Professional Documents
Culture Documents
CCN1 Sensitizes Esophageal Cancer Cells To TRAIL Me - 2017 - Experimental Cell R
CCN1 Sensitizes Esophageal Cancer Cells To TRAIL Me - 2017 - Experimental Cell R
A R T I C L E I N F O A B S T R A C T
Keywords: TRAIL is one of the best anti-cancer molecules in our body. It kills a variety of cancer cells that are resistant to
TRAIL conventional chemotherapy, without causing much negative impact on normal cells, because its death receptors
CCN1 are almost exclusively found on cancer cells. However, some cancer cells are not sensitive to TRAIL treatment,
Apoptosis even though they express its death receptors. A second molecule is needed to help TRAIL to complete its mission.
Esophageal cancer
Finding such molecules now becomes a top priority in cancer research. Our study shows that CCN1 is such a
Receptor
molecule. CCN1 was highly expressed in the esophageal epithelium of the patients suffering from gastro-
esophageal reflux disease, but faded away as the situation worsened towards adenocarcinoma. Treating the
tumor cells with CCN1 resulted in apoptosis, while the same treatment to the normal cells only nourished cell
growth. It was TRAIL that mediated this process. Apparently, CCN1 altered the expression profile of TRAIL and
its receptors in tumor cells, namely, activating TRAIL and its death receptors and shutting down its decoy
receptors. CCN1 and TRAIL worked as a team to put the cancer cells to death, as elimination of either one failed
apoptosis.
Abbreviations: TRAIL, Tumor necrosis factor-Related Apoptosis-Inducing Ligand; EAC, esophageal adenocarcinoma; BE, Barrett's Esophagus; GERD, gastroesophageal reflux disease;
TNF, tumor necrosis factor; FADD, Fas-Associated protein with Death Domain
⁎
Corresponding author at: College of Medicine, University of California Irvine, CA 92697, USA.
E-mail address: Jianyuan.chai@gmail.com (J. Chai).
http://dx.doi.org/10.1016/j.yexcr.2017.10.015
Received 7 September 2017; Received in revised form 11 October 2017; Accepted 17 October 2017
Available online 18 October 2017
0014-4827/ © 2017 Elsevier Inc. All rights reserved.
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
surgery, but neither is a clear winner thus far. In fact, increasing evi- done in Lipofectamine-2000 (Invitrogen, Carlsbad) and selected using
dence shows multiple side-effects associated with this line of drugs, Neomycin or Puromycin.
such as decreased absorption of vitamins/minerals [9], susceptibility to For cell treatment, cells were plated either in Petri-dishes (for RNA
infections [10], bone fracture [11], and even elevated risk of devel- or protein isolation), or on collagen-coated coverslips (for staining).
oping cancer [12]. For these reasons, the Food and Drug Administration Cells were synchronized and treated accordingly with recombinant
of the United States has repeatedly issued warnings on the use of acid human CCN1, or TRAIL (Peprotech, Rocky Hill) or combinations as
suppressive drugs. For people who have responded to medication but indicated.
continue to experience GERD symptoms, surgery to reconstruct the
esophageal sphincter is usually an option. However, only 5% of GERD 2.2. Apoptosis assays
patients undergo surgery and follow-up study found that almost two-
thirds of the surgery patients were back on medication [13]. Cells were cultured on coverslips and treated as indicated.
A good anti-cancer drug should not only be able to kill cancer cells, Following the manufacturer's protocol, cells were then incubated in a
but also be harmless to the normal tissue. TRAIL (Tumor necrosis binding buffer containing FITC-conjugated Annexin V and propidium
factor-Related Apoptosis-Inducing Ligand) is such an example. As a iodide (Abcam, Cambridge) for 5 min. The number of apoptotic cells in
member of TNF family, TRAIL can selectively kill a variety of cancer 5 microscopic fields from each of 3 coverslips per treatment was re-
cells that are resistant to conventional chemotherapy while leaving corded under a Nikon fluorescence microscope (n=15). The apoptotic
normal cells unharmed, because its death receptors are almost ex- index (%) was calculated by dividing the number of apoptotic cells by
clusively found on cancer cells [14]. When an adequate amount of the total number of cells in control (verified by DAPI staining) under
TRAIL is in present, cancer cells will usually go into apoptosis. How- the same magnification.
