Synthesis of the essential core of the human glycosylphosphatidylinositol (GPI) anchor

Article abstract of DOI:10.1016/j.bioorg.2010.12.002

The biological role of GPI anchors is of paramount importance; however, we are still far from fully understanding the structure-function relationship of these molecules. One major limiting factor has been the tiny quantities available from natural sources

Full text of DOI:10.1016/j.bioorg.2010.12.002

Bioorganic Chemistry 39 (2011) 88–93
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Synthesis of the essential core of the human glycosylphosphatidylinositol (GPI)
anchor
Barbara Richichi a, Lucio Luzzatto b,1, Rosario Notaro b, Giancarlo la Marca c,2, Cristina Nativi a,d,3,
⇑
a Dipartimento di Chimica, Universita’ di Firenze, via della Lastruccia, 3,13 I-50019 Sesto F.no, Italy
b Istituto Toscano Tumori, ITT, via T. Alderotti, 26 N, I-50100 Firenze, Italy
c Dipartimento di Farmacologia, Universita’ di Firenze, AOU Meyer, I-50139 Firenze, Italy
d FiorGen, University of Florence, via Sacconi, 6 I-50019 Sesto F.no, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 November 2010
Available online 17 December 2010
The biological role of GPI anchors is of paramount importance; however, we are still far from fully under-
standing the structure–function relationship of these molecules. One major limiting factor has been the
tiny quantities available from natural sources; obtaining homogeneous and well-defined GPI structures
by synthesis, is both a challenge and an attractive goal. We report here the convergent synthesis of the
essential core of the human GPI anchor 1, exploiting a common precursor to obtain the trisaccharidic
donor 2 and a novel protecting groups sequence. The final product, prepared for the first time, is biolog-
ically active.
Keywords:
GPI anchor
Glycosylations
PNH disease
Ó 2011 Published by Elsevier Inc.
1. Introduction
biological studies. Several efforts have been made in the organic
synthesis of GPIs [9–20]: however, due to molecular complexity,
Glycosyl phosphatidylinositols (GPIs) are a family of glycolipids
defined by the structural motif Man 1-4 GlcN 1-6 myo-inositol-
these are not easily amenable to structural modifications. As a mat-
ter of fact, the lack/insertion of a monosaccharidic unit or of a
phospholipidic chain often mean a re-designing of the all synthe-
sis. For this reason the availability of new strategies to get homo-
geneous and well-defined GPIs and related structures remains an
attractive target.
Recently, it has been proposed that an autoimmune attack
against GPI anchors may be a key factor to explain the expansion
of hematopoietic cells deficient in all GPI-linked proteins (GPIꢀ)
in the serious rare blood disorder paroxysmal nocturnal hemo-
globinuria (PNH) [21]. PNH is characterized by complex clinical
manifestations, of which the most spectacular, intravascular
haemolysis producing hemoglobinuria, is a direct consequence
of the deficiency from the red cell membrane of the two comple-
ment-regulatory proteins CD59 and CD55 [21,22]. Studies of
the pathogenetic mechanisms of PNH have been hampered by
the fact that the mammalian GPI molecule is not available.
This has been the main goal driving the work reported in this
paper.
a
a
P-lipid. GPIs anchor a variety of functionally diverse hydrophilic
molecules (proteins, proteoglycans and phosphoglycans) to cell
membranes in eukaryotic cells [1,2]. GPIs are found in all eukary-
otes, but their structures vary; in fact, although the carbohydrate
backbone is universally conserved, additional phosphoethanol-
amine carbohydrate side chains and different phospholipid tails
differentiate the more than 20 GPI anchors that are known.
Approximately 150 proteins are modified by GPI, which is vital
for embryonic development, immune response and neurogenesis
in mammals [3].
Although GPI anchors play a pivotal role in numerous biological
functions, a full understanding of their structure–function relation-
ship is still elusive, largely because of the extremely limited quanti-
ties of GPI anchors available from natural sources [4–6]. Thus, for
example, while the role of the GPI lipid moiety is clearly related to
embedding the anchor in the lipid bilayer of the membrane, the
function of the glycan moiety is not yet completely understood [7,8].
Chemical synthesis provides a valuable tool for tackling these
issues, as it can supply native and non-native GPI anchors for
The choice of the specific target was not easy and, to some ex-
tent arbitrary. We have selected structure 1 (Fig. 1) based on the
following considerations.
⇑
Corresponding author. Fax: +39 0554573570.
E-mail addresses: lucio.luzztto@ittumori.it (L. Luzzatto), giancarlolamarca@
(a) Most GPI-anchored proteins on the cell surface do not have
phosphoethanolamine (EtNP) on Man2 of GPI [7]; we there-
fore thought that the GPI glycan core of structure 1 might be
the major precursor for protein attachment in the endoplas-
matic reticulum.
unifi.it (G.la Marca), cristina.nativi@unifi.it, mail@fiorgen.net (C. Nativi).
1
Fax: +39 0554385213.
Fax: +39 0555662849.
Fax: +39 0554574225.
