Synthesis and cytotoxic activity of two novel 1-dodecylthio-2-decyloxypropyl-3-phosphatidic acid conjugates with gemcitabine and cytosine arabinoside

Article abstract of DOI:10.1021/jm020571x

Cytosine arabinoside (ara-C) and gemcitabine (dFdC) are two standard chemotherapy drugs used in the treatment of patients with various cancers. To alter the pharmacokinetic and pharmacodynamic properties of these molecules, we conjugated a synthetic phosp

Full text of DOI:10.1021/jm020571x

J . Med. Chem. 2003, 46, 4205-4208
4205
Brief Articles
Syn th esis a n d Cytotoxic Activity of Tw o Novel
1-Dod ecylth io-2-d ecyloxyp r op yl-3-p h osp h a tid ic Acid Con ju ga tes w ith
Gem cita bin e a n d Cytosin e Ar a bin osid e
Richard L. Alexander,† Susan L. Morris-Natschke,‡ Khalid S. Ishaq,‡ Ronald A. Fleming,§ and
Gregory L. Kucera*,#
Department of Physiology and Pharmacology, Wake Forest University School of Medicine, Medical Center Boulevard,
Winston-Salem, North Carolina 27517, Department of Medicinal Chemistry and Natural Products, School of Pharmacy,
University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, Kucera Pharmaceutical Company,
111 N. Chestnut Street, Suite 104, Victoria Hall, Winston-Salem, North Carolina 27101, and Department of Internal Medicine,
Section on Hematology/ Oncology, Wake Forest University School of Medicine, Medical Center Boulevard,
Winston-Salem, North Carolina 27517
Received December 18, 2002
Cytosine arabinoside (ara-C) and gemcitabine (dFdC) are two standard chemotherapy drugs
used in the treatment of patients with various cancers. To alter the pharmacokinetic and
pharmacodynamic properties of these molecules, we conjugated a synthetic phospholipid to
both ara-C and dFdC and investigated their chemotherapeutic potential. The dFdC conjugate
had greater cytotoxic activity compared with the ara-C conjugate and demonstrated notable
cytotoxicity against all human cell lines tested.
In tr od u ction
demonstrated oral bioavailability,13 and previous struc-
ture-activity relationship studies showed that the
phospholipid molecule chosen had a favorable toxicity
profile in vitro.14
The deoxycytidine analogue cytosine arabinoside (ara-
C) is effective against several types of hematological
cancers such as leukemia and lymphoma1 but is not as
effective against solid tumors.2 Once inside the cell,
ara-C is phosphorylated by deoxycytidine kinase to yield
ara-C monophosphate (ara-CMP), which is further
phosphorylated to the active form, ara-C triphosphate
(ara-CTP),3 which is a potent inhibitor of DNA synthesis
causing DNA chain termination and DNA strand breaks.4
In comparison, gemcitabine (dFdC) is a deoxycytidine
analogue5 that shows activity against leukemias6 and
various solid tumors.7,8 Like ara-C, gemcitabine is also
phosphorylated by deoxycytidine kinase to the mono-
phosphate dFdC-MP.9 The drug is further phosphor-
ylated to the active form dFdC triphosphate (dFdC-TP),
which can be incorporated into DNA and inhibit DNA
polymerase, ultimately promoting apoptosis.10
To increase the uptake and efficacy of ara-C and
dFdC, we chose to conjugate these drugs to a synthetic
C-1 thioether, C-2 oxyether phospholipid, which would
not be subject to metabolism at the C-1 and C-2 alkyl
side chains in the GI tract. To our knowledge, this is
the first report of the synthesis of a phospholipid
conjugated to gemcitabine. The resulting conjugates
have shown promising results in vitro in MTS assays.
Furthermore, on the basis of our data, we believe the
lipid-dFdC conjugate is a drug that is superior to the
lipid-ara-C conjugate and should be further investi-
gated.
