Design, synthesis, and biological activity of a family of novel ceramide analogues in chemoresistant breast cancer cells

Article abstract of DOI:10.1021/jm9009668

Resistance to chemotherapy and endocrine therapy is a major cause of breast cancer treatment failure. We have synthesized six novel analogues using C8-ceramide as the lead analogue and studied their effect on hormone therapy resistant (MDA-MB-231) and che

Full text of DOI:10.1021/jm9009668

5748 J. Med. Chem. 2009, 52, 5748–5752
DOI: 10.1021/jm9009668
Design, Synthesis, and Biological Activity of a Family of Novel Ceramide Analogues in Chemoresistant
Breast Cancer Cells
James W. Antoon,†,§ Jiawang Liu,‡,§ Matthew M. Gestaut,† Matthew E. Burow,† Barbara S. Beckman,† and
Maryam Foroozesh*,‡
†Department of Pharmacology and Department of Medicine, Tulane University School of Medicine, 1430 Tulane Avenue, New Orleans, Louisiana
70112, and ‡Department of Chemistry, Xavier University of Louisiana, 1 Drexel Drive, New Orleans, Louisiana 70125. §These authors contributed
equally to this work.
Received May 11, 2009
Resistance to chemotherapy and endocrine therapy is a major cause of breast cancer treatment failure.
We have synthesized six novel analogues using C8-ceramide as the lead analogue and studied their effect
on hormone therapy resistant (MDA-MB-231) and chemoresistant (MCF-7TN-R) breast cancer cells.
Pharmacologic intervention using these ceramide analogues inhibited clonogenic survival and induced
apoptosis, with one analogue being more effective than C8-ceramide. Our results show ceramide-based
therapy has therapeutic potential in treating drug resistant breast cancer.
Introduction
These analogues were tested for their effect on breast cancer
clonogenic survival and proapoptotic activity in the chemo-
sensitive MCF-7, endocrine-resistant MDA-MB-231, and
chemoresistant MCF-7TN-R breast cancer cell lines. The last
two of these cell lines represent highly aggressive, metastatic,
and drug resistant forms of human breast cancer. The MDA-
MB-231 cell line is triple-negative (ER-R/PR/HER2 negative)
and resistant to hormone and endocrine therapies. The MCF-
7TN-R cell line was derived from MCF-7 cells grown in
increasing concentrations of tumor necrosis factor until re-
sistance was established.11 These cells exhibit increased resis-
tance to several chemotherapeutic agents including etoposide,
paclitaxel, and doxorubicin. In all cell lines tested, we found
that 3 was the most effective analogue and was more effica-
cious than ceramide in inducing apoptosis and preventing
survival of chemoresistant breast cancer cells.
Breast cancer remains the predominant form of carcinoma
affecting American women today, and resistance to chemo-
therapy and endocrine therapy is a major cause of treatment
failure. Recent research suggests that pathogenic alterations in
endogenous ceramide levels contribute to breast cancer chemo-
resistance.1-5 Targeting the bioactive sphingolipid ceramide is
a promising approach because of its regulation of cellular
apoptosis and survival. Several chemotherapeutic agents, in-
cluding paclitaxel and doxorubicin, induce apoptosis through
the induction of ceramide signaling as a mechanism of action.4,6
In fact, decreased synthesis and increased metabolism of
ceramide have been shown to be important components of
chemoresistance in human breast cancer.7-9
Our laboratory previously showed that ceramide anal-
ogues, particularly 4,6-dieneceramide, have therapeutic poten-
tial in the treatment of chemosensitive and -resistant breast
cancer.10 Here, we used C8-ceramide as the lead analogue for
the design and synthesis of six novel ceramides analogues
(Figure 1). In these analogues, we have incorporated a new
amide functional group in the sphingosine backbone, replacing
the original enol functional group (the C3 OH and the C4-C5
double bond of C8-Cer). This enol double bond is believed to
be important in ceramide’s biological activity. The presence of
this new additional amide functional group is expected to
change the polarity while testing the importance of the pre-
sence of C3 OH and C4-C5 double bond for optimum
ceramide activity. The carbon chain length in the backbone
was kept approximately the same as the known ceramides to
maintain optimum lipid solubility and facility of passage
through membranes. Other structural variations of our anal-
ogues include the presence and absence of the original amide
carbonyl group and different sizes and shapes of the substi-
tuents on the original nitrogen.