2
3
0045-2068/$ - see front matter Ó 2011 Published by Elsevier Inc.
doi:10.1016/j.bioorg.2010.12.002

B. Richichi et al. / Bioorganic Chemistry 39 (2011) 88–93
89
the
a
1–6 dimannoside 11 (95%) (Scheme 2). Dimannoside 11
1–2), under the same
OAc
O
TIPSO
BnO
O
was deacetylated [30] and glycosidated (
a
O-
+H3N
BnO
P
reaction conditions reported above, with donor 3 giving the trim-
annoside 12 (83%). Sequential deprotection of the allyl group under
standard conditions (94%) and activation as trichloroacetimidate
afforded the desired trisaccharidic donor 2 (70%) (Scheme 2).
The pseudo-disaccharide 13 (Scheme 3) was obtained reacting
the activated derivative 14 [31] and allylinositol 15 [26,32] at room
temperature with trimethylsilyl triflate in dichloromethane
[14,25,33].
O
OH
O
O
BnO
BnO
BnO
O
O
HO
HO
Man3
O
O
HO
HO
HO
O
OBz
O
BnO
BnO
Man2
O
O
OH
O
NH
After acetyl removal, the obtained azide 16 (a anomer, 53%) was
HO
Man1 HO
OH
O
Cl3C
firstly treated with benzaldehyde dimethylacetal, then the acetal
was reductively opened using an acid and sodium cyanoboronhy-
dride to afford the pseudo-disaccharidic acceptor 13 (see Scheme
3). The assembly of the pseudo-pentasaccharide 17 [24] was ob-
tained in excellent yield by coupling the trisaccharide 2 with the
glucosamine-inositol pseudo-disaccharide 13 (Scheme 4).
Acetyl and benzoyl groups of the pseudo-pentasaccharide 17
[24] were converted into a benzyl protecting group; the allyl resi-
due on inositol moiety was then removed by treatment with palla-
dium (II) chloride in the presence of acetic acid leading to the
pseudo-pentasaccharide 20 [34] (67%). Coupling with the
phospholipid 21 [25] afforded the phosphate derivative 22. Mild
cleavage of the triisopropyl group with scandium (III) triflate [25]
gave 23, to which the phosphoethanol amine complex 24 [25,28]
was attached, giving 25 (Scheme 5).
As previously reported by Fraser-Reid for a similar molecule
[35], the final global debenzylation and azide reduction, performed
in a single step, proved to be rather unpredictable. In our hands,
the following approach was successful (any variation of this proce-
dure was rather unsatisfactory, producing partially deprotected
material or causing loss of the phospholipid tail): starting with
dichloromethane/isopropanol (1:1) mixture as solvent, in the pres-
ence of palladium (II) hydroxide (20% on charcoal) and an excess of
formic acid [25], with balloon pressure for eight hours at room
temperature (see Section 4).
O
2
1
HO
HO
OH
OH
OH
+H3N
O
O
P
O
O-
O
OCOC17H35
OCOC17H35
Fig. 1. Structure of GPI anchor 1 and trisaccharidic donor 2.
(b) The glycolipid residues are required for mooring the entire
anchor to the cell membrane. Since the precise structural
requirements (e.g. length) are still unclear [7], we have
opted for two C17H35 acyl residues.
(c) We have omitted any acyl group on inositol, because this
does not appear in the mammalian GPI biosynthetic path-
way [7,23].
Although we have capitalized on previous syntheses of GPI an-
chors, the molecule described herein is novel.
2. Results and discussion
After filtration and evaporation of the solvent, the recovered
material was re-dissolved in
a mixture of dichloromethane/
The synthetic strategy we have adopted has some features in
common with previous approaches to the total synthesis of GPI
[11,24–28]; however the protecting groups strategy, the selection
of glycosylation methods, as well as the choice of some key inter-
mediates have been specifically devised for compound 1.
The trisaccharidic donor 2 [24] (Fig. 1) represents the first target
of the proposed total synthesis. Trichloroacetimidates donors 3 and
4 and mannopyranoside acceptor 5 have been chosen (Scheme 1).
The three mannopyranosides were obtained from a common pre-
cursor, the commercially available orthoester 6, in high yielding
steps (Scheme 1).
isopropanol/water (1:1:0.2) and further treated with palladium
(II) hydroxide (20% on charcoal) and an excess of formic acid
for additional 8 h at room temperature. The crude product puri-
fied by HPLC (see Fig. S1) afforded the desired GPI anchor 1 in
75% yield. A flow injection analysis of 1 in methanol/formic acid
in a triple quadrupole MS instrument equipped with an electro-
spray source, gave m/z 1638 [M+H]+ as quasi-molecular ion.
ESI-HRMS confirmed the structure of 1 depicted in Fig. 1 (see Sec-
tion 4).
