Concerning the reactivities of the C-1, C-2 and C-6 hydroxy groups of myo-inositol

Article abstract of DOI:10.1039/a907318c

Regioselectivities in the reactions of the three contiguous free hydroxy groups at C-6, C-1, and C-2 of 3,4,5-tri-O-benzyl-D-myo-inositol have been examined. Stannylene activation permits selective alkylation and esterification at C-1; however, acyl migra

Full text of DOI:10.1039/a907318c

View Article Online / Journal Homepage / Table of Contents for this issue
Concerning the reactivities of the C-1, C-2 and C-6 hydroxy
groups of myo-inositol
Gopinadhan nair Anilkumar, Zhaozhong J. Jia, Ralf Kraehmer and Bert Fraser-Reid*
Natural Products and Glycotechnology Research Institute, Inc., 4118 Swarthmore Rd, Durham,
North Carolina 27707, USA
Received (in Cambridge, UK) 9th September 1999, Accepted 21st October 1999
Regioselectivities in the reactions of the three contiguous free hydroxy groups at C-6, C-1, and C-2 of 3,4,5-tri-O-
benzyl--myo-inositol have been examined. Stannylene activation permits selective alkylation and esterification
at C-1; however, acyl migration back and forth between C-1 and C-2 leads to unpredictable ratios of the isolated
regioisomers. With the stable C1-alkylated products, further alkylation is regioslective for the axial C-2–OH,
whereas acylation is regioselective for the equatorial C-6–OH. In most cases the ‘other’ regioisomer is not observed,
the by-products being those of dialkylation or diacylation.
However after mannosylation, the inositol’s acyl group is lost
Introduction
in mammals, but is retained in (some) parasites, a circum-
The discovery in 1983 of inositol 1,3,4-trisphosphate, 1, as an
stance which might provide an opening for parasite-specific
intracellular second messenger which potentiates calcium
therapeutic agents.7
release was of landmark biological significance.1 Interest in
Unusual differences in reactivity have also been observed in
phosphoinositides intensified with subsequent speculation
laboratory synthetic manipulations of myo-inositol. In their
that there is a connection between inositolphosphoglycans
synthesis of a bis-mannosylated phosphoinositol from myco-
(IPGs) and insulin,2,3 and with the identification of glycosyl-
bacteria, van Boom and co-workers found that mannosylations
phosphatidylinositol (GPI) membrane anchors, summarized as
leading to 5 must be done in the order C-2 then C-6.9,10
2, in several pathogens.4,5 The selective functionalization of the
inositol moiety that is observed in 1 and 2 prompts speculation
about the relative reactivities of myo-inositol’s various hydroxy
Results and discussion
groups.5 This speculation surfaces again in connection with the
biosynthesis of GPI membrane anchors, where a phosphoryl-
ated inositol becomes glycosylated at the C-6–OH by N-acetyl-
glucosamine to give the pseudo-disaccharide 3. The latter is
then de-N-acetylated to give a free amine, as in 4 (Scheme 1),
which is then coupled to the first (of several) mannose residues,
leading to 2.6
The foregoing summary therefore raises questions about the
relative reactivities of the hydroxy groups of myo-inositol and
its derivatives. Our interest in this matter11 arose from recent
observations in our laboratory12 concerning the regioselective
reactions of triol 6b, prepared conveniently from acetate 6a
which was obtained by the elegant route of Bender and
Budhu.13 In keeping with David’s precedent,14 we anticipated
that stannylene formation would occur preferentially at the cis-
diol moiety, and that the intermediate15 so formed, 7 (Scheme
2), would undergo alkylation at the equatorial site as is nor-
mally observed.16 Monoalkylation of the diol obtained thereby,
8,12 was expected to favor the equatorial C-6–OH to give 9; but
this expectation was not fulfilled, since the regioisomer 10a was
The subsequent biosynthetic steps now depend on the organ-
ism concerned. Thus in mammalian systems and yeast, acyl-
ation of the inositols C-2–OH, as in 4 (R = acyl), is a necessary
prelude to mannosylation.7,8 This difference in biosynthesis is
somewhat surprising in view of the fact that the GPI core
itself is conserved throughout evolution,4a and a conserved
biosynthetic pathway might therefore have been expected.
Scheme 1
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596
This journal is © The Royal Society of Chemistry 1999
3591

View Article Online
Table 1 C-2–OH versus C-6–OH: Regioselectivity in some partially protected myo-inositols
Entry
Substrate
Reaction conditions
Major product (%)
i
ii
8
8
BnBr (1.2 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h
AllBr (1.2 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h
AllBr (1.2 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h
BnBr (1.1 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h
Benzyl trichloroacetimidate, TfOH, CH2Cl2, RT, 60 h
Ac2O (3.5 equiv.), DMAP, Pyr, 0 ЊC, 30 min
PalmCl (3 equiv.), DMAP, Pyr, 0 ЊC, 1 h
10a (66)
10b (89)
13 (76)
16a (78)
16a (80)
17a (77)
17b (73)
17c (71)
19 (75)
iii
iv
v
vi
vii
viii
ix
12
12
12
12
12
12
6a
PalmCl (3 equiv.), NaH, Bu4NI, DMF, 0 ЊC–RT, 14 h
PalmCl (3 equiv.), DMAP, Pyr, RT, 60 h
Scheme 2 Reagents and conditions: i, ref. 12; ii, RBr (1.2 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h; iii, BnBr (3 equiv.), Bu4NI, 70 ЊC, 2 h; iv, AllBr (1.2
equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h; v, BnBr (1.1 equiv.), NaH, Bu4NI, DMF, 0 ЊC, 1 h.
obtained as the major product. Indeed, 9 was not observed at
all, the by-product being the penta-O-benzylated material 11.
