Rapid and practical synthesis of D-myo-inositol 1,4,5-trisphosphate

Article abstract of DOI:10.1039/a801874j

A concise synthetic sequence to biologically active O-myo-inositol 1,4,5-trisphosphate is described involving just five steps from myo-inositol and minimal chromatography with a key transformation of orthoacetate into acetate protection.

Full text of DOI:10.1039/a801874j

View Article Online / Journal Homepage / Table of Contents for this issue
Rapid and practical synthesis of D-myo-inositol 1,4,5-trisphosphate
Shane W. Garrett, Changsheng Liu, Andrew M. Riley and
Barry V. L. Potter*,†
Wolfson Laboratory of Medicinal Chemistry, Department of Pharmacy and Pharmacology,
University of Bath, Claverton Down, Bath, UK BA2 7AY
A concise synthetic sequence to biologically active D-myo-
inositol 1,4,5-trisphosphate is described involving just
five steps from myo-inositol and minimal chromatography
with a key transformation of orthoacetate into acetate
protection.
protecting group that could be later transformed by partial
hydrolysis into an acetate ester of defined regiochemistry such
as 8‡ (Scheme 1). Building upon the above we have now
developed a synthesis of chiral Ins(1,4,5)P3 involving only
five steps from myo-inositol. Furthermore, the methodology
developed has potential application in the synthesis of other
inositol phosphates. myo-Inositol orthoacetate 4§ was prepared
by modifying methodology established for the synthesis of
the orthoformate orthoester of myo-inositol.8 myo-Inositol was
treated with triethyl orthoacetate and PTSA in DMF. The acid
was removed by precipitation as a salt. After filtration and
evaporation in vacuo the residue was dissolved in hot methanol
and filtered to remove any unreacted myo-inositol. Recrystal-
lisation from methanol–chloroform–light petroleum followed
by further recrystallisation from methanol gave myo-inositol
orthoacetate 4 of sufficient purity for use in the next step, in 60–
65% yield. Orthoacetate 4 was treated with 2.1 equivalents of
1S-(Ϫ)-camphanoyl chloride and a catalytic amount of DMAP
in dichloromethane to yield the two diastereoisomeric di-
camphanates 5¶ and 6. In our synthesis of the enantiomers
of myo-inositol 1,3,4,5-tetrakisphosphate,6 careful chrom-
atography was required to separate the analogous 2,4- and 2,6-
diastereoisomeric dicamphanate esters of myo-inositol ortho-
formate. As a consequence of the orthoacetate functionally, the
corresponding diesters 5 and 6 were easier to crystallise and
thus were separable without chromatography. Crystallisation
from ethyl acetate gave the required 1-2,6-diester 5 in 37%
yield. Control over the selectivity of the desymmetrisation could
be achieved by changing the reaction conditions. An excess
-myo-Inositol 1,4,5-trisphosphate [Ins(1,4,5)P3, 1] is a cellular
second messenger first identified in 1983.1 Phospholipase C
catalyses hydrolysis of phosphatidylinositol 4,5-bisphosphate
to Ins(1,4,5)P3 and diacylglycerol. Ins(1,4,5)P3 interacts specif-
ically with a tetrameric receptor-operated Ca2ϩ channel on the
endoplasmic reticulum to mobilise Ca2ϩ stores in stimulated
cells.2 Ins(1,4,5)P3 mediates the agonist-induced response
via this rise in intracellular Ca2ϩ concentration. There is con-
tinuing interest in the biology of Ins(1,4,5)P3 and the many
other related inositol polyphosphates, and in the synthesis of
analogues that may offer the prospect of pharmacological
intervention in this signalling pathway.3
OH
OH
OH
–2O3PO
–2O3PO
–2O3PO
–2O3PO
–2O3PO
2–
2–
OPO3
OPO3
HO
HO
1
2
Structures of -Ins(1,4,5)P3 1 and -Ins(1,3,4,5)P4 2
Since Ozaki and co-workers first prepared optically active
Ins(1,4,5)P3 in 1986,4 many routes have been described for the
synthesis of enantiomerically pure Ins(1,4,5)P3 from diverse
starting materials.3 They are generally time-consuming long
linear sequences involving extensive chromatography. The
critical strategic points in most routes to chiral inositol phos-
phates are desymmetrisation of myo-inositol (a meso com-
pound) and resolution of an intermediate. The most rapid route
to chiral 1 hitherto described is probably that of Salamonczyk
and Pietrusiewicz.5 This route, however, involves a difficult
precipitation-driven equilibrium and is not readily modified to
provide materials leading to other inositol phosphate deriv-
atives. Here, we describe a shorter synthetic sequence which
obviates the need for tedious chromatographic separations and
long sequences of protecting group manipulations and which,
by modification of reaction conditions and/or isolation pro-
cedures, can provide material for alternative targets. The route
is also applicable to relatively large scale preparations.