Experimental Section
General. All the chemicals were purchased from Sigma Che-
mical Co. All coupling and deprotection reactions were carried
out under anhydrous conditions. The purities of the intermedi-
ates and the products were confirmed by thin layer chromato-
graphy (Whatman 250 μm layer, UV254, flexible plates for TLC
and silica gel IB2-F). Chromatography was performed on
1
HPTLC silica gel 60 F254. H NMR spectra were recorded on
a Varian EM 360A spectrometer, and 13C NMR spectra were
recorded on Varian 300 MHz spectrometer. Mass spectral data
were determined by Micromass Quattro Micro API and Agilent
6890 GC with 5973 MS.
(S)-tert-Butyl 3-hydroxy-1-oxo-1-(tetradecylamino)propan-2-
ylcarbamate Intermediate. An amount of 1.32 g (9.76 mmol) of 1-
hydroxybenzotriazole hydrate (HOBt) was added at 0 °C to a
solution of 2.00 g (9.76 mmol) of N-Boc-L-serine in 60 mL of
anhydrous tetrahydrofuran (THF). The pH of the solution was
adjusted to 8-9 with 4-methylmorpholine. After the mixture was
stirred for 5 min, 2.21 g (10.74 mmol) of N,N0-dicyclohexyl-
carbodiimide (DCC) was added. A solution of 2.08 g (9.76 mmol)
of tetradecylamine in 3 mL of anhydrous THF was added to a
*To whom correspondence should be addressed. Phone: 504-520-
5078. Fax: 504-520-7942. E-mail: mforooze@xula.edu.
r
pubs.acs.org/jmc
Published on Web 08/20/2009
2009 American Chemical Society

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 18 5749
60 MHz) δ/ppm=7.780 (t, J=5.1 Hz, 1H), 7.304 (m, 5H), 4.726
(br, 1H), 3.790-2.869 (m, 7H), 1.566-0.706 (m, 27H). 13C NMR
(DMSO-d6, 75 MHz) δ/ppm=172.425, 141.041, 128.788, 128.651,
127.361, 64.638, 63.150, 52.097, 38.902, 31.979, 29.853, 29.717,
29.413, 29.382, 27.014, 22.763, 14.609. Anal. Calcd for
C24H42N2O2: C, 73.80; H, 10.84; N, 7.17; O, 8.19. Found: C,
71.64; H, 10.63; N, 6.75; O, 9.10.
Figure 1. Structures of C8-ceramide, C8-4,6-dieneceramide, and
target analogues studied.
Analogue 3: (S)-2-(Benzylideneamino)-3-hydroxy-N-tetrade-
cylpropanamide. A mixture of 0.50 g (1.20 mmol) of (S)-2-
amino-3-hydroxy-N-tetradecylpropanamide, 0.05 g (1.20 mmol)
of NaOH, 0.13 g (1.20 mmol) of benzaldehyde, and 10 mL of
methanol was stirred at room temperature for 8 h. The solvent
was then evaporated under reduced pressure, and the residue
was washed with cooled methanol to yield 0.32 g (68.7%) of (S)-
2-(benzylideneamino)-3-hydroxy-N-tetradecylpropanamide
solution of N-Boc-L-serine, and the mixture was stirred at 0 °C for
2 h and at room temperature overnight. On evaporation the
residue was dissolved in 60 mL of ethyl acetate. The solution was
washed successively with saturated sodium bicarbonate, 5%
potassium bisulfate, and saturated sodium chloride, and the
organic phase was separated and dried over anhydrous magne-
sium sulfate for 2 h. After filtration and evaporation under
reduced pressure 3.57 g (91.5%) of (S)-tert-butyl 3-hydroxy-1-
oxo-1-(tetradecylamino)propan-2-ylcarbamatewas obtainedasa
powder. 1H NMR (DMSO-d6, 60 MHz) δ/ppm=7.708 (t, J=5.1
Hz, 1H), 6.507 (d, J=7.9 Hz, 1H), 4.765 (t, J=5.1 Hz, 1H), 3.881
(m, 1H), 3.525 (m, 2H), 3.076 (m, 2H), 1.401 (s, 9H), 1.589-0.740
(m, 36H). 13C NMR (DMSO-d6, 75 MHz) δ/ppm = 170.725,
155.784, 78.683, 62.558, 57.548, 39.161, 31.994, 29.732, 29.504,
29.413, 28.790, 26.953, 22.778, 14.548.
