Synthesis of lipids for development of multifunctional lipid-based drug-carriers

Article abstract of DOI:10.1016/j.bmcl.2011.08.103

A simple approach to synthesize phospholipids to modulate drug release and track lipid-based particulate drug-carriers is described. We synthesized two ether lipids, 1 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phosphocholine (C ₃₁PC) and 2 1-O-

Full text of DOI:10.1016/j.bmcl.2011.08.103

Bioorganic & Medicinal Chemistry Letters 21 (2011) 6370–6375
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of lipids for development of multifunctional lipid-based drug-carriers
⇑
Guodong Zhu, Yahya Alhamhoom, Brian S. Cummings, Robert D. Arnold
Department of Pharmaceutical and Biomedical Sciences, College of Pharmacy, University of Georgia, 215 W. Green Street, Rm 220 Athens, GA 30602 2352, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A simple approach to synthesize phospholipids to modulate drug release and track lipid-based particu-
Received 26 May 2011
Revised 22 August 2011
Accepted 25 August 2011
Available online 8 September 2011
late drug-carriers is described. We synthesized two ether lipids, 1 1-O-hexadecyl-2-pentadenoyl-sn-glyc-
erol-3-phosphocholine (C31PC) and
2 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phosphomethanol
(C31PM), and examined their ability to alter enzymatically triggered release of 6-carboxyfluorescein from
liposomes incubated in TRIS buffer or fetal bovine serum solutions. Further, we demonstrated that odd-
chain lipids, for example, C31PC, could be identified in rat plasma without interference of endogenous lip-
ids. This approach can be adapted to synthesize a variety of lipids for use in developing and optimizing
multifunctional drug-carriers.
Keywords:
Liposomes
Phospholipids
Secretory phospholipase A2
Synthesis
Ó 2011 Elsevier Ltd. All rights reserved.
Lipid based drug-carriers, such as liposomes, can alter the phar-
macokinetics and improve the efficacy of a variety of therapeutic
agents.1 However, mechanisms to ‘control’ or ‘tune’ their drug re-
lease kinetics and track drug-carrier disposition in vivo are limited.
A variety of physical and physiological approaches have been
examined to control drug release from drug-carriers, including
exposure to light,2–4 heat,5–7 and use of ultrasound.8,9 However,
the clinical use of these strategies has been limited because of their
ability to induce rapid ‘burst’ drug-release profiles and the inacces-
sibility of some tissue to light or heat.
Enzymatic approaches exploiting elevated expression of en-
zymes, such as esterases, in some disease states have the potential
to modulate drug-release selectively. Secretory phospholipase A2
(sPLA2) is an esterase that preferentially hydrolyzes glycerophos-
pholipids, such as phosphatidylcholine, at the sn-2 ester bond,
releasingafatty acid(FA) andalysophospholipid (LP).10 Recentstud-
ies demonstrated that the expression and catalytic activity of sPLA2
is increased in prostate,11–14 breast15–17 and pancreatic18–20 cancers.
We, and others, have developed sPLA2 responsive liposomes
(SPRL) with enhanced drug release.21–24 SPLA2-mediated degrada-
tion of phospholipids results in the formation of FA and LP; these
increase membrane fluidity and result in a transition of the lipid
bilayer from a gel-like to a liquid-crystalline phase. Alterations in
membrane fluidity are believed responsible for the enhanced diffu-
sion and release of contents entrapped within the aqueous-core of
liposomes; however, the complete mechanism is not fully known.
Our recent studies used electrospray ionization mass spectrometry
to examine the selectivity of sPLA2 for various lipids and quantify
sPLA2-mediated degradation of prototype liposomes formula-
tions.24 Lipid degradation was correlated to enhanced release of
6-CF, a fluorescent probe. We, and others, have hypothesized that
slight modifications of existing lipids or synthesis of novel lipids
and lipid-prodrugs could be used to further tune drug release
and improve sPLA2-activity.21
The specificity and activity of sPLA2 is greatest for phospholipids
with anionic head groups and shorter FA acyl chains.25–27 However,
optimal formulations need to balance drug retention properties
with their circulation half-life, tissue distribution, and drug release
kinetics. The use of high-phase transition, saturated phospholipids
with neutral head groups have been shown, in combination with
cholesterol (CHOL) and hydrophilic coatings (e.g., polyethylene gly-
col), to produce long-circulating, sterically-stabilized liposomes
(SSL). Unfortunately, the clinical utility of these formulations is lim-
ited, in part, because some drugs, such as doxorubicin or vincristine,
are entrapped stably and the rate of drug release is slow, whereas for
other drugs, such as topotecan or paclitaxel, the rate of release is
fast.28 The synthesis of novel lipids, based on existing lipids, may al-
low for the modulation, or tuning, of drug release.
