Design and synthesis of indole derivatives of adenophostin A. A entry into subtype-selective IP3 receptor ligands

Article abstract of DOI:10.1016/j.tetlet.2009.12.045

Indole derivatives 3a and 3b of adenophostin A (2) in which the adenine of 2 was replaced with indole or 4-fluoroindole was designed as potential inositol trisphosphate receptor ligands. These target compounds were successfully synthesized from the key di

Full text of DOI:10.1016/j.tetlet.2009.12.045

Tetrahedron Letters 51 (2010) 977–979
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Design and synthesis of indole derivatives of adenophostin A.
A entry into subtype-selective IP3 receptor ligands
Tetsuya Mochizuki a, Akihiko Tanimura b, Akihiro Nezu b, Mika Ito c, Hiroshi Abe c, Yoshihiro Ito c,
Mitsuhiro Arisawa a, Satoshi Shuto a,
*
a Faculty of Pharmaceutical Sciences, Hokkaido University, Kita-12, Nishi-6, Kita-ku, Sapporo 060-0812, Japan
b School of Dentistry, Health Sciences University of Hokkaido, Ishikari-Tobetsu 061-0293, Japan
c RIKEN, Advanced Science Institute, 2-1, Hirosawa, Wako, Saitama 351-0198, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Indole derivatives 3a and 3b of adenophostin A (2) in which the adenine of 2 was replaced with indole or
4-fluoroindole was designed as potential inositol trisphosphate receptor ligands. These target compounds
were successfully synthesized from the key disaccharide unit 6. Biological evaluation showed that 3b
selectively activates IP3R1, a subtype of IP3 receptors.
Received 30 October 2009
Revised 9 December 2009
Accepted 10 December 2009
Available online 16 December 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Adenophostin A (Fig. 1, 2), isolated from Penicillium brevicom-
pactum,1 is a very potent myo-inositol 1,4,5-trisphosphate (IP3, 1)
receptor agonist, which is 10–100 times more potent than the
endogenous ligand IP3 in stimulating Ca2+ release and in binding
to IP3 receptors.1b,1c Due to the high potency, its total synthesis2
and also structure–activity relationship studies3,4 have been exten-
sively investigated. These studies have demonstrated that the ade-
nine or the corresponding aromatic ring at the 10b-position is
essential for the high potency as an IP3 receptor ligand.3,4
have been known so far. Thus, we focused our attention on the
development of adenophostin analogs with IP3 receptor subtype
selectivity.
Recently, Potter and co-workers investigated the binding mode
of 2 to the IP3 receptor by molecular modeling using the X-ray
crystal structure of IP3R1-binding core to suggest that the adenine
of adenophostin A seems to associate with the guanidium side
chain of Arg504 via a cation–
p interaction in the binding pocket
of IP3R1.3d Based on these modeling results, we designed an indole
derivative 3a of adenophostin A in which the adenine of adenopho-
stin A is replaced by indole, as shown in Figure 1. The shape of in-
dole is similar to that of adenine as a 5/6-fused aromatic ring
system, and it has efficient electron-donating feature,6 which
As a second messenger, IP3 functions generally in many organs,5
and at least three types of IP3 receptor subtypes, those are, IP3R1,
IP3R2, and IP3R3, are known.5b IP3 receptor ligand, which is selec-
tively active to one of the receptor subtypes, can be effectively used
as a biological tool for investigating cellular signal transduction via
IP3. However, none of the subtype-selective IP3 receptor ligands
might be suitable for the cation–p interaction speculated by the
modeling. In addition, the electron-donating potency of the indole
OH
OH
2
2
O3HPO
O3PO
O3PO
O3PO
O
O
OH
OH
2
2
1"
OH
OH
HO
HO
HO
2
O3PO
2
O
O
2
OPO3
O
N
O
N
N
O3PO
3'
N
1'
2
2
O3PO
O3PO
4
X
NH2
N
IP3 (1)
3a: X = H
3b: X = F
adenophostin A (2)
Figure 1. IP3 (1), adenophostin A (2), and the target indole derivatives 3a and 3b.
* Corresponding author. Tel.: +81 11 706 3769; fax: +81 11 706 3769.
E-mail address: shu@pharm.hokudai.ac.jp (S. Shuto).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.12.045

978
T. Mochizuki et al. / Tetrahedron Letters 51 (2010) 977–979
followed by acetylation of the resulting free hydroxyls afforded
the 1,2,30,40-tetra-O-acetate 8.
Indole nucleosides have been synthesized via glycosidation
with indoline as a glycosyl acceptor and subsequent oxidative aro-
matization of indoline ring into indole ring.9 Thus, using the tetra-
O-acetate 8 as a glycosyl donor, when a solution of 8 and indoline
(3 equiv) was heated in AcOH/EtOH (1:10) under reflux, the b-
selective glycosidation occurred to give glycosidation product 9a.