‡ Data for compound 8: mp 226–229 ЊC; [α]D ϩ5.2 (c 1.9, DMF)
(Found: C, 57.5; H, 6.5. Calc. for C28H38O13: C, 57.63; H, 6.74%);
δH(400 MHz; [2H]7DMF; TMS) 0.96, 1.02, 1.05, 1.10, 1.12, 1.17 (18H,
6 × s, camph CH3), 1.55–1.67 (2H, m, camph CH2), 1.94–2.14 (4H, m,
camph CH2), 2.02 (3H, s, COCH3), 2.49–2.61 (2H, m, camph CH2),
3.61 (1H, dt, H-5, J 5.9, 9.3 Hz D2O, exchange gives t, J 9.8 Hz) 3.86
(1H, dt, H-4, J 4.9, 9.8 Hz, D2O exchange gives t, J 9.8 Hz), 4.15–4.17
(1H, m, H-1, D2O exchange gives dd, J 2.8, 10.4 Hz), 4.96 (1H, dd, H-3,
J 2.8, 10.1 Hz), 5.29 (1H, t, H-6, J 10.1 Hz), 5.58 (1H, d, 5-OH, J 5.4
Hz), 5.60 (1H, m, 4-OH), 5.62–5.63 (1H, m, H-2, D2O exchange gives
t, J 2.8 Hz), 5.75 (1H, d, 1-OH, J 5.9 Hz); m/z (FABϩ) [Found:
(M ϩ H)ϩ, 583.2394. C28H39O13 requires 583.2391].
§ Data for compound 4: mp 185–187 ЊC (with softening at 165 ЊC)
(Found: C, 47.0; H, 6.0. Calc. for C8H12O6: C, 47.06; H, 5.92%); δH(400
MHz; [2H]6DMSO) 1.28 (3H, s, CH3), 3.94–4.0 (4H, m, Ins-H), 4.21–
4.27 (2H, m, Ins-H), 5.19 (1H, s, 2-OH), 5.38 (2H, d, 4-OH and 6-OH,
J 5.2 Hz); δC(100 MHz; [2H]6DMSO) 24.30 (CH3), 57.70, 67.24, 69.24,
74.98 (Ins C), 107.66 (CCH3); m/z (FABϩ) 205 [(M ϩ H)ϩ, 100%].
¶ Data for compound 5: mp 228–231 ЊC; [α]D24 ϩ17 (c = 1, CH2Cl2)
(Found: C, 59.6; H, 6.45. Calc. for C28H36O12: C, 59.59; H, 6.43%);
δH(400 MHz; CDCl3; TMS) 0.98, 0.99, 1.08, 1.09, 1.10, 1.11 (18H, 6 × s,
camph CH3), 1.44 (3H, s, O3CCH3), 1.67–1.76 (2H, m, camph CH2),
1.91–2.10 (4H, m, camph CH2), 2.39–2.53 (2H, m, camph CH2), 3.23
(1H, d, 2-OH, J 6.4 Hz), 4.33–4.36 (1H, m, H-3), 4.39–4.43 (1H, m, H-1),
4.45–4.50 (1H, m, H-5), 4.57–4.62 (1H, m, H-4), 5.21–5.25 (1H, m, H-
2), 5.52–5.57 (1H, m, H-6); δC(100 MHz; CDCl3) 9.59, 9.66, 16.47,
16.54, 16.65 (camph CH3), 23.94 (O3CCH3), 28.76, 28.90, 30.44, 30.86
(camph CH2), 54.41, 54.51, 54.80, 54.89, 90.86, 91.03 (camph C), 63.72,
66.78, 68.15, 69.14, 69.30, 71.86 (ins C), 108.8 (O3C), 166.15, 166.81,
177.73, 178.02 (camph CO); m/z (FABϩ) 565 [(M ϩ H)ϩ, 100%].
We have recently published6 a rapid synthesis of -myo-
inositol 1,3,4,5-tetrakisphosphate [-Ins(1,3,4,5)P4, 2] and
its enantiomer -Ins(1,3,4,5)P4 by a chiral desymmetrisation
approach. In acid catalysed deprotection of intermediates utilis-
ing orthoformate protection7 we often observed small quan-
tities of formate esters as products of partial deprotection.
While these esters were not stable enough to be useful syn-
thetically, this finding did suggest that a different orthoester
such as an orthoacetate might be useful as an intermediate
† E-Mail: B.V.L.Potter@Bath.ac.uk
J. Chem. Soc., Perkin Trans. 1, 1998
1367

View Article Online
to make Ins(1,4,5)P3, could be crystallised directly from the
CH3
OH
OH
mixture in 25% yield using methanol–light petroleum (total
yield 31%). Pure 7 could be obtained in 40% yield by chrom-
atography of the mother liquor (RF 8 0.32, 7 0.47, [EtOAc]).
Structure determination of these acetate esters was straight-
O
O
i
HO
HO
O
OH
OH
HO
OH
OH
1
forward by H-1H COSY NMR experiments. No derivative
3
with a 5-acetate was obtained. Phosphitylation of 8 using
dibenzyl N,N-diisopropyl phosphoramidite9 followed by oxid-
ation of the intermediate phosphite triester with MCPBA gave
the protected phosphate 9 in 72% yield. One-pot deprotection
by hydrogenolysis at atmospheric pressure over 10% Pd/C,
then ester hydrolysis with ammonia at 60 ЊC gave Ins(1,4,5)P3.
Deprotection in this manner ensured that any risk of phosphate
migration was minimised. For biological evaluation, the
Ins(1,4,5)P3 1 was further purified by ion exchange chrom-
atography on Q-Sepharose Fast Flow resin eluting with a gradi-
ent of triethylammonium hydrogen carbonate buffer (0–100%
1 ) to give the triethylammonium salt in 70% yield (quantified
by Briggs phosphate assay10). A typical preparation generated
60 mg of this highly active target compound. Clearly the 2,4-
diester should provide the now well-established biological con-
trol -Ins(1,4,5)P311 via an analogous synthetic route. The other
readily available regioisomeric acetate 7 could find use in the
synthesis of further inositol polyphosphates or phospholipids.
Biological evaluation using saponin-permeabilised hepato-
cytes showed that synthetic 1 prepared via this route was indis-
4
ii
CH3
O
CH3
O
O
O
O
+
O
Ocamph
Ocamph
vi
Ocamph
OH
OH
Ocamph
6
5
Ocamph
OH
HO
iii
HO
HO
camphO
10
Ocamph
Ocamph
AcO
HO
HO
HO
HO
HO
+
OH
OAc
camphO
camphO
8
7
iv
tinguishable in its intracellular Ca2ϩ release profile (EC50
121 n) from sample of commercially available
Ins(1,4,5)P3 (EC50 = 122 8 n).
=
2
a
Ocamph
OR
OH
OH
AcO
v
–2O3PO
–2O3PO
RO
RO
In summary, we have described a novel route to this highly
potent second messenger in chiral form with considerable
advantages over currently established routes in terms of facility,
speed and potential for large scale synthesis.
2–
OPO3
camphO
HO
9 R = PO(OBn)2
1
O
Acknowledgements
We thank the Wellcome Trust (Programme Grant 045491) for
financial support and Dr C. W. Taylor, University of
Cambridge, for biological evaluation of synthetic material.
camph =
O
O
Scheme 1 Reagents and conditions: i, Triethyl orthoacetate, PTSA,
DMF, 90–100 ЊC, 4 h; ii, 1S-(Ϫ)-camphanoyl chloride, DMAP, dichloro-
methane, 0 ЊC then RT, 2 h; iii, 80% aqueous trifluoroacetic acid, 7 days;
iv (a), (BnO)2PNPri2, 1H-tetrazole, dichloromethane, RT, 1 h; (b),
MCPBA, dichloromethane, Ϫ78 ЊC to RT; v (a), H2, 10% Pd on C,
MeOH–H2O, RT, 12 h; (b), conc. aqueous NH3, 60 ЊC, 4 h; vi, 5  HCl–
MeOH 1:11, reflux, 14 h
References
1 H. Streb, R. F. Irvine, M. J. Berridge and I. Schulz, Nature, 1983,
306, 67.
2 M. J. Berridge, Nature, 1993, 361, 315.
3 B. V. L. Potter and D. Lampe, Angew. Chem., Int. Ed. Engl., 1996,
34, 1933.
of the 2,4-diester 6 was gained by omitting the DMAP, and
crystallisation from methanol gave 6 in 40–45% yield. The
regiochemistry of the required 2,6-dicamphanate 5 was deter-
mined by removing the orthoacetate completely by refluxing in
methanol–HCl to give tetrol 10. The [α]D value of 10 was found
to be identical to that of authentic material {[α]D25 Ϫ8 (c 1,
DMF)} synthesised from the analogous 2,6-dicamphanate
with orthoformate protection, a compound whose structure
had been unambiguously determined previously by X-ray
crystallography.6
Selective transformation of 5 to 8 was required to produce
the appropriately protected intermediate for the synthesis of 1.
Whereas acidic hydrolysis of the analogous orthoformate, as
described in our synthesis of 2, resulted in complete loss of the
protection afforded by this functionality, treatment of 5 with
80% aqueous trifluoroacetic acid at room temperature for one
week yielded the acetate esters 7 and 8. Regioisomer 8, required
4 S. Ozaki, Y. Watanabe, T. Ogasawara, Y. Kondo, N. Shiotani,
H. Nishii and T. Matsuki, Tetrahedron Lett., 1986, 27, 3157.
5 G. M. Salamonczyk and K. M. Pietrusiewicz, Tetrahedron Lett.,
1991, 32, 6167.
6 A. M. Riley, M. F. Mahon and B. V. L. Potter, Angew. Chem., Int.
Ed. Engl., 1997, 36, 1472.
7 H.-W. Lee and Y. Kishi, J. Org. Chem., 1985, 50, 4402.
8 S. Ozaki, Y. Koga, Y. Watanabe, Y. Kimura and M. Hirata, Bull.
Chem. Soc. Jpn., 1994, 67, 1058.
9 K. L. Yu and B. Fraser-Reid, Tetrahedron Lett., 1988, 29, 979.
10 A. P. Briggs, J. Biol. Chem., 1922, 53, 13.
11 J. Strupish, A. M. Cooke, B. V. L. Potter, R. Gigg and S. R.
Nahorski, Biochem. J., 1988, 253, 901.
Paper 8/01874J
Received 6th March 1998
Accepted 6th March 1998
1368
J. Chem. Soc., Perkin Trans. 1, 1998