Thapsigargin sensitizes human melanoma cells to TRAIL-induced apoptosis by up-regulation of TRAIL-R2 through the unfolded protein response
Li Hua Cheny
Chen Chen Jiangy
Yu Fang Wang
Xu Dong Zhang
To whom correspondence should be addressed. Tel:
Email: Correspondence may also be addressed to Dr Xu Dong Zhang. Tel:
Immunology and Oncology Unit, David Maddison Clinical Sciences Building, Newcastle Misericordiae Hospital
Cnr. King & Watt Streets, Newcastle, New South Wales 2300
We have previously reported that sensitivity of melanoma cells to tumor necrosis factor-related apoptosis-inducing ligand (TRAIL)induced apoptosis is largely correlated with the levels of expression of TRAIL death receptors, in particular, TRAIL-R2 on the cell surface. However, fresh melanoma isolates and melanoma tissue sections express, in general, low levels of death receptors for TRAIL. We show in this study that the endoplasmic reticulum stress inducer, thapsigargin (TG), selectively up-regulated TRAIL-R2 and enhanced TRAIL-induced apoptosis in melanoma cells. However, the TRAIL-R2 pathway did not appear to be involved in induction of apoptosis by TG alone. Up-regulation of TRAIL-R2 appeared to be cooperatively mediated by the inositolrequiring transmembrane kinase and endonuclease 1a (IRE1a)and activation of transcription factor (ATF)-6-signaling pathways of the unfolded protein response (UPR) and the transcription factor CCAAT/enhancer-binding protein-homologous protein (CHOP). The latter played a critical role in the initial phase of the increase in TRAIL-R2 as small interfering RNA (siRNA) knockdown of CHOP blocked up-regulation of TRAIL-R2 only at a relatively early stage (16 h) after exposure to TG. In contrast, IRE1a and ATF6 appeared to be crucial in maintaining the increased levels of TRAIL-R2 in that siRNA knockdown of IRE1a or ATF6 had no effect on the increase in TRAIL-R2 at the initial phase, but blocked TRAIL-R2 up-regulation at a relatively late stage (36 h). Our results indicate that modulation of the UPR may be useful in sensitizing melanoma cells to TRAIL-induced apoptosis by up-regulation of TRAIL-R2.
Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is
a member of the tumor necrosis factor (TNF) family that appears to be
a promising candidate for cancer therapeutics because of its selective
cytotoxicity against malignancies (13