Subsequent oxidative treatment of 9a with MnO2 in toluene
yielded the desired b-indole nucleoside derivative 10a10 in 69%
yield from 8.
The 50-O-allyl-protecting group was replaced with a TBS group
at this stage.11 Thus, long time heating of 10a with (PPh3)3RhCl
in toluene/EtOH under reflux and subsequent treatment of the
resulting product with TBSCl/DMAP in pyridine gave the 50-O-silyl
ether 12a. The three O-acetyl groups of 12a were removed simul-
taneously with NaOMe/MeOH to give 13a. The phosphate units
were introduced into the 20-, 300-, and 400-hydroxyls using the phos-
phoramidite method with o-xylene N,N-diethylphosphoramidite
(XEPA).12 Thus, 13a was treated with XEPA in the presence of tet-
razole as a promoter in CH2Cl2, followed by oxidation with m-CPBA
to form the desired 20,300,400-trisphosphate derivative 14a in 97%
yield. After the removal of the 50-O-TBS group with TBAF/THF,
the protecting groups of the phosphate moieties were finally re-
moved under hydrogen atom transfer conditions with Pd(OH)2
and cyclohexene in MeOH to furnish the target trisphosphate 3a
as a sodium salt, after treatment with ion-exchange resin.
Another target compound 3b was synthesized by employing 4-
fluoroindoline as an acceptor instead of indoline in the glycosida-
tion step with the donor 8, as shown in Scheme 2.
Figure 2. The optimized structures (Ia and IIa) and their electronic potential
surfaces (Ib and IIb) of the model compounds I and II by DFT calculation.
ring may be controlled by introducing an electron-donating substi-
tuent or electron-withdrawing substituent.
Based on these findings and considerations, we also designed
another indole derivative 3b in which an electron-withdrawing
fluoro group is substituted at the indole 4-position. We thought
that if the cation–p interaction is indeed important for the binding
of adenophostin A to the receptor, affinity for the receptor might be
changed depending on the electron-donating feature or electron-
withdrawing feature of the substituent introduced at the indole
moiety.
To understand the electronic properties of indole derivatives,
quantum chemical calculation was carried out using 1-methylin-
dole (I) and 4-fluoro-1-methylindole (II) as model compounds
(Fig. 2).7 The model compounds I and II were fully optimized based
on density functional theory (DFT) giving the stable structures Ia
and IIa, respectively. Electrostatic potential surface of the opti-
mized structures Ia and IIa showed that 1-methylindole is elec-
tron-richer at the center of the aromatic ring than the
corresponding 4-fluoro derivative, which is shown as Ib and IIb,
respectively, in Figure 2. Furthermore, ionization energies of Ia
and IIa were calculated at 7.0725 eV and 7.1424 eV, respectively.
These calculation data indicated that 1-carbon-substituted indole
is more electron-donating than the corresponding 4-fluoro conge-
ner, as we expected.
We have previously developed a method for the identification
of ligands of IP3Rs using fluorescent biosensors,13,14 and estab-
lished subtype-specific IP3 biosensors LIBRAvI and LIBRAvII for
investigating binding affinity of IP3 receptor ligands for IP3R1 and
IP3R2, respectively.15 Thus, the binding abilities of adenophostin
A and the target compounds 3a and 3b were examined with these
biosensors. Figure 3 indicates the increase in emission ratios of LIB-
RAvI and LIBRAvII with a same concentration (1 lM) of these com-
pounds. Effect of adenophostin A, 3a, and 3b on LIBRAvI was ꢀ80%,
ꢀ40%, and ꢀ35% of the maximal response, respectively. These re-
sults indicate that the potency of adenophostin A for the activation
of IP3R1 is higher than that of compounds 3a and 3b. Although sim-
ilar effects of adenophostin A and 3a were observed on the emis-
sion ratio of LIBRAvII compared to those on LIBRAvI, the effect of
target compounds 3b on LIBRAvII was significantly small. Thus,
their selectivity index (LIBRAvI/LIBRAvII) is 0.97 for adenophostin
A, 1.1 for 3a, and 8.4 for 3b, respectively. These results indicate that
4-fluoroindole derivative 3b has a higher selectivity for IP3R1 over
IP3R2. To our knowledge, this kind of subtype-selective IP3 receptor
ligands has not been reported so far.
As shown in Scheme 1, we planned to synthesize the target
compounds 3a and 3b via the
vided by the -selective glycosidation with a sulfoxide donor 4 and
an acceptor 5.4d The
-disaccharide 6 has been used previously as
a-disaccharide 6, which can be pro-
a
a
an effective synthetic intermediate for synthesizing various deriv-
atives of adenophostin A.4f
Synthesis of the target compound 3a from 6 was accomplished
as summarized in Scheme 2. After the removal of the 5-O-TBS
group of 6, the 5-hydroxyl was re-protected with an allyl group8
to give 7. Acidic removal of the ketal-protecting groups of 7
These results suggest that the indole ring in 3b might not effec-
tively work for the expected cation–p interaction in its binding to
the receptor. However, significantly weak binding of 3b to the
IP3R2 compared with that to the IP3R1 suggested that the subtype
OBn
OMe
O
O
OBn
OMe
O
D-glucose
D-xylose
O
SPh
4
O
OTBS
O
Tf2O, DTBMP
Et2O, -78 °C
O
BnO
OMe
BnO
OMe
O
3a and 3b
O
O
+
78%
TBSO
O
O
O
6
O
5
HO
Scheme 1.

T. Mochizuki et al. / Tetrahedron Letters 51 (2010) 977–979
979
OBn
O
OBn
O
OBn
O
OMe
O
AcO
O
O
O
AcO
AcO
O
O
AcO
c
d
BnO
BnO
b
a
BnO
OMe
6
O
O
O
O
O
O
9a: X = H
9b: X = F
7
8
O
N
AcO
OAc
AcO
X
OBn
O
OBn
O
OBn
R2O
R2O
OR1
OTBS
AcO
AcO
O
AcO
O
BnO
BnO
AcO
O
e
O
g
i
BnO
O
N
O
N
3a and 3b
O
O
R2O
X
AcO
X
N
AcO
X
13a: X = H, R2 = H
13b: X = F, R2 = H
11a: X = H, R1 = H
11b: X = F, R1 = H
10a: X = H
10b: X = F
h
f
14a: X = H, R2 =
O
P
O
12a: X = H, R1 = TBS
12b: X = F, R1 = TBS
14b: X = F, R2
=
O
Scheme 2. Reagents and conditions: (a) (1) TBAF, THF, rt; (2) CH2@CHCH2Br, Ca(OH)2, CaO, DMF, rt, 97%; (b) (1) 90% aq TFA, rt; (2) Ac2O, DMAP, Et3 N, MeCN, rt, 77%; (c)
indoline, AcOH, EtOH, reflux; (d) MnO2, toluene, rt, 69% (10a) from 8, 77% (10b) from 8; (e) (PPh3)3RhCl, toluene/EtOH, reflux, 68% (11a), 72% (11b); (f) TBSCl, DMAP, pyridine,
rt, 64% (12a), 79% (12b); (g) NaOMe, MeOH, rt, quant (13a), quant (13b); (h) XEPA, tetrazole, CH2Cl2, ꢁ40 °C to rt, then m-CPBA, ꢁ40 °C, 97% (14a), 71% (14b); (i) (1) TBAF,
THF, rt; (2) Pd(OH)2, cyclohexene, MeOH, reflux, 97% (3a), 97% (3b).
2. Total synthesis of adenophostin A: (a) Hotoda, H.; Takahashi, M.; Tanzawa, K.;
Takahashi, S.; Kaneko, M. Tetrahedron Lett. 1995, 36, 5037–5040; (b) van
Straten, N. C. R.; van der Marel, G. A.; van Boom, J. H. Tetrahedron 1997, 53,
6509–6522; (c) Marwood, R. D.; Correa, V.; Taylor, C. W.; Potter, B. V. L.
Tetrahedron: Asymmetry 2000, 11, 397–403.
3. Synthesis of adenophostin A analogs see: (a) de Kort, M.; Regenbogen, A. D.;
Overkleeft, H. S.; Challis, J.; Iwata, Y.; Miyamoto, S.; van der Marel, G. A.; van
Boom, J. Tetrahedron 2000, 56, 5915–5928. and references therein; (b) Chretien,
F.; Moitessier, N.; Roussel, F.; Mauger, J. P.; Chapleur, Y. Curr. Org. Chem. 2000, 4,
513–534; (c) Roussel, F.; Moitessier, N.; Hilly, M.; Chrétien, F.; Mauger, J.-P.;
Chapleur, Y. Bioorg. Med. Chem. 2002, 10, 759–768; (d) Rosenberg, H. J.; Riley, A.
M.; Laude, A. J.; Taylor, C. W.; Potter, B. V. L. J. Med. Chem. 2003, 46, 4860–4871;
(e) Rossi, A. M.; Riley, A. M.; Tovey, S. C.; Rahman, T.; Dellis, O.; Taylor, E. J.;
Veresov, V. G.; Potter, B. V.; Taylor, C. W. Nat. Chem. Biol. 2009, 5, 631–639. and
references therein.