After deacetylation under basic conditions, 6 was monosilylated
at O-6 and dibenzylated to afford 7. Reaction of the orthoester 7
[29] with allyl alcohol in the presence of a Lewis acid formed the
3. Conclusion
Over the past decade, the arduous task of de novo synthesis has
been achieved for GPI anchors resembling those of several organ-
isms (Trypanosoma brucei [11,13], Plasmodium falciparum [35], Tox-
oplasma gondii [14], Saccharomyces cerevisiae [36], Trypanosoma
cruzi [28,37], Plasmodium falciparum [38]); but in only two cases
those of mammals (rat brain Thy [10], and human sperm CD52
[17]). However, in these two cases (a pseudo heptasaccharide with
two EtNP and a GPI-anchored peptide lacking the glycerolipid res-
idue), the complexity of the molecules makes these not easily ame-
nable to structural modifications. The work described here is the
first report of a tailored synthesis of the essential core of the hu-
man GPI anchor structure 1. To satisfactorily obtain GPI 1 we used
a common precursor to produce the trisaccharidic donor 2 and
developed a convenient protection/deprotection strategy. The final
crucial step (removal of 15 benzyl groups, one benzyloxy residue
corresponding allyl glycoside 8, as
a anomer, in 95% yield. The axial
acetyl group of 8 was removed and replaced, using standard condi-
tions, by a benzoyl group. Subsequent deprotection of the triisopro-
pylsilyl ether gave the orthogonally protected acceptor 5 in 98%
yield (Scheme 1). Treatment of the orthoester 7 with para-toluen-
sulfonic acid afforded the mannopyranoside 9, which was acti-
vated as trichloroacetimidate under Schmidt’s conditions,
affording donor 3 (Scheme 1). Donor 4, in turn, was prepared from
the tribenzyl orthoester 10 which, after treatment with para-tolu-
ensulfonic acid (98%) and activation as trichloroacetimidate under
Schmidt’s conditions, gave the suitably protected donor 4 (91%,
a
anomer) (see Scheme 1).
Glycosylation of 5 with donor 4 by treatment with trimethyl-
silyl triflate in dichloromethane at room temperature, afforded

90
B. Richichi et al. / Bioorganic Chemistry 39 (2011) 88–93
O
O
AcO
AcO
AcO
1) K2CO3, TIPSO
O
O
O
O
MeOH
BnO
O
O
BnO
2) TIPSCl,
AllOH
7
6
Imidazole, DMF
BF3 OEt2
95%
87% (over 2 steps)
3) BnBr, NaH
DMF, 84%
1) NaOMe
HO
BnO
BnO
OBz
O
TIPSO
OAc
O
MeOH, CH2Cl2
BnO
2) BzCl, NEt3
CH2Cl2, 90%
(over 2 steps)
3) AcCl, CH2Cl2
MeOH, 98%
BnO
OAll
5
8
OAll
p
TsOH
Cl3CCN
TIPSO
TIPSO
BnO
BnO
OAc
O
OAc
DME-H2O
98%
DBU, CH2Cl2
O
BnO
7
BnO
92%
OH
9
O
NH
NH
3
Cl3C
1) pTsOH
1) K2CO3
BnO
BnO
BnO
O
OAc
DME-H2O
98%
BnO
O
2) BnBr, NaH,
O
O
DMF
96%
BnO
O
6
BnO
2)
Cl3CCN
10
O
(over 2 steps)
DBU, CH2Cl2
91%
4
Cl3C
Scheme 1. Synthesis of donors 3 and 4 and acceptor 5.
TIPSO
BnO
OAc
O
BnO
BnO
BnO
OAc
O
BnO
1) PdCl2, NaOAc
AcOH, H2O, 94%
1) AcCl, CH2Cl2
BnO
BnO
BnO
O
4
MeOH, 80%
O
O
5
OBz
O
2
2) Cl3CCN,DBU
CH2Cl2 , 70%
TMSOTf
CH2Cl2
95%
2) 3, TMSOTf,
BnO
BnO
CH2Cl2
O
OBz
83%
OAll
O
BnO
BnO
11
12
OAll
Scheme 2. Synthesis of trisaccharidic donor 2.
AcO
AcO
AcO
AcO
BnO
OBn
O
NH
TMSOTf
CH2Cl2
OBn
OBn
O
HO
BnO
+
BnO
OBn
AllO
BnO
N3
O
OBn
OBn
CCl3
O
N3
α β
/
= 2.4/1
15
14
AllO
NaOMe, MeOH
1) PhCH(OMe)2
CH3CN,CSA, 92%
2) NaBH3CN, THF,
HCl, 60%
α
(
anomer, 53%
BnO
HO
over 2 steps)
O
HO
BnO
O
BnO
HO
OBn
BnO
OBn
N
3O
BnO
OBn
OBn
OBn
OBn
O
N3
AllO
13
AllO
16
Scheme 3. Synthesis of pseudo-disaccharide 13.

B. Richichi et al. / Bioorganic Chemistry 39 (2011) 88–93
91
and reduction of the azido group), was optimized and enabled us to
obtain a sample of 1 suitable for biological tests (see Section 4).
With respect to syntheses reported for different GPIs structures,
the strategy we followed gave the desired essential core of human
GPI, employing orthogonal protecting groups and decreasing the
protection–deprotection steps.
This compound is currently investigated for in vitro studies,
aiming to identifying the target molecules of a T cell-mediated
autoimmune process underlying the pathogenesis of PNH. At pres-
ent, we have found that after loading CD1d glycoprotein with 1 we
can detect in the peripheral blood of normal subjects a discrete
(and very small) population of T lymphocytes that presumably rec-
ognize the GPI molecule. This indicates that our synthetic GPI is
biologically active. Biological tests and activity will be published
in detail in the due corse.
OAc
O
TIPSO
BnO
BnO
O
BnO
BnO
BnO
O
O
OBz
O
BnO
TMSOTf
BnO
OBn
O
CH2Cl2
O
2
+ 13
BnO
- 40 °C
80%
BnO
OBn
OBn
OBn
N3
O
AllO
OR
O
TIPSO
BnO
BnO
17 R = Ac, R' = Bz
NaOMe, MeOH,
CH2Cl2 ,94%,
O
BnO
BnO
BnO
4. Experimental
O
4.1. Materials and methods
O
OR'
O
All solvents were of reagent grade quality and purchased com-
mercially. All starting materials were purchased commercially and
used without further purification. NMR spectra used for character-
ization of products and binding experiments were recorded on a
Varian Inova and Mercury 400 instruments and Bruker 700 instru-
ment. The NMR spectra were referenced to solvent. Mass spectra
were recorded on an Agilent Technologies 6110 Quadrupole LC/
MS. ESI-MS analysis was performed both in positive or negative
ion mode. HRMS were performed on a LTQ-IT-Orbitrap with a
spray voltage of 2.10 kV and a resolution of 100,000. C, H and N ele-
mental analysis was performed on a Perkin–Elmer 2400 elemental
analyser.
BnO
BnO
OBn
O
O
BnO
BnO
OBn
OBn
OBn
N3
O
AllO
18 R = R' = H
OBn
O
TIPSO
BnO
BnO
BnBr, NaH
DMF, 98%
19 R = R' = Bn
O
BnO
BnO
BnO
O
PdCl2
AcOH
O
H2O, 67%
OBn
O
BnO
BnO
OBn
O
4.2. Synthesis of (3,4-di-O-benzyl-6-O-triisopropylsilyl-
mannopyranosyl)-(1 ? 2)-(3,4,6-tri-O-benzyl- -mannopyranosyl)-
(1 ? 6)-(3,4-di-O-benzyl- -mannopyranosyl)-(1 ? 4)-(2-azido-
3,6-di-O-benzyl-2-deoxy- -glucopyranosyl)-(1 ? 6)-1-O-allyl-
2,3,4,5-tetra-O-benzyl- -myo-inositol (18)
a-D-
O
a-D
BnO
BnO
OBn
a
-D
OBn
OBn
N3
O
a
-D
HO
D
20
Scheme 4. Assembly of pseudo-pentasaccharide 20.
To a stirred solution of 17 (1.8 g, 0.760 mmol) in a mixture of
CHCl3/MeOH 1:2 (13 mL), 4.2 mL of 0.5 M solution of MeONa in
OBn
O
RO
BnO
BnO
H
O
O
O
+NHEt3
P
BnO
BnO
O-
1) PivCl, Py
20
+
BnO
O
2) I2, Py-H2O
> 90%
OCOC17H35
O
OBn
O
BnO
BnO
OCOC17H35
OBn
O
21
O
BnO
BnO
O
OBn
OBn
OBn
N
3O
O
O P
O
O-
O
O-
P
NHCbz
NHEt3
+NHEt3
H
+
OCOC17H35
OCOC17H35
Sc(OTf)3
22, R = TIPS
23, R = OH
24
AcCN-H2O
50 °C, 50%
1) 24, PivCl, Py
2) I2, Py-H2O
86%
25, R = Et3NH+ -HPO3(CH2)2NHCbz
Scheme 5. Synthesis of derivative 25.

92
B. Richichi et al. / Bioorganic Chemistry 39 (2011) 88–93
MeOH at rt were added. The mixture was warmed at 45 °C for 24 h
then the pH adjusted to neutrality with HCl (10% in MeOH). Evap-
oration of the solvent under vacuum gave a crude which was dis-
solved in CH2Cl2 (100 mL) and washed with brine (1 ꢁ 20 mL).
The organic phase was dried over Na2SO4 and concentrated to dry-
ness. The crude product was purified by flash chromatography on
silica gel (petroleum ether:EtOAc 2:1) to give 18 (1.58 g, 94%) as
glass solid. ½aꢂ2D5 + 30.81 (c 0.65 CHCl3); 1H NMR (400 MHz, CDCl3):
d 7.42–7.04 (m. 65H), 5.98–5.88 (m, 1H), 5.75 (d, J = 4.0 Hz, 1H),
5.30–5.25 (m, 1H), 5.23 (d, J = 1.6 Hz, 1H), 5.21 (d, J = 1.6 Hz, 1H),
5.21–5.17 (m, 1H), 4.99–4.78 (m, 8H), 4.73–4.25 (m, 19H), 4.13–
3.96 (m, 9H), 3.88–3.72 (m, 10H), 3.69–3.65 (m, 2H), 3.58–3.28
(m, 9H), 3.23–3.17 (m, 2H), 2.19 (d, J = 4.0 Hz, 1H), 2.05 (d,
J = 2.4 Hz, 1H), 1.05–1.04 (m, 21H); 13C NMR (100 MHz, CDCl3): d
138.9, 138.8, 138.74, 138.72, 138.58, 138.54, 138.3, 138.25,
138.21, 138.1, 134.1, 128.49, 128.45, 128.40, 128.36, 128.34,
128.26, 128.21, 128.18, 128.13, 127.97, 127.92, 127.86, 127.80,
127.7, 127.66, 127.63, 127.5, 127.49, 127.47, 127.44, 127.41,
127.3, 127.26, 127.22, 127.05, 127.02, 117.1, 101.5, 100.1, 99.1,
97.4, 81.8, 81.4, 80.8, 80.3, 79.9, 79.8, 79.5, 75.7, 75.3, 75.1, 75.0,
74.9, 74.56, 74.52, 74.3, 74.0,73.7, 73.5, 73.1, 72.9, 72.79, 72.71,
72.4, 72.0, 71.9, 71.7, 71.6, 71.1, 70.7, 69.5, 69.0, 68.6, 68.5, 65.8,
63.3, 62.4, 18.06, 18.02, 11.9; HRMS-ESI (m/z): Calcd for
[M+H+NH4]2+ C133H158O25N4Si, 1119.54867. Found: 1119.54870.
CHCl3 (100 mL) and washed with a saturated solution of Na2S2O3
(1 ꢁ 10 mL). The organic phase was washed with TEAB buffer
(3 ꢁ 20 mL), dried over Na2SO4 and concentrated to dryness.
The crude product was purified by flash chromatography on sil-
ica gel (CH2Cl2:MeOH 18:1 + 1%NEt3) to give 22 (0.635 g, 99%) as
glass solid.
½
aꢂ2D5 + 43.2 (c 0.5 CH2Cl2); 1H NMR (400 MHz,
CDCl3): d 7.41–6.99 (m, 75H), 5.91 (d, J = 3.6 Hz, 1H), 5.28–5.20
(m, 2H), 5.03–5.00 (A part of an AB system, J = 12.0 Hz, 1H),
4.94–4.83 (m, 7H), 4.78–4.74 (m, 4H), 4.68–4.30 (m, 22H),
4.25–3.99 (m, 11H), 3.94–3.30 (m, 21H), 3.12 (dd, J = 10.4 Hz,
3.6 Hz, 1H), 2.91 (q, J = 7.2 Hz, 6H), 2.26–2.21 (m, 4H), 1.54 (bs,
4H), 1.26–1.17 (m, 65H), 1.05–1.03 (m, 21H), 0.88 (t, J = 6.8 Hz,
6H); 13C NMR (100 MHz, CDCl3): d 173.3, 173.0, 139.7, 139.1,
138.9, 138.8, 138.75, 138.72, 138.6, 138.58, 138.55, 138.50,
138.39, 138.36, 137.9, 128.5, 128.36, 138.32, 128.28, 128.23,
128.20, 128.14, 128.12, 128.0, 127.99, 127.96, 127.90, 127.8,
127.71, 127.6, 127.4, 127.39, 127.32, 127.2, 127.1, 127.0,
126.99, 126.9, 126.8, 100.9, 99.2, 98.4, 96.4, 81.7, 80.9, 80.0,
79.6, 78.2, 75.9. 75.6, 74.9, 74.8, 74.6, 74.5, 74.3, 73.7, 73.6,
73.1, 72.9, 72.7, 72.2, 72.1, 72.0, 71.9, 71.8, 71.4, 70.5, 69.2,
68.9, 66.4, 63.6, 63.1, 62.7, 45.3, 34.2, 34.0, 31.9, 29.7, 29.6,
29.5, 29.3, 29.1, 27.1, 24.88, 24.80, 22.6, 18.07, 18.02, 14.1,
12.0, 8.4; 31P NMR (81 MHz, CDCl3): d ꢀ1.74; HRMS-ESI (m/z):
[M+Na+K]2+ Calcd for C183H236N3O32PSiNaK, 1554.29813. Found:
1554.29705.
4.3. Synthesis of (2,3,4-tri-O-benzyl-6-O-triisopropylsilyl-
mannopyranosyl)-(1 ? 2)-(3,4,6-tri-O-benzyl- -mannopyranosyl)-
(1 ? 6)-(2,3,4-tri-O-benzyl- -mannopyranosyl)-(1 ? 4)-(2-azido-
3,6-di-O-benzyl-2-deoxy- -glucopyranosyl)-(1 ? 6)-2,3,4,5-tetra-
O-benzyl- -myo-inositol (20)
a-D-
a-
D
4.5. Synthesis of triethylammonium-(2,3,4-tri-O-benzyl-
mannopyranosyl)-(1 ? 2)-(3,4,6-tri-O-benzyl- -mannopyranosyl)-
(1 ? 6)-(2,3,4-tri-O-benzyl- -mannopyranosyl)-(1 ? 4)-(2-azido-
3,6-di-O-benzyl-2-deoxy- -glucopyranosyl)-(1 ? 6)-2,3,4,5-tetra-
a-D-
a
D
-
D
a-D
a-
a-D
a-D
D
O-benzyl-1-O-(1,2-di-O-octadecanoyl-sn-glyceryl-phosphonato)-D-
To an ice cooled solution of 18 (1.28 g, 0.575 mmol) in dry DMF
(16.0 mL), BnBr (0.590 g, 0.410 mL, 3.45 mmol) and NaH (0.115 g,
4.60 mmol) were added. The mixture was warmed to rt and stirred
for 2.0 h. After this time, 0.5 mL of MeOH and a saturated solution
of NH4Cl (15 mL) were added. The mixture was then diluted with
Et2O (150 mL) and the organic phase was washed with brine
(2 ꢁ 20 mL) and dried over Na2SO4. Evaporation of the solvent un-
der vacuum gave a crude which was purified by flash chromatog-
raphy on silica gel (petroleum ether:EtOAc 5:1) to give 19
(1.35 g, 98%) as an oily. To a stirred solution of 19 (1.33 g,
0.553 mmol) in AcOH (15 mL), H2O (0.761 mL), NaOAc (0.636 g,
7.75 mmol) and PdCl2 (0.686 g, 3.87 mmol) were added. The mix-
ture was stirred at rt for 6.0 h then diluted with AcOEt and washed
with H2O (2 ꢁ 10 mL), with a saturated solution of NaHCO3
(3 ꢁ 10 mL) and brine (1 ꢁ 10 mL). The organic phase was dried
over Na2SO4 and concentrated to dryness. The crude product was
purified by flash chromatography on silica gel (petroleum ether:E-
tOAc 4:1) to give 20 (0.875 g, 67%) as an oily. The spectroscopic and
analytical data were in agreement with those reported previously.
myo-inositol (23)
A solution of 22 (0.442 g, 0.143 mmol) in a mixture of CHCl3/
MeOH (1:1) was passed through a column with a cation-ex-
changer resin (Amberlite IR-120, Na+ form). The eluate was col-
lected and concentrated to dryness. The residue was then
dissolved in CH3CN (28 mL) and water (14 lL) and Sc(OTf)3
(0.283 g, 0.575 mmol) were added. The reaction mixture was
warmed at 50 °C for 10 h then a few drops of pyridine were
added. The resulting mixture was diluted with CHCl3 (50 mL)
and washed with brine (2 ꢁ 10 mL). The organic phase was
dried over Na2SO4 and concentrated to dryness. The crude prod-
uct was purified by flash chromatography on silica gel
(CHCl3:MeOH 30:1 + 0.5%NEt3) to give 23 (0.214 g, 50%) as a
glass solid.
½
aꢂ2D5 + 39.2 (c 0.75 CH2Cl2); 1H NMR (400 MHz,
CDCl3): d 7.41–7.40(m, 2H), 7.32–7.00 (m, 73H), 5.85 (bs, 1H),
5.25–5.22 (m, 2H), 5.043 (d, J = 2.0 Hz, 1H), 5.02–4.99 (A part
of an AB system, J = 11.6 Hz, 1H), 4.95–4.55 (m, 16H), 4.49–
4.30 (m, 16H), 4.25–4.22 (m, 1H), 4.17–4.02 (m, 7H), 3.97–
3.37 (m, 24H), 3.17 (dd, J = 10.4 Hz, J = 4.0 Hz, 1H), 2.94–2.88
(m, 6H), 2.26–2.21 (m, 4H), 2.00 (bs, 5H), 1.54 (m, 4H), 1.32–
1.18 (m, 60H), 0.88 (t, J = 6.8 Hz, 6H); 13C NMR (100 MHz,
4.4. Synthesis of triethylammonium-(2,3,4-tri-O-benzyl-6-O-
triisopropylsilyl-
-mannopyranosyl)-(1 ? 6)-(2,3,4-tri-O-benzyl-
mannopyranosyl)-(1,4)-(2-azido-3,6-di-O-benzyl-2-deoxy-
a-D-manno pyranosyl)-(1 ? 2)-(3,4,6-tri-O-benzyl-
a-
D
a
-
D
-
CDCl3): d 173.3, 172.9, 139.6, 138.9, 138.65, 138.62, 138.5,
a-
D-
138.4, 138.38, 138.34, 138.0, 137.8, 128.5, 128.4, 128.3,
128.2,8, 128.24, 128.22, 128.18, 128.16, 128.0, 127.9, 127.8,
127.76, 127.7, 127.65, 127.61, 127.5, 127.44, 127.42, 127.36,
127.32, 127.25, 127.22, 127.15, 127.10, 127.0, 126.9, 126.8,
100.7, 99.5, 99.3, 96.5, 81.7, 81.6, 80.9, 79.7, 79.2, 75.7, 75.6,
75.1, 74.95, 74.90, 74.7, 74.5, 74.3, 74.1, 73.7, 73.2, 73.0, 72.7,
72.3, 72.2, 72.19, 72.13, 72.0, 71.9, 71.7, 70.3, 69.5, 69.0, 68.7,
66.0, 63.0, 62.6, 62.1, 45.3, 34.2, 34.0, 31.9, 29.69, 29.64,
29.52, 29.34, 29.30, 29.13, 24.87, 24.84, 22.66, 17.6, 14.0, 12.2,
8.3; 31P NMR (81 MHz, CDCl3): d ꢀ1.66; HRMS–ESI (m/z):
[M+NH4]+ Calcd for C174H220N4O32P, 2908.54428. Found:
2908.53941.
