 0.8H), 5.28 (m, 1H), 5.08 (d, 0.8H, J )
9.9), 5.02 (m, 1H), 4.93 (m, 1H), 4.74 (d, 0.20H, J ) 3.3), 4.57-
4.39 (m, 4H), 4.23 (m, 1H), 4.15 (m, 1H), 3.86 (m, 1H), 3.55 (dd,
1H, J ) 3.3, 9.9), 3.35 (m, 2H), 2.60 (t, 2H, J ) 6.4), 2.54-2.31
(m, 4H), 1.85-1.46 (m, 4H), 1.13-0.98 (m, 24H); FAB-HRMS
calcd for C49H64O13Na, 883.4244; found, 883.4274 (MNa+).
N6-Benzoyl-9-[5-benzyl-5-deoxy-3-O-(2,6-di-O-benzyl-3,4-di-
O-isobutyryl-R-D-glucopyranosyl)-2-O-isobutyryl-â-D-ribofur-
anosyl]adenine (7b). Compound 7b (169 mg, 67% as a solid) was
prepared from 9b (200 mg, 0.24 mmol) by the procedure described
for the synthesis of 7a (Table 2, entry 5):11 [R]20.7D +35.5 (c 1.58,
CHCl3); 1H NMR (CDCl3, 400 MHz) δ 8.99 (br s, 1H, NH), 8.80
(s, 1H, H-2), 8.10 (s, 1H, H-8), 8.30 (m, 2H, o-Bz), 7.61 (m, 3H,
m-, p-Bz), 7.32-7.11 (m, 15H, Ph × 3), 6.19 (d, 1H, H-1′, J )
6.4), 5.99 (dd, 1H, H-2′, J ) 5.3, 6.4), 5.49 (dd, 1H, H-3′′, J )
9.8, 10.2), 5.09 (dd, 1H, H-4′′, J ) 9.8, 9.8), 4.92 (d, 1H, H-1′′, J
) 3.2), 4.59-4.49 (m, 4H, H-3′, OCH2Ph × 3), 4.43 (d, 1H, OCH2-
Ph, J ) 11.1), 4.37 (m, 1H, H-4′), 3.97 (ddd, 1H, H-5′′, J ) 2.0,
2.0, 10.2), 3.58 (dd, 1H, H-2′′, J ) 3.6, 10.2), 3.36 (dd, 1H, H-6′′a,
J ) 4.5, 10.2), 3.30 (dd, 1H, H-6′′b, J ) 2.0, 10.2), 2.27, 2.62
(each m, each 1H, H-6′a, 6′b), 2.51-2.37 (m, 3H, COCH(CH3)2
× 3), 2.19, 2.10 (each m, each 1H, H-5′a, 5′b), 1.26 (t, 3H, COCH-
(CH3)2), 1.21-1.00 (m, 15H, COCH(CH3)2 × 3); 13C NMR (CDCl3,
100 MHz) δ 155.9, 152.6, 149.5, 144.3, 141.3, 139.7, 138.4, 138.4,
131.5, 128.1, 128.1, 128.0, 127.9, 127.2, 127.1, 125.6, 119.1, 110.3,
96.1, 95.3, 87.0, 81.9, 78.9, 77.3, 72.3, 72.1, 71.8, 71.3, 70.3, 69.7,
41.0, 40.1, 39.9, 35.2, 32.4, 31.2, 29.8, 28.4, 23.3, 22.4, 21.2, 18.9,
16.36, 13.9, 11.2, 10.8.
