Design and synthesis of lipid-coupled inositol 1,2,3,4,5,6-hexakisphosphate derivatives exhibiting high-affinity binding for the HIV-1 MA domain

Article abstract of DOI:10.1039/c4ob00350k

The precursor of Gag protein (Pr55^Gag) of human immunodeficiency virus, the principal structural component required for virus assembly, is known to bind d-myo-phosphatidylinositol 4,5-bisphosphate (PIP₂). The N-terminus of Pr55^G

Full text of DOI:10.1039/c4ob00350k

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Design and synthesis of lipid-coupled inositol
1,2,3,4,5,6-hexakisphosphate derivatives exhibiting
high-affinity binding for the HIV-1 MA domain†
Cite this: Org. Biomol. Chem., 2014,
12, 5006
Hiroshi Tateishi,‡a Kensaku Anraku,‡b Ryoko Koga,a Yoshinari Okamoto,a
Mikako Fujitac and Masami Otsuka*a
The precursor of Gag protein (Pr55Gag) of human immunodeficiency virus, the principal structural com-
ponent required for virus assembly, is known to bind D-myo-phosphatidylinositol 4,5-bisphosphate (PIP2).
The N-terminus of Pr55Gag, the MA domain, plays a critical role in the binding of Pr55Gag to the plasma
membrane. Herein, we designed and synthesized myo-phosphatidylinositol 2,3,4,5,6-pentakisphosphate
(PIP5) derivatives comprising highly phosphorylated inositol and variously modified diacylglycerol to
examine the MA-binding properties. The inositol moiety was synthesized starting with myo-inositol and
assembled with a hydrophobic glycerol moiety through a phosphate linkage. The Kd value for MA-binding
of the PIP5 derivative 2 (Kd = 0.25 µM) was the lowest (i.e., highest affinity) of all derivatives, i.e., 70-fold
Received 15th February 2014,
Accepted 13th May 2014
lower than the Kd for the PIP2 derivative 1 (Kd = 16.9 µM) and 100-fold lower than the Kd for IP6 (Kd
=
25.7 µM), suggesting the possibility that the PIP5 derivative blocks Pr55Gag membrane binding by compet-
DOI: 10.1039/c4ob00350k
www.rsc.org/obc
ing with PIP2 in MA-binding.
ribosomal synthesis, Pr55Gag is directed to the plasma mem-
brane, where it is assembled with other components to form
1. Introduction
The development of anti-human immunodeficiency virus type immature budding virions. The N-terminus of Pr55Gag, the MA
1 (HIV-1) drugs has achieved marked success in the past two domain, plays a critical role in the binding of Pr55Gag to the
decades as envisaged by reverse transcriptase inhibitors, pro- plasma membrane.3 Recent studies have shown that D-myo-
tease inhibitors, entry inhibitors, and integrase inhibitors. phosphatidylinositol 4,5-bisphosphate (PIP2) is the binding
However, because the use of these drugs has encountered target of the basic patch of the MA domain.4–6
limitations because of the emergence of resistant viral vari-
We previously developed a highly sensitive in vitro assay to
ants, the development of new drugs based on novel mecha- determine the binding affinity of Pr55Gag/MA for phosphoino-
nisms has become urgent. This study focused on the sitide derivatives by employing a surface plasmon resonance
membrane targeting of the HIV-1 precursor of Gag protein (SPR) sensor in which a synthetic biotinylated inositol phos-
(Pr55Gag) at the stage of virus assembly, exploiting the possi- phate was immobilized.7–9 The SPR experiments comparing
bility to block the virus assembly by small molecules that the Pr55Gag/MA affinity of IP3 and PIP2 suggested that both the
compete at the membrane binding of Pr55Gag
.
divalent phosphate groups and the acyl chains of PIP2 are
HIV-1 genome-encoded Pr55Gag protein is the principal essential for tight binding to Pr55Gag/MA.
structural component required for virus assembly.1,2 Following
Because the PIP2-binding region of the MA domain con-
tains many basic residues that interact with acidic phosphate
groups of the inositol,2,10,11 the MA-binding affinity of phos-
phatidylinositol derivatives would be increased by increasing
the number of phosphate groups. This, together with several
previously published studies,2,10,11 would provide the basis for
the molecular design of novel competitors that would block
the PIP2-Pr55Gag binding.
Herein, we performed an SPR analysis of the MA domain
binding of highly phosphorylated inositol phosphates, myo-
inositol 1,2,3,4,5,6-hexakisphosphate (IP6), D-myo-inositol
aDepartment of Bioorganic Medicinal Chemistry, Faculty of Life Sciences,
Kumamoto University, 5-1 Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan.
E-mail: motsuka@gpo.kumamoto-u.ac.jp; Fax: +81-96-371-4620;
Tel: +81-96-371-4620
bDepartment of Medical Technology, Kumamoto Health Science University,
325 Izumi-machi, Kita-ku, Kumamoto 861-5598, Japan
cResearch Institute for Drug Discovery, School of Pharmacy, Kumamoto University,
5-1 Oe-honmachi, Chuo-ku, Kumamoto 862-0973, Japan
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4ob00350k
1,4,5-trisphosphate (IP3), and
a synthetic PIP2 derivative
‡These authors contributed equally to this work.
5006 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
had previously constructed.7 An expression vector for MA
having a FLAG tag at the C-terminus was used. Proteins were
purified from transfected 293 T cells using anti-FLAG agarose
beads employing the FLAG tag affinity method. Purified pro-
teins were quantified by SDS-PAGE analysis, and their concen-
tration was estimated by comparing the band intensity with
that of the protein marker. After purification, the solution in
which each protein was dissolved was exchanged with flow
buffer in the SPR system through dialysis. Flow buffer was sup-
plemented with 0.5 mg mL−1 BSA to inhibit non-selective
binding to the biotin-modified control surface, followed by 2%
(v/v) glycerol to prevent protein destabilization.12 Contrary to
the previous SPR analysis,7 5% dimethylsulfoxide was also sup-
plemented with analysis buffer to dissolve complexes in this
experiment (ESI 2†). Association was followed for 3 min and
dissociation was measured at a flow rate of 20 µl min−1 at
25 °C, after which the surfaces were regenerated by injecting
dilute NaOH solution. As shown in Fig. 1b, the injection of
0.24, 0.48, 0.64, and 0.96 μM MA into immobilized D-myo-inosi-
tol 1,3,4,5-tetrakisphosphate (IP4) showed a concentration-
dependent response unit (RU).
The dissociation constants (Kd) of MA–IP3, MA–IP6, and
MA–1 complexes were calculated via a competition assay. Solu-
tions containing varying concentrations of each competitor
were preincubated with MA and passed over the immobilized
IP4 surface. The competition curves were obtained by setting
the concentration of competitors upon the horizontal axis and
the response of free MA, determined based on the concen-
tration of MA bound to immobilized-IP4, upon the vertical
axis. The RU curves for competition between MA and the
various competitors are shown in Fig. 2a,c and e; the corres-
ponding Kd values are shown in Fig. 2b,d and f. The Kd value
for MA in competition with IP3 was 272 µM (Fig. 2b), indicat-
ing that IP3 binds MA weakly. It was noteworthy that IP6
showed Kd (25.7 µM) (Fig. 2d) comparable to that of 1 (16.9
µM) (Fig. 2f), although IP6 does not possess the diacylglycerol
moiety. These findings suggested that MA-affinity would be
further increased by introducing a diacylglycerol into IP6.
Fig. 1 Structures of IP3, IP6, and the PIP2 derivative 1 (a). Binding activity
of 0.24, 0.48, 0.64 and 0.96 µM MA proteins to biotinylated IP4. Each
protein was injected over a biotinylated IP4-immobilized sensor chip at a
flow rate of 20 µl min−1 for 180 s (b).
2.2. Design and synthetic strategy of PIP5 derivatives
We designed PIP5 derivatives having a modified glycerol
moiety (Fig. 3). To compare the influence of the aliphatic
chain structure of the glycerol group, both acyl (compound 2)
and alkyl ether (compound 4) derivatives were designed. To
confirm that the 2′-acyl chain participates in PIP2-MA binding
and the 1′-acyl does not,5 1′-O-methyl-2′-acyl/alkyl derivatives
(compounds 3 and 4) were designed. Our synthetic strategy for
the PIP5 derivatives (Fig. 3) was to differentiate the six hydroxyl
groups of myo-inositol through the diacetal intermediate,13
and the suitably protected intermediate was coupled with an
acyl/alkyl-glycerol moiety by a bifunctional phosphorylating
agent.14 A 1,5-dihydro-2,4,3-benzodioxaphosphepin-3-yl group
was employed for the synthesis of the acyl derivatives (i.e., 12),
having non-natural C8 acyl chains 1 (Fig. 1a) and found that
IP6 bound MA strongly, demonstrating the significance of the
number of phosphate groups. Further, we designed and syn-
thesized lipid-coupled IP6 derivatives, namely myo-phospha-
tidylinositol 2,3,4,5,6-pentakisphosphate (PIP5) derivatives,
expecting that their MA binding would be stronger than PIP2,
leading to the blockade of the Pr55Gag membrane target.
2. Results and discussion
2.1. SPR analysis of MA-interaction of IP3, IP6, and PIP2
To compare the relative MA-binding affinity of IP6, IP3, and the whereas the 2-cyanoethyl group was used for the phosphoryla-
PIP2 derivative 1 (Fig. 1a), we performed an SPR assay that we ting agent of the alkyl ether derivatives (i.e., 12).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5007

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 2 Competition assay and calculation of the equilibrium dissociation constants (Kd) for MA-competitor complexes. The equilibrium mixtures of
MA and competitors IP3 (a), IP6 (c), and the PIP2 derivative 1 (e) were injected over the biotinylated IP4-immobilized sensor chip at a flow rate of 20 µl
min−1 for 180 s. The average response unit (RU) for the increasing concentration of each competitor was measured at 160–170 s, and each RU
datum was converted to a concentration of uncompetitive MA protein used for the construction of competition curves between uncompetitive MA
and IP3 (b), IP6 (d), and the PIP2 derivative 1 (f). Calculated Kd values are shown. Each experiment was performed in duplicate.
5008 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 3 Design and synthetic strategy of PIP5 derivatives.