Ch em istr y
To increase the efficacy of ara-C, Ryu et al. linked
ara-C with natural phospholipids.11 These compounds
demonstrated promising in vitro and in vivo results;
however, they were subject to degradation in the GI
tract by phospholipase A2 and other lipases that degrade
ester-containing lipid molecules.12 However, previous
animal model experiments using 1-dodecylthio-2-decyl-
oxypropyl-3-phosphatidic acid as a carrier for AZT
The phospholipid was synthesized as previously pub-
lished (see Supporting Information for Scheme 1).14,15
Briefly, 3-mercapto-1,2-propanediol (1) was reacted with
bromododecane in ethanol/KOH to yield 1-S-dodecyl-2,3-
glycerol (2). The primary OH was protected by reacting
2 with triphenylmethyl chloride in pyridine. Next, the
product 3 was refluxed with bromodecane and sodium
hydride in tetrahydrofuran (THF) to give 4. The tri-
phenylmethyl protecting group was removed using
p-toluenesulfonic acid in CHCl3/MeOH, and the result-
ing dialkylglycerol 5 was reacted with phosphorus
oxychloride in pyridine and THF to yield 1-S-dodecyl-
2-O-decylthioglycero-3-phosphatidic acid (6).
* To whom correspondence should be addressed. Phone: (336) 716-
6348. Fax: (336) 716-0255. E-mail: gkucera@wfubmc.edu.
† Department of Physiology and Pharmacology, Wake Forest Uni-
versity School of Medicine.
‡ University of North Carolina at Chapel Hill.
§ Kucera Pharmaceutical Company.
To direct conjugation to the 5′-OH, the other reactive
groups were chemically protected (see Supporting In-
#
Department of Internal Medicine, Wake Forest University School
of Medicine.
10.1021/jm020571x CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/09/2003

4206 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
Brief Articles
Ta ble 1. Effects of dFdC and ara-C Conjugates on Different Cell Lines
IC50 ( SD, n ) 3 (µM)
ara-C conjugate
67.3 ( 26
IC50 ( SD, n ) 3 (µM)
ara-C
HL-60 (human, promyelocytic leukemia)
0.76 ( 0.16
dFdC
dFdC conjugate
HL-60 (human, promyelocytic leukemia)
CEM-SS (human T-4 lymphoblastoid clone)
HeLa (human, epitheliod carcinoma, cervix)
HT-29 (human, colon, adenocarcinoma)
MCF7/WT-2′ (human, breast carcinoma)
U373-MG (human, glioblastoma)
BxPC3 (human, pancreas, adenocarcinoma)
SK-LU-1 (human, lung, adenocarcinoma)
SCC-25 (human, squamous cell carcinoma, tongue)
BG-1 (human, breast, adenocarcinoma)
0.33 ( 0.07
0.03 ( 0.02
0.008 ( 0.0012
0.043 ( 0.015
1.66 ( 1.1
0.13 ( 0.03
0.16 ( 0.02
0.055 ( 0.01
0.047 ( 0.026
0.68 ( 0.37
1.8 ( 0.57
0.6 ( 0.13
0.69 ( 0.33
0.75 ( 0.07
1.73 ( 1.1
0.29 ( 0.043
1.59 ( 0.41
1.12 ( 0.17
0.59 ( 0.39
1.84 ( 0.73
that the dFdC conjugate is able to cross the lipid
membrane of the cell better than the ara-C conjugate.
These results would also suggest a reason for why the
dFdC conjugate is more cytotoxic than the ara-C con-
jugate. The specific experimental conditions can be
found in the Supporting Information.
Using an MTS assay, we confirmed that the dFdC
conjugate was more cytotoxic than the ara-C conjugate
in all cell lines tested. The resulting IC50 values can be
seen in Table 1. The ara-C conjugate did show cytotox-
icity in HL-60 cells; however, it was much less cytotoxic
in the other cell lines tested. In contrast, the dFdC
conjugate was cytotoxic in every cell line tested. There-
fore, we believe that the dFdC conjugate warrants
further study into its metabolism and mechanism of
action.
We also used electrospray ionization mass spectros-
copy to investigate the metabolism of these two conju-
gate molecules. When the conjugates were incubated in
media over 72 h, we observed <4% reduction of the
parent conjugate. However, when the ara-C conjugate
was incubated in the presence of HL-60 cells (50 µM)
for 72 h, there was a slight reduction in the parent drug
(10-20%). When the dFdC conjugate was incubated in
the presence of HL-60 cells (10 µM) for 72 h, there was
a statistically significant (p < 0.05, one-way ANOVA,
with Dunnett’s multiple comparison post test) reduction
in the parent compound (50%). The specific experimen-
tal conditions can be found in the Supporting Informa-
tion.