Abbreviations: sPLA2, secretory phospholipase A2; C31PC, 1-O-hexadecyl-2-penta-
denoyl-sn-glycerol-3-phosphocholine;
C31PM, 1-O-hexadecyl-2-pentadenoyl-sn-
glycerol-3-phosphomethanol; ESI-MS, electrospray ionization-mass spectrometry;
PC, phosphatidylcholine; PM, phosphatidylmethanol; PG, phosphatidylglycerol; PE,
phosphatidylethanolamine, DPPC, 1,2-dipalmityl-sn-glycero-3-phosphatidylcho-
line; DSPC, 1,2-distearoyl-sn-glycero-3-phosphatidylcholine; DSPG, 1,2-distearoyl-
sn-glycerol-3-phosphatidylglycerol; DSPE, 1,2-distearoyl-sn-glycero-3-phosphati-
dylethanolamine; SSL, sterically-stabilized liposome; 6-CF, 6-carboxyfluorescein;
LP, lysophospholipid; FA, fatty acid; DSPE-PEG, 1,2-distearoyl-sn-glycero-3-phos-
phoethanolamine-N-[poly(ethyleneglycol)2000]; CSI, chlorosulfonyl isocyanate;
BBr3, boron tribromide; DCM, dichloride methylene; THF, tetrahydrofuran; DMF,
dimethylformamide; HPLC, high performance liquid chromatography; SPRL, sPLA2
responsive liposome; satd, saturated.
Another challenge associated with evaluating sPLA2-mediated
liposome degradation and drug release is the ability to quantify
⇑
Corresponding author. Tel.: +1 706 542 6813; fax: +1 706 542 5358.
E-mail address: rarnold@rx.uga.edu (R.D. Arnold).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.08.103

G. Zhu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6370–6375
6371
Table 1
Structures of existing and novel sPLA2-targeted ether lipids
a
Compound
R1 (linkage)
R2 (linkage)
R3
Tm (°C)
DSPC
DPPC
DSPE
DSPG
1
18:0 (Ester)
16:0 (Ester)
18:0 (Ester)
18:0 (Ester)
16:0 (Ether)
16:0 (Ether)
18:0 (Ester)
16:0 (Ester)
18:0 (Ester)
18:0 (Ester)
15:0 (Ester)
15:0 (Ester)
Choline
Choline
Ethanolamine
Glycerol
Choline
49.6
38.9
74.3
54.8
34.4
52.2
2
Methanol
(1) 1-O-C31PC; (2) 1-O-C31PM.
Figure 1. Structure of ether lipids.
a
Tm was determined using individual lipids hydrated in dd-water at 40 mM.
the degradation of lipids from in vivo samples. The majority of lip-
ids used to prepare drug-carriers cannot be separated from similar
lipids found endogenously. Therefore, we sought to synthesize and
use lipids containing odd FA acyl chain lengths, not commonly
found in vivo, thus, permitting direct assessment of liposome
tissue distribution and their degradation.
The aim of this study was to develop a simple and rapid ap-
proach to synthesize glycerophospholipids with sn-1 ether and
sn-2 ester linked FA (Fig. 1) that could be used to tune drug-release
and be identified in biological samples without interference from
endogenous lipids. Specifically we synthesized two ether lipids, 1
1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phosphocholine
(C31PC) and 2 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phos-
phomethanol (C31PM), and examined their ability to alter enzy-
matically triggered release of 6-carboxyfluorescein (6-CF) from
liposomes incubated in TRIS buffer or fetal bovine serum (FBS)
solutions, and evaluated their ability to be identified from lipids
found endogenously in plasma.
Using R-(O)-benzyl glycidol as a starting material, intermediate
6 (Scheme 1) was synthesized29 by a three-step reaction, (i) regio-
selective opening of epoxide ring with ether30, (ii) Steglich esterifi-
cation to its sn-2 position with fatty acid31 and (iii) deprotection
reaction of benzyl group using Lewis acid boron tribromide.25,32
Intermediate 6 was then used as the starting material for the
synthesis of both 1-O lipids 1 and 2 (Table 1) by conjugating either
phosphocholine or phosphomethanol (Scheme 2). H2/Pd-C has
been used successfully to deprotect the benzyl group, resulting in
high product yield. However, this method is limited in that it can
only be applied for saturated lipid synthesis. Therefore, we used
alternative methods to remove the benzyl group. First chlorosulfo-
nyl isocyanate (CSI) was tried,33 but the result was not desirable.