Similarly, treatment of the monoallyl compound 8 with one
additional equivalent of allyl bromide gave the bis(allyl ether)
10b.
The greater reactivity of the C-2–OH of diol 8 recalls the
biosynthetic pathway 3 → 4 →2 in Scheme 1, and invited
speculation that the fatty acylation step was designed to ‘block’
the more reactive axial hydroxy group. In the hope of providing
insight into this aspect, we have examined some alkylation and
acylation experiments of related inositol diols and triols, and
we report the results herein.
For a start, we established that benzylation of 6b, without
prior stannylene activation, led to a complex mixture. However,
the stannylene intermediate 7 could also be trapped by benzyl-
ation to give the tetra-O-benzyl diol 12, which underwent
selective allylation at C-2 leading to 13, the regioisomer of 10a.
Similarly, treatment of 12 with one equivalent of benzyl
bromide gave the penta-O-benzylated material 16a as the
major product.The minor products in these reactions were
not the regioisomers 15a and 15b, but the dialkylated analogues
14 and 16b, identified by independant preparation from 13
and 16a, respectively.
acetic anhydride (Scheme 3). That the monoester obtained in
77% yield was the C-6 derivative, 17a, was immediately appar-
ent from the splittings of 10 and 10 Hz, for the proton which
is shifted downfield to δ 5.53–5.48. Diacetate 18a was obtained
as the minor product. Notably, the C-2 monoacetate was not
observed.
With palmitoyl chloride in pyridine and DMAP under
standard conditions, the corresponding C-6 palmitate 17b was
obtained in 73% yield. However, in this case, the regioisomeric
product, 18b, was also observed (21%).
The regioselectivities observed for esterifications of 12 (Table
1, entries vi, vii) are seen to favor the C-6–OH, and therefore are
clearly different from alkylations which seemed to favor C-2–
OH. Are the above regioselectivities at C-2 and C-6 dependent
on the nature of the C-1 functionality? This question is relevant
since the starting material for the study is the C-1 acetate 6a, the
direct product of the Bender–Budhu process.13 We wished to
include the corresponding mono benzoyl analogue 20 in this
study; but reaction of the stannylene intermediate 7 with benz-
oyl chloride was not selective for C-1. Thus, with 1.1 equiv-
alents of the reagent, both esters 20 and 21 were obtained. In
subsequent experiments it was shown, beginning with pure 20
or 21, that benzoyl migration to give the other regioisomer
occurred readily. Accordingly, the ratio of these benzoates
varies widely from one experiment to another.
It was now of interest to see whether acylation followed a
similar pattern as the above described alkylations. The tetra-O-
benzyl diol 12 was therefore treated with 3.5 equivalents of
The ease of ester migration between 20 and 21 would clearly
3592
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596

View Article Online
Scheme 3 Reagents and conditions: ia, Ac2O (3.5 equiv.), DMAP, Pyr, 0 ЊC, 30 min; ib, PalmCl (3 equiv.), DMAP, Pyr, 0 ЊC, 1 h; ic, BzCl (3 equiv.),
DMAP, Pyr, 0 ЊC, 1 h; id, PalmCl (3 equiv.), NaH, Bu4NI, DMF, 0 ЊC–RT, 14 h; ii, PalmCl (3 equiv.), DMAP, Pyr, RT, 60 h; iii, BzCl (1.1 equiv.),
Bu4NI, RT, 1 h.
each of the three contiguous hydroxy groups of compound 6b
which is readily obtained by the Bender–Budhu synthesis.13
Stannylene mediated reactions allow excellent C-1–OH selec-
tivities, affording products such as 8, 12 and 20. However, in
the case of esterifications, the possibility of reversible acyl
migration might cause complications. Further examination of
these selectivities is underway and will be reported in due
course.
Scheme 4
Experimental
General methods
complicate any attempts at NaH-induced alkylation of C-1
esters such as 20 or 6a. However, DMAP-induced palmitoyl-
ation of 6a was as regioselective as for the C-1 benzyl analogue
12 (compare entries vii and ix).
Solvents of commercial anhydrous grade have been used. All
reactions were conducted under an inert argon atmosphere.
