Cytotoxic effects of combination of oxidosqualene cyclase inhibitors with atorvastatin in human cancer cells

Article abstract of DOI:10.1021/jm300256z

Ten oxidosqualene cyclase inhibitors with high efficacy as cholesterol-lowering agents and of different chemical structure classes were evaluated as potential anticancer agents against human cancer cells from various tissue origins and nontumoral human-brain-derived endothelial cells. Inhibition of cancer cell growth was demonstrated at micromolar concentrations, comparable to the concentrations of statins necessary for antitumor effect. Human glioblastoma cells were among the most sensitive cells. These compounds were also able to decrease the proliferation of angiogenic brain-derived endothelial cells, as a model of tumor-induced neovasculation. Additive effects in human glioblastoma cells were also demonstrated for oxidosqualene cyclase inhibitors in combination with atorvastatin while maintaining selectivity against endothelial cells. Thus, not only statins targeting the 3-hydroxy-3- methylglutaryl coenzyme A reductase but also inhibitors of oxidosqualene cyclase decrease tumor growth, suggesting new therapeutic opportunities of combined anti-cholesterol agents for dual treatment of glioblastoma.

Full text of DOI:10.1021/jm300256z

Article
pubs.acs.org/jmc
Cytotoxic Effects of Combination of Oxidosqualene Cyclase
Inhibitors with Atorvastatin in Human Cancer Cells
Davide Staedler,†,‡ Catherine Chapuis-Bernasconi,† Henrietta Dehmlow,§ Holger Fischer,§
Lucienne Juillerat-Jeanneret,*,† and Johannes D. Aebi*,§
†Centre Hospitalier Universitaire Vaudois (CHUV) and University of Lausanne (UNIL), CH-1011 Lausanne, Switzerland
‡Institute of Chemical Sciences and Engineering, Ecole Polytechnique Fed
́ ́
erale de Lausanne, CH-1015, Lausanne, Switzerland
§F. Hoffmann-La Roche Ltd., Pharmaceutical Division, CH-4070 Basel, Switzerland
S
* Supporting Information
ABSTRACT: Ten oxidosqualene cyclase inhibitors with high
efficacy as cholesterol-lowering agents and of different
chemical structure classes were evaluated as potential
anticancer agents against human cancer cells from various
tissue origins and nontumoral human-brain-derived endothe-
lial cells. Inhibition of cancer cell growth was demonstrated at
micromolar concentrations, comparable to the concentrations
of statins necessary for antitumor effect. Human glioblastoma
cells were among the most sensitive cells. These compounds
were also able to decrease the proliferation of angiogenic
brain-derived endothelial cells, as a model of tumor-induced neovasculation. Additive effects in human glioblastoma cells were
also demonstrated for oxidosqualene cyclase inhibitors in combination with atorvastatin while maintaining selectivity against
endothelial cells. Thus, not only statins targeting the 3-hydroxy-3-methylglutaryl coenzyme A reductase but also inhibitors of
oxidosqualene cyclase decrease tumor growth, suggesting new therapeutic opportunities of combined anti-cholesterol agents for
dual treatment of glioblastoma.
myeloblastic leukemia, and colorectal tumors.8,10 Statins have
INTRODUCTION
Cholesterol is an essential lipid in the body, in particular for the
■
also shown antimetastatic activity11−14 and can act additively or
synergistically with anticancer chemotherapeutic drugs.6,7,10
stability and function of cell membranes and cell organelles and
for cellular trafficking.1,2 An excess of cholesterol causes lipid
However, these effects were not always consistent, and it was
disorders, such as hypercholesterolemia.3,4 The rate-limiting
suggested that the hydrophobicity of the statins is important for
antitumor activity.15
enzyme for cholesterol biosynthesis is 3-hydroxy-3-methylglu-
taryl coenzyme A (HMG-CoA) reductase, and inhibitors of this
Statin concentrations as anticancer agents is much higher
than statin concentrations necessary as antihyperlipidemic
enzyme, the statins, are used for the treatment of lipid
disorders.3,4 In addition to their effects on hypercholester-
agents. In the past years, novel synthetic inhibitors of HMG-
CoA reductase activity that inhibit tumor cell proliferation and
olemia, statins were recently associated with a reduced risk of
are structurally unrelated to existing statins have also been
cancer progression, including melanoma, prostate, and color-
evaluated.16 Alternatively, new molecules that inhibit other
enzymes of cholesterol biosynthesis, including inhibitors of the
oxidosqualene/lanosterol cyclase (OSC) have been developed,
also at Hoffmann-La Roche.16−23 However, until now these
inhibitors have been evaluated only for their anti-cholesterol
potential. In this work we evaluated whether inhibitors of OSC,
like the statins, may also be potential anticancer agents, either
alone or in combination with the statin atorvastatin.
ectal cancers.5,6 These effects were in part due to the anti-
inflammatory properties of statins, their inhibition of the
mevalonate pathway, and HMG-CoA reductase-independent
effects, such as the inhibition of protein−protein interaction.5
Inhibition of cholesterol synthesis by statins modifies the
composition of the cell membrane via the reduction of the
isoprenoids farnesyl pyrophosphate and gerenylgeranyl pyro-
phosphate formation and the subsequent prenylation of small
GTP-binding proteins. Statins also affect multiple cellular
signaling pathways, such as the mitogen-activated protein
kinases (MAPKs), phosphatidiylinsotiol 3 kinase (PI3K), AKT
(protein kinase B), and the epidermal growth factor receptor
(EGFR).7,8 In vitro statins cause tumor-cell death and growth
inhibition, and induce cell cycle arrest.9,10 In vivo, statins inhibit
cancer growth, including neuroblastoma, glioma, acute
OSC is an enzyme downstream of mevalonate synthesis
involved in the biosynthesis of steroids and chlolesterol. In
cancer, the effects of statins are postulated to be related to the
inhibition of the biosynthesis of mevalonate-derived farnesyl
Received: November 14, 2011
Published: April 26, 2012
© 2012 American Chemical Society
4990
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
and geranylgeranyl groups required for isoprenylation of
oncogenic proteins. Mevalonate is the initial molecule of
diverse pathways important for signaling in cell growth,
proliferation, migration, and survival. The inhibition of HMG-
CoA reductase by statins results in a few hours in an increased
level of the protein as a net result of a feedback mechanism,
thus reducing the efficacy of the statins. The enzymatic-
mediated biosynthesis of steroids, downstream of HMG-CoA
reductase, originates with farnesyl pyrophosphate through
reductive dimerization to squalene, then conversion of squalene
to lanosterol through epoxidation of squalene by squalene
epoxidase to produce 2,3-oxidosqualene. OSC then generates
the tetracyclic precursor of cholesterol, lanosterol. Synthesis of
24(S),25-epoxycholesterol is favored over cholesterol synthesis
under conditions of partial OSC inhibition, whereas complete
OSC inhibition results in decreased synthesis of cholesterol and
24(S),25-epoxycholesterol. Because 24(S),25-epoxycholesterol
not only represses HMG-CoA reductase activity but also
enhances its degradation, this should result in the generation of
a synergistic, self-limited, negative regulatory feedback
loop.17,24−28 Therefore, the first aim of our work was to
evaluate new OSC inhibitors (OSCi) as in vitro anticancer
agents, either as single agents or in combination with the statin
atorvastatin, against human cancer cells of diverse origins, then
to focus on human glioblastoma cells and human angiogenic
brain-derived endothelial cells, hypothesizing that mevalonate
depletion in tumor cells or endothelial cells may contribute to
inhibition of tumor growth and angiogenesis.
imidazole-1-carboxylic acid 4-trifluoromethylphenyl ester to
give carbamate 21. After reaction of 21 with ethyl-(2-hydroxy-
ethyl)amine, compound 7 was converted to its citric acid salt
and isolated by crystallization. The preparation of amide 10 is
shown in Scheme 3. Building block 19 was oxidized to
carboxylic acid 22 using Sharpless conditions,34 and N-Boc
deprotection yielded amino acid 23 as the hydrochloride salt.
After heating in hexamethyldisilazane, the persilylated amino
acid was reacted with 4-chlorophenyl chloroformate. Hydrolysis
of the silyl ester during workup and coupling with 4-hydroxy-
piperidine gave amide 10.
RESULTS AND DISCUSSION
■
All compounds selected for the present experiments demon-
strated the highest efficacy out of their congeners in the
respective chemical class as cholesterol lowering agents in the
hamster model of hypercholesterolhemeia (results not shown).
The most advanced OSCi, compound 5, had been previously
shown to decrease very low density lipoproteins (VLDL) and
low density lipoproteins (LDL) apoB100 in miniature pigs.22
For in vivo evaluations and the present experiments, the
metabolic stability of the OSCi was determined in rat hepato-
cytes, assessing the efficacy and potency of each compound in
reducing cholesterol production while increasing monooxidos-
qualene (MOS) levels.19 The newly synthesized carbamates 5,
6, and 7 showed residual inhibitory activies greater than or
equal to 60% after 24 h at 300 nM (Supporting Information
Table S2). Amide 10, which was structurally the least related
OSCi, was less stable (76% inhibition after 8 h and 8% inhibi-
tion after 24 h). We also investigated the chemical stability at
pH 4 and 7.4 at 37 °C and found no degradation of the OSCi.