F igu r e 1. Chemical structures of the two compounds syn-
thesized: (A) ara-C conjugate; (B) dFdC conjugate.
formation for Scheme 2). The ara-C 7 was refluxed with
benzoic anhydride in ethanol to form the amide deriva-
tive 8.16 The 5′-OH was protected using tert-butyldi-
methylsilyl chloride (TBDMS) in dimethyl formamide
(DMF) to give 9.17 Next, the 3′-OH and 2′-OH were
protected using acetic anhydride in pyridine.18 Finally,
the TBDMS group of 10 was removed using tetrabutyl-
ammonium fluoride in THF.17
The protected ara-C 11 was reacted with the phos-
phatidic acid 6 using 2,4,6-triisopropylbenzenesulfonyl
chloride in pyridine at 40-50 °C to generate 12. The
amide and ester protecting groups were removed using
2 N NH3-MeOH to yield the final lipid-ara-C conjugate
molecule 13.18 The dFdC conjugate was synthesized in
the same manner as described above. Both conjugate
molecules can be seen in Figure 1. For synthetic
schemes and NMR chemical shifts, please see Support-
ing Information.
Resu lts a n d Discu ssion
Con clu sion
The conjugates were tested for their solubility in
aqueous solution. When sonicated and subjected to
centrifugation for 20 min in a microfuge, approximately
50% of the ara-C conjugate formed lipid aggregates that
were soluble in phosphate-buffered saline (PBS). On the
other hand, the dFdC conjugate was about 80% soluble
in the PBS under similar conditions. When the two
conjugates were subjected to a 1-octanol and PBS
mixture, the ara-C conjugate was found almost equally
in both layers (partition coefficient of 1.17). Conversely,
the dFdC conjugate favored the octanol layer with a
partition coefficient of 3.47. Taken together, these
results suggest that the dFdC conjugate forms a lipid
aggregate in PBS that remains in solution better than
the ara-C conjugate. Furthermore, the dFdC conjugate
favors the octanol layer, which supports our hypothesis
The synthetic ara-C and dFdC lipid conjugates show
cytotoxic activity in several different tumor cell lines.
The dFdC conjugate showed better activity than the
ara-C conjugate across all cell lines tested (ara-C data
not shown). These results are supported by the fact that
the dFdC conjugate more easily formed soluble lipid
aggregates in PBS than the ara-C conjugate. Addition-
ally, the dFdC conjugate had a greater partition coef-
ficient in octanol than the ara-C conjugate. These results
suggest that the dFdC conjugate may be superior to the
ara-C conjugate because it can more readily pass
through the lipid membrane of the cell.
The results of the metabolism studies suggest that
the presence of cells is necessary for the metabolism/
activation of the compounds. Furthermore, the ara-C

Brief Articles
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19 4207
molecule is incorporated very poorly into cells as shown
by the slight reduction in the parent ara-C conjugate
over time. Conversely, the dFdC conjugate was more
readily taken up by the cells and metabolized. These
results may also indicate why the dFdC conjugate is
superior to the ara-C conjugate.
and Dr. Kerry A. Pickin Paumi. We also thank Dr.
George DuBay from the Duke University Center for
Mass Spectrometry who performed the high-resolution
mass spectral analyses. This work was supported in part
by the North Carolina Biotechnology Center and Kucera
Pharmaceutical Company.
We also have hypothesized that the conjugate mol-
ecules take longer to enter the cell and be cleaved by a
phospholipase C-like enzyme.19 Although the ara-C and
dFdC molecules are transported across the cell mem-
brane via a transporter, we believe that the conjugates
cross the membrane by passive diffusion because of the
increased lipophilicity of the conjugate molecules com-
pared to the nucleosides alone. On the basis of our initial
data, the lipid-dFdC conjugate has altered pharmaco-
kinetic properties compared with the parent nucleoside,
may have possible clinical efficacy, and warrants further
investigation.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and synthetic schemes. This material is available free
of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Mastrianni, D. M.; Tung, N. M.; Tenen, D. G. Acute myelogenous
leukemia: current treatment and future directions. Am. J . Med.
1992, 92, 286-295.
(2) Chabner, B. A.; Myers, C. E. Clinical pharmacology of cancer
chemotherapy. In Cancer: Principles and Practice of Oncology;
DeVita, V. T., J r., Hellman, S., Rosenberg, S. A., Eds.; Lippin-
cott: Philadelphia, PA, 1989; pp 349-395.
(3) Kessel, D.; Hall, T. C.; Wodinsky, I. Transport and phosphoryl-
ation as factors in the antitumor action of cytosine arabinoside.