2.3. Syntheses of the IP6 moiety
benzodioxaphosphepin-3-yl)diethylamine18 and 1H-tetrazole
and subsequent oxidation with MCPBA in 75% yield. Oxidative
The syntheses of the IP6 moiety for acyl derivatives were per- cleavage of the p-methoxybenzyl group with CAN19 gave the
formed as shown in Scheme 1. The starting material DL-3-O- desired IP6 fragment 12, accompanying a phosphate migration
benzyl-1,2:4,5-di-O-cyclohexylidene-myo-inositol
6 was pre- product 13 in which the O-xylyl protected phosphate group at
pared according to the method of Billington et al.13 Benzyla- the 2-phosphate group migrated to the 1-phosphate allocating
tion of the alcohol 6 provided 7, which was further treated a stable conformation of myo-inositols.18 Because compounds
with p-toluenesulfonic acid and H2O to give deacetalized 8 in 12 and 13 could not be separated, the mixture was used for the
76% yield (for 2 steps). The cis-1,2-diol of 8 was regioselectively next coupling reaction without separation.
p-methoxybenzylated by means of the dibutyltin oxide pro-
The synthesis of the IP6 moiety for alkyl ether derivatives
cedure.15,16 Thus, the tin complex of the 1,2-diol was reacted was performed as shown in Scheme 2. The 2,3,4,5,6-penta-
with p-methoxybenzyl chloride in the presence of cesium fluo- hydroxy compound 10 was converted to the corresponding
ride to give regioselectively protected 9 in 89% yield. The selec- pentakisphosphonate 14 by treatment with bis(2-cyanoethyl)-
tive deprotection of the benzyl group of 9 by the method of N,N-diisopropylphosphoramidite20 and 1H-tetrazole and sub-
Oikawa et al.17 gave 10 in 45% yield. The 2,3,4,5,6-penta- sequent oxidation with MCPBA in 73% yield. Oxidative clea-
hydroxy compound 10 was converted to the corresponding vage of the p-methoxybenzyl group with CAN19 gave the IP6
pentakisphosphonate 11 by treatment with (1,5-dihydro-2,4,3- fragment 15 in 68% yield.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5009

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 Reagents and conditions: (i) benzyl bromide, NaH, DMF, rt, overnight, 94%; (ii) TsOH, THF–H2O, reflux, 5 h, 81%; (iii) (a) Bu2SnO, toluene,
reflux, 3 h; (b) CsF, MPM-Cl, DMF, −40 °C then rt, 48 h, 89%; (iv) H2/W-2 RANEY®-Ni, MeOH, 50 °C, 3 h, 45%; (v) (a) (1,5-dihydro-2,4,3-benzodioxa-
phosphepin-3-yl)diethylamine, 1H-tetrazole, CH2Cl2, rt, overnight; (b) MCPBA, CH2Cl2, −40 °C then rt, 1 h, 75%; (x) CAN, CH3CN–H2O, rt, 1 h.
Scheme 2 Reagents and conditions: (i) (a) bis(2-cyanoethyl)-N,N-diisopropylaminophosphoramidite, 1H-tetrazole, CH2Cl2, rt, 1.5 h; (b) MCPBA,
CH2Cl2, −78 °C then rt, 5 min, 73%; (ii) CAN, CH3CN–H2O, rt, 1.5 h, 68%.
2.4. Syntheses of di/mono-acylglycerol and di/mono-
alkylglycerol moieties
available starting material (R)-3-benzyloxy-1,2-propanediol 16
was reacted with heptanoyl chloride under basic conditions to
give compound 17 in 86% yield. The deprotection of the
benzyl group of 17 gave 18 in 96% yield. Compound 20 was
obtained by dialkylation of 16 followed by the benzyl deprotec-
tion in 59% yield (for 2 steps).
The syntheses of diacylglycerol and dialkylglycerol moieties
were performed as shown in Scheme 3. The commercially
The syntheses of the monoacylglycerol and monoalkyl-
glycerol moieties were performed as shown in Scheme 4. Com-
pound 16 was regioselectively methylated by means of the
dibutyltin oxide procedure. The tin complex of the 1,2-diol was
reacted with methyl iodide in the presence of cesium fluoride
to give 21 in 71% yield, accompanying a small amount of 2-O-
methyl product. Acylation of the 2-hydroxyl of 21 with hepta-
noyl chloride gave 22 in 93% yield. The deprotection of the
benzyl group of 22 gave 23 in 93% yield. Alkylation of the
2-hydroxyl of 21 with hexyl chloride gave 24 in 92% yield.
Finally, compound 24 was treated with H2/10% palladium
carbon to afford the debenzylated product 25 in 89% yield.
Scheme 3 Reagents and conditions: (i) heptanoyl chloride, DMAP, pyri-
dine, CH2Cl2, overnight, 86%; (ii) H2/Pd–C, CH2Cl2, overnight, 96%; (iii)
hexyl bromide, NaH, DMF, rt, overnight, 70%; (iv) H2/Pd–C, CH2Cl2, 24 h,
84%.
2.5. Coupling of IP6 and glycerol fragments
The coupling of acylated glycerol moieties and IP6 fragments
was performed as shown in Scheme 5. The glycerol moiety 18
5010 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
The monoacylglycerol derivatives, 3 and its isomer 3′ as triethy-
lammonium salts, were synthesized by the same procedure.
The coupling reaction of the IP6 fragment and the alkylated
glycerol moieties was performed as shown in Scheme 6. The
glycerol moiety 20 or 25 was reacted with the bifunctional
phosphorylating agent (2-cyanoethyl)-N,N,N′,N′-tetraisopropyl-
phosphoramidite14 and 1H-tetrazole to yield a rather labile
phosphoramidite. This compound was condensed with the IP6
fragment 20 or 25 without further purification. Oxidation of
the condensed product with tert-BuOOH gave 1,2-O-dihexylgly-
ceryl or 1-O-methyl-2-O-hexyl IP6 30 or 31 in 41% and 63%
yield, respectively. Finally, protecting groups were removed by
reaction with NH3 to give water-soluble PIP5 derivatives that
were purified by reverse phase chromatography followed by
cation-exchange chromatography to give 4 and 5 as triethylam-
monium salts in 64% and 31% yield, respectively.
Scheme 4 Reagents and conditions: (i) (a) Bu2SnO, toluene, reflux, 3 h;
(b) CsF, methyl iodide, DMF, −40 °C then rt, 2 days, 71%; (ii) heptanoyl
chloride, DMAP, pyridine, CH2Cl2, overnight, 93%; (iii) H2/Pd–C, CH2Cl2,
overnight, 93%; (iv) hexyl-Br, NaH, DMF, rt, overnight, 92%; (v) H2/Pd–C,
CH2Cl2, 24 h, 89%.
2.6. SPR analysis of MA complexes of PIP5 derivatives
Kd values of the MA complex of PIP5 derivatives were calculated
by the competition assay as described above. The RU curves
for competition between MA and the various competitors are
shown in Fig. 4a,c,e,g,i and k; the corresponding Kd values are
shown in Fig. 4b,d,f,h,j and l. As illustrated in Fig. 5, which
shows the Kd of the MA complex of IP3, IP6, the PIP2 derivative
1, and PIP5 derivatives with structure, the Kd values for MA in
competition with 2 (Kd = 0.25 µM) (Fig. 4b) were the lowest
(i.e., highest affinity) of all PIP5 derivatives, which was 70-fold
lower than the Kd for 1 (16.9 µM) and 100-fold lower than the
Kd for IP6 (25.7 µM). Therefore, the Kd value of the 2-MA
complex showed that PIP5 derivatives having both IP6 and
was reacted with benzyl-N,N,N′,N′-tetraisopropylphosphorami-
dite14 and 1H-tetrazole and subsequently condensed with the
IP6 fragment mixture of 12 and 13. Oxidation with tert-BuOOH
gave diheptanoyl glyceryl IP6 26 and 27 in 22% and 45% yield,
respectively. Finally, the protecting groups were removed by
hydrogenolysis with palladium carbon to give diheptanoyl gly-
ceryl PIP5 derivatives. These PIP5 derivatives were purified by
cation-exchange chromatography to give 2 and its isomer 2′ as
triethylammonium salts in 34% and 35% yield, respectively.
Scheme 5 Reagents and conditions: (i) (a) benzyl-N,N,N’,N’-tetraisopropylphosphoramidite, 1H-tetrazole, CH2Cl2, rt, 15 min; (b) 18 or 23, 1H-tetra-
zole, CH2Cl2, rt, 24 h; (c) tert-BuOOH, CH2Cl2, rt, 5 min, 26 (22%), 27 (45%), 28 (63%), 29 (11%); (ii) H2/Pd–C, tBuOH–H2O, 24 h, 2 (34%), 2’ (35%),
3 (44%), 3’ (22%).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5011

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 6 Reagents and conditions: (i) (a) (2-cyanoethyl)-N,N,N’,N’-tetraisopropylphosphoramidite, 1H-tetrazole, CH2Cl2, rt, 1.5 h; (b) 20 or 25,
1H-tetrazole, CH2Cl2, rt, 2 h; (c) tert-BuOOH, CH2Cl2, rt, 5 min, 30 (41%) and 31 (63%); (ii) aq. NH3, MeOH, 55 °C, 10 h, 4 (64%) and 5 (31%).
diacylglycerol moiety interact with MA tightly. The binding binding pocket of the MA–1 complex revealed that both inosi-
affinity of 2′ was 7.60 µM (Fig. 4d), which was 3-fold lower than tol and 2′-acyl group of 1 are accommodated in the MA
that of the 3-MA complex (Kd = 2.04 µM) (Fig. 4f), and almost binding pocket. In contrast, the 1′-acyl chain is located outside
the same as that of the 2′-MA complex (Kd = 9.01 µM) (Fig. 4h). the binding pocket (Fig. 6a). Although a similar calculated
These data showed that the phosphate isomers 2′ and 3′ result was obtained for the MA–2 complex, the outside orien-
bound MA more weakly than 1-phosphate derivatives 2 and 3. tation of the 1′-acyl chain was more pronounced (Fig. 6c). As
In contrast, the MA-binding affinity of 4 having an alkyl chain shown in Fig. 6b, the 1-phosphate interacts with Arg22 (2.9 Å:
at the glycerol moiety was 1.37 µM (Fig. 4j), which was 18-fold NH2, ionic; 3.0, 3.7 Å: NH, H-acceptor). The 4-phosphate inter-
lower than that of the PIP2 derivative 1, and was 5-fold higher acts with Lys98 (2.6, 2.9, 3.8 Å: NH2, ionic; 2.6 Å: NH2, H-accep-
than that of the diacyl derivative 2 (Kd = 0.25 µM). These data tor), whereas the 5-phosphate interacts with Arg76 (3.0 Å: NH2,
revealed that the diacyl glycerol structure is better than the 2.6, 3.6, 3.9 Å: NH, ionic; 3.0 Å: NH2, 2.6 Å: NH, H-acceptor). In
dialkyl glycerol structure in MA binding. The Kd value for the the case of 2 (Fig. 6d), the 2′-acyl carbonyl oxygen of 2 interacts
5-MA complex was 7.98 µM (Fig. 4l), which was almost the with Lys27 (2.9 Å: NH2, H-acceptor). The 1-phosphate interacts
same as that of 2′ and 3′-MA complex. In SPR analyses, all PIP5 with Arg22 (2.8, 3.0 Å: NH2, 3.5 Å: NH2, ionic; 3.0, 3.7 Å: NH,
derivatives bound MA more tightly than the PIP2 derivative 1, H-acceptor). The 2-phosphate interacts with Arg22 (2.6, 3.5 Å:
IP6 and IP3. The order of Kd was 2 < 4 < 3 < 5 = 2′ < 3′ < 1 < IP6 NH2, ionic; 2.6 Å: NH2, 3.0 Å: CH2, H-acceptor). The 3-phos-
< IP3. The structure–activity relationship of these compounds phate interacts with Lys98 (2.7, 2.7 Å: NH2, ionic; 2.7, 2.7 Å:
revealed that a highly phosphorylated inositol structure and NH2, H-acceptor). The 4-phosphate interacts with Lys98 (2.6,
diacyl (not monoacyl) glycerol at 1-position of inositol are 2.8 Å: NH2, ionic; 2.6, 2.8 Å: NH2, H-acceptor). The 5-phos-
important for MA domain binding.