(S)-2-Amino-3-hydroxy-N-tetradecylpropanamide Intermedi-
ate. At 0 °C, to a solution of 1.00 g (2.50 mmol) of (S)-tert-butyl
3-hydroxy-1-oxo-1-(tetradecylamino)propan-2-ylcarbamate in
20 mL of dichloromethane (DCM), 3.0 mL of trifluoro-
acetic acid was added. After the mixture was stirred for 4 h, the
solvent was removed under vacuum pressure, and the residue was
crystallized in diethyl ether and petroleum ether to give 0.68 g
(65.7%) of (S)-2-amino-3-hydroxy-N-tetradecylpropanamide as
a powder. 1H NMR (DMSO-d6, 60 MHz) δ/ppm=8.358 (t, J=
6.0 Hz, 1H), 8.112 (m, 3H), 4.583 (br, 1H), 3.715 (m, 3H), 3.093
(m, 2H), 1.531-0.677 (m, 27H). 13C NMR (DMSO-d6, 75 MHz)
δ/ppm=167.141, 61.040, 55.042, 39.449, 31.979, 29.747, 29.701,
29.489, 29.398, 26.968, 22.763, 14.563.
1
as colorless powder. ESI/MS (m/e) 389 [M þ H]þ. H NMR
(DMSO-d6, 60 MHz) δ/ppm = 8.281 (s, 1H), 7.818 (m, 3H),
7.515(m, 3H), 4.795 (br, 1H), 3.768 (m, 3H), 3.103 (m, 2H),
1.584-0.702 (m, 27H). 13C NMR (DMSO-d6, 75 MHz) δ/ppm=
170.497, 163.239, 136.532, 131.597, 129.198, 129.061, 76.254,
63.788, 39.039, 31.964, 29.808, 29.701, 29.382, 26.983, 22.763,
14.609. Anal. Calcd for C24H40N2O2: C, 74.18; H, 10.38; N,
7.21; O, 8.23. Found: C 72.83; H, 10.44; N, 7.14; O, 8.77.
Analogue 4: (S)-3-Hydroxy-2-pivalamido-N-tetradecylpropa-
namide. At 0 °C to a solution of 0.10 g (0.98 mmol) of trimethyl-
acetic acid in anhydrous THF (60 mL), 0.13 g (0.96 mmol) of
HOBt and 0.40 g (0.97 mmol) of (S)-2-amino-3-hydroxy-N-
tetradecylpropanamide were added. After 5 min, 0.23 g
(1.12 mmol) of DCC was added and the pH of the solution was
adjusted to 8-9 with 4-methylmorpholine. The mixture was
stirred at 0 °C for 2 h and at room temperature overnight. On
evaporation the residue was dissolved in 100 mL of ethyl acetate.
The solution was washed successively with saturated sodium
bicarbonate, 5% potassium bisulfate, and saturated sodium
chloride, and the organic phase was separated and dried over
anhydrous magnesium sulfate for 2 h. After filtration and
evaporation under reduced pressure 0.28 g (72.9%) of (S)-3-
hydroxy-2-pivalamido-N-tetradecylpropanamide was obtained
as a powder. ESI/MS (m/e) 385 [M þ H]þ. 1H NMR (DMSO-
d6, 90 MHz) δ/ppm=7.642(t, J=5.1Hz, 1H), 7.088(d, J=7.9 Hz,
1H), 4.818 (t, J =5.1 Hz, 1H), 4.198 (m, 1H), 3.552(m, 2H), 3.037
(m, 2H), 1.124 (s, 9H), 1.572-0.728 (m, 36H). 13C NMR
(DMSO-d6, 75 MHz) δ/ppm=177.861, 170.558, 62.406, 55.908,
39.161, 31.979, 29.732, 29.701, 29.443, 29.398, 27.910, 26.938,
22.778, 14.609. Anal. Calcd for C22H44N2O3: C, 68.70; H, 11.53;
N, 7.28; O, 12.48. Found: C 68.25; H, 11.51; N, 7.10; O, 13.09.
