Practical unequivocal synthesis of phosphatidyl-myo-inositols

Article abstract of DOI:10.1016/S0040-4039(99)02207-8

The direct phosphatidylation of 1D-2,3,4,5,6-penta-O-benzyl-myo-inositol with sn-3-phosphatidic acid and subsequent hydrogenolytic debenzylation produces 1D-1-(sn-3-phosphatidyl)-myo-inositol in excellent yield (>90%) and unequivocal structural and stereo

Full text of DOI:10.1016/S0040-4039(99)02207-8

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 847–850
Practical unequivocal synthesis of phosphatidyl-myo-inositols
Rajindra Aneja ∗ and Sarla G. Aneja
Functional Lipids Division, Nutrimed Biotech, Langmuir Laboratory, Cornell University Research Park, Ithaca,
NY 14850-1257, USA
Received 8 October 1999; revised 22 November 1999; accepted 23 November 1999
Abstract
The direct phosphatidylation of 1D-2,3,4,5,6-penta-O-benzyl-myo-inositol with sn-3-phosphatidic acid and sub-
sequent hydrogenolytic debenzylation produces 1D-1-(sn-3-phosphatidyl)-myo-inositol in excellent yield (>90%)
and unequivocal structural and stereochemical purity, and, is readily adaptable for large scale production. © 2000
Elsevier Science Ltd. All rights reserved.
1D-1-(1-Fattyacyl0-2-fattyacyl00-sn-glycero-3-phospho)-myo-inositols 1a based on diverse long chain
fattyacyls, commonly referred to as phosphatidylinositols (PtdIns), constitute an important group of
biological phospholipids.1 As the parent participant in the phosphatidylinositol metabolic and signaling
cycle, PtdIns regulate vital normal cellular functions including growth, multiplication, survival, and
apoptosis, as well as abnormal states exemplified by cancers and diabetes.2 PtdIns are important also
for the development of diagnostics and therapeutics based on signaling modalities,3 and bio-compatible
amphiphiles for skincare and intravenous drug delivery formulations.4 Large quantities of individual
molecular species of PtdIns with saturated fattyacyls are required for the latter applications. Hence
there is a considerable interest and all recent syntheses have targeted PtdIns with saturated fattyacyls.5
However, no practical approach adaptable for economical large scale synthesis is available. Moreover, the
molar rotation values of some synthetic PtdIns differ significantly from the bench mark value provided
by natural PtdIns.
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02207-8
tetl 16135

848
We now report a new convergent synthesis based on the direct phosphatidylation of a penta-O-
protected myo-inositol with sn-phosphatidic acid as the lipid synthon. This contrasts with the previous
syntheses which all use 1,2-diacyl-sn-glycerol, a synthon prone to racemization via neighboring acyl mi-
grations, and, introduce the phosphate group in multiple steps using trivalent and pentavalent phosphorus
reagents.6 The three key steps in the new synthesis are outlined in Scheme 1. The starting material
1D-1-O-allyl-2,3:5,6-di-O-cyclohexylidene-myo-inositol 2 was obtained from myo-inositol in four steps
as described.7 The reaction of 2 successively with (i) acetic acid–water (80:20) at 100°C, evaporation
to dryness, (ii) complete benzylation in DMF using NaH/BnBr and evaporation under 2 millitorr to
remove all volatiles, and (iii) deallylation by successive heating with t-BuOK in DMSO followed by
acetic acid–water (80:20), in a one pot procedure, gave (−)-2,3,4,5,6-penta-O-benzyl-myo-inositol 3 in
95% yield. The absolute configuration of 3 has been established unequivocally as 1D-2,3,4,5,6-penta-O-
benzyl-myo-inositol.8 The chiral lipid synthons 1,2-difattyacyl-sn-glycero-3-phosphoric acids 4a–c (sn-
3-PA) were prepared from the corresponding sn-phosphatidylcholines by lipolysis with phospholipase
D.9 The crucial esterification reaction of the secondary 1-hydroxyl group in 3 with the phosphoric acid
in 4 gave less than 50% yield when phosphatidylation was conducted under conditions recommended
for primary alcohols10 but a dramatic, highly reproducible increase in yield to 90% was achieved by the
following protocol. Typically, 0.15 mmole of sn-3-PA 4 was added gradually over 1 h at 35–37°C to
a stirred solution of the alcohol 3 (0.1 mmole) and triisopropylbenzene sulfonyl chloride (TPSCl, 0.4
mmole) in anhydrous pyridine. Thus, reaction with dioctanoyl sn-3-PA 4b, work-up by addition of water
and purification by flash chromatography on silica, gave 1D-1-(1,2-dioctanoyl-sn-glycero-3-phospho)-
2,3,4,5,6-penta-O-benzyl-myo-inositol 5b,11 [α]D +12.79 (c 1.04, CHCl3), in 90% yield. Reaction of
5b in tert-butanol with H2 at 45 psi and Pd–C catalyst caused complete debenzylation and produced
a quantitative yield of the short acyl chain 1D-1-(1,2-dioctanoyl-sn-glycero-3-phospho)-myo-inositol
(dioctanoyl PtdIns) 1b,12 [α]D +8.80 (c 0.45, CHCl3:CH3OH 4:1), molar rotation [Φ] +51.57. Similarly,
use of distearoyl sn-3-PA 4c in Scheme 1 yielded the long acyl chain distearoyl PtdIns 1c, [α]D +5.97 (c
0.55, CHCl3:CH3OH 4:1), [Φ] +51.70, in yield comparable with 1b.
The absolute configurations of 3 and 4 are well established and are expected to be delivered unam-
biguously to the target sn-3-phosphatidyl-D-myo-inositols during coupling by esterification, and re-
tained during hydrogenolysis, leading unequivocally to 1D-1-(sn-3-phosphatidyl)-myo-inositol absolute
1
stereochemistry for 1b and 1c. The H, 13C and 31P NMR spectra are consistent with this. The newly
prepared compounds 1b and 1c and other PtdIns differ in fattyacyl type and hence molecular weight. To
normalize these differences in the homologous series, we compared the molar rotations of our synthetics
with the values for natural PtdIns as the bench mark for 1D-1-(sn-3-phosphatidyl)-myo-inositol absolute
configuration. The molar rotation values of 1b [Φ] +51.57 and 1c [Φ] +51.70 are virtually identical with
the values for PtdIns from soybean [Φ] +51.77, [α]D +6.20 (c 0.51, CHCl3:CH3OH 4:1), and bovine
brain [Φ] +51.01, [α]D +5.75 (c 0.40, CHCl3:CH3OH 4:1).