However, we observed lower recovery rates for compounds 2,
4, 6, and 8 at pH 7.4, which is probably due to difficulties in
keeping the compound in solution at this pH. However no
degradation products were identified (Supporting Information
Table S2).
CHEMISTRY
■
The syntheses and characterization of compounds 1, 2, 3, 4, 8,
and 917−20,29 have been previously described (structures shown
in Table 1). Here we describe the syntheses of the alkyne-
cyclohexyl inhibitors (compounds 5 and 6, Scheme 1) and of
the cyclohexylethyl inhibitor (compound 7, Scheme 2) in
which the cyclohexyl moiety of the already known compound
829 was moved closer to the amine. Then the replacement of
this important amine by an amide (compound 10, Scheme 3)
was done with only a minimal loss of inhibitory activity.
Compound 1,20 bearing a rigid triphenyl ketone system, mimics
the high energy intermediate of oxidosqualene cyclization to
lanosterol.30,31 The benzylamine moiety of 1 was replaced by
an oxyalkylamine (compound 2),17 and the benzophenone was
replaced by a heteroaryl (compounds 3 and 4),19,29 by an
alkyne cyclohexylcarbamate (compounds 5 and 6), or an alkyl-
cyclohexyl system (compounds 8 and 9).29
First we evaluated the OSCi in human cells, 11 cancer cell
lines from various tissue origins, and 1 nontumoral human
brain-derived endothelial cell line (HCEC). Only compounds
1, 2, 3, 4, and 8 displayed efficient antiproliferative properties
toward the majority of the cell lines after 24, 48 or 72 h of
exposure (Supporting Information Table S3). The efficacy
increased with the time of exposure, starting already after 24 h
of exposure. In human cancer cells, the IC50 values were
dependent on the particular cell line considered (Table 2 and
Figures 1 and 2). In human brain-derived LN18 and LN229
glioblastoma cells and HCEC brain-derived endothelial cells
compounds 1, 2, 3, and 4 were the most efficient compounds,
displaying IC50 values lower than 10 μM for at least one tumor
cell line (Table 2 and Figures 1 and 2), compound 1 being the
most selective compound for tumor cells compared to
nontumoral HCEC cells. Atorvastatin (AT) also displayed
some selectivity for human brain-derived tumor cells compared
to endothelial cells but became active only after 72 h of cell
exposure. The effect of OSCi or AT on DNA synthesis by these
cells exposed to the compounds for 24 h was also determined
using the thymidine incorporation assay. As for cell survival
assays, OSCi compounds 2, 3, and 4 were the most active
inhibitors of DNA synthesis of the series against the three cells
(Figure 3 and Supporting Information Table S4). OSCi com-
pounds 1 and 5−9 decreased DNA synthesis with less efficacy
than OSCi compounds 2, 3, and 4, but the effects were com-
parable in the three cells, whereas AT and OSCi 10 were less
The synthesis of the alkyne OSCi is depicted in Scheme 1.
Starting from trans-aminocyclohexanecarboxylic acid 11, the
aldehyde 12 was synthesized via the formation of the Weinreb
amide, N-methylation of the Boc-protected amine, and lithium
aluminum hydride reduction of the Weinreb amide. Corey−
Fuchs alkyne introduction32 was followed by reaction with
the appropriate halogenide to give intermediate 14. Selective
O-deprotection, mesylation, and cleavage of the Boc-protecting
group using TFA gave amine 15, which was converted to
carbamate 16. Amination and salt formation gave the two
alkynes 5 and 6. The synthesis of cyclohexylethyl linked OSCi
7 is described in Scheme 2. Silyl protected trans-cyclohexyl 1733
was hydrogenated and deprotected with platinum oxide in
ethanol/chloroform to form hydrochloride 18, which after Boc
protection and N-methylation gave alcohol 19. Mesylation and
deprotection provided amine 20 as the hydrochloride which
could be reacted with the in situ generated N-methylated
4991
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Table 1. Structure and in Vitro Properties of the OSCi and AT
a
b
All potency measurements were performed in triplicate. The standard error of the mean (sem) ranged between 10% and 20%. All measurements
were done in triplicate, and the reproducibility was 0.1%. Apparent permeability Pe from the PAMPA assay at pH 6.5;35 All measurements were
done in triplicate, and the reproducibility was 4%. Percent of the compound binding to the PAMPA lipid membrane; all measurements were done
in triplicate, and the reproducibility was 4%. Human LN229, LN18, and HCEC cells were exposed to increasing concentrations (0, 2, 5, 10, 20,
c
d
e
and 30 μM) of the OSCi compounds 1−10 or AT for 72 h. Then the MTT assay was performed and the IC50 values were calculated. Results are the
mean (standard deviations not shown, ≤10%) of triplicates of two independent experiments.
efficient in inhibiting DNA synthesis in endothelial cells com-
pared to glioblastoma cells (Figure 3 and Supporting Information
Table S4). These results suggest an initial cytostatic effect of both
OSCi and AT in human glioblastoma cells.
Compound 1 with AT combinations did not enhance the
cytotoxic effects of each drug separately (data not shown). In
LN18 cells exposed to compound 2, a cytotoxicity-enhancing
effect of AT was seen at 2 and 4 μM compound 2. In LN229
cells exposed to 2 μM compound 2, a cytotoxicity-enhancing
effect was detected well below the IC50 values of both
inhibitors. In HCEC cells, only at the highest concentrations
of the OSCi (4 and 10 μM) was the AT effect on cell death
seen (Figure 4). In LN18 cells exposed to compound 3, the
addition of AT only decreased the survival of cells after
exceeding the IC50 of AT. In LN229 cells exposed to 2 μM
compound 3, the addition of 2.5 μM AT decreased cell survival
by 80%. With higher concentration of compound 3 in LN18
and in LN229 cells, only the cytotoxic effect of the increased
concentrations of the OSCi was seen. In HCEC cells no
Thus, for further experiments, combining an OSCi with AT,
we selected the most active OSCi compounds 1, 2, 3, and 4 and
the LN18 and LN229 cells as models of human glioblastoma
cells and the nontumoral HCEC endothelial cells as model of
nontumoral angiogenic brain-derived endothelial cells. Com-
pounds 1, 2, 3, and 4 at three different concentrations (2, 4, and
10 μM) were combined with three different concentrations of
AT (2.5, 5, and 10 μM) for 72 h because for this treatment
time, AT as single agent dose-dependently reduced cell survival
more in LN18 and LN229 glioblastoma cancer cells than in
nontumoral HCEC cells.
4992
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of the Alkyne OSCi Compounds 5 and 6
a
Reagents and conditions: (a) MeOMeNH·HCl, HOBT, NMM, EDCI, CH2Cl2, 0 °C to rt, quantitative; (b) NaH (55% in oil), MeI, DMF, 0 °C to
rt, 84%; (c) LAH, THF, 0−15 °C, quantitative; (d) Ph3P, CBr4, Et3N, CH2Cl2, 0 °C to rt, 62%; (e) (1) n-BuLi, THF, −78 °C; (2) DMPU; (3) 2-(2-
bromoethoxy)tetrahydro-2H-pyran, THF, −78 °C to rt, 40%; (f) PPTs, MeOH, 55 °C, 77%; (g) MeSO2Cl, pyridine, DMAP, CH2Cl2, 0 °C to rt,
quantitative; (h) TFA, CH2Cl2, 0 °C, quantitative; (i) 4-ClPhOCOCl, i-Pr2NEt, CH2Cl2, 0 °C to rt, quantitative; (k) (1) ethyl-(2-
hydroxyethyl)amine, MeOH, 65 °C; (2) HCl in dioxane, 59%; (l) (1) N-methylpropylamine, NaI, MeOH, 60 °C; (2) fumaric acid, EtOAc, EtOH,
76%.
a
Scheme 2. Synthesis of the Methylethylcyclohexyl OSCi 7
a
Reagents and conditions: (a) PtO2, H2 (1 atm), EtOH/9% CHCl3, 97%; (b) (1) (Boc)2O, Et3N, CH2Cl2, rt; (2) Ac2O, pyridine, CH2Cl2, rt; (3)
MeI, NaH (55% in oil), DMF, 0 °C to rt, 68%, three steps; (c) K2CO3, MeOH, rt, 86%; (d) MeSO2Cl, Et3N, CH2Cl2, 0 °C to rt, quantitative; (e) 4
M HCl in dioxane, rt, 92%; (f) (1) 4-trifluoromethylphenol, 1,1′-carbonyldiimidazol, CH Cl , rt; (2) (CH ) SO , Hunig's base, CH CN, 80 °C; (3)
̈
2
2
3
2
4
3
20, Hunig's base, CH CN, rt, 89%; (g) (1) ethyl-(2-hydroxyethyl)amine, DMA, 60 °C; (2) citric acid monohydrate, EtOAc, i-PrOH, n-pentane,
̈
3
62%.
a
Scheme 3. Synthesis of the Amide OSCi 10
a
Reagents and conditions: (a) NaIO4, RuCl3.H2O, CH3CN, CCl4, H2O, rt, 96%; (b) 4 M HCl in dioxane, rt, quantitative; (c) (1) HMDS, 145 °C;
(2) 4-ClPhOCOCl, i-Pr2EtN, THF, 0 °C to rt, 57%; (d) (1) (COCl)2/DMF, CH2Cl2, 0 °C to rt; (2) 4-hydroxypiperidine, CH2Cl2, 97%.
cytotoxicity-enhancing effect of compound 3 with AT was
observed (Figure 5). In LN18 cells exposed to compound 4 a
cytotoxicity-enhancing effect with AT was seen. In LN229 cells
exposed to 4 μM compound 4, a pronounced cytotoxicity-
enhancing effect with AT was demonstrated. In HCEC cells
only the OSCi inhibitory effect was apparent (Figure 6).
