University of Southern California Dissertations and Theses

The differential effects of selective COX-2 inhibitors on cell proliferation and induced ER stress in glioblastoma and pancreatic carcinoma cell lines

(USC Thesis Other)

The differential effects of selective COX-2 inhibitors on cell proliferation and induced ER stress in glioblastoma and pancreatic carcinoma cell lines

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

THE DIFFERENTIAL EFFECTS OF SELECTIVE COX-2 INHIBITORS ON CELL
PROLIFERA TION AND INDUCED ER STRESS IN GLIOBLASTOMA AND
PANCREA TIC CARCINOMA CELL LINES

by

Huan-Ching Chuang

A Thesis Presented to the
FACULTY OF THE GRADUA TE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(BIOCHEMISTRY & MOLECULAR BIOLOGY)

May 2008

Copyright 2008 Huan-Ching Chuang

ii
Acknowledgments
I especially would like to thank my advisor, Dr. Axel H. Schönthal, for giving me
advice on the thesis and giving valuable feedback on all the experiments. I also want
to thank Dr. Adel Kardosh and Dr. Peter Pyrko for instructions on lab procedures and
providing sample theses; Dr. Hee-Yeon Cho for experimental feedback and
troubleshooting problems; and Dr. Encouse Golden for technology instruction. The
USC Glioma Research Group is acknowledged for stimulating discussion.

iii
Table of Contents
Acknowledgments ii
List of Figures iv

Abstract v

Introduction 1
Hypothesis and Objective 7

Materials and Methods 9

Results 14

Discussion 22

References 31

iv
List of Figures
Figure 1. Survival and Cell Death by ER Stress and Unfolded Protein Response 5

Figure 2. Model of ER Stress and Its Connection to Cancer and Chemotherapy 4
Figure 3. Chemical Structures of CXB, UMC, and DMC 14
Figure 4. MTT Assays of Three Glioblastoma Cell Lines 16
Figure 5. Western Blot of ER Specific Markers in U251 Cell Line 17
Figure 6. Western Blot of ER Specific Marker and Survivin in U251 Cell Line 18
Figure 7. MTT Assays of Two Pancreatic Cell Lines 19
Figure 8. Western Blot of ER Specific Marker in Pancreatic Cell Lines 20
Figure 9. Total Cell Number after Long-Term Treatment with CXB 21

v
Abstract
Because many reports now challenge the notion that celecoxib (Celebrex
®
)
execute its antitumor effect in the absence of COX-2 inhibitory activity, the aim of the
project is to investigate COX-2 involvement with respect to its antitumor potential.
We have studied two close structural analogs of celecoxib, DMC and UMC, which
differentially display COX-2 inhibitory activity. UMC has COX-2 inhibitory function,
while DMC has antitumor potential. In our results, DMC exhibited the most potent
antitumor activity, while UMC was the least effective. During acute treatment, the
three drugs and all tested cell lines were able to trigger endoplasmic reticulum stress,
as indicated by the induction of CHOP, GRP78, and caspase-7 activation. Thus, we
concluded that the antitumor activity of celecoxib is independent of COX-2 inhibitory
activity and relies on the ability to trigger ER stress. This finding confirms previous
studies and can be beneficial to anti-cancer therapy.

1
Introduction
Cyclooxygenase inhibitors have been implicated in both treatment and
prevention of cancer (16, 34). These inhibitors, traditionally known as the
non-steroidal anti-inflammatory drugs (NSAID) including aspirin, ibuprofen, and the
new COX-2 inhibitors, are used to treat patients with rheumatoid arthritis, fever, and
pain. Yet their use is somewhat limited due to their unwanted side effects particularly
in the gastrointestinal (GI) tract, as well as cardiovascular complications (10, 32).
The mechanism by which these NSAIDs exert their effects is the inhibition of
prostaglandin synthesis via cyclooxygenase (COX or more properly called
prostaglandin H synthase) enzymes, which are required to catalyze the conversion of
arachidonic acid to prostaglandins. There are at least two distinct isoforms of COX,
which are COX-1 and COX-2. Recent studies have shown that COX-1 is
constitutively expressed in many tissues such as the GI tract and kidney, and is
involved in the production of prostaglandin that participates in the normal cell activity.
On the other hand, COX-2 is the inducible form and is responsible for synthesizing
prostaglandin primarily at the sites of inflammation in response to cytokines and
growth factors (11, 14, 26, 29, 31, 45). In addition to elevated COX-2 levels at
inflammatory sites, the COX-2 level is greatly induced in the tissues such as
post-implanted ovaries, or in colon adenoma and carcinoma cells (29). Moreover, the
level of COX-2 is elevated in various types of cancer cell such as colon and
pancreatic cancer (13, 44). Thus, overexpression of COX-2 might lead to the
development of cancer, which involves an increase in cell division, decrease in
apoptotic process, promotion of metastasis, and growth of neovascularization (13, 18,
30, 49).

2
The elevated levels of COX-2 in cancer cells suggest the need for compounds
that solely block COX-2 without altering the function of COX-1. Many NSAIDs are
being evaluated for their antitumor effect. The traditional NSAIDs such as sulindac
and indomethacin inhibit both COX-1 and COX-2, whereas new selective NSAID
classes such as celecoxib (Celebrex
®
), valdecoxib (Bextra
®
) and rofecoxib (Vioxx
®
)
specifically inhibit only COX-2. The highly selective COX-2 inhibitors, referred to as
coxibs, promise to offer the therapeutic benefit of traditional NSAIDs with less of the
associated side effects (24).
Celecoxib, also known as Celebrex
®
, was first developed as an
anti-inflammatory drug and analgesic agent. It is now U.S. FDA-approved and used to
relieve symptoms of osteoarthritis rheumatoid and to treat familial adenomatous
polyposis (FAP) (14, 37). The efficacy of celecoxib in reducing colorectal
adenomatous polyps has been promising (3, 5, 47). Several studies have suggested the
possibility of celecoxib in prevention and treatment of other types of cancer (16, 24).
In addition, celecoxib has been shown to have its antitumor effect in vivo, as seen in
different animal models with great reduction of tumor growth (9).
Despite these promising results, the mechanism by which celecoxib exerts its
cell-killing function has remained controversial due to an increasing number of
reports that challenge the notion of celecoxib exerting its anti-proliferative effect in
the absence of COX-2 (15, 23, 41). For example, four selective COX-2 inhibitors
applied to oral squamous cell carcinoma all show increased levels of COX-2 protein
expression, but only celecoxib appears to induce apoptosis, as it is evident in
caspase-3/7 activation, DNA fragmentation and poly(ADP-ribose) polymerase (PARP)
cleavage (27).

3
In addition to its COX-2 inhibitory function, CXB perhaps possesses a second
function, which is the antitumor effect; these two functions might work independent
of each other (40, 42). To further investigate the relationship between these two
functions of CXB, a close structural analog of CXB called 2,5-dimethyl-celecoxib
(DMC) was developed first by the group of Ching-Shih Chen at Ohio State University
(46). To compare the chemical structures, celecoxib is designated as 4-[5-(4-
methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benzenesulfonamide and has a
methyl group at position 4 of the terminal phenyl ring. In contrast, DMC is designated
as 4-[5-(2,5-dimethylphenyl)-3 trifluoromethyl)-1H-pyrazol-1-yl]benzenesulfonamide
and has two methyl groups positioned at 2 and 5 of the terminal phenyl ring (Figure 3,
50). From the crystal structure COX-2 complex, CXB with one methyl group is fits
well into the active site of the COX-2 enzyme. However, the extra methyl group on
DMC makes the molecule too bulky, resulting in no binding to the enzyme (12). This
evidence explains the lack of COX-2 inhibitory function in DMC.
Despite the lack of COX-2 inhibitory activity, DMC is able to potently promote
the antitumor effect in a similar fashion as CXB in both vitro and vivo. DMC,
however, is able to inhibit tumor growth more potently than CXB (22). Both CXB and
DMC exert their antitumor effect by inhibiting certain signal transduction pathways,
blocking cell proliferation, and inducing apoptosis at a much lower concentration than
the traditional NSAIDs (4, 20, 22).
A number of papers have provided explanations to why CXB and DMC have
such strong and potent antitumor effects as opposed to the standard NSAIDs. One of
the best-explained mechanisms by which CXB and DMC exert this function is
through sarcoplasmic/endoplasmic reticulum (ER) calcium ATPase (SERCA). This
transmembrane protein resides in the ER and is responsible for transporting Ca
2+
from

4
cytosol into the ER, thereby maintaining the Ca
2+
gradient between ER compartment
and cytosol (17, 36).
As described in an earlier study, the elevated level of intracellular Ca
2+
([Ca
2+
]
i
)
is the way in which celecoxib triggers rapid apoptosis in tumor cells. The increase in
Ca
2+
in ER indicates the blockage of SERCA by CXB (17, 48). Disruption of ER
homeostasis via Ca
2+
leakage affects protein folding and causes ER stress. ER
stress-induced cell apoptosis by CXB is evident in animal tumor model (25, 36).
CHOP/GADD153 (CCAAT/enhancer binding protein homologous transcription
factor/growth arrest and DNA damage-inducible gene 153) is one of the components
in the ER stress-mediated apoptosis pathway (33). Besides CHOP marker, the other
main executioner of the pro-apoptotic arm of ER stress is caspase-7, which are
stimulated by CXB (25). In addition, the unfolded protein response is triggered by ER
stress and this response is in the early phase of ER stress for cell survival and
long-term ER stress adaptation. A chaperone protein, called glucose-regulated protein
of 78 KDa (GRP78), resides in the ER and binds to misfolded proteins under stress.
Its action is required for reversing ER stress and facilitating protein folding (28, 38).
A cell will determine to either induce GRP78 for survival pathway or induce CHOP
for apoptotic pathway depending on the degree of severity of ER stress. Because
cancer cells are in chronic stress, the target to SERCA by CXB has a great advantage
to kill tumor cells (Figure 1 and 2). Moreover, CXB is favored over SERCA blockers
such as thapsigargin because it is less cytotoxic and will less likely to both normal and
tumor cells (36). Taken all together, the inactivation of the function of SERCA,
subsequent ER stress and unfolded protein response all point to the idea of a
celecoxib-mediated antitumor effect in the absence of COX-2 activity.

5

Figure 1. Survival and Cell Death by ER Stress and Unfolded Protein Response. Factors such as
hypoxia, low glucose, elevated Ca
2+
, and increased levels of misfolded proteins will induce ER stress
response. Depending on severity, cell will decide to either induce GRP78 for survival or CHOP for
apoptosis (model created by Axel H. Schönthal, PhD, 2007).

Since both CXB and DMC are apoptotic inducers, survivin will be a protein of
interest to study the antitumor effect in these two drugs. Survivin is a member of the
inhibitors of apoptosis (IAP) family of proteins and plays a role in cell division and
apoptosis. Its main function in mitosis is to preserve mitotic apparatus, allowing
proper mitosis progression (1). In addition, its function is not limited to cell division
but is implicated with apoptosis inhibition. From a study, the overexpression of
survivin has been associated with apoptosis inhibition, as seen in mostly human
cancer cells such as lung and prostate (2). Its anti-apoptotic function is achieved
through the inhibition of caspase-3 and caspase-7 activity. Moreover, survivin appears
to be associated with increased resistance to conventional therapy (1, 7). Thus, it is an
important marker to be investigated in terms of antitumor function in CXB and DMC.
Besides survivin, Poly(ADP-ribosyl)ation (PARP) is another protein of interest
during the experiment, since inactivation of this protein leads to apoptosis. PARP
maintains normal cell function and viability. It is the first characterized substrate of

6
caspase-3 that shares a homologous region with a large subunit of Replication Factor
C (39). It is implicated that caspase-3 can mediate cell cycle arrest. The cleavage of
PARP inhibits normal function of DNA replication. It is also the key effector of DNA
repair, replication and transcription reactions (6, 8). Thus, it is important to investigate
the level of PARP in order to study cells that are undergoing apoptosis.

Figure 2. Model of ER Stress and Its Connection to Cancer and Chemotherapy. There are three
major activity levels of the ER stress response system as summarized in Figure 2. (1) The “No ER
Stress” condition (left) is the default situation in normal cells. Here, very low levels of GRP78 in the
ER lumen bind to PERK, IRE1, and A TF6, and keep these ER stress components in their inactive states.
(2) In cancer cells, chronic stress (low glucose, hypoxia, misfolded proteins) generates the “Low ER
Stress” condition (middle). Here, low levels of continuous stress lead to partial activation of the ER
stress response system, with emphasis on elevated levels of GRP78 (thickened arrows), which provide
the protective component of the ER stress system and furthermore increase chemoresistance of tumor
cells. In this scenario, the ER transmembrane components PERK, IRE1, and A TF6 display low levels
of activity, which presumable are fine-tuned and adjusted by GRP78. Elevated levels of GRP78 help to
neutralize deleterious effects of the initial stress condition, such as through the ability to act as a
chaperone for misfolded proteins. (3) Persistent, high level stress generates the “Severe ER Stress”
situation (right), which is characterized by severe (but transient) inhibition of protein synthesis via
PERK-mediated phosphorylation (=inactivation) of translation initiation factor 2 alpha (eIF2α) (thick
lines). Under these conditions, the protective effort of GRP78 is overwhelmed, and the activation of
pro-apoptotic CHOP and subsequent initiation of cell death dominates (very thick arrows). Our work
has demonstrated that during treatment of tumor cells with DMC, the “Severe ER Stress” scenario
applies. In the continuous presence of this drug, the cell is unable to neutralize drug-induced stress,
despite elevated levels of GRP78; instead, strong induction of CHOP and activation of caspase 4
initiate apoptosis, and the cell dies (model created by Axel H. Schönthal, PhD, 2007).

7
To further confirm the mechanism by which CXB exerts its antitumor effect
without the involvement of COX-2 activity, we have obtained another critical close
structural analog of CXB with in vitro antitumor potential and slightly higher COX-2
inhibitory activity. This analog is called unmethylated celecoxib (UMC) and lacks the
methyl group at position 4. It is shown to have about 20% higher COX-2 inhibitory
activity than CXB (35). UMC is applied to cell culture in order to further generalize
the relationship between antitumor effect and COX-2 inhibitory activity.
The main objective of this paper is to investigate the differential effects and
response of CXB, DMC and UMC in vitro tumor cells. We hypothesize that UMC
will mediate apoptosis through the ER stress mechanism and COX-2 independent
pathway, as previously shown in both CXB and DMC. Since the role of UMC has not
yet been explored, its role in antitumor activity will be critical and might be important
for chemoprevention in terms of its degree of side effects. We employ all three drugs
in the same cell lines under the same conditions in parallel in order to obtain a more
accurate result. Moreover, since we already have much compelling evidence of CXB’s
main executing mechanism and well-established results from DMC, UMC will be our
main experimental factor to discover the role of COX-2 inhibitory activity in
ER-medicated cell death in short-term treatment, and suppression of focus formation
in long-term treatment.
To further establish the mechanism by which the three drugs engage in their
antitumor activity, we used two pancreatic cell lines that have been characterized by
their COX-2 level. Bx-PC3 is overexpressed with COX-2, whereas MIA-PaCa-2 is
negative for COX-2 (43). With previous studies indicating COX-2 inhibition is not
required for exerting an antitumor effect, we would expect that the results in these two
cell lines should be similar. Indeed, our results indicate that the difference between

8
these two cell lines is negligible, and thus it can be concluded that COX-2 inhibitory
function does not play a significant role.
In our short-term studies (less than 48 hours), our results showed that DMC is
the most potent anti-proliferative and cell-killing mediated compound, whereas UMC
is the least effective compound in terms of its antitumor function (DMC>CXB>UMC).
The trend of their antitumor function is exactly the opposite of their COX-2 inhibitory
function, such that UMC is the strongest COX-2 inhibitor and DMC has almost no
COX-2 inhibitory function. Our studies also point out that despite their differential
effects on COX-2 activity, ER stress appears to be more important to their antitumor
effects. This is evident in the increased expression of ER stress specific markers such
as CHOP and caspase-7 in short-term treatment. In addition to short-term treatment,
we looked at the effects of CXB, DMC, and UMC in long-term treatment (up to 3
months) at sub-toxic concentrations in order to evaluate the change in phenotypic
expression. We found out that at these concentrations, cells are able to restore contact
inhibition and inhibit the formation of foci in tumor cells. In conclusion, the COX-2
inhibitory function appears to be insignificant in its involvement with ER
stress-medicated apoptosis in short-term drug application or focus suppression in
long-term treatment.

9
Materials and Methods
Materials
Celecoxib is 4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]
benzenesulfonamide (35). DMC is the first close structural analog, where the 5-aryl
moiety has been altered by replacing 4-methylphenyl with 2,5-dimethylphenyl,
resulting in 4-[5-(2,5-dimethylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]
benzenesulfonamide (22). UMC is the second close structural analog, where the
5-aryl moiety has been altered by eliminating 4-methylphenyl group, resulting in
4-[3-(trifluoromethyl)-1 H-pryzaol-1-yl]benzenesulfonamide (35). Celebrex
®
capsules
were either obtained from the pharmacy or were synthesized in our laboratory
according to previously published procedures. DMC and UMC are also synthesized in
our laboratory according to previous literature (22, 35). Each drug was dissolved in
DMSO at 100 mM (stock solution) and added to the cell culture medium in a manner
that kept the final concentration of solvent (DMSO) below 0.1%.
Cell lines and culture conditions
The glioblastoma cell lines: U251, LN229, and T98G were generously provided
by Dr. Frank B. Furnari and Dr. Webster K. Cavenee (Ludwig Institute of Cancer
Research, La Jolla, CA); and the pancreatic carcinoma cell lines MIA PaCa-2 and
BxPC-3 were obtained from Dr. Guido Eibl (UCLA, Los Angeles, CA). All cells
were cultured in Dulbecco’s modified Eagle’s medium (DMEM; GIBCO BRL, Grand
Island, NY) supplemented with 10% fetal bovine serine, 100 U/ml penicillin, and 0.1
mg/ml streptomycin (p/s; GIBCO) in a humidified incubator at 37°C and a 5% CO
2

atmosphere (21).

10
Immunoblots
After cells were cultured with specific drugs for various time periods, cells were
harvested by scraping directly on Petri dish, washing with PBS, and centrifuging at
14,000 rpm. Both floating and adhering cells were harvested to allow the detection of
caspase 7 activation. Total cellular lysates were prepared by lysing the collected cells
with RIPA buffer. The amount of RIPA buffer added depends on the size of the
pellets. Lysed cells were vortexed three times with 10-minute interval to ensure
complete cell lysis. Cells were kept on ice for all the above processes. Cells were
centrifuged for 15 minutes at 4°C. The pellets were discarded after centrifugation.
Protein concentrations were determined using the bicinchoninic acid (BCA) protein
assay reagent (Pierce, Rockford, IL). A standard curve was plotted (R
2
value greater
than 0.985) on excel to determine the volumes of each sample at 50 μg. For Western
blot analysis, a combination of 50 μg from each lysate, LaemmLi sample buffer, and
RIPA buffer was prepared. The mixture was vortexed and heated in a water bath at
95°C for 6 to 8 minutes to allow complete disruption of three-dimensional structure of
proteins. The mixture was then spun down before loading onto a gel. Various proteins
were separated by SDS-PAGE gel.
Antibodies
Chemiluminescence
The primary and secondary antibodies including anti-survivin, anti-CHOP,
anti-GRP78, anti-mouse, and anti-rabbit were purchased from Santa Cruz
Biotechnology, Inc. (Santa Cruz, CA). Anti-caspase-7 and anti-PARP were obtained
from Cell Signaling (Beverly, MA). According to manufacturer’s recommendations,
primary antibodies from Santa Cruz were diluted 1:500 in 5% nonfat dry milk in
PBS-Tween-20 (0.1%, v/v), and antibodies from Cell Signaling 1:1000 in 5% BSA in

11
PBS-Tween-20 (0.1%, v/v). Membranes were probed with specific diluted antibodies
and were incubated on shaker in sealed bags at 4°C overnight.
Primary-probed membranes were washed three times with PBS-Tween-20 on
shaker. Fresh PBS-Tween-20 was used for each wash. Only little amount of
PBS-Tween-20 was needed just to cover the surface of the membranes. The
secondary antibodies were diluted 1:5000 in 5% milk for Santa Cruz and in 5% BSA
for Cell Signaling. Membranes were incubate with secondary antibodies on shaker for
at least 45 minutes at 4°C. Membranes were washed three times with PBS-Tween-20
in the same way as described earlier.
To detect signals, the secondary antibodies were coupled to horseradish
peroxidase. Signals were detected by chemiluminescence using the SuperSignal West
substrate from Pierce. The membranes were first incubated with Pico
chemiluminescence solution for one minute before exposure. Kimwipes were used to
remove the excessive solution after one minute of incubation. Membranes were held
between two pieces of transparency and were placed in autoradiography cassette for
exposure. CL-XPosure films from Pierce were placed on top of the membranes and
this process was carried out in the dark room. Both long and short exposures were
necessary in order to obtain a clean signal. If signal came out relatively weak after
developing, certain percentage of Femto chemiluminescence was added to increase
signals. Several exposures were obtained to ensure consistency of the results.
Fluorescent Detection
Primary antibodies were the same as the ones described in chemiluminescence
section except these were diluted in Odyssey™ Blocking Buffer (LI-COR
Biosciences, Lincoln, NE). All procedures were followed according to protocol
supplied by the manufacturer (www.licor.com). With the exception of Odyssey™

12
Blocking Buffer, all the procedures were the same before adding secondary antibodies.
IR-labeled secondary antibodies were purchased from Li-Cor and were diluted
1:15000 in Odyssey™ Blocking Buffer. Since secondary antibodies are light sensitive,
the membranes had to be protected from light until they were scanned. The
membranes were incubated in dark at room temperature for least 45 minutes.
Membranes were washed with PBS-Tween-20 for three times before exposure. With
an Odyssey
TM
infrared imaging system, membranes were scanned in the appropriate
channels (700 nm or 800 nm) depending on the host species of the primary antibodies.
The level of signals was adjusted with different intensity settings. Several scans were
performed to ensure consistency of the results.
MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay
Both glioblastoma and pancreatic carcinoma cell lines were seeded 4.0x10
3
cells
per well in 96 well plates. The number of cells per unit volume was determined using
hemocytometer. The following day, seeded cells were supplemented with various
concentrations of drug-contained media. At least two wells were prepared for
background and contained no cells with fresh media. After 48 hours, 10 μL of MTT
dye was added to all wells. The reaction was terminated with 100 μL of solubilization
solution after four hours of incubation. Cell proliferation was measured with an
ELISA plate reader at 490 nm. At least triplicates were required to calculate the mean
values with percent errors.
Focus Formation Assays
Cells were seeded in a 6-well plates about 30% confluent. In the next day,
increasing concentration of drugs were added to the original plate. After about four
days, the control cells became fully confluent and thus its monolayer began to form.
At this time, all cells were supplemented with fresh media with drugs or no drugs

13
every two days until focus formation became apparent. Focus formation with CXB
treatment was quantitated by counting the number of foci per microscopic field.

14
Results
Figure 3 presents the actual chemical structures and properties of celecoxib and
its two close structural analogs. The COX-2 inhibitory function and antitumor
function of each compound appear to reside in different domains in the molecules.
The parent compound is CXB (celecoxib), which has one methyl group at position 4
of the terminal phenyl ring. UMC and DMC are the derivative compounds of
celecoxib. For DMC (2,5-dimethyl-celecoxib), an extra methyl group is added to
celecoxib, making 2 methyl groups at positions 2 and 5 of the phenyl ring. For UMC
(unmethylated-celecoxib), the methyl group is removed from position 4 of the
celecoxib. The COX-2 inhibitory potency is shown in this figure with IC50 of CXB at
40 nM, IC50 of UMC at 32 nM and IC50 of DMC at greater than 100,000 nM (35).

Figure 3. Chemical Structures of CXB, UMC, and DMC. CXB has one methyl group at the C-4 (p)
position of its terminal phenyl ring; this substitution is lacking in UMC; DMC has two methyl groups
(at the 2- and 5-positions). The listed COX-2 inhibitory potency (IC50) of these compounds is derived
from two earlier studies that used human recombinant COX-2 in vitro (35, 40)

In order to determine cell growth and survival, we treat three different
glioblastoma cell lines with increasing concentrations of CXB, UMC, and DMC and
determine cell viability by MTT assays. In all cases, DMC greatly reduces cell

15
viability and its IC50 is in the range of 35-45 μM. IC50 for CXB ranges from 55-65
μM. In contrast, UMC is the least effective and its IC50 is equal and greater than 100
μM. In all cases, DMC appears to be the most effective drug and CXB is about
20-30% less effective. UMC is the least potent drug in all three cell lines (Figure 4).
The trend of the cytotoxicity of each compound does not show any correlation to its
COX-2 inhibitory function.
As it was previously reported that ER stress might be involved in the mechanism
by which these drugs exerts their cell-killing function, ER stress-specific markers
GRP-78 and CHOP were investigated at sixteen-hour time point. At this time point,
earlier events in ER stress response can be revealed. Induction of GRP-78 level
indicates a protective mechanism cells use to adapt to ER stress (28, 38). The level of
GRP-78 is elevated in all cases compared to the control. The GRP-78 induction
occurs at 50 μM CXB, and higher induction at 75 μM CXB and 100 μM CXB.
However, GRP-78 levels for the two higher concentrations of CXB are about the
same. For UMC, the level of GRP-78 is induced at 50 μM and a little higher at 100
μM. Again, DMC displays the strongest effect, since its GRP-78 induction at 50 μM is
as effective as CXB at 75 μM. The second ER stress marker CHOP shows similar
trends among the three drugs. The level of CHOP induction at 40 μM DMC is about
the same as that at 60 μM CXB; 100 μM UMC shows the weakest effect (Figure 5).
As previously mentioned, the antitumor effect indeed correlates with the drugs’ ability
to induce ER stress.

16

Figure 4. MTT Assays of Three Glioblastoma Cell Lines. Three glioblastoma cell lines, U251,
LN229, and T98G , are treated with DMC, CXB, and UMC. Cell growth and survival is determined
after 48 hours by MTT assay. The yellow color indicates death cells, whereas the purple color
designates viable cells. 4*10
3
cells per well are seeded in a 96-well plate and cells are supplemented
with increasing drug concentrations as indicated above. MTT dye is added after 48 hours of incubation
with drugs and another 4 hours of incubation with MTT dye is needed before the final termination of
the reaction with solubilization solution. Cell viability is measured with an ELIZA plate reading at 490
nm. The values shown are the mean (±SD) of three measurements.

To investigate the mechanism by which celecoxib and its derivatives induce
apoptosis, we chose several markers based on previous reported results. We determine
the effects of these drugs on the levels of survivin, activated caspase-7, and caspase-7
induced PARP cleavage by Western blot analysis at a 48-hour time point. This time

17
point reveals latter events of ER stress-induced apoptosis. We first looked at the
expression of survivin. The function of survivin in cell cycle is to maintain the mitotic
apparatus and ensure normal cell progression. It also has an anti-apoptotic function,
which is executed via caspase inhibition (1). The results show that survivin expression
for all drug treatments is downregulated, as shown in Figure 6. The complete
downregulation of survivin occurs at 60 μM DMC, 75 μM CXB, and 100 μM UMC.
DMC treatment has the strongest effect, whereas UMC treatment has the weakest
effect. The survivin has the ability to promote caspase activation. Thus, when survivin
is at the lowest level, caspase-7 is activated.

Figure 5. Western Blot of ER Specific Markers in U251 Cell Line. Drug treatments with Celecoxib
(CXB), UMC, and DMC up-regulate GRP78 and CHOP protein. The cell line U251 is cultured in the
presence of the three drugs for 16 hours. Total cellular lysates are prepared for Western blot analysis
with specific antibodies against GRP78, CHOP , and actin protein. In the GRP78 blot, there is a
non-specific band below the actual band itself.

As shown in Figure 6, all drugs have minimal effect on the basal level of
caspase-7, and thus the inactive form of caspase-7 appears to be at a constant level.
However, there is an apparent change in the active form of caspase-7. Figure 6 shows
that as the survivin decreases, there is a comparative increase in caspase-7 activation.
At the level of complete downregulation of survivin, there is a total upregulation of

18
active caspase 7. In the case of caspase-7 activation, 60 μM of DMC has an effect
similar to that of 75 μM CXB; and UMC display weaker effects even at 100 μM. We
further investigated the level of PARP (poly-ADP-ribose polymerase), which is the
substrate for caspase-7 that involved in the formation of poly(ADP-rose), DNA repair,
replication and transcription reactions (39). The cleavage of PARP indicates cells
undergoing apoptosis. The level of PARP cleavage increases with increasing
concentrations of all drugs. However, the effect of drug potency varies between drugs.
DMC still displays the strongest effect and UMC has the least effect on PARP
cleavage; CXB has a slightly weaker effect than DMC. Overall, the effects of these
three drugs on ER stress and apoptosis agree with their ability to trigger antitumor
effects, not COX-2 inhibitory effects.

Figure 6. Western Blot of ER Specific Marker and Survivin in U251 Cell Line. Protein level of
survivin is down-regulated and caspase 7 and PARP are activated in all cases. The cell line, U251
(glioblastoma), is cultured in the presence of the indicated three drugs, Celecoxib(CXB), DMC, and
UMC with increasing concentrations for 48 hours. Total cellular lysates are prepared for Western blot
analysis with specific antibodies against survivin, caspase 7, PARP , and actin (a loading control).
DMSO (a solvent to dissolve drugs) is used as another control in addition to untreated cells.

To further examine the role of COX-2 in these three drugs, we applied the drugs
to two distinct pancreatic carcinoma cell lines that had been characterized by their
COX-2 level. The cell line BX-PC-3 possesses a high level of COX-2, while
MIA-PaCa-2 cells are negative for COX-2 expression (46, 50). Both cell lines were
treated with increasing concentrations of each drug, and MTT assay was performed to

19
determine cell viability and proliferation. As shown in Figure 7, both cells display
similar results regardless of their COX-2 levels. IC50 ranges for both cells are very
close to glioblastoma cell lines, as seen before in Figure 7. Cell viability at IC50 for
DMC ranges from 40 to 50 μM and indicates it is the most potent cytotoxic drug
surveyed. The IC50 for CXB ranges from 60 to 70 μM and is less potent than DMC.
UMC displays the weakest effect on cell proliferation and viability, and its IC50 is
slightly greater than 100 μM. The similar results of the two cell lines show that the
endogenous level of COX-2 does not influence the drugs’ ability to execute
cytotoxicity.

Figure 7. MTT Assays of Two Pancreatic Cell Lines. Two pancreatic carcinoma cell lines,
MIA-PaCa-2 and Bx-PC-3, are treated with DMC, CXB, and UMC. Cell growth and survival is
determined after 48 hours by MTT assay. The yellow color indicates death cells, whereas the purple
color designates viable cells. 4*10
3
cells per well are seeded in a 96-well plate and cells are
supplemented with increasing drug concentrations as indicated above. MTT dye is added after 48 hours
of incubation with drugs and another 4 hours of incubation with MTT dye is needed before the final
termination of the reaction with solubilization solution. Cell viability is measured with an ELISA plate
reading at 490 nm. The values shown are the mean (±SD) of three measurements.

20
The next experiment we performed was the application of DMC, CXB, and
UMC to MIA-PaCa-2 and BxPC-3, whose characterized features have been discussed
in the MTT assay section. With these cell lines, we looked at ER stress-specific
marker CHOP. In both cell lines, with increasing concentrations of drugs after 48
hours, there is induction of CHOP at 75 μM CXB and 60 μM DMC, as shown in
Figure 8. There is no significant induction of CHOP at 75 μM UMC. As previously
mentioned, DMC is the most potent compound among these drugs. Thus, a lower
concentration of DMC was applied to the cells. There is no difference in CHOP
induction between the two cell lines. The expression of COX-2 does not play a role in
ER stress response.

Figure 8. Western Blot of ER Specific Marker in Pancreatic Cell Lines. There is induction of
CHOP for both Celecoxib (CXB) and DMC treated cells in both cell lines, Mia-PACA-2 and Bx-PC-3.
Both cell lines are pancreatic carcinoma. Mia-PACA-2 is known as COX-2 positive, whereas Bx-Pc-3
is known as COX-2 negative. Both cells are incubated with three drugs for 20 hours. Total cellular
lysates are collected for Western blot analysis with specific antibodies against CHOP and actin protein.
The level of CHOP induction between two cell lines is about the same.

21
To give a quantitative figure for the focus inhibition experiment, both pancreatic
cell lines were treated with increasing concentrations of CXB for up to 4 weeks. Once
CXB is added to the monolayer, cell proliferation is reduced since there is observable
decrease in cell numbers per surface area. Cells are less crowded. In addition, contact
inhibition might be restored, as there are fewer cells shed into the medium (not
shown). The result shows that critical reduction in cell numbers occurs at 30 μM CXB
(Figure 9). Again, DMC shows itself as the most effective drug and UMC is evidently
the least potent compound in the long-term drug treatment (data not shown). There is
no significant difference in cell numbers between the two cell lines. The COX-2
expression is not involved in the change in phenotypic response, and thus CXB does
not utilize its COX-2 function to achieve these outcomes.

Figure 9. Total Cell Number after Long-Term Treatment with CXB. Cells were treated with
increasing concentrations of CXB and the total number of cells per well (6-well plate) was determined
after 4 weeks (shown is the average of two counts). In all instances, the phenotypic changes were
similar in the case of CXB, UMC, or DMC (not shown for all treatment conditions), except that DMC
was the most potent, and UMC the least potent, compound. These experiments were repeated with very
similar outcomes.

22
Discussion
The question of whether CXB executes its antitumor effect in the absence of
COX-2 inhibition remains controversial. It is important to understand the basic
mechanism of CXB that can be beneficial to many cancer therapeutic applications. In
the previous studies, there were many evidences that supported the notion of CXB
being able to suppress tumor growth without the involvement of its COX-2 inhibition
(15). In addition, the chemical properties of CXB suggest that the two functions in
this molecule might be separated (12). In this sense, it is interesting to discover close
structural analogs of CXB with different intrinsic functions. It is thought that these
side effects developed during chemotherapy might be due to long-term inhibition of
COX-2 activity. These CXB derivatives, such as DMC and UMC with separated
functions, should have advantages over CXB with fewer unwanted side effects, and
thus, are worthwhile discoveries for cancer therapeutic applications (50).
To compare the COX-2 inhibitory activity and antitumor activity side-by-side,
two close structural analogs of CXB with differential COX-2 expression were studied.
The first analog, DMC, is devoid of COX-2 inhibitory function and solely possesses
only antitumor function with stronger effect. On the other hand, UMC has a weaker
antitumor effect overall, but its COX-2 inhibitory function is about 20% more potent
than that of CXB (35). Thus, it will be interesting to study the separate functions for
antitumor purposes.
So far, only the functions of UMC have been revealed. However, the mechanism
by which UMC induces apoptosis has remained unsolved. Because UMC has a much
stronger COX-2 inhibitory function than others, an effort to closely examine its
apoptotic pathway and involvement of COX-2 is essential. As mentioned before, one
of the best-explained mechanisms that agree with the COX-2 independent pathway is

23
through ER stress-induced apoptosis (17, 36). Several ER-specific markers are
investigated to study the pathway of ER stress-induced cell death. From the results,
we conclude that despite the weak antitumor effect of UMC, it consistently induces
ER stress under drug treatments as both CXB and DMC, suggesting its role in ER
stress-induced apoptosis pathways (Figure 5 and 6).
Since many reports have proposed and established the COX-2 independent drug
treatment in short-term (25, 50), the focus should shift to long-term treatment and
address the question of whether COX-2 inhibition plays a role during chronic drug
treatment. Thus, in addition to our short-term drug exposure, we applied CXB and its
derivatives to tumor cells at their sub-toxic concentrations for up to 3 months and
observed morphological changes. Our results allow us to conclude that under this
condition, all drugs are able to restore contact inhibition and deter focus formation
(Figure 9 and other figures not shown).
In order to determine cell proliferation and survival, we applied CXB, UMC, and
DMC in parallel to glioblastoma cell lines at a 48-hour time point. Our results show
that the IC50 of DMC that inhibits cell growth is much lower than that of UMC and
CXB. This result indicates that DMC has the most potent antitumor activity and is
consistent with the previous study on DMC in both vivo and vitro (19, 22). The IC50
of CXB is in a range that is 20% higher than that of DMC, meaning the
anti-proliferative effect in this molecule is not as strong. As for UMC, its
anti-proliferative function is not as effective as that of CXB and DMC. This
experiment shows the differential effects on their anti-proliferative properties. Perhaps
this might suggest the role of COX-2 inhibitory function along with antitumor activity.
Regarding different cell lines, there is no significant difference among these three
glioblastoma cell lines (Figure 4). This suggests that cell lines from the same origin

24
lead to similar outcomes. Although MTT assay is not a very sensitive method, the
results are confirmed by comparison to previous studies, and indeed, the resulting data
are in a similar range (Figure 4 and 22).
To investigate the pathway by which UMC induces cell death, cells are incubated
with this compound for short-term treatment. Due to the previous studies that suggest
that perturbed ER homeostasis will lead to ER stress (17, 36), a few ER specific
pro-apoptotic proteins such as GRP78, CHOP, and caspase-7, were studied. From our
results, there is an induction of GRP78 and CHOP. Since GRP78 is a protective
protein that binds to unfolded proteins and is able to revert ER stress (28, 38), the
increased expression of GRP78 suggests that cells are placed in a more stressful
environment as the drug concentrations increase (Figure 5). However, the level of
GRP78 remains the same at higher concentrations. At 75 μM CXB and 100 μM, cells
are overwhelmed by these concentrations, and a higher concentration of either CXB
or UMC will have no differential effects. As for CHOP marker, the expression is
sequentially increasing as the drug concentration increases. CHOP is specific for ER
and is an indicator of cells undergoing apoptosis (33). The level of CHOP starts to
appear at 75 μM CXB, 75 μM UMC, and 50 μM DMC. As the drug concentrations
increase, there is a comparative increase in CHOP expression. From Figure 5, this
result is to be expected and it can be concluded that a higher drug dose is more
effective in promoting apoptosis. As for caspase-7, although there is no difference in
the pro-caspase-7 level, there is a significant increase in the active form of caspase-7
as concentration increases (Figure 6). Cleaved caspase-7 activates downstream
effector PARP, leading to DNA degradation or fragmentation (6, 8). The results imply
that the cells are undergoing ER stress and follow the ER-mediated cell death
pathway in a predictable manner.

25
Besides the ER stress pathway, it is important to examine the role of survivin and
its associated pathway leading to apoptosis during short-term drug treatment. As
previously mentioned, survivin is involved in maintaining proper cell progression and
essential mitotic apparatus (1). The level of survivin is completely suppressed at the
highest concentrations of each drug, indicating there is a decrease in cell survival and
its anti-apoptotic function is strongly inhibited (Figure 6). Since survivin is
downregulated, there will be no inhibition of caspases’ activities. As shown in Figure
6, the highest level of caspase-7 and lowest expression of survivin appear to be at the
same concentration. This result suggests that survivin has the ability to regulate or
suppress the activity of caspase-7. Moreover, PARP is the downstream effector of
caspase-7. There is a noticeable decrease in full-length PARP and a decrease in active
PARP, indicating that the full-length PARP is cleaved and converted to active PARP.
The result shows that the level of cleaved PARP corresponds to the level of active
caspase-7, meaning active caspase-7 effectively cleaves PARP, which leads to
apoptosis. Taken together, Figures 5 and 6 both show: 1) cell apoptosis via ER stress
mechanism; and 2) the response to UMC as consistent with what we expected, such
that cells undergo apoptosis through ER stress in a manner similar to that produced by
CXB and DMC. Although the antitumor effect of UMC is much weaker, UMC is still
able to induce ER stress and mediate cell death through this pathway.
In order to settle the controversial issue of COX-2 involvement in cell killing, we
have developed an experiment that allows us to further generalize the role of COX-2
during cell treatment. We use two pancreatic cell lines that have different
characterized levels of COX-2. MIA-PaCa-2 is negative for COX-2 expression,
whereas BX-PC-3 is positive for COX-2 expression. If these drugs execute
cell-killing effects in the absence of COX-2, then we would expect that there will be

26
no difference between the outcomes between these two cell lines. From the MTT
assays, the IC50, which shows cell viability at 50%, is in approximately the same
range for both cell lines (Figure 7). There is an insignificant difference between them.
In addition, to investigate the ER stress mechanism, we look at CHOP induction. The
presence of CHOP in both CXB and DMC treated lysates indicates ER stress (Figure
8). The absence of the signal in UMC might be due to its weaker effect in antitumor
function. The results from these two cell lines appear to be approximately the same,
challenging the notion of COX-2 involvement in antitumor effect. Taken together,
from the earlier discussed issue with ER stress and COX-2 examination, we can
deduce that the antitumor effect is independent of COX-2 inhibitory activity.
In addition, the results from Figures 7 and 8 agree with the role of DMC. DMC
lacks COX-2 inhibitory function, yet potently inhibits cell growth and survival.
Because of the lack of COX-2 inhibitory function, this compound helped previous
studies to establish the ER stress mechanism (20, 40). Not only is DMC effective in
cell culture experiments, but there is evidence from in vivo studies that DMC is able
to consistently work in the same way as in vitro response through ER
stress-medicated cytotoxicity. However, the drug concentrations in vivo are
comparatively lower, such as less than 10 μM in serum (36). Since both in vitro and in
vivo studies show the ER stress-mediated apoptosis mechanism, as well as the
unnoticeable difference between the two pancreatic cell lines, it is suggested that
COX-2 activity is more likely to be absent during antitumor activity executed by CXB,
UMC, and DMC.
The discussions on short-term treatment conclude with an undetectable level of
COX-2 involvement. However, long-term drug application has remained unanswered.
We suspect that the COX-2 independent mechanism through ER stress might only be

27
favorable in short-term treatment, and that cells might undergo the COX-2 dependent
pathway when cells do not undergo immediate apoptosis. For this reason, we applied
these three drugs to cells with sub-toxic concentrations, which will not induce ER
stress or cytotoxicity. After chronic treatment, the cells were able to restore contact
inhibition and inhibit the formation of focus developed by typical tumor cells (data
not shown). Figure 9 shows that the critical concentration of reduction in cell numbers
is between 20 μM and 30 μM and gives the quantitative figure. This is the range
where no ER stress will take place in short-term treatment. Indeed, at this sub-toxic
concentration, cell number is greatly reduced, suggesting the suppression of focus
formation.
To minimize the variations and external factors in MTT assays, all three drugs
are applied to the cells under the same conditions. One 96-well plate is employed for
each cell line, and different setups on a plate are essential due to the fact that
orientation might give different O.D. readings. We are limited in the number of
available wells, so many repetitions are necessary to obtain reliable results. Figures 4
and 7 represent the best results performed in a single experiment. A better, smoother
S-curve can be obtained by combining and averaging between all trials (not shown).
Despite the limitation on numbers of wells, a more important factor in variability
comes from the number of cells seeded in each well. In order to improve and
minimize this problem, a statistical technique is used. Cells need to be counted at least
three times, and cells are seeded randomly throughout the plate. Both repetitions and
random cell seeding are critical to ensure reliability and consistency of the data.
Troubleshooting with Western blot is important to any attempt to duplicate these
experiments in the future. The most critical factor is correct timing. During the gel
running process, a low voltage should be set in order to ensure that all lysates in each

28
lane run at the same speed. During the transfer process, for smaller proteins such as
survivin and caspase-7, it is best to transfer within a shorter time (60 minutes). Longer
transferring time increases the level of background. When dealing with
fluorescence-conjugated secondary antibodies, it is important to protect membrane
from light; otherwise, the background signal will increase significantly. Moreover,
bent or folded membranes that create a crease will result in higher background levels.
The membranes should be incubated with secondary antibodies in the shortest time
possible to reduce background signal. Timing is critical to obtaining a clean and
unquestionable Western blot result.
Some of the questions have remained unsolved, and thus more experiments can
be expanded in the future. Our results were obtained under in vitro conditions, and
this represents a weak and poor model to study clinical cancer therapeutic applications.
A future direction lies in animal models for drug treatment to avoid cell culture
artifacts. In addition, with the discovery of focus inhibition and restored contact
inhibitory ability during long-term conditions, the application of sub-toxic
concentrations of these drugs could possibly contribute to the reduction of tumor size
in animals with implanted tumors (22, 36). Moreover, various pathways that mediate
cell death should be studied, such that caspase independent pathway is just as
important as the ER stress-mediated pathway.
In conclusion, UMC is able to induce ER stress and undergo cell apoptosis in a
manner similar to CXB and DMC, namely through the COX-2 independent pathway.
This held true in all our tested cell lines (glioblastoma and pancreatic cell lines). DMC
and UMC only possess one common property; however, both of them are able to
trigger ER stress response. Moreover, the potency of COX-2 inhibitory function
among these three drugs is in inverse proportion to their anti-proliferative efficacy.

29
Due to the lack of correlation between these two functions, we thus conclude that
inhibition of COX-2 is not required to trigger growth-inhibitory and cytotoxic
response. This conclusion is consistent with previous suggested COX-2 independent
targets (15, 23, 41), as well as ER stress response induction during short-treatment (17,
22, 25, 33, 36, 48). This ER stress model adequately supported the notion of the
COX-2 independent pathway during acute treatment, while chronic treatment is
explained through the restoration of contact inhibition and inhibition of focus
development. This project indeed addressed the mechanism by which UMC induces
apoptosis, the ability of all drugs to execute antitumor effect without the involvement
of COX-2, and long-term treatment at sub-toxic concentrations that might suggest
fewer associated side effects. All these outcomes are clinically relevant and can be
further strengthened by the performance of suggested possible future experiments.

30
References
1. Altieri, D. C. 2003. Survivin, versatile modulation of cell division and
apoptosis in cancer. Oncogene 22:8581-9.
2. Ambrosini, G., C. Adida, and D. C. Altieri. 1997. A novel anti-apoptosis
gene, survivin, expressed in cancer and lymphoma. Nat Med 3:917-21.
3. Arber, N., C. J. Eagle, J. Spicak, I. Racz, P. Dite, J. Hajer, M. Zavoral, M.
J. Lechuga, P. Gerletti, J. Tang, R. B. Rosenstein, K. Macdonald, P.
Bhadra, R. Fowler, J. Wittes, A. G . Zauber, S. D. Solomon, and B. Levin.
2006. Celecoxib for the prevention of colorectal adenomatous polyps. N Engl
J Med 355:885-95.
4. Backhus, L. M., N. A. Petasis, J. Uddin, A. H. Schonthal, R. D. Bart, Y.
Lin, V . A. Starnes, and R. M. Bremner. 2005. Dimethyl celecoxib as a novel
non-cyclooxygenase 2 therapy in the treatment of non-small cell lung cancer. J
Thorac Cardiovasc Surg 130:1406-12.
5. Bertagnolli, M. M., C. J. Eagle, A. G. Zauber, M. Redston, S. D. Solomon,
K. Kim, J. Tang, R. B. Rosenstein, J. Wittes, D. Corle, T. M. Hess, G. M.
Woloj, F. Boisserie, W. F. Anderson, J. L. Viner, D. Bagheri, J. Burn, D. C.
Chung, T. Dewar, T. R. Foley, N. Hoffman, F. Macrae, R. E. Pruitt, J. R.
Saltzman, B. Salzberg, T. Sylwestrowicz, G. B. Gordon, and E. T. Hawk.
2006. Celecoxib for the prevention of sporadic colorectal adenomas. N Engl J
Med 355:873-84.
6. Boulares, A. H., A. G . Yakovlev, V. Ivanova, B. A. Stoica, G . Wang, S. Iyer,
and M. Smulson. 1999. Role of poly(ADP-ribose) polymerase (PARP)
cleavage in apoptosis. Caspase 3-resistant PARP mutant increases rates of
apoptosis in transfected cells. J Biol Chem 274:22932-40.
7. Cohen, C., C. M. Lohmann, G . Cotsonis, D. Lawson, and R. Santoianni.
2003. Survivin expression in ovarian carcinoma: correlation with apoptotic
markers and prognosis. Mod Pathol 16:574-83.
8. D'Amours, D., S. Desnoyers, I. D'Silva, and G . G. Poirier. 1999.
Poly(ADP-ribosyl)ation reactions in the regulation of nuclear functions.
Biochem J 342 ( Pt 2):249-68.

31
9. Diperna, C. A., R. D. Bart, E. M. Sievers, Y. Ma, V . A. Starnes, and R. M.
Bremner. 2003. Cyclooxygenase-2 inhibition decreases primary and
metastatic tumor burden in a murine model of orthotopic lung adenocarcinoma.
J Thorac Cardiovasc Surg 126:1129-33.
10. Drazen, J. M. 2005. COX-2 inhibitors--a lesson in unexpected problems. N
Engl J Med 352:1131-2.
11. Dubois, R. N., S. B. Abramson, L. Crofford, R. A. Gupta, L. S. Simon, L.
B. Van De Putte, and P. E. Lipsky. 1998. Cyclooxygenase in biology and
disease. FASEB J 12:1063-73.
12. Flower, R. J. 2003. The development of COX2 inhibitors. Nat Rev Drug
Discov 2:179-91.
13. Fosslien, E. 2000. Biochemistry of cyclooxygenase (COX)-2 inhibitors and
molecular pathology of COX-2 in neoplasia. Crit Rev Clin Lab Sci
37:431-502.
14. Geis, G. S. 1999. Update on clinical developments with celecoxib, a new
specific COX-2 inhibitor: what can we expect? J Rheumatol Suppl 56:31-6.
15. Grosch, S., T. J. Maier, S. Schiffmann, and G. Geisslinger. 2006.
Cyclooxygenase-2 (COX-2)-independent anticarcinogenic effects of selective
COX-2 inhibitors. J Natl Cancer Inst 98:736-47.
16. Howe, L. R., and A. J. Dannenberg. 2002. A role for cyclooxygenase-2
inhibitors in the prevention and treatment of cancer. Semin Oncol 29:111-9.
17. Johnson, A. J., A. L. Hsu, H. P. Lin, X. Song, and C. S. Chen. 2002. The
cyclo-oxygenase-2 inhibitor celecoxib perturbs intracellular calcium by
inhibiting endoplasmic reticulum Ca2+-A TPases: a plausible link with its
anti-tumour effect and cardiovascular risks. Biochem J 366:831-7.
18. Kakiuchi, Y., S. Tsuji, M. Tsujii, H. Murata, N. Kawai, M. Yasumaru, A.
Kimura, M. Komori, T. Irie, E. Miyoshi, Y. Sasaki, N. Hayashi, S.
Kawano, and M. Hori. 2002. Cyclooxygenase-2 activity altered the
cell-surface carbohydrate antigens on colon cancer cells and enhanced liver
metastasis. Cancer Res 62:1567-72.

32
19. Kardosh, A., M. Blumenthal, W. J. Wang, T. C. Chen, and A. H.
Schonthal. 2004. Differential effects of selective COX-2 inhibitors on cell
cycle regulation and proliferation of glioblastoma cell lines. Cancer Biol Ther
3:55-62.
20. Kardosh, A., N. Soriano, Y. T. Liu, J. Uddin, N. A. Petasis, F. M. Hofman,
T. C. Chen, and A. H. Schonthal. 2005. Multitarget inhibition of
drug-resistant multiple myeloma cell lines by dimethyl-celecoxib (DMC), a
non-COX-2 inhibitory analog of celecoxib. Blood 106:4330-8.
21. Kardosh, A., N. Soriano, P. Pyrko, Y. T. Liu, M. Jabbour, F. M. Hofman,
and A. H. Schonthal. 2007. Reduced survivin expression and tumor cell
survival during chronic hypoxia and further cytotoxic enhancement by the
cyclooxygenase-2 inhibitor celecoxib. J Biomed Sci 14:647-62.
22. Kardosh, A., W. Wang, J. Uddin, N. A. Petasis, F. M. Hofman, T. C. Chen,
and A. H. Schonthal. 2005. Dimethyl-celecoxib (DMC), a derivative of
celecoxib that lacks cyclooxygenase-2-inhibitory function, potently mimics
the anti-tumor effects of celecoxib on Burkitt's lymphoma in vitro and in vivo.
Cancer Biol Ther 4:571-82.
23. Kashfi, K., and B. Rigas. 2005. Non-COX-2 targets and cancer: expanding
the molecular target repertoire of chemoprevention. Biochem Pharmacol
70:969-86.
24. Keller, J. J., and F. M. Giardiello. 2003. Chemoprevention strategies using
NSAIDs and COX-2 inhibitors. Cancer Biol Ther 2:S140-9.
25. Kim, S. J., Z. Zhang, E. Hitomi, Y. C. Lee, and A. B. Mukherjee. 2006.
Endoplasmic reticulum stress-induced caspase-4 activation mediates apoptosis
and neurodegeneration in INCL. Hum Mol Genet 15:1826-34.
26. Kismet, K., M. T. Akay, O. Abbasoglu, and A. Ercan. 2004. Celecoxib: a
potent cyclooxygenase-2 inhibitor in cancer prevention. Cancer Detect Prev
28:127-42.
27. Ko, S. H., G . J. Choi, J. H. Lee, Y. A. Han, S. J. Lim, and S. H. Kim. 2008.
Differential effects of selective cyclooxygenase-2 inhibitors in inhibiting
proliferation and induction of apoptosis in oral squamous cell carcinoma.
Oncol Rep 19:425-33.
28. Lee, A. S. 2005. The ER chaperone and signaling regulator GRP78/BiP as a
monitor of endoplasmic reticulum stress. Methods 35:373-81.

33
29. Lipsky, P. E. 1999. Specific COX-2 inhibitors in arthritis, oncology, and
beyond: where is the science headed? J Rheumatol Suppl 56:25-30.
30. Masferrer, J. L., K. M. Leahy, A. T. Koki, B. S. Zweifel, S. L. Settle, B. M.
Woerner, D. A. Edwards, A. G. Flickinger, R. J. Moore, and K. Seibert.
2000. Antiangiogenic and antitumor activities of cyclooxygenase-2 inhibitors.
Cancer Res 60:1306-11.
31. Needleman, P., and P. C. Isakson. 1997. The discovery and function of
COX-2. J Rheumatol Suppl 49:6-8.
32. Nussmeier, N. A., A. A. Whelton, M. T. Brown, R. M. Langford, A. Hoeft,
J. L. Parlow, S. W. Boyce, and K. M. Verburg. 2005. Complications of the
COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engl J
Med 352:1081-91.
33. Oyadomari, S., and M. Mori. 2004. Roles of CHOP/GADD153 in
endoplasmic reticulum stress. Cell Death Differ 11:381-9.
34. Parente, L., and M. Perretti. 2003. Advances in the pathophysiology of
constitutive and inducible cyclooxygenases: two enzymes in the spotlight.
Biochem Pharmacol 65:153-9.
35. Penning, T. D., J. J. Talley, S. R. Bertenshaw, J. S. Carter, P. W. Collins, S.
Docter, M. J. Graneto, L. F. Lee, J. W. Malecha, J. M. Miyashiro, R. S.
Rogers, D. J. Rogier, S. S. Yu, AndersonGd, E. G. Burton, J. N. Cogburn,
S. A. Gregory, C. M. Koboldt, W. E. Perkins, K. Seibert, A. W. Veenhuizen,
Y. Y. Zhang, and P. C. Isakson. 1997. Synthesis and biological evaluation of
the 1,5-diarylpyrazole class of cyclooxygenase-2 inhibitors: identification of
4-[5-(4-methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benze
nesulfonamide (SC-58635, celecoxib). J Med Chem 40:1347-65.
36. Pyrko, P., A. Kardosh, Y. T. Liu, N. Soriano, W. Xiong, R. H. Chow, J.
Uddin, N. A. Petasis, A. K. Mircheff, R. A. Farley, S. G . Louie, T. C. Chen,
and A. H. Schonthal. 2007. Calcium-activated endoplasmic reticulum stress
as a major component of tumor cell death induced by 2,5-dimethyl-celecoxib,
a non-coxib analogue of celecoxib. Mol Cancer Ther 6:1262-75.
37. Rao, C. V ., and B. S. Reddy. 2004. NSAIDs and chemoprevention. Curr
Cancer Drug Targets 4:29-42.

34
38. Reddy, R. K., C. Mao, P. Baumeister, R. C. Austin, R. J. Kaufman, and A.
S. Lee. 2003. Endoplasmic reticulum chaperone protein GRP78 protects cells
from apoptosis induced by topoisomerase inhibitors: role of A TP binding site
in suppression of caspase-7 activation. J Biol Chem 278:20915-24.
39. Rheaume, E., L. Y. Cohen, F. Uhlmann, C. Lazure, A. Alam, J. Hurwitz, R.
P. Sekaly, and F. Denis. 1997. The large subunit of replication factor C is a
substrate for caspase-3 in vitro and is cleaved by a caspase-3-like protease
during Fas-mediated apoptosis. EMBO J 16:6346-54.
40. Schonthal, A. H. 2006. Antitumor properties of dimethyl-celecoxib, a
derivative of celecoxib that does not inhibit cyclooxygenase-2: implications
for glioma therapy. Neurosurg Focus 20:E21.
41. Schonthal, A. H. 2007. Direct non-cyclooxygenase-2 targets of celecoxib and
their potential relevance for cancer therapy. Br J Cancer 97:1465-8.
42. Schonthal, A. H. 2007. Induction of apoptosis by celecoxib in cell culture: an
uncertain role for cyclooxygenase-2. Cancer Res 67:5575-6; author reply
5576.
43. Schonthal, A. H., T. C. Chen, F. M. Hofman, S. G. Louie, and N. A. Petasis.
2008. Celecoxib analogs that lack COX-2 inhibitory function: preclinical
development of novel anticancer drugs. Expert Opin Investig Drugs
17:197-208.
44. Shiff, S. J., and B. Rigas. 1999. The role of cyclooxygenase inhibition in the
antineoplastic effects of nonsteroidal antiinflammatory drugs (NSAIDs). J Exp
Med 190:445-50.
45. Smith, W. L., D. L. DeWitt, and R. M. Garavito. 2000. Cyclooxygenases:
structural, cellular, and molecular biology. Annu Rev Biochem 69:145-82.
46. Song, X., H. P. Lin, A. J. Johnson, P. H. Tseng, Y. T. Yang, S. K. Kulp, and
C. S. Chen. 2002. Cyclooxygenase-2, player or spectator in cyclooxygenase-2
inhibitor-induced apoptosis in prostate cancer cells. J Natl Cancer Inst
94:585-91.
47. Steinbach, G ., P. M. Lynch, R. K. Phillips, M. H. Wallace, E. Hawk, G . B.
Gordon, N. Wakabayashi, B. Saunders, Y. Shen, T. Fujimura, L. K. Su,
and B. Levin. 2000. The effect of celecoxib, a cyclooxygenase-2 inhibitor, in
familial adenomatous polyposis. N Engl J Med 342:1946-52.

35
48. Tanaka, K., W. Tomisato, T. Hoshino, T. Ishihara, T. Namba, M. Aburaya,
T. Katsu, K. Suzuki, S. Tsutsumi, and T. Mizushima. 2005. Involvement of
intracellular Ca2+ levels in nonsteroidal anti-inflammatory drug-induced
apoptosis. J Biol Chem 280:31059-67.
49. Tang, X., Y. J. Sun, E. Half, M. T. Kuo, and F. Sinicrope. 2002.
Cyclooxygenase-2 overexpression inhibits death receptor 5 expression and
confers resistance to tumor necrosis factor-related apoptosis-inducing
ligand-induced apoptosis in human colon cancer cells. Cancer Res 62:4903-8.
50. Zhu, J., X. Song, H. P. Lin, D. C. Young, S. Yan, V. E. Marquez, and C. S.
Chen. 2002. Using cyclooxygenase-2 inhibitors as molecular platforms to
develop a new class of apoptosis-inducing agents. J Natl Cancer Inst
94:1745-57.

Abstract (if available)

Abstract Because many reports now challenge the notion that celecoxib (Celebrex®) execute its antitumor effect in the absence of COX-2 inhibitory activity, the aim of the project is to investigate COX-2 involvement with respect to its antitumor potential. We have studied two close structural analogs of celecoxib, DMC and UMC, which differentially display COX-2 inhibitory activity. UMC has COX-2 inhibitory function, while DMC has antitumor potential. In our results, DMC exhibited the most potent antitumor activity, while UMC was the least effective. During acute treatment, the three drugs and all tested cell lines were able to trigger endoplasmic reticulum stress, as indicated by the induction of CHOP, GRP78, and caspase-7 activation. Thus, we concluded that the antitumor activity of celecoxib is independent of COX-2 inhibitory activity and relies on the ability to trigger ER stress. This finding confirms previous studies and can be beneficial to anti-cancer therapy.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Enhanced Burkitt 's lymphoma cell killing by the combination treatments of bortezomib with celecoxib and 2,5-dimethyl-celecoxib (DMC)

PDF

Examining perillyl alcochol in glioblastoma treatment and relating its effect to endoplasmic reticulum stress response to further advance its usage in combination treatment with dimethyl-celecoxib

PDF

Identification of the targets required for the anti-tumor effect of celecoxib and its non-COX-2 inhibitory derivative

PDF

Anti-cancer effects of novel glidobactin type proteasome inhibitors

PDF

Glioblastoma treatment with 2,5-dimethyl-celecoxib (DMC) in vitro: - effects of additional chemotherapeutic drugs and tumor microenvironment, - inconsistencies among commonly used in vitro cell g...

PDF

Construction of plasmids for constitutive and induced expression of secreted alkaline phosphatase in mammalian ovarian carcinoma cells

PDF

The effects of methionine restriction on hepatitis B virus replication

PDF

The effect of estradiol on TAZ in stromal cell line ST2 in osteoblast differentiation and ER+breast cancer cell

Asset Metadata

Creator Chuang, Huan-Ching (author)

Core Title The differential effects of selective COX-2 inhibitors on cell proliferation and induced ER stress in glioblastoma and pancreatic carcinoma cell lines

School Keck School of Medicine

Degree Master of Science

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2008-05

Publication Date 04/17/2008

Defense Date 03/26/2008

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag caspase 7,CHO,cyclooxygenase-2,endoplasmic,GPR 78,OAI-PMH Harvest,reticulum stress,survivin

Language English

Advisor Schönthal, Axel H. (committee chair), Hong, Young Kwon (committee member), Tokes, Zoltan A. (committee member)

Creator Email florachuang@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1144

Unique identifier UC175522

Identifier etd-Chuang-20080417 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-58020 (legacy record id),usctheses-m1144 (legacy record id)

Legacy Identifier etd-Chuang-20080417.pdf

Dmrecord 58020

Document Type Thesis

Rights Chuang, Huan-Ching

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu