Article

Inhibition of Fatty Acid Synthase by Orlistat Accelerates Gastric Tumor Cell Apoptosis in Culture and Increases Survival Rates in Gastric Tumor Bearing Mice In Vivo

Biochemistry, A.T. Still University of the Health Sciences, Kirksville, USA.
Lipids (Impact Factor: 1.85). 05/2009; 44(6):489-98. DOI: 10.1007/s11745-009-3298-2
Source: PubMed

ABSTRACT

Orlistat, an anti-obesity drug, is a potent inhibitor of fatty acid synthase (FAS) and tumor cell viability. It can also induce apoptotic cancer cell death. We examined the effects of Orlistat on cultured NUGC-3 gastric cancer cells. We identified that inhibition of FAS via Orlistat exposure results in rapid cellular damage preceded by a direct but short-lived autophagic response. The Orlistat induced damage can be reversed through the addition of lipid containing media in a process that normally leads to cell death. By limiting exogenous lipid availability and inhibiting FAS using Orlistat, we demonstrated both a greater sensitivity and amplified cancer cell death by activation of apoptosis. We have identified "windows of opportunity" at which time apoptosis can be aborted and cells can be reversed from the death pathway. However, when challenged beyond the window of recovery, cell death becomes all but certain as the ability to be rescued decreases considerably. In vivo examination of Orlistat's ability to inhibit gastrointestinal cancer was examined using heterozygous male C57BL/6J APC-Min mice, which spontaneously develop a fatal gastrointestinal cancer. Mice were fed either a high fat (11%) or low fat (1.2%) diet containing no Orlistat or 0.5 mg Orlistat/g of chow. Orlistat treated mice fed the high fat, but not low fat diet, survived 7-10% longer than the untreated controls.

Full-text

Available from: James L Cox, Oct 27, 2015
ORIGINAL ARTICLE
Inhibition of Fatty Acid Synthase by Orlistat Accelerates Gastric
Tumor Cell Apoptosis in Culture and Increases Survival Rates
in Gastric Tumor Bearing Mice In Vivo
Shawn Dowling Æ James Cox Æ Richard J. Cenedella
Received: 24 October 2008 / Accepted: 16 March 2009
Ó AOCS 2009
Abstract Orlistat, an anti-obesity drug, is a potent
inhibitor of fatty acid synthase (FAS) and tumor cell via-
bility. It can also induce apoptotic cancer cell death. We
examined the effects of Orlistat on cultured NUGC-3
gastric cancer cells. We identified that inhibition of FAS
via Orlistat exposure results in rapid cellular damage pre-
ceded by a direct but short-lived autophagic response. The
Orlistat induced damage can be reversed through the
addition of lipid containing media in a process that nor-
mally leads to cell death. By limiting exogenous lipid
availability and inhibiting FAS using Orlistat, we demon-
strated both a greater sensitivity and amplified cancer cell
death by activation of apoptosis. We have identified
‘windows of opportunity’ at which time apoptosis can be
aborted and cells can be reversed from the death pathway.
However, when challenged beyond the window of recov-
ery, cell death becomes all but certain as the ability to be
rescued decreases considerably. In vivo examination of
Orlistat’s ability to inhibit gastrointestinal cancer was
examined using heterozygous male C57BL/6J APC-Min
mice, which spontaneously develop a fatal gastrointestinal
cancer. Mice were fed either a high fat (11%) or low fat
(1.2%) diet containing no Orlistat or 0.5 mg Orlistat/g of
chow. Orlistat treated mice fed the high fat, but not low fat
diet, survived 7–10% longer than the untreated controls.
Keywords Thin layer chromatography
Analytical Techniques Lipid biochemistry
General Area Lipases Lipoproteins Metabolism
Cancer Physiology Fatty acids Specific Lipids
Sterols
Introduction
Fatty acid synthase (FAS) functions to form palmitic acid
from acetyl and malonyl coenzymeA. It is highly expressed
in many tumors that are dependent on de novo synthesis to
supply the fatty acid needed for growth and proliferation
[15]. The capacity of FAS inhibitors to kill cancer cells
suggests a new approach to treatment of human cancers [6].
Orlistat is one of these FAS inhibitors.
Orlistat, an over-the-counter anti-obesity drug, decrea-
ses absorption of dietary fat by inhibiting gastric and
pancreatic lipases through covalent modification of the
enzymes [7]. It is also a potent inhibitor of FAS [8]. Using
an activity-based proteomics screen for serine hydrolases,
Kridel et al. [8] discovered that Orlistat irreversibly inhibits
the thioesterase activity of FAS. The capacity of Orlistat to
inhibit the proliferation of prostatic cancer cells cultured in
serum free media appeared to be due to restricting the
availability of fatty acids for growth, since the addition of
palmitic acid to the media reversed the inhibition [8].
A decrease in DNA synthesis, arrest of cell progression
through the G1/S boundary, and apoptotic cell death are all
consequences of inhibiting FAS in cultured cancer cells
with Orlistat [6, 9].
The effects of Orlistat on proliferation and viability of
tumor cells have typically been examined after 24–72 h [8,
10] and even after 4–5 days [10] of continuous exposure to
the drug. This approach assumes that the Orlistat has a
linear-accumulative effect on tumor cells. Possible imme-
diate or very rapid effects of Orlistat on tumor cells have
not been reported. We searched for such effects and
S. Dowling J. Cox R. J. Cenedella (&)
Biochemistry, A.T. Still University of the Health Sciences,
Kirksville, USA
e-mail: rcenedella@atsu.edu
123
Lipids
DOI 10.1007/s11745-009-3298-2
Page 1
observed that as little as one-half hour exposure of NUGC-
3 cells to 100–500 lM Orlistat induced transient autoph-
agy. In addition, we observed that 4 h exposure followed
by removal of the drug reduced viability when examined at
hour 48. The presence of Orlistat for the first 8 h of culture
resulted in near total loss of viability when measured 40 h
later. Acute-irreversible damage appeared to occur at the
interface between 4 and 8 contact hours with Orlistat. We
probed the mechanism of reduced viability by examining
NUGC-3 cells for the emergence of autophagy and apop-
tosis following short-term exposure to the drug. Rapid
onset of apoptosis was attributed to nutritional stress, in
which altering the availability of nutrients places a stress
on the cell to acquire alternative sources of nutrients to
meet the cells biogenesis needs. In this case, nutritional
stress was caused by the deprivation of fatty acids since the
apoptosis could be arrested by addition of whole serum to
the culture media.
Although Kridel et al. [8] demonstrated that intraperi-
toneally injected Orlistat suppressed the growth of
implanted tumor cells, use of Orlistat to treat systemic
tumors is limited by the fact that it is essentially unab-
sorbed following oral dosage [11]. However, since orally
administered Orlistat would directly contact the gastroin-
testinal mucosa, the drug might be of value in treating
gastrointestinal tumors exposed to the intestinal lumen. We
tested this possibility by examining the effect of orally
given Orlistat on the survival of C57BL/6J APC-Min mice
(Min). The Min mice provide an excellent in vivo model
for human gastrointestinal cancer, since the adenomas that
develop in these animals result from inactivation of the
same tumor suppressor gene associated with most human
colon cancers [12]. The tumors develop due to a mutation
in the murine AP gene, a homolog of the human APC gene
(JAX MICE literature, The Jackson Laboratory). However,
Min mice develop a heavy tumor load in the small rather
than large intestines which results in death at several
months of age [1315].
Materials and Methods
Cell Culture and Reagents
All reagents were from Sigma Chemical Co. (St. Louis,
MO) unless otherwise noted. NUGC-3 gastric cancer cells
(human stomach) were obtained from the Health Science
Research Resource Bank (Osaka, Japan). Bovine lens
epithelial cells (local source) were used as a normal cell
control. Cells were seeded at 30,000–40,000 per well in 96
well dishes and grown at 37°C and 5% carbon dioxide in
RPMI-1640 media containing 10% fetal calf serum (FCS)
plus 0.25% antibiotic/antimycotic containing mix. This is
called whole media (WM). Experiments were started
1–2 days later with cell layers 70–90% confluent. Orlistat
from 5 to 120 mg Zenical
TM
capsules (obtained from a local
pharmacy) was extracted by homogenization in 3–5 ml
aliquots of 1:1 (v/v) chloroform:methanol and transferred
to a weighed test tube. After evaporation of the solvent
under nitrogen and further drying overnight under
vacuum, the recovered Orlistat was dissolved in ethanol at
50 mg/ml. Aliquots were added to WM or serum deficient
media (DM) containing 0.1% FCS to yield 500 lM. Serial
dilutions provided media containing 200, 100, 50, and
25 lM Orlistat. The capacity of Orlistat to decrease cell
viability was also examined in non-cancerous (normal)
bovine lens epithelial cells. Culture conditions were as we
described before [16].
Cell Viability
Cell layers in groups of 5 or 6 wells were incubated for
4–48 h at 37°C with 200 ll of WM or DM containing
0–500 lM Orlistat. Cell viability was assessed by the MTT
(thiazolyl blue tetrazolium bromide) assay essentially as
described by Carmichael et al. [17]. Test media was
replaced with 200 ll of WM, 25 ll of MTT solution
(5 mg/ml water) was added, and the samples incubated for
4 h at 37°C. The MTT solution was replaced with 100 llof
DMSO and the optical density at 570 nm measured 1 h
later. Absorbance for each group was expressed as the
mean ± SEM. Statistical significance of differences
(P B 0.05) was assessed by one-way ANOVA.
Measurement of Fatty Acid and Synthesis
Near confluent layers of NUGC-3 cells were washed twice
in serum free media, after which 200 ll of plain RPMI-
1640 media was added followed by 1 llof
14
C-acetate
(55 mCi/mmol, ARC Inc, St. Louis, MO). Serum free
media was used to avoid dilution of the labeled acetate
with competing sources of unlabeled acetate. Following
4-hour incubation in the absence or presence of varying
concentrations of Orlistat, cell layers were washed twice
with PBS. The cell layer in each well was dissolved in
200 ll of 1 N NaOH in 50% ethanol containing 100 lgof
triolein and 100 lg of cholesterol added as carrier. Each
well was washed with 4 additional 200 ll-aliquots of the
1 N NaOH and pooled into screw capped test tubes with
the first 200 ll. Samples were saponified for 2 h at 100°C,
diluted with 1 ml of water, acidified and the non-saponi-
fiable plus saponifiable lipids extracted with 2–4 ml
aliquots of hexane. The hexane extracts were washed with
an equal volume of water to remove any water-soluble
radiolabel. The hexane extracts were evaporated and the
lipids separated by thin layer chromatography in silica gel
Lipids
123
Page 2
using a solvent of hexane:diethyl ether:glacial acetic acid
(73:25:2: v/v/v). The sterol (cholesterol) band and the fatty
acid band were recognized by exposure to iodine vapor and
marked. After loss of the iodine by sublimation, the cho-
lesterol and fatty acid bands were recovered and the
radiolabel measured by scintillation counting.
We accounted for variation in incorporation from well to
well in any given group by expressing the level of fatty
acid synthesis as a ratio of incorporation of
14
C into fatty
acid to cholesterol. Cholesterol controls were run to
determine Orlistat’s effect on total sterol synthesis. Sig-
nificant differences were not seen in the disintegrations per
minute (dpm) of
14
C-acetate incorporated into cholesterol
between control and Orlistat treated cell layers. Decreases
in the fatty acid:cholesterol ratio reflects inhibition of fatty
acid synthesis, since Orlistat has no known effects on
cholesterol biosynthesis.
Western Blotting: Changes in the Relative Mass
of Fatty Acid Synthase (FAS)
NUGC-3 cells were grown in RPMI-1640 on 35 mm
plastic dishes to assure adequate protein. Cell layers
exposed or unexposed to Orlistat under varying condition
were washed twice with PBS and dissolved in 0.50 ml of
lysis buffer [18]. Samples were assayed for protein (ali-
quots containing 50 lg protein were separated on 8%
PAGE gels 94% stacking gel), transferred to membranes
which were then probed with 0.8 lg/ml rabbit-polyclonal,
antihuman FAS-antiserum (Santa Cruz Biotechnology) and
HRP-conjugated goat-antirabbit IgG secondary antibody
(1/60,000 dilution). Immunoreactive protein was detected
by enhanced chemiluminescences and relative concentra-
tions estimated by densitometric scanning of the exposed
X-ray film as done before [18].
Detection of Autophagy and Apoptosis
Autophagic vacuoles were labeled with monodansylcada-
verine (MDC) using a modified protocol as described by
Biederbick et al. [19]. Briefly, cells were grown on cov-
erslips, incubated at 37°C in WM, and DM ± 500 lM
Orlistat for varying time intervals. Media was removed;
cells were washed once with PBS, and replaced with PBS
supplemented with 0.1 mM MDC. Cells were allowed to
incubate for 10 min at 37°C. Following incubation, cells
were washed three times with PBS, fixed using 4% para-
formaldehyde, and mounted using Fluoromount. Cells were
examined using a NIKON Eclipse 80i upright fluorescent
microscope equipped with a V2-A filter system (excitation
340 nm, barrier 514 nm). Images were captured using a
NIKON DS-Qi1 digital camera and were processed using
Image J.
Autophagic vacuole formation was measured using
methods described by Munafo and Colombo [20]. Briefly,
cells were grown in 60 mm dishes and treated using
varying concentrations of WM and DM ± Orlistat. To
identify autophagy activity and duration, MDC incorpora-
tion was measured after 0.5, 1, 2, 4, 6, 8, and 10-hour
Orlistat exposure. After varying times the media was
removed and PBS was added with the addition of 0.1 mM
of MDC and incubated at 37°C for 10 min. Following
incubation the cells were washed three times with PBS and
collected in 10 mM Tris–HCl, pH 8 with 0.1% Triton
X-100. Intracellular MDC incorporation was measured
using a BioTek FLx 800 Microplate Fluorescent Reader
equipped with an excitation filter of 365/20 nm and an
emissions filter of 528/20 nm. The number of cells per well
were normalized by the addition of ethidium bromide to a
final concentration of 0.1 lM to each well followed by the
measuring of DNA fluorescence (excitation filter 528/
20 nm and emission filter of 590/20 nm). The MDC mea-
sured was expressed as the percent of autophagic activity
relative to whole media control.
The mechanism of cell death (apoptosis) was deter-
mined through the use of fluorochrome inhibitor of
caspases (FLICA) (Immunochemistry Technologies, LCC).
Activated caspase is an enzyme found active only in cells
undergoing apoptosis. FLICA irreversibly binds to casp-
ases (caspase-1, -3, -4, -5, -6, -7, -8, -9) allowing apoptosis
to be measured [21, 22]. Caspase activity was visually
examined using the suggested manufactures protocol for
labeling caspases with FLICA. Briefly, cells were grown on
coverslips and treated with various concentrations of
Orlistat substituted in DM. Cells were then treated with
FLICA solution in a 1:30 ratio (10 ll FLICA:290 ll
media) and incubated for 1 h. Next, media was removed
and replaced with fresh media (300 ll) containing 1.5 llof
200 lg/ml stock Hoechst stain solution (Immunochemis-
try). After 5 min cells were rinsed with wash buffer
(Immunochemistry) and examined immediately using a
NIKON Eclipse 80i upright fluorescent microscope
equipped with a V2-A filter system (excitation 340 nm,
barrier 514 nm). Images were captured using a NIKON
DS-Qi1 digital camera and were processed using Image J.
Animals and Treatments
Heterozygous male C57BL/6J APC-Min (multiple intesti-
nal neoplasia) mice were purchased from The Jackson
Laboratory (Bar Harbor, ME). Heterozygous mice were
used for these studies because homozygous mice die in
utero. Mice in groups of 4–8 were fed either a high fat
(11% fat) or low fat (1.2%) diet containing no Orlistat or
Orlistat at 0.5 mg/g chow. Mice were started on the diets at
either 50–60 days (Experiments 1 and 3) or 33–34 days
Lipids
123
Page 3
(Experiment 2) of age. Age at death varied from 104–
229 days of age. Mice were housed in the Animal Care
Facility at KCOM/ATSU and maintained according to the
‘Guide of the Care and Use of Laboratory Animals’ from
the National Research Council.
Recovery of Orlistat and Preparation of Test Diets
The contents of 4–6 Zenical capsules, each containing
120 mg of Orlistat, were pooled and extracted three times
with 5 ml of 1:1 chloroform–methanol (v/v) using a
Polytron-type homogenizer. The solvent from the pooled
extracts recovered after centrifugation was evaporated
under nitrogen. The residue was dried overnight to constant
weight under high vacuum, and dissolved in ethanol to give
a stock solution of 50 mg/ml. Mice in two studies were fed
ground Purina Mouse Diet 5015 that contained 11% fat
plus 0 (control) or 0.5 mg Orlistat/g chow. In a third study,
mice were fed a customized powered diet (AIN-93M,
Purina Mills TESTDIET, Richmond, IN) containing 0.2%
fat, from saturated and monounsaturated fatty acids, to
which we added 1% linoleic acid to guard against essential
fatty acid deficiency. Orlistat was added at 0 (control) or
0.5 mg/g (test) to this very low fat diet.
HPLC Analysis of Orlistat
The HPLC method of Souri et al. [11] was used to confirm
the concentration of Orlistat in chow and to estimate the
distribution of Orlistat in the mouse gut following oral
dosage. Samples of mouse chow (11% fat) with Orlistat
and sections of mouse gut removed at 4 h after oral dosage
were extracted with 1:1 chloroform–methanol, and the
recovered lipids were fractionated by thin layer chromato-
graphy on Silica Gel G plates using a solvent of hexane:
diethyl ether:glacial acetic acid (73:25:2). Orlistat, visual-
ized by exposure to iodine vapor, migrated with an Rf of
between 0.1 and 0.2. After sublimation of the iodine, the
TLC zone containing Orlistat was recovered and extracted
twice with 3 ml aliquots of methanol. The residue
remaining after evaporation of the methanol was dissolved
in several hundred microliters of acetonitrile:0.1% ortho-
phosphoric acid in water (90:10), centrifuged and aliquots
injected onto a Waters C18 reverse phase column (10 lm,
3.9 9 300 mm, 125 A lBondapak) eluted at 1 ml/min with
the 90:10 acetonitrile-acidic water and absorbance moni-
tored at 205 nm. Ten micrograms of Orlistat standard
(injected in 100 ll) eluted at about 6 min.
Survival Statistical Analyses
To determine the effects of Orlistat on mouse survival, the
end point analyzed was the difference in day of death
between untreated (control) and Orlistat treated animals
(test). The Kaplan–Meier procedure [18] was used to
estimate the probability of surviving a given time. Since all
untreated animals die prematurely (4–7 months) compared
to wild type mice (1.5–2 years), even slight prolongation of
life due to treatment can be recognized. Using a statistical
computer program (GraphPad Prism version 5.0 for Win-
dows, GraphPad Software, San Diego, CA, USA) survival
curves were generated by the Kaplan–Meier procedure [23]
and analyzed by the Log-Rank (Mantel–Cox) and the
Gehan–Wilcoxon Tests. The Log-Rank test is more
powerful and gives equal weight to all time points.
Results
Potential use of Orlistat as an anticancer drug would likely
be restricted to treatment of gastrointestinal tumors since it
is essentially unabsorbed following oral dosage [2].
Therefore, we chose to examine the effects of Orlistat on a
human derived gastric cancer cell line, NUGC-3 [24].
Viability of cells cultured for 48 h in serum deficient
media, DM (0.10% serum), was decreased by more than
70% when continuously exposed to 25 lM Orlistat
(Fig. 1). Near complete inhibition was seen at 100 lM.
The dose response curve was markedly shifted to the right
by including 10% serum in the media, called whole media
(WM).
Although cell viability assays are typically performed
following 48 h or more of continuous exposure to an
inhibitor, we considered the possibility that the decreased
viability observed at 48 h was due in part or total to cell
injury induced soon after adding Orlistat rather than to a
Fig. 1 Effect of Orlistat on the viability of NUGC-3 gastric tumor
cells. Cells were cultured for 48 h in RMPI-1640 media containing
10% fetal calf serum (whole media, WM) or 0.10% serum (deficient
media, DM) and varying concentrations of Orlistat. Cell viability was
assessed by the MTT assay. Each point is the mean ± SEM of 6 cell
wells
Lipids
123
Page 4
sustained accumulation of injury. Cells were incubated for
4, 8 or 12 h in DM containing 0 (control), 100 or 500 lM
Orlistat, and then incubated to hour 48 in drug free media.
Cell viability was significantly reduced from only 4-hour
exposure to 500 lM Orlistat and an 8-hour exposure to
100 lM Orlistat (Fig. 2a). Four-hour exposure to 500 lM
Orlistat decreased viability by about 65%, while 8-hour
exposure to 100 lM decreased viability by about 40%.
These brief exposure periods accounted for about two-
thirds and one-half of the total decrease seen for cells
cultured the entire 48 h in DM with 500 and 100 lM
Orlistat, respectively. Addition of WM rather than DM to
the NUGC-3 cell layers after 4 h exposure to 500 lM
Orlistat and after 8 h exposure to 100 lM totally rescued
the cells; that is, there was no decrease in viability com-
pared to cells cultured in the absence of Orlistat measured
at 48 h (Fig. 2b). However, there were limits to the
capacity of WM to rescue Orlistat exposed cell. For
example, 12 h of exposure to 500 lM Orlistat in DM
committed the cells to detachment and disintegration.
Orlistat toxicity likely begins with inhibition of fatty
acid synthesis. Incubation of NUGC-3 cells with 200 lM
Orlistat for 4 h decreased incorporation of
14
C-acetate into
total fatty acids by about 75% (Fig. 3a). We expressed
effects on fatty acid synthesis as a decrease in the ratio of
incorporation of
14
C-acetate into fatty acid to total sterol.
Since Orlistat had no effect on sterol synthesis, this
expression should account for potential differences in cell
numbers from well to well. Not only does Orlistat inhibit
fatty acid synthesis, but also once exposed to this drug the
inhibition seems permanent. Fatty acid synthesis continued
to be inhibited after Orlistat was removed from the media.
NUGC-3 cells were incubated for 4 h in media containing
200 lM Orlistat, the cell layers were washed and then
incubated an additional 4 h in Orlistat free media con-
taining
14
C-acetate. Synthesis was still inhibited by 80%
(Fig. 3b). This finding supports the idea that Orlistat binds
irreversibly to the thioesterase domain of FAS [8]. Up
regulation of FAS expression might be a means to cir-
cumvent the near total inhibition of fatty acid synthesis;
however, FAS protein levels (assessed by Western blot-
ting) were not increased following exposure of Orlistat
(data not shown). Thus, once exposed to Orlistat fatty acid
synthesis appears permanently inhibited. This would make
the cells dependent upon uptake of lipids from the media as
the only source of fatty acids to sustain membrane turnover
and cell proliferation.