TLC plates (Riedel-de Haen, coated with silica gel 60 F 254),
were illuminated by UV light. Silica gel (Spectrum SIL 58, 230–
400 mesh, grade 60) was used for column chromatography. All
NMR spectra were recorded at 25 ЊC at 400 MHz (1H) or 100
MHz (13C), and chemical shifts are reported relative to internal
TMS. J-values are given in Hz. Accurate mass measurements
were made using FAB at 10 K resolution, and elemental anal-
yses were conducted by Atlantic Microlab, Norcross, GA. 1-O-
Allyl-3,4,5-tri-O-benzyl--myo-inositol 8 and 1-O-allyl-2,3,4,5-
tetra-O-benzyl--myo-inositol 10a were prepared as previously
described.12
The results in Table 1 show clearly that, for the 2,6-diols,
alkylation is selective for the axial C-2–OH, and acylation for
the equatorial C-6–OH. In an effort to rationalize these selec-
tivities, we considered the H-bonding pattern shown for 12 in
Scheme 4. It was possible that upon treatment with NaH, the
chelate A would be preferred to B. The observed C-2-selective
alkylation could therefore be chelation driven. In order to test
this idea, we examined a mechanistically different benzylating
reagent where chelation should not be a factor. Upon treatment
of 12 with benzyl trichloromethylacetimidate17 [entry (v)], the
major product was also shown to be 16a. This implies that
chelation was not essential for the selectivity.
1,2-Di-O-allyl-3,4,5-tri-O-benzyl-D-myo-inositol 10b
With regard to esterification, would chelation-driven acyl-
ation be shown to be C-2-selective? Palmitoylation of 12 was
therefore examined under the standard alkylation conditions
using NaH–Bu4NI–DMF. However, instead of a palmitate, the
formyl ester 17c was obtained in 71% yield, with the formation
of the equatorial palmitate 17b in minor amounts. Although
17c is the result of a Vilsmeier reaction,18 its formation con-
forms to the trend of C-6-selectivity observed in esterification
reactions.
A mixture of diol 8 (20 mg, 0.04 mmol), NaH (60% in mineral
oil; 4 mg, 0.1 mmol) and tetrabutylammonium iodide (15 mg,
0.04 mmol) in anhydrous DMF (1 mL) under Ar was chilled in
an ice-bath. To the stirred mixture was added allyl bromide
(4.14 µL, 0.048 mmol). After 1 h the reaction mixture was
quenched with drops of water and solvent was removed under
reduced pressure. The residue was flash chromatographed (1:4
EtOAc–hexane) to afford 10b (19 mg, 89%) as a colorless semi-
1
solid, Rf (1:2 EtOAc–hexane) 0.65; H NMR (CDCl3) δ 7.38–
We then turned to some early work of Angyal and Tate, on
the vicinal cis-1,2-diol of the 3,4,5,6-tetra-O-benzylinositol.
Reaction with one equivalent of benzyl chloride or methyl
iodide led to overwhelming C-1 (equatorial) alkylation with
≈1% axial regioisomer. However, with benzyloxymethyl
chloride and dihydropyran, the ratio of regioisomeric products
was 1:1. Angyal and Tate suggested that reaction at the
equatorial site is preferred in cases “which involve an inter-
mediate or a transition state” — such as esterifications and SN2
displacements. On the other hand, reactions which proceed
through “carbonium ions, and the subsequent rapid reactions
of the bond-deficient carbon atoms are not sensitive to steric
hindrance”.19
7.22 (m, 15H), 5.98–5.86 (m, 2H, allyl), 5.31–5.29 (m, 1H, allyl),
5.27–5.25 (m, 1H, allyl), 5.21–5.14 (m, 2H, allyl), 4.92–4.80 (m,
4H, Bn), 4.70–4.69 (d, 2H, Bn), 4.31–4.27 (m, 2H), 4.16–4.10
(m, 1H), 4.08–3.99 (m, 4H), 3.96 (dd, J 2, 2.1, 1H), 3.38–3.33
(m, 2H), 3.09–3.06 (ddd, J 12, 2, 0.8, 1H), 2.46 (br s, 1H, OH);
13C NMR δC 139.16, 139.11, 135.89, 134.77, 128.70, 128.61,
128.34, 128.16, 127.98, 127.88, 127.82, 117.72, 117.13, 83.67,
81.70, 81.26, 80.00, 76.13, 75.63, 73.60, 73.33, 73.20, 72.94,
71.44; FABMS m/z (relative intensity) (Mϩ ϩ 1) 531 (60), 529
(100), 439 (72), 391 (66), 363 (51), 338 (45).
1,3,4,5-Tetra-O-benzyl-D-myo-inositol 12
Our studies have shown that selective access can be had to
A mixture of triol 6b (990 mg, 2.2 mmol) and dibutyltin oxide
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596
3593

View Article Online
(660 mg, 2.65 mmol) in benzene (25 mL) was refluxed for 20 h
with a Dean–Stark trap. The temperature was reduced to 70 ЊC,
tetrabutylammonium iodide (815 mg, 2.2 mmol) and benzyl
bromide (780 µL, 6.6 mmol) were added, and the mixture was
stirred under Ar for 2 h. The solvent was evaporated off under
reduced pressure and the residue, on flash chromatography (1:3
EtOAc–hexane), afforded 12 as a colorless solid (455 mg, 82%)
[based on consumed triol 6b (530 mg, 1.17 mmol recovered)], Rf
78%) as a colorless semi-solid. The hexabenzyl product 16b was
obtained as the minor product.