ever, more and more studies found that some cancer cells are not sen-
sitive to TRAIL treatment, despite expression of the death receptors, and
2.3. Real-time RT-PCR and array analysis
require a second agent to precondition them in order for TRAIL to do its
job [15,16]. Therefore, finding the right sensitizers for TRAIL now be-
Total RNA was extracted using RNeasy kit (Qiagen, Valencia) fol-
comes a top priority in cancer research. Based on our study, CCN1
lowing the manufacturer's protocol. Reverse transcription was done
appears to be such a molecule, at least for EAC.
following the procedure: 25 °C/10 min – 55 °C/30 min – 85 °C/5 min –
CCN1 is a matricellular protein that essentially supports cell adhe-
4 °C/∞. 96-well plates of human gastroesophageal cancer tissue cDNA
sion, migration and survival. It is one of the targets of serum response
array, apoptosis array and TNF array were purchased from Origene.
factor (SRF), a transcription factor that we have been studying for
Real-time PCR was performed following the two-step program using
nearly two decades [17,18]. Cells attaching to CCN1 usually activate
SYBR Green master mix (SABiosciences, Frederick). Data were gener-
cyto-protective pathways against apoptosis. However, a number of
ated from at least 5 independent experiments and analyzed according to
studies have revealed the other edge of this molecule [19–21], i.e.,
the ΔΔCt method. Briefly, ΔCt was calculated by subtracting the Ct value
triggering apoptosis sometimes. The mechanisms are still not fully un-
of GAPDH from the Ct value for each gene; and then ΔΔCt was calcu-
derstood, but recent data have drawn a line to TNF cytokines. It was
lated by subtracting the ΔCt of the control from the ΔCt of the treat-
shown that incubation of fibroblasts with either CCN1 or TNFα alone
ment; and finally the fold change was calculated using the formula:
promoted cell proliferation, but in the presence of both, cells underwent
Fold Change = 2(-ΔΔCt).
apoptosis [22]. Using CCN1 with other TNF molecules like FasL or
TRAIL produced a similar effect [19,23]. In prostate cancer cells, CCN1
2.4. Western blot analysis
was found to be able to do both protective and destructive work, all
depending on the availability of TRAIL [23]. Yet, these cells express
Western blot analyses were done as described in our earlier studies
CCN1 naturally and therefore, do not need a sensitizer to engage in
[24,25]. Antibodies included CCN1 and GAPDH (Santa Cruz Bio-
TRAIL-mediated apoptosis.
technology, Santa Cruz). Signals were quantified with the TotalLab
The TNF superfamily now includes 19 identified members which
TL100 software based on at least five replicates.
function through one or more of the 29 receptors. Some of the receptors
(e.g. Fas, TRAILR1) contain death domains in their cytoplasmic tails,
which can trigger apoptosis upon ligand binding, while others (e.g. 2.5. Immunofluorescence and immunohistochemistry
TRAILR3, TRAILR4) do not have such a structure and therefore cannot
provide intracellular signals for apoptosis. However, this second group Cells were cultured till desired confluence on coverslips that have
of receptors can effectively compete with the first group for their cor- been pre-coated with type I collagen. After serum starvation and in-
responding ligands and abort the apoptotic effort of the first group, and dicated treatment, cells were fixed for 10 min in 4% paraformaldehyde.
for this reason, they are often referred to as the decoy receptors in After incubation with a primary antibody for 2 h, cells were washed in
contrast to the first group, the death receptors. PBS and incubated with a FITC-conjugated secondary antibody
(Abcam) for an hour. Nuclei were counter-stained with Propidium
2. Materials and methods Iodide (Invitrogen).
For immunohistochemistry, tissue specimens were fixed in 10%
2.1. Cell culture, transfection and treatment Formalin and embedded in paraffin. Tissue sections were de-paraffi-
nized, re-hydrated, and retrieved in a microwave oven by pause-
Human EAC cells OE33, OE19 and FLO-1 (Sigma-Aldrich, St Louis) heating. Slides were incubated with the primary antibody of interest for
were cultured in RPMI medium plus 10% fetal bovine serum, while 2 h in a humidified chamber. After 3 × 10 min wash in PBS, signals
normal human esophageal epithelial cells Het1A (American Type were detected using a LSAB+ kit (Agilent) containing a universal sec-
Culture Collection, Manassas) were maintained in culture in KBM-2 ondary linker and HRP-conjugated streptavidin with AEC chromogen as
medium with supplements (Lonza, Walkersville). substrate. Nuclei were counter-stained with hematoxylin.
The following plasmids were used for transfection: pCMV6 carrying
one of the following open reading frames: CCN1, TRAIL, TRAILR2, 2.6. Statistical analysis
TRAILR3, TRAILR4, and OPG; and pRS with shRNA against each one of
them. pRS with negative control shRNA or pCMV6 without an insert Numerical data were analyzed by single classification one-way
were used as controls (Origene, Rockville). All cell transfections were ANOVA and P < 0.05 was considered as significant.
164
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
165
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
166
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
Table 2 induced TRAIL expression, but also activated TRAILR2, the second
The effects of CCN1 transfections on the expression of BIRC5 and TP53 in EAC cells death receptor, pushing the cancer cells more towards death. In addi-
(OE33) versus normal esophageal epithelial cells (Het1A), based on apoptosis array
tion, all the three decoy receptors were shut down in the presence of
analysis. The numbers represent relative levels of the gene after CCN1 manipulation in
comparison to its normal expression. CCN1, meaning that the EAC cells doomed to death. In Het1A cells, on
the other hand, CCN1 treatment stopped the expression of TRAIL li-
Cell lines CCN1 BIRC5 TP53 gand, but activated two of the three decoy receptors, TRAILR4 and
OPG, while the two death receptors remained in silence, that is to say,
OE33 1.00 1.00 1.00
OE33CCN1- 0.04 3.13 × 107 0.18 the normal cells were completely immune to TRAIL apoptotic attacks as
OE33CCN1+ 29.86 0.15 3.06 × 104 long as CCN1 is around. To verify this result, we stained all three types
Het1A 1.00 1.00 1.00 of AEC cells and Het1A cells before and after CCN1 treatment for all the
Het1ACCN1- 0.09 0.26 2.10 × 103 TRAIL receptors. In agreement with the RT_PCR outcomes, the decoy
Het1ACCN1+ 64.00 6.81 × 106 0.93
receptors were found activated in Het1A cells but shut down in EAC
cells in the presence of CCN1. Fig. 2E-H show the immunofluorescence
To verify the results obtained from using recombinant CCN1, we staining for TRAILR4.
transfected the cells with plasmids carrying either CCN1 open reading To follow this trend of thought, next, we wanted to know how im-
frame (CCN1+) or shRNA against CCN1 (CCN1-), and then assessed portant TRAIL was to CCN1-induced EAC cell apoptosis. We incubated
apoptosis using the same array analysis. Overexpression of CCN1 both OE33 and Het1A cells on coverslips for 6 h with one of the fol-
caused a similar effect in gene expression as using recombinant CCN1 lowing combinations: (1) recombinant TRAIL at 30 ng/ml, (2) re-
protein, while knockdown of CCN1 promoted survival in OE33 cells but combinant TRAIL at 30 ng/ml plus recombinant CCN1 at 1 μg/ml, (3)
suppressed it in Het1A cells. Table 2 shows the effects on two re- recombinant CCN1 alone at 1 μg/ml, (4) recombinant CCN1 at 1 μg/ml
presentative genes, BIRC5 and TP53. plus TRAIL antibody at 50 μg/ml, or (5) BSA. Apoptosis was then as-
sayed using Annexin V. As shown in Fig. 3A, using recombinant TRAIL
alone did not cause a significant change in terms of apoptosis compared
3.3. CCN1 requires TRAIL to induce apoptosis in EAC cells to control (BSA), but addition of CCN1 to this experiment (TRAIL+C-
CN1) raised the number of apoptotic OE33 cells by 7 fold (P < 0.01),
As mentioned above, CCN1 is commonly known for supporting cell indicating that, TRAIL cannot kill EAC cells without the presence of
survival, therefore, its impact on Het1A cells was expected. However, CCN1. Using recombinant CCN1 alone, on the other hand, induced
its apoptotic effect on OE33 cells made us think that EAC cells might apoptosis at almost the same magnitude as TRAIL and CCN1 combi-
express something that could convert CCN1 from pro-survival to pro- nation, suggesting that extrinsic TRAIL was unnecessary for CCN1 to
death. According to Lau's experiments [19–23], who discovered CCN1, induce EAC cell death. However, when we added TRAIL antibody to
CCN1 cannot cause apoptosis in its own, and it requires the presence of keep any available TRAIL neutralized (CCN1+TRAIL.Ab), CCN1 lost all
TNF cytokines to become a cell killer. In his experiment, combining its apoptotic power, indicating that CCN1-induced TRAIL expression is
500 ng/ml CCN1 with 100 ng/ml FASLG doubled the number of an essential requirement for EAC cell apoptosis.