849
Scheme 1.
Unequivocal synthesis of PtdIns has been a problem. For instance, we have shown previously8 that
myo-inositol synthons belonging to the 1L- absolute configuration series were used in some recent
syntheses5g,5h which claimed 1D-1 series PtdIns products. Consistent with this, the molar rotation values
calculated from the published specific rotations of ditetradecanoyl PtdIns [Φ] +101.10, [α]D +13.4,5g and
diheptanoyl PtdIns [Φ] +83.21, [α]D +14.2,5h differ significantly from the bench mark value [Φ] +51 for
(sn-3-phosphatidyl)-myo-inositols. Rotation data for synthetic PtdIns were not reported in other recent
literature.5d,5e,5i
Large quantities of the starting materials 2 and 4a–c are readily obtained in high structural and optical
purity, and the revised protocol for TPSCl mediated condensation gives very high yields. The unit
operations of Scheme 1 are simple and well suited for scale up, and the TPSCl reagent is inexpensive.
Thus, the synthesis described herein provides the very first practical unequivocal synthesis of saturated
fattyacyl type PtdIns adaptable for large scale production. In contrast with the published syntheses of
PtdIns which require oxidation of H-phosphonate or phosphite intermediates to phosphate, Scheme 1 is
potentially adaptable for unsaturated fattyacyl type PtdIns using analogues of 3 with alternative protecting
groups which do not require hydrogenolysis. Work on scale up and extension to unsaturated fattyacyl type
PtdIns will be reported.
Acknowledgements
This work was supported by PHS/NIH grant GM59550 (to R.A.). We thank Wenyan Zhu and Jed-
Sian Cheng for experimental assistance, Dave Fuller (Baker Laboratory, Cornell) for NMR and Brandon
Roach (Phyton, Inc.) and Robert W. Sherwood (Biotechnology, Cornell) for MS data.

850
References
1. Carter, H. E.; Johnson, P.; Weber, E. J. Annu. Rev. Biochem. 1965, 34, 109–119.
2. For reviews, see: (a) Berridge, M. J. Nature 1993, 361, 315. (b) Lee, S. B.; Rhee, S. G. Curr. Opin. Cell Biol. 1995, 7,
183–185. (c) Lipid Second Messengers-Handbook of Lipid Research; Bell, R. M.; Exton, J. H.; Prescott, S. M., Eds.; Plenum
Press: New York, NY 1996; Vol. 8. (d) Brodbeck, U. Biol. Chem. 1998, 379, 1041–1044. (e) Rameh, L. E.; Cantley, L. C.
J. Biol. Chem. 1999, 274, 8347–8350. (f) Katan, M. Biochim. Biophys. Acta 1998, 1436, 5–17. (g) Exton, J. H. Biochim.
Biophys. Acta 1998, 1436, 105–115. (h) Fruman, D. A.; Meyers, R. E.; Cantley, L. C. Annu. Rev. Biochem. 1998, 67,
481–507. (i) Bottomley, M. J.; Salim, K.; Panayotou, G. Biochim. Biophys. Acta 1998, 1436, 165–183.
3. (a) Jett, M.; Chudzik, J.; Alving, C. R.; Stanacev, N. Z. Cancer Res. 1985, 45, 4810–4815. (b) Stralfors, P. Bioessays 1997,
19, 327–335. (c) Leung, L. W.; Lin, W. Y.; Richard, C.; Bittman, R.; Arthur, G. J. Liposome Res. 1998, 8, 213–218.
4. Lee, K. D.; Hong, K.; Papahadjopoulos, D. Biochim. Biophys. Acta 1992, 1103, 185–197.
5. For syntheses of PtdIns, see: (a) Klyashchitskii, B. A.; Zhelvakova, E. G.; Shvets, V. I.; Evstigneeva, R. P.; Preobranzhenskii,
N. A. Tetrahedron Lett. 1970, 587–590. (b) Molotkovsky, J.-G.; Bergelson, L. D. Chem. Phys. Lipids 1973, 11, 135–147.
(c) Shvets, V. I.; Klyashchitskii, B. A.; Stepanov, A. E.; Evstigneeva, R. P. Tetrahedron 1973, 29, 331–340. (d) Jones, M.;
Rana, K. K.; Ward, J. G. Tetrahedron Lett. 1989, 30, 5353–5356. (e) Young, R. C.; Downes, C. P.; Eggleston, D. S.; Jones,
M.; Macphee, C. H.; Rana, K. K.; Ward, J. G. J. Med. Chem. 1990, 33, 641–646. (f) Salamonczyk, M.; Bruzik, K. S.
1990, 2015–2016. (g) Murakata, C.; Ogawa, T. Tetrahedron Lett. 1991, 32, 101–104. (h) Garigapati, V. R.; Roberts, M. F.
Tetrahedron Lett. 1993, 29, 769–772. (i) Leung, L. W.; Vilcheze, C.; Bittman, R. Tetrahedron Lett. 1998, 39, 2921–2924.
6. For a discussion, see: (a) Aneja, S. G.; Parra, A.; Stoenescu, C.; Xia, W.; Aneja, R. Tetrahedron Lett. 1997, 38, 803–807.
(b) Aneja, R.; Aneja, S. G. In Advances in Phosphoinositides; Bruzik, K. S., Ed.; ACS Symposium Series 718; American
Chemical Society: Washington, DC, 1998; pp. 224–231.
7. Aneja, R.; Aneja, S. G.; Parra, A. Tetrahedron Lett. 1996, 37, 5081–5082. Aneja, R. US Patent Application 08/862,865
Allowed May 1999.
8. Systematic name 1L-1,2,4,5,6-penta-O-benzyl-myo-inositol; Aneja, R.; Aneja, S.; Pathak, V. P.; Ivanova, P. T. Tetrahedron
Lett. 1994, 35, 6061–6062.
9. Aneja, R. Biochem. Soc. Trans. 1974, 2, 38–42.
10. Aneja, R.; Chadha, J. S.; Davies, A. P. Tetrahedron Lett. 1969, 4183–4186. Aneja, R.; Chadha, J. S.; Davies, A. P. Biochim.
Biophys. Acta 1970, 218, 102–110.
11. All new compounds were characterized fully including high resolution and multidimensional 1H, 13C and 31P NMR,
ESMS, and optical rotation data. [α]D is expressed without units, and the actual units, deg mL/(g dm), are understood;
all measurements are at rt (20–22°C). Molar rotations of natural PtdIns are based on molecular weights calculated from their
fattyacyl compositions.
12. DiC8:0PtdIns (+)-1b: ES(−) MS m/z 585.0 (M−H)−; 1H NMR (400 MHz, CDCl3:CD3OD, 2:1) δ ppm 0.90 (t, 6H, CH3),
1.29 (br s, 16H, (CH2)8), 1.61 (m, 4H, CH2CH2C_O), 2.29–2.36 (m, 4H,CH2C_O), 3.30 (t, 1H, inositol 5-H), 3.46–3.48 (dd,
1H, inositol 3-H), 3.67(m, 1H, inositol 4-H), 3.80 (m, 1H, inositol 6-H), 3.92 (t, 1H, inositol 2-H), 4.05–4.17 (m, 2H, sn-3
CH2), 4.28, 4.45 (m, 2H, sn-1 CH2), 4.28 (t, 1H, inositol 1-H), 5.26 (m, 1H, sn-2 CH); 13C NMR (100 MHz, CDCl3:CD3OD,
2:1) 173.69 and 173.35 (2 C_O), 76.24 (inositol C-2), 74.12 (inositol C-5), 72.10 (inositol C-4), 71.35 (inositol C-6), 71.19
and 70.98 (inositol C-1), 70.14 and 70.06 (sn-2 glycerol C), 63.46 (sn-3 glycerol C), 62.39 (sn-1 glycerol C), 33.82 (acyl
chain sn-2 α-CH2), 33.68 (acyl chain sn-1 α-CH2), 31.29, 31.28, 24.51 and 24.45 ((CH2)n), 22.19 (acyl chain ω-CH2),
13.46 (acyl chain CH3); 31P NMR (162 MHz, CDCl3:CD3OD, 2:1) δ ppm (external H3PO4 ref.) −0.258 (s).