Data for 16a.—1H NMR (CDCl3) δ 7.38–7.24 (m, 25H, Bn),
4.92–4.77 (m, 6H, Bn), 4.68–4.51 (m, 4H, Bn), 4.19–4.15
(dd, J 9.6, 9.6, 1H), 4.08–4.02 (m, 2H), 3.39–3.35 (m, 2H),
3.20–3.17 (dd, J 9.6, 2, 1H), 2.55 (s, 1H, OH); 13C NMR
δC 138.80, 138.78, 138.75, 138.27, 137.90, 128.43, 128.34,
128.32, 128.24, 128.10, 127.98, 127.78, 127.76, 127.67, 127.65,
127.58, 127.54, 127.50, 127.44, 127.31, 83.41, 81.35, 81.08,
81.05, 75.73, 75.29, 74.01, 73.63, 72.88, 72.79, 72.64; FABMS
m/z (relative intensity) (Mϩ ϩ 1) 631 (100), 629 (89), 607 (36);
HRFABMS [Calc. for C41H41O6 (Mϩ Ϫ H): m/z 629.2903.
Found m/z 629.2901].
1
(1:2 EtOAc–hexane) 0.5; H NMR (CDCl3) δ 7.34–7.24 (m,
20H, Bn), 4.90–4.87 (dd, J 10.8, 3.2, 2H, Bn), 4.83–4.79 (dd,
J 10.8, 1.6, 2H, Bn), 4.69–4.61 (m, 4H, Bn). 4.18 (dd, J 2, 2.1,
1H), 4.09–4.04 (ddd, J 9.6, 9.6, 2, 1H) 3.98–3.93 (dd, J 9.6, 9.6,
1H), 3.38–3.35 (dd, J 10, 2.8, 1H), 3.34–3.30 (dd, J 9.2, 9.6, 1H),
3.21–3.17 (dd, J 9.6, 2.8, 1H), 2.65 (s, 1H, OH), 2.59 (s, 1H,
OH); 13C NMR: δC 138.59, 137.77, 137.68, 128.40, 128.32,
128.28, 128.19, 127.82, 127.76, 127.73, 127.72, 127.49, 127.41,
82.76, 80.75, 79.78, 78.99, 75.65, 75.27, 72.47, 72.31, 72.14,
66.88. FABMS m/z (relative intensity) (Mϩ ϩ 1) 541 (100), 539
(68), 449 (30), 391 (15), 359 (15).
Method B. The diol 12 (50 mg, 0.09 mmol) dissolved in dry
methylene dichloride (2 mL) was chilled in an ice-bath. Benzyl
2,2,2-trichloroacetimidate (66 µL, 0.36 mmol) and triflic acid
(2 µL, 0.02 mmol) were added. The temperature was increased
to room temp. and the mixture was stirred under Ar for 60 h.
The solvents was evaporated off and the residue was flash
chromatographed (1:4 EtOAc–hexane) to afford 16a (40 mg,
80%) as the major product [the yield based on consumed diol
12 (7 mg, 0.013 mmol recovered)].
2-O-Allyl-1,3,4,5-tetra-O-benzyl-D-myo-inositol 13 and 2,6-di-
O-allyl-1,3,4,5-tetra-O-benzyl-D-myo-inositol 14
A mixture of diol 12 (100 mg, 0.18 mmol), NaH (60% in min-
eral oil; 15 mg, 0.36 mmol) and tetrabutylammonium iodide
(66 mg, 0.18 mmol) in anhydrous DMF (3 mL) under Ar was
chilled in an ice-bath. To the stirred mixture was added allyl
bromide (20 µL, 0.22 mmol). After 1 h the reaction mixture was
quenched with drops of water and solvent was removed under
reduced pressure. The residue was flash chromatographed (1:4
EtOAc–hexane) to afford 13 (80 mg, 76%) as a colorless semi-
solid, Rf (1:2-EtOAc–hexane) 0.7. The diallyl product 14 (10
mg, 9%) was obtained as the minor product.
6-O-Acetyl-1,3,4,5-tetra-O-benzyl-D-myo-inositol 17a and
2,6-di-O-acetyl-1,3,4,5-tetra-O-benzyl-D-myo-inositol 18a
A mixture of diol 12 (20 mg, 0.04 mmol) and 4-(dimethyl-
amino)pyridine (5 mg, 0.04 mmol) in anhydrous pyridine
(1 mL) under Ar was chilled in an ice-bath. To the stirred mix-
ture was added acetic anhydride (13 µL, 0.14 mmol). After 30
min the reaction mixture was quenched with drops of water and
the solvent was evaporated off under reduced pressure. The
residue was flash chromatographed (1:4 EtOAc–hexane) to
afford 17a (17 mg, 77%) as colorless semi-solid, Rf (1:2 EtOAc–
hexane) 0.52. The diacetyl product 18a (5 mg, 20%) was
obtained as the minor product.