The failure of this reaction may have resulted from the low selec-
tivity of CSI for ether bonds or the instability of intermediate 6 in
the resulting harsh chemical environment, that is, use of sodium
hydroxide and the long reaction time (up to 21 h). The instability
of intermediate 6 is most likely the result of acyl migration that
is accelerated at elevated temperatures and the non-neutral pH
environment of this reaction. With this in mind, we adapted an
existing method using boron tribromide (BBr3).33 The reaction
was performed at À78 °C and was completed in 5 min.32 BBr3 dis-
played high selectivity over the benzyl ether group, resulting in an
almost complete (ꢀ100%) formation of 6 from 5. Furthermore, the
short reaction time has also been shown to reduce acyl migra-
tion.34 The primary drawback of this reaction is the use of large
quantities of DCM solvent, that is, 0.1 g of 5 needs 100 mL of
DCM to dissolve at À78 °C.
Synthesis of 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phos-
phocholine (1)35 and 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-
phosphomethanol (2)36 were completed using the conditions
developed by Hirth, G.37 Initial purification by chromatography
using a TMD-8 resin column and a silica gel column according to
the literature did not initially produce a pure product. However,
purity was achieved using a gradient method in flash chromatogra-
phy. It should be mentioned in the preparation of 1-O-hexadecyl-
2-pentadenoyl-sn-glycerol-3-phosphomethanol (2) satd NaHCO3
solution was used in place of water because of the resulting forma-
tion of the sodium salt form of 2 facilitated purification, compared
to its free acid form.
The effect of incorporation of 1 or 2 on the release of 6-CF from
prototypical SSL samples (0.05
lmol/mL) was determined in the
presence and absence of sPLA2, as previously described by us24
;
SSL formulations contained 1,2-distearoyl-sn-glycero-3-phosphati-
dylcholine (DSPC), cholesterol and 1,2-distearoyl-sn-glycero-3-
phosphoethanolamine-N-[poly(ethylene glycol)2000] (DSPC/CHOL/
PEG) in a 9:5:1 mole ratio. Incorporation of 10 or 30 lipid mol %
of C31PM resulted in enhanced sPLA2-mediate release of 6-CF com-
pared to SSL formulations (Fig. 2A). Preparations using 90% or 30
lipid mol % C31PC had greater and equal release to those incorpo-
rating 10 lipid mol % DSPG (Fig. 2B). DSPG is an anionic lipid that
is degraded rapidly by sPLA2, but results in liposome formulations
that display poor drug retention.24 Together, these data suggest
that shortening of FA acyl chains from C18:0/18:0 to C15:0/16:0,
using an ester/ether linked acyl-chains and controlling the per-
centage of C31PM or C31PC can be used to tune the release from
SSL-like formulations. This represents an advancement over the
use of other anionic lipids, such as DSPG, that are very sensitive
to sPLA2, but have bulky head groups that decrease their stability
R1O
a
R1OH
OBn
R1O
R2COO
OBn
b
OBn
R2COOH
HO
O
5
4
3
c
R1O
R2COO
OH
6
Scheme 1. Synthesis of intermediate 6. (a) R1OH, NaH, THF, DMF, 16 h, 80 °C (b) R2COOH, DCC, DMAP, DCM, 24 h, room temperature (c) BBr3, DCM, 5 min, À78 °C.

6372
G. Zhu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6370–6375
O
P
R1O
R2COO
O
O
N+
O-
a
1
R1O
R2COO
OH
O
P
b
R1O
R2COO
O
O
O-
2
Scheme 2. Synthesis of products 1 and 2. (a) (i) POCl3, Et3N, DCM, 1 h, room temperature; (ii) pyridine, choline tosylate, 16 h, room temperature; (iii) H2O,1 h (b) (i)
MeOPOCl2, TMP, 24 h, room temperature; (ii) satd NaHCO3, 2 h.
and an anionic charge that can reduce the circulation half-life of
liposomes.
