Synthesis of D- and L-myo-inositol 2,4,5-trisphosphate and trisphosphorothioate: structural analogues of D-myo-inositol 1,4,5-trisphosphate.

Article abstract of DOI:10.1016/S0008-6215(02)00289-6

The preparation of D- and L-myo-inositol 2,4,5-trisphosphate is described, together with the phosphorothioate counterparts. The known chiral diols D- and L-1,4-di-O-benzyl-5,6-bis-O-p-methoxybenzyl-myo-inositol were regioselectively protected at the 3-position using a benzyl group via a 2,3-O-stannylene acetal. Removal of the p-methoxybenzyl groups of each enantiomer gave D- and L-1,3,6-tri-O-benzyl-myo-inositol. Phosphitylation with bis(benzyloxy)diisoproplyaminophosphine and 1H-tetrazole gave the trisphosphite intermediate for each enantiomer. Oxidation with 3-chloroperoxybenzoic acid gave the fully protected D- and L-myo-inositol 2,4,5-trisphosphates. Sulphoxidation of the D- and L-2,4,5-trisphosphite intermediates gave the fully protected D- and L-myo-inositol 2,4,5-trisphosphorothioate compounds. The fully protected trisphosphates were deblocked using hydrogenolysis and the phosphorothioates were deprotected using sodium in liquid ammonia. The individual compounds were then purified using ion exchange chromatography to afford pure D- and L-myo-inositol 2,4,5-trisphosphates together with the corresponding phosphorothioates.

Full text of DOI:10.1016/S0008-6215(02)00289-6

Carbohydrate Research 337 (2002) 1795–1801
www.elsevier.com/locate/carres
Synthesis of
D
- and -myo-inositol 2,4,5-trisphosphate and
L
trisphosphorothioate: structural analogues of
1,4,5-trisphosphate
D-myo-inositol
Stephen J. Mills, Changsheng Liu, Barry V. L. Potter*
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology, Uni6ersity of Bath, Cla6erton Down,
Bath BA2 7AY, UK
Received 24 May 2002; accepted 14 August 2002
Abstract
The preparation of
The known chiral diols
3-position using a benzyl group via a 2,3-O-stannylene acetal. Removal of the p-methoxybenzyl groups of each enantiomer gave
- and -1,3,6-tri-O-benzyl-myo-inositol. Phosphitylation with bis(benzyloxy)diisoproplyaminophosphine and 1H-tetrazole gave
the trisphosphite intermediate for each enantiomer. Oxidation with 3-chloroperoxybenzoic acid gave the fully protected - and
-2,4,5-trisphosphite intermediates gave the fully protected
-myo-inositol 2,4,5-trisphosphorothioate compounds. The fully protected trisphosphates were deblocked using hydrogenoly-
D
- and
L
-myo-inositol 2,4,5-trisphosphate is described, together with the phosphorothioate counterparts.
D
- and
L
-1,4-di-O-benzyl-5,6-bis-O-p-methoxybenzyl-myo-inositol were regioselectively protected at the
D
L
D
L
-myo-inositol 2,4,5-trisphosphates. Sulphoxidation of the
D
- and
L
D-
and
L
sis and the phosphorothioates were deprotected using sodium in liquid ammonia. The individual compounds were then purified
using ion exchange chromatography to afford pure - and -myo-inositol 2,4,5-trisphosphates together with the corresponding
phosphorothioates. © 2002 Elsevier Science Ltd. All rights reserved.
D
L
Keywords: Benzyl ether;
D- and L-myo-Inositol 2,4,5-trisphosphate; D- and L-myo-Inositol 2,4,5-trisphosphorothioate; Chiral intermediates;
Phosphitylation
1. Introduction
It is well established that
D-myo-inositol 1,4,5-
trisphosphate [Ins(1,4,5)P3, 1] is a second messenger.1
This hydrophilic product interacts with specific recep-
tors that are recognised as a family of tetrameric lig-
and-gated Ca2+ channels. When Ins(1,4,5)P3,
1
interacts with a site on the N-terminal portion of its
receptor, Ca2+ ions are released from intracellular
stores. There are three fully characterised Ins(1,4,5)P3
receptor subtypes (I, II and III) known to date. From
the protein monomers, a tetrameric Ca2+ channel is
assembled, to give homo- and heteromeric receptors.
Most mammalian cells express a mixture of the
Ins(1,4,5)P3 receptor subtypes in a given cell line or
tissue.1,2
* Corresponding author. Tel.: +44-1225-386639; fax +44-
1225-386114
E-mail address: prsbvlp@bath.ac.uk (B.V.L. Potter).
0008-6215/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0008-6215(02)00289-6

1796
S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
Once Ca2+ ions have been released from the stores,
chromatography. Gigg and co-workers16 synthesised a
fully protected -myo-inositol 2,4,5-trisphosphate
derivative that was only partially deblocked. Two syn-
theses of racemic Ins(2,4,5)P3 have also appeared in the
literature, using myo-inositol as the starting mate-
rial,17,18 and a novel route using benzene has also been
published.19
the signal is then ‘switched off’ providing inactive or
less active compounds. Deactivation of Ins(1,4,5)P3 can
occur by a specific 5-phosphatase enzyme, which re-
moves the phosphate group at C-5 of Ins(1,4,5)P3 to
generate myo-inositol 1,4-bisphosphate [Ins(1,4)P2, 2]
and turn the signal off. Second, Ins(1,4,5)P3 may be
phosphorylated by a specific 3-kinase, to afford myo-
inositol 1,3,4,5-tetrakisphosphate, [Ins(1,3,4,5)P4, 3].
The role of the latter compound remains controversial;
however, recent studies imply that it regulates the fre-
quency of Ca2+ oscillations in vivo in conjunction with
the 3-kinase.3 Compound 3 is a potent bimodal regula-
tor of cellular sensitivity4 to Ins(1,4,5)P3 and is the most
potent 5-phosphatase inhibitor known, with an IC50
value of 0.15 mM.4
D
Early biological studies by Putney Jr.20 formally
demonstrated that Ins(2,4,5)P3 is a poor substrate for
3-kinase and a full agonist like Ins(1,4,5)P3, albeit with
a longer lifetime. Like Ins(1,4,5)P3, 4a can fully activate
sustained Ca2+ entry as well as Ca2+ release in
lacrimal acinar cells. Racemic myo-inositol 2,4,5-
trisphosphate was evaluated for Ca2+ mobilisation and
found to be 30-fold weaker than Ins(1,4,5)P3 in some
cell types.11,13 Racemic Ins(2,4,5)P3 has also been used
as a metabolically resistant analogue11 because it is only
a weak substrate for Ins(1,4,5)P3 5-phosphatase and a
poor substrate for Ins(1,4,5)P3 3-kinase.11,21 In other
Since the first synthesis of
D-Ins(1,4,5)P3, numerous
analogues have been synthesised and evaluated in struc-
ture–activity studies for their ability to mimic the ac-
tion of the natural ligand.5 One of our goals is to
develop analogues based on Ins(1,4,5)P3 to probe the
polyphosphate pathway of signal transduction. We
studies,
D
-Ins(2,4,5)P3 was 6-fold,10 13-fold12 or 25-
fold14 less potent than Ins(1,4,5)P3 for Ca2+ mobilisa-
tion. In experiments carried out by Tegge and
co-workers, 4a and 4b were tested in saponin perme-
abilised rat basophilic leukemic (RBL) cells.14 The RBL
cells consist of type II and type III Ins(1,4,5)P3 receptor
subtypes and 4a was found to be 25-fold less potent
than Ins(1,4,5)P3, while 4b was 650-fold less active than
Ins(1,4,5)P3 for Ca2+ release. This is the only study
wanted to synthesise
[Ins(2,4,5)P3, 4a], a regioisomer of Ins(1,4,5)P3, and its
enantiomer, -myo-inositol 2,4,5-trisphosphate
Ins(2,4,5)P3, 4b] together with their new phosphoroth-
ioate counterparts, -myo-inositol 2,4,5-trisphos-
phorothioate [ -Ins(2,4,5)PS3, 5a] and -myo-inositol
2,4,5-trisphosphorothioate [ -Ins(2,4,5)PS3, 5b]. Previ-
D-myo-inositol 2,4,5-trisphosphate
L
[L-
D
D
L
L
ously, when phosphorothioate groups replace phos-
phate moieties in inositol phosphate analogues,5–8 they
have provided compounds that are resistant to 5-phos-
phatase and 3-kinase and can display partial agonist
activity when tested for Ca2+ release under specific
experimental conditions.6–8
where both
D- and L-Ins(2,4,5)P3 have been tested,
albeit in one cell line. Racemic Ins(2,4,5)P3 was tested
for Ca2+ release in permeabilised rat hepatocytes22 and
found to be a partial agonist in these cells. Racemic
Ins(2,4,5)P3 released some 65% of the total Ca2+ com-
pared to Ins(1,4,5)P3. Chock and co-workers3 used
Ins(2,4,5)P3 to stimulate Ca2+ release in HeLa cells.
However, only Ins(1,4,5)P3 and Ins(1,3,4,5)P4 enhanced
the frequency of Ca2+ oscillations, and this activity
distinguished the natural ligand from the Ins(2,4,5)P3
derivative. Ins(2,4,5)P3 has also been used at submaxi-
mal levels in Ca2+ release experiments to show that
pure synthetic Ins(1,3,4,5)P4 does not release Ca2+ ions
from intracellular stores.4 These examples illustrate the
importance of myo-inositol 2,4,5-trisphosphate as a
powerful tool for investigating Ins(1,4,5)P3 metabolism
and Ca2+ signalling and our new route to both enan-
tiomers of this compound and their phosphorothioate
derivatives should enhance the potential of this impor-
tant tool.
D-Ins(2,4,5)P3 4a has been extracted from several
tissues.9–13 However, it is unclear whether 4a is synthe-
sised in vivo, or is due to migration of the phosphate
group from position C-1-OH to C-2-OH under certain
conditions. Ins(2,4,5)P3 4a is an inositol polyphosphate
regioisomer of Ins(1,4,5)P3 1 which has been the target
of fewer synthetic approaches compared with the natu-
ral receptor ligand,5 Ins(1,4,5)P3. Racemic and chiral
Ins(2,4,5)P3, were made in the first few years of intense
synthetic work which culminated in two syntheses14,15
of chiral Ins(2,4,5)P3. The synthesis by Tegge and co-
workers14 delivered both enantiomers. The
tiomer 4a was derived from -pinitol and the
-enantiomer 4b from -quebrachitol after several pro-
D-enan-
D
L
L
tecting and deprotecting steps. However, the starting
materials required for the synthesis of 4a and 4b are
expensive compared to myo-inositol. The first synthesis
2. Results and discussion
of
D
-Ins(2,4,5)P3 4a was achieved by Watanabe and
co-workers15 who resolved a racemic myo-inositol
derivative using a chiral auxillary then separated the
two diastereoisomers using medium pressure silica gel
Previously,23 we have described the synthesis of the
chiral diols,
D-1,4-di-O-benzyl-5,6-bis-O-p-methoxy-

S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
1797
benzyl-myo-inositol 6b, and its
L
-enantiomer 6a.† They
which were then purified by ion exchange chromatogra-
phy on Q Sepharose Fast Flow in a similar manner for
4a and 4b.
were synthesised when the racemic 2,3-diol 6ab was
reacted with the chiral acid, (S)-(+)-O-acetylmandelic
acid and the resulting diastereoisomers were separated
by flash chromatography. The chiral auxillary was then
removed by methanolysis to provide the individual
diols 6a and 6b (Scheme 1). Selective tin-mediated
benzylation24 of 6a and 6b at the equatorial 3-hydroxy
was accomplished via the cis-2,3-O-dibutylstannylene
in the presence of caesium fluoride in N,N-dimethylfor-
mamide (DMF) to provide the intermediates 7a and 7b
in 85 and 80% yield, respectively. The key 2,4,5-triol
intermediates 8a and 8b were obtained by deprotection
of the p-methoxybenzyl groups in a mixture of trifl-
uoroacetic acid in CH2Cl2 (1:10) at room temperature
for 30 min. Under these conditions, the benzyl groups
remained intact to give a suitably protected 2,4,5-triol
derivative. The two triols 8a and 8b have equal and
opposite specific rotations (+20°) and (−20°), respec-
tively, which were similar to the literature value16 [h]D
There is no doubt that the source and purity of
D
-Ins(2,4,5)P3 plays a pivotal role in the quality of data
arising from biological experiments. We have synthe-
sised pure -Ins(2,4,5)P3 4a and the -enantiomer 4b
D
L
from a racemic precursor, which was then resolved
using a chiral auxillary and then transformed into the
named target compounds. Furthermore, pure
D-
Ins(2,4,5)PS3 5a, which has previously been unavail-
able, may release less Ca2+ than its phosphate
counterpart, and behave as a low intrinsic activity
partial agonist at the Ins(1,4,5)P3 receptor6–8 under
specific experimental conditions. Synthetic Ins(2,4,5)P3
+16.2° (for the
D-enantiomer of 1,3,6-tri-O-benzyl-
myo-inositol) and possessed the same liquid crystalline
properties which have also been described by Gigg and
co-workers.16
The phosphate groups were introduced by a PIII
method followed by oxidation. Thus, a mixture of
bis(benzyloxy)diisopropylaminophosphine25 and 1H-
tetrazole was stirred in CH2Cl2 at room temperature. A
tetrazolide intermediate23 (lP +127 ppm) was observed
by 31P NMR and the individual triols 8a and 8b were
added in separate experiments. The intermediate
trisphosphite was stirred for a short time and oxidised
with m-chloroperoxybenzoic acid (MCPBA) at −78 °C
to yield the fully protected phosphate triesters (9a and
9b), after work-up and purification. Deprotection of the
benzyl protective groups was accomplished by hy-
drogenolysis in the presence of 10% palladium on car-
bon as the catalyst, to afford the corresponding
D- and
L-2,4,5-trisphosphates (4a and 4b), respectively. These
compounds were further purified by ion exchange chro-
matography on Q Sepharose Fast Flow where the
compounds eluted between 500 and 600 mmol dm−3
triethylammonium hydrogencarbonate (TEAB) buffer,
and isolated as pure glassy triethylammonium salts and
quantified by a modification of the Briggs phosphate
analysis.26
The phosphorothioate counterparts 5a and 5b were
synthesised in a similar way; however, the intermediate
PIII trisphosphites were oxidised using sulphur in a
mixed solvent of pyridine and DMF7 to afford the fully
blocked trisphosphorothioate triesters 10a and 10b.
Treatment of 10a or 10b with sodium in liquid ammo-
nia furnished the trisphosphorothioates 5a and 5b,
Scheme 1. Reagents and conditions: (i) Dibutyltin oxide,
PhCH3, Dean–Stark apparatus; then CsF, BnBr, DMF (85%
for 7a, 80% for 7b); (ii) CH2Cl2–trifluoroacetic acid (10:1), 30
min (85% for 8a, 87% for 8b); (iii) (BnO)2PNPri2, 1H-tetra-
zole, CH2Cl2, rt, 15 min, then add triols 8a or 8b in separate
experiments, 10 min, then MCPBA, 0 °C, 30 min (73% for 9a,
83% for 9b); (iv) (BnO)2PNPri2, 1H-tetrazole, CH2Cl2, rt,
triols 8a or 8b in separate experiments, then add S8/DMF/
pyridine, 15 min (90% for 10a, 93% for 10b); (v) H2 10%
Pd/C, 30 psi, (overnight), followed by purification by ion-ex-
change chromatography; (94% for 4a, 51% for 4b). (vi) Na/
NH3 (−78 °C), then purification by ion exchange
chromatography (72% for 5a, 70% for 5b).
† After compounds 6a and 6b there is a switch of
nomenclature. Thus, -“ - and -“ -.
D- and L-
D
L
L
D

1798
S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
was tested in a preliminary fashion for its Ca2+ mobil-
ising properties in a clam homogenate assay (A.
Galione, S. J. Mills and B. V. L. Potter, unpublished
data), where it was found to have potent activity, albeit
around an order of magnitude weaker than
Ins(1,4,5)P3.
Recently, Ins(2,4,5)P3 synthesised by the route de-
scribed here, was used as a biological tool to investigate
its effects in leech photoreceptors.27 It was more effec-
tive than Ins(1,4,5)P3 at Ca2+ release, since it is not
immediately metabolised by deactivating enzymes. Fur-
ther experimental findings also demonstrated that
Ins(2,4,5)P3 mimicked the effect of excitation and adap-
tation, produced in the same way as light. However,
Ins(1,4,5)P3 could not depolarise intact leech photore-
ceptors in the same way. Many biologists have used
Ins(2,4,5)P3 in their Ca2+ signalling experiments since
it behaves as a long lived agonist in most cells and a
surrogate for Ins(1,4,5)P3 at its receptor. We have
tested compound 4a in competitive binding experiments
using recombinant mammalian type I, II and III
The 31P NMR shifts were measured in ppm relative to
external 85% phosphoric acid. Mps (uncorrected) were
determined using a Reichert–Jung Thermo Galen
Kofler block. Microanalysis was carried out by the
University of Bath Microanalysis Service. Mass spectra
were recorded by the University of Bath Mass Spec-
trometry Service using positive and negative ion fast
atom bombardment (FAB) with 3-nitrobenzyl alcohol
(NBA) as the matrix. Optical rotations were measured
using an Optical Activity Ltd. AA-10 polarimeter, and
[h]D-values are given in units of 10−1 deg cm2 g−1 and
were measured at ambient temperature. Ion exchange
chromatography was performed on an LKB-Pharmacia
medium-pressure ion exchange chromatograph using Q
Sepharose Fast Flow with gradients of triethylammo-
nium hydrogen carbonate (TEAB) as eluent. Column
fractions containing inositol polyphosphates and phos-
phorothioates were assayed for total phosphate and
phosphorothioate by a modification of the Briggs test.26
(Diisopropylamino)dichlorophosphine was prepared
by the method of Tanaka and co-workers30 by adding
two mole equivalents of diisopropylamine to a solution
of phosphorus trichloride in dry diethyl ether at −
78 °C. The crude product was purified by distillation
under reduced pressure (lP +169.4 ppm) and could be
stored as a crystalline solid at −20 °C. Two equiva-
lents of benzyl alcohol in the presence of triethylamine
were then reacted with the purified product in CH2Cl2,
to afford bis(benzyloxy)(diisopropylamino)phosphine25
(lP +147.9 ppm) which was pure by 31P NMR, (Rf
0.78, hexane–triethylamine 10:1).
Ins(1,4,5)P3 receptors expressed in
a baculovirus/
Spodoptera frugiperda 9 (Sf9) cell system.28 It was
found that Ins(2,4,5)P3 bound tightest to type II and
type III receptors; Kd of 131 nM for type II (cf
Ins(1,4,5)P3 is 11 nM), and type III; Kd, 194 nM (cf
Ins(1,4,5)P3 is 17 nM), which is around 11 fold weaker
than Ins(1,4,5)P3, for both type II and type III receptor.
However, binding was weakest in the type I receptor,
where Kd was 615 nM for Ins(2,4,5)P3 and 24 nM for
Ins(1,4,5)P3, which is 24 fold weaker in this receptor
subtype. We have thus achieved a practical route to
D
-1,3,6-Tri-O-benzyl-4,5-bis-O-(p-methoxybenzyl)-
biologically active
D
-Ins(2,4,5)P3, its
L
-enantiomer and
myo-inositol 7a.—A mixture of compound23 6a (1.88 g,
3.13 mmol) and dibutyltin oxide (0.85 g, 3.4 mmol)
were heated at reflux in toluene (250 mL) using a
Dean–Stark apparatus for 2.5 h. The reaction mixture
was cooled and the toluene was evaporated to give a
syrup that was dried under vacuum for a further 2 h.
Caesium fluoride (1.89 g, 7.82 mmol) and dry DMF (30
mL) were added to the syrup under an atmosphere of
nitrogen, together with benzyl bromide (1.07 g, 6.26
mmol), and the reaction was stirred overnight at room
temperature. The reaction was complete by TLC
(CH2Cl2–Et2O, 15:1) and showed a single product (Rf
0.30). The solvents were evaporated under reduced
pressure and the syrupy residue was extracted with
CH2Cl2 (100 mL), and the extract was washed with
water (100 mL) and stirred with sodium hydrogen
carbonate solution (10% w/v) for 30 min, washed with
water again, dried over MgSO4, then filtered through a
pad of Celite and the solvent was evaporated. The
crude product was purified by flash chromatography
(CH2Cl2–Et2O, 15:1) to give the title compound 7a as a
solid. Yield (1.83 g, 85%); mp 106–108 °C (from
EtOAc–hexane); [h]D −9° (c 1, CHCl3); 1H NMR
(CDCl3): l 7.20–7.34 (m, 19 H, 3×CH2Ph and 2×0.5
the corresponding phosphorothioate counterparts with-
out contamination of other trisphosphates or phospho-
rothioates. The compounds will be of use for
investigating signal transduction in the polyphospho-
inositide pathway.
3. Experimental
General methods.—Chemicals were purchased from
Aldrich, Fluka and Lancaster. Sodium hydride was
used as a 60% dispersion in mineral oil. Thin-layer
chromatography (TLC) was performed on precoated
plates (Merck TLC aluminium sheets silica 60 F254);
products were visualised by spraying with phospho-
molybdic acid in methanol, which was then heated for
thirty seconds at high temperature. Flash chromatogra-
phy refers to the procedure developed by Still and
co-workers29 and was carried out on Sorbsil C60 silica
gel.
NMR spectra (proton frequency 270 or 400 MHz)
were referenced to internal SiMe4 or deuterium oxide.
Samples recorded in D2O were approximately pH 4–5.

S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
1799
CH2C6H4OMe), 6.84 (d, 2 H, J 8.8 Hz, 0.5×
CH2C6H4OMe), 6.83 (d, 2 H, J 8.6 Hz, 0.5×
CH2C6H4OMe), 4.69–4.92 (m, 10 H, 3×CH2Ph, and
2×CH2C6H4OMe), 4.21 (dd, 1 H, J2,3 2.25 Hz, H-2),
3.97 (2×dd overlapping, 2 H, J4,5 and J1,6 9.5 Hz, H-4
and H-6), 3.79 (s, 3 H, CH2C6H4OMe), 3.78 (s, 3 H,
CH2C6H4OMe), 3.40 (m, 3 H, overlapping, H-1, H-3
and H-5), 2.48 (br s, 1 H, D2O, exchangeable HO-2);
13C NMR (CDCl3): l 158.96 (Cq, CH2PhOMe), 138.69,
137.91, 137.86 (Cq, CH2Ph), 132.53, 130.86, 130.78,
129.38, 128.23, 128.12, 127.71, 127.65, 127.58, 127.31
(CH2Ph), 113.55 (CH2PhOMe), 82.79, 81.05, 80.83,
79.70, 79.53, 67.16 (6×myo-inositol ring carbons),
75.65, 75.36, 72.44, 72.41 (CH2PhOMe and CH2Ph),
55.02 (CH2PhOMe); Anal. Calcd for C43H46O8: C,
74.76; H 6.71. Found: C, 74.5; H, 6.72.
bons), 75.43, 74.19, 72.39 (CH2Ph); Anal. Calcd for
C27H30O6: C, 71.98; H, 6.71. Found: C, 72.0; H, 6.70.
L
-1,3,6-Tri-O-benzyl-myo-inositol 8b.—Compound
7b (1.05 g, 1.52 mmol) was stirred in a mixture of
CH2Cl2–trifluoroacetic acid (55 mL, 10:1) for 30 min.
After this time TLC showed a product Rf 0.20 (ether)
and a deprotected p-methoxybenzyl derivative (Rf
0.70). Work-up and purification, as for the D-derivative
8a, provided the title compound (0.59 g, 87%); mp
109–112 °C (liquid crystal), 126–127 °C, (clear liquid,
from EtOAc–hexane); [h]D −20° (c 1, CHCl3). The
NMR data were identical with those of the
D-enan-
tiomer; Anal. Calcd for C27H30O6: C, 71.98; H, 6.71.
Found: C, 71.7; H, 6.68.
D
- 1,3,6 - Tri - O - benzyl - 2,4,5 - tris - O - [di(benzyloxy)-
phosphoryl]-myo-inositol 9a.—A mixture of bis(benzy-
loxy)(diisopropylamino)phosphine25 (0.69 g, 2 mmol)
and 1H-tetrazole (0.42 g, 6 mmol) in dry CH2Cl2 (10
L
-1,3,6-Tri-O-benzyl-4,5-bis-O-(p-methoxybenzyl)-
myo-inositol 7b.—A mixture of compound 6b (1.42 g,
2.36 mmol) and dibutyltin oxide (0.65 g, 2.6 mmol)
were heated at reflux in toluene (250 mL) using a
Dean–Stark apparatus for 2.5 h. The reaction mixture
was cooled and the toluene was evaporated to provide
a syrup and dried under vacuum for 2 h. Caesium
fluoride (1.075 g, 7.08 mmol) and dry DMF (30 mL)
were added to the syrup under an atmosphere of nitro-
gen, together with benzyl bromide (0.81 g, 6.26 mmol),
and the reaction was stirred overnight at room temper-
ature. Work-up and purification, as for 7a, furnished
7b. Yield (1.30 g, 80%); mp 106–108 °C (from EtOAc–
hexane); [h]D +9° (c 1, CHCl3); NMR data were
mL), was stirred for 15 min at rt. D-1,3,6-Tri-O-benzyl-
myo-inositol 8a (0.22 g, 0.49 mmol) was then added to
the reaction mixture which was stirred for a further 15
min. The solution was cooled to −78 °C and MCPBA
(1.00 g, 2.81 mmol) was added and the mixture was
allowed to warm-up over 30 min. The reaction mixture
was then partitioned between CH2Cl2 and a 10% aq
solution of sodium metabisulfite (50 mL of each). The
organic layer was then washed with brine and water (50
mL of each), dried (MgSO4) and the solvent was evap-
orated to give the crude product, which was purified by
flash chromatography (CHCl3–acetone 10:1, Rf 0.40) to
give the title compound 9a as a syrup. Yield (0.44 g,
73%); [h]D −9.2° (c 6.2, CHCl3); Lit.16 [h]D −9.0° (c
identical with those of the
D-enantiomer; Anal. Calcd
for C43H46O8: C, 74.76; H, 6.71. Found: C, 74.9; H,
6.70.
1
1, CHCl3); H NMR (CDCl3): l 6.98–7.45 (m, 45 H,
D
-1,3,6-Tri-O-benzyl-myo-inositol 8a.—Compound
9×CH2Ph), 5.38 (br, ddd, 1 H, JP,H 9.2 Hz, H-2),
4.52–5.15 (m, 20 H, H-4, H-5, 9×CH2Ph), 3.89 (dd, 1
H, J1,6 9.5 Hz, H-6), 3.52 (dd, 1 H, J3,4 9.8 Hz, H-3),
3.51 (dd, 1 H, J1,2 2.1 Hz, H-1); 13C NMR (CDCl3): l
138.38, 136.79, 136.52, 136.04, 135.99, 135.91, 135.84,
135.75 (Cq, CH2Ph), 128.66, 128.51, 128.34, 128.25,
128.07, 128.00, 127.94, 127.77, 127.72, 127.68, 127.61,
127.23, 127.16, 127.08 (CH2Ph), 79.04, 78.27, 77.96,
77.50, 75.55, 72.46 (6×myo-inositol ring carbons),
74.56, 72.51, 72.18, 69.44, 69.40, 69.35, 69.15, 69.09,
69.02 (CH2Ph); 31P NMR (CDCl3): l −1.57, −1.65,
−1.87 (31P–1H decoupled).
7a (1.52 g, 2.20 mmol) was stirred in a mixture of
CH2Cl2–trifluoroacetic acid (55 mL, 10:1) for 30 min.
TLC (ether) showed the product Rf 0.20 and the depro-
tected p-methoxybenzyl group (Rf 0.70). The reaction
mixture was partitioned between H2O and CH2Cl2 (100
mL of each), dried (MgSO4), then evaporated to give
the crude product. The title compound 8a was purified
by flash chromatography (CHCl3–EtOAc, 1:1). Yield
(0.84 g, 85%); mp 109–112 °C (liquid crystal), 126–
127 °C (clear liquid, from EtOAc–hexane); [h]D +20°
(c 1, CHCl3), Lit.16 mp 112 °C (liquid crystal) 127 °C
(clear liquid); [h]D +16.2° (c 1, CHCl3); 1H NMR
(CDCl3): l 7.24–7.36 (m, 15 H, 3×CH2Ph), 4.76, 5.00
(AB, 2 H, JAB 11.2 Hz, CH2Ph), 4.62, 4.70 (AB, 4 H,
JAB 11.35 Hz, 2×CH2Ph), 4.20 (dd, 1 H, J2,3 2.6 Hz,
H-2), 3.94 (dd, 1 H, J1,6 9.5 Hz, H-6), 3.81 (dd, 1 H, J4,5
9.3 Hz, H-4), 3.34–3.41 (m, 2 H, overlapping, H-5,
H-3), 3.21 (dd, 1 H, J1,2 2.75 Hz, H-1), 2.85 (br s, 3 H,
exchangeable HO-2, HO-4 and HO-5); 13C NMR
(CDCl3): l 138.64, 137.73, 137.67 (Cq, CH2Ph), 128.51,
128.44, 128.39, 127.91, 127.83, 127.66 (CH2Ph), 80.41,
79.66, 78.96, 71.86, 66.91 (6×myo-inositol ring car-
L
-1,3,6-Tri-O-benzyl-2,4,5-tris-O-[di(benzyloxy)phos-
phoryl]-myo-inositol 9b.—A mixture of bis(benzy-
loxy)(diisopropylamino)phosphine25 (0.69 g, 2 mmol)
and 1H-tetrazole (0.42 g, 6 mmol) in dry CH2Cl2 (10
mL), was stirred for 15 min. L-1,3,6-Tri-O-benzyl-myo-
inositol 8b (0.22 g, 0.49 mmol), was then added to the
reaction mixture which was stirred for a further 15 min.
The solution was cooled to −78 °C and MCPBA (1.00
g, 2.81 mmol) was added and the mixture was allowed
to warm-up over 30 min. Work-up and purification was
carried out as for the
D-enantiomer. Yield (0.50 g,

1800
S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
83%); [h]D +9.3° (c 6.34, CHCl3). The NMR data were
identical with those of the -enantiomer.
-myo-Inositol 2,4,5-trisphosphate 4a.—D-1,3,6-Tri-
by flash chromatography (light petroleum–EtOAc, 5:1)
to furnish the title compound 9a as a syrup. Yield (0.25
g, 90%); [h]D −9.5° (c 1.7, CH2Cl2); 1H NMR
(CDCl3): l 6.88–7.40 (m, 45 H, 9×CH2Ph), 5.53 (br
ddd, JH,P 12.6 Hz, H-2), 4.45–5.33 (m, 20 H, H-4 and
H-5, 9×CH2Ph), 3.94 (dd, 1 H, J1,6 9.6 Hz, H-6), 3.60
(dd, 2 H, J3,4 9.5 Hz, H-1 and H-3); 13C NMR (CDCl3):
l 138.74, 136.81, 136.58, 136.06, 136.00, 135.92, 135.85,
135.61, 135.48 (Cq, CH2Ph), 128.78, 128.31, 128.23,
127.99, 127.94, 127.87, 127.84, 127.81, 127.74, 127.19,
126.84, 126.30 (CH2Ph), 79.89, 78.64, 78.51, 75.67,
73.97, 73.42 (6×myo-inositol ring carbons), 73.98,
72.90, 71.97, 69.70, 69.64, 69.56, 69.43, 69.34 (CH2Ph);
31P NMR (CDCl3): l 67.51, 68.29, 68.82; MS: (FAB)
m/z Calcd for C69H69O12P3S3 [M−H]− 1279.324.
Found 1279.326.
D
D
O-benzyl-2,4,5-tris-O-[di(benzyloxy)phosphoryl]-myo-
inositol 9a (0.115 g, 0.0935 mmol) was hydrogenolysed
in a mixture of methanol–water (4:1, 50 mL) over
palladium on carbon (10%, 0.20 g) at 30 psi overnight.
