Synthesis of mycobacterial triacylated phosphatidylinositol dimannoside containing an acyl lipid chain at 3-O of inositol

Article abstract of DOI:10.1021/ol1008137

A seven-step synthesis of triacylated phosphatidylinositol dimannoside is described from myo-inositol 1,3,5-orthoformate. It proceeded in 31% overall yield via a highly regioselective and stereoselective 2,6-di-O-d-mannosylation as the key step.

Full text of DOI:10.1021/ol1008137

ORGANIC
LETTERS
2010
Vol. 12, No. 11
2618-2621
Synthesis of Mycobacterial Triacylated
Phosphatidylinositol Dimannoside
Containing an Acyl Lipid Chain at 3-O
of Inositol
Pratap S. Patil and Shang-Cheng Hung*
Genomics Research Center, Academia Sinica, 128, Section 2, Academia Road, Taipei
11529, Taiwan, and Department of Chemistry, National Tsing Hua UniVersity, 101,
Section 2, Kuang-Fu Road, Hsinchu 30043, Taiwan
schung@gate.sinica.edu.tw
Received April 9, 2010
ABSTRACT
A seven-step synthesis of triacylated phosphatidylinositol dimannoside is described from myo-inositol 1,3,5-orthoformate. It proceeded in
31% overall yield via a highly regioselective and stereoselective 2,6-di-O-D-mannosylation as the key step.
Mycobacterium tuberculosis (Mtb), an acid-fast Gram-
positive bacterium that persists in infected macrophages for
a long-term period, is responsible for the development of
the active disease tuberculosis (TB).1 Although the current
BCG vaccine is effective for preventing the incidence of
miliary TB in children, it is not effective at protecting adults
from TB infection. Due to the spread of multidrug-resistant
Mtb strains and the increase of patients with lethal HIV-TB
co-infection, the development of new drugs or vaccines to
combat TB is indeed important.
The Mtb cell envelope, consisting of outer capsule, cell
wall, and plasma membrane,2 is responsible for the initial
stage of host cell infection. It forms a thick protective barrier
to prevent antibiotic penetration3 and manipulates the host
immune system in a way that develops a favorable environ-
ment for bacterial survival and growth.4 The covalently ester-
linked mycolic acids and peptidoglycan-arabinogalactans are
the major components of cell wall, while phosphatidylinositol
D-mannosides (PIMs, Figure 1), lipomannan (LM), lipoara-
binomannan (LAM), and mannose-capped lipoarabinoman-
nan (ManLAM) are noncovalently attached to the plasma
membrane or outer capsule through their phosphatidylinositol
anchor containing palmitic, stearic, and tuberculostearic acid
residues. Among these components, PIMs are ubiquitously
distributed in both pathogenic and nonpathogenic species of
mycobacteria and play significant roles in cell-wall biogen-
esis5 and in many immunomodulatory events, including
neutralization of potentially cytotoxic O2 free radicals,6
(3) (a) Bhowruth, V.; Alderwick, L. J.; Brown, A. K.; Bhatt, A.; Besra,
G. S. Biochem. Soc. Trans. 2008, 36, 555–565. (b) Jozefowski, S.; Sobota,
A.; Kwiatkowska, K. BioEssays 2008, 30, 943–954.
(4) (a) Rojas, R. E.; Thomas, J. J.; Gehring, A. J.; Hill, P. J.; Belisle,
J. T.; Harding, C. V.; Boom, W. H. J. Immunol. 2006, 177, 2959–2968. (b)
Russell, D. G. Nat. ReV. Microbiol. 2007, 5, 39–47.
(1) Bloom, B. R.; Murray, C. J. L. Science 1992, 257, 1055–1064.
(2) The Mycobacterial Cell EnVelope; Daffe, M., Reyrat, J.-M., Eds.;
ASM Press: Washington DC, 2008.