As shown (Fig. 2), there was a narrow window of
opportunity to rescue NUGC-3 cell from permanent injury
and death when cultured in DM containing Orlistat. The
response of the cells to this nutritional stress, no source of
fatty acids, was extremely rapid induction of autophagy
(Fig. 4). The autophagic response was Orlistat dose depen-
dent, peaked at one-half hour of incubation and returned to
baseline by 2 h. Thus, it seems unlikely that autophagy
accounts for the irreversible injury and consequent cell
Fig. 2 Early effects of Orlistat on cell viability. a NUGC-3 cell
layers were incubated in serum deficient media containing 0, 100 or
500 lM Orlistat. After 4, 8 or 12 h of incubation the media was
replaced with Orlistat free DM and the incubation continued to hour
48 at which time cell viability was measured. Controls were cultured
for 48 h in Orlistat free DM. The ‘48’ hour groups were cells
incubated with either 100 or 500 lM Orlistat-DM. b Controls are
cells cultured for 48 h in Orlistat free WM. Cells were incubated for
4 h in DM containing 500 lM Orlistat or 8 h with 100 lM Orlistat
and then switched to Orlistat free WM and the culture continued to
hour 48 at which time viability was measured
Fig. 3 Irreversible inhibition of fatty acid synthesis by Orlistat in
NUGC-3 tumor cells. a Subconfluent cell layers in 96 well dishes
were incubated for 4 h in serum free RMPI-1640 media containing
1 lCi of
1-14
C-acetate and 0 (Control C) or 200 lM Orlistat (Treated
T). Total fatty acids and sterols recovered after saponification were
separated by TLC, fractions isolated, the
14
C-content measured and
results expressed as the ratio of incorporation into fatty acids to
sterols. Each value is the mean ± SEM of 5 cell layers. b Cell layers
were incubated for 4 h with 0 or 200 lM Orlistat in the absence of
14
C-acetate; the media was then replaced with Orlistat free media
containing the
14
C-acetate and incubated for a second 4-hour interval.
Values are the mean ± SEM of five cell layers
Lipids
123
Page 5
death seen with longer-term exposure. Rather, cell death
could be due to induction of apoptosis. The onset and pro-
gression of apoptosis paralleled the loss of cell viability.
Apoptosis was seen after 8 h exposure to 100 lM Orlistat in
DM (Fig. 5e), but not at 4 h (Fig. 5c). Loss of cell viability
occurred after 8 but not 4 h exposure to 100 lM (Fig. 2 a).
At the higher drug concentration (500 lM), apoptosis was
evident after 4 h of incubation (Fig. 5d) and markedly
increased by the 8 h of drug exposure (Fig. 5f). The pro-
gression of apoptosis over the 4–8 h interval was largely
halted by exchanging the Orlistat-DM media for drug-free
WM at hour four (Fig. 5g). This parallels the time-interval
boundaries for rescue of NUGC-3 cell viability from expo-
sure to 500 lM Orlistat (Fig. 2).
Applying the information garnered from in vitro
experiments, the potential of Orlistat to inhibit in vivo
gastrointestinal cancer was evaluated by examining the
effect of orally administered Orlistat on the survival of
APC-Min mice that spontaneously develop a fatal gastro-
intestinal cancer. Mice were fed a high fat (11%) or low fat
(1.2%) diet containing 0 or 500 mg Orlistat/kg chow. Each
mouse weighed approximately 25 g and consumed an
average of 3 g of chow per day, resulting in delivery of
about 60 mg of Orlistat/kg body weight. The fecal fat
content of Orlistat treated mice increased 4-to-6-fold
compared to untreated control mice when the animals were
fed chow containing 11% fat (Table 1). In animals fed
chow containing 1.2% fat, fecal fat increased about
2.5-fold in Orlistat treated mice versus untreated controls
(Table 1). The increased fecal fat obviously reflects the
inhibition of gastric lipases by the drug.
Orally administered Orlistat moves quickly through the
GI tract. Four hours after giving a single oral 50 mg/kg
dose of Orlistat (in aqueous emulsion) by gavage, the drug
was detected almost exclusively in the large intestines
(Fig. 6). Similar results were seen in a replicate experi-
ment. This observation is consistent with GI transit times in
the mouse being 4–6 h [25].
Considering the browsing feeding pattern of mice, the
gut should receive multiple exposures to Orlistat during the
day. The 500 mg of Orlistat/kg of chow corresponds to
about 1 mM (MW = 497.5), a concentration in great
excess of that needed to inhibit the viability of cultured,
human–derived, NUGC-3 gastrointestinal cancer cells
(Fig. 1). Fifty lM Orlistat in serum deficient media (0.1%
serum) and 200 lM Orlistat in whole media (10% serum)
both inhibited cell viability by about 80% (Fig. 1). Similar
results were seen in replicate experiments.
Three experiments were conducted to evaluate the effect
of orally administered Orlistat on survival times of the
Fig. 4 Rapid induction and
disappearance of the autophagic
response to Orlistat.
Fluorescence microscopy
showing autophagic vesicles in
NUGC-3 cells unexposed to
Orlistat (a), exposed to 500 lM
Orlistat for 30 min (b)or6h
(c). The graph shows the time
course of autophagic response
to varying concentrations of
Orlistat. Values are the mean
± SEM of 3 cell layers.
Differences in cell numbers
between wells normalized using
ethidium bromide [21]
Lipids
123
Page 6
APC-min mice (Fig. 7). Median survival times were
increased 7–10% (25.5 and 9 days, experiments 1 and 2,
respectively) in Orlistat fed mice (Fig. 7 top and middle
panels). The difference was statistically significant by the
log-rank test (P = 0.047) and the Wilcoxon test
(P = 0.0318) in experiment 1 and was not significant in
experiment 2 (Table 2). Experiment 3, using a low fat
(1.2%) diet produced no significant results (Table 2),
however, control mice fed the low fat diet produced a
greater median survival time than their treated counterparts
(Table 1).
Discussion
As stated above, cancer cells can place a great demand on
FAS to supply the fatty acids required for membrane
turnover and cell proliferation. NUGC-3 cells cultured in
serum deficient media (0.1% serum) grew as well as when
cultured in media containing 10% serum, indicating that
lipogenesis can sustain growth. In contrast to cancer cells,
viability of a normal cell line (bovine lens epithelial)
decreased by about 50% when cultured in serum deficient
media, indicating a major reliance on extracellular lipids
(Fig. 8). Also, the normal bovine lens epithelial cells were
virtually resistant to the growth inhibition mediated by high
concentrations of Orlistat in WM, indicating little reliance
on fatty acid synthesis (Fig. 8). The general toxicity of
Orlistat is also addressed by this data. Since the viability of
lens epithelial cells cultured in WM containing no Orlistat
and high levels of Orlistat (200 and 500 lM) were similar,
Orlistat has no apparent toxicity to these cultured cells
independent of its effect on FAS. This would appear to also
include toxicity due to inhibition of triacylglycerol lipases,
the basis of its use as an anti-obesity drug [7].