Analogue 5: (S)-N-(3-Hydroxy-1-oxo-1-(tetradecylamino)pro-
pan-2-yl)-3,5-dimethoxybenzamide. At0 °Ctoa solutionof0.18 g
(1.0 mmol) of 3,5-dimethoxybenzoic acid in anhydrous THF
(40 mL), 0.14 g (1.0 mmol) of HOBt and 0.30 g (1.0 mmol) of (S)-
2-amino-3-hydroxy-N-tetradecylpropanamide were added. Then
0.23 g (1.1 mmol) of DCC was added, and the pH of the solution
was adjusted to 8 with 4-methylmorpholine. The mixture was
stirred at 0 °C for 2 h and at room temperature overnight. On
evaporation the residue was dissolved in 100 mL of ethyl acetate.
The solution was washed successively with saturated sodium
bicarbonate, 5% potassium bisulfate, and saturated sodium
chloride, and the organic phase was separated and dried
over anhydrous magnesium sulfate for 2 h. After filtration and
evaporation under reduced pressure, 0.38 g (82.0%) of (S)-N-(3-
hydroxy-1-oxo-1-(tetradecylamino)propan-2-yl)-3,5-dimethoxy-
benzamide was obtained as a powder. GC/MS with BSTFA/
TMCS (m/e) 518 [M þ 1TMS - H2O]þ, 662 [M þ 3TMS -
H2O]þ. 1H NMR (DMSO-d6, 60 MHz) δ/ppm=8.186 (d, J=7.9
Hz, 1H), 7.830 (t, J=5.1 Hz, 1H), 7.054 (d, J=2.4 Hz, 2H), 6.700
(t, J=2.4 Hz, 1H), 4.874 (t, J =5.1 Hz, 1H), 4.413 (m, 1H), 3.788
Analogue 1: (S)-N-(3-Hydroxy-1-oxo-1-(tetradecylamino)pro-
pan-2-yl)cyclohexanecarboxamide. At 0 °C, to a solution of 0.16
g (1.20 mmol) of cyclohexanecarboxylic acid in anhydrous THF
(20 mL) and DMF (5 mL), 0.16 g (1.20 mmol) of HOBt and 0.50
g (1.20 mmol) of (S)-2-amino-3-hydroxy-N-tetradecylpropan-
amide were added. After 5 min, 0.30 g (1.45 mmol) of DCC was
added, and the pH of the solution was adjusted to 8-9 with 4-
methylmorpholine. The mixture was stirred at 0 °C for 2 h and at
room temperature overnight. On evaporation the residue was
dissolved in 100 mL of ethyl acetate. The solution was washed
successively with saturated sodium bicarbonate, 5% potassium
bisulfate, and saturated sodium chloride, and the organic phase
was separated and dried over anhydrous magnesium sulfate for
2 h. After filtration and evaporation under reduced pressure
0.40 g (81.3%) of (S)-N-(3-hydroxy-1-oxo-1-(tetradecylamino)-
propan-2-yl)cyclohexanecarboxamide was obtained as powder.
1
ESI/MS (m/e) 411 [M þ H]þ. H NMR (DMSO-d6, 60 MHz)
δ/ppm=7.559 (m, 2H), 4.762 (t, J=5.1 Hz, 1H), 4.195 (m, 1H),
3.495 (m, 3H), 3.020 (m, 2H), 1.912-0.711 (m, 37H). 13C NMR
(DMSO-d6, 75 MHz) δ/ppm=175.841, 170.649, 62.452, 55.604,
44.368, 39.145, 34.029, 31.979, 29.884, 29.717, 29.458, 29.382,
26.938, 26.164, 26.012, 25.146, 22.778, 14.609. Anal. Calcd for
C24H46N2O3: C, 70.20; H, 11.29; N, 6.82; O, 11.69. Found: C,
69.34; H, 11.25; N, 6.52; O, 12.42.
Analogue 2: (S)-2-(Benzylamino)-3-hydroxy-N-tetradecylpro-
panamide. To a solution of 0.20 g (0.52 mmol) of (S)-2-(ben-
zylideneamino)-3-hydroxy-N-tetradecylpropanamide in 10 mL of
methanol, 32 mg (0.52 mmol) of sodium borohydride (NaBH4)
was added. The mixture was stirred at room temperature for 10 h
before evaporation under vacuum. The residue was purified
by thin layer chromatography to give 115 mg (57.5%) of (S)-
2-(benzylamino)-3-hydroxy-N-tetradecylpropanamide as a color-
less powder. ESI/MS (m/e) 391 [M þ H]þ. 1H NMR (DMSO-d6,
(s, 6H), 3.674 (m, 2H), 3.054 (m, 2H), 1.708-0.705 (m, 27H). 13
C
NMR (DMSO-d6, 75 MHz) δ/ppm=170.345, 166.580, 160.946,
137.017, 106.150, 103.857, 62.361, 56.971, 56.120, 39.252, 31.979,
29.747, 29.701, 29.443, 29.398, 26.968, 22.778, 14.609. Anal.

5750 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 18
Scheme 1. Synthesis Scheme for the Ceramide Analoguesa
Antoon et al.
a (i) DCC/HOBT in THF 0 °C; (ii) TFA/CH2Cl2 (0 °C), NaCO3/ethyl acetate; (iii) DCC/HOBT in THF 0 °C and different acids; (iv) NaOH/
benzaldehyde in methanol; v) NaBH3CN/methanol.
Table 1. IC50 of Ceramide Analogues (μM)a
Calcd for C26H44N2O5: C, 67.21; H, 9.54; N, 6.03; O, 17.22.
Found: C 65.08; H, 9.59; N, 6.05; O, 17.23.
compd
MCF-7
MDA-MB-231
MCF-7-TNR
Analogue 6: (S)-2-Acetamido-3-hydroxy-N-tetradecylpropan-
amide. At 0 °C to a solution of 0.026 mL (0.48 mmol) of acetic
acid in anhydrous THF (50 mL), 32 mg (0.24 mmol) of HOBt
and 0.10 g (0.24 mmol) of (S)-2-amino-3-hydroxy-N-tetradecyl-
propanamide were added. Then 53 mg (0.26 mmol) of DCC in
5 mL of THF was added, and the pH was adjusted to 7-8 with 4-
methylmorpholine. The mixture was stirred at 0 °C for 2 h and at
room temperature overnight. On evaporation the residue was
dissolved in 80 mL of ethyl acetate. The solution was washed
successively with saturated sodium bicarbonate, 5% potassium
bisulfate, and saturated sodium chloride, and the organic phase
was separated and dried over anhydrous magnesium sulfate for
2 h. After filtration and evaporation under reduced pressure
crude product was obtained and recrystallized using ethyl
acetate to obtain 0.06 g (73.0%) of (S)-2-acetamido-3-hydro-
DH-Cer
C8-Cer
31.7
4.2
41.4
3.0
49.6
8.6
Analog 1
Analog 2
Analog 3
Analog 4
Analog 5
Analog 6
7.2
5.0
11.5
10. 5
2.4
23.5
15.2
7.7
4.7
6.7
6.1
6.6
11.1
15.0
52.7
11.2
44.7
75.1
a The IC50 values were calculated from experiments shown in Figure 1,
Supporting Information, using the formulas described in the Materials
occurred in mild conditions, and ceramide analogues were
produced with acceptable yields.
Analogue 3 Displays the Greatest Antisurvival Effect in
MDA-MB-231 and MCF-7TN-R Cells. The six ceramide
analogues were compared for their inhibitory effect on breast
cancer clonogenic survival. Cells were treated with varying
concentrations of analogue, ranging from 0.1 to 20 μM for 10
days. Cell colonies were then fixed, stained, and recorded as
percent vehicle control. The IC50 values for the analogues are
listed in Table 1. All analogues displayed antisurvival prop-
erties in drug-resistant and drug-sensitive breast cancers.
All six analogues showed biological activity in the MCF-7
cell line. Analogs 2 and 3, both missing the original amide
carbonyl group, had the greatest efficacy in MCF-7 cells,
with IC50 of 5.0 μM (p < 0.01) and 4.7 μM (p < 0.001),
respectively. Interestingly, these analogues also showed
activity in endocrine-resistant MDA-MB-231 and chemo-
resistant MCF-7TN-R cells. Analog 3 was the most potent
antisurvival agent and more effective than C8-ceramide in
treating drug-resistant breast cancer, with IC50 of 2.4 μM in
MDA-MB-231 cells and 7.7 μM in MCF-7TN-R cells, with
p of 0.037and 0.019, respectively. Analogue 6 was the least
effective in all of the cell lines tested.
xy-N-tetradecylpropanamide as
a powder. GC/MS with
BSTFA/TMCS (m/e) 414 [M þ 1TMS]þ, 468 [M þ 2TMS -
H2O] þ. 1H NMR (DMSO-d6, 60 MHz) δ/ppm=7.730 (m, 2H),
4.775 (t, J=5.1 Hz, 1H), 4.155 (m, 1H), 3.495 (m, 2H), 3.020 (m,
2H), 1.846 (s, 3H), 1.740-0.632 (m, 27H). 13C NMR (DMSO-
d6, 75 MHz) δ/ppm=168.538, 61.617, 56.849, 39.297, 31.964,
29.717, 29.686, 29.580, 29.382, 26.923, 22.763, 14.609. Anal.