apoptotic cells caused by FASLG alone; while increasing CCN1 to 2 μg/
ml tripled the effect. Lau's work led us to speculate that EAC cells might 3.4. The shutdown of TRAIL decoy receptors is the key for CCN1-induced
have a different expression profile of TNF and/or TNFR family members EAC cell apoptosis
that makes them more sensitive to CCN1 apoptotic potential.
To test this hypothesis, at first, we screened these two families using Our previous experiments indicated that CCN1 and TRAIL need
a RT_PCR array containing gene sequences for 18 TNF ligands and 27 each other to kill EAC cells, as eliminating either one failed to cause
their receptors. EAC cells were found to express more of these genes apoptosis. Our next question was why TRAIL itself could not kill the
than the normal cells, namely, 7 ligands and 16 receptors in OE33 cells, cancer cells even when all its death receptors were available. We
8 ligands and 16 receptors in OE19 cells, 8 ligands and 15 receptors in thought that the expressions of TRAIL decoy receptors might be the
FLO-1 cells, versus 6 ligands and 8 receptors in Het1A cells, suggesting reason, because the decoy receptors have a dominant effect over the
that EAC cells were more vulnerable to TNF apoptotic attacks than the death receptors; they can form either homo-oligomers with themselves
normal cells. What even more interesting was the expressions of TRAIL or hetero-oligomers with the death receptors to disable the apoptotic
and its receptors: OE33 cells expressed four out of five TRAIL receptors, power of TRAIL.
including the death receptor TRAILR1 and three decoy receptors To test this hypothesis, we evaluated the contribution of each in-
TRAILR3, TRAILR4 and OPG, but not TRAIL ligand; while Het1A cells dividual TRAIL receptor to CCN1-induced apoptosis. At first, we
expressed TRAIL ligand but none of its receptors (Table 3). These results transfected OE33 cells with various plasmid combinations to establish
prompted us to shift our attention immediately to TRAIL and its re- cell lines with the following expression profiles: (1) OE33TRAILR2+, (2)
ceptors. OE33TRAILR3-, (3) OE33TRAILR4-, (4) OE33OPG-, (5) OE33DCR-, (6)
Next, we probed the same type of array using RNA extracts after OE33TRAILR2+/DCR-, and (7) OE33. “DCR-” represents shutdown of all
CCN1 treatment. As shown in Table 3, in OE33 cells, CCN1 not only three decoy receptors. The knockdowns were created using shRNA
vectors and selected for at least 80% drop in the expression of the
Table 3 targeted gene. All the cells were then incubated with recombinant
The effect of CCN1 on the expression of TRAIL and its receptors in EAC cells (OE33)
versus normal esophageal epithelial cells (Het1A), based on TNF/TNFR array analysis. In
TRAIL at 30 ng/ml for 6 h. Apoptosis was examined using Annexin V.