Perhaps the major findings of this work are the speed
with which Orlistat can insult the cultured cancer cells and
the capacity of whole media to rescue the cells from this
insult, at least up to a point. The rapid untoward effects are
attributed to near total restriction of fatty acids due to
inhibition of synthesis coupled with the absence of media
lipids. Because membrane phospholipids are highly
dynamic and can turnover with half-lives in the minutes
[26], a block in the availability of fatty acids could rapidly
compromise membrane integrity. In view of this rapid
turnover and because the restriction of fatty acids consti-
tutes nutritional stress, it is not surprising that autophagy
was activated within minutes of adding Orlistat. However,
it was surprising that the autophagic response was over
Fig. 5 Antagonism of Orlistat induced apoptosis. Images show
overlap fluorescence of apoptotic caspase stained (green) and nucleic
acid stained (blue) cells. All cell layers, except b and g, were
incubated in DM. a Control. b Staurosporine, 500 nM in WM for 3 h.
c Orlistat, 100 lM for 4 h. d Orlistat, 500 lM for 4 h. e Orlistat,
100 lM for 8 h. f Orlistat, 500 lM for 8 h. g Cell layers were
incubated for 4 h with 500 lM Orlistat, the media was replaced with
Orlistat free WM and the incubation continued an additional 4 h.
These results are representative of triplicate experiments (color figure
online)
Table 1 Effect of Orlistat on fecal fat content and median survival
times
Experiment Group Fecal fat content
a
(mg fat/g dry
feces)
Starting
age
(days)
Median
survival
(days)
1 High fat Control 45.3 50–60 147
Treated 197.8 172.5
2 High fat Control 63.7 33–40 122
Treated 388.8 131
3 Low fat Control 22.4 58 194.5
Treated 52 136
a
Controls were fed diets with either 11% high fat (Experiment 1 and
2) or 1.2% low fat (Experiment 3). Treated mice received the same
diet but containing 0.5 mg Orlistat/g chow. Feces was collected from
the cages of the various groups
Lipids
123
Page 7
within 2 h. This could reflect that there is a limited pool of
cellular substrates for the autophagic response. The extent
to which the autophagy damaged the cells is unclear, but it
is clear that apoptosis was activated by 4–8 h of culture,
depending on the concentration of Orlistat. Early signs of
apoptosis, cellular blebbing, were observed in as little as
4 h following treatment with 500 lM Orlistat substituted
DM and 12 h exposure to 100 lM Orlistat substituted DM.
The apoptosis is linked to the nutritional deficiency since
adding a source of fatty acids to the media suppressed its
progression and reversed the signs of cellular blebbing.
Apparently the cells can recover, re-attain normal viability,
from initiated apoptosis. However, there is clearly a
threshold of apoptosis inducing cell injury beyond which
the cells cannot recover.
Knowles and coworkers [9] describe in a paper that the
apoptosis induced in mammary and prostate cell tumor
lines by Orlistat was dependent on inhibition of FAS but
independent of fatty acid availability. This paradox is
explained by a pleiotropic consequence of inhibiting FAS
that involves up-regulation of a stress responsive gene,
DDIT4, which inhibits the mTOR pathway. Our findings
with the NUGC-3 gastric tumor cells are more consistent
with apoptosis arising from nutritional stress due to
decreased availability of fatty acids. Addition of lipid-
containing media to NUGC-3 cells exposed to Orlistat for
4 h halted progression of apoptosis over the next 4 h. Even
though Orlistat was removed from the media in the second
incubation, FAS continued to be inhibited (Fig. 2b). Thus,
in the face of continued inhibition of FAS, whole media
alone (lipids) halted apoptosis. The differences between
our findings and those of Knowles et al. [9] could be related
to differences in the duration of incubation with Orlistat.
Eight hours exposure of NUGC-3 cells to 100 lM Orlistat
clearly induced apoptosis (Fig. 5e). Perhaps a higher level
of stress develops with 24 and 48 h of continuous exposure
that activated the DDIT4 initiated apoptosis. Although we
used higher concentrations of Orlistat, their observations at
50 lM and ours at 100 lM should be similar. Our use of a
gastric cell line may also have contributed to differences
between our findings. As stated above, if Orlistat is to have
a role in treating cancer it would be most likely for gas-
trointestinal tumors that could come in direct contact with
the orally administered drug.
The greater sensitivity to inhibition of NUGC-3 cells in
serum deficient media was likely due to the low lipid
content of this media, since viability was greater in WM.
This observation is consistent with FAS being Orlistat’s
target in cancer cells and Orlistat being more toxic to the
cancer cells when they are wholly dependent on FAS to
supply the fatty acids needed for phospholipid synthesis
and, therefore, membrane formation.
Because increased cell death was seen in cultured cancer
cells treated with Orlistat in DM, we applied this infor-
mation to one of our in vivo experiments. Orlistat was fed
in a very low fat diet in experiment 3. Animals were
58 days of age at the start. Previous work in our laboratory
had showed that when added to a lipid emulsion, Orlistat
partitioned into the lipid phase (data not shown). Given this
result, we reasoned that Orlistat might similarly partition
into the increased intestinal fat content generated by inhi-
bition of lipases, and therefore, be less available to the
tumors. Therefore, Orlistat was administered in a very low
fat diet in the third experiment with the goal of enhancing
drug exposure to gastrointestinal tumors. No statistically
significant differences were seen between control and
Orlistat treated groups (Fig. 7 bottom panel, Table 2).