30 mol % of C31PM to SSL formulations reduced the overall rate of
6-CF release compared to SSL formulations in the presence and ab-
sence of 10 mol % DSPE (Fig. 3A). This is not unexpected given the
higher phase transition (Tm 52.2 °C, Table 1) of hydrated C31PM
compared to DSPC (49.6 °C). Inclusion of DSPE (10 mol %) was
Serum can alter release from SSL and SPLR formulations.24 The
effect of 10% (v/v) fetal bovine serum (FBS) on sPLA2-mediated 6-
CF release was determined (Fig. 3). The addition of 10 or
Figure 2. sPLA2-mediated release of 6-CF from A. prototype SSL and C31PM (10 and
30 mol %) modified SSL and B. C31PC (90 or 30 mol %) and DSPG (10 mol %) modified
SSL formulations. Liposome samples were treated with 0 (control) or 2.5 lg/mL
sPLA2 for 0–48 h at 25 °C in TRIS buffer (pH 7.4) and 6-CF release was determined by
quantifying fluorescence at an excitation wavelength of 480 nm and an emission
intensity at 510 nm. Data are represented as the mean SEM of a least three
separate experiments, (n = 5/study).
Figure 3. Effect of 10% fetal bovine serum on sPLA2-mediated 6-CF release from A.
SSL, and DSPE (10 mol %) and C31PM (10 and 30 mol %) modified SSL formulations
and B. C31PC (10, 30, and 90 mol %) and DSPG (10 mol %) modified SSL. The effect of
serum on sPLA2-mediated release of 6-CF from formulations was determined over
216 h at 37 °C by quantifying fluorescence at an excitation wavelength of 480 nm
and an emission intensity at 510 nm. Data are represented as the mean SEM of a
least three separate experiments.

G. Zhu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6370–6375
6373
similar to the C31PM (10 mol %). DSPE had the greatest phase tran-
sition temperature (74 °C, Table 1), that would be expected to im-
prove membrane stability, but also has a bulkier head group
relative to the phosphomethanol head group on C31PM. Further,
lipids such as DSPE and DSPG. SSL formulations containing
10 mol % DSPG or 10 mol % C31PC had similar release profiles (Fig
3B). Whereas, incorporation of C31PC at 30 and 90 mol % resulted
in a stepwise decrease in 6-CF release, relative to 10 mol % DSPG
(Fig 3B). Although the mechanisms underlying these effects are
not fully known, DSPG is an anionic lipid with a phase transition
C31PM is anionic, its relatively small head group may be protected
from sPLA2-mediated degradation, relative to other bulkier anionic
Figure 4. Mass spectra of blank rat plasma A. or plasma spiked DSPC B., P31PC C. or SSL formulation containing DSPC and C31PC in a 9:1 mole ratio D. DSPC (791?608 m/z) and
31PC (707?523 m/z) were identified based on their ion-pair as determined using MRM.
C

6374
G. Zhu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6370–6375
of 54.8 °C, Table 1), whereas C31PC is zwitterionic in nature. Most
importantly is that the shorter acyl-chain FA used in 1 and 2
showed concentration dependent alterations in 6-CF release,
suggesting that the rate and extent of release may be modulated.
These studies suggest that this approach can be used to synthe-
size lipids and modulate the release of intra-luminal constituents
from liposomes. Although each formulation will have to be tailored
to the physicochemical properties of a target drug(s), this approach
provides a platform for making lipid modifications based on drug-
carrier release kinetics.
Another challenge optimizing lipid based drug-carriers is our
ability to track their disposition in vivo due to the presence of
endogenous lipids. To overcome this challenge we synthesized
odd chain lipids, not normally found in nature. Although a variety
of fluorescent and radio-labeled probes exist, their effect on mem-
brane fluidity and particulate disposition is not well known and
these probes are generally not approved for human use. This is
one reason why a goal of this work was to prepare lipids that were
multifunctional, that is, have the ability to tune release and be
identified in biological samples.
a King Khalid University Fellowship to Y.A., and an American Foun-
dation for Pharmaceutical Education–Wyeth New Investigator
Grant to R.D.A.
References and notes
1. Drummond, D. C.; Meyer, O.; Hong, K. L.; Kirpotin, D. B.; Papahadjopoulos, D.
Pharmacol. Rev. 1999, 51, 691.
2. Bondurant, B.; Mueller, A.; O’Brien, D. F. Biochim. Biophys. Acta 2001, 1511, 113.
3. Shum, P.; Kim, J. M.; Thompson, D. H. Adv. Drug Delivery Rev. 2001, 53, 273.
4. Spratt, T.; Bondurant, B.; O’Brien, D. F. Biochim. Biophys. Acta 2003, 1611, 35.
5. Yatvin, M. B.; Weinstein, J. N.; Dennis, W. H.; Blumenthal, R. Science 1978, 202,
1290.
6. Kong, G.; Anyarambhatla, G.; Petros, W. P.; Braun, R. D.; Colvin, O. M.;
Needham, D.; Dewhirst, M. W. Cancer Res. 2000, 60, 6950.
7. Needham, D.; Anyarambhatla, G.; Kong, G.; Dewhirst, M. W. Cancer Res. 2000,
60, 1197.
8. Huang, S. L.; MacDonald, R. C. Biochim. Biophys. Acta 2004, 1665, 134.
9. Schroeder, A.; Avnir, Y.; Weisman, S.; Najajreh, Y.; Gabizon, A.; Talmon, Y.; Kost,
J.; Barenholz, Y. Langmuir 2007, 23, 4019.
10. Cummings, B. S.; McHowat, J.; Schnellmann, R. G. J. Pharmacol. Exp. Ther. 2000,
294, 793.
11. Dong, Q.; Patel, M.; Scott, K. F.; Graham, G. G.; Russell, P. J.; Sved, P. Cancer Lett.
2006, 240, 9.
12. Kallajoki, M.; Alanen, K. A.; Nevalainen, M.; Nevalainen, T. J. Prostate 1998, 35,
263.
SSL formulations in the presence and absence of
C31PC
13. Jiang, J.; Neubauer, B. L.; Graff, J. R.; Chedid, M.; Thomas, J. E.; Roehm, N. W.;
Zhang, S.; Eckert, G. J.; Koch, M. O.; Eble, J. N.; Cheng, L. Am. J. Pathol. 2002, 160,
667.
14. Graff, J. R.; Konicek, B. W.; Deddens, J. A.; Chedid, M.; Hurst, B. M.; Colligan, B.;
Neubauer, B. L.; Carter, H. W.; Carter, J. H. Clin. Cancer Res. 2001, 7, 3857.
15. Yamashita, S.; Yamashita, J.; Ogawa, M. Br. J. Cancer 1994, 69, 1166.
16. Yamashita, S.; Ogawa, M.; Sakamoto, K.; Abe, T.; Arakawa, H.; Yamashita, J. Clin.
Chim. Acta 1994, 228, 91.
17. Yamashita, S.; Yamashita, J.; Sakamoto, K.; Inada, K.; Nakashima, Y.; Murata, K.;
Saishoji, T.; Nomura, K.; Ogawa, M. Cancer 1993, 71, 3058.
18. Kiyohara, H.; Egami, H.; Kako, H.; Shibata, Y.; Murata, K.; Ohshima, S.; Sei, K.;
Suko, S.; Kurano, R.; Ogawa, M. Int. J. Pancreatol. 1993, 13, 49.
19. Kuopio, T.; Ekfors, T. O.; Nikkanen, V.; Nevalainen, T. J. APMIS 1995, 103, 69.
20. Oka, Y.; Ogawa, M.; Matsuda, Y.; Murata, A.; Nishijima, J.; Miyauchi, K.; Uda, K.;
Yasuda, T.; Mori, T. Enzyme 1990, 43, 80.
21. Jensen, S. S.; Andresen, T. L.; Davidsen, J.; Hoyrup, P.; Shnyder, S. D.; Bibby, M.
C.; Gill, J. H.; Jorgensen, K. Mol. Cancer Ther. 2004, 3, 1451.
22. Davidsen, J.; Vermehren, C.; Frokjaer, S.; Mouritsen, O. G.; Jorgensen, K. Int. J.
Pharm. 2001, 214, 67.
(10 mol %) were spiked into rat plasma, extracted using a modified
Bligh–Dyer assay38, and lipids were identified via their unique ion-
pairs by LC/MS/MS using mixed reaction monitoring (MRM) mode
as previously described by us.24 As expected, analysis of blank plas-
ma demonstrated background levels of ion-pairs corresponding to
DSPC (791?608 m/z) and C31PC (707?523 m/z) (Fig. 4). Spiking of
plasma with DSPC (Fig. 4B), C31PC (Fig. 4C) and SSL formulations
containing DSPC and C31PC in a 9:1 mole ratio (Fig. 4D) resulted
in an increased intensity for peaks corresponding to both ion-pairs.