glucopyranosyl)-(1 ? 6)-2,3,4,5-tetra-O-benzyl-1-O-(1,2-di-O-
octadecanoyl-sn-glyceryl-phosphonato)-D-myo-inositol (22)
H-phosphonate 21 (0.900 g, 1.14 mmol) and 20 (0.408 g,
0.203 mmol) were co-evaporated with anhydrous pyridine
(2 ꢁ 5 mL) and dried under vacuum for 2.0 h. After this time,
the mixture was dissolved in pyridine (20 mL) and pivaloyl chlo-
ride (0.293 g, 0.298 mL, 2.43 mmol) was added. The resulting
mixture was stirred for 3.0 h then iodine (0.309 g, 1.22 mmol)
in a mixture of pyridine/water (19:1, 1.7 mL) was added. The
reaction mixture was further stirred for 1.5 h, then diluted with

B. Richichi et al. / Bioorganic Chemistry 39 (2011) 88–93
93
4.6. Synthesis of bistriethylammonium-(2,3,4-tri-O-benzyl-6-O-(2-(N-
benzyloxycarbonyl) aminoethyl-phosphonato)- -mannopyranosyl)-
(1 ? 2)-(3,4,6-tri-O-benzyl- -mannopyranosyl)-(1 ? 6)-(2,3,
4-tri-O-benzyl- -mannopyranosyl)-(1 ? 4)-(2-azido-3,6-di-O-
benzyl-2-deoxy- -glucopyranosyl)-(1 ? 6)-2,3,4,5-tetra-O-benzyl-
References
a-D
[1] V.L. Stevens, Biochem. J. 310 (1995) 361–370.
[2] P. Gerold, V. Eckert, R.T. Schwarz, Trend. Glycosci. Glycotechnol. 8 (42) (1996)
265–277.
[3] K. Hwa, Glycosyl phosphatidylinositol-linked glycoconjugates: Structure,
biosynthesis and function, in: A.M. Wu (Ed.), Molecular Immunology of
Complex Carbohydrate-2, Khuwer Academic, Plenum, 2001, pp. 207–214.
[4] S. Mayor, A.K. Menon, G.A.M. Cross, J. Cell Biol. 14 (1) (1991) 61–71.
[5] M. Beucher, W.S.F. Meira, V. Zegarra, L.M.C. Galvao, E. Chiari, K.A. Norris,
Protein Expres. Purif. 27 (1) (2003) 19–26.
[6] D.K. Sharma, T.K. Smith, A. Crossman, J.S. Brimacombe, M.A.J. Ferguson,
Biochem. J. 328 (1997) 171–177.
[7] M. Fujita, Y. Maeda, M. Ra, Y. Yamaguchi, R. Taguchi, T. Kinoshita, Cell 139 (2)
(2009) 352–365.
[8] M.G. Paulick, C.R. Bertozzi, Biochemistry 47 (27) (2008) 6991–7000.
[9] A. Crossman, J.S. Brimacombe, M.A.J. Ferguson, T.K. Smith, Carbohydr. Res. 321
(1–2) (1999) 42–51.
[10] A.S. Campbell, B. Fraser-Reid, J. Am. Chem. Soc. 117 (41) (1995) 10387–10388.
[11] D.K. Baeschlin, A.R. Chaperon, L.G. Green, M.G. Hahn, S.J. Ince, S.V. Ley, Chem. –
A Eur. J. 6 (1) (2000) 172–186.
[12] M. Martin-Lomas, N. Khiar, S. Garcia, J.L. Koessler, P.M. Nieto, T.W.
Rademacher, Chem. – A Eur. J. 6 (19) (2000) 3608–3621.
[13] C. Murakata, T. Ogawa, Carbohydr. Res. 235 (1992) 95–114.
[14] K. Pekari, D. Tailler, R. Weingart, R.R. Schmidt, J. Org. Chem. 66 (2001) 7432–
7442.
[15] F. Kamena, M. Tamborrini, X.Y. Liu, Y.U. Kwon, F. Thompson, G. Pluschke, P.H.
Seeberger, Nat. Chem. Biol. 4 (4) (2008) 238–240.
[16] Y.U. Kwon, X.Y. Liu, P.H. Seeberger, Chem. Commun. (17) (2005) 2280–2282.
[17] J. Xue, N. Shao, Z.W. Guo, J. Org. Chem. 68 (10) (2003) 4020–4029.
[18] M. Lahmann, P.J. Garegg, P. Konradsson, S. Oscarson, Can. J. Chem. – Rev. Can.
Chim. 80 (8) (2002) 1105–1111.
a
-D
a
a
-D
-D
1-O-(1,2-di-O-octadecanoyl-sn-glyceryl-phosphonato)-D-myo-
inositol (25)
H-phosphonate 24 (0.149 g, 0.414 mmol) and 23 (0.080 g,
0.026 mmol) were co-evaporated with anhydrous pyridine
(2 ꢁ 3 mL) and dried under vacuum for 2.0 h. After this time, the
mixture was dissolved in pyridine (4.3 mL) and pivaloyl chloride
(0.100 g, 0.102 mL, 0.829 mmol) was added. The resulting mixture
was stirred for 3.0 h then iodine (0.105 g, 0.414 mmol) in a mixture
of pyridine/water (19:1, 0.413 mL) was added. The reaction mix-
ture was further stirred for 1.5 h, then diluted with CHCl3
(50 mL) and washed with
a saturated solution of Na2S2O3
(1 ꢁ 5 mL). The organic phase was washed with TEAB buffer
(3 ꢁ 20 mL), dried over Na2SO4 and concentrated to dryness. The
crude product was purified by flash chromatography on silica gel
(CH2Cl2:MeOH 18:1 + 0.5%NEt3) to give 25 (0.061 g, 86%) as glass
solid. ½aꢂ2D5 + 37.84 (c 0.65 CHCl3); 31P NMR (81 MHz, CDCl3): d
ꢀ1.79,+1.07;
HRMS–ESI
(m/z):
[M+Na+K]2+
Calcd
for
C
184H228N4O37P2KNa, 1604.75407. Found: 1604.75175.