phate interacts with Arg76 (2.8 Å: NH2, 2.6, 3.4 Å: NH, ionic;
To confirm the regiochemistry of 2 and 2′, we synthesized 2 2.8 Å: NH2, 2.6, 3.4 Å: NH, H-acceptor). The MA–2 complex
again by an independent route using dibenzyl N,N-diethylphos- showed a greater number of amino acid interactions compared
phoramidite that does not cause phosphate migration. In fact, with MA–1, owing to the greater number of phosphates of 2.
compound 2 was obtained as a sole product without the Although 1-, 4-, and 5-phosphate of both 1 and 2 interact with
accompanying isomer 2′. The newly synthesized 2 showed a Kd Arg22, Lys98, and Arg76, respectively, 2- and 3-phosphate of 2
value virtually identical to that obtained before (scheme 5), additionally interact with Arg22 and Lys98, respectively. In this
verifying the regiochemistry of 2 (ESI 2†).
context, judging from the results of the docking score based
on the electric interaction, van der Waals attraction and strain
energy of the ligand, the MA–2 complex was more stable than
the MA–1 complex (−374.7 kcal and −250.2 kcal as the U_dock
values, respectively). This is in agreement with SPR data
(0.25 µM and 16.9 µM as the Kd values, respectively).
Saad et al. demonstrated an “extended lipid” conformation
of the MA–1 complex, in which the glycerol 2′-acyl chain is
accommodated in the MA cleft and the glycerol 1′-acyl remains
buried in the membrane.5 Thus, the 1′-acyl does not contribute
to MA binding. However, in our study, although the MOE ana-
2.7. Theoretical binding analysis of MA–1 or MA–2
complexes
A molecular docking study (MOE) was adapted to the MA–1
and MA–2 complexes. The structures of complexes around the
binding pocket are shown in Fig. 6a and c, and the detailed
structures are shown in Fig. 6b and d, wherein lime green
lines (ionic interaction) and light blue lines with cylinder solid
(H-acceptor) indicate that the interaction between amino acids
and 1 (or 2) is shorter than 4.0 Å, respectively. The surrounded
5012 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 4 Competition assay and calculation of the equilibrium dissociation constants (Kd) for MA-competitor complexes. The sensorgrams of MA and
competitors, 2 (a), 2’ (c), 3 (e), 3’ (g), 4 (i), and 5 (k) are shown. The competition curves between uncompetitive MA and 2 (b), 2’ (d), 3 (f), 3’ (h), 4 ( j),
and 5 (l) are shown. Calculated Kd values are shown. Each experiment was performed in duplicate.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5013

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 5 Dissociation constant (Kd) of MA complexed with IPs, PI, and PIP5 derivatives.
lysis of the MA–2 complex indicates that the 1′-acyl was located single amino acid substitutions into MA and analyzed the
outside the binding pocket, 3 (without the 1′-acyl) did not effects on a variety of aspects of virus life cycles. They observed
bind MA (Kd = 2.04 µM) as strongly as 2 (Kd = 0.25 µM) did, as that a single amino acid mutation near the terminus of MA
revealed by the SPR analysis. It is hypothesized that the differ- and in the vicinity of residues 55 and 85 caused virus assembly
ence of Kd values between 3 and 2 is caused not only by the defects. Furthermore, they identified that a highly basic
interaction between the 2′-acyl chain and hydrophobic region domain between MA residues 17 and 31 (16 and 30 in the
of MA but also by the interaction between primordial carbons MOE number) is implicated in membrane binding. In this
of the 1′-acyl chain of 2 and the hydrophobic region of MA, MOE analysis, not only Arg22 at a highly basic region but also
which was not observed in MOE analysis.
the amino acids which have never been investigated, Arg76
Freed et al.2,21 demonstrated the role of the MA in the and Lys98, are implicated in MA–1 binding.
HIV-1 replication and mapped the functional domains within
HIV-1 is a retrovirus, which is a family of enveloped viruses
this protein by site-directed mutagenesis to introduce over 80 that replicate in a host cell through the process of reverse tran-
5014 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
scription. Retroviruses have Gag, Pol, and Env proteins. Chan
et al.22 examined the possible role of PIP2 in Gag-membrane
interaction of the alpharetrovirus Rous sarcoma virus (RSV)
and showed that neither membrane localization of RSV Gag-
GFP nor release of virus-like particles was affected by phospha-
tase-mediated depletion of PIP2 in transfected avian cells.
Furthermore, Inlora et al.23 determined the role of the MA–
PIP2 interaction in Gag localization and membrane binding of
a deltaretrovirus, human T-lymphotropic virus type 1 (HTLV-1).
They demonstrated that, unlike HIV-1 Gag, subcellular localiz-
ation of Gag and virus-like particles released by HTLV-1 was
minimally sensitive to polyphosphoinositide 5-phosphatase IV
(5ptaseIV) overexpression. These results suggest that the inter-
action of HTLV-1 MA with PIP2 is not essential for HTLV-1 par-
ticle assembly. Accordingly, MA–PIP2 binding might be
significant only in HIV-1 among retroviruses, and our findings
of MA-binding of PIP5 derivatives may be HIV-1 specific.
Although PIP5 derivatives bind MA tightly, when highly
charged these derivatives would not permeabilize the cell
membrane in spite of the fact that the viral assembly occurs
inside the cell. We intend to use a membrane carrier or syn-
thesize a phosphate prodrug compound to improve the cell
membrane permeability in the future.
3. Materials and methods
3.1. General methods
Chemicals were purchased from Aldrich, Fluka, Kanto Chemi-
cal, Nacalai tesque, and Wako. Thin layer chromatography
(TLC) was performed on precoated plates (Merck TLC sheets
silica 60 F254): products were visualized by spraying phospho-
molybdic acid in EtOH or basic potassium permanganate and
heated at high temperature. Chromatography was carried out
on Silica Gel 60 N (40–100 mesh). Reverse phase chromato-
graphy was performed using a C18 column (Cole-Parmer, USA).
Cation exchange chromatography was performed using Dowex
50WX8 (H+, 100–200 mesh). NMR spectra (JEOL JNM-AL300)
were referenced to SiMe4 or HDO. Infra-red spectra were
recorded on a JASCO FT/IR-410. The samples were prepared as
KBr discs or thin films between sodium chloride discs. Micro-
analysis was carried out using a Yanaco MT-5S. High resolu-
tion MS (HRMS) were recorded with a JEOL JMS-DX303HF by
using positive and negative FAB with 3-nitrobenzyl alcohol
(NBA) (containing HMPA or not) as the matrix.
3.2. DL-3,6-Di-O-benzyl-1,2:4,5-di-O-cyclohexylidene-
myo-inositol (7)
To a solution of DL-1,2:4,5-di-cyclohexylidene-myo-inositol 6
(2.27 g, 6.67 mmol) in DMF (10 ml) was added NaH (0.676 g,
28.1 mmol) followed by benzyl bromide (2.0 ml, 16.9 mmol),
and the resulting mixture was stirred at room temperature
under argon for 24 h. The reaction was quenched with MeOH,
and concentrated under reduced pressure, and the residue was
Fig. 6 Docking studies of MA–1 (a, b) and MA–2 (c, d) complexes. The
lime green lines (ionic interaction) and light blue lines with cylinder solid
(H-acceptor) indicate the interaction between amino acids and 1 (b) or 2 diluted with AcOEt. The organic phase was washed with H2O
(d) shorter than 4.0 Å, respectively.
and saturated aqueous NaCl, dried over Na2SO4, and then
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5015

View Article Online
Paper
Organic & Biomolecular Chemistry
concentrated under reduced pressure. The crude product was (CDCl3) δ: 55.2, 67.0, 71.9, 72.0, 72.2, 74.2, 75.3, 79.0, 79.4,
purified by silica gel column chromatography (hexane–AcOEt = 80.4, 113.8, 127.6, 127.9, 127.9, 128.4, 128.5, 129.5,
5 : 1) to afford 7 (3.25 g, 94%) as a white solid.
129.9, 137.8, 137.9, 138.7, 159.4, 162.5. IR (KBr) 3460, 2880,
1H NMR (CDCl3) δ: 1.25–1.69 (20H, m, CH2 × 10), 3.33 (1H, 1610, 1520, 1450, 1180, 1100, 810, 750 cm−1. Mp. 154 °C.
t, J = 9.3 Hz, CH), 3.62–3.67 (1H, dd, J = 10.6, 6.6 Hz, CH), MS (FAB) m/z 503 (M + Na)+. Anal. Calcd for C28H32O7: C,
3.71–3.76 (1H, dd, J = 4.2, 10.2 Hz, CH), 3.98 (1H, d, J = 9.7 Hz, 69.98; H, 6.71. Found: C, 70.02; H, 6.76. TLC; Rf 0.50 (CH2Cl2–
CH), 4.02–4.06 (1H, d, J = 5.1, 6.4 Hz, CH), 4.33 (1H, t, J = MeOH = 10 : 1).
4.5 Hz, CH), 4.78–4.90 (4H, m, CH2 × 2), 7.22–7.43 (10H, m,
C6H5 × 2). 13C NMR (CDCl3) δ: 23.9, 24.2, 24.3, 24.4, 25.4, 25.5,
3.5. DL-1-O-(p-Methoxybenzyl)-myo-inositol (10)
35.7, 36.9, 37.8, 72.0, 72.3, 75.0, 76.6, 77.2, 79.1, 80.3, 81.0,
To a solution of 9 (1.86 g, 3.87 mmol) in MeOH (25 ml) was
110.8, 113.1, 127.8, 128.1, 128.4, 128.5, 128.6, 128.7, 138.5,
added W-2 RANEY® Nickel (0.20 g, 3.03 mmol), and the result-
138.7. IR (KBr) 3030, 2935, 2860, 1500, 1165, 1110, 850, 830,
ing mixture was stirred at 50 °C under hydrogen for 3 h. The
740 cm−1. MS (FAB) m/z 521 (M + H)+. Mp. 123 °C. Anal. Calcd
mixture was filtered through a pad of celite and concentrated
for C32H40O6: C, 73.82; H, 7.74. Found: C, 73.87; H, 7.98. TLC;
under reduced pressure. The residue was washed with heated
Rf 0.42 (hexane–AcOEt = 5 : 1).
AcOEt, and the resulting crystals were filtered. Drying of the
crystals under reduced pressure afforded 10 (0.52 g, 45%) as a
white solid.