N6-Benzoyl-9-[5-deoxy-3-O-(2,6-di-O-benzyl-3,4-di-O-isobu-
tyryl-R-D-glucopyranosyl)-2-O-isobutyryl-5-phenethyl-â-D-ribo-
furanosyl]adenine (7c). Compound 7c (30 mg, 60% as a solid)
was prepared from 9c (110 mg, 0.11 mmol) by the procedure
described for the synthesis of 7a (Table 2, entry 5):11 1H NMR
(CDCl3, 400 MHz) δ 9.02 (br s, 1H, NH), 8.75 (s, 1H, H-2), 8.06
(s, 1H, H-8), 8.03 (m, 2H, o-Bz), 7.53 (m, 3H, m-, p-Bz), 7.34-
7.09 (m, 15H, Ph × 3), 6.14 (d, 1H, H-1′, J ) 5.0), 5.91 (dd, 1H,
H-2′, J ) 5.0, 5.0), 5.49 (dd, 1H, H-3′′, J ) 9.8, 9.8), 5.09 (dd,
1H, H-4′′, J ) 9.8, 9.8), 4.90 (d, 1H, H-1′′, J ) 3.6), 4.59-4.28
(m, 6H, H-3′, -4′, OCH2Ph × 4), 3.99 (m, 1H, H-5′′), 3.58 (dd,
1H, H-2′′, J2′′, 1′′ ) 3.6, J2′′, 3′′ ) 9.8), 3.38 (m, 2H, H-6′′a, 6′′b),
2.59 (t, 2H, H-7′, J ) 7.6), 2.51-2.41 (m, 3H, COCH(CH3)2 × 3),
1.82-1.60 (m, 4H, H-5′a,b, 6′a,b), 1.12-1.02 (m, 18H, COCH-
(CH3)2 × 3); 13C NMR (CDCl3, 125 MHz) δ 176.0, 175.4, 152.6,
151.3, 149.5, 142.0, 133.5, 132.8, 132.3, 130.8, 128.8, 128.7,
128.40, 128.3, 128.3, 128.0, 127.8, 127.7, 125.8, 123.6, 98.3, 87.8,
83.3, 79.6, 77.6, 73.6, 73.4, 73.2, 71.2, 69.7, 68.5, 67.92, 35.6, 34.1,
33.9, 33.7, 19.1, 18.9, 18.8; FAB-HRMS calcd for C57H66N5O12,
1012.4708; found, 1012.4750 (MH+).
9-[5-Benzyl-5-deoxy-3-O-(2,6-di-O-benzyl-R-D-glucopyrano-
syl)-â-D-ribo-pentofuranosyl] adenine (16b). Compound 16b (72
mg, 96% as a solid) was prepared from 7b (110 mg, 0.11 mmol)
by the procedure described for the synthesis of 16a:11 [R]202.3
D
1
+50.1 (c 0.57, CH30H); H NMR (DMSO-d6, 400 MHz) δ 8.36
(s, 1H), 8.14 (s, 1H), 7.41-7.10 (m, 17H), 5.93 (d, 1H, J ) 6.4),
5.31 (d, 1H, J ) 3.2), 5.29 (d, 1H, J ) 5.8), 5.01 (m, 1H), 4.79,
4.60 (each d, each 1H, J ) 11.4), 4.49, 4.45 (each d, each 1H, J )

5′-Modified Adenophostin A Analogues
Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19 5757
21 (2 mg, 8% as a foam) and 22 (11 mg, 60% as a foam). 21: 1H
NMR (CDCl3, 400 MHz) δ 7.33-7.21 (m, 15H), 5.42 (dd, 1H, J
) 9.5, 9.5), 5.25 (m, 1H), 5.03 (dd, 1H, J ) 9.5, 9.5), 5.00 (d, 1H,
J ) 3.2), 4.64, 4.56, 4.49, 4.44 (each d, each 1H, J ) 12.1), 4.63
(m, 1H), 4.08 (dd, 1H, J ) 5.1, 10.1), 4.00 (dd, 1H, J ) 5.3, 6.9),
3.88 (ddd, 1H, J ) 3.4, 6.9, 10.1), 3.81 (dd, 1H, J ) 3.7, 10.1),
3.56 (dd, 1H, J ) 3.6, 10.1), 3.41 (m, 2H), 3.07 (dd, 1H, J ) 4.0,
14.1), 2.75 (dd, 1H, J ) 8.1, 14.1), 2.00, 1.90, 1.89 (each s, each
3H); FAB-HRMS calcd for C37H43O11, 663.2806; found, 663.2831
(MH+). 22: 1H NMR (CDCl3, 400 MHz) δ 7.36-7.17 (m, 15H),
5.42 (dd, 1H, J ) 9.8, 9.8), 5.05 (dd, 1H, J ) 9.8, 9.8), 4.80 (d,
1H, J ) 3.2), 4.67, 4.58, 4.56, 4.42 (each d, each 1H, J ) 12.1),
4.18 (m, 1H), 4.07 (m, 1H), 4.98 (dd, 1H, J ) 5.1, 10.1), 3.83-
3.80 (m, 2H), 3.69 (dd, 1H, J ) 3.7, 10.1), 3.64-3.61 (m, 2H),
3.37 (m, 2H), 3.36 (dd, 1H, J ) 4.0, 14.1), 2.85 (dd, 1H, J ) 8.1,
14.1), 2.00, 1.90, (each s, each 3H); FAB-HRMS calcd for
C5H41O10, 621.2700; found, 621.2710 (MH+).