Calcd for C19H38N2O3: C, 66.63; H, 11.18; N, 8.18; O, 14.01.
Found: C 66.75; H, 10.39; N, 7.56; O, 13.46.
Results
Synthesis of Analogues 1-6. The synthetic route of anal-
ogues 1-6 is described in Scheme 1. Boc-protecting L-serine
was coupled with tetradecylamine, giving (S)-tert-butyl-3-
hydroxy-1-oxo-1-(tetradecylamino)propan-2-ylcarbamate in
91.5% yield. After removal of the tert-butoxycarbonyl (Boc)
group with trifluoroacetic acid in dichloromethane, the com-
mon intermediate (S)-2-amino-3-hydroxy-N-tetradecylpro-
panamide was formed in 65.7% yield. Analogs 1, 4, 5, and 6
were produced through the coupling of (S)-2-amino-3-hydro-
xy-N-tetradecylpropanamide with cyclohexanecarboxylic
acid (in 81.3% yield), pivalic acid (in 72.9% yield), 3,5-
dimethoxybenzoic acid (in 82.0% yield), and acetic acid (in
73.0%), respectively. In the presence of sodium hydroxide,
condensation with benzaldehyde in methanol provided 3 in
68.7% yield, which was reduced by NaBH3CN to give 2 in
57.5% yield. These results show that all of these reactions
Since 3 showed the greatest efficacy compared to the other
newly synthesized analogues, we tested the ability of this
drug to induce apoptosis in MCF-7 cells. Using a cell death
detection ELISA (Roche), we measured the amount of
apoptosis with IC50 treatments of 3 and C8-ceramide. As
seen in Figure 2, 3 caused a 5.4-fold induction (p=0.034) in

Brief Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 18 5751
Figure 2. Effect of 3 on breast cancer apoptosis. MCF-7 cells were
treated with double IC50 concentrations of ceramide (8.3 μM) and 3
(9.4 μM) for 24 h. The values are the mean ( SE of three
independent experiments.
apoptosis compared to a 5.0-fold induction caused by C8-
ceramide (p=0.008).
Analogue 3 Inhibits Proliferation and Sphingosine-1-phos-
phate Levels in Chemoresistant Breast Cancer Cells. Both 3
and C8-ceramide have similar survival IC50 values in MCF-7
cells, and the disparity in proapoptotic ability led us to
evaluate the ability of 3 to block proliferation as an anti-
survival mechanism in addition to apoptosis. The capacity of
novel drugs to inhibit long-term survival and proliferation is
essential in determining the therapeutic potential of any new
treatment.12,13 We found that 3 has a greater antiprolifera-
tive effect than C8-ceramide in MCF-7, MDA-MB-231, and
MCF-7TN-R cell lines, with IC50 of 4.5, 1.8, and 1.2 μM,
respectively (Figure 3).
Selective toxicity for cancer, but not normal, cells is
essential in the development of targeted cancer experimental
therapeutics. To determine the selectivity of 3 and C8-
ceramide to inhibit cell growth of cancer cells, we performed
proliferation assays using MCF10A (normal breast
epithelial) cells (Figure 4). Interestingly, 3 and C8-ceramide
showed little effect on normal breast epithelial proliferation,
even at 100 μM, which is greater than 50-fold of the IC50 in
chemoresistant cells.
Figure 3. Effect of 3 on breast cancer proliferation. (A) MCF-7,
(B) MDA-MB-231, and (C) MCF-7TN-R cells were treated with
varying concentrations of ceramide analogue (0.1-100 μM) for 24 h.
The values are the mean ( SE of three independent experiments.