controls, OE33 and Het1A, expression levels were all set to be 1 arbitrarily, the values in As shown in Fig. 3B, expression of TRAILR2 without changing anything
CCN1-treated groups are fold changes (P < 0.01) relative to their corresponding controls. else (OE33TRAILR2+) approximately doubled the rate of apoptosis
caused by TRAIL alone, but the total number of apoptotic cells was still
Gene Protein OE33 OE33+CCN1 Het1A Het1A+CCN1
below 25%. Knockdown of each individual decoy receptor increased
TNSF10 TRAIL 1 6.21 × 10 3
1 0 the rate to 25–45%. Among them, knockdown of TRAILR4 had the best
TNSRSF10A TRAILR1 1 1.25 1 0 effect. Simultaneous knockdown of all three decoy receptors (OE33DCR-)
TNFRSF10B TRAILR2 1 14.93 1 0 raised the apoptosis rate to 65%, while expressing TRAILR2 at the same
TNFRSF10C TRAILR3 1 0 1 0
time caused a massive cell death, almost equivalent to the effect of
TNFRSF10D TRAILR4 1 0 1 3.10 × 103
TNFRSF11B OPG 1 0 1 1.46 × 107 TRAIL and CCN1 combination as in the previous experiment. Taken
together, this experiment demonstrated that each receptor contributed
167
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
4. Discussion
168
T. Dang et al. Experimental Cell Research 361 (2017) 163–169
and some of them are already under development for chemotherapeutic 1257–1267.
[23] C.A. Franzen, C.C. Chen, V. Todorović, V. Juric, R.I. Monzon, L.F. Lau, Matrix
drugs [46]. protein CCN1 is critical for prostate carcinoma cell proliferation and TRAIL-induced
apoptosis, Mol. Cancer Res. 7 (2009) 1045–1055.
Acknowledgements [24] J. Chai, M. Norng, C. Modak, K. Reavis, W. Mouazzen, J. Pham, CCN1 Induces a
reversible epithelial-mesenchymal transition in gastric epithelial cells, Lab.
Investig. 90 (8) (2010) 1140–1151.
This study was supported by the VA Research Foundation in the [25] J. Chai, M. Norng, A.S. Tarnawski, J. Chow, A critical role of serum response factor
United States and Baotou Medical College in China, and did not receive in myofibroblast differentiation during experimental oesophageal ulcer healing in
rat, Gut 56 (5) (2007) 621–630.
any specific grant from funding agencies in the public, commercial, or [26] M.T. Lin, C.C. Chang, S.T. Chen, H.L. Chang, J.L. Su, Y.P. Chau, et al., Cyr61 ex-
not-for-profit sectors. pression confers resistance to apoptosis in breast cancer MCF-7 cells by a me-
chanism of NF-kappaB-dependent XIAP up-regulation, J. Biol. Chem. 279 (2004)
24015–24023.
Conflict of interest
[27] S. Gery, D. Xie, H. Gabra, C. Miller, H. Wang, D. Scott, W.S. Yi, M.L. Popoviciu,
J.W. Said, H.P. Koeffer, Ovarian carcinomas: CCN genes are aberrantly expressed
None. and CCN1 promotes proliferation of these cells, Clin. Cancer Res. 11 (2005)
7243–7254.
[28] S.E. Holloway, A.W. Beck, L. Girard, M.R. Jaber, C.C. Jr Barnett, R.A. Brekken,
References J.B. Fleming, Increased expression of Cyr61 (CCN1) identified in peritoneal me-
tastases from human pancreatic cancer, J. Am. Coll. Surg. 200 (2005) 371–377.
[1] Global Burden of Disease Cancer Collaboration, C. Fitzmaurice, D. Dicker, A. Pain, [29] C.C. Chen, K.H. Kim, L.F. Lau, The matricellular protein CCN1 suppresses hepato-
H. Hamavid, M. Moradi-Lakeh, M.F. MacIntyre, C. Allen, G. Hansen, R. Woodbrook, carcinogenesis by inhibiting compensatory proliferation, Oncogene 35 (2016)
et al., The Global burden of cancer 2013, JAMA Oncology 1 (2015) 505–527. 1314–1323.
[2] J. Chai, M.M. Jamal, Esophageal malignancy – a growing concern, World J. [30] W. Chien, T. Kumagai, C.W. Miller, J.C. Desmond, J.M. Frank, J.W. Said, et al.,
Gastroenterol. 18 (2012) 6521–6526. Cyr61 suppresses growth of human endometrial cancer cells, J. Biol. Chem. 279
[3] N. Pandeya, P.M. Webb, S. Sadeghi, A.C. Green, D.C. Whiteman, Australian cancer (2004) 53087–53096.
study. gastro-oesophageal reflux symptoms and the risks of oesophageal cancer: are [31] X. Tong, D. Xie, J. O'Kelly, C.W. Miller, C. Muller-Tidow, H.P. Koeffler, Cyr61, a
the effects modified by smoking, NSAIDs or acid suppressants? Gut 59 (2010) member of CCN family, is a tumor suppressor in non-small cell lung cancer, J. Biol.