Science 1967, 156, 1240-1241.
Exp er im en ta l Section
(4) Fram, R. J .; Kufe, D. W. DNA strand breaks caused by inhibitors
of DNA synthesis: 1-beta-D-arabinofuranosylcytosine and aphid-
icolin. Cancer Res. 1982, 42, 4050-4053.
(5) Plunkett, W.; Huang, P.; Xu, Y. Z.; Heinemann, V.; Grunewald,
R.; Gandhi, V. Gemcitabine: metabolism, mechanisms of action,
and self-potentiation. Semin. Oncol. 1995, 22, 3-10.
(6) Grunewald, R.; Kantarjian, H.; Du, M.; Faucher, K.; Tarassoff,
P.; Plunkett, W. Gemcitabine in leukemia: a phase I clinical,
plasma, and cellular pharmacology study. J . Clin. Oncol. 1992,
10, 406-413.
(7) Lund, B.; Hansen, O. P.; Theilade, K.; Hansen, M.; Neijt, J . P.
Phase II study of gemcitabine (2′,2′-difluorodeoxycytidine) in
previously treated ovarian cancer patients. J . Natl. Cancer Inst.
1994, 86, 1530-1533.
(8) Casper, E. S.; Green, M. R.; Kelsen, D. P.; Heelan, R. T.; Brown,
T. D.; Flombaum, C. D.; Trochanowski, B.; Tarassoff, P. G. Phase
II trial of gemcitabine (2,2′-difluorodeoxycytidine) in patients
with adenocarcinoma of the pancreas. Invest. New Drugs 1994,
12, 29-34.
Proton NMR spectra were recorded on a Bruker-Spectrospin
300 MHz spectrometer. The chemical shifts are reported in
parts per million from internal TMS. Melting points were
determined on a Laboratory Devices Mel-Temp II melting
point apparatus, and mass spectral analysis was performed
on an Agilent 1100 series LC/MSD trap. High-resolution mass
spectral analysis was performed on a J EOL, J MS-SX102A
mass spectrometer. Silica gel 60 plates (Merck F254 or What-
man) were used for thin-layer chromatography. Purity of all
samples was ascertained by thin-layer chromatography in the
indicated solvent systems; compounds were visualized by UV
light, iodine, H2SO4 spray, or phosphomolybdate spray. Col-
umn chromatography was performed with silica gel (70-230
mesh). All chemicals and solvents were purchased from Sigma-
Aldrich and Fisher Scientific except dFdC, which was pur-
chased from the Wake Forest University Baptist Medical
Center Pharmacy. NMR data of the lipid intermediates were
consistent with those previously reported.14,15 All reactions
were carried out under N2 and extremely dry conditions.
Aqueous solubility and partition coefficient measurements
were tested as previously described.20
(9) Bouffard, D. Y.; Laliberte, J .; Momparler, R. L. Kinetic studies
on 2′,2′-difluorodeoxycytidine (Gemcitabine) with purified human
deoxycytidine kinase and cytidine deaminase. Biochem. Phar-
macol. 1993, 45, 1857-1861.
(10) Huang, P.; Plunkett, W. Induction of apoptosis by gemcitabine.
Semin. Oncol. 1995, 22, 19-25.
In Vitr o Cytotoxicity Stu d ies. The CEM-SS, HL-60, and
BG-1 cells were maintained in RPMI-1640 medium supple-
mented with 10% FBS. HeLa cells were maintained in DMEM
medium supplemented with 10% FBS. U373-MG and SK-LU
were maintained in MEM medium supplemented with 10%
FBS. BxPC3 were maintained in RPMI-1640 supplemented
with 10% FBS, 10 mM HEPES, 1.5 g/L NaHCO3, and sodium
pyruvate. HT-20 cells were maintained in McCoy’s 5A Medium
Modified supplemented with 10% FBS. MCF7/WT-2′ cells were
maintained in D-MEM/F-12 with 10% FBS and 10 µg/mL
insulin. SCC-25 cells were maintained in DMEM with 10%
FBS and 400 ng/mL hydrocortisone. All cells were maintained
in log phase and kept in a humidified atmosphere of 5% CO2/
95% air at 37 °C. All media also contained penicillin-
streptomycin (1%). The CellTiter 96 aqueous nonreactive cell
proliferation assay (Promega Life Sciences, Madison, WI) was
used according to the manufacturer’s details, and the results
were read on a precision microplate reader (Molecular Devices,
Sunnyvale, CA). The optical densities of the drug-treated cells
were compared to those of untreated control cells, and the data
were plotted in GraphPad Prizm. Nonlinear regression analy-
sis was used to determine IC50 values. For a more detailed
explanation of the biological assays, please see Supporting
Information.