Although no significant difference was observed, the
Fig. 6 Distribution of Orlistat in the gastrointestinal tract estimated
by HPLC analysis. Relative concentration of Orlistat in the GI tract at
4 h after a mouse was given (by gavage) 50 mg/kg of Orlistat in
aqueous emulsion. Nearly all of the drug was found in the large
intestine and colon. Similar results were seen in a replicate
experiment. The Orlistat standard (10 lg injected in 100 ll) eluted
at about 6 min. Zones: 1 Stomach, 2 Upper small intestine, 3 Lower
small intestines, 4 Large intestines and colon
Lipids
123
Page 8
median survival time for untreated mice fed the low fat diet
exceeded that of the Orlistat treated group. This data sug-
gest that limiting exogenous sources of fatty acids via diet
restrictions may offer potential in treating gastrointestinal
cancers. However, a significant difference was seen when
Orlistat was supplemented into the high fat diet. This data
suggest that when supplemented into a high fat diet, Orli-
stat produced more beneficial results than supplemented
into an already fat limited diet.
In conclusion, Orlistat appeared to slightly prolong the
survival of mice with a fatal-genetic gastrointestinal can-
cer. There was no evidence that Orlistat accelerated death
in these tumor-bearing animals. If the findings are appli-
cable to humans, they support the idea that Orlistat inhibits
this type of cancer. Long-term survival studies in rats and
other mice strains are needed to further evaluate the
potential of Orlistat to inhibit GI cancer.
Acknowledgments This work was funded by grants from the A.T.
Still University of Health Sciences.
References
1. Kuhajda FP (2006) Fatty acid synthase and cancer: new appli-
cation of an old pathway. Cancer Res 66:5977–5980
2. Kuhajda FP (2000) Fatty-acid synthase and human cancer: new
perspectives on its role in tumor biology. Nutrition 16:202–208
3. Milgraum LZ, Witters LA, Pasternack GR, Kuhajda FP (1997)
Enzymes of the fatty acid synthesis pathway are highly expressed
in in situ breast carcinoma. Clin Cancer Res 3:2115–2120
4. Pizer ES, Pflug BR, Bova GS, Han WF, Udan MS, Nelson JB
(2001) Increased fatty acid synthase as a therapeutic target in
androgen-independent prostate cancer progression. Prostate
47:102–110
Fig. 7 Survival curves of control and Orlistat treated APC-min/
? mice. In experiments 1 and 2 mice were fed an 11% fat diet
containing no Orlistat or 0.5 mg Orlistat/g chow. In experiment 3
mice were fed a 1.2% fat diet with and without 0.5 mg Orlistat/g
chow
Table 2 Statistical analyses of experiments
Experiment 1 Experiment 2 Experiment 3
Log-Rank (Mantel–Cox) Test
Chi Square 4.609 0.3990 0.8191
P Value 0.0318 0.5276 0.3654
Gehan–Breslow–Wilcoxon Test
Chi Square 3.938 2.201 3.530
P Value 0.0472 0.1379 0.0602
Fig. 8 Effect of Orlistat on the viability of bovine lens epithelial
cells. Cells in 96 well plates were cultured for 48 h with varying
concentrations of Orlistat added in either serum deficient DMEM
(0.1% calf serum), DM, or DMEM containing 10% serum, WM. Cell
viability was measured at hour 48 by the MTT assay. Values are mean
± SEM of 4 wells
Lipids
123
Page 9
5. Rossi S, Graner E, Febbo P, Weinstein L, Bhattacharya N, Onody
T, Bubley G, Balk S, Loda M (2003) Fatty acid synthase
expression defines distinct molecular signatures in prostate can-
cer. Mol Cancer Res 1:707–715
6. Pizer ES, Chrest FJ, DiGiuseppe JA, Han WF (1998) Pharma-
cological inhibitors of mammalian fatty acid synthase suppress
DNA replication and induce apoptosis in tumor cell lines. Cancer
Res 58:4611–4615
7. (2006) PDR guide to drug interactions, side effects, and indica-
tions. Thomson PDR, Montvale, NJ
8. Kridel SJ, Axelrod F, Rozenkrantz N, Smith JW (2004) Orlistat is
a novel inhibitor of fatty acid synthase with antitumor activity.
Cancer Res 64:2070–2075
9. Knowles LM, Axelrod F, Browne CD, Smith JW (2004) A fatty
acid synthase blockade induces tumor cell-cycle arrest by down-
regulating Skp2. J Biol Chem 279:30540–30545
10. Menendez JA, Vellon L, Lupu R (2005) Orlistat: from anti-
obesity drug to anticancer agent in Her-2/neu (erbB-2)-over-
expressing gastrointestinal tumors? Exp Biol Med (Maywood)
230:151–154
11. Souri E, Jalalizadeh H, Kebriaee-Zadeh A, Zadehvakili B (2007)
HPLC analysis of orlistat and its application to drug quality
control studies. Chem Pharm Bull (Tokyo) 55:251–254
12. Jacoby RF, Seibert K, Cole CE, Kelloff G, Lubet RA (2000) The
cyclooxygenase-2 inhibitor celecoxib is a potent preventive and
therapeutic agent in the min mouse model of adenomatous pol-
yposis. Cancer Res 60:5040–5044
13. Chiu CH, McEntee MF, Whelan J (1997) Sulindac causes rapid
regression of preexisting tumors in Min/?mice independent of
prostaglandin biosynthesis. Cancer Res 57:4267–4273
14. Moser AR, Dove WF, Roth KA, Gordon JI (1992) The Min
(multiple intestinal neoplasia) mutation: its effect on gut epithe-
lial cell differentiation and interaction with a modifier system.