The increased potency of 3 in hormone therapy resistant
and chemoresistant breast cancer cells compared to drug
sensitive MCF-7 cells led us to evaluate the effect of this
analogue on endogenous sphingolipid levels compared to
C8-ceramide (Figure 5). We demonstrate that dihydrosphin-
gosine is elevated following 3 treatment, with almost a
50-fold increase in MCF-7TN-R cells. Sphingosine-1-phos-
phate, the proliferative metabolite of ceramide, decreased in
all three cell lines. Interestingly, C8-ceramide had the oppo-
site effect and increased sphingosine-1-phosphate levels.
These results correlate with the increased potency of 3 in
reducing proliferation in MDA-MB-231 and MCF-7TN-R
cells.
Figure 4. Effect of 3 and C8-ceramide on normal breast prolifera-
tion. MCF10A cells were treated with varying concentrations of
drug (0.1-100 μM) for 24 h. The values are the mean ( SE of four
independent experiments.
Discussion
ceramide analogue. The presence of the new amide functional
group and the length of the carbon backbone cause a change
in polarity and improve the facility of passage through
membranes. Other structural variations are the presence and
absence of the original amide carbonyl group and different
sizes and shapes of the substituents on the original nitrogen.
Analogues 1, 4, 5, and 6 contain the original amide group of
ceramides but are not potent antisurvival agents. Analogues 2
All six novel ceramide analogues studied were designed to
havethe same long chain backbone containing 17 carbons and
1 amide nitrogen. This new amide functional group replaced
the allyl alcohol functional group of the C8-ceramide. It was
previously believed that the enol trans double bond contri-
butes to ceramide’s biological activity. However, our results
show that 3, lacking this enol functional group, is an effective

5752 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 18
Antoon et al.
pharmacological intervention using ceramide-based therapy
has therapeutic potential in treating drug resistant breast
cancer.
Acknowledgment. We thank Steven Elliott for technical
expertise in performing apoptosis assays and Dr. James
Robinson’s laboratory at Tulane for help in quantifying
apoptosis and MTT results. We also thank Dr. Jacek
Bielawski and the M.U.S.C. Lipidomics Core for their assis-
tance and expediency in quantifying cellular lipid levels. This
research was supported by a grant from the Louisiana Cancer
Research Consortium (Grant 631324).
Supporting Information Available: Experimental conditions
and analytical data for analogues 1-6. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Simstein, R.; Burow, M.; Parker, A.; Weldon, C.; Beckman, B.
Apoptosis, chemoresistance, and breast cancer: insights from the
MCF-7 cell model system. Exp. Biol. Med. (Maywood, NJ, U. S.)
2003, 228, 995–1003.
(2) Shabbits, J. A.; Mayer, L. D. P-Glycoprotein modulates ceramide-
mediated sensitivity of human breast cancer cells to tubulin-bind-
ing anticancer drugs. Mol. Cancer Ther. 2002, 1, 205–213.
(3) Gouaze-Andersson, V.; Yu, J. Y.; Kreitenberg, A. J.; Bielawska,
A.; Giuliano, A. E.; Cabot, M. C. Ceramide and glucosylceramide
upregulate expression of the multidrug resistance gene MDR1 in
cancer cells. Biochim. Biophys. Acta 2007, 1771, 1407–1417.
(4) Liu, Y. Y.; Yu, J. Y.; Yin, D.; Patwardhan, G. A.; Gupta, V.;
Hirabayashi, Y.; Holleran, W. M.; Giuliano, A. E.; Jazwinski,
S. M.; Gouaze-Andersson, V.; Consoli, D. P.; Cabot, M. C. A role
for ceramide in driving cancer cell resistance to doxorubicin.
FASEB J. 2008, 22, 2541–2551.
(5) Schiffmann, S.; Sandner, J.; Birod, K.; Wobst, I.; Angioni, C.;
Ruckhaberle, E.; Kaufmann, M.; Ackermann, H.; Lotsch, J.;
Schmidt, H.; Geisslinger, G.; Grosch, S. Ceramide synthases and
ceramide levels are increased in breast cancer tissue. Carcinogenesis
2009, 30, 745–752.
(6) Qiu, L.; Zhou, C.; Sun, Y.; Di, W.; Scheffler, E.; Healey, S.;
Wanebo, H.; Kouttab, N.; Chu, W.; Wan, Y. Paclitaxel and
ceramide synergistically induce cell death with transient activation
of EGFR and ERK pathway in pancreatic cancer cells. Oncol. Rep.
2006, 16, 907–913.