4. (a) Shuto, S.; Tatani, K.; Ueno, Y.; Matsuda, A. J. Org. Chem. 1998, 63, 8815–
8824; (b) Abe, H.; Shuto, S.; Matsuda, A. J. Org. Chem. 2000, 65, 4315–4325; (c)
Shuto, S.; Yahiro, Y.; Ichikawa, S.; Matsuda, A. J. Org. Chem. 2000, 65, 5547–
5557; (d) Mochizuki, T.; Kondo, Y.; Abe, H.; Taylor, C. W.; Potter, B. L. V.;
Matsuda, A.; Shuto, S. Org. Lett. 2006, 8, 1455–1458; (e) Terauchi, M.; Abe, H.;
Tovey, S. C.; Dedos, S. G.; Taylor, C. W.; Paul, M.; Trusselle, M.; Potter, B. V. L.;
Matsuda, A.; Shuto, S. J. Med. Chem. 2006, 49, 1900–1909; (f) Mochizuki, T.;
Kondo, Y.; Abe, H.; Tovey, S. C.; Dedos, S. G.; Taylor, C. W.; Paul, M.; Potter, B. V.
L.; Matsuda, A.; Shuto, S. J. Med. Chem. 2006, 49, 5750–5758. and references
therein.
5. (a) Berridge, M. J. Nature (London) 1993, 361, 315–325; (b) Joseph, S. K. Cell.
Figure 3. Binding affinities of adenophostin A (2) and its indole derivatives 3a and
3b for LIBRAvI and LIBRAvII. Effects of compounds on the emission ratio (480/
535 nm) of LIBRAvI and LIBRAvII are expressed as % of maximal response. The
Signalling 1996, 8, 1–7.
6. (a) Mecozzi, S.; West, A. P.; Dougherty, D. A. J. Am. Chem. Soc. 1996, 118, 2307–
2308; (b) Yorita, H.; Otomo, K.; Hiramatsu, H.; Toyama, A.; Miura, T.; Takeuchi,
H. J. Am. Chem. Soc. 1996, 118, 2307–2308; (c) Inoue, Y.; Sugio, S.; Andzelm, J.;
Makamura, N. J. Phys. Chem. A 1998, 102, 646–648.
maximal response was determined by the application of 30
each experiment.
lM IP3 at the end of
*
7. The preoptimized geometries by HF/3-21G were taken to be the input
*
geometries for final optimization by B3LYP/6-31G . Single point energies and
all electronic properties were calculated by B3LYP/6-31G*.
8. Preliminary experiments suggested that in the glycosidation of a-disaccharide
selectivity may be manipulated by changing the substituent at the
indole 4-position in indole derivatives of adenophostin A.
In conclusion, the indole derivatives 3a and 3b designed as a no-
vel IP3 receptor ligands were successfully synthesized from the key
disaccharide unit 6. While binding affinities of 3a and 3b for IP3R1
and IP3R2 are weaker than adenophostin A, the 4-fluoroindole
derivative 3b was shown to be an IP3R1-selective ligand. Therefore,
3a may be an effective lead for developing subtype-selective IP3
receptor ligands, which can be useful biological tools.
donors at the 1b-position of the ribose moiety with indoline as an acceptor, a
bulky protecting group, such as TBS group, at the 5-primary hydroxyl of the
donor might prevent the glycosidation to proceed.
9. Preobrazheskaya, M. N.; Vigdorchik, M. M.; Suvorov Tetrahedron 1967, 23,
4653–4660.
10. The anomeric b-configuration was confirmed from an NOE correlation between
H-2 of the indole moiety and H-20 of the ribose moiety.
11. When deprotection of the 50-O-allyl group was examined after the introduction
of the o-xylene phosphate units at the 20, 300, and 400-hydroxyls, it was
unsuccessful.
12. Watanabe, Y.; Komoda, Y.; Ebisuya, K.; Ozaki, S. Tetrahedron Lett. 1990, 31,
255–256.
13. Nezu, A.; Tanimura, A.; Morita, T.; Shitara, A.; Tojyo, Y. Biochim. Biophys. Acta
Gen. Subj. 2006, 1760, 1274–1280.
References and notes
1. (a) Takahashi, M.; Kagasaki, T.; Hosoya, T.; Takahashi, S. J. Antibiot. 1993, 46,
1643–1647; (b) Takahashi, M.; Tanzawa, K.; Takahashi, S. J. Biol. Chem. 1994,
269, 369–372; (c) Hirota, J.; Michikawa, T.; Miyawaki, A.; Takahashi, M.;
Tanzawa, K.; Okura, I.; Furuichi, T.; Mikoshiba, K. FEBS Lett. 1995, 368, 248–252.
14. Zhang, L.; Huang, W.; Tanimura, A.; Morita, T.; Harihar, S.; De Wald, D. B.;
Prestwich, G. D. Chem. Med. Chem. 2007, 2, 1281–1289.
15. Tanimura, A.; Morita, T.; Nezu, A.; Shitara, A.; Hashimoto, N.; Tojyo, Y. J. Biol.
Chem. 2009, 284, 8910–8917.