The reaction mixture was then filtered through a bed of
Celite to remove the solid components, and the remain-
ing solvents were evaporated to give a syrup. The
residue was then dissolved in MilliQ water (150 mL)
and purified by ion exchange chromatography on Q
Sepharose Fast flow, eluting with a gradient of TEAB
buffer (0–1000 mmol) at pH 8.6. The triethylammo-
nium salt of 4a eluted at ca. 500 mmol buffer. Yield
(0.090 mmol, 94%); [h]D −7.9° (c 1.52, MeOH) (tri-
ethylammonium salt); Lit.14 [h]546 −8.05° (H2O) (hex-
L
-1,3,6-Tri-O-benzyl-2,4,5-tris-O-[di(benzyloxy)thio-
1
akiscyclohexylammonium salt); H NMR (D2O): l 4.53
phosphoryl]-myo-inositol 10b.—Compound 10b was ob-
tained in an identical fashion to that described for 10a.
Yield (0.26 g, 93%); [h]D +10.0° (c 1.8, CH2Cl2); MS:
(FAB) m/z Calcd for C69H69O12P3S3 [M−H]−
1279.324. Found 1279.329.
(d, 1 H, JP,H 8.1 Hz, H-2), 4.26 (ddd, 1 H, J4,5 9.3 Hz,
H-4), 3.95 (ddd, 1 H, J5,6 9.0 Hz, H-5), 3.81 (dd, 1 H,
J1,6 9.7 Hz, H-6), 3.68 (br d, 1 H, J3,4 9.9 Hz, H-3), 3.57
(br d, 1 H, J1,2 9.9 Hz, H-1); 31P NMR (D2O): l 2.18
(d, 1 P, JP,H 6.7 Hz), 2.08 (d, 1 P, JP,H 10.1 Hz), 1.96 (d,
1 P, JP,H 6.7 Hz); MS: (FAB) m/z Calcd for C6H14O15P3
[M−H]− 418.9545. Found 418.9542.
D
-myo-Inositol 2,4,5-trisphosphorothioate 5a.—Am-
monia was condensed into a three-neck flask at
−78 °C. An excess of sodium was added to the liquid
ammonia then distilled into a second three-neck flask
and kept at −78 °C. Sodium was added until the
solution remained blue. Compound 10a (0.10 g, 0.08
mmol) was dissolved in dry 1,4-dioxane (2 mL) and
added to the blue solution of sodium in liquid ammo-
nia. The mixture was stirred for 10 min then quenched
with ethanol. The ammonia and ethanol were evapo-
rated off under a stream of nitrogen and the crude
product was purified by ion-exchange chromatography
on Q Sepharose Fast Flow with a gradient of TEAB
buffer (0.1–1.0 mol dm−3), pH 8.0, to give the triethy-
lammonium salt of the trisphosphorothioate 5a. Yield
L
-myo-Inositol 2,4,5-trisphosphate 4b.—L-1,3,6-Tri-O
- benzyl - 2,4,5 - tris - O - [di(benzyloxy)phosphoryl] - myo-
inositol 9b (0.215 g, 0.175 mmol) was hydrogenolysed
in a mixture of methanol–water (4:1, 50 mL) over
palladium on carbon (10%, 0.20 g) at 30 psi overnight.
The reaction mixture was then filtered over a bed of
Celite to remove the solid components, and the remain-
ing solvents were evaporated to give a syrup. Work-up
and purification was carried out in the same way as for
the
D-enantiomer. The title compound 4b eluted at ca.
600 mmol of TEAB buffer. Yield (0.090 mmol, 51%);
[h]D +8° (c 1.5, MeOH) (triethylammonium salt);
Lit.14 [h]546 +9.5° (H2O) (hexakiscyclohexylammo-
nium) salt. The mass spectrum and NMR data were
1
(0.062 g, 72%); [h]D −5.2° (c 0.8, MeOH); H NMR
(D2O): l 4.69 (br s, 1 H, H-2), 4.52 (ddd, 1 H, J4,5 9.6
Hz, JP,H 11.0 Hz, H-4), 4.14 (ddd, 1 H, J5,6 8.9, 9.1 Hz,
JP,H 10.9 Hz, H-5), 3.87 (dd, 1 H, J1,6 9.3 Hz, H-6), 3.68
(d, 1 H, J3,4 9.8 Hz, H-3), 3.51 (dd, 1 H, J1,2 3.4 Hz,
H-1); 31P NMR (D2O): l 44.33, 44.62, 44.88; MS:
(FAB) m/z Calcd for C6H14O12P3S3 [M−H]− 466.886.
Found 466.885.
identical with those of the
D-enantiomer; MS: (FAB)
m/z Calcd for C6H14O15P3 [M−H]− 418.954. Found
418.953.
D
-1,3,6-Tri-O-benzyl-2,4,5-tris-O-[di(benzyloxy)thio-
phosphoryl]-myo-inositol 10a.—A mixture of -1,3,6-
D
tri-O-benzyl-myo-inositol (0.10 g, 0.22 mmol), bis(ben-
zyloxy)diisopropylaminophosphine25 (0.46 g, 1.3 mmol)
and 1H-tetrazole (0.15 g, 2.14 mmol) in dry CH2Cl2 (10
mL) was stirred at room temperature for 30 min. The
solvent was evaporated and dry DMF (3 mL), dry
pyridine (1 mL), and sulphur (0.46 g) were added to the
residue and the mixture was stirred for a further 15
min. After this time the solvents were evaporated in
vacuo and the reaction mixture was dissolved in CHCl3
(30 mL) then washed with brine and water (30 mL of
each), dried (MgSO4), and the solvent was evaporated
to give the crude product. Purification was carried out
L
-myo-Inositol 2,4,5-trisphosphorothioate 5b.—Com-
pound 5b was obtained in an identical manner to that
described for compound 5a. Yield (0.06 g, 70%); [h]D
+5.9° (c 0.6, MeOH); MS: (FAB) m/z Calcd for
C6H14O12P3S3 [M−H]− 466.886. Found 466.885.
Acknowledgements
We thank the Wellcome Trust for Programme Grant
support (060554), Dr. A. Galione for a preliminary

S.J. Mills et al. / Carbohydrate Research 337 (2002) 1795–1801
1801
14. Tegge, W.; Denis, G. V.; Ballou, C. E. Carbohydr. Res.
1991, 217, 107–116.
15. Watanabe, Y.; Ogasawara, T.; Nakahira, H.; Matsuki,
T.; Ozaki, S. Tetrahedron Lett. 1988, 29, 5259–5262.
16. Desai, T.; Gigg, J.; Gigg, R.; Mart´ın-Zamora, E. Carbo-
hydr. Res. 1994, 262, 59–77.
biological assay and Dr Andrew Riley for useful
discussions.
References
17. (a) DeSolms, S. J.; Vacca, J. P.; Huff, J. R. Tetrahedron
Lett. 1987, 28, 4503–4506;
1. Berridge, M. J. Nature (London) 1993, 361, 315–325.
2. Wilcox, R. A.; Primrose, W. U.; Nahorski, S. R.; Chal-
liss, R. A. J. Trends Pharmacol. Sci. 1998, 19, 467–475.
3. Zhu, D. M.; Tekle, E.; Huang, C. Y.; Chock, P. B. J.
Biol. Chem. 2000, 275, 6063–6066.
4. Hermosura, M. C.; Takeuchi, H.; Fleig, A.; Riley, A. M.;
Potter, B. V. L.; Hirata, M.; Penner, R. Nature (London)
2000, 408, 735–740.
5. Potter, B. V. L.; Lampe, D. Angew. Chem., Int. Ed. Engl.
1995, 34, 1933–1972.
6. Liu, C.; Al-Hafidh, J.; Westwick, J.; Potter, B. V. L.
Bioorg. Med. Chem. 1994, 2, 253–257.
7. Mills, S. J.; Riley, A. M.; Murphy, C. T.; Bullock, A. J.;
Westwick, J.; Potter, B. V. L. Bioorg. Med. Chem. Lett.
1995, 5, 203–208.
8. Murphy, C. T.; Riley, A. M.; Mills, S. J.; Lindley, C. J.;
Potter, B. V. L.; Westwick, J. J. Mol. Pharmacol. 2000,
57, 595–601.
9. Irvine, R. F.; Brown, K. D.; Berridge, M. J. Biochem. J.
1984, 222, 269–272.
10. Burgess, G. M.; Irvine, R. F.; Berridge, M. J.; McKinney,
J. S.; Putney, J. W., Jr. Biochem. J. 1984, 224, 741–746.
11. Polokoff, M. A.; Bencen, G. H.; Vacca, J. P.; deSolms, S.
J.; Young, S. D.; Huff, J. R. J. Biol. Chem. 1988, 263,
11922–11927.
(b) DeSolms, S. J.; Vacca, J. P.; Huff, J. R.; Billington,
D. C.; Baker, R.; Kulagowski, J.; Mawer, I. Tetrahedron
1989, 45, 5679–5702.
18. Chung, S.-K.; Chang, Y.-Y.; Sohn, K.-W. J. Chem. Soc.,
Chem. Commun. 1996, 163–164.
19. Carless, H. A. J.; Busia, K. Tetrahedron Lett. 1990, 31,
3449–3452.
20. Bird, G. St. J.; Rossier, M. F.; Hughes, A. R.; Shears, S.
B.; Armstrong, D. L.; Putney, J. W., Jr. Nature (London)
1991, 352, 162–165.
21. Ryu, S. H.; Lee, S. Y.; Lee, K.-Y.; Rhee, S. G. FASEB J.
1987, 1, 388–393.
22. Marchant, J. S.; Chang, Y.-T.; Chung, S.-K.; Irvine, R.
F.; Taylor, C. W. Biochem. J. 1997, 321, 573–576.
23. Mills, S. J.; Potter, B. V. L. J. Chem. Soc., Perkin Trans.
1 1997, 1279–1286.
24. David, S.; Hanessian, S. Tetrahedron 1985, 41, 643–663.
25. Yu, K.-L.; Fraser-Reid, B. Tetrahedron Lett. 1988, 29,
979–982.
26. Lampe, D.; Liu, C.; Potter, B. V. L. J. Med. Chem. 1994,
37, 907–912.
27. Ukhanov, K.; Mills, S. J.; Potter, B. V. L.; Walz, B. Cell
Calcium 2001, 29, 335–345.
28. Nerou, P. E.; Riley, A. M.; Potter, B. V. L.; Taylor, C.
W. Biochem. J. 2001, 355, 59–69.
12. Willcocks, A. L.; Strupish, J.; Irvine, R. F.; Nahorski, S.
R. Biochem. J. 1989, 257, 297–300.
13. Hirata, M.; Watanabe, Y.; Ishimatsu, T.; Ikebe, T.;
Kimura, Y.; Yamaguchi, K. S.; Ozaki, S.; Kaga, T. J.
Biol. Chem. 1989, 264, 20303–20308.
29. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43,
2923–2925.
30. Tanaka, T.; Tamatsukuri, S.; Ikehara, M. Tetrahedron
Lett. 1986, 27, 199–202.