31–38. Chem. 276 (2001) 47709–47714.
[4] J. Lagergren, Adenocarcinoma of oesophagus: what exactly is the size of the pro- [32] J. Chai, C. Modak, Y. Ouyang, S.Y. Wu, M.M. Jamal, CCN1 induces β-catenin
blem and who is at risk? Gut 54 (2005) (Suppl 1: i1-5). translocation in esophageal squamous cell carcinoma through integrin α11, ISRN
[5] D.A. Corley, Obesity and the rising incidence of oesophageal and gastric adeno- Gastroenterol. (2012) 207235.
carcinoma: what is the link? Gut 56 (11) (2007) 1493–1494. [33] H.N. LeBlanc, A. Ashkenazi, Apo2L/TRAIL and its death and decoy receptors, Cell
[6] M.M. Finucane, G.A. Stevens, M.J. Cowan, et al., National, regional, and global Death Differ. 10 (2003) 66–75.
trends in body-mass index since 1980: systematic analysis of health examination [34] F.C. Kischkel, D.A. Lawrence, A. Chuntharapai, P. Schow, K.J. Kim, A. Ashkenazi,
surveys and epidemiological studies with 960 country-years and 9.1 million parti- Apo2L/TRAIL-dependent recruitment of endogenous FADD and caspase-8 to death
cipants, Lancet 377 (2011) 557–567. receptors 4 and 5, Immunity 12 (2000) 611–620.
[7] C.J. Ruhm, Current and future prevalence of obesity and severe obesity in the [35] M.J. Dyer, M. MacFarlane, G.M. Cohen, Barriers to effective TRAIL-targeted therapy
United States. Forum for Health Economics & Policy; 10 (2), article 6, 2007. of malignancy, J. Clin. Oncol. 25 (2007) 4505–4506.
[8] M. Ng, T. Fleming, M. Robinson, et al., Global, regional, and national prevalence of [36] J.P. Sheridan, S.A. Marsters, R.M. Pitti, et al., Control of TRAIL-induced apoptosis
overweight and obesity in children and adults during 1980–2013: a systematic by a family of signaling and decoy receptors, Science 277 (1997) 818–821.
analysis for the Global burden of disease study 2013, Lancet 384 (2014) 766–781. [37] S. Neumann, J. Hasenauer, N. Pollak, P. Scheurich, Dominant negative effects of
[9] T. Ito, R.T. Jensen, Association of long-term proton pump inhibitor therapy with tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) receptor 4
bone fractures and effects on absorption of calcium, vitamin B12, iron, and mag- on TRAIL receptor 1 signaling by formation of heteromeric complexes, J. Biol.
nesium, Curr. Gastroenterol. Rep. 12 (2010) 448–457. Chem. 289 (2014) 16576–16587.
[10] J.S. Bajaj, S.M. Ratliff, D.M. Heuman, K.L. Lapane, Proton pump inhibitors are as- [38] A.D. Sanlioglu, E. Dirice, C. Aydin, N. Erin, S. Koksoy, S. Sanlioglu, Surface TRAIL
sociated with a high rate of serious infections in veterans with decompensated decoy receptor-4 expression is correlated with TRAIL resistance in MCF7 breast
cirrhosis, Aliment. Pharmacol. Ther. 36 (2012) 866–874. cancer cells, BMC Cancer 5 (2005) 54.
[11] L.A. Fraser, W.D. Leslie, L.E. Targownik, A. Papaioannou, J.D. Adachi, CaMos re- [39] C. Aydin, A.D. Sanlioglu, B. Karacay, G. Ozbilim, L. Dertsiz, O. Ozbudak, C.A. Akdis,
search Group. The effect of proton pump inhibitors on fracture risk: report from the S. Sanlioglu, Decoy receptor-2 small interfering RNA (siRNA) strategy employing
Canadian multicenter osteoporosis study, Osteoporos. Int. 24 (2013) 1161–1168. three different siRNA constructs in combination defeats adenovirus-transferred
[12] D.A. Corley, A. Kubo, Body mass index and gastroesophageal reflux disease: a tumor necrosis_factor-related apoptosis-inducingligand resistance in lung cancer
systematic review and meta-analysis, Am. J. Gastroenterol. 101 (2006) 2619–2628. cells, Hum. Gene Ther. 18 (2007) 39–50.