3.3. DL-3,6-Di-O-benzyl-myo-inositol (8)
To a solution of 7 (3.95 g, 7.58 mmol) in THF–H2O (5 : 1,
60 ml) was added p-toluenesulfonic acid monohydrate (1.90 g,
10.0 mmol). The resulting mixture was refluxed for 5 h, and
then neutralized with Et3N, and concentrated under reduced
pressure. The crude product was washed with a heated AcOEt,
and the resulting crystals were filtered. Drying the crystal
under reduced pressure afforded 8 (2.22 g, 81%) as a white
solid.
1H NMR (DMSO) δ: 2.91–2.94 (1H, m, CH), 3.03–3.06 (2H,
m, CH), 3.33–3.36 (1H, m, CH), 3.48–3.52 (1H, m, CH), 3.73
(3H, s, CH), 3.91 (1H, s, CH), 4.36–4.57 (7H, m, OH × 5, CH2),
6.87 (2H, d, J = 8.8 Hz, CH3OC6H5(CH × 2)), 7.31 (2H, d, J =
8.4 Hz, CH3OC6H5(CH × 2)). 13C NMR (DMSO) δ: 55.0, 69.3,
70.3, 71.7, 72.0, 72.4, 75.4, 79.6, 113.4, 129.0, 131.2, 158.5.
IR (KBr) 3390, 2910, 1610, 1590, 1510, 1250, 1120, 890,
820 cm−1. Mp. 183 °C. MS (FAB) m/z 299 (M − H)+. Anal. Calcd
for C14H20O7: C, 55.99; H, 6.71. Found: C, 56.06; H, 6.72. TLC;
Rf 0.39 (CH2Cl2–MeOH = 3 : 1).
1H NMR (DMSO) δ: 2.49 (3H, bs, OH × 3), 3.12 (2H, t, J =
9.9 Hz, CH × 2), 3.28 (1H, d, J = 7.3 Hz, CH), 3.59 (2H, t, J = 9.5
Hz, CH × 2), 3.95 (1H, s, CH), 4.53–4.79 (4H, m, CH2),
7.21–7.42 (10H, m, C6H5 × 2). 13C NMR (CDCl3) δ: 69.8, 70.8,
71.4, 72.3, 73.4, 75.0, 79.8, 81.8, 126.9, 127.1, 127.5, 127.8,
128.0, 139.3, 139.9. IR (KBr) 3750, 3030, 2905, 1500, 1450,
1110, 900, 740 cm−1. Mp. 204 °C. MS (FAB) m/z 383 (M + Na)+.
3.6. DL-1-O-(p-Methoxybenzyl)-2,3,4,5,6-penta-O-[(1,5-di-
hydro-2,4,3-benzodioxaphosphepin-3-yl)phosphoryl]-
myo-inositol (11)
Anal. Calcd for C20H24O6: C, 66.65; H, 6.71. Found: C, 66.40; To a suspension of 10 (0.050 g, 0.166 mmol) in CH2Cl2 (10 ml)
H, 6.83. TLC; Rf 0.48 (CH2Cl2–MeOH = 7 : 1).
was added MS4A, and the resulting suspension was stirred at
room temperature under argon for 15 min. To the mixture was
added (1,5-dihydro-2,4,3-benzodioxaphosphepin-3-yl)diethyl-
amine (0.358 ml, 1.66 mmol) followed by 1H-tetrazole (0.116 g,
3.4. DL-3,6-Di-O-benzyl-1-O-(p-methoxybenzyl)-myo-inositol
(9)
A mixture of 8 (2.10 g, 5.66 mmol) and dibutyltin oxide (1.74 g, 1.66 mmol), the resulting mixture was stirred overnight at
7.00 mmol) in toluene (100 ml) was refluxed for 3 h in a Dean– room temperature under argon. To the mixture was added
Stark apparatus to remove water. The mixture was concentrated m-chloroperbenzoic acid (0.336 g, 1.50 mmol) in small
under reduced pressure. To the residue was added cesium portions, and the resulting mixture was stirred at −40 °C to
fluoride (1.06 g, 7.00 mmol), and the mixture was suspended room temperature for 1 hour. The mixture was purified by
in heated DMF (30 ml) at 100 °C. To the resulting suspension silica gel column chromatography (AcOEt–hexane = 15 : 1) to
was added p-methoxybenzyl chloride (0.887 ml, 6.20 mmol) at afford 11 (0.151 g, 75%) as a white yellow solid.
−78 °C, and the mixture was stirred at room temperature
1H NMR (CDCl3) δ: 3.82 (3H, s, OCH3), 3.92 (1H, d, J = 8.6
under argon for 48 h. After concentration of the reaction Hz, CH), 4.52 (1H, d, J = 10.4 Hz, CH), 4.72–5.80 (26H, m, CH2,
mixture under reduced pressure, the residue was purified by C6H4(CH2)2 × 5, CH × 4), 6.90 (2H, d, J = 8.4 Hz, CH3OC6H4(CH
silica gel column chromatography (CH2Cl2–MeOH = 10 : 1) to × 2)), 6.96 (20H, m, C6H4 × 5), 7.46 (2H, d, J = 8.4 Hz,
afford 9 (2.40 g, 89%) as a white solid.
CH3OC6H4(CH × 2)). 13C NMR (CDCl3) δ: 55.1, 68.0, 68.9, 69.2,
1H NMR (CDCl3) δ: 2.48 (1H, bs, OH), 2.65 (2H, bs, OH), 74.4, 75.4, 76.6, 77.0, 77.2, 77.4, 113.5, 128.4, 128.5, 128.6,
3.19–3.23 (1H, dd, J = 2.7, 9.5 Hz, CH), 3.39 (1H, t, J = 9.3 Hz, 128.7, 128.8, 128.8, 129.0, 129.0, 129.2, 129.4, 129.8, 134.3,
CH), 3.76–3.82 (4H, m, OCH3, CH), 3.95 (1H, t, J = 9.3 Hz, CH), 135.1, 135.2, 135.5, 135.6, 159.1. IR (KBr) 1610, 1510, 1460,
4.16 (1H, s, CH), 4.61–4.70 (4H, m, CH2 × 2), 4.75 (1H, d, J = 1380, 1290, 1020, 860, 730 cm−1. Mp 165 °C. HRMS(FAB) m/z
11.2 Hz, C6H5CH2(CH)), 4.93 (1H, d,
J
=
11.2 Hz, calcd for C54H56O22P5 (M + H)+ 1211.2022. Found: 1211.1870.
C6H5CH2(CH)), 6.85 (2H, d, J = 8.8 Hz, CH3OC6H4(CH × 2)), Anal. Calcd for C54H56O22P5: C, 53.56; H, 4.58. Found:
7.23–7.36 (12H, m, C6H5 × 2, CH3OC6H5(CH × 2)). 13C NMR C, 53.21; H, 4.72. TLC; Rf 0.55 (CH2Cl2–MeOH = 10 : 1).
5016 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
3.7. DL-2,3,4,5,6-Penta-O-[(1,5-dihydro-2,4,3-benzodioxa-
phosphepin-3-yl)phosphoryl]-myo-inositol (12) and
DL-1,3,4,5,6-penta-O-[(1,5-dihydro-2,4,3-benzodioxa-
phosphepin-3-yl)phosphoryl]-myo-inositol (13)
m/z calcd for C36H47N10O21P5 (M + Na)+ 1133.1503. Found:
1133.1545. Rf 0.25 (CH2Cl2–MeOH = 7 : 1).
3.10. (R)-1-Benzyloxy-2,3-bis(heptanoyl)propane (17)
To a solution of 11 (0.070 g, 0.0578 mmol) in CH3CN–H2O
(9 : 1,
5 ml) was added diammonium cerium(IV) nitrate
To a mixture of (R)-3-benzyloxy-1,2-propandiol (16) (0.10 g,
0.549 mmol) in CH2Cl2 (5 ml) was added pyridine (0.11 ml,
1.37 mmol) followed by dimethylaminopyridine (0.0036 g,
0.27 mmol) and the resulting mixture was cooled to 0 °C. To
the mixture was added heptanoyl chloride (0.20 ml,
1.26 mmol) and the resulting mixture was stirred overnight at
room temperature under argon. The reaction was quenched
with H2O (25 ml), and the resulting water phase was extracted
with CH2Cl2. The organic layer was washed with 2 M aqueous
hydrogen chloride (20 ml) and H2O (25 ml). The resulting
organic phase was further washed with brine (30 ml) and
dried over Na2SO4, and then concentrated under reduced
pressure. The crude product was purified by silica gel column
chromatography (hexane–AcOEt = 9 : 1) to afford 17 (0.193 g,
86%) as a colorless oil.
(0.158 g, 0.288 mmol) and the resulting mixture was stirred at
room temperature for 1 hour. The mixture was concentrated
under reduced pressure, and the residue was purified by silica
gel column chromatography (CH2Cl2–MeOH = 10 : 1) to afford
the mixture of 12 and 13. Compounds 12 and 13 were used for
the next coupling reaction without further purification.
Rf values of compounds 12 and 13 were 0.37 and 0.29,
respectively (CH2Cl2–MeOH = 10 : 1).
3.8. DL-1-O-(p-Methoxybenzyl)-2,3,4,5,6-penta-
O-[bis(2-cyanoethyl)phosphoryl]-myo-inositol (14)
To a suspension of 10 (0.050 g, 0.166 mmol) in CH2Cl2 (10 ml)
was added MS4A, and the resulting suspension was stirred at
room temperature under argon for 15 min. To the mixture was
1H NMR (CDCl3) δ: 0.86–0.90 (6H, m, CH3 × 2), 1.28–1.36
(12H, m, CH2 × 6), 1.54–1.66 (4H, m, CH2 × 2), 2.25–2.34 (4H,
m, CH2 × 2), 3.59 (2H, d, J = 5.1 Hz, CH2OCH2C6H5), 4.15–4.22
(1H, dd, J = 6.2, 11.7 Hz, CH2OCO), 4.32–4.37 (1H, dd, J = 3.8,
11.9 Hz, CH2OCO), 4.49–4.58 (2H, dd, J = 12.1, 15.2 Hz,
C6H5CH2), 5.20–5.27 (1H, ddt, J = 3.9, 5.1, 6.2 Hz, CH2CHCH2),
7.26–7.37 (5H, m, C6H5). 13C NMR (CDCl3) δ: 14.0, 22.4, 24.8,
24.9, 28.7, 28.8, 31.4, 34.1, 34.3, 62.6, 68.3, 70.0, 73.3, 127.6,
127.7, 128.4, 137.7, 173.1, 173.4. IR (KBr) 2820, 1740, 1460,
1160, 1100, 740, 700 cm−1. HRMS(FAB) m/z calcd for C24H39O5
(M + H)+ 407.2797. Found: 407.2760. Anal. Calcd for C24H39O5:
C, 70.90; H, 9.42. Found: C, 70.61; H, 9.62. TLC; Rf 0.35
(hexane–AcOEt = 9 : 1).
added
bis(2-cyanoethyl)-N,N-diisopropylphosphoramidite
(0.383 ml, 1.50 mmol) followed by 1H-tetrazole (0.105 g,
1.50 mmol), the resulting mixture was stirred at room tempera-
ture under argon for 4 h. To the mixture was added m-chloro-
perbenzoic acid (0.336 g, 1.50 mmol) in small portions, and
the resulting mixture was stirred at −78 °C to room tempera-
ture for 1 hour. The mixture was purified by silica gel column
chromatography (CH2Cl2–MeOH = 7 : 1) to afford 14 (0.15 g,
73%) as a colorless oil.