These data suggest that odd-chain lipids, for example, C31PC, can
be identified in rat plasma without interference of endogenous lip-
ids. Further, the presence of a C15:0 FA on C31PC may facilitate the
tracking of its degradation and metabolism in vivo. Currently,
determining liposome degradation in vivo is difficult without the
use of tracers or radiolabels due to the high endogenous levels of
even-chained lipids and their fatty acid metabolites found in bio-
logical tissues.
23. Linderoth, L.; Peters, G. H.; Jorgensen, K.; Madsen, R.; Andresen, T. L. Chem.
Phys. Lipids 2007, 146, 54.
24. Zhu, G.; Mock, J. N.; Aljuffali, I.; Cummings, B. S.; Arnold, R. D. J. Pharm. Sci.
2011, 100, 3146.
In conclusion, a simple approach is presented that permits syn-
thesis of a variety of ether phospholipids from a few intermediates
while decreasing acyl migration. We demonstrated the synthesis of
two odd-chain ether lipids with a PC or PM head-group that could
be used to alter drug release. We also demonstrated that the odd
chain lipid was able to be identified after extraction from complex
biological samples, for example, plasma. These data suggest that
this approach may be used to make slight modifications to existing
lipids or create novel lipids to tune the release of therapeutic
agents from lipid based particulate carriers and track their lipids
in vivo. Further, we believe this approach can be extended to use
incorporation of other head groups with functional groups suitable
for attaching targeting species (e.g., antibodies or peptides) or
imaging probes.
25. Andresen, T. L.; Davidsen, J.; Begtrup, M.; Mouritsen, O. G.; Jorgensen, K. J. Med.
Chem. 2004, 47, 1694.
26. Andresen, T. L.; Jensen, S. S.; Kaasgaard, T.; Jorgensen, K. Curr. Drug Deliv. 2005,
2, 353.
27. Jensen, S. S.; Andresen, T. L.; Davidsen, J.; Hoyrup, P.; Shnyder, S. D.; Bibby, M.
C.; Gill, J. H.; Jorgensen, K. Mol. Cancer Ther. 2004, 3, 1451.
28. Drummond, D. C.; Noble, C. O.; Hayes, M. E.; Park, J. W.; Kirpotin, D. B. J. Pharm.
Sci. 2008, 97, 4696.
29. A syringe-septum technique was used for moisture-sensitive reactions. THF
was prepared freshly by distilling with sodium and benzophenone ketyl. Other
solvents like DFM, DCM3, Et3N, and pyridine were mixed with 3 Å molecular
sieves before use. Reagents were purchased from Sigma–Aldrich Chemical Co.
and used without further purification. Silica gel (35–70 lm, 200–430 mesh)
was used for column chromatography. Products were visualized on TLC using
either iodine vapor or Hanessian’s stain method. Low resolusion MS was
obtained on
a HPLC-LC/MSD trap XCT Ultra Plus system (Agilent), high
resolution MS was obtained on a LCT Premier Orthogonal Acceleration TOF
Mass Spectrometer and 400 MHz 1H NMR spectra were used for chemical
identifications. A Mettler Toledo DSC 1 Star System was used to determine
melting temperature of DSPC, DPPC, and products 1 and 2. DSPC, DSPG, DSPE,
and DSPE–PEG were purchased from Avanti Polar Lipids Inc. (Alabaster, USA).
sPLA2 was purchased from Cayman Chemical Company (Ann Arbor, MI) and
Genway Biotech Inc (San Diego, CA). F-12k cell culture media and FBS were
purchased from Hyclone (Rockford, Illinois). CHOL was purchased from Sigma–
Aldrich (St. Louis, Missouri). Acetonitrile and methanol were of HPLC grade
from Fisher Scientific (Pittsburgh, PA). All other chemicals and solvents were of
analytical grade, obtained from commercial sources and used without further
Conflicts of Interest
No conflicts of interest
Acknowledgments
purification. All experiments used ultrapure water (>3 M
Millipore Milli-Q synthesis system (Billerica, MA).