[13] A.H. Merry, L.J. Schouten, R.A. Goldbohm, P.A. van den Brandt, Body mass index, [40] R. Riccioni, L. Pasquini, G. Mariani, E. Saulle, A. Rossini, D. Diverio, E. Pelosi,
height and risk of adenocarcinoma of the oesophagus and gastric cardia: a pro- A. Vitale, A. Chierichini, M. Cedrone, R. Foà, F. Lo Coco, C. Peschle, U. Testa, TRAIL
spective cohort study, Gut 56 (2007) 1503–1511. decoy receptors mediate resistance of acute myeloid leukemia cells to TRAIL,
[14] F.C. Kimberley, G.R. Screaton, Following a TRAIL: update on a ligand and its five Haematologica 90 (2005) 612–624.
receptors, Cell Res. 14 (2004) 359–372. [41] D. Lane, I. Matte, C. Laplante, P. Garde-Granger, C. Rancourt, A. Piché,
[15] M.A. Hall, J.L. Cleveland, Clearing the TRAIL for cancer therapy, Cancer Cell 12 Osteoprotegerin (OPG) activates integrin, focal adhesion kinase (FAK), and Akt
(2007) 4–6. signaling in ovarian cancer cells to attenuate TRAIL-induced apoptosis, J. Ovarian
[16] F.A. Kruyt, TRAIL and cancer therapy, Cancer Lett. 263 (2008) 14–25. Res. 6 (2013) 82.
[17] C. Modak, J. Chai, Serum response factor: look into the gut, World J. Gastroenterol. [42] M. Irmler, M. Thome, M. Hahne, et al., Inhibition of death receptor signals by
16 (2010) 2195–2201. cellular FLIP, Nature 388 (1997) 190–195.
[18] J. Chai, Roles of SRF in endothelial cells during hypoxia, in: J. Chai (Ed.), Research [43] A.D. Schimmer, K. Welsh, C. Pinilla, et al., Small-molecule antagonists of apoptosis
Directions in Tumor Angiogenesis, 2013 Intech Publisher, Rijeka, 2013, pp. 29–45. suppressor XIAP exhibit broad antitumor activity, Cancer Cell 5 (2004) 25–35.
[19] V. Juric, C.C. Chen, F. Lau, Fas-mediated apoptosis is regulated by the extracellular [44] S. Hinz, A. Trauzold, L. Boenicke, et al., Bcl-XL protects pancreatic adenocarcinoma
matrix protein CCN1 in vitro and in vivo, Mol. Cell Biol. 29 (2009) 3266–3279. cells against CD95- and TRAIL-receptor-mediated apoptosis, Oncogene 19 (2000)
[20] C.C. Chen, L.F. Lau, Deadly liaisons: fatal attraction between CCN matricellular 5477–5486.
proteins and the tumor necrosis factor family of cytokines, J. Cell Commun. Signal. [45] T.M. Grzeszkiewicz, D.J. Kirschling, N. Chen, L.F. Lau, CYR61 stimulates human
4 (2010) 63–69. skin fibroblast migration through integrin αVβ5 and enhances mitogenesis through
[21] V. Todorovic, C.C. Chen, N. Hay, L.F. Lau, The matrix protein CCN1 (CYR61) in- integrin αVβ3, independent of its carboxyl-terminal domain, J. Biol. Chem. 276
duces apoptosis in fibroblasts, J. Cell Biol. 171 (2005) 559–568. (2001) 21943–21950.
[22] C.C. Chen, J.L. Young, R.I. Monzon, N. Chen, V. Todorović, L.F. Lau, Cytotoxicity of [46] D. Cox, M. Brennan, N. Moran, Integrins as therapeutic targets: lessons and op-
TNFalpha is regulated by integrin-mediated matrix signaling, EMBO J. 26 (2007) portunities, Nat. Rev. Drug Discov. 9 (10) (2010) 804–820.
169