1
Data for 13. H NMR (CDCl3) δ 7.36–7.25 (m, 20H, Bn),
5.98–5.88 (m, 1H, allyl), 5.30–5.24 (ddd, J 17.2, 3.2, 1.6, 1H,
allyl), 5.17–5.13 (ddd, J 10.4, 2.8, 1.2, 1H, allyl), 4.91–4.80 (m,
4H, Bn), 4.70–4.57 (m, 4H, Bn), 4.34–4.21 (m, 2H, allyl). 4.13–
4.08 (dd, J 9.6, 9.6, 1H), 4.02–3.97 (dd, J 9.6, 9.6, 1H), 3.96–
3.95 (dd, J 2.4, 2.4, 1H), 3.37–3.32 (m, 2H), 3.16–3.13 (dd, J 9.6,
2.4, 1H), 2.48 (s, 1H, OH); 13C NMR: δC 138.82, 138.79, 138.23,
137.90, 135.55, 128.47, 128.36, 128.34, 128.26, 127.97, 127.82,
127.70, 127.63, 127.53, 127.46, 116.69, 83.36, 81.36, 80.87,
79.98, 75.76, 75.30, 73.32, 73.28, 72.79, 72.30; FABMS m/z
(relative intensity) (Mϩ ϩ 1) 581 (100), 579 (48), 531 (17), 489
(15), 391 (13) (Calc. for C37H40O6: C, 76.53; H, 6.94. Found: C,
76.49; H, 6.97%).
1
Data for 17a. H NMR (CDCl3) δ 7.36–7.22 (m, 20H, Bn),
5.53–5.48 (dd, J 10, 10, 1H), 4.92–4.48 (m, 8H, Bn), 4.22–4.21
(dd, J 2.8, 2.4, 1H), 4.10–4.05 (dd, J 9.6, 9.6, 1H), 3.42–3.36 (m,
2H), 3.30–3.27 (dd, J 9.6, 2.4, 1H), 2.51 (s, 1H, OH), 1.89 (s,
3H, Ac); 13C NMR δC 169.89, 138.54, 138.36, 137.84, 137.51,
128.47, 128.46, 128.34, 128.07, 127.94, 127.89, 127.86, 127.71,
127.61, 127.57, 80.99, 80.87, 79.37, 75.95, 75.32, 72.66, 72.62,
72.02, 66.97, 20.99; FABMS m/z (relative intensity) (Mϩ ϩ 1)
605 (Mϩ ϩ Na), 583 (100), 581 (32), 491 (21), 475 (16), 385 (13)
(Calc. for C36H38O7: C, 74.21, H, 6.57. Found: C, 74.41, H,
6.51%).
1
Data for 14. H NMR (CDCl3) δ 7.22–7.38 (m, 20H, Bn),
6.01–5.89 (m, 2H, allyl), 5.29–5.22 (m, 2H, allyl), 5.17–5.11 (m,
2H, allyl), 4.89–4.79 (m, 4H, Bn), 4.69–4.61 (m, 4H, Bn), 4.40–
4.28 (m, 4H), 3.99–3.94 (dd, J 9.2, 9.6, 1H), 3.93–3.92 (dd, J 2.4,
2.4, 1H), 3.89–3.84 (dd, J 9.6, 9.6, 1H), 3.41–3.36 (dd, J 9.2, 9.2,
1H), 3.31–3.28 (dd, J 9.6, 1.6, 1H), 3.26–3.23 (dd, J 10, 2, 1H);
13C NMR δC 139.96, 139.89, 139.50, 139.38, 136.83, 136.47,
129.42, 129.38, 129.35, 129.07, 129.02, 128.69, 128.65, 128.60,
128.57, 128.53, 117.77, 117.59, 84.71, 82.65, 82.48, 81.77, 81.69,
77.01, 76.89, 75.61, 75.03, 74.40, 73.86, 73.81; FABMS m/z
(relative intensity) (Mϩ ϩ 1) 621 (100), 619 (30), 529 (9), 460
(13), 391 (12).
1
Data for 18a. H NMR (CDCl3) δ 7.35–7.21 (m, 20H, Bn),
5.83–5.82 (dd, J 2.8, 2.8, 1H), 5.44–5.39 (dd, J 10, 10, 1H),
4.89–4.38 (m, 8H, Bn), 3.96–3.91 (dd, J 9.2, 9.6, 1H), 3.47–3.40
(m, 2H), 3.36–3.32 (dd, J 10, 2.8, 1H), 2.16 (s, 3H, Ac), 1.92 (s,
3H, Ac); FABMS m/z (relative intensity) (Mϩ ϩ 1) 625 (82), 623
(35), 517 (100), 475 (58), 439 (62), 427 (71), 391 (45), 383 (45).
1,3,4,5-Tetra-O-benzyl-6-O-palmitoyl-D-myo-inositol 17b and
1,3,4,5-tetra-O-benzyl-2-O-palmitoyl-D-myo-inositol 18b
A mixture of diol 12 (50 mg, 0.09 mmol) and 4-(dimethyl-
amino)pyridine (11 mg, 0.09 mmol) in anhydrous pyridine
(2 mL) was chilled in an ice-bath. To the stirred mixture was
added freshly prepared palmitoyl chloride (92 µL, 0.27 mmol).
After 1 h the reaction mixture was quenched with drops of
water and the solvent was evaporated off under reduced pres-
sure. The residue was flash chromatographed (1:4 EtOAc–
hexane) to afford 17b (52 mg, 73%) as colorless semi-solid,
Rf (1:2 EtOAc–hexane) 0.6. The axial palmitoyl ester 18b
(15 mg, 21%) was obtained as the minor product.