4993
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
a
Table 2. IC50 after 72 h of Cell Exposure to Compounds 1−10 and AT
IC50 (μM)
A549
CaCo2
HeLe
KB
LnCap
LNZ308
MCF-7
Me300
OVCAR
1
6.19
11.3
5.14
6.06
13.59
8.41
22.7
6.93
6.09
>30
>30
3.81
2.99
1.04
1.97
4.81
6.75
10.0
6.3
2.69
2.89
6.37
8.85
8.64
6.82
10.0
1.06
4.53
21.8
4.22
2.03
2.88
4.66
8.52
5.86
4.53
10.0
1.02
4.28
7.58
1.97
5.07
>30
>30
19.5
23.6
18.7
>30
4.62
>30
5.55
>30
14.4
5.26
4.7
25.7
6.77
5.58
8.3
13.9
>30
6.2
>30
>30
16.7
>30
>30
>30
>30
12.94
>30
>30
>30
2
3
4
4.4
13.4
26.2
18.9
>30
16.0
20.9
25.0
1.2
5
15.3
12.6
23.2
2.1
>30
>30
>30
20.7
12.3
>30
>30
6
7
8
9
2.67
>30
18.7
>30
>30
7.44
10
AT
a
Cells were treated with increasing concentrations (0, 2, 5, 10, 20, and 30 μM) of the OSCi compounds 1−10 and AT for 72 h. Then the MTT assay
was performed and the IC50 values were calculated. Results are the mean (sd not shown, ≤10% of the mean) of triplicates of two independent
experiments. IC50 values for LN18, LN229, and HCEC cells are provided in Table 1 and Figures 1 and 2.
Acetate is a precursor in the biosynthesis of cellular lipids,
including sterols, lipoproteins, and glycolipids; thus, we
evaluated the incorporation of radioactice acetate in trichloro-
acetic acid (TCA)precipitable high molecular weight cellular
molecules and/or acid insoluble lipids in glioblastoma cells
exposed to increasing concentrations of AT and OSCi 2 or 3
(Figure 7). AT dose-dependently decreased 14C-acetate incor-
poration in the LN229 cells and to a lesser extent in LN18 cells.
Both OSCi compounda 2 and 3 increased the 14C-acetate
incorporation in both cell lines. From the known different
mode of action of these molecules, this result was not un-
expected and we can propose the following hypothesis. Statins
inhibit at an early step of the biosynthesis of sterols, preventing
accumulation of insoluble radioactive intermediates, while OSC
inhibition would result in the accumulation of insoluble 2,3-
oxidosqualene25 and/or additional high molecular weight
molecules, such as lipoproteins. In the glioblastoma cells, the
increase in the incorporation of 14C-acetate induced by OSCi 2
was inhibited by AT addition, while for compound 3 the
addition of AT had only a marginal influence (Figure 7). Thus,
also in human glioblastoma cells different cellular pathways can
be selectively inhibited by an OSCi and a statin, as previously
shown for other human cancer cells.35
Then the evaluation of possible additivity or synergism of
OSCi and AT combinations was examined for combinations
statistically significant using first for screening purposes the
median effect model proposed by Chou.36 This method
proposes that a calculated combination index (CI) of ∼1
defines additivity and that CI < 1 defines synergism. First, the
CI factor was calculated to select combinations with possible
additive or synergistic effects (Supporting Information Table S5).
An additive effect was shown for a combination of AT with low
concentrations of OSCi compounds 2, 3, and 4 in LN18 cells and
OSCi compounds 2 and 3 in LN229 cells but only with high
concentrations of OSCi 2 in HCEC cells (Supporting
Information Table S5). Combination of AT with low concen-
tration of OSCi 3 in LN229 cells provided CI < 0.5. Thus, for this
combination, the computational method proposed by Lee et al.37
was used to determine a possible synergism. According to this
model, the interaction index (II) should be <1 and its confidence
interval should not include 1 to prove synergism; otherwise, the
combined effect is additive. For the combination of 3 with AT in
LN229 human glioblastoma cells, the confidence interval included
1 (Supporting Information Figure S1); thus, this combination did
not suggest a possible synergism. Therefore, the additive effect of
OSCi with AT in LN18 cells and LN229 cells does not appear to
be dependent on the chemical structure class of the compounds,
since these effects were observed for two different classes of OSCi
but may depend on the physicochemical and/or biochemical
characteristics, i.e., membrane permeability and hydrophobicity,
which were determined (Table 1) according to published pro-
cedures.38,39 From this information, it can be ruled out that
limited cell membrane permeability was responsible for the
observed effect, since the PAMPA (parallel artificial membrane
permeation assay) membrane permeability (Pe PAMPA) was
medium to high for all compounds (Table 1). The only
correlation we found was the correlation between the ratios of
their cytotoxicity potency for the human brain-derived endothelial
cells to the two human glioblastoma cells (IC50(HCEC)/
IC50(LN229) or IC50(HCEC)/IC50(LN18)) (Figure 8). For only
log D at pH 7.4 an inverse correlation between the IC50 ratios of
HCEC to LN18 and LN229 cells (Figure 8) was observed.
However, the log D of AT (0.98) exerted a large influence on the
plot. Thus, more polar OSCi compounds are more selective for
inhibiting LN229 or LN18 growth, compared to nontumoral
HCEC cells. It is interesting to note that the membrane fraction
(Table 1) measured in the PAMPA experiment did not correlate
with the selectivity ratio (IC50(HCEC)/IC50(LN229), r2 = 0.05)
or the selectivity ratio (IC50(HCEC)/IC50(LN18), r2 = 0.09).
Hence, it can be excluded that the difference in unspecific binding
to either of the cell membranes governs the observed selectivity.
This suggests that log D (pH 7.4) must be taken into account for
further development of selective anticancer therapeutic com-
pounds.
CONCLUSION
■
We have evaluated several chemical classes of our best
inhibitors of OSC, an enzyme downstream of the HMG-CoA
reductase in the mevalonate pathway. We show here that OSCi
compounds have the potential to decrease the growth of human
cancer cells of diverse tissue origin at micromolar concen-
trations, comparable to the concentrations of statins necessary
for antitumor effect. Human glioblastoma and brain-derived
endothelial cells, as a model of tumor-induced angiogenic endo-
thelial cells, were among the most sensitive cells. It has been
shown that the brain is the most cholesterol-rich organ of the
body and the intensive cholesterol metabolism in brain tumors
cells, higher than in nontumoral cells,40−44 suggests that the
4994
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Figure 1. Cytotoxic effects of a 72 h exposure of human glioblastoma and brain-derived endothelial cells to OSCi compounds 1−6. Human LN18,
LN229, and HCEC cells were exposed to increasing concentrations of compounds 1−6 for 72 h. Then the MTT assay was performed and IC50
values were calculated. Results are the mean sd of triplicates of two independent experiments. IC50 values are the mean (sd not shown, ≤10% of
the values of the means) of triplicates of two independent experiments.
cholesterol pathway is a possible target for a specific pharma-
cological treatment of these tumors, thus opening interesting
perspectives for the treatment of these very aggressive brain
tumors. Previous studies have suggested that statins may have
some antitumoral effect in brain cancers.45−50 A phase I−II trial
of lovastatin for anaplastic astrocytoma and glioblastoma
multiforme showed that high doses of lovastatin did not induce
toxicity to the central nervous system (CNS).51 However,
the concentrations of statins necessary to achieve anticancer
effects were shown to be 10−100 times higher than their
concentrations as cholesterol-lowering drugs, suggesting that
drug combinations must be developed. We found that not only
statins, targeting the HMG-CoA reductase, but also inhibitors
of the OSC decrease tumor cell growth and that their com-
bination enhances their antitumoral efficacy. As statins are
known to up-regulate HMG-CoA reductase expression and as
24(S),25-epoxycholesterol, a product of the OSC, decreases
HMG-CoA reductase activity and enhances its degradation,
OSC was postulated to be an interesting target in combination
with a statin, since this should result in the generation of a
4995
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Figure 2. Cytotoxic effects of a 72 h exposure of human glioblastoma and brain-derived endothelial cells to OSCi compounds 7−10 and AT. Human
LN18, LN229, and HCEC cells were exposed to increasing concentrations of compounds 7−10 and AT for 72 h. Then the MTT assay was
performed and IC50 values were calculated. Results are the mean sd of triplicates of two independent experiments. IC50 values are the mean
(sd not shown, ≤10% of the values of the means) of triplicates of two independent experiments.
Figure 3. DNA synthesis in glioblastoma and endothelial cells exposed to OSCi or AT. Human LN18, LN229, and HCEC cells were exposed for
24 h to increasing concentrations of OSCi compounds 2−4 and AT. Then [3H]T incorporation was performed to evaluate DNA synthesis. Results
◇
□
▲
△
are the mean of triplicates (sd not shown, <10% of the values of the means): , 2; , 3; , 4; , AT.
4996
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Figure 4. Cytotoxic effects of a 72 h cell exposure to 2−AT
combination. Human LN18 (A), LN229 (B), and HCEC (C) cells
were treated with 2, 4, and 10 μM OSCi 2 in combination with 2.5, 5,
and 10 μM AT for 72 h. Then the MTT assay was performed. Results
Figure 5. Cytotoxic effects of a 72 h cell exposure to 3−AT
combination. Human LN18 (A), LN229 (B), and HCEC (C) cells
were treated with 2, 4, and 10 μM OSCi 3 in combination with 2.5, 5,
and 10 μM AT for 72 h. Then the MTT assay was performed. Results
are the mean
sd of triplicates of two independent experiments.
are the mean
sd of triplicates of two independent experiments.
Treated cells were compared to untreated cells using the Student’s
t-test: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001.
Treated cells were compared to untreated cells using the Student’s
t-test: ∗, p < 0.05; ∗∗, p < 0.01.
received. All reactions were followed by TLC (TLC plates F254,
Merck). Proton and carbon NMR spectra were obtained on a Bruker
300, 400, or 600 MHz instrument. The respective carbon NMR
spectra were 75, 100, and 150 MHz. The chemical shifts (δ in ppm)
are reported relative to tetramethylsilane as internal standard. NMR
abbreviations are as follows: s, singlet; d, doublet; t, triplet; q,
quadruplet; quint, quintuplet; m, multiplet; br, broadened. Purity was
analyzed by H NMR. In the H NMR spectra, the acidic protons of
the fumarate, HCl, and citrate salts were not seen unless listed. Mass
spectra were recorded on an SSQ7000 (Finnigan-MAT) spectrometer
for electron impact ionization. LC−MS (liquid chromatography−mass
spectrometry) data were recorded on an Agilent 1290 LC with CTC
PAL coupled to Agilent 6520 QTOF. HPLC was performed on
Zorbax Eclipse Plus C18 2.1 mm × 50 mm, 1.8 μm (Agilent catalog
no. 959757-902). Column temperature was 55°C. A gradient from 5%
to 99% of eluent B in eluent A was used at a flow rate of 1 mL/min.
Eluent B consisted of 80% acetonitrile/20% isopropanol/0.01% formic
acid. Eluent A was 0.02% formic acid in water. UV detector was DAD,
220 5 nm or 265 35 nm. All compounds were characterized by 1H
NMR and MS and final compounds with microanalyses. Elemental
analyses were performed by Solvias AG (Mattenstrasse, Postfach, CH-
4002 Basel, Switzerland). Results of elemental analyses were within
0.4% of theoretical values. Solvents (1H NMR) and H2O content (Karl
Fischer determination) were incorporated in the formula. Melting
synergistic, self-limited, negative regulatory loop to deplete
cancer cells from cholesterol. In support of this hypothesis
additive effects of OSCi in combination with AT were
demonstrated in human glioblastoma cells while maintaining
selectivity against endothelial cells. Our results also suggest that
the lipophilicity of the OSCi must be taken into account for
selectivity, meaning that a more polar OSCi is more selective
for inhibiting the growth of cancer cells and for further
development of anti-cholesterol compounds in the context of
cancer therapy. In summary, we have shown that not only
statins targeting HMG-CoA reductase but also inhibitors of
OSC decrease tumor cell growth, suggesting new therapeutic
opportunities of combined anti-cholesterol agents for dual
treatment of glioblastoma.
1
1
EXPERIMENTAL SECTION
■
Chemistry. General. The syntheses and characterizations of
compounds 1−4, 8, and 917−20,29 have been previously described.
The detailed synthesis and characterization of intermediate com-
pounds 12−16 and 18−23 are provided as Supporting Information.
Reactions were carried out under an atmosphere of argon. Solvents
and reagents were obtained from commercial sources and were used as
4997
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
1H NMR (400 MHz, DMSO-d6, rt) δ 9.36 (broad s, 1H), 7.43 and
7.15 (AA′-BB′-system, JAB = 8.8 Hz, 4H), 5.32 (broad s, 1H), 3.9 and
3.8 (broad m, 1H, 3:1 mixture of rotamers), 3.70 (broad s, 2H), 3.2−3.0
(broad s, 6H), 2.88 and 2.79 (broad s, 3H, 3:1 mixture of rotamers), 2.23
(m, 3H), 1.95 (m, 2H), 1.85−1.55 (m, 6H), 1.39 (m, 2H), 1.18 (m, 3H).
1H NMR (400 MHz, DMSO-d6, 120 °C) δ 7.33−7.40 (m, 2H), 7.08−
7.16 (m, 2H), 3.79−3.90 (m, 1H), 3.76 (t, J = 5.4 Hz, 2H), 3.06−3.22
(m, 6H), 2.86 (s, 3H), 2.25 (td, J = 6.9, 2.0 Hz, 2H), 2.15−2.22 (m, 1H),
1.93−2.04 (m, 2H), 1.77−1.88 (m, 2H), 1.68−1.75 (m, 2H), 1.60 (qd,
J = 12.5, 3.4 Hz, 2H), 1.34−1.47 (m, 2H), 1.25 (t, J = 7.2 Hz, 3H). 13C
NMR (151 MHz, DMSO-d6) δ 156.6, 153.2, 150.1, 129.1, 123.7, 85.0,
78.4, 55.2, 53.8, 50.6, 47.6, 40.0, 32.2, 28.8, 28.2, 28.0, 22.3, 15.4. MS m/
z: 421.2 (M + H+, 1Cl). Anal. (C23H33ClN2O3·1.08HCl·1.6H2O) C, H,
N, Cl.
trans-Methyl-{4-[5-(methylpropylamino)pent-1-ynyl]cyclohexyl}-
carbamic acid 4-Chloro-Phenyl Ester Fumarate (6). A solution of
trans-methanesulfonic acid 5-(4-methylaminocyclohexyl)pent-4-ynyl
ester trifluoroacetate (16) (0.84 mg, 1.96 mmol) and N-methylpropyl-
amine (3.02 mL, 29.44 mmol, 15 equiv) in MeOH (15 mL) was
treated with sodium iodide (0.03 g, 0.2 mmol, 0.1 equiv), and the
mixture was stirred overnight at 60 °C. The appropriate portion was
extracted with aqueous saturated NaHCO3 solution/Et2O (3×) and
aqueous 10% NaCl solution, and the organic phases were dried over
Na2SO4, filtered, and evaporated. Purification by flash column
chromatography on silica gel (CH2Cl2, then CH2Cl2/MeOH, 95:5
to 9:1) gave trans-methyl-{4-[5-(methylpropylamino)pent-1-ynyl]-
cyclohexyl}carbamic acid 4-chlorophenyl ester (0.75 g, 94%). MS
m/z: 405.5 (M + H+, 1Cl).
The amine (0.7 g, 1.73 mmol) in EtOAc and fumaric acid (0.21 g,
1.81 mmol, 1.05 equiv) in EtOH were mixed and evaporated with
EtOAc (3×). Slow (24 h) cooling of an EtOAc solution to 4 °C gave
crystals. Filtration and washing with a small amount of EtOAc gave the
desired product as white crystals (0.73 g, 81%). Mp: 95−98 °C (dec).
1H NMR (300 MHz, DMSO-d6) δ 7.43 and 7.15 (AA′-BB′-system,
JAB = 8.8 Hz, 4H), 6.55 (s, 2H), 3.82 (broad s, 1H), 2.88 and 2.78 (s,
3H, 3:1 mixture of rotamers), 2.55−2.4 (broad m, 4H), 2.29 (s, 3H),
2.16 (t, J = 6.8 Hz, 3H), 1.92 (m, 2H), 1.75−1.25 (m, 10H), 0.85 (t,
1
J = 7.3 Hz, 3H). H NMR (300 MHz, DMSO-d6 + D2O) δ 7.43 and
7.15 (AA′-BB′-system, JAB = 8.8 Hz, 4H), 6.49 (s, 2), 3.80 (broad s,
1H), 2.95−2.70 (m, 7H), 2.57 (s, 3H), 2.20 (t, J = 6.0 Hz, 3H), 1.95
(m, 2H), 1.80−1.50 (m, 8H), 1.50−1.30 (m, 2H), 0.89 (t, J = 7.4 Hz,
3H). 13C NMR (101 MHz, DMSO-d6, 120 °C) δ 165.9, 153.3, 150.5,
133.8, 129.1, 128.9, 123.2, 84.0, 79.9, 59.1, 55.8, 55.2, 41.7, 32.6, 29.1,
28.8, 28.4, 26.4, 19.8, 16.0, 11.4. MS m/z: 405.5 (M + H+, 1Cl). Anal.
(C23H33ClN2O2·C4H4O4·0.15H2O) C, H, N, Cl.
trans-[2-(4-{[Ethyl-(2-hydroxyethyl)amino]methyl}cyclohexyl)-
ethyl]methylcarbamic Acid 4-Trifluoromethylphenyl Ester Citrate
(7). A solution of trans-methanesulfonic acid 4-{2-[methyl-(4-
trifluoromethylphenoxycarbonyl)amino]ethyl}cyclohexylmethyl ester
(21) (0.58, 1.33 mmol) and ethyl-(2-hydroxyethyl)amine (1.29 mL,
13.26 mmol, 10 equiv) in DMA (6 mL) was stirred for 3 days at 60 °C.
The appropriate portion was extracted with aqueous 0.25 M NaOH/
Et2O (3×), water, and aqueous saturated NaCl solution, and the organic
phases were dried over MgSO4, filtered, and evaporated. Purification by
flash column chromatography on silica gel (CH2Cl2/MeOH, 95:5) gave
trans-[2-(4-{[ethyl-(2-hydroxyethyl)amino]methyl}cyclohexyl)ethyl]-
methylcarbamic acid 4-trifluoromethylphenyl ester (0.38 g, 67%). MS
m/z: 431.3 (M + H+).
The amine (2.8 g, 6.50 mmol) and citric acid monohydrate (1.37 g,
6.50 mmol) were dissolved in hot EtOAc (28 mL)/i-PrOH (1.6 mL),
then slowly (48 h) cooled to −20 °C. After 24 h additional n-pentane
(10 mL) was mixed in to give after filtration and washing with a small
amount of n-pentane the desired product as white crystals (3.76 g,
93%). Mp: 64 °C (slow dec). 1H NMR (400 MHz, DMSO-d6,
120 °C) δ 8.08−10.06 (br s, <3H) 7.61−7.78 (m, 2H), 7.23−7.39 (m,
2H), 3.57 (t, J = 6.0 Hz, 2H), 3.38 (dd, J = 8.1, 7.3 Hz, 2H), 2.75−2.86
(m, 4H), 2.59 (d, JAB = 15.3 Hz, 2H), 2.68 (d, JAB = 15.3 Hz, 2H), 2.55
(d, J = 6.7 Hz, 2H), 2.60 (br s, 3H), 1.72−1.86 (m, 4H), 1.42−1.59
(m, 3H), 1.19−1.33 (m, 1H), 1.07 (t, J = 7.1 Hz, 3H), 0.82−1.04
(m, 4H). 13C NMR (101 MHz, DMSO-d6, 120 °C) δ ppm 170.8,
Figure 6. Cytotoxic effects of a 72 h cell exposure to 4−AT
combination. Human LN18 (A), LN229 (B), and HCEC (C) cells
were treated with 2, 4, and 10 μM OSCi 4 in combination with 2.5, 5,
and 10 μM AT for 72 h. Then the MTT assay was performed. Results
are the mean
sd of triplicates of two independent experiments.
Treated cells were compared to untreated cells using the Student’s
t-test: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001.
points (uncorrected) were determined using a Buchi 510 apparatus
̈
and are uncorrected. Silica gel 60 (0.04−0.063 mm, Merck) was used
for flash chromatography. The purities of final compounds 5, 6, 7, and
10 were analyzed by LC−MS and by elemental analysis and were
found to be above 95%.
trans-(4-{5-[Ethyl-(2-hydroxyethyl)amino]pent-1-ynyl}-
cyclohexyl)methylcarbamic Acid 4-Chlorophenyl Ester Hydrochlor-
ide (5). A solution of trans-methanesulfonic acid 5-(4-methyl-
aminocyclohexyl)pent-4-ynyl ester trifluoroacetate (16) (0.57 g, 1.33
mmol) and ethyl-(2-hydroxyethyl)amine (1.29 mL, 13.27 mmol, 10
equiv) in MeOH (13 mL) was stirred overnight at 65 °C. The
appropriate portion was extracted with aqueous saturated NaHCO3
solution/Et2O (3×) and aqueous 10% NaCl solution, and the organic
phases were dried over Na2SO4, filtered, and evaporated. Purification
by flash column chromatography on silica gel (CH2Cl2, then CH2Cl2/
MeOH, 95:5 to 9:1) gave trans-(4-{5-[ethyl-(2-hydroxyethyl)amino]-
pent-1-ynyl}cyclohexyl)methylcarbamic acid 4-chlorophenyl ester
(0.35 g, 63%). MS m/z: 421.4 (M + H+, 1Cl). The amine was
dissolved in dioxane (1 mL) and treated with 4 M HCl in dioxane
(0.17 mL, 0.69 mmol, 1.1 equiv), filtered, and evaporated. Dissolving
in CH2Cl2 and evaporation (3×) gave after drying in high vacuum the
desired compound as a yellow viscous oil (0.36 g, 59% over two steps).
4998
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Figure 7. High molecular weight and acid-insoluble lipid synthesis in glioblastoma cells exposed to OSCi and AT. Human LN18 or LN229 cells were
exposed for 24 h to increasing concentrations of AT or to 2 μM OSCi 2 or 3 in combination with increasing concentration of AT. Then 14C-acetate
incorporation was performed to evaluate high molecular weight lipid synthesis. Results are the mean of triplicates. Treatments were compared using
○
○
○○
, p < 0.01; ∗∗∗ and
the Student’s t-test: ∗, treated cells vs untreated cells; , AT−OSCi combination vs OSCi alone; ∗ and , p < 0.05; ∗∗ and
○○○
, p < 0.001.
154.6, 153.1, 126.4 (q, J = 3.8), 122.1, 72.0, 60.6, 57.9, 56.1, 49.0, 47.0,
43.7, 35.2, 35.0, 34.6, 32.2, 30.9, 10.4. MS m/z: 431.3 (M + H+). Anal.
(C22H33F3N2O3·C6H8O7·0.06H2O) C, H, N, F.
A solution of trans-4-{2-[(4-chlorophenoxycarbonyl)methylamino]-
ethyl}cyclohexanecarboxylic acid (7.46 g, 21.95 mmol) in CH2Cl2
(200 mL) was treated at room temperature with 2 drops of DMF,
followed by oxalyl chloride (0.28 mL, 2.08 mmol, 1.1 equiv) within
5 min, and stirring was continued for 90 min. The solution was
evaporated, redissolved in CH2Cl2 (200 mL), and added to a cooled
solution (0 °C) of 4-hydroxypiperidine (6.66 g, 65.86 mmol, 3 equiv)
and Et3N (15.30 mL, 109.76 mmol, 5 equiv) in CH2Cl2 (120 mL).
The mixture was kept at room temperature for 1.5 h and then parti-
tioned between Et2O (3×) and aqueous 10% KHSO4 solution. The
organic phases were washed once with aqueous 10% NaCl solution,
dried over Na2SO4, filtered, and evaporated to give the desired product
(9.00 g, 97%). 1H NMR (300 MHz, DMSO-d6) δ 7.43 and 7.13 (AA′-
BB′-system, two rotamers, 4H), 4.71 (d, J = 10.5 Hz, 1H), 3.82 (m,
1H), 3.66 (m, 2H), 3.39 (m, 1H), 3.30 (m, 1H), 3.15 (m, 1H), 3.00
and 2.88 (s, 3H, 4:5 mixture of rotamers), 2.94 (m, 1H), 1.84−1.58
(m, 6H), 1.56−0.90 (m, 10H). 1H NMR (400 MHz, DMSO-d6, 120 °C)
δ 7.30−7.45 (m, 2H), 7.03−7.17 (m, 2H), 4.21 (d, J = 4.3 Hz, 1H),
3.74−3.83 (m, 2H), 3.66−3.74 (m, 1H), 3.31−3.40 (m, 2H),
3.03−3.14 (m, 2H), 2.95 (s, 3H), 2.49−2.54 (m, 1H), 1.75−1.84 (m,
2H), 1.61−1.75 (m, 4H), 1.45−1.55 (m, 2H), 1.22−1.45 (m, 5H),
0.98−1.11 (m, 2H). 13C NMR (101 MHz, DMSO-d6, 120 °C) δ
173.4, 153.5, 150.5, 129.1, 123.2, 65.7, 46.8, 34.7, 34.5, 34.2, 31.9, 29.1.
trans-{2-[4-(4-Hydroxypiperidine-1-carbonyl)cyclohexyl]ethyl}-
methylcarbamic Acid 4-Chlorophenyl Ester (10). A mixture of trans-
4-(2-methylaminoethyl)cyclohexanecarboxylic acid·HCl (23) (9.75 g,
43.9 mmol) in hexamethyldisilazane (115 mL, 552 mmol, 12 equiv)
was heated under reflux to 145 °C for 2.5 h. The solution was
evaporated. The residue was suspended in THF (200 mL) and treated
with 4-chlorophenyl chloroformate (6.77, 48.4 mmol, 1.1 equiv) at
0 °C, and the mixture was stirred at room temperature overnight.