Acknowledgment. We wish to acknowledge financial sup-
port from Ministry of Education, Culture, Sports, Science and
Technology, Japan for Scientific Research 17659027 and
17390027 (to S.S.), the Japan Society for Promotion of Science
for Creative Scientific Research 13NP0401 (to A.M.), and the
Wellcome Trust for Program Grants 039662 (to C.W.T.) and
060554 (to B.V.L.P.).
Supporting Information Available: Purities of compounds by
combustion analysis and HPLC (1H NMR charts of 5a, 5b, 5c,
and 6.
References
(1) This report constitutes Part 243 of Nucleosides and Nucleotides: For
Part 242: Inoue, N.; Minakawa, N.; Matsuda, A. Synthesis and
properties of 4′-thioDNA: unexpected RNA-like behavior of 4′-
thioDNA. Nucleic Acids Res. 2006, 34, 3476-3483.
(2) Wilcox, R. A.; Primrose, W. U.; Nahorski, S. R.; Challiss, R. A. J.
New developments in the molecular pharmacology of the myo-inositol
1,4,5-trisphosphate receptor. Trends Pharmacol. Sci. 1998, 19, 467-
475.
(2R,3R,4S)-2-Benzyl-tetrahydrofuran-3-yl 2,6-Di-O-benzyl-R-
D-glucopyranoside (23). Compound 23 (7 mg, 86% as a solid)
was prepared from 22 (10 mg, 11 µmol) by the procedure described
for the synthesis of 16a:11 1H NMR (DMSO-d6, 400 MHz) δ 7.38-
7.14 (m, 15H), 5.14 (br s, 1H), 5.07 (d, 1H, J ) 3.4), 4.73, 4.66
(each d, each 1H, J ) 11.7), 4.49, 4.46 (each d, each 1H, J )
12.6), 4.33 (d, 1H), 4.12 (m, 1H), 3.95 (m, 1H), 3.84 (dd, 1H, J )
5.1, 9.3), 3.79 (dd, 1H, J ) 5.7, 5.7), 3.64-3.57 (m, 3H), 3.52-
3.46 (m, 3H), 3.22 (dd, 1H, J ) 3.4, 9.4), 3.14 (dd, 1H, H-4, J )
9.3, 9.3), 2.90 (dd, 1H, J ) 3.7, 13.9), 2.63 (dd, 1H, J ) 7.9, 13.9);
FAB-HRMS calcd for C31H37O8, 537.2487; found, 537.2473
(MH+).
Trisphosphate Derivative 24. Compound 24 (38 mg, 85% as a
solid) was prepared from 23 (22 mg, 0.040 mmol) by the procedure
described for the synthesis of 17a:11 1H NMR (CDCl3, 400 MHz)
δ 7.42-7.11 (m, 27H), 5.47-5.32 (m, 2H), 5.24-4.74 (m, 15H),
4.66, 4.60, 4.53 (each d, each 1H, J ) 11.9), 4.32 (m, 1H), 4.06
(dd, 1H, J ) 3.4, 10.5), 4.00 (dd, 1H, J ) 4.2, 10.5), 3.93 (m, 1H),
3.82 (m, 1H), 3.76-3.67 (m, 2H), 3.64 (dd, 1H, J ) 3.6, 9.8),
3.05 (dd, 1H, J ) 10.0, 14.2), 2.76 (dd, 1H, J ) 3.5, 14.2); FAB-
HRMS calcd for C55H58O17P3, 1083.2887; found, 1083.2880 (MH+).