1,2,3,4,5-Penta-O-benzyl-D-myo-inositol 16a
Method A. A mixture of diol 12 (50 mg, 0.09 mmol), NaH
(60% in mineral oil; 7 mg, 0.18 mmol) and tetrabutylam-
monium iodide (33 mg, 0.09 mmol) in anhydrous DMF (2 mL)
under Ar was chilled in an ice-bath. To the stirred mixture was
added benzyl bromide (13 µL, 0.1 mmol). After 1 h the reaction
mixture was quenched with drops of water and the solvent was
evaporated off under reduced pressure. The residue was flash
chromatographed (1:4 EtOAc–hexane) to afford 16a (45 mg,
3594
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596

View Article Online
1
Data for 17b. H NMR (CDCl3) δ 7.34–7.21 (m, 20H, Bn),
1-O-Benzoyl-3,4,5-tri-O-benzyl-D-myo-inositol 20 and
5.57–5.52 (dd, J 10, 10, 1H), 4.91–4.49 (m, 8H, Bn), 4.21–4.20
(dd, J 2.4, 2.4, 1H), 4.10–4.05 (dd, J 9.6, 9.2, 1H), 3.43–3.36 (m,
2H), 3.32–3.29 (dd, J10, 2.4, 1H), 2.50 (s, 1H, OH), 2.16–2.11
(m, 2H), 1.63–1.21 (m, 26H), 0.89–0.86 (t, 3H); 13C NMR
δC 172.63, 138.53, 138.35, 137.84, 137.46, 128.42, 128.31,
128.24, 128.05, 127.88, 127.82, 127.58, 127.55, 127.51, 127.46,
80.96, 80.91, 79.33, 77.48, 75.91, 75.17, 72.59, 72.28, 71.99,
66.90, 34.39, 31.89, 29.66, 29.65, 29.62, 29.41, 29.33, 24.81,
24.68, 22.66, 14.09; FABMS m/z (relative intensity) (Mϩ ϩ 1)
779 (100), 777 (60), 685 (78), 671 (62), 595 (81), 581 (71), 533
(62) (Calc. for C50H66O7: C, 77.09; H, 8.54. Found; C, 77.17; H,
8.58%).
2-O-benzoyl-3,4,5-tri-O-benzyl-D-myo-inositol 21
A mixture of the triol 6b (80 mg, 0.18 mmol) and dibutyltin
oxide (54 mg, 0.22 mmol) was refluxed for 20 h in benzene
(3 mL) with a Dean–Stark trap. The temperature was reduced
to ambient, tetrabutylammonium iodide (66 mg, 0.18 mmol)
and benzoyl chloride (25 µL, 0.2 mmol) were added, and the
mixture was stirred under Ar for 1 h. The solvent was evapor-
ated off under reduced pressure and the residue on flash
chromatography (1:3 EtOAc–hexane) afforded 20 (55 mg, 65%)
as white semi-solid [Rf (2:1 EtOAc–hexane) 0.75] and the
regioisomer 21 (10 mg, 2%) [Rf (2:1 EtOAc–hexane) 0.5] as
colorless semi-solid [yield based on consumed triol 6b (12 mg,
0.026 mmol recovery)].
1
Data for 18b. H NMR (CDCl3) δ 7.36–7.22 (m, 20H, Bn)
5.89–5.88 (dd, J 2.8, 2.8, 1H), 4.90–4.46 (m, 8H, Bn), 4.00–3.95
(dd, J 9.2, 9.6, 1H), 3.87–3.82 (dd, J 9.6, 9.6, 1H), 3.52–3.48
(dd, J 9.6, 2.4, 1H), 3.43–3.38 (dd, J 9.2, 9.2, 1H), 3.32–3.29
(dd, J 10, 2.4, 1H), 2.36–1.22 (m, 29H), 0.90–0.86 (t, 3H, Palm);
FABMS m/z (relative intensity) (Mϩ ϩ 1) 779 (50), 777 (80), 671
(75), 581 (100), 491 (40), 439 (45).
1
Data for 20. H NMR (CDCl3) δ 8.10–8.08 (m, 2H, ArH),
7.57–7.24 (m, 18H, ArH), 4.98–4.95 (dd, J 10.4, 2.8, 1H), 4.96–
4.68 (m, 6H, Bn), 4.39–4.38 (dd, J 2.8, 2.8, 1H), 4.30–4.25
(ddd, J 10, 10, 2.8, 1H), 4.01–3.96 (dd, J 9.6, 9.2, 1H), 3.63–
3.60 (dd, J 9.6, 2.4, 1H), 3.47–3.42 (dd, J 9.6, 9.2, 1H), 2.53 (s,
1H, OH), 2.34 (d, 1H, OH); 13C NMR δC 166.22, 138.47,
138.39, 137.38, 133.27, 129.85, 129.64, 128.53, 128.39, 128.38,
128.05, 127.91, 127.84, 127.65, 82.95, 80.88, 80.15, 75.80,
75.64, 73.60, 72.82, 70.54, 67.81; FABMS m/z (relative inten-
sity) (Mϩ ϩ 1) 555 (100), 553 (86), 463 (65), 447 (20), 373 (26),
357 (30).