X) obtained from a
We thank Jeremy L Grove, Jennifer A Haley, and Dr. Yang Geng
for their technical assistance related to the synthesis. We also
thank Dr. Timothy Long for helpful discussions. This research was
funded in part by Georgia Cancer Coalition Distinguished Scholar
Grants to R.D.A./B.S.C., an NIH NIBIB (EB08153) to R.D.A./B.S.C., a
University of Georgia Graduate Fellowship Stipend Award to G.Z.,
30. Synthesis of
4 was initiated by washing a dispersion of NaH (0.0765 g,
3.20 mmol) in (60% mineral oil) using a 50 mL N2 protected flask with dry
petroleum ether (20 mL Â 3).28 Cetyl alcohol (0.74 g, 3.05 mmol) was added
followed by dry THF (10 mL) at 0 °C. The reaction mixture was refluxed at 80 °C
for 1 h. (R)-O-benzyl glycidol (3) (0.25 g, 1.53 mmol) was added followed by
addition of DMF (25 mL) drop-wise over 5 min. The reaction mixture was

G. Zhu et al. / Bioorg. Med. Chem. Lett. 21 (2011) 6370–6375
stirred overnight at 80 °C. The mixture was cooled to room temperature and (quintet, 2H, b-CH2),
6375
1.26 (br, s, 48H, 24 Â CH2),
r
1.56 (quintet, 2H, b-CH2),
r
r
the reaction stopped by addition of water (2.5 mL) and solvents removed under
vacuum. The residue was dissolved with diethyl ether, washed in brine three
times and dried over Na2SO4. The product was purified using two rounds of
column chromatography using ether/DCM (1:4) then ether/hexane (1:1). 1-O-
Hexadecyl-3-O-benzyl-sn-glycerol (4) has a Rf 0.31 (ether/hexane 1:4), 1H
0.886 (t, 6H, CH3 Â 2), and ESI-MS+ of 541.3 m/z (MW = 540.9).
33. Kim, J. D.; Han, G.; Zee, O. P.; Jung, Y. H. Tetrahedron Lett. 2003, 44, 733.
34. Pluckthun, A.; Dennis, E. A. Biochemistry 1982, 21, 1743.
35. To a flame-dried, N2 protected 50 mL flask, 18
l
L POCl3 (0.19 mmol) and
L Et3N (0.19 mmol) and 6
1.5 mL DCM were added. Then 3 mL DCM with 27
l
NMR(400 MHz, CDCl3):
(quintet, 1H, CH), 3.52 (m, 6H, CH2CHCH2, O–CH2),
(quintet, 2H, b-CH2),
1.26 (br s, 26H, 13 Â CH2),
MS+ of 407.1 m/z (MW = 406.3).
r
7.35 (m, 5H, Ph),
r
4.57(s, 2H, CH2Ph),
r
r
4.00
1.57
(0.16 mg, 0.14 mmol) were added to the flask drop-wise over 20 min. This
mixture was stirred for 1 h at room temperature. After this, 0.2 mL pyridine
(1.2 mmol) and 128 mg choline tosylate (0.223 mmol) were added. The
reaction was allowed to stand for 24 h at room temperature. 0.2 ml water
was added to quench the reaction. The reaction mixture was concentrated and
then passed through a TMD-8 resin column using THF/H2O (9:1) as mobile
phase. The resulting crude product was purified on a silica gel column using
three mobile phases in a row (100 mL DCM/MeOH = 85:15, 200 mL MeOH and
then 100 mL DCM/MeOH/H2O, 65:25:4). Product yield (from 5 to 1, two steps)
was 42%. 1-O-hexadecyl-2-pentadenoyl-sn-glycerol-3-phosphocholine (1) has
r
r
1.73(s, OH),
r
r 0.887 (t, 3H, CH3), and ESI-
31. Three grams of 4 (7.4 mmol), 2.7 g pentadecanoic acid (11.1 mmol), 0.17 g
DMAP (1.48 mmol), 150 mL DCM and 3.1 g DCC (14.8 mmol) were added to a
flame-dried, N2 protected 250 mL flask. The reaction was cooled to 0 °C for
15 min and allowed to stand at room temperature for 24 h. The reaction was
stopped by adding 5 mL acetic acid and stirred for 0.5 h. The sample was stored
at a À20 °C overnight and the resulting precipitate removed by filtering. The
solvent was evaporated and the resulting residues were dissolved in EtOAC,
washed with 1 M HCl (30 mL Â 3), satd NaHCO3 (30 mL Â 3), brine (30 mL Â 3)
a
Rf 0.35 (DCM/MeOH/H2O, 65:25:4), 1H NMR (400 MHz, CDCl3):
r
5.13
(quintet, 1H, CH),
3.88 (s, 2H, POCH2CH2N+),
9H, N+(CH3)3),
(quintet, 2H, b-CH2),
HRMS via TOF ES MS+ 706.5803 m/z (MW = 706.0).