(2R,3R,4S)-2-Benzyl-4-hydroxytetrahydrofuran-3-yl 3,4-Di-
R-D-glucopyranoside Trisphosphate (6). Compound 6 (15 mg,
87% as a solid) was prepared from 24 (31 mg, 0.029 mmol) by the
procedure described for the synthesis of 5a:11 1H NMR (D2O, 400
MHz) δ 7.30-7.19 (m, 5H, Ph), 5.04 (d, 1H, H-1, J ) 3.6), 4.76
(m, 1H, H-4′), 4.31 (m, 1H, H-2′), 4.32 (m, 1H, H-3), 4.07 (dd,
1H, H-3′, J ) 5.3, 5.3), 4.01-3.89 (m, 3H, H-4, H-5′a,b), 3.61-
3.55 (m, 2H, H-2, H-6a), 3.49 (m, 1H, H-6b), 3.06 (m, 1H, H-5),
2.92 (m, 2H, CH2Ph); 31P NMR (D2O, 202 MHz, H-decoupled) δ
0.51, 0.35, 0.02; FAB-HRMS calcd for C17H26 O17P3, 595.0377;
found, 595.0380 [(M - H)-].
Biological Methods. The effects of compounds on intracellular
Ca2+ stores of a stable cell line expressing rat type 1 IP3 receptor
(DT40-IP3R1 cells) were measured using a low-affinity Ca2+
indicator trapped within the intracellular stores of permeabilized
cells as described previously.24
(3) (a) Potter, B. V. L.; Lampe, D. Chemistry of inositol lipid mediated
cellular signaling. Angew. Chem., Int. Ed. Engl. 1995, 34, 1933-
1972. (b) Riley, A. M.; Morris, S. A.; Nerou, E. P.; Correa, V.; Potter,
B. V. L.; Taylor, C. W. Interactions of inositol 1,4,5-trisphosphate
(IP3) receptors with synthetic poly(ethylene glycol)-linked dimers of
IP3 suggest close spacing of the IP3-binding sites. J. Biol. Chem.
2002, 277, 40290-40295. (c) Riley, A. M.; Laude, A. J.; Taylor, C.
W.; Potter, B. V. L. Dimers of D-myo-inositol 1,4,5-trisphosphate:
design, synthesis, and interaction with Ins(1,4,5)P3 receptors. Bio-
conjugate Chem. 2004, 15, 278-289 and references therein.
(4) (a) Takahashi, M.; Kagasaki, T.; Hosoya, T.; Takahashi, S. Adeno-
phostins A and B: potent agonists of inositol 1,4,5-trisphosphate
receptor produced by Penicillium breVicompactum. J. Antibiot. 1993,
46, 1643-1647. (b) Takahashi, M.; Tanzawa, K.; Takahashi, S.
Adenophostins, newly discovered metabolites of Penicillium breVi-
compactum, act as potent agonists of the inositol 1,4,5-trisphosphate
receptor. J. Biol. Chem. 1994, 269, 369-372. (c) Hirota, J.;
Michikawa, T.; Miyawaki, A.; Takahashi, M.; Tanzawa, K.; Okura,
I.; Furuichi, T.; Mikoshiba, K. Adenophostin-mediated quantal Ca2+
release in the purified and reconstituted inositol 1,4,5-trisphosphate
receptor type 1. FEBS Lett. 1995, 368, 248-252.
(5) Correa, V.; Nerou, E. P.; Riley, A. M.; Marwood, R. D.; Shuto, S.;
Rosenberg, H. J.; Horne, G.; Potter, B. V. L.; Taylor, C. W. Structural
determinants of adenophostin A activity at inositol trisphosphate
receptors. Mol. Pharmacol. 2001, 59, 1206-1215.
(6) (a) Tatani, K.; Shuto, S.; Ueno, Y.; Matsuda, A. Synthesis of 1-O-
[(3S,4R)-3-hydroxytetrahydrofuran-4-yl]-R-D-glucopyranoside 3,4,3′-
trisphosphate as a novel potent IP3 receptor ligand. Tetrahedron Lett.
1998, 39, 5065-5068. (b) Shuto, S.; Tatani, K.; Ueno, Y.; Matsuda,
A. Synthesis of adenophostin analogues lacking the adenine moiety
as novel potent IP3 receptor ligands: Some structural requirements
for the significant activity of adenophostin A. J. Org. Chem. 1998,
63, 8815-8824. (c) Kashiwayanagi, M.; Tatani, K.; Shuto, S.;
Matsuda, A. Inward current responses to IP3 and adenophostin-
analogues in turtle olfactory sensory neurons. Eur. J. Neurosci. 2000,
12, 606-612. (d) Abe, H.; Shuto, S.; Matsuda, A. Synthesis of the
C-glycosidic analogue of adenophostin A, a potent IP3 receptor
agonist, using a temporary silicon-tethered radical coupling reaction
as the key step. Tetrahedron Lett. 2000, 41, 2391-2394. (e) Abe,
H.; Shuto, S.; Matsuda, A. Synthesis of the C-glycosidic analogue
of adenophostin A and its uracil congener as potential IP3 receptor
ligands. Stereoselective construction of the C-glycosidic structure by
a temporary silicon-tethered radical coupling reaction. J. Org. Chem.
2000, 65, 4315-4325. (f) Shuto, S.; Yahiro, Y.; Ichikawa, S.;
Matsuda, A. Synthesis of 3,7-anhydro-D-glycero-D-ido-octitol 1,5,6-
trisphosphate as an IP3 receptor ligand using a radical cyclization
reaction with a vinylsilyl tether as the key step. Conformational
restriction strategy using steric repulsion between adjacent bulky
protecting groups on a pyranose ring. J. Org. Chem. 2000, 65, 5547-
5557. (g) Terauchi, M.; Yahiro, T.; Abe, H.; Ichikawa, S.; Tovey, S.
C.; Dedos, S. G.; Taylor, C. W.; Potter, B. V. L.; Matsuda, A.; Shuto,
S. Synthesis of 4,8-anhydro-D-glycero-D-ido-nonanitol 1,6,7-trispho-
sphate as a novel IP3 receptor ligand using a stereoselective radical
cyclization reaction based on a conformational restriction strategy.
Tetrahedron 2005, 61, 3697-3707.
Molecular Modeling. Ligand structures were built in MOE
starting from unconstrained cyclopentane. Constraints were added
to this ring to ensure that the final structure had the 2′-endo
conformation. The structure was built up to give adenophostin A,
minimizing at each stage (i.e., adding the glucose, then the
phosphates and finally the adenine ring) using the MMFF94 force
field. Once the adenophostin A molecule had been built, the 5′-
modified analogues were derived by adding substituents at the 5′
position using (CH2)nC6H5, where n ) 1, 2, 3. These structures
were then placed in the IP3 receptor (1N4K) by aligning the 3,4-
trans-bisphosphate on the glucose ring with the 4,5-trans-bispho-
sphate of IP3 from the crystal structure.27 Following removal of
IP3 from the structure, the heavy atoms of the receptor and
associated water molecules were fixed, and the structure was
allowed to minimize.
(7) (a) Rosenberg, H. J.; Riley, A. M.; Laude, A. J.; Taylor, C. W.; Potter,
B. V. L. Synthesis and Ca2+-mobilizing activity of purine-modified
mimics of adenophostin A: a model for the adenophostin-Ins(1,4,5)-
P3 receptor interaction. J. Med. Chem. 2003, 46, 4860-4871 and

5758 Journal of Medicinal Chemistry, 2006, Vol. 49, No. 19
Mochizuki et al.
references therein. (b) Wilcox, R. A.; Erneux, C.; Primrose, W. U.;
Gigg, R.; Nahorski, S. R. 2-Hydroxyethyl D-glucopyranoside 2,3′,4′-
trisphosphate, a novel, metabolically resistant, adenophostin A and
myo-inositol 1,4,5-trisphosphate analogue potently interacts with the
myo-inositol 1,4,5-trisphosphate receptor. Mol. Pharmacol. 1995, 47,
1201-1211. (c) de Kort, M.; Regenbogen, A. D.; Overkleeft, H. S.;
Challis, J.; Iwata, Y.; Miyamoto, S.; van der Marel, G. A.; van Boom,
J. Synthesis and biological evaluation of cyclophostin: a 5′,6′-tethered
analogue of adenophostin A. Tetrahedron 2000, 56, 5915-5928 and
references therein. (d) Roussel, F.; Moitessier, N.; Hilly, M.; Chre´tien,
F.; Mauger, J.-P.; Chapleur, Y. Bioorg. Med. Chem. 2002, 10, 759-
768 and references therein. (e) Hotoda, H.; Murayama, K.; Miyamoto,
S.; Iwata, Y.; Takahashi, M.; Kawase, Y.; Tanzawa, K.; Kaneko,
M. Molecular recognition of adenophostin, a very potent Ca2+
inducer, at the D-myo-inositol 1,4,5-trisphosphate receptor, Biochem-
istry 1999, 38, 9234-9241.