1,3,4,5-Tetra-O-benzyl-6-O-formyl-D-myo-inositol 17c
A mixture of diol 12 (50 mg, 0.09 mmol), NaH (60% in mineral
oil; 7 mg, 0.18 mmol) and tetrabutylammonium iodide (33 mg,
0.09 mmol) in anhydrous DMF (2 mL) under Ar was chilled in
an ice-bath. To the stirred mixture was added freshly prepared
palmitoyl chloride (92 µL, 0.27 mmol). The temperature was
slowly increased to room temperature and the mixture was
stirred for 14 h. The reaction mixture was quenched with
drops of water and the solvent was evaporated off. The residue
was flash chromatographed (1:4 EtOAc–hexane) to afford 17c
(30 mg, 71%) as a colorless, viscous liquid, Rf (1:2 EtOAc–
hexane) 0.55 [yield based on consumed diol 12 (10 mg, 0.018
mmol recovery)]. The minor product obtained was 17b (15 mg,
26%).
1
Data for 21. H NMR (CDCl3) δ 8.02–8.00 (m, 2H, ArH),
7.50–7.22 (m, 18H, ArH), 6.89–6.87 (dd, J 2.8, 2.8, 1H), 4.95–
4.50 (m, 6H, Bn), 3.95–3.89 (m, 2H), 3.58–3.55 (dd, J 9.6, 2.8,
1H), 3.54–3.50 (m, 1H), 3.37–3.32 (dd, J 9.2, 9.6, 1H), 3.21 (d,
1H, OH), 3.10 (d, 1H, OH); 13C NMR δC 166.16, 138.43,
138.30, 137.56, 133.12, 129.82, 129.73, 128.51, 128.36, 128.26,
128.11, 128.03, 127.93, 127.89, 127.64, 127.54, 82.58, 81.40,
78.54, 75.72, 75.68, 73.24, 71.92, 70.26, 70.21; FABMS m/z
(relative intensity) (Mϩ ϩ 1) 555 (100), 553 (35), 463 (20), 447
(36), 357 (42).
Data for 17c. 1H NMR (CDCl3) δ 8.065 (s, 1H, formyl), 7.35–
7.22 (m, 20H, Bn), 5.21–5.16 (dd, J 9.6, 10, 1H), 4.93–4.54 (m,
8H, Bn), 4.21–4.20 (dd, J 2.8, 2.4, 1H), 4.06–4.02 (dd, J 9.6, 9.6,
1H), 3.49–3.44 (dd, J 9.6, 9.6, 1H), 3.41–3.38 (dd, J 10, 2.8, 1H),
3.37–3.34 (dd, J 10, 2.8, 1H), 2.49 (s, 1H, OH); 13C NMR
δC 161.34, 138.41, 137.82, 137.63, 137.10, 128.54, 128.50,
128.37, 128.11, 128.07, 127.96, 127.93, 127.90, 127.83, 127.82,
127.68, 80.86, 80.29, 79.31, 76.74, 75.94, 75.63, 72.75, 72.35,
66.94; FABMS m/z (relative intensity) (Mϩ ϩ 1) 569 (100), 567
(48), 477 (24), 391 (40).
Acknowledgements
We are grateful to Professor Laurens Anderson of the
University of Wisconsin, particularly with regard to the early
work of Angyal and Tate, and to Professor A. Vasella of the
ETH for helpful discussions. This work was supported by
grants from the NIH (GM40171) and the Research in Tropical
Diseases (TDR) program of the World Health Organization.
R. K. is grateful to the Deutscheforschungsgemeinshaft for a
Post Doctoral Fellowship.
1-O-Acetyl-3,4,5-tri-O-benzyl-6-O-palmitoyl-D-myo-inositol 19
A mixture of diol 6a (50 mg, 0.1 mmol) and 4-(dimethyl-
amino)pyridine (12 mg, 0.1 mmol) in anhydrous pyridine
(2 mL) was chilled in an ice-bath. To the stirred mixture was
added palmitoyl chloride (104 µL, 0.3 mmol). The temperature
was increased to ambient and the mixture was stirred for 60 h
before being quenched with drops of water and the solvent was
evaporated off under reduced pressure. The residue was flash
chromatographed (1:4 EtOAc–hexane) to afford 19 (20 mg,
75%) as colorless-semi solid [yield based on consumed diol 6a
(32 mg, 0.065 mmol recovered)], 1H NMR (CDCl3) δ 7.36–7.22
(m, 15H, Bn), 5.64–5.59 (dd, J 10.4, 10, 1H), 4.95–4.47 (m, 7H),
4.26–4.25 (dd, J 2.4, 2.4, 1H), 4.05–4.00 (dd, J 9.2, 9.6, 1H),
3.56–3.48 (m, 2H), 2.16–2.12 (m, 2H), 2.07 (s, 3H, Ac), 1.64–
1.21 (m, 26H), 0.89–0.86 (t, 3H); 13C NMR δC 172.49, 170.30,
138.13, 137.35, 128.55, 128.37, 128.32, 128.07, 128.01, 127.89,
127.69, 127.60, 80.97, 80.79, 79.59, 76.02, 75.46, 72.86, 71.37,
70.64, 67.79, 34.27, 31.90, 29.68, 29.58, 29.42, 29.34, 29.24,
29.06, 24.93, 24.71, 22.67, 14.10; FABMS m/z (relative inten-
sity) (Mϩ Ϫ 1) 729 (5), 713 (10), 623 (10), 533 (13), 443 (7), 368
(12), 334 (43), 316 (100).