(11) Ryu, E. K.; Ross, R. J .; Matsushita, T.; MacCoss, M.; Hong, C.
I.; West, C. R. Phospholipid-nucleoside conjugates. 3. Syntheses
and preliminary biological evaluation of 1-â-D-arabinofuranosyl-
cytosine 5′-monophosphate-L-1,2-dipalmitin and selected 1-â-D-
arabinofuranosylcytosine 5-diphosphate-L-1,2-diacylglycerols. J .
Med. Chem. 1982, 25, 1322-1329.
(12) Lambert, D. M. Rationale and applications of lipids as prodrug
carriers. Eur. J . Pharm. Sci. 2000, 11, S15-S27.
(13) Herrmann, D. B. J .; Opitz, H.-G.; Kucera, L. S. In vitro and in
vivo evaluation of the new anti-AIDS compound Fozivudine.
Naunyn-Schmiedeberg’s Arch. Pharmacol. 1997, 356, R44.
(14) Piantadosi, C.; Marasco, C. J ., J r.; Morris-Natschke, S. L.; Meyer,
K. L.; Gumus, F.; Surles, J . R.; Ishaq, K. S.; Kucera, L. S.; Iyer,
N.; Wallen, C. A.; et al. Synthesis and evaluation of novel ether
lipid nucleoside conjugates for anti-HIV-1 activity. J . Med. Chem.
1991, 34, 1408-1414.
(15) Kucera, L. S.; Iyer, N.; Morris-Natschke, S. L.; Chen, S. Y.;
Gumus, F.; Ishaq, K.; Herrmann, D. B. In vitro evaluation and
characterization of newly designed alkylamidophospholipid ana-
logues as anti-human immunodeficiency virus type 1 agents.
Antiviral Chem. Chemother. 1998, 9, 157-165.
(16) Lin, T.-S.; Prusoff, W. H. 1-(5-Azido-5-deoxy-â-d-arabinofurano-
syl) cytosine and 1-(5-Amino-5-deoxy-â-D-arabinofuranosyl) cy-
tosine. In Nucleic Acid Chemistry; Townsend, L. B., Tipson, R.
S., Eds.; J ohn Wiley & Sons: New York, 1978; pp 261-272.
(17) Ryu, E. K.; Ross, R. J .; Matsushita, T.; MacCoss, M.; Hong, C.
I.; West, C. R. Phospholipid-nucleoside conjugates. 3. Syntheses
and preliminary biological evaluation of 1-â-D-arabinofuranosyl-
cytosine 5′-monophosphate-L-1,2-dipalmitin and selected 1-â-D-
arabinofuranosylcytosine 5-diphosphate-L-1,2-diacylglycerols. J .
Med. Chem. 1982, 25, 1322-1329.
Ack n ow led gm en t. We thank the following people:
Dr. J an Hes, Gilda Saluta, Dr. Marcus Wright, J . T.
Tomlinson, Dr. J ian Dai, Nathan Iyer, Bryan Greene,
(18) Brachwitz, H.; Bergmann, J .; Thomas, Y.; Wollny, T.; Langen,
P. Synthesis and antiproliferative potency of 9-beta-D-arabino-

4208 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 19
Brief Articles
furanosyl-2-fluoroadenine phospholipid adducts. Bioorg. Med.
Chem. 1999, 7, 1195-1200.
(19) Kucera, G. L.; Goff, C. L.; Iyer, N.; Morris-Natschke, S.; Ishaq,
K. S.; Wyrick, S. D.; Fleming, R. A.; Kucera, L. S. Cellular
metabolism in lymphocytes of a novel thioether-phospholipid-
AZT conjugate with anti-HIV-1 activity. Antiviral Res. 2001, 50,
129-137.
(20) Hong, C. I.; Nechaev, A.; Kirisits, A. J .; Vig, R.; West, C. R.
Nucleoside conjugates. 13. Synthesis and antitumor activity of
1-â-D-arabinofuranosylcytosine conjugates of thioether lipids
with improved water solubility. J . Med. Chem. 1993, 36,
1785-1790.
J M020571X