1H NMR (CD3COCD3) δ: 2.65–2.91 (20H, m, CH2CH2CN ×
10), 3.68 (3H, s, OCH3), 3.95 (1H, d, J = 9.3 Hz, CH), 4.11–4.51
(21H, m, CH2CH2CN × 10, CH), 4.65–4.80 (5H, m, CH2, CH × 3),
5.36 (1H, d, J = 9.2 Hz, CH), 6.84 (2H, d, J = 8.8 Hz,
CH3OC6H5(CH × 2)), 7.39 (2H, d, J = 8.63 Hz, CH3OC6H5(CH × 2)).
IR (KBr) 3300, 2890, 2255, 1610, 1470, 1415, 1280, 1040, 820,
795, 765 cm−1. HRMS(FAB) m/z calcd for C44H55N10O22P5 3.11. 1,2-O-Diheptanoyl-sn-glycerol (18)
(M + Na)+ 1253.2078. Found: 1253.2029. TLC; Rf 0.28 (CH2Cl2–
To a solution of 17 (0.193 g, 0.475 mmol) in CH2Cl2 (10 ml)
MeOH = 10 : 1).
was added 10% Pd–C (0.126 g, 0.119 mmol), and the resulting
mixture was stirred overnight at room temperature under
hydrogen. The mixture was filtered through a pad of celite,
and the resulting filtrate was concentrated under reduced
3.9. DL-2,3,4,5,6-Penta-O-[bis(2-cyanoethyl)phosphoryl]-
myo-inositol (15)
To a solution of 14 (0.073 g, 0.059 mmol) in CH3CN–H2O (9 : 1, pressure. The crude product was purified by silica gel column
10 ml) was added diammonium cerium(IV) nitrate (0.208 g, chromatography (hexane–AcOEt = 2 : 1) to afford 18 (0.144 g,
0.379 mmol) and the resulting mixture was stirred at room 96%) as a colorless oil.
temperature for 1.5 h. The mixture was concentrated under
1H NMR (CDCl3) δ: 0.89 (6H, t, J = 6.8 Hz, CH3 × 2),
reduced pressure, and the residue was purified by silica gel 1.21–1.37 (12H, m, CH2 × 6), 1.50–1.68 (4H, m, CH2 × 2), 2.12
column chromatography (CH2Cl2–MeOH = 7 : 1 to 3 : 1) to (1H, bs, OH), 2.30–2.37 (4H, dd, J = 7.1, 14.5 Hz, CH2 × 2), 3.38
afford 15 (0.055 g, 68%) as a colorless oil.
(2H, bs, HOCH2), 4.20–4.26 (1H, dd, J = 5.7, 11.9 Hz,
1H NMR (CD3COCD3
+
D2O) δ: 2.93–3.02 (20H, m, OCOCHH), 4.30–4.35 (1H, dd, J = 4.6, 11.9 Hz, OCOCHH),
CH2CH2CN × 10), 4.22 (1H, s, CH), 4.41–4.53 (20H, m, 5.00–5.12 (1H, m, CH). 13C NMR (CDCl3) δ: 14.0, 22.4, 22.5,
CH2CH2CN × 10), 4.64–4.94 (4H, m, CH × 4), 5.20 (1H, d, J = 24.8, 24.9, 28.7, 28.8, 31.4, 34.1, 34.3, 61.5, 62.0, 173.4, 173.6.
9.0 Hz, CH). 13C NMR (CD3COCD3) δ: 19.8, 19.9, 19.9, 20.0, IR (KBr) 3590, 3140, 2930, 2860, 1740, 1160, 1100 cm−1. HRMS
20.0, 63.9, 64.0, 64.1, 64.2, 64.3, 64.3, 64.6, 68.8, 74.5, 76.1, (FAB) m/z calcd for C17H32O5 (M + Na)+ 339.2147. Found:
76.8, 79.0, 79.2, 79.2, 118.3, 118.4, 118.6. IR (film) 3020, 2910, 339.2154. Anal. Calcd for C17H32O5: C, 64.53; H, 10.19. Found:
2255, 1635, 1470, 1415, 1340, 1280, 1040 cm−1. HRMS(FAB) C, 64.33; H, 10.22. TLC; Rf 0.45 (hexane–AcOEt = 2 : 1).
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5017

View Article Online
Paper
Organic & Biomolecular Chemistry
3.12. (R)-1-Benzyloxy-2,3-bis(hexyloxy)propane (19)
C11H16O3 (M + Na)+ 219.0997. Found: 219.1012. TLC; Rf 0.58
(hexane–AcOEt = 1 : 2).
To a mixture of 16 (0.366 g, 2.03 mmol) in DMF (10 ml) was
added NaH (0.406 g, 16.9 mmol) followed by bromohexane
(0.708 ml, 5.0 mmol), and the resulting mixture was stirred at
room temperature under argon for 24 h. The reaction was
quenched with MeOH, and concentrated under reduced
pressure, and then the residue was diluted with AcOEt. The
organic phase was washed with H2O and saturated aqueous
NaCl, dried over Na2SO4, and then concentrated under
reduced pressure. The crude product was purified by silica gel
column chromatography (hexane–AcOEt = 5 : 1) to afford 19
(0.506 g, 70%) as a colorless oil.
3.15. (R)-1-Benzyloxy-2-heptanoyl-3-methoxypropane (22)
21 (0.119 g, 0.608 mmol) was allowed to react under the same
conditions as described for the preparation of 17 to give 22
(0.175 g, 93%) as a colorless oil.
1H NMR (CDCl3) δ: 0.85–0.90 (3H, m, CH3), 1.25–1.36 (6H,
m, CH2 × 3), 1.57–1.67 (2H, m, CH2), 2.34 (2H, t, J = 7.5 Hz,
CH2), 3.35 (3H, s, OCH3), 3.55–3.57 (2H, d, J = 5.1 Hz,
CH3OCH2), 3.61–3.62 (2H, d, J = 5.0 Hz, C6H5CH2OCH2),
4.50–4.59 (2H, dd, J = 12.1, 12.3 Hz, C6H5CH2), 5.16–5.22 (1H,
m, CH2CHCH2), 7.25–7.37 (5H, m, C6H5). 13C NMR (CDCl3) δ:
14.0, 22.4, 24.9, 28.7, 31.4, 34.3, 59.2, 68.6, 71.0, 71.3, 73.2,
127.6, 127.6, 128.3, 138.0, 173.4. IR (KBr) 3290, 2990, 2850,
1740, 1500, 1460, 1370, 1100, 740, 700 cm−1. HRMS(FAB) m/z
calcd for C18H29O4 (M + H)+ 309.2066. Found: 309.2068. TLC;
Rf 0.23 (hexane–AcOEt = 9 : 1).
1H NMR (CDCl3) δ: 0.88 (6H, m, CH3 × 2), 1.29 (12H, bs,
CH2 × 6), 1.52–1.59 (2H, m, CH2 × 2), 3.40–3.59 (9H, m,
CH2OCH2 × 3, CH2OCH2C6H5, CH2CHCH2), 4.55 (2H, s,
C6H5CH2), 7.25–7.34 (5H, m, C6H5). 13C NMR (CDCl3) δ: 14.0,
22.5, 25.7, 25.7, 29.5, 30.0, 31.6, 70.2, 70.5, 70.6, 71.6, 73.2,
77.8, 127.4, 127.5, 128.2, 138.3. IR (KBr) 3070, 3030, 2970,
2850, 1600, 1455, 1380, 1270, 1115, 730, 700 cm−1. MS (FAB)
m/z 351 (M + H)+. HRMS(FAB) m/z calcd for C22H39O3 (M + H)+
351.2889. Found: 351.2892. TLC; Rf 0.58 (hexane–AcOEt =
5 : 1).
3.16. 2-O-Heptanoyl-1-O-methyl-sn-glycerol (23)
22 (0.390 g, 1.27 mmol) was allowed to react under the same
conditions as described for the preparation of 18 to give 23
(0.258 g, 93%) as a colorless oil.
1H NMR (CDCl3) δ: 0.88 (3H, t, J = 6.8 Hz, CH3), 1.26–1.35
(6H, m, CH2 × 3), 1.59–1.65 (2H, m, CH2), 2.32–2.39 (3H, m,
OH, CH2), 3.38 (3H, s, OCH3), 3.55–3.60 (1H, dd, J = 4.8, 10.6
Hz, CH3OCH2(CH)), 3.59–3.64 (1H, dd, J = 4.9, 10.4 Hz,
CH3OCH2(CH)), 3.79 (2H, d, J = 4.4 Hz, CH2OH), 5.00–5.03
(1H, m, CH). 13C NMR (CDCl3) δ: 14.0, 22.4, 24.9, 28.7, 31.4,
34.3, 59.3, 62.5, 71.6, 72.7, 173.7. IR (KBr) 3630, 3240, 2810,
1735, 1460, 1110 cm−1. HRMS(FAB) m/z calcd for C11H23O6
(M + H)+ 219.1596. Found: 219.1590. TLC; Rf 0.44 (hexane–
AcOEt = 1 : 1).
3.13. 1,2-O-Dihexyl-sn-glycerol (20)
19 (0.406 g, 1.13 mmol) was allowed to react under the same
conditions as described for the preparation of 18 to give 20
(0.285 g, 84%) as a colorless oil.
1H NMR (CDCl3) δ: 0.87 (3H, t, J = 6.7 Hz, CH3 × 2), 1.30
(12H, m, CH2 × 6), 1.54–1.57 (4H, m, CH2 × 2), 2.30 (1H, bs,
OH), 3.42–3.71 (9H, m, CH2OCH2
× 3, CH2OCH2C6H5,
CH2CHCH2). 13C NMR (CDCl3) δ: 14.0, 22.6, 25.7, 29.5, 30.0,
31.6, 31.6, 63.0, 70.3, 70.9, 71.8, 78.2. IR (KBr) 3440, 2960,
2930, 1465, 1380, 1120 cm−1. MS (FAB) m/z 261 (M + H)+.
HRMS(FAB) m/z calcd for C15H33O3Na (M + Na)+ 283.2249. 3.17. (R)-1-Benzyloxy-2-hexyloxy-3-methoxypropane (24)
Found: 283.2252. TLC; Rf 0.53 (hexane–AcOEt = 2 : 1).
21 (0.120 g, 0.611 mmol) was allowed to react under the same
conditions as described for the preparation of 19 to give 24
3.14. (R)-1-Benzyloxy-3-methoxypropan-2-ol (21)
(0.157 g, 92%) as a colorless oil.
