Facile and efficient synthesis of bolaamphiphilic tetraether phosphocholines

Article abstract of DOI:10.1021/jo9803176

A facile synthesis of ether-linked bolaform phospholipids in good yields has been developed. Triflic acid catalyzed oxirane ring opening of benzyl-protected rac-glycidyl with long-chain l,ω-alkanediols (n = 16, 20) produced 1,1′-diglycerol diethers in 80-90% yield. Double alkylation of the secondary hydroxy groups with 1-bromooctane or 1-bromodecane gave the corresponding benzyl-protected tetraethers in 66% yield. Hydrogenolysis of the benzyl groups in the presence of Pd/C (55-66% yield) followed by phosphorylation with 2 equiv of 2-chloro-2-oxo-1,3,2-dioxaphospholane and amination with excess trimethylamine produced the tetraether bolaform bisphosphocholines as white powders in ～75% yield. This approach provides a reliable and efficient method for preparing a wide variety of symmetrical bolaform phospholipids on a multigram scale.

Full text of DOI:10.1021/jo9803176

7180
J . Org. Chem. 1998, 63, 7180-7182
F a cile a n d Efficien t Syn th esis of Bola a m p h ip h ilic Tetr a eth er
P h osp h och olin es
So¨nke Svenson† and David H. Thompson*
Department of Chemistry, Purdue University, 1393 Brown Building, West Lafayette, Indiana 47907-1393
Received February 19, 1998
A facile synthesis of ether-linked bolaform phospholipids in good yields has been developed. Triflic
acid catalyzed oxirane ring opening of benzyl-protected rac-glycidyl with long-chain 1,ω-alkanediols
(n ) 16, 20) produced 1,1′-diglycerol diethers in 80-90% yield. Double alkylation of the secondary
hydroxy groups with 1-bromooctane or 1-bromodecane gave the corresponding benzyl-protected
tetraethers in 66% yield. Hydrogenolysis of the benzyl groups in the presence of Pd/C (55-66%
yield) followed by phosphorylation with 2 equiv of 2-chloro-2-oxo-1,3,2-dioxaphospholane and
amination with excess trimethylamine produced the tetraether bolaform bisphosphocholines as
white powders in ∼75% yield. This approach provides a reliable and efficient method for preparing
a wide variety of symmetrical bolaform phospholipids on a multigram scale.
In tr od u ction
materials are generally quite low. We sought a new
synthesis that would utilize readily available starting
materials that could be converted in good yields to
phosphocholine-type bolaamphiphiles in a minimum
number of steps. Previously, we reported the reaction
of solketal with long-chain 1,ω-dibromoalkanes, which
required the synthesis of 1,16-dibromohexadecane or 1,-
20-dibromoeicosane from the corresponding diol or diacid,
respectively.6 This pathway produced 1,1′-polymethylene
diglycerols with four hydroxy groups that required selec-
tive protection prior to installation of the secondary ether
substituents. Efficient and selective protection of the
primary hydroxy groups, however, proved to be difficult.
Protection with triphenylmethyl chloride gave the desired
product in only 30% yield, with the triply protected
diglycerol recovered as the main product. Unfortunately,
replacement of trityl chloride with tert-butyldiphenyl-
chlorosilane (TBDPS-Cl) did not improve the yield; about
50% of the silyl ether was hydrolyzed to the silanol, even
when the reactions employed lyophilized 1,1′-polymeth-
ylene diglycerol precursors in the presence of molecular
sieves,20 thus limiting the yield of the doubly TBDPS
protected diglycerol to 20-30%. Reactions employing an
excess of silyl chloride resulted in an increased production
of the triply protected diglycerol.