[19] A. Ali, R.A. Vishwakarma, Tetrahedron 66 (24) (2010) 4357–4369.
[20] B.M. Swarts, Z.W. Guo, J. Am. Chem. Soc. 132 (19) (2010) 6648–6650.
[21] A. Poggi, S. Negrini, M.R. Zocchi, A.M. Massaro, L. Garbarino, S. Lastraioli, L.
Gargiulo, L. Luzzatto, R. Notaro, Blood 106 (7) (2005) 2399–2408.
[22] W.F. Rosse, R.E. Ware, Blood 86 (9) (1995) 3277–3286.
[23] Y. Maeda, R. Watanabe, C.L. Harris, J.J. Hong, K. Ohishi, K. Kinoshita, T.
Kinoshita, EMBO J. 20 (1–2) (2001) 250–261.
4.7. Synthesis of (6-O-aminoethylphosphonato-
(1 ? 2)- -manno pyranosyl-(1 ? 6)- -mannopyranosyl-
(1 ? 4)-(2-amino-2-deoxy- -glucopyranosyl)-(1 ? 6)-1-O-(1,2-di-
O-octadecanoyl-sn-glyceryl-phosphonato)- -myo-inositol (1)
a-D-mannopyranosyl)-
a
-
D
a-D
a-D
D
To a stirred solution of 25 (93 mg, 0.029 mmol) in a mixture of
CH2Cl2/iPrOH 1:1 (28 mL), Pd(OH)2/C (20% Pd content) (93 mg) and
formic acid (6 lL, 0.145 mmol) were added. The mixture was stir-
[24] Y.U. Kwon, R.L. Soucy, D.A. Snyder, P.H. Seeberger, Chem. – A Eur. J. 11 (8)
(2005) 2493–2504.
[25] X.Y. Liu, Y.U. Kwon, P.H. Seeberger, J. Am. Chem. Soc. 127 (14) (2005) 5004–
5005.
red at rt for 8.0 h then filtered through a pad of Celite. Evaporation
of the solvent under vacuum gave a crude which was dissolved in a
mixture of CH2Cl2/iPrOH/H2O 1:1:0.2 (22 mL) and Pd(OH)2/C (20%
[26] X.Y. Liu, B.L. Stocker, P.H. Seeberger, J. Am. Chem. Soc. 128 (11) (2006) 3638–
3648.
[27] X.M. Wu, Z.W. Guo, Org. Lett. 9 (21) (2007) 4311–4313.
[28] D.V. Yashunsky, V.S. Borodkin, M.A.J. Ferguson, A.V. Nikolaev, Angew. Chem. –
Int. Ed. 45 (3) (2006) 468–474.
[29] A. Ravida, X.Y. Liu, L. Kovacs, P.H. Seeberger, Org. Lett. 8 (9) (2006) 1815–1818.
[30] N.E. Byramova, M.V. Ovchinnikov, L.V. Backinowsky, N.K. Kochetkov,
Carbohydr. Res. 124 (1983) C8–C11.
[31] R. Madsen, U.E. Udodong, C. Roberts, D.R. Mootoo, P. Konradsson, B. Fraser-
Reid, J. Am. Chem. Soc. 117 (5) (1995) 1554–1565.
[32] Z.J. Jia, L. Olsson, B. Fraser-Reid, J. Chem. Soc. – Perkin Trans. 1 (4) (1998) 631–
632.
[33] R.A. Vishwakarma, A.K. Menon, Chem. Commun. (4) (2005) 453–455.
[34] C.F.W. Becker, X.Y. Liu, D. Olschewski, R. Castelli, R. Seidel, P.H. Seeberger,
Angew. Chem. – Int. Ed. 47 (43) (2008) 8215–8219.
Pd content) (100 mg) and formic acid (6 lL, 0.145 mmol) were
added. The mixture was stirred at rt for 8.0 h then filtered through
a pad of Celite. Evaporation of the solvent under vacuum gave 1
(40 mg, 75%) as glassy solid. HPLC purification of 1 (10 mg) [Phe-
nomenex Synergi 4 lm Polar-RP 80A, from 90% water (+0.1% for-
mic acid) and 10 methanol (+0.1% formic acid), to 10% water
(+0.1% formic acid) and 90% (+0.1% formic acid)] allowed to obtain
1 (8 mg) as pure compound.
[35] J. Lu, K.N. Jayaprakash, U. Schlueter, B. Fraser-Reid, J. Am. Chem. Soc. 126 (24)
(2004) 7540–7547.
Acknowledgments
[36] T.G. Mayer, B. Kratzer, R.R. Schmidt, Angew. Chem., Int. Ed. Eng. 33 (21) (1994)
2177–2181.
[37] A. Ali, D.C. Gowda, R.A. Vishwakarma, Chem. Commun. (9) (2005) 1224.
[38] M.C. Hewitt, D.A. Snyder, P.H. Seeberger, J. Am. Chem. Soc. 124 (45) (2002)
13434–13436.
Authors thank Istituto Toscano Tumori (ITT), Florence, Italy for
financial support (cap. 26/6, 2007) and Prof. S. Oscarson (University
College, Dublin, Ireland) for intermediates’ samples provided.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bioorg.2010.12.002.