A mixture of 16 (0.50 g, 2.74 mmol) and dibutyltin oxide
1H NMR (CDCl3) δ: 0.88 (3H, m, CH3), 1.29 (6H, bs, CH2 ×
(0.697 g, 2.80 mmol) in toluene (50 ml) was refluxed for 3 h in 3), 1.53–1.60 (2H, m, CH2), 3.35 (3H, s, OCH3), 3.45–3.62 (7H,
a Dean–Stark apparatus to remove water. The mixture was con- m, CH3OCH2, CH2OCH2C6H5, CH2CHCH2, OCH2), 4.55 (2H, s,
centrated under reduced pressure. To the residue was added C6H5CH2), 7.25–7.34 (5H, m, C6H5). 13C NMR (CDCl3) δ: 14.0,
cesium fluoride (0.759 g, 5.0 mmol), and the mixture was sus- 22.6, 25.7, 30.0, 31.6, 59.1, 70.0, 70.5, 72.7, 73.3, 77.7, 127.4,
pended in heated DMF (30 ml) at 100 °C. To the resulting sus- 127.5, 128.2, 138.3. IR (KBr) 3285, 3065, 2960, 1600, 1455,
pension was added methyl iodide (0.311 ml, 10.0 mmol) at 1270, 1200, 1100, 700 cm−1. MS (FAB) m/z 281 (M + H)+. Anal.
−78 °C, and the mixture was stirred at room temperature Calcd for C11H16O3: C, 72.82; H, 10.06. Found: C, 72.67; H,
under argon with light shielding for 48 h. After concentration 10.28. TLC; Rf 0.58 (hexane–AcOEt = 5 : 1).
of the reaction mixture under reduced pressure, the residue
was purified by silica gel column chromatography (hexane–
3.18. 2-O-Hexyl-1-O-methyl-sn-glycerol (25)
AcOEt = 1 : 2) to afford 21 (0.386 g, 71%) as a colorless oil.
24 (0.153 g, 0.54 mmol) was allowed to react under the same
1H NMR (CDCl3) δ: 2.71 (1H, bs, OH), 3.36 (3H, s, OCH3), conditions as described for the preparation of 20 to give 25
3.38–3.56 (4H, m, CH3OCH2, CH2OH), 3.98 (1H, d, J = 4.4 Hz, (0.092 g, 89%) as a colorless oil.
CH2CHCH2), 4.54 (2H, s, C6H5CH2), 7.25–7.32 (5H, m, C6H5).
1H NMR (CDCl3) δ: 0.89 (3H, t, J = 6.8 Hz, CH3), 1.30–1.37
13C NMR (CDCl3) δ: 59.0, 69.2, 71.2, 73.3, 73.7, 127.6, 128.3, (6H, m, CH2 × 3), 1.54–1.61 (2H, m, CH2), 2.35 (1H, bs, OH),
137.8. IR (KBr) 3450, 3060, 3030, 2890, 1500, 1450, 1360, 1330, 3.37 (3H, s, OCH3), 3.46–3.70 (7H, m, CH3OCH2, CH2OH,
1200, 1100, 970, 740, 700 cm−1. HRMS(FAB) m/z calcd for CH2CHCH2, OCH2). 13C NMR (CDCl3) δ: 14.0, 22.5, 25.7, 30.0,
5018 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
31.6, 59.2, 62.6, 70.3, 72.6, 78.3. IR (KBr) 3310, 2935, 1455, 1581.3473. Found: 1581.3490 (M + Na)+. Rf 0.67 (CH2Cl2–
1104 cm−1. MS (FAB) m/z 281 (M + H)+. HRMS(FAB) m/z calcd MeOH = 10 : 1).
for C11H22O3Na (M + Na)+ 213.1467. Found: 213.1466. TLC; Rf
3.20. DL-1-O-(1,2-O-Diheptanoyl-sn-glyceryl) hydrogen
0.58 (hexane–AcOEt = 1 : 2).
phosphoryl]-myo-inositol 2,3,4,5,6-pentakis(hydrogen-
phosphate): 2
3.19. DL-2,3,4,5,6-Penta-O-[(1,5-dihydro-2,4,3-benzodio-
xaphosphepin-3-yl) phosphoryl]-myo-inositol 1-{[1,2-O-
diheptanoyl-sn-glyceryl](benzyl)phosphate} (26) and DL-1,3,4,5,6-
penta-O-[(1,5-dihydro-2,4,3-benzodioxaphosphepin-3-yl)-
phosphoryl]-myo-inositol 2-{[1,2-O-diheptanoyl-sn-glyceryl]-
(benzyl)phosphate} (27)
To a solution of 26 (0.030 g, 0.019 mmol) in tBuOH (8 ml) and
H2O (1.5 ml) was added 10% Pd–C (0.15 g, 0.14 mmol), and
the resulting mixture was stirred at room temperature under
hydrogen for 24 h. The mixture was filtered through a pad of
celite, and then the celite pad was washed with H2O. The
resulting filtrate was lyophilized. The residue was dissolved in
To a mixture of 18 (0.117 g, 0.54 mmol) in CH2Cl2 (5 ml) was H2O (2 ml), and filtered through the cation-exchange resin. To
added
benzyl-N,N,N′,N′-tetraisopropylphosphoramidite the resulting filtrate (0.009 g, 0.009 mmol) was added triethyl-
(0.20 ml, 0.54 mmol) followed by MS4A (0.20 g), and the result- amine (0.014 ml, 0.10 mmol), and concentrated under reduced
ing mixture was stirred at room temperature under argon for pressure. The resulting residue was dissolved in H2O, and lyo-
15 min. To the mixture was added 1H-tetrazole (0.038 g, philized to afford 2 (0.010 g, 34% from compound 26) as a
0.54 mmol), and the resulting mixture was stirred at room white solid.
temperature under argon for 10 min. To the mixture was
1H NMR (D2O) δ: 0.70 (6H, bs, CH3 × 2), 1.12 (12H, bs,
added, completely dissolved, a mixture of compounds 12 and CH2 × 6), 1.42 (4H, bs, CH2 × 2), 2.06–2.30 (4H, m, CH2 × 2),
13 (0.118 g, 0.108 mmol) in CH2Cl2 (10 ml) with MS4A, fol- 3.96–4.47 (10H, m, CH × 6, CH2OP, CH2OCO), 5.22 (1H, bs,
lowed by adding 1H-tetrazole (0.076 g, 1.08 mmol), and the CH2CHCH2). HRMS(FAB) m/z calcd for C23H47O28P6 957.0680.
resulting mixture was stirred at room temperature for further Found: 957.0623 (M − H)+.
24 h. To the mixture was added tert-butylhydroperoxide
(0.082 ml, 0.818 mmol), and stirred at room temperature for
further 5 min. The mixture was purified by silica gel column
chromatography (CH2Cl2–MeOH = 20 : 1) to afford compound
3.21. DL-2-O-(1,2-O-Diheptanoyl-sn-glyceryl) hydrogen
phosphoryl]-myo-inositol 1,3,4,5,6-pentakis(hydrogen-
phosphate): 2′
26 (0.056 g, 22%) as a white solid and compound 27 (0.092 g, 27 (0.045 g, 0.029 mmol) was allowed to react under the same
45%) as a white solid.
3.19.1 Compound 26. 1H NMR (CDCl3) δ: 0.70–0.80 (6H, (0.008 g, 39% from an acid form of 2′) as a white solid.
m, CH3 × 2), 1.01–1.18 (12H, m, CH2 × 6), 1.35–1.40 (4H, m,
1H NMR (D2O) δ: 0.70 (6H, bs, CH3 × 2), 0.98–1.22 (12H, m,
conditions as described for the preparation of 2 to give 2′
CH2 × 2), 1.91–2.14 (4H, m, CH2 × 2), 3.97–4.03 (2H, dd, J = 5.1, CH2 × 6), 1.43 (4H, bs, CH2 × 2), 2.23–2.28 (4H, m, CH2 × 2),
5.7 Hz, CH2OP), 4.16–4.33 (3H, m, CH, CH2OCO), 4.68–5.69 3.34 (1H, bs, CH), 3.60 (1H, bs, CH), 3.77 (1H, bs, CH),
(28H, m, CH × 5, C6H5CH2, CH2CHCH2, C6H4(CH2)2 × 5), 4.05–4.30 (7H, m, CH × 3, CH2OP, CH2OCO), 5.20 (1H, bs,
6.91–7.53 (25H, m, C6H4 × 5, C6H5). 13C NMR (CDCl3) δ: 13.9, CH2CHCH2).
22.3, 24.5, 28.6, 31.3, 33.9, 61.7, 66.5, 68.4, 68.9, 69.0, 69.1,
1H NMR (D2O) δ: 0.66–0.68 (6H, m, CH3 × 2), 1.03–1.24
69.2, 69.3, 69.4, 69.5, 70.0, 70.2, 73.8, 76.2, 76.7, 76.9, 77.0, (111H, m, CH2 × 6, NCH2CH3 × 33), 1.28–1.43 (4H, m, CH2 ×
77.2, 77.3, 127.7, 128.3, 128.4, 128.7, 128.8, 128.9, 129.0, 129.1, 2), 1.90–2.28 (4H, m, CH2 × 2), 2.86–3.05 (66H, m, NCH2CH3 ×
129.2, 129.3, 129.4, 134.9, 135.1, 135.4, 135.6, 135.7, 172.6, 33), 3.57–3.59 (1H, m, CH), 3.82 (1H, t, J = 5.6 Hz, CH), 3.99
173.1. IR (KBr) 2930, 1740, 1460, 1380, 1300, 1160, 1020, 860, (2H, bs, CH × 2), 4.08–4.16 (4H, m, CH × 2, CH2OCO),
770, 730 cm−1. HRMS(FAB) m/z calcd for C70H84O28P6Na 4.26–4.40 (2H, m, CH2OP), 5.13 (1H, bs, CH2CHCH2). HRMS
1581.3473. Found: 1581.3435 (M + Na)+. Mp 98 °C. Anal. Calcd (FAB) m/z calcd for C23H47O28P6 957.0680. Found: 957.0756
for C70H84O28P6: C, 5.57; H, 53.92. Found: C, 5.57; H, 54.37. (M − H)+.
Rf 0.46 (CH2Cl2–MeOH = 10 : 1).