Tetraether bolaamphiphiles have been the focus of
several ongoing investigations in this laboratory.1-7 Our
interest in these materials stems from their ability to
form stable membrane vesicles with lamellar thicknesses
on the order of 2.4 nm,3,5 a dimension that is ideally
suited for studies of coupled electron and ion transport
phenomena across vesicle membranes using vectorially
inserted gramicidin-based diads and triads.1,2 Large
quantities of bisphosphocholine bolaamphiphiles are
required for these studies because they are used as the
host membrane matrix. Synthetic pathways for bispho-
sphoric6 and bisphosphocholine7 bolaamphiphiles 8 and
9, as well as other bolaform derivatives, have been
described;8-19 however, the overall yields of purified
* Fax: 765-496-2592. E-mail: davethom@chem.purdue.edu.
† Current address: Division of Biological Materials, Northwestern
University, 10-019 Ward Building, 311 E. Chicago Avenue, Chicago,
IL 60611. E-mail: s-svenson@nwu.edu.
(1) Thompson, D. H.; Kim, J .-M. Mater. Res. Soc. Symp. Proc. 1992,
277, 93-98.
(2) Thompson, D. H.; Kim, J .-M.; Di Meglio, C. Proc. SPIE Int. Soc.
Opt. Eng. 1993, 1853, 142-147.
(3) Thompson, D. H.; Wong, K.; Humphry-Baker, R.; Wheeler, J .;
Kim, J .-M.; Rananavare, S. B. J . Am. Chem. Soc. 1992, 114, 9035-
9042.
(4) DiMeglio, C.; Svenson, S.; Haynes, R.; Thompson, D. H., manu-
script in preparation.
(5) DiMeglio, C.; Rananavare, S. B.; Thompson, D. H., manuscript
in preparation.
(6) Kim, J .-M.; Thompson, D. H. Langmuir 1992, 8, 637-644.
(7) Thompson, D. H.; Svendsen, C. B.; DiMeglio, C.; Anderson, V.
C. J . Org. Chem. 1994, 59, 2945-2955.
(8) Duivenvoorde, F. L.; Feiters, M. C.; van der Gaast, S. J .;
Engberts, J . F. B. N. Langmuir 1997, 13, 3737-3743.
(9) Fyles, T. M.; Loock, D.; van Straaten Nijenhus, W. F.; Zhou, X.
J . Org. Chem. 1996, 61, 8866-8874.
(10) Eguchi, T.; Arakawa, K.; Terachi, T.; Kakinuma, K. J . Org.
Chem. 1997, 62, 1924-1933.
(11) Berkowitz, W. F.; Pan, D.; Bittman, R. Tetrahedron Lett. 1993,
34, 4297-4300.
(12) Bellucci, G.; Catelani, G.; Chiappe, C.; D’Andrea, F.; Grigo, G.
Tetrahedron: Asymmetry 1997, 8, 765-773.
(13) Velty, R.; Benvegnu, T.; Plusquellec, D. Synlett 1996, 817-819.
(14) Shimizu, T.; Masuda, M. J . Am. Chem. Soc. 1997, 119, 2812-
2818.
(15) Menger, F. M.; Littau, C. A. J . Am. Chem. Soc. 1993, 115,
10083-10090.
(16) Lafont, D.; Boullanger, P.; Chevalier, Y. J . Carbohydr. Chem.
1995, 14, 533-550.
Our previously reported synthesis of 8 and 9 required
six steps starting from oxiranemethanol 3-nitrobenzene-
sulfonate (NBS) and long-chain 1,ω-alkanediols.7 The
primary limitations of this pathway are (i) the second
alkylation step which uses alkyl triflates in the presence
of 1,8-bis(dimethylamino)naphthalene, (ii) the cleavage
of the NBS protecting group under conditions that do not
generate either elimination or oligomeric products, and
(iii) the relatively high costs of chiral glycidyl sulfonates
(17) Elferink, M. G. L.; de Wit, J . G.; Driessen, A. J . M.; Konings,
W. N. Biochim. Biophys. Acta 1994, 1193, 247-254.
(18) Ladika, M.; Fisk, T. E.; Wu, W. W.; J ons, S. D. J . Am. Chem.
Soc. 1994, 116, 12093-12094.
(19) Song, L. D.; Rosen, M. J . Langmuir 1996, 12, 1149-1153.