J Cell Biol 116:1517–1526
15. Taketo MM (2006) Mouse models of gastrointestinal tumors.
Cancer Sci 97:355–361
16. Hitchener WR, Cenedella RJ (1985) Absolute rates of sterol
synthesis estimated from [3H]water for bovine lens epithelial
cells in culture. J Lipid Res 26:1455–1463
17. Carmichael J, DeGraff WG, Gazdar AF, Minna JD, Mitchell JB
(1987) Evaluation of a tetrazolium-based semiautomated colori-
metric assay: assessment of chemosensitivity testing. Cancer Res
47:936–942
18. Sexton PS, Neely AR, Cenedella RJ (2004) Distribution of
caveolin-1 in bovine lens and redistribution in cultured bovine
lens epithelial cells upon confluence. Exp Eye Res 78:75–82
19. Biederbick A, Kern HF, Elsasser HP (1995) Monodansylcada-
verine (MDC) is a specific in vivo marker for autophagic vacu-
oles. Eur J Cell Biol 66:3–14
20. Munafo DB, Colombo MI (2001) A novel assay to study
autophagy: regulation of autophagosome vacuole size by amino
acid deprivation. J Cell Sci 114:3619–3629
21. Darzynkiewicz Z, Bedner E, Smolewski P, Lee BW, Johnson GL
(2002) Detection of caspases activation in situ by fluorochrome-
labeled inhibitors of caspases (FLICA). Methods Mol Biol
203:289–299
22. Grabarek J, Amstad P, Darzynkiewicz Z (2002) Use of fluores-
cently labeled caspase inhibitors as affinity labels to detect acti-
vated caspases. Hum Cell 15:1–12
23. Daniel WW (2005) Biostatistics: a foundation for analysis in the
health sciences. Wiley, Hoboken, NJ
24. Akiyama S, Amo H, Watanabe T, Matsuyama M, Sakamoto J,
Imaizumi M, Ichihashi H, Kondo T, Takagi H (1988) Charac-
teristics of three human gastric cancer cell lines, NU-GC-2,
NU-GC-3 and NU-GC-4. Jpn J Surg 18:438–446
25. Garcia SB, Barros LT, Turatti A, Martinello F, Modiano P,
Ribeiro-Silva A, Vespucio MV, Uyemura SA (2006) The anti-
obesity agent Orlistat is associated to increase in colonic prene-
oplastic markers in rats treated with a chemical carcinogen.
Cancer Lett 240:221–224
26. Hao M, Maxfield FR (2000) Characterization of rapid membrane
internalization and recycling. J Biol Chem 275:15279–15286
Lipids
123
Page 10
    • "Orlistat that was developed as an anti-obesity drug is an efficient irreversible inhibitor of FAS as well. Orlistat exerts anti-cancer effects both in in vitro and in vivo models of different tumour histotypes (prostate, gastric, melanoma, breast)5657585960 , however its use still has some pharmacological limitations due to low solubility, cell permeability and poor selectivity. Another class of FAS inhibitors is small molecule cerulenin and its analogue C75, both of which have shown potent anti-tumour activity. "
    [Show abstract] [Hide abstract] ABSTRACT: Cancer cells metabolically adapt to undergo cellular proliferation. Lipids, besides their well-known role as energy storage, represent the major building blocks for the synthesis of neo-generated membranes. There is increasing evidence that cancer cells show specific alterations in different aspects of lipid metabolism. The changes of expression and activity of lipid metabolizing enzymes are directly regulated by the activity of oncogenic signals. The dependence of tumour cells on the deregulated lipid metabolism suggests that proteins involved in this process could be excellent chemotherapeutic targets for cancer treatment. Due to its rare side effects in non-cancerous cells, metformin has been recently revaluated as potential anti-tumorigenic drug, which negatively affects lipid biosynthetic pathways. In this review we summarised the emerging molecular events linking the anti-proliferative effect of metformin with lipid metabolism in cancer cells.
    No preview · Article · Apr 2014 · Biochimica et Biophysica Acta (BBA) - Reviews on Cancer
  • Source
    • "FASN expression is up-regulated in the early steps of breast cancer and represents a therapeutic target for breast cancer metastasis [40,41] and liposarcoma [42]. Inhibition of FASN suppressed the growth of cancer stem-like cells in breast cancer [43] and colon cancer [44], and induced apoptosis in diffuse large B-cell lymphoma [45] and in gastric-tumor-bearing mice [46]. CD44 knockdown was also associated with down-regulation of heat shock transcription factor 1 (HSF1) to a level similar to that seen in non-BCSCs. "
    [Show abstract] [Hide abstract] ABSTRACT: Breast cancer stem cells (BCSCs) are the source of breast tumors. Compared with other cancer cells, cancer stem cells show high resistance to both chemotherapy and radiotherapy. Targeting of BCSCs is thus a potentially promising and effective strategy for breast cancer treatment. Differentiation therapy represents one type of cancer stem-cell-targeting therapy, aimed at attacking the stemness of cancer stem cells, thus reducing their chemo- and radioresistance. In a previous study, we showed that down-regulation of CD44 sensitized BCSCs to the anti-tumor agent doxorubicin. This study aimed to determine if CD44 knockdown caused BCSCs to differentiate into breast cancer non-stem cells (non-BCSCs). We isolated a breast cancer cell population (CD44+CD24- cells) from primary cultures of malignant breast tumors. These cells were sorted into four sub-populations based on their expression of CD44 and CD24 surface markers. CD44 knockdown in the BCSC population was achieved using small hairpin RNA lentivirus particles. The differentiated status of CD44 knock-down BCSCs was evaluated on the basis of changes in CD44+CD24- phenotype, tumorigenesis in NOD/SCID mice, and gene expression in relation to renewal status, metastasis, and cell cycle in comparison with BCSCs and non-BCSCs. Knockdown of CD44 caused BCSCs to differentiate into non-BCSCs with lower tumorigenic potential, and altered the cell cycle and expression profiles of some stem cell-related genes, making them more similar to those seen in non-BCSCs. Knockdown of CD44 is an effective strategy for attacking the stemness of BCSCs, resulting in a loss of stemness and an increase in susceptibility to chemotherapy or radiation. The results of this study highlight a potential new strategy for breast cancer treatment through the targeting of BCSCs.
    Full-text · Article · Dec 2011 · Journal of Translational Medicine
  • [Show abstract] [Hide abstract] ABSTRACT: First Page of the Article
    No preview · Conference Paper · Mar 1986
Show more