3.22. DL-2,3,4,5,6-Penta-O-[(1,5-dihydro-2,4,3-benzodioxa-
3.19.2 Compound 27. 1H NMR (CDCl3) δ: 0.75–0.82 (6H,
phosphepin-3-yl)phosphoryl]-myo-inositol 1-{[2-O-heptanoyl-
1-O-methyl-sn-glyceryl](benzyl)phosphate} (28) and
DL-1,3,4,5,6-penta-O-[(1,5-dihydro-2,4,3-benzodioxa-
m, CH3 × 2), 1.12–1.19 (12H, m, CH2 × 6), 1.40–1.58 (4H, m,
CH2 × 2), 2.17–2.25 (4H, m, CH2 × 2), 3.99–4.37 (6H, m, CH × 2,
CH2OP, CH2OCO), 4.48–4.64 (2H, m, CH × 2), 4.70–5.77 (25H,
phosphepin-3-yl)phosphoryl]-myo-inositol 2-{[2-O-heptanoyl-1-
m, CH × 2, CH2CHCH2, C6H4(CH2)2 × 5, CH2C6H5), 7.17–7.44
O-methyl-sn-glyceryl](benzyl)phosphate} (29)
(25H, m, C6H4 × 5, C6H5). 13C NMR (CDCl3) δ: 14.1, 22.6, 24.8,
28.9, 31.6, 34.1, 61.8, 61.9, 65.8, 67.1, 67.2, 69.3, 69.6, 69.7, 22 (0.117 g, 0.54 mmol) was allowed to react under the same
69.8, 70.4, 70.9, 71.0, 73.5, 76.3, 76.7, 77.4, 128.0, 128.3, 128.4, conditions as described for the preparation of 27 to give 28
128.5, 128.6, 128.9, 129.0, 129.1, 129.2, 129.3, 129.4, 129.6, (0.098 g, 63%) as a white solid and compound 29 (0.018 g,
129.7, 129.8, 134.7, 135.0, 135.1, 135.6, 135.7, 135.8, 135.9, 11%) as a white solid.
136.0, 173.0, 173.4. IR (KBr) 2930, 1740, 1460, 1300, 1020, 860,
770, 730 cm−1. HRMS(FAB) m/z calcd for C70H84O28P6Na m, CH3), 1.19–1.28 (6H, m, CH2 × 3), 1.42–1.63 (2H, m, CH2),
3.22.1 Compound 28. 1H NMR (CDCl3) δ: 0.77–0.88 (3H,
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5019

View Article Online
Paper
Organic & Biomolecular Chemistry
2.21–2.27 (2H, m, CH2), 3.17–3.31 (5H, m, OCH3, CHCH2), added completely dissolved compound 12 (0.061 g,
3.45–4.53 (2H, m, CH2CH), 4.25–4.38 (2H, m, CH × 2), 0.0549 mmol) in CH2Cl2 (10 ml) and CH3CN (5 ml) with
4.88–5.75 (29H, m, CH × 4, CH2C6H5, CH2OP, CH2CHCH2, MS4A, followed by adding 1H-tetrazole (0.035 g, 0.50 mmol),
(CH2)2C6H5 × 5), 7.14–7.48 (25H, m, C6H4 × 5, C6H5). IR (KBr) and the resulting mixture was stirred at room temperature for
2930, 1740, 1460, 1380, 1290, 1230, 860, 730, 700 cm−1. HRMS further 24 h. To the mixture was added tert-butylhydroperoxide
(FAB) m/z calcd for
C64H74O27P6Na 1483.2741. Found: (0.058 ml, 0.40 mmol), and stirred at room temperature for
1483.2659 (M + Na)+. Rf 0.63 (AcOEt–CH2Cl2–MeOH = 15 : 5 : 1). further 5 min. The mixture was purified by silica gel column
3.22.2 Compound 29. 1H NMR (CDCl3) δ: 0.80–0.86 (3H, chromatography (CH2Cl2–MeOH = 7 : 1 to 5 : 1) to afford crude
m, CH3), 1.20–1.30 (6H, m, CH2 × 3), 1.49–1.74 (2H, m, CH2), compound 30 (0.025 g, 31%) as a colorless oil.
2.24–2.33 (2H, m, CH2), 3.28–3.37 (5H, m, OCH3, CHCH2),
1H NMR (CD3OD) δ: 0.79–0.84 (6H, m, CH3 × 2), 1.10–1.23
3.45–3.59 (2H, m, CH2CH), 4.17–4.37 (2H, m, CH × 2), (12H, m, CH2 × 6), 1.47 (4H, bs, CH2 × 2), 2.51–2.89 (22H, m,
4.90–5.66 (27H, m, CH × 4, CH2C6H5, CH2OP, CH2CHCH2, CH2CH2CN × 11), 3.34–3.71 (7H, m, CH2 × 3, CH), 4.22–4.68
(CH2)2C6H5 × 5), 7.16–7.52 (25H, m, C6H4 × 5, C6H5). IR (KBr) (27H, m, CH2CH2CN × 11, CH × 5), 4.68–4.84 (2H, m, CH2),
3000, 2880, 1740, 1460, 1300, 1020, 860, 730 cm−1. HRMS(FAB) 5.33 (1H, bs, CH). HRMS(FAB) m/z calcd for C54H81N11O26P6
m/z calcd for C64H74O27P6Na 1483.2741. Found: 1483.2697 (M 1508.3678. Found: 1508.3728 (M + Na)+. TLC; Rf 0.46 (CH2Cl2–
+ Na)+. Rf 0.72 (AcOEt–CH2Cl2–MeOH = 15 : 5 : 1).
MeOH = 7 : 1).
3.23. DL-1-O-[(2-O-Heptanoyl-1-O-methyl-sn-glyceryl)
hydrogen phosphoryl]-myo-inositol 2,3,4,5,6-pentakis-
(hydrogenphosphate): 3
3.26. DL-1-O-(1,2-O-Dihexyl-sn-glyceryl) hydrogen phosphoryl]-
myo-inositol 2,3,4,5,6-pentakis(hydrogenphosphate): 4
To a solution of 30 (0.025 g, 0.0168 mmol) in MeOH (5 ml)
28 (0.098 g, 0.0671 mmol) was allowed to react under the same was added 25% NH4OH (5 ml, 66.4 mmol), and the resulting
conditions as described for the preparation of 2 to give 3 mixture was stirred at 55 °C for 12 h. The mixture was concen-
(58.2 mg, 44%) as a white solid.
trated under reduced pressure, and the residue was adapted to
1H NMR (D2O) δ: 0.22 (3H, t, J = 6.4 Hz, CH3), 0.42–0.66 reverse phase chromatography (C18 column, 5 g, 50% CH3CN
(60H, m, CH2 × 3, NCH2CH3 × 18), 0.98 (2H, t, J = 6.8 Hz, CH2), to 100% CH3CN). The resulting eluted fraction was concen-
1.80 (2H, m, CH2), 2.55–2.57 (36H, m, NCH2CH3 × 18), 2.74 trated under reduced pressure. The residue was dissolved in
(3H, s, OCH3), 3.06 (2H, t, J = 6.0 Hz, CHCH2), 3.33 (2H, J = H2O (2 ml), and filtered through the cation-exchange resin. To
5.5 Hz, CH2CH), 3.44–3.54 (1H, m, CH), 3.61–3.70 (3H, m, the resulting filtrate was added triethylamine (0.0460 ml,
CH × 3), 3.81–3.94 (2H, m, CH × 2), 4.55–4.64 (1H, bs, 0.337 mmol), and concentrated under reduced pressure. The
CH2CHCH2). HRMS(FAB) m/z calcd for C16H37O26P6 858.9948. resulting residue was dissolved in H2O, and lyophilized to
Found: 859.0034 (M − H)+.
afford 4 (0.016 g, 64%) as a colorless oil.
1H NMR (D2O) δ: 0.76 (6H, bs, CH3), 1.14–1.19 (66H, m,
CH2 × 6, NCH2CH3 × 18), 1.38–1.47 (4H, m, CH2 × 2), 3.05–3.12
(36H, m, NCH2CH3 × 18), 3.41–3.67 (7H, m, CH2 × 3, CH),
3.92–4.19 (5H, m, CH × 5), 4.43–4.88 (3H, m, CH2OP, CH).
3.24. DL-2-O-[(2-O-Heptanoyl-1-O-methyl-sn-glyceryl)
hydrogen phosphoryl]-myo-inositol 1,3,4,5,6-pentakis-
(hydrogenphosphate): 3′
29 (0.018 g, 0.0121 mmol) was allowed to react under the same HRMS(FAB) m/z calcd for C21H47O26P6 901.0781. Found:
conditions as described for the preparation of 2 to give 3′ 901.0793 (M − H)+.
(0.0051 g, 22%) as a white solid.
3.27. DL-2,3,4,5,6-Penta-O-[bis(2-cyanoethyl)phosphoryl]-
1H NMR (D2O) δ: 0.74 (3H, t, J = 6.2 Hz, CH3), 1.05–1.18
myo-inositol 1-{[2-O-hexyl-1-O-methyl-sn-glyceryl](2-cyanoethyl)-
phosphate} (31)
(114H, m, CH2 × 3, NCH2CH3 × 36), 1.50 (2H, t, J = 7.3 Hz,
CH2), 2.29–2.35 (2H, m, CH2), 2.93–3.19 (72H, m, NCH2CH3 ×
36), 3.22–3.33 (5H, s, OCH3, CH2CH), 3.56–3.57 (2H, m, 25 (0.090 g, 0.473 mmol) was allowed to react under the same
CH2CH), 3.86–3.89 (1H, m, CH), 4.22–4.48 (5H, m, CH × 5), conditions as described for the preparation of 30 to give 31
5.03–5.13 (1H, m, CH2CHCH2). HRMS(FAB) m/z calcd for (0.023 g, 41%) as a colorless oil.
C16H37O26P6 858.9948. Found: 858.9951 (M − H)+.
1H NMR (CD3OD) δ: 0.87 (3H, bs, CH3), 1.10–1.30 (6H, m,
CH2 × 3), 1.56 (2H, bs, CH2), 2.99 (22H, bs, CH2CH2CN × 11),
3.29–3.74 (10H, m, OCH3, CH2 × 3, CH), 4.30–4.49 (22H, m,
CH2CH2CN × 11), 4.74–4.97 (5H, m, CH × 5), 5.41 (1H, s, CH).
HRMS(FAB) m/z calcd for C49H71N11O26P6 1438.2895. Found:
3.25. DL-2,3,4,5,6-Penta-O-[bis(2-cyanoethyl)phosphoryl]-myo-
inositol 1-{[1,2-O-dihexyl-sn-glyceryl](2-cyanoethyl)phosphate}
(30)
To a solution of 20 (0.098 g, 0.378 mmol) in CH2Cl2 (5 ml) was 1438.2861 (M + Na)+. TLC; Rf 0.35 (CH2Cl2–MeOH = 7 : 1).
added (2-cyanoethyl)-N,N,N′,N′-tetraisopropylphosphoramidite
3.28. DL-1-O-[(2-O-Hexyl-1-O-methyl-sn-glyceryl) hydrogen
phosphoryl]-myo-inositol 2,3,4,5,6-pentakis(hydrogenphosphate): 5
(0.150 ml, 0.473 mmol) followed by MS4A (0.10 g), and the
resulting mixture was stirred at room temperature under argon
for 15 min. To the mixture was added 1H-tetrazole (0.026 g, 31 (0.023 g, 0.0164 mmol) was allowed to react under the same
0.378 mmol), and the resulting mixture was stirred at room conditions as described for the preparation of 4 to give 5
temperature under argon for 10 min. To the mixture was (0.0147 g, 63%) as a colorless oil.