Figure 5. Effect of 3 on cellular sphingolipid species. (A) MCF-7,
(B) MDA-MB-231, and (C) MCF-7TN-R cells were treated with
ceramide (10 μM) and 3 (10 μM) for 24 h. The values are the mean (
SE of two independent experiments.
and 3, both lacking the original amide carbonyl group, show
the greatest efficacy in MCF-7 cells, endocrine-resistant
MDA-MB-231 cells, and chemoresistant MCF-7TN-R cells.
This indicates that this modification at the original N atom of
ceramide increases the antisurvival effect of the ceramide
analogues. Since the main metabolic pathway of ceramide is
based on the original amide functional group, the increased
antisurvival effect of these ceramide analogues in comparison
with ceramide is most probably due to a change in meta-
bolism. Analogues 2 and 3 also share another important
structural feature, a benzyl substituent, not present in
the other analogues. Their only structural difference is the
presence of a carbon-nitrogen double bond in 3 in place of a
carbon-nitrogen single bond in 2. This minor structural
difference has led to significant changes in the antiprolifera-
tion activity between these analogues. The benzylideneamino
group in 3 is harder to hydrolyze than the benzylamino group
in 2. Ceramides cause apoptosis through their interactions
with caspases and other apoptotic signaling pathways. Since
the hydrolysis of ceramides and ceramide analogues by the
enzyme ceramidase leads to sphingosine or sphingosine anal-
ogues that are metabolized to form the cell growth-promoting
sphingosine-1-phosphate or similar analogues, slower meta-
bolism by hydrolysis is expected to lead to increased ceramide
or ceramide analogue concentrations in the cells and higher
apoptotic activity.10 Taken together, our results show that
(7) Lucci, A.; Giuliano, A. E.; Han, T. Y.; Dinur, T.; Liu, Y. Y.;
Senchenkov, A.; Cabot, M. C. Ceramide toxicity and metabolism
differ in wild-type and multidrug-resistant cancer cells. Int.
J. Oncol. 1999, 15, 535–540.
(8) Liu, Y. Y.; Han, T. Y.; Giuliano, A. E.; Cabot, M. C. Expression of
glucosylceramide synthase, converting ceramide to glucosylcera-
mide, confers adriamycin resistance in human breast cancer cells.
J. Biol. Chem. 1999, 274, 1140–1146.
(9) Ameyar, M.; Atfi, A.; Cai, Z.; Stancou, R.; Shatrov, V.; Bettaieb,
A.; Chouaib, S. Analysis of human breast adenocarcinoma MCF7
resistance to tumor necrosis factor-induced cell death. Lack of
correlation between JNK activation and ceramide pathway. J. Biol.
Chem. 1998, 273, 29002–29008.
(10) Struckhoff, A. P.; Bittman, R.; Burow, M. E.; Clejan, S.; Elliott, S.;
Hammond, T.; Tang, Y.; Beckman, B. S. Novel ceramide analogs
as potential chemotherapeutic agents in breast cancer. J. Pharma-
col. Exp. Ther. 2004, 309, 523–532.
(11) Weldon, C. B.; Parker, A. P.; Patten, D.; Elliott, S.; Tang, Y.;
Frigo, D. E.; Dugan, C. M.; Coakley, E. L.; Butler, N. N.; Clayton,
J. L.; Alam, J.; Curiel, T. J.; Beckman, B. S.; Jaffe, B. M.; Burow,
M. E. Sensitization of apoptotically-resistant breast carcinoma
cells to TNF and TRAIL by inhibition of p38 mitogen-activated
protein kinase signaling. Int. J. Oncol. 2004, 24, 1473–1480.
(12) Yacoub, A.; Mitchell, C.; Brannon, J.; Rosenberg, E.; Qiao, L.;
McKinstry, R.; Linehan, W. M.; Su, Z. S.; Sarkar, D.; Lebedeva, I.
V.; Valerie, K.; Gopalkrishnan, R. V.; Grant, S.; Fisher, P. B.;
Dent, P. MDA-7 (interleukin-24) inhibits the proliferation of renal
carcinoma cells and interacts with free radicals to promote cell
death and loss of reproductive capacity. Mol. Cancer Ther 2003, 2,
623–632.
(13) Meeran, S. M.; Katiyar, S.; Elmets, C. A.; Katiyar, S. K. Inter-
leukin-12 deficiency is permissive for angiogenesis in UV radiation-
induced skin tumors. Cancer Res. 2007, 67, 3785–3793.