(20) The four hydroxy groups presumably trap water molecules, via
hydrogen bonding, between lamellar plates formed by association of
the polymethylene segments in a manner similar to that reported for
calixarene solids. See: Atwood, J . L.; Bott, S. G. In Calixarenes:
A
Versatile Class of Macrocyclic Compounds; Vicens, J .; Bo¨hmer, V., Eds.;
Kluwer Academic: Dordrecht, Holland, 1991; pp 199-210.
S0022-3263(98)00317-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/25/1998

Bolaamphiphilic Tetraether Phosphocholines
J . Org. Chem., Vol. 63, No. 21, 1998 7181
F igu r e 2. 1H and 13C APT (attached proton test) NMR spectra
of bolaform phosphocholine 8.
aphospholane/trimethylamine route,27-29 to give the bo-
laform tetraether phosphocholines 8 and 9 as white
powders in ∼75% isolated yield. The structures of the
phosphocholine bolaamphiphiles 8 and 9 were confirmed
F igu r e 1. Synthesis pathway to bolaamphiphilic tetraether
phosphocholines.
1
by H and 13C NMR spectroscopy (Figure 2).
Con clu sion s
and alkyl triflates. The tedious chromatographic puri-
fications required after each step and the low overall
yields observed also made this synthesis impractical for
large scale preparations.
A facile synthesis of ether-linked bolaform phospho-
lipids has been developed. The approach is efficient and
utilizes readily purified intermediates, making it ame-
nable to the large scale production of these bolaam-
phiphiles. Application of enantiomerically pure glyceryl
synthons, in principle, should also enable the synthesis
of chiral bolaform phospholipids.
Resu lts a n d Discu ssion
The synthesis described here (Figure 1) utilizes benzyl
glycidyl ether (1), formed by coupling benzyl alcohol
with epichlorohydrin as the glycerol precursor in >90%
yield.21,22 Triflic acid catalyzed oxirane ring opening
with long-chain 1,ω-alkanediols gave the benzyl-protected
diether diols 2 and 3 in 80-90% yield.23 It proved
important to use a 3:1 epoxide/alkanediol ratio in this
step because lower ratios (e.g., 2:1) reduced the yields to
30-40% and left some unreacted alkanediol which was
difficult to chromatographically separate from the prod-
ucts 2 and 3. Also, very small catalytic amounts of triflic
acid (∼2 mol %) were required, as higher concentrations
of catalyst were found to decrease the yield of the diether
diols 2 and 3 via addition of a third equivalent of 1.
Subsequent alkylation of the secondary hydroxy groups,
carried out via Williamson synthesis using sodium hy-
dride and the appropriate 1-bromoalkane,24,25 gave the
benzyl-protected tetraethers 4 and 5 in 66% yield.26
Hydrogenolysis of the benzyl groups was conducted in
the presence of palladium on carbon and gave the
tetraether diols 6 and 7 in moderate yield (55-66%).27
The purified diols 6 and 7 were then phosphorylated and
aminated via the established 2-chloro-2-oxo-1,3,2-diox-
Exp er im en ta l Section
All yields reported are for isolated products (>95% pure).
Melting points were determined with a Perkin-Elmer DSC-7/
intracooler.6 All products are characterized by 1H and 13C
NMR; satisfactory mass spectra and elemental analyses were
obtained where appropriate. NMR chemical shifts are re-
ported in ppm with tetramethylsilane and solvent peaks as
internal standards for the proton and carbon spectra, respec-
tively.
Ma ter ia ls. All reagents were used as received. Solvents
were purified by distillation from an appropriate desiccant.