r
4.36 (s, 2H, POCH2CH2N+),
r
3.98 (m, 2H, CH2CHCH2), r
and dried over Na2SO4. The intermediate
chromatography using petroleum ether/DCM (1:4) as
5
was purified using column
mobile phase.
r
3.55 (m, 4H, CH2CHCH2, OCH2C15H31),
r
3.44 (s,
2.32 (t, 2H, CH2COO), 1.53
1.26 (br, s, 48H, 24 Â CH2), r 0.887 (t, 6H, CH3 Â 2), and
a
r
r 1.60 (quintet, 2H,b-CH2), r
Product yield was 69%. 1-O-Hexadecyl-2-pentadenoyl-3-O-benzyl-glycerol
(5) has a Rf 0.25 (petroleum ether/DCM 1:4), a 631.3 m/z (MW = 630.5) by
r
ESI-MS and 1H NMR (400 MHz, CDCl3):
r
7.33 (m, 5H, Ph),
3.63 (d, 2H, CH2CHCH2),
3.42 (m, 2H, OCH2C15H31), 2.34 (t, 2H, CH2COO),
(quintet, 2H, b-CH2), 1.53 (quintet, 2H, b-CH2),
1.26 (br s, 48H, 24 Â CH2),
0.887 (t, 6H, CH3 Â 2), and ESI-MS+ of 631.3 m/z (MW = 630.5).
r
5.18 (quintet, 1H,
3.58 (d, 2H,
1.62
36. To a solution of MOPOCL (3.4 mmol) and TMP (1.82 mmol) in dried toluene
(2 mL) under N2 protected at À20 °C, 0.4 g of 6 (0.74 mmol) in toluene (10 mL)
was added drop wise. This mixture was stirred for 24 h at room temperature.
After this, 2 mL satd NaHCO3 was added and stirred for 2 h. This reaction
mixture was concentrated by azeotropic distillation (with ethanol and
toluene). The resulting crude product was purified on a silica gel column
(DCM/MeOH, 4:1). Product yield (from 5 to 2, two steps) was 40%. 1-O-
hexadecyl-2-pentadenoyl-sn-glycerol-phosphomethanol (2) has a Rf 0.4 (DCM/
CH),
r
4.54–4.56 (AB, 2H, CH2Ph),
r
r
CH2CHCH2),
r
r
r
r
r
r
32. Intermediate 5 (0.2 g, 0.32 mmol) was added to a flame-dried, N2 protected
250 mL flask with 200 mL DCM. The flask was bathed into a mixture of dry ice
and acetone, which produced a temperature of –78 °C. BBr3 (1 M BBr3 in DCM,
0.64 mmol) was injected into the flask and allowed to stand for 5 min. The
reaction was quenched with 20 mL satd. NaHCO3 and 20 mL diethyl ether. The
solvent was then washed with satd NaHCO3 (30 mL Â 3), water (30 mL Â 3),
brine (30 mL Â 3) and dried over Na2SO4. Product 6 was concentrated under
MeOH = 4:1), 1H NMR (400 MHz, CDCl3):
r
5.18 (quintet, 1H, CH),
3.61–3.57 (m, 5H, CH2CHCH2, POCH3), 3.40 (m, 2H,
2.32 (t, 2H, CH2COO), 1.60 (quintet, 2H, b-CH2), 1.52
0.887 (t, 6H, CH3 Â 2), and
r 4.00 (d, 2H,
CH2CHCH2),
r
r
OCH2C15H31),
r
r
r
(quintet, 2H, b-CH2),
r 1.26 (br, s, 48H, 24 Â CH2), r
vacuum yielding
glycerol (6) had
a
crude white solid. 1-O-hexadecyl-2-pentadenoyl-sn-
TOF MS+, 657.5034 m/z, M+Na+ adduct (MW = 633.5).
37. Hirth, G.; Barner, R. Helv. Chim. Acta 1982, 65, 1059.
38. Bligh, E. G.; Dyer, W. J. Can. J. Biochem. Physiol. 1959, 37, 911.
a
Rf 0.71 (DCM/diethyl ether, 10:1), 1H NMR (400 MHz,
CDCl3):
r
5.00 (quintet, 1H, CH),
r
3.81 (d, 2H, CH2CHCH2),
r
3.63 (m, 2H,
1.63
CH2CHCH2),
r
3.45 (m, 2H, OCH2C15H31),
r
2.36 (t, 2H, CH2COO), r