(5) Alderwick, L. J.; Birch, H. L.; Mishra, A. K.; Eggeling, L.; Besra,
G. S. Biochem. Soc. Trans. 2007, 35, 1325–1328.
10.1021/ol1008137  2010 American Chemical Society
Published on Web 05/05/2010

Scheme 1. Retrosynthesis of the Triacylated PIM2 1
Figure 1. Structures of phosphatidylinositol dimannoside (PIM2)
and phosphatidylinositol hexamannoside (PIM6). R1, R2: palmitic
acid, stearic acid, or tuberculostearic acid. R3, R4: H, palmitic acid
or stearic acid.
induction of cytokines,7 phagocytosis of organism by binding
with the mannose receptors,8 and growth of organism in the
host macrophages.9 Different structures of PIMs impact the
recognition and response of the host cell and influence the
intercellular fate of pathogens.10 PIMs also interact with
VLA-5 on CD4+ T lym-phocytes and activate integrins for
promoting adhesion to the extracellular matrix glycopro-
teins.11 A recent report has revealed that PIM6 can stimulate
the CD1b-restricted T cells, and the partial digestion of the
oligomannose moiety assisted by CD1e for this process is
essential.12
PIM2, the smallest precursor for the biosynthesis of highly
D-mannosylated molecules, has a diacylated phosphatidyl
lipid moiety at the 1-O position of myo-inositol and two
R-linked D-mannose sugars at the 2-O and 6-O positions.
Elongation at the 6′-O position of the 6-O-linked D-mannosyl
unit by an R1f6 D-mannosyl core, branched with R1f2-
linked D-mannose, forms the higher PIMs and LM, which
can be further extended by an arabinan domain to give LAM
and ManLAM. The degree of acylation could vary the
structures and functions of PIMs, LM, LAM, and Man-
LAM.13 Acylations at the primary hydroxy group of the 2-O-
linked D-mannosyl residue and/or at the 3-O position of myo-
inositol residue leads to the triacylated and tetraacylated
derivatives.
cance of the acyl lipid chains.14 Owing to the structural
complexity and potential immunotherapeutics,15 the synthesis
of diacylated and triacylated PIMs and LM with a lipid chain
at 6′-O of the D-mannosyl ring has been documented in the
literature.15a,16 No report has addressed the preparation of
triacylated and tetraacylated PIMs or LM containing an acyl
chain at the 3-O position of myo-inositol. In continuation of
our interest in the biosynthesis of PIMs and the development
of anti-TB vaccines, we herein report the first synthesis of
the triacylated PIM2 1.
The generation of optically pure myo-inositol derivatives
with appropriate protecting groups is a challenging task.17
Typical procedures include multistep synthesis from D-
glucose via Ferrier reaction,18 resolution of racemic myo-
Alkaline hydrolysis of PIMs and LAM abolishing many
immunoregulatory effects has clearly indicated the signifi-
(14) Gilleron, M.; Ronet, C.; Mempel, M.; Monsarrat, B.; Gachelin, G.;
Puzo, G. J. Biol. Chem. 2001, 276, 34896–34904.
(15) (a) Boonyarattanakalin, S.; Liu, X.; Michieletti, M.; Lepenies, B.;
Seeberger, P. H. J. Am. Chem. Soc. 2008, 130, 16791–16799. (b) Parlane,
N. A.; Denis, M.; Severn, W. B.; Skinner, M. A.; Painter, G. F.; La Flamme,
A. C.; Ainge, G. D.; Larsen, D. S.; Buddle, B. M. Immunol. InVest. 2008,
37, 129–142.
(6) Chan, J; Fan, X; Hunter, S. W.; Brennan, P. J.; Bloom, B. R. Infect.
Immun. 1991, 59, 1755–1761.
(7) Barnes, P. F.; Chatterjee, D.; Abrams, J. S.; Lu, S.; Moskowitz, N.;
Zhang, Y.; Rom, W. N. J. Immunol. 1992, 149, 541–547.
(8) Kang, B. K.; Schlesinger, L. S. Infect. Immun. 1998, 66, 2769–2777.
(9) Hestvik, A. L. K.; Hmama, Z.; Av-Gay, Y. FEMS Microbiol. ReV.
2005, 29, 1041–1050.
(16) (a) Elie, C. J. J.; Verduyn, R.; Dreef, C. E.; Brounts, D. M.;
Vandermarel, G. A.; van Boom, J. H. Tetrahedron 1990, 46, 8243–8254.
(b) Watanabe, Y.; Yamamoto, T.; Ozaki, S. J. Org. Chem. 1996, 61, 14–
15. (c) Watanabe, Y.; Yamamoto, T.; Okazaki, T. Tetrahedron 1997, 53,
903–918. (d) Jayaprakash, K. N.; Lu, J.; Fraser-Reid, B. Bioorg. Med. Chem.
Lett. 2004, 14, 3815–3819. (e) Stadelmaier, A.; Biskup, M. B.; Schmidt,
R. R. Eur. J. Org. Chem. 2004, 3292–3303. (f) Jayprakash, K. N.; Lu, J.;
Fraser-Reid, B. Angew. Chem., Int. Ed. 2005, 44, 5894–5898. (g) Liu, X.;
Stocker, B. L.; Seeberger, P. H. J. Am. Chem. Soc. 2006, 128, 3638–3648.
(h) Fraser-Reid, B.; Lu, J.; Jayprakash, K. N.; Lopez, C. Tetrahedron:
Asymmetry 2006, 17, 2449–2463. (i) Ho¨lemann, A; Stocker, B. L.;
Seeberger, P. H. J. Org. Chem. 2008, 71, 8071–8088. (j) Fraser-Reid, B.;
Chaudhuri, S. R.; Jayaprakash, K. N.; Lu, J.; Ramamurty, C. V. S. J. Org.
Chem. 2008, 73, 9732–9743. (k) Patil, P. S.; Hung, S.-C. Chem.sEur. J.
2009, 15, 1091–1094. (l) Ali, A.; Wenk, M. R.; Lear, M. J. Tetrahedron
(10) Torrelles, J. B.; Azad, A. K.; Schlesinger, L. S. J. Immunol. 2006,
177, 1805–1816.
(11) Rojas, R. E.; Thomas, J. J.; Gehring, A. J.; Hill, P. J.; Belisle, J. T.;
Harding, C. V.; Boom, W. H. J. Immunol. 2006, 177, 2959–2968.
(12) de la Salle, H.; Mariotti, S.; Angenieux, C.; Gilleron, M.; Garcia-
Alles, L.-F.; Malm, D.; Berg, T.; Paoletti, S.; Maitre, B.; Mourey, L.;
Salamero, J.; Cazenave, J. P.; Hanau, D.; Mori, L.; Puzo, G.; De Libero,
G. Science 2005, 310, 1321–1324.
(13) (a) Chatterjee, D.; Khoo, K.-H. Glycobiology 1998, 8, 113–120.
(b) Nigou, J.; Gilleron, M.; Puzo, G. Biochimie 2003, 85, 153–166. (c)
Karakousis, P. C.; Bishai, W. R.; Dorman, S. E. Cell. Microbiol. 2004, 6,
105–116. (d) Briken, V.; Porcelli, S. A.; Besra, G. S.; Kremer, L. Mol.
Microbiol. 2004, 53, 391–403.
Lett. 2009, 50, 5664–5666
.