1-O-Ben zyl-r a c-glycid ol (1). A mixture of aqueous so-
dium hydroxide (20 mL, 50% w/w), racemic epichlorohydrin
(14.2 g, 153.4 mmol), and tetrabutylammonium hydrogen
sulfate (0.5 g, 1.47 mmol) was vigorously stirred at room
temperature. Benzyl alcohol (3.3 g, 30.9 mmol) was then
added so that the temperature did not exceed 25 °C and the
mixture stirred at room temperature for 20 h. Cold water (80
mL) was added, the aqueous phase extracted (2 × 100 mL)
with ether, and the extract washed (50 mL) with a brine
solution. The ether phase was dried over magnesium sulfate
and the solvent removed by rotary evaporation. Filtration of
the remaining oil through a 3-cm plug of silica gel (6-cm
diameter) using 5:1 hexane/ether as eluent gave 1 in 92%
yield (4.65 g, 28.3 mmol). TLC: Rf ) 0.25, 5:1 hexane/ether.
1H NMR (CDCl3): 2.63 (dd, J AM ) 2.6 Hz, J AB ) 5.0 Hz, 1H),
2.81 (dd, J BM ) 4.4 Hz, J AB ) 5.0 Hz, 1H), 3.20 (m, 1H),
(21) Mouzin, G.; Cousse, H.; Rieu, J .-P.; Duflos, A. Synthesis 1983,
117-119.
(22) Bittman and co-workers have also reported the synthesis of
chiral benzyl glycidyl ethers from chiral glycidyl tosylates. See: Byun,
H.-S.; Bittman, R. Tetrahedron Lett. 1989, 30, 2751-2754.
(23) Similar efficiencies were observed in the BF3‚Et2O-catalyzed
addition of ω-chloroalkan-1-ols to rac-benzyl glycidyl ether. See: ref
11.
3.44 (dd, J A′M ) 5.8 Hz, J A′B′ ) 11.4 Hz, 1H), 3.78 (dd, J B′M
3.0 Hz, J A′B′ ) 11.4 Hz, 1H), 4.60 (d, 2H), 7.36 (m, 5H).
)
(27) Menger, F. M.; Chen, X. Y.; Brocchini, S.; Hopkins, H. P.;
Hamilton, D. J . Am. Chem. Soc. 1993, 115, 6600-6608.
(28) Yamauchi, K.; Moriya, A.; Kinoshita, M. Biochim. Biophys. Acta
1989, 1003, 151-160.
(29) Yamauchi, K.; Sakamoto, Y.; Moriya, A.; Yamada, K.; Hosoka-
wa, T.; Higushi, T.; Kinoshita, M. J . Am. Chem. Soc. 1990, 112, 3188-
3191.
(24) Feuer, H.; Hooz, J . In The Chemistry of the Ether Linkage; Patai,
S., Ed.; Wiley-Interscience: New York, 1967; pp 445-498.
(25) Bhatia, S. K.; Hajdu, J . Synthesis 1989, 16-20.
(26) The method of Berkowitz et al.11 required an additional three
steps to effect the synthesis of closely related benzyl-protected tetra-
ethers in lower net yields.

7182 J . Org. Chem., Vol. 63, No. 21, 1998
Svenson and Thompson
1,1′-Di-O-h exa d eca m et h ylen e-3,3′-O-d ib en zyld i-r a c-
glycer ol (2). Triflic acid (50 µL) was added to a solution of 1
(4.0 g, 24.4 mmol) and 1,16-hexadecanediol (2.0 g, 7.8 mmol)
in 100 mL of dry CHCl3. The mixture was heated at reflux
for 48 h and the solvent removed by evaporation. The
remaining oil was purified by filtration through a 3-cm plug
of silica gel (6-cm diameter) using 2:1 hexane/ethyl acetate as
eluent to give the benzyl-protected diether diol 2 in 90% yield
(4.12 g, 7.0 mmol). TLC: Rf ) 0.21, 2:1 hexane/ethyl acetate.
1H NMR (CDCl3): 1.28 (s, 24H), 1.58 (m, 4H), 2.38 (s, 2H),
3.34-3.75 (m, 12H), 3.98 (m, 2H), 4.56 (d, 4H), 7.34 (m, 10H).
13C NMR (CDCl3): 26.1, 29.6, 69.5, 71.7, 73.4, 127.7, 128.4,
138.0. MS (pos. FAB): m/z 609 (35%, [M + Na]+), 587 (100%,
[M + H]+).