Water (10 mL) was added at room temperature followed by aqueous
1 M NaOH (110 mL, 2.5 equiv). Stirring was continued for 1 h at
room temperature. The organic solvent was evaporated and the
residue partitioned between n-hexane (3×) and H2O. The water phase
was acidified with 1 M HCl and extracted with EtOAc (3×). This
organic phase was dried over Na2SO4, filtered, and evaporated to yield
the crude product. This residue was again partitioned between Et2O
(3×) and aqueous saturated NaHCO3 solution. The aqueous phase
was acidified with aqueous 1 N HCl and extracted with EtOAc (3×).
Drying over Na2SO4 and evaporation gave pure trans-4-{2-[(4-
chlorophenoxycarbonyl)methylamino]ethyl}cyclohexanecarboxylic
acid as a brown oil (8.56 g, 57%). MS m/z: 338.1 (M − H−, 1Cl).
4999
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Figure 8. Correlation between log D and IC50. (A) Correlation of the selectivity ratio IC50(HCEC)/IC50(LN229) with log D at pH 7.4. Black line
represents the regression line (r2 = 0.91; IC50(HCEC)/IC50(LNZ229) = (4.23 0.17) − (log D)(1.33 0.15), and dotted lines represent the 95%
confidence interval. Cross-validation coefficient (leave-one-out method) is q2 = 0.88. (B) Correlation of the selectivity ratio IC50(HCEC)/
IC50(LN18) with log D at pH 7.4. Black line represents the regression line (r2 = 0.69; IC50(HCEC)/IC50(LNZ229) = (4.13 0.32) − (log D)-
(0.87 0.21), and dotted lines represent the 95% confidence interval. Cross-validation coefficient (leave-one-out method) is q2 = 0.66.
MS m/z: 423.2 (M + H+, 1Cl). Anal. (C22H31ClN2O4·0.55H2O·
0.12C2H6O) C, H, N, Cl.
and neck cancer cells were from M. Barberi Heyob, University of
Nancy, France. The human-brain-derived HCEC microvascular endo-
thelial cell line was a kind gift of D. Staminirovic (Ottawa, Canada). All
cells were grown in Dulbecco’s modified Eagle medium (DMEM)
medium containing 4.5 g/L glucose, 10% heat-inactivated fetal calf
serum (FCS), and penicillin/streptomycin (all cell culture reagents
were obtained from Invitrogen, Basel, Switzerland). Unless otherwise
specified, cells were grown for 24 h in 48-well plates (Costar, Corning,
NY, U.S.). Then the compounds were added for the indicated times
and concentrations. Compounds 2, 3, 5, 6, 7, 8, and 9 were dissolved
at 50 mM in water and compounds 1, 4, 10, and AT in DMSO, and
then they were diluted to 1 mM in cell culture medium containing
FCS immediately before use.
Determination of Cytotoxicity. Following exposure to the
compounds, cell viability was assessed using the MTT assay (3-(4,5-
dimethyl-2-thiazoyl)-2,5-diphenyltetrazolium bromide, Sigma-Aldrich,
200 μg/mL final concentration), essentially as previously described.52
Absorbance at 540 nm was measured in a multiwell plate reader (iEMS
Reader, Labsystems, Bioconcept, Allschwil, Switzerland), and the
absorbance values of treated cells were compared to the absorbance
values of untreated cells. Experiments were conducted in triplicate
wells and repeated twice. The mean standard deviation (sd) values
were calculated.
Evaluation of DNA Synthesis. Following cell exposure to the com-
ponds for 48 h, tritiated thymidine ([3H]T) (Amersham-Pharmacia,
Glattbrugg, Switzerland, 400 nCi/mL final concentration) was added
to the cells for 2−4 h. Then the cell layers were precipitated with 10%
TCA and dissolved in 0.1% sodium dodecyl sulfate (SDS) in 0.1 N
NaOH, and scintillation cocktail (Optiphase HI-Safe, PerkinElmer,
Beaconsfield, U.K.) was added. Radioactivity was counted with a
β-counter (WinSpectra, Wallac, Germany). The radioactivity counts of
treated cells were compared to the radioactivity counts of untreated
cells. Experiments were conducted in triplicate wells, and the mean
sd values were calculated.
Acetate Incorporation. Cells were seeded in a 48-well plate for
24 h and exposed to the chemicals for 24 h. Then [14C]sodium acetate
(American Radiolabeled Chemicals, St. Louis, MO, U.S.; 10 nCi/mL
final concentration) was added to the cells for 7.5 h. The cell layers
were precipitated with 10% TCA and dissolved in 0.1% SDS in 0.1 N
NaOH and scintillation cocktail (Optiphase HI-Safe). Radioactivity
was counted in a β-counter (WinSpectra). The radioactivity counts of
treated cells were compared to the radioactivity counts of untreated
cells. Experiments were conducted in triplicate wells, and the mean
sd values were calculated.
Physicochemical Characterization of the Inhibitors. log D
Determination. The high-throughput assay method is derived from
the conventional “shake flask” method. The compound of interest is
distributed between a 50 mM aqueous 3-[[1,3-dihydroxy-2-(hydroxy-
methyl)propan-2-yl]amino]-2-hydroxypropane-1-sulfonic acid
(TAPSO) buffer at pH 7.4 and 1-octanol. The distribution coefficient
is then calculated from the difference in concentration in the aqueous
phase before and after partitioning and the volume ratio of the two
phases. A detailed description of the method can be found elsewhere.38
PAMPA. For PAMPA, a “sandwich” is formed from a 96-well filter
plate and a 96-well in-house-made Teflon plate such that each well is
divided into two chambers: donor at the bottom and acceptor at the
top, separated by a microfilter coated with a 10% (w/v) egg-phos-
phatidylcholine and 0.5% (w/v) cholesterol dissolved in dodecane.
The distribution of a compound between the top compartment, the
artificial membrane, and the bottom compartment is then determined.
A detailed description of the method can be found elsewhere.39
Biology. Evaluation of OSC Inhibition. Inhibition of human liver
microsomal 2,3-OSC was used to evaluate the inhibitory potency of
the new molecules, essentially as described previously.19 Briefly, OSC
activity was measured in human liver microsomes in sodium phosphate
buffer−EDTA−dithtiothreitol, pH 7.4. The inhibitors were dissolved
in DMSO and diluted to the desired concentration in the buffer. [14C]
́
R,S-MOS (12.8 mCi/mmol, 1 mM final concentration, 20 nCi/iL) in
phosphate buffer/1% BSA was mixed with the microsomes, and the
reaction proceeded for 1 h at 37 °C. The reaction was stopped by the
addition of 10% KOH/MeOH. Then n-hexane/Et2O containing non-
radioactive MOS and lanosterol as carriers was added. After shaking,
the upper phase was evaporated and suspended in n-hexane/Et2O and
applied to a silica gel plate for chromatographic separation using
n-hexane/Et2O as the eluent. Radioactive MOS and lanosterol were
detected on the silica gel plate using a phosphor imager (Molecular
Dynamics). The ratio of MOS to lanosterol was determined from the
radioactive bands in order to determine the yield of the reaction and
OSC inhibition.
Cells and Cell Treatments. Human A549 lung, CaCo2 colon, HeLa
cervix, LnCap prostate, MCF-7 breast, OVCAR ovarian, and LN229
glioblastoma cancer cells are available from the ATCC (American
Tissue Culture Collection, Manassas, VA, U.S.). Human LN18 and
LNZ308 glioblastoma cells were a kind gift from AC Diserens, CHUV,
Lausanne, Switzerlande. Human Me300 melanoma cells were from
D. Rimoldi, Ludwig Institute, Lausanne branch, Switzerland, and KB head
5000
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
Evaluation of Results. Each experiment was repeated in triplicate
wells at least twice. The mean and standard deviation values were
calculated, and statistical significance was evaluated using the Student’s
t-test.
Squared cross-validation coefficients according to the leave-one-out
method (q2) and the squared correlation coefficients (r2) were
calculated with SIMCA, version 12 (Umetrics Inc., Umea, Sweden).
REFERENCES
■
(1) Maxfield, F. R.; Tabas, I. Role of cholesterol and lipid
organization in disease. Nature 2005, 438, 612−621.
(2) Raposo, G.; Stenmark, H. Membranes and organelles. Curr. Opin.
Cell Biol. 2008, 20, 357−359.
(3) Nissen, S. E.; Tuzcu, E. M.; Schoenhagen, P.; Crowe, T.; Sasiela,
W. J.; Tsai, J.; Orazem, J.; Magorien, R. D.; O’Shaughnessy, C.; Ganz,
P. Statin therapy, LDL cholesterol, C-reactive protein, and coronary
artery disease. N. Engl. J. Med. 2005, 352, 29−38.
(4) Mills, E. J.; Rachlis, B.; Wu, P.; Devereaux, P. J.; Arora, P.; Perri,
D. Primary prevention of cardiovascular mortality and events with
statin treatments: a network meta-analysis involving more than 65,000
patients. J. Am. Coll. Cardiol. 2008, 52, 1769−1781.
(5) Demierre, M. F.; Higgins, P. D.; Gruber, S. B.; Hawk, E.;
Lippman, S. M. Statins and cancer prevention. Nat. Rev. Cancer 2005,
5, 930−942.