11882. (b) Tamura, S.; Abe, H.; Matsuda, A.; Shuto, S. Control of
the R/â-stereoselectivity in Lewis acid-promoted C-glycosidation
using a controlling anomeric effect based on the conformational
restriction strategy. Angew. Chem., Int. Ed. 2003, 42, 1021-1023.
(c) Abe, H.; Terauchi, M.; Matsuda, A.; Shuto, S. A study on the
conformation-anomeric effect-stereoselectivity relationship in ano-
meric radical reactions using conformationally restricted glucose
derivatives as substrates. J. Org. Chem. 2003, 68, 7439-7447.
(16) (a) Montchamp, J. L.; Tian, F.; Hart, M. E.; Frost, J. W. Butane
2,3-bisacetal protection of vicinal diequatorial diols. J. Org. Chem.
1996, 61, 3897-3899. (b) Hense, A.; Ley, S. V.; Osborn, H. M. I.;
Owen, D. R.; Poisson, J.-F.; Warriner, S. L.; Wesson, K. E. Direct
preparation of diacetals from 1,2-diketones and their use as 1,2-diol
protecting groups. J. Chem. Soc., Perkin Trans. 1 1997, 2023-2031.
(17) Parr, I. B.; Horenstein, B. A. New electronic analogues of the sialyl
cation: N-functionalized 4-acetamido-2,4-dihidroxypiperidines. In-
hibition of bacterial sialidases. J. Org. Chem. 1997, 62, 7489-7494.
(18) Nonstereoselective examples of Vorbru¨ggen glycosylation, see:
Azuma, T.; Isono, K. Transnucleosidation: An improved method for
transglycosylation from pyrimidines to purines. Chem. Pharm. Bull.
1977, 25, 3347-3353.
(19) Nomura, M.; Sato, T.; Washinosu, M.; Tanaka, M.; Asao, T.; Shuto,
S.; Matsuda, A. Practical large-scale synthesis of 1-(3-C-ethynyl-â-
ribo-pentofuranosyl)cytosine (ECyd), a potent antitumor nucleoside.
Isobutyryloxy group as an efficient anomeric leaving group in the
Vorbru¨ggen glycosylation reaction. Tetrahedron 2002, 58, 1279-
1288.
(20) Watanabe, Y.; Komoda, Y.; Ebisuya, K.; Ozaki, S. An efficient
phosphorylation method using a new phosphitylating agent, 2-di-
ethylamino-1,3,2-benzodioxaphosphepane. Tetrahedron Lett. 1990,
31, 255-256.
(21) Hayakawa, Y.; Kataoka, M. Facile synthesis of oligodeoxyribo-
nucleosides via the phosphoroamidite method without nucleoside base
protection. J. Am. Chem. Soc. 1998, 120, 12395-12401.
(22) A review of organocopper reagents: Bruce, H. L.; Sengupta, S.
Organocopper reagents: substitution, conjugate addition, carbo/
metallocupration, and other reactions. Org. React. 1992, 41, 135-
631.
(23) Kakefuda, A.; Shuto, S.; Nagahata, T.; Seki, J.; Sasaki, T.; Matsuda,
A. Synthesis and biological evaluations of ring-expanded oxetanocin
analogues: purine and pyrimidine analogues of 1,4-anhydro-2-deoxy-
D-arabitol and 1,4-anhydro-2-deoxy-3-hydroxymethyl-D-arabitol. Tet-
rahedron 1994, 50, 10167-10182.
(24) Laude, A. J.; Tovey, S. C.; Dedos, S. G.; Potter, B. V. L.; Lummis,
S. C. R.; Taylor, C. W. Rapid functional assays of recombinant IP3
receptors. Cell Calcium 2005, 38, 45-51.