1,1′-Di-O-eicosa m eth ylen e-3,3′-O-d iben zyld i-r a c-glyc-
er ol (3). This was prepared as described for 2, except 1,20-
eicosanediol (2.5 g, 8.1 mmol) was used as substrate. Yield:
80% (4.18 g, 6.5 mmol). TLC: Rf ) 0.24, 2:1 hexane/ethyl
acetate. 1H NMR (CDCl3): 1.28 (s, 32H), 1.58 (m, 4H), 2.30
(s, 2H), 3.30-3.75 (m, 12H), 3.98 (m, 2H), 4.56 (d, 4H), 7.34
(m, 10H). MS (pos. FAB): m/z 643 (100%, [M + H]+).
1,1′-Di-O-h exadecam eth ylen e-2,2′-O-dioctyl-3,3′-O-diben -
zyld i-r a c-glycer ol (4). The diether diol 2 (6.0 g, 10.2 mmol),
dissolved in 60 mL of dry THF, was added to a suspension of
sodium hydride (2.1 g, 52.5 mmol; 60% dispersion in mineral
oil) in 90 mL of dry THF. When the gas evolution ceased,
1-bromooctane (19.0 g, 98.4 mmol) was added and the mixture
heated at reflux for 3 days. The mixture was filtered through
a 2-cm plug of silica gel and the filtrate concentrated by
evaporation. The remaining viscous oil was purified by
filtration through a 3-cm plug of silica gel (12-cm diameter)
using 5:1 hexane/ethyl acetate as eluent to give the benzyl-
protected tetraether 4 in 66% yield (5.4 g, 6.7 mmol). TLC:
Rf ) 0.58, 5:1 hexane/ethyl acetate. 1H NMR (CDCl3): 0.88
(t, 6H), 1.28 (s, 44H), 1.58 (m, 8H), 3.30-3.70 (m, 18H), 4.56
(d, 4H), 7.34 (m, 10H).
1,1′-Di-O-eicosa m eth ylen e-2,2′-O-d id ecyl-3,3′-O-d iben -
zyld i-r a c-glycer ol (5). This was prepared as described for
4, except 1-bromodecane was used as substrate. Yield: 66%.
TLC: Rf ) 0.59, 5:1 hexane/ethyl acetate. 1H NMR (CDCl3):
0.88 (t, 6H), 1.30 (s, 60H), 1.58 (m, 8H), 3.35-3.75 (m, 18H),
4.56 (d, 4H), 7.34 (m, 10H). 13C NMR (CDCl3): 14.0, 22.6, 25.6,
26.0, 29.2-29.9, 31.8, 32.6, 70.1, 70.6, 71.5, 73.2, 77.8, 127.4,
128.2, 138.2.
1,1′-Di-O-h exa deca m eth ylen e-2,2′-O-d ioctyld i-r a c-glyc-
er ol (6). The benzyl protected tetraether 4 (1.3 g, 1.6 mmol)
and palladium (1.0 g; 10 wt % on carbon) were added to 80
mL of 5:1 ethanol/acetic acid. Hydrogenolysis was conducted
at room temperature at slight positive pressure using hydro-
gen. The reaction was quenched after 20 h by adding ether
(70 mL) to the mixture and filtering it through a 1-cm plug of
Celite. The Celite plug was washed (1 × 100 mL) with ether,
and the solvent was removed by evaporation. The remaining
oil was purified by filtration through a 3-cm plug of silica gel
(6-cm diameter) using 2:1 hexane/ethyl acetate as eluent. The
tetraether diol 5 was isolated in 66% yield (670 mg, 1.06 mmol)
as an oil that slowly solidified at room temperature. TLC: Rf
) 0.33, 2:1 hexane/ethyl acetate. 1H NMR (CDCl3): 0.88 (t,
6H), 1.28 (s, 44H), 1.58 (m, 8H), 2.38 (br s, 2H), 3.35-3.80 (m,
18H).