(6) Chan, K. K.; Oza, A. M.; Siu, L. L. The statins as anticancer
agents. Clin. Cancer Res. 2003, 9, 10−19.
(7) Cemeus, C.; Zhao, T. T.; Barrett, G. M.; Lorimer, I. A.;
Dimitroulakos, J. Lovastatin enhances gefitinib activity in glioblastoma
cells irrespective of EGFRvIII and PTEN status. J. Neuro-Oncol. 2008,
90, 9−17.
(8) Guo, D.; Prins, R. M.; Dang, J.; Kuga, D.; Iwanami, A.; Soto, H.;
Lin, K. Y.; Huang, T. T.; Akhavan, D.; Hock, M. B.; Zhu, S.; Kofman,
A. A.; Bensinger, S. J.; Yong, W. H.; Vinters, H. V.; Horvath, S.;
Watson, A. D.; Kuhn, J. G.; Robins, H. I.; Mehta, M. P.; Wen, P. Y.;
DeAngelis, L. M.; Prados, M. D.; Mellinghoff, I. K.; Cloughesy, T. F.;
Mischel, P. S. EGFR signaling through an Akt-SREBP-1-dependent,
rapamycin-resistant pathway sensitizes glioblastomas to antilipogenic
therapy. Sci. Signaling 2009, 2, ra82.
ASSOCIATED CONTENT
■
S
* Supporting Information
Detailed protocols for chemical synthesis and characterization
of compounds 12−16 and 18−23; in vitro inhibitory stability
and chemical stability of OSCi 1−10; cytotoxicity of
compounds 1−10 and AT after 24−72 h of cell exposure;
IC50 values for inhibition of DNA synthesis after 24 h of
exposure of cells to compounds 1−10 and AT; evaluation of
synergism or additivity of the effects of combination of
compounds with AT in human endothelial and glioblastoma
cells; evaluation of synergism of the effects of combination of
compound 3 with AT in human LN229 glioblastoma cells. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*For L.J.-J.: phone, +41 21 314 7173; fax, +41 21 314 7115;
e-mail, lucienne.juillerat@chuv.ch. For J.D.A.: phone, +41 61 688
7738; fax, +41 61 688 8367; e-mail, johannes.aebi@roche.com.
(9) van de Donk, N. W.; Kamphuis, M. M.; Lokhorst, H. M.; Bloem,
A. C. The cholesterol lowering drug lovastatin induces cell death in
myeloma plasma cells. Leukemia 2002, 16, 1362−1371.
Notes
The authors declare no competing financial interest.
(10) Graaf, M. R.; Richel, D. J.; van Noorden, C. J.; Guchelaar, H. J.
Effects of statins and farnesyltransferase inhibitors on the development
and progression of cancer. Cancer Treat. Rev. 2004, 30, 609−641.
(11) Vincent, L.; Chen, W.; Hong, L.; Mirshahi, F.; Mishal, Z.;
Mirshahi-Khorassani, T.; Vannier, J. P.; Soria, J.; Soria, C. Inhibition of
endothelial cell migration by cerivastatin, an HMG-CoA reductase
inhibitor: contribution to its anti-angiogenic effect. FEBS Lett. 2001,
495, 159−166.
ACKNOWLEDGMENTS
■
We sincerely thank Dr. Jean Ackermann, Dr. Christian Jenny,
Dr. Francois Montavon, Felix Gruber, Marie-Paule Imhoff,
Isabelle Kaufmann, Fritz Koch, Hans-Jakob Krebs, and Anke
Kurt for their excellent and dedicated assistance; Dr. Olivier
Morand, Gisela Flach, Corinne Handschin, and Elisabeth Von
Der Mark for the OSC inhibition data; Dr. Caroline Wyss
Gramberg, Dr. Inken Plitzko, Dr. Josef Schneider, and Christian
Bartelmus for spectroscopic determination and analysis;
Isabelle Parilla, Bjoern Wagner, and Severin Wendelspiess for
determining the MDO data.
(12) Park, H. J.; Kong, D.; Iruela-Arispe, L.; Begley, U.; Tang, D.;
Galper, J. B. 3-Hydroxy-3-methylglutaryl coenzyme A reductase
inhibitors interfere with angiogenesis by inhibiting the geranylger-
anylation of RhoA. Circ. Res. 2002, 91, 143−150.
(13) Weis, M.; Heeschen, C.; Glassford, A. J.; Cooke, J. P. Statins
have biphasic effects on angiogenesis. Circulation 2002, 105, 739−745.
(14) Dulak, J.; Jozkowicz, A. Anti-angiogenic and anti-inflammatory
effects of statins: relevance to anti-cancer therapy. Curr. Cancer Drug
Targets 2005, 5, 579−594.
(15) Jakobisiak, M.; Golab, J. Statins can modulate effectiveness of
antitumor therapeutic modalities. Med. Res. Rev. 2010, 30, 102−135.
(16) Perchellet, J. P.; Perchellet, E. M.; Crow, K. R.; Buszek, K. R.;
Brown, N.; Ellappan, S.; Gao, G.; Luo, D.; Minatoya, M.; Lushington,
G. H. Novel synthetic inhibitors of 3-hydroxy-3-methylglutaryl-
coenzyme A (HMG-CoA) reductase activity that inhibit tumor cell
proliferation and are structurally unrelated to existing statins. Int. J.
Mol. Med. 2009, 24, 633−643.
(17) Morand, O. H.; Aebi, J. D.; Dehmlow, H.; Ji, Y. H.; Gains, N.;
Lengsfeld, H.; Himber, J. Ro 48-8.071, a new 2,3-oxidosqualene:la-
nosterol cyclase inhibitor lowering plasma cholesterol in hamsters,
squirrel monkeys, and minipigs: comparison to simvastatin. J. Lipid
Res. 1997, 38, 373−390.
ABBREVIATIONS USED
■
AT, atorvastatin; Boc, tert-butyloxycarbonyl; BSA, bovine
serum albumin; BuLi, butyllithium; CI, combination index;
CNS, central nervous system; DMEM, Dulbecco’s modified
Eagle Mmedium; EDCI, N-(3-dimethylaminopropyl)-N′-ethyl-
carbodiimide hydrochloride; FCS, fetal calf serum; HCEC,
human cerebral endothelial cell; HMG-CoA, 3-hydroxy-3-
methylglutaryl coenzyme A; HOBT, 1-hydroxybenzotriazole
hydrate; [3H]T, tritiated thymidine; IC50, concentration
inhibiting 50% of enzymatic activity; II, interaction index;
LN18, LN229, human glioblastoma cells; MOS, monooxidos-
qualene; MTT, 3-(4,5-dimethyl-2-thiazoyl)-2,5-diphenyltetra-
zolium bromide; OSC, oxidosqualene cyclase; OSCi, oxidos-
qualene cyclase inhibitor; PAMPA, parallel artificial membrane
permeation assay; Pe, permeability; i-Pr2NEt, N-ethyldiisopro-
pylamine; SDS, sodium dodecyl sulfate; TAPSO, 3-[[1,3-
dihydroxy-2-(hydroxymethyl)propan-2-yl]amino]-2-hydroxy-
propane-1-sulfonic acid; TCA, trichloroacetic acid
(18) Balliano, G.; Dehmlow, H.; Oliaro-Bosso, S.; Scaldaferri, M.;
Taramino, S.; Viola, F.; Caron, G.; Aebi, J.; Ackermann, J.
Oxidosqualene cyclase from Saccharomyces cerevisiae, Trypanosoma
cruzi, Pneumocystis carinii and Arabidopsis thaliana expressed in yeast: a
model for the development of novel antiparasitic agents. Bioorg. Med.
Chem. Lett. 2009, 19, 718−723.
5001
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002

Journal of Medicinal Chemistry
Article
(19) Dehmlow, H.; Aebi, J. D.; Jolidon, S.; Ji, Y. H.; von der Mark, E.
M.; Himber, J.; Morand, O. H. Synthesis and structure−activity studies
of novel orally active non-terpenoic 2,3-oxidosqualene cyclase
inhibitors. J. Med. Chem. 2003, 46, 3354−3370.
(20) Lenhart, A.; Reinert, D. J.; Aebi, J. D.; Dehmlow, H.; Morand,
O. H.; Schulz, G. E. Binding structures and potencies of oxidosqualene
cyclase inhibitors with the homologous squalene−hopene cyclase. J.
Med. Chem. 2003, 46, 2083−2092.
(21) Rowe, A. H.; Argmann, C. A.; Edwards, J. Y.; Sawyez, C. G.;
Morand, O. H.; Hegele, R. A.; Huff, M. W. Enhanced synthesis of the
oxysterol 24(S),25-epoxycholesterol in macrophages by inhibitors of
2,3-oxidosqualene:lanosterol cyclase: a novel mechanism for the
attenuation of foam cell formation. Circ. Res. 2003, 93, 717−725.
(22) Telford, D. E.; Lipson, S. M.; Barrett, P. H.; Sutherland, B. G.;
Edwards, J. Y.; Aebi, J. D.; Dehmlow, H.; Morand, O. H.; Huff, M. W.
A novel inhibitor of oxidosqualene:lanosterol cyclase inhibits very low-
density lipoprotein apolipoprotein B100 (apoB100) production and
enhances low-density lipoprotein apoB100 catabolism through marked
reduction in hepatic cholesterol content. Arterioscler., Thromb., Vasc.
Biol. 2005, 25, 2608−2614.
(36) Chou, T. C. Theoretical basis, experimental design, and
computerized simulation of synergism and antagonism in drug
combination studies. Pharm. Rev. 2006, 58, 621−681.
(37) Lee, J. J.; Kong, M.; Ayers, G. D.; Lotan, R. Interaction index
and different methods for determining drug interaction in combination
therapy. J. Biopharm. Stat. 2007, 17, 461−480.
(38) Wuitschik, G.; Carreira, E. M.; Wagner, B.; Fischer, H.; Parrilla, I.;
Schuler, F.; Rogers-Evans, M.; Muller, K. Oxetanes in drug discovery:
structural and synthetic insights. J. Med. Chem. 2010, 53, 3227−3246.
(39) Kansy, M.; Senner, F.; Gubernator, K. Physicochemical high
throughput screening: parallel artificial membrane permeability assay
in the description of passive absorption processes. J. Med. Chem. 1998,
41, 1007−1010.
(40) Bjorkhem, I.; Meaney, S. Brain cholesterol: long secret life
behind a barrier. Arterioscler., Thromb., Vasc. Biol. 2004, 24, 806−815.
(41) Dietschy, J. M.; Turley, S. D. Cholesterol metabolism in the
central nervous system during early development and in the mature
animal. J. Lipid Res. 2004, 45, 1375−1397.
(42) Natali, F.; Siculella, L.; Salvati, S.; Gnoni, G. V. Oleic acid is a
potent inhibitor of fatty acid and cholesterol synthesis in C6 glioma
cells. J. Lipid Res. 2007, 48, 1966−1975.
(43) Prasanna, P.; Thibault, A.; Liu, L.; Samid, D. Lipid metabolism
as a target for brain cancer therapy: synergistic activity of lovastatin
and sodium phenylacetate against human glioma cells. J. Neurochem.
1996, 66, 710−716.
(44) Brown, R. C.; Cascio, C.; Papadopoulos, V. Pathways of
neurosteroid biosynthesis in cell lines from human brain: regulation of
dehydroepiandrosterone formation by oxidative stress and beta-
amyloid peptide. J. Neurochem. 2000, 74, 847−859.
(45) Bababeygy, S. R.; Polevaya, N. V.; Youssef, S.; Sun, A.; Xiong,
A.; Prugpichailers, T.; Veeravagu, A.; Hou, L. C.; Steinman, L.; Tse, V.
HMG-CoA reductase inhibition causes increased necrosis and
apoptosis in an in vivo mouse glioblastoma multiforme model.
Anticancer Res. 2009, 29, 4901−4908.
(46) Bouterfa, H. L.; Sattelmeyer, V.; Czub, S.; Vordermark, D.;
Roosen, K.; Tonn, J. C. Inhibition of Ras farnesylation by lovastatin leads
to downregulation of proliferation and migration in primary cultured
human glioblastoma cells. Anticancer Res. 2000, 20, 2761−2771.
(47) Chan, D. Y. L.; Chen, G. G.; Poon, W. S.; Liu, P. C. Lovastatin
sensitized human glioblastoma cells to TRAIL-induced apoptosis. J.
Neuro-Oncol. 2008, 86, 273−283.
(23) Gao, J.; Chu, X.; Qiu, Y.; Wu, L.; Qiao, Y.; Wu, J.; Li, D.
Discovery of potent inhibitor for farnesyl pyrophosphate synthase in
the mevalonate pathway. Chem. Commun. 2010, 46, 5340−5342.
(24) Mark, M.; Muller, P.; Maier, R.; Eisele, B. Effects of a novel 2,3-
oxidosqualene cyclase inhibitor on the regulation of cholesterol
biosynthesis in HepG2 cells. J. Lipid Res. 1996, 37, 148−158.
(25) Eisele, B.; Budzinski, R.; Muller, P.; Mark, M. Effects of a novel
̈
2,3-oxidosqualene cyclase inhibitor on cholesterol biosynthesis and
lipid metabolism in vivo. J. Lipid Res. 1997, 38, 564−575.
(26) Dollis, D.; Schuber, F. Effects of 2,3-oxidosqualene-lanosterol
cyclase inhibitor, 2,3:22,23-dioxidosqualene and 24,25-epoxycholester-
ol on the regulation of cholesterol biosynthesis in human hepatoma
cell line Hep-G2. Biochem. Pharmacol. 1994, 48, 49−57.
(27) Peffley, D. M.; Gayen, A. K.; Morand, O. H. Down-regulation of
3-hydroxy-3-methylglutaryl coenzyme A reductase mRNA levels and
synthesis in syrian hamster C100 cells by the oxidosqualene cyclase
inhibitor [4′-(6-allyl-ethyl-amino-hexyloxy)-2′-fluoro-phenyl]-(4-bro-
mophenyl)-methanone (RO0488071): comparison to simvastatin.
Biochem. Pharmacol. 1998, 56, 439−449.
(28) Gardner, R. G.; Shan, H.; Matsuda, S. P. T.; Hampton, R. Y. An
oxysterol-derived positive signal for 3-hydroxy-3-methylglutaryl-CoA
reductase degradation in yeast. J. Biol. Chem. 2001, 276, 8681−8694.
(29) Dehmlow, H.; Ackermann, J.; Aebi, J.; Blum-Kaelin, D.;
Chucholowski, A.; Coassolo, P.; Hartman, P.; Kansy, M.; Maerki, H.
P.; Morand, O.; Von der Mark, E.; Panday, N.; Ruf, A.; Thoma, R.;
Schulz-Gasch, T. Oxidosqualene cyclase (OSC) inhibitors for the
treatment of dyslipidemia. Chimia 2005, 59, 72−76.
(30) Jolidon, S.; Polak-Wyss, A.; Hartman, P. G.; Guerry, P. 2,3-
Oxidosqualenelanosterol Cyclase: An Attractive Target for Antifungal
Drug Design. In Recent Advances in the Chemistry of Anti-Infective
Agents; Bentley, P. H., Ponsford, R, Eds.; Bookcraft Ltd.: Bath, U.K.,
1993; pp 223−233.
(31) Thoma, R.; Schulz-Gasch, T.; D’Arcy, B.; Benz, J.; Aebi, J.;
Dehmlow, H.; Hennig, M.; Stihle, M.; Ruf, A. Insight into steroid
scaffold formation from the structure of human oxidosqualene cyclase.
Nature 2004, 432, 118−122.
(32) Corey, E. J.; Fuchs, P. L. Synthetic method for conversion of formyl
groups into ethynyl groups. Tetrahedron Lett. 1972, 36, 3769−3772.
(33) Li, P.; Yamamoto, H. Amino acid salt catalyzed intramolecular
Robinson annulation. Chem. Commun. 2009, 36, 5412−5414.
(34) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. A
greatly improved procedure for ruthenium tetroxide catalyzed
oxidations of organic compounds. J. Org. Chem. 1981, 46, 3936−3938.
(35) Brusselmans, K.; Timmermans, L.; Van de Sande, T.; Van
Veldhoven, P. P.; Guan, G.; Schechter, I.; Claessens, F.; Verhoeven,
G.; Swinnen, J. V. Squalene synthase, a determinant of Raft-associated
cholesterol and modulator of cancer cell proliferation. J. Biol. Chem.
2007, 282, 18777−18785.
(48) Jiang, Z.; Zheng, X.; Lytle, R. A.; Higashikubo, R.; Rich, K. M.
Lovastatin-induced up-regulation of the BH3-only protein, Bim, and
cell death in glioblastoma cells. J. Neurochem. 2004, 89, 168−178.
(49) Schmidt, F.; Groscurth, P.; Kermer, M.; Dichgans, J.; Weller, M.
Lovastatin and phenylacetate induce apoptosis, but not differentiation in
human malignant glioma cells. Acta Neuropathol. 2001, 101, 217−224.
(50) Tapia-Perez, J. H.; Kirches, E.; Mawrin, C.; Firsching, R.;
Schneider, T. Cytotoxic effect of different statins and thiazolidine-
diones on malignant glioma cells. Cancer Chemother. Pharmacol. 2011,
67, 1193−1201.
(51) Larner, J.; Jane, J.; Laws, E.; Packer, R.; Myers, C.; Shaffrey, M.
A phase I−II trial of lovastatin for anaplastic astrocytoma and
glioblastoma multiforme. Am. J. Clin. Oncol. 1998, 21, 579−583.
(52) Berger, Y.; Dehmlow, H.; Blum-Kaelin, D.; Kitas, E. A.; Loffler,
̈
B. M.; Aebi, J. D.; Juillerat-Jeanneret, L. Endothelin-converting-
enzyme-1 inhibition and growth of human glioblastoma cells. J. Med.
Chem. 2005, 48, 483−498.
5002
dx.doi.org/10.1021/jm300256z | J. Med. Chem. 2012, 55, 4990−5002