References
1 (a) H. Strebv, R. F. Irvine, M. J. Berridge and I. Schulz, Nature,
1983, 306, 67; (b) M. J. Berridge and R. F. Irvine, Nature, 1984, 312,
3151.
2 (a) E. Kilgour, Cell. Signalling, 1993, 5, 97; (b) G. N. Gaulton and
J. C. Pratt, Semin. Immunol., 1994, 6, 97; (c) T. W. Rademacher,
H. Caro, S. Kunjuara, D. Y. Wang, A. L. Greenbaum and
P. MacLean, Braz. J. Med. Biol. Res., 1994, 27, 327; (d) J. M. Mato,
Cell. Signal, 1989, 1, 143; (e) G. Romero and J. Larner, Adv.
Pharmacol., 1993, 24, 21; ( f ) A. R. Saltiel, FASEB J., 1994, 8, 1034.
3 A. Kessler, B. Muller, S. Wied, A. Crecelius and J. Eckel, Biochem.
J., 1998, 330, 277.
4 See for example: (a) M. J. McConville and M. A. J. Ferguson,
Biochem. J., 1993, 294, 305; (b) P. Gerold, V. Eckert and R. T.
Schwarz, Trends Glycosci. Glycotechnol., 1996, 8, 265; (c) L.
Proudfoot, A. V. Nikolaev, G.-J. Feng, X.-Q. Wei, M. A. J.
Ferguson, J. S. Brimacombe and F. Y. Liew, Proc. Natl. Acad. Sci.
U.S.A., 1996, 93, 10984; (d) N. Stahl, D. R. Borchelt, K. Haiso and
S. B. Prusiner, Cell, 1987, 51, 229; (e) L. Schofield, S. Novakovic,
P. Gerold, R. T. Schwarz, J. J. McConville and S. D. Tachado,
J. Immunol., 1996, 156, 1886; ( f ) M. A. J. Ferguson, Philos. Trans.
R. Soc. London, Ser. B, 1997, 352, 1295.
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596
3595

View Article Online
5 Y. Watamabe, in Studies in Natural Products Chemistry, ed. A.
Rahman, Elsevier, Amsterdam, 1996, vol. 12, p. 35.
6 (a) A. K. Menon, N. A. Baumann, W. Van’t Hof and J. Vidugiriene,
Glycoconjugate Biosynth., 1993, 861; (b) P. T. Englund, Annu. Rev.
Biochem., 1993, 62, 121.
12 Z. J. Jia, L. Olsson and B. Fraser-Reid, J. Chem. Soc., Perkin Trans.
1, 1998, 631.
13 S. L. Bender and R. J. Budhu, J. Am. Chem. Soc., 1991, 113, 9883.
14 S. David, A. Thieffry and A. Veyrieres, J. Chem. Soc., Perkins Trans.
1, 1981, 1796.
7 V. L. Stevens and H. Zhang, J. Biol. Chem., 1994, 269, 31397.
8 (a) M. L. S. Guther and M. A. J. Ferguson, EMBO J., 1995, 14,
3080; (b) W. T. Doerrler, J. Ye, J. R. Falck and M. A. Lehrman,
J. Biol. Chem., 1996, 271, 27031.
9 (a) C. J. J. Elie, C. E. Dreef, R. Verduyn, G. A. van der Marel and
J. H. van Boom, Tetrahedron, 1989, 45, 3477; (b) C. J. J. Elie,
R. Verduyn, C. E. Dreef, D. M. Brounts, G. A. van der Marel and
J. H. van Boom, Tetrahedron, 1990, 46, 8243.
15 S. David, in Preparative Carbohydrate Chemistry, ed. S. Hanessian,
Marcel Dekker, New York, 1996, ch. 4.
16 T. S. Cameron, P. K. Bakshi, R. Thangarasa and T. B. Grindley,
Can. J. Chem., 1992, 70, 1623.
17 H.-P. Wessel, T. Iversen and D. R. Bundle, J. Chem. Soc., Perkin
Trans. 1, 1985, 2247.
18 (a) A. Vilsmeier and A. Haack, Ber. Dtsch. Chem. Ges., Abt. B, 1927,
60, 119; (b) K. Morita, S. Noguchi and M. Nishikawa, Chem.
Pharm. Bull., 1959, 7, 896.
10 C. J. J. Elie, R. Verduyn, C. E. Dreef, G. A. van der Marel and
J. H. van Boom, J. Carbohydr. Chem., 1992, 11, 715.
19 S. J. Angyal and M. E. Tate, J. Chem. Soc., 1965, 6949.
11 L. Schofield, M. J. McConville, D. Hansen, A. S. Campbell,
B. Fraser-Reid, M. J. Grusby and S. D. Tachado, Science, 1999, 283,
225.
Paper 9/07318C
3596
J. Chem. Soc., Perkin Trans. 1, 1999, 3591–3596