1,1′-Di-O-eicosa m et h ylen e-2,2′-O-d id ecyld i-r a c-glyc-
er ol (7). This was prepared as described for 6. Yield: 55%.
TLC: Rf ) 0.30, 2:1 hexane/ethyl acetate. Mp 38-40 °C. 1H
NMR (CDCl3): 0.90 (t, 6H), 1.30 (s, 60H), 1.58 (m, 8H), 2.16
(s, 2H), 3.35-3.85 (m, 18H). 13C NMR (CDCl3): 14.4, 22.8,
26.4, 29.2-30.4, 32.2, 33.2, 63.2, 70.6, 71.2, 72.2, 78.6.
1,1′-Di-O-h exadecam eth ylen e-2,2′-O-dioctylbis(r a c-glyc-
er o-3-p h osp h och olin e) (8). 2-Chloro-2-oxo-1,3,2-dioxaphos-
pholane (1.07 g, 7.5 mmol) was added dropwise at 10 °C to a
solution of the tetraether diol 6 (1.9 g, 3.0 mmol) and dry
triethylamine (871.0 mg, 8.6 mmol) in 30 mL of dry benzene.
The mixture was stirred for 10 min. at 10 °C and overnight at
room temperature. Crystalline Et3N‚HCl was removed by
filtration and the solvent evaporated to give the phosphory-
lated bolaamphiphile as a viscous oil. The residue was
transferred to a pressure tube with 25 mL of anhydrous CH3-
CN. The pressure tube was then cooled in a dry ice-acetone
bath, and cold (0°) anhydrous trimethylamine (4-5 mL) was
poured into the tube. Then the tube was sealed and heated
in an oil bath for 48 h at 65 °C. A white powder formed during
the reaction. The mixture was cooled in an ice bath, the excess
trimethylamine gas released, and the solution poured into a
round-bottom flask. Volatile components were removed by
evaporation and the remaining oil purified by column chro-
matography on silica gel using CHCl3 with an increasing
amount of CH3OH as eluent. The bolaform phosphocholine 7
was isolated as a white powder in 73% yield (2.1 g, 2.18 mmol).
TLC: Rf ) 0.35, 8:6:4:3 chloroform/acetone/methanol/NH4OH.
1H NMR (CD3OD): 0.88 (t, 6H), 1.28 (s, 44H), 1.50 (m, 8H),
3.16 (s, 18H), 3.35-3.65 (m, 18H), 3.82 (t, 4H), 4.18 (br m,
4H). 13C NMR (CD3OD): 14.6, 23.8, 27.4, 30.2-31.0, 33.2,
54.8, 60.5, 66.4, 67.6, 71.4, 72.8, 79.4. Anal. Calcd for
tetrahydrate: C, 55.79; H, 10.73; N, 2.71; P, 6.00. Found: C,
55.7; H, 10.89; N, 3.09; P, 6.46.
1,1′-Di-O-eicosa m eth ylen e-2,2′-O-d id ecylbis(r a c-glyc-
er o-3-p h osp h och olin e) (9). This was prepared as described
for 8. Yield: 76%. TLC: Rf ) 0.34, 8:6:4:3 chloroform/acetone/
methanol/NH4OH. 1H NMR (CD3OD): 0.88 (t, 6H), 1.30 (s,
60H), 1.58 (m, 8H), 3.24 (s, 18H), 3.40-3.75 (m, 18H), 3.88 (t,
4H), 4.28 (br m, 4H). 13C NMR (CD3OD): 14.6, 23.8, 27.4,
30.2-31.0, 33.2, 54.8, 60.5, 66.4, 67.6, 71.4, 72.8, 79.4. Anal.
Calcd for tetrahydrate: C, 58.71; H, 11.09; P, 5.41. Found:
C, 58.21; H, 11.06; P, 5.46.
Ack n ow led gm en t. The authors acknowledge the
financial support of the National Science Foundation
(MCB-9319099).
Su p p or tin g In for m a tion Ava ila ble: NMR spectra of
compounds 1-9 (19 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
J O9803176