(8) (a) Inoue, T.; Kikuchi, K.; Hirose, K.; Iino, M.; Nagano, T. Synthesis
and evaluation of 1-positon-modified inositol 1,4,5-trisphosphate
analogues. Bioorg. Med. Chem. Lett. 1999, 9, 1697-1702. (b) Inoue,
T.; Kikuchi, K.; Hirose, K.; Iino, M.; Nagano, T. Small molecule-
based laser inactivation of inositol 1,4,5-trisphosphate receptor. Chem.
Biol. 2000, 8, 9-15. (c) Nakanishi, W.; Kikuchi, K.; Inoue, T.;
Hirose, K.; Iino, M.; Nagano, T. Hydrophobic modifications at
1-phosphate of inositol 1,4,5-trisphosphate analogues enhance recep-
tor binding. Bioorg. Med. Chem. Lett. 2002, 12, 911-913.
(9) Davis, A. M.; Teague, S. J. Hydrogen bonding, hydrophobic
interactions and failure of the rigid receptor hypothesis. Angew.
Chem., Int. Ed. 1999, 38, 736-749.
(10) (a) Ahang, M.-Q.; Leurs, R.; Timmerman, H. Histamine H1-receptor
antagonists. In Burger’s medicinal chemistry and drug discoVery,
5th edition; Wolff, M. E., Eds.; John Wiley & Sons: New York,
1997; Vol. 5, pp 495-559. (b) Timmerman, H.; Smith, R. D. Major
classes of currently established antihypertensive drugs. In Burger’s
medicinal chemistry and drug discoVery, 5th edition; Wolff, M. E.,
Eds.; John Wiley & Sons: New York, 1997; Vol 2, pp 278-299.
(11) A preliminary account on the synthesis of 5a has appeared:
Mochizuki, T.; Kondo, Y.; Abe, H.; Taylor, C. W.; Potter, B. L. V.;
Matsuda, A.; Shuto, S. Design and synthesis of 5′-deoxy-5′-
phenyadenophostin A, a highly potent IP3 receptor ligand. Org. Lett.
2006, 8, 1455-1458.
(12) Niedballa, U.; Vorbru¨ggen, H. A general synthesis of N-glycosides.
I. Synthesis of pyrimidine nucleosides. J. Org. Chem. 1974, 39,
3654-3660.
(13) (a) Kahne, D.; Walker, S.; Cheng, Y.; Van Engen, D. Glycosylation
of unreactive substrates. J. Am. Chem. Soc. 1989, 111, 6881-6882.
(b) Yan, L.; Kahne, D. Generalizing glycosylation: synthesis of the
blood group antigens Lea, Leb, and Lex using a standard set of reaction
conditions. J. Am. Chem. Soc. 1996, 118, 9239-9248.
(14) (a) Crich, D.; Sun, S. Formation of â-mannopyranosides of primary
alcohols using the sulfoxide method. J. Org. Chem. 1996, 61, 4506-
4507. (b) Crich, D.; Sun, S. Are glycosyl triflates intermediates in
the sulfoxide glycosylation method? A chemical and 1H, 13C, and
19F NMR spectroscopic investigation. J. Am. Chem. Soc. 1997, 119,
11217-11223.
(15) (a) Abe, H.; Shuto, S.; Matsuda, A. Highly R- and â-selective radical
C-glycosylation reactions based on the conformational restriction
strategy using a controlling anomeric effect. A study on the
conformation-anomeric effect-stereoselectivity relationship in the
anomeric radical reactions. J. Am. Chem. Soc. 2001, 123, 11870-
(25) Altona, C.; Sundaralingam, M. Conformational analysis of the sugar
ring in nucleosides and nucleotides. Improved method for the
interpretation of proton magnetic resonance coupling constants. J.
Am. Chem. Soc. 1974, 95, 2333-2344.
(26) Preference of a C2′-endo form (C2′-endo/C3′-endo ) ca. 7/3) in
adenophostin A has been shown by the previous studies: ref 7e.
(27) Bosanac, I.; Alattia, J.-R.; Mal, T. K.; Chan, J.; Talarico, S.; Tong,
F. K.; Tong, K. I.; Yoshikawa, F.; Furuichi, T.; Iwai, M.; Michikawa,
T.; Mikoshiba, K.; Ikura, M. Structure of the inositol 1,4,5-
trisphosphate receptor binding core in complex with its ligand. Nature
2002, 420, 696-700.
JM060310D