5020 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
1H NMR (D2O) δ: 0.72 (3H, bs, CH3), 1.11–1.16 (60H, m, sensor chip. Association was followed for 3 min, and dis-
CH2 × 3, NCH2CH3 × 18), 1.44 (2H, bs, CH2), 3.01–3.09 (36H, sociation was measured at a flow rate of 20 μL min−1 at 25 °C.
m, NCH2CH3 × 18), 3.25 (3H, s, OCH3), 3.43–3.65 (5H, m, The surfaces were regenerated by injecting three 15 s pulses of
CH2 × 2, CH), 3.97–4.09 (5H, m, CH × 5), 4.36–4.72 (3H, m, 50 mM NaOH in 1 M NaCl, three 15 s pulses of 50 mM NaOH,
CH2OP, CH). HRMS(FAB) m/z calcd for C16H37O26P6 830.9999. and then a single 15 s pulse of 10 μM IP4. The resulting sur-
Found: 830.9959 (M − H)+.
faces were post conditioned by injecting three 15 s pulses of
10 mM NaOH. Analysis of the response was performed using
evaluation software supplied with the instrument (BIAevalua-
3.29. Plasmids, cells, and transfection
The designated pEF-Gag (p17) cFLAG was used for expression tion version 3.1). To eliminate small bulk refractive change
vectors of the MA domain. 293 T cells24 were cultured in Dul- differences at the beginning and end of each injection,
becco’s modified Eagle’s medium supplemented with 10% binding responses were referenced by subtracting the response
heat-inactivated FBS. The calcium phosphate coprecipitation generated across a surface modified with biotin.
method25 was used for the transfection of 293 T cells. Trans-
fected cells were cultured at 37 °C for 48 h before use in
protein purification.
3.33. Equilibrium-binding measurement
To determine Kd values, 1.96 μM MA was mixed with various
concentrations of inositol phosphates, phosphatidylinositols.
After reaching equilibrium (less than 30 min in all cases at
25 °C), 60 μL of each mixture was injected over the IP4 surface
at 20 μL min−1 to quantify the free MA remaining in the equili-
brium mixture. The Kd was obtained by fitting the data to a
solution affinity model using BIA evaluation 3.1: Afree = 0.5(B −
A − Kd) + (0.25(A + B + Kd)2 − AB)0.5, where A = initial concen-
tration of proteins, Afree = concentration of unbound proteins
remaining in the equilibrium mixture, and B = initial concen-
tration of IP4.
3.30. Protein purification
Vector-transfected 293 T cells were lysed with TNE buffer
(10 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1% NP-40, and
10 μg mL−1 aprotinin, pH 7.8) containing 1 mM dithiothreitol
(DTT). After centrifugation (12 000 rpm, 4 °C, 5 min), the
supernatant was mixed with Sepharose CL-4B (Sigma-Aldrich,
St. Louis, MO), and the resulting suspension was incubated for
2 h at 4 °C. This incubation was repeated twice, and the final
supernatant was treated with mouse anti-FLAG M2 affinity gel
(Sigma-Aldrich, St. Louis, MO) and 0.5 ng mL−1 1× FLAG
peptide (Sigma-Aldrich, St. Louis, MO), to remove nonspecific
components interacting with the FLAG antibody, and incu-
bated for 8 h at 4 °C. The beads were washed five times with
TNE buffer plus 1 mM DTT. A solution of 150 μg mL−1 3×
FLAG peptide (Sigma-Aldrich, St. Louis, MO) in TBS buffer
(50 mM Tris-HCl and 150 mM NaCl, pH 7.4) with 1 mM DTT
was loaded onto the beads and incubated for 30 min at 4 °C.
Following centrifugation, the resulting supernatant was used
for the SPR assay.
3.34. Molecular docking methodology
Docking studies were performed using MOE 2012.10. The
crystal structure of myr-MA (PDB code: 1UPH)26 was obtained
from the Protein Data Bank to prepare protein for docking
studies. The docking procedure was followed using the stan-
dard protocol implemented in MOE 2012.10. To the structure
was added hydrogen atoms and electric charge by Protonate
3D, and the resulting structure was optimized by Amber12:
EHT, and then the dummy atoms were disposed in the
docking site using Site finder (Alpha Site Setting; probe radius
1: 1.4 Å, probe radius 2: 1.8 Å, isolated donor/acceptor: 3 Å,
connection distance: 2.5 Å, minimum size: 3 Å, and radius:
2 Å). The docking simulation was carried out by ASEDock. The
targeting ligands were assigned in ASEDock, and the confor-
mations were integrated by LowModeMD based on the algor-
ithm of conformation analysis (Step 1; cutoff: 4.5 Å, RMS (root
mean square) gradient: 10 kcal mol−1 Å−1, energy threshold:
500 kcal mol−1, Step 2; optimize 5 lowest energy or 5 best score
conformation, cutoff: 8 Å, RMS gradient: 0.1 kcal mol−1 Å−1).
3.31. Protein quantification
The cFLAG proteins were resolved by SDS-PAGE followed by
Coomassie Brilliant Blue (CBB) staining. Each gel band was
quantified using ImageJ (version 1.38x) software, and protein
concentrations were determined by comparing the intensity of
protein bands with the intensity of a protein marker.
3.32. SPR studies
A
BIACORE2000 (GE Healthcare, BIACORE AB, Uppsala,
Sweden) was used as the surface plasmon resonance bio-
sensor. To prepare the IP4 immobilized sensor chip surface for
9
the BIACORE, biotinylated IP4 in HEPES buffer (50 mM
HEPES, 500 mM NaCl, 3.4 mM EDTA, and 0.005% Tween 20,
pH 7.4) was injected over streptavidin covalently immobilized
4. Conclusion
upon the sensor chip surface (Sensor Chip SA, GE Healthcare, In this study, lipid-coupled myo-inositol 1,2,3,4,5,6-hexa-
BIACORE AB, Uppsala, Sweden) until a suitable level was kisphosphate (IP6) derivatives having both IP6 and diacylgly-
achieved. The flow buffer contained 10 mM HEPES, 150 mM cerol moiety that could interact with the HIV-1 MA domain
NaCl, 3.4 mM EDTA, 0.005% Tween 20, 2% (v/v) glycerol, and were designed and synthesized. These compounds, in fact,
0.5 mg mL−1 BSA (pH 7.8). Purified proteins were dialyzed bound to the MA domain more tightly than the PIP2 derivative
against flow buffer and injected over the immobilized IP4 1 or IP6 does and may provide the structural basis of the
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5006–5022 | 5021

View Article Online
Paper
Organic & Biomolecular Chemistry
molecular design of novel anti-HIV agents that block the mem-
9 K. Anraku, T. Inoue, K. Sugimoto, K. Kudo, Y. Okamoto,
T. Morii, Y. Mori and M. Otsuka, Bioorg. Med. Chem., 2011,
19, 6833–6841.
brane localization of Pr55Gag
.
10 E. O. Freed, J. M. Orenstein, A. J. Buckler-White and
M. A. Martin, J. Virol., 1994, 68, 5311–5320.
11 W. Zhou, L. J. Parent, J. W. Wills and M. D. Resh, J. Virol.,
1994, 68, 2556–2569.
12 C. G. Ferguson, R. D. James, C. S. Bigman, D. A. Shepard,
Y. Abdiche, P. S. Katsamba, D. G. Myszka and
G. D. Prestwich, Bioconjugate Chem., 2005, 16, 1475–1483.
13 D. C. Billington, R. Baker, J. J. Kulagowski and I. M. Mawer,
J. Chem. Soc., Chem. Commun., 1987, 314–316.
14 W. Bannwarth and A. Trzeciak, Helv. Chim. Acta, 1987, 70,
175–186.
Acknowledgements
This work was supported in part by a Grant-in-Aid for Explora-
tory Research (B) (23390028) (to M.O.), a Grant-in-Aid for
Young Scientists (B) (24790124) (to K.A.), and a Grant-in-Aid
for Scientific Research (C) (19659025) (to M.F.) from the Japan
Society for the Promotion of Science, and with the aid of a
special fellowship for culture, education and science (to K.A.)
granted by Kumamoto Health Science University.
15 N. Nagashima and M. Ohno, Chem. Lett., 1987, 141–144.
16 C. Liu and B. L. Potter, J. Org. Chem., 1997, 62, 8335–8340.
17 Y. Oikawa, T. Tanaka, K. Horita and O. Yonemitsu, Tetra-
hedron Lett., 1984, 25, 5397–5400.
References
18 S. Ozaki, Y. Kondo, N. Shiotani, T. Ogasawara and
Y. Watanabe, J. Chem. Soc., Perkin Trans. 1, 1992, 729–737.
19 B. Classon, P. Garegg and B. Samuelsson, Acta Chem.
Scand., Ser. B, 1984, 38, 419–422.
1 J. W. Wills and R. C. Craven, AIDS, 1991, 5, 639–654.
2 E. O. Freed, Virology, 1998, 251, 1–15.
3 M. R. Conte and S. Matthews, Virology, 1998, 246,
191–198.
20 E. Uhlmann and J. Engels, Tetrahedron Lett., 1986, 27,
1023–1026.
21 E. O. Freed, J. M. Orenstein, A. J. Buckler-White and
M. A. Martin, J. Virol., 1994, 68, 5311–5320.
22 J. Chan, R. Dick and V. M. Vogt, J. Virol., 2011, 85, 10851–
10860.
4 A. Ono, S. D. Ablan, S. J. Lockett, K. Nagashima and
E. O. Freed, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 14889–
14894.
5 J. S. Saad, J. Miller, J. Tai, A. Kim, R. H. Ghanam and
M. F. Summers, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
11364–11369.
23 J. Inlora, V. Chukkapalli, D. Derse and A. Ono, J. Virol.,
2011, 85, 3802–3810.
24 J. S. Lebkowski, S. Clancy and M. P. Calos, Nature, 1985,
317, 169–171.
25 A. Adachi, H. E. Gendelman, S. Koenig, T. Folks, R. Willy,
A. Rabson and M. A. Martin, J. Virol., 1986, 59, 284–291.
26 C. Tang, E. Loeliger, P. Luncsfold, I. Kinde, D. Beckett and
M. F. Summers, Proc. Natl. Acad. Sci. U. S. A., 2004, 101,
517–522.
6 N. Shkriabai, S. A. Datta, Z. Zhao, S. Hess, A. Rein
and M. Kvaratskhelia, Biochemistry, 2006, 45, 4077–
4083.
7 K. Anraku, R. Fukuda, N. Takamune, S. Misumi,
Y. Okamoto, M. Otsuka and M. Fujita, Biochemistry, 2010,
49, 5110–5116.
8 K. Anraku, T. Inoue, K. Sugimoto, Y. Okamoto, T. Morii,
Y. Mori and M. Otsuka, Org. Biomol. Chem., 2008, 6, 1822–
1830.
5022 | Org. Biomol. Chem., 2014, 12, 5006–5022
This journal is © The Royal Society of Chemistry 2014