Org. Lett., Vol. 12, No. 11, 2010
2619

inositol compounds using chiral auxiliaries,19 intramolecular
olefin metathesis of the 1,7-octadiene derivatives followed
by dihydroxylation,20 and desymmetrization of myo-inositol
employing peptide-based catalysts.21 We recently found that
sugars were useful chiral agents for desymmetrization in
natural product synthesis.16k,22 Conceptually, two D-mannosyl
units could be simultaneously installed at the 2-O and 6-O
positions of the meso myo-inositol-derived 2,4,6-triol. This
desymmetrized approach would straightforwardly afford the
chiral pseudotrisaccharide core. Scheme 1 illustrates our
retrosynthetic design for the triacylated PIM2 1. The target
molecule 1 could be obtained by coupling of the H-
phosphonate 2 with the 1-alcohol 3 followed by global
deprotection of all benzyl groups. Due to the steric effect
created by the bulkiness of two adjacent 2-O- and 6-O-linked
D-mannosyl rings, the environment of the 1-C-hydroxyl
group in the myo-inositol ring of 1,3-diol 4 is more hindered
than the hydroxyl at the 3-C position. The stearyl group could
be regioselectively introduced at the 3-C-hydroxyl of 4 to
get the 1-alcohol 3. In addition, the 2-O-attached mannosyl
ring and the 3-C-hydroxyl orient toward the same ꢀ-face and
the 4-C- and 5-C-hydroxyls are more reactive because of
the steric effect. Regioselective 4,5-di-O-benzylation of
1,3,4,5-tetraol 5 would yield the desired diol 4. The prepara-
tion of tetraol 5 could be carried out via hydrolysis of the
orthoformate group in the 4-alcohol 6 under mild acidic
conditions. Regioselective and stereoselective coupling of
the D-mannosyl trichloroacetimidate 723 with commercially
available myo-inositol 1,3,5-orthoformate 8 would furnish
the pseudotrisaccharide 6 in a one-pot manner. Theoretically,
the first D-mannosyl unit could be assembled at the 2-C-
equatorial hydroxyl followed by desymmetric installation of
the second D-mannosyl ring at the 6-C-axial hydroxyl.
Although the strategy for direct glycosidation of 2,4,6-
triol 8 is promising,24 its low solubility in most common
organic solvents, e.g., dichloromethane, ethyl ether, aceto-
nitrile, toluene, and nitromethane, is a big concern. Com-
Scheme 2. Desymmetrization of the Meso Triol 8
pound 8 is slightly soluble in THF and completely dissolves
in 1,4-dioxane. Since the melting point of dioxane is 11 °C
and the temperature for most sugar coupling reactions is
below -20 °C, a mixed solvent of dioxane with dichlo-
romethane or tetrahydrofuran was selected for further
investigation. As outlined in Scheme 2, BF3·OEt2-activated
coupling of 8 with 1.2 equiv of the D-mannosyl donor 7 in
a 1:1 ratio of 1,4-dioxane and dichloromethane at -40 °C
for 1 h gave the 2-O-mannosylated 4,6-diol 9, the 2,6-di-
O-mannosylated 4-alcohol 6, and its diastereoisomer 10 in
67%, 5%, and 3% isolated yields, respectively. This fact has
revealed that the 2-C-equatorial hydroxy group of 8 is more
reactive than the others at 4-C and 6-C. When 2 equiv of
the donor 7 was used in the reaction, the yield of 6 slightly
increased, but the mono-D-mannosylated compound 9 was
acquired as the major adduct. To realize the lower reactivity
of the axial hydroxyls, excess donor and higher reaction
temperature were employed to improve the results. The
optimized 64% yield of 6, having identical spectral data with
a previous report,16k was obtained by treatment of 8 with 8
equiv of the donor 7 in the presence of BF3·OEt2 as the
promoter at -40 °C and then warming to -20 °C. Under
these conditions, 9 and 10 were isolated in 5% and 19%
yields, respectively. Different promoters (AgOTf and TM-
SOTf), D-mannosyl donors (thioglycoside, glycosyl phos-
phate, and glycosyl chloride), and mixed solvent combina-
tions (1,4-dioxane/tetrahydrofuran and pure tetrahydrofuran)
were also examined, but the yields of the expected 2,6-di-
O-mannosylated product 6 were not satisfactory. With the
4-alcohol 6 in hand, removal of the 1,3,5-orthoformate group
was further investigated. p-Toluenesulfonic acid (p-TSA)-
catalyzed methanolysis of 6 was carried out at room
temperature, and the corresponding 1,3,4,5-tetraol 5 was
afforded in excellent yield (99%).
(17) (a) Potter, B. V. L.; Lampe, D. Angew. Chem., Int. Ed. 1995, 34,
1933–1972. (b) Conway, S. J.; Miller, G. J. Nat. Prod. Rep. 2007, 24, 687–
707.
(18) Estevez, V. A.; Prestwich, G. D. J. Am. Chem. Soc. 1991, 113,
9885–9887.
(19) (a) Murakata, C.; Ogawa, T. Carbohydr. Res. 1992, 235, 95–114.
(b) Bruzik, K. S.; Tsai, M. D. J. Am. Chem. Soc. 1992, 114, 6361–6374.
(c) Painter, G. F.; Grove, S. J. A.; Gilbert, I. H.; Holmes, A. B.; Raithby,
P. R.; Hill, M. L.; Hawkins, P. T.; Stephens, L. R. J. Chem. Soc., Perkin
Trans. 1 1999, 923–935. (d) Stadelmaier, A.; Schmidt, R. R. Carbohydr.
Res. 2003, 338, 2557–2569. (e) Conway, S. J.; Gardiner, J.; Grove, S. J. A.;
Johns, M. K.; Lim, Z. Y.; Painter, G. F.; Robinson, D.; Schieber, C.;
Thuring, J. W.; Wong, L. S. M.; Yin, M. X.; Burgess, A. W.; Catimel, B.;
Hawkins, P. T.; Ktistakis, N. T.; Stephens, L. R.; Holmes, A. B. Org.
Biomol. Chem. 2010, 8, 66–76.
(20) Conrad, R. M.; Grogan, M. J.; Bertozzi, C. R. Org. Lett. 2002, 4,
1359–1361.
(21) (a) Wang, Y. L. K.; Chen, W.; Blair, D.; Pu, M. M.; Xu, Y. J.;
Miller, S. J.; Redfield, A. G.; Chiles, T. C.; Roberts, M. F. J. Am. Chem.
Soc. 2008, 130, 7746–7755. (b) Sculimbrene, B. R.; Xu, Y. J.; Miller, S. J.
J. Am. Chem. Soc. 2004, 126, 13182–13183. (c) Morgan, A. J.; Wang, Y. K.;
Roberts, M. F.; Miller, S. J. J. Am. Chem. Soc. 2004, 126, 15370–15371.
(d) Sculimbrene, B. R.; Morgan, A. J.; Miller, S. J. J. Am. Chem. Soc.
2002, 124, 11653–11656. (e) Sculimbrene, B. R.; Miller, S. J. J. Am. Chem.
Soc. 2001, 123, 10125–10126.
(22) Luo, S. -Y.; Jang, Y.-J.; Liu, J.-Y.; Chu, C.-S.; Liao, C.-C.; Hung,
S.-C. Angew. Chem., Int. Ed. 2008, 47, 8082–8085
(23) Zhu, X.; Schimidt, R. R. Angew. Chem., Int. Ed. 2009, 48, 1900–
1934.
.
(24) (a) Uhlmann, P.; Vasella, A. HelV. Chim. Acta 1992, 75, 1979–
1994. (b) Uriel, C.; Go´mez, A. M.; Lo´pez, J. C.; Fraser-Reid, B. Eur. J.
Org. Chem. 2009, 403–411.
2620
Org. Lett., Vol. 12, No. 11, 2010

Scheme 3
.
Regioselective Benzylation and Acylation
Scheme 4. Synthesis of PIM2 1
protons of two hydroxy groups, respectively. In addition, the
2-H and 4-H protons also showed NOE correlation with the
anomeric protons of the two mannosyl rings. Regioselective
esterification of 1,3-diol 4 with stearic acid via a combination
of dicyclohexylcarbodiimide (DCC) and 4-N,N-dimethy-
laminopyridine (DMAP) provided the corresponding 1-al-
cohol 3 as a single diastereoisomer in 86% yield. The high
regioselectivity was attributed to the steric effect of two
mannosyl moieties that blocked the nucleophilic acylation
of the 1-C-hydroxyl. The installation of the stearic acyl chain
at the 3-O position of compound 3 was confirmed by the
correlation of 1-H with the hydroxy proton and the significant
downfield shift of the 3-H proton from δ 3.25 (1H spectrum
of 4 in CDCl3) to 4.76 ppm (1H spectrum of 3 in CDCl3).
Finally, the synthesis of the target molecule 1 is depicted
in Scheme 4. Pivaloyl chloride mediated coupling of the
1-alcohol 3 with the H-phosphonate 216g followed by iodine
oxidation in situ gave the corresponding phosphate, which
underwent counterion exchange on Dowex WX8 Na+ resin
to furnish the fully protected phosphatidyl dimannoside 13
in 92% yield. Cleavage of all benzyl groups in 13 under
hydrogenolytic conditions afforded the expected PIM2 1
(87%).
The next challenges included not only regioselective di-
O-benzylation of the 4,5-dihydroxyls at the myo-inositol
subunit of 5 but also regioselective introduction of an acyl
group at the 3-O position (Scheme 3). Williamson’s etheri-
fication of 5 under basic conditions (NaH, BnBr), even at
low temperature, resulted in a mixture of different O-
benzylated isomers, which were very difficult to separate and
identify. Alternative reagent combinations employing Ag2O/
BnBr (neutral conditions) and BF3·OEt2/BnOC(dNH)CCl3
(acidic conditions) also failed to provide the desired 1,3-
diol 4. Recently, we had developed a two-step procedure to
introduce benzyl-type protecting groups on carbohydrates in
a highly regioselective manner via per-O-trimethylsilylation
of the starting alcohol followed by TMSOTf-catalyzed
Et3SiH-reductive etherification.25 Treatment of compound 5
with trimethylsilyl chloride and triethylamine led to the
1,3,4,5-tetra-O-TMS ether 11 (quantitative yield), which was
subjected to 3 equiv of benzaldehyde and Et3SiH in the
presence of TMSOTf at -78 °C and then warming to -20
°C, yielding the 4,5-di-O-benzylated 1,3-diol 4 and the 3,4,5-
tri-O-benzylated 1-alcohol 12 in 72% and 5% yields,
respectively. The configuration of compound 4 was deter-
mined through a series of NMR experiments, including 1H,
In summary, we have developed a straightforward route
to prepare the triacylated PIM2 1 from myo-inositol 1,3,5-
orthoformate 8 in seven steps with an overall yield of 31%.
The immuno properties of this PIM2 molecule and its
application to the enzymatic biosynthesis of higher D-
mannosylated PIMs, LM, and LAM derivatives will be
further studied.
13C, DEPT, H-1H COSY, H-13C COSY, and H-1H
NOESY spectra (Supporting Information) in CDCl3. All
protons on the myo-inositol ring contained at least one big
trans-diaxial coupling constant, except the 2-H proton which
had a triplet splitting with a small coupling constant J ) 2.2
Hz. The chemical shift of this characteristic 2-H signal was
easily identified at δ 4.27 ppm in the 1H spectrum. The 1-H
and 3-H protons, which individually exhibited a correlation
with the 2-H proton, were found to have couplings with the
1
1
1
Acknowledgment. We thank the National Science Council
(NSC 97-2113-M-001-033-MY3, NSC 98-2119-M-001-008-
MY2) and Academia Sinica for the support.
(25) (a) Wang, C.-C.; Lee, J.-C.; Luo, S.-Y.; Fan, H.-F.; Pai, C.-L.; Yang,
W.-C.; Lu, L.-D.; Hung, S.-C. Angew. Chem., Int. Ed. 2002, 41, 2360–
2362. (b) Wang, C.-C.; Lee, J.-C.; Luo, S.-Y.; Kulkarni, S. S.; Huang, Y.-
W.; Lee, C.-C.; Chang, K.-L; Hung, S.-C. Nature 2007, 446, 896–899. (c)
Wang, C.-C.; Kulkarni, S. S.; Lee, J.-C.; Luo, S.-Y.; Hung, S.-C. Nat.
Protoc. 2008, 3, 97–113.
Supporting Information Available: Experimental pro-
cedure and 1H NMR spectra. This material is available free
of charge via the Internet at http://pubs.acs.org.
OL1008137
Org. Lett., Vol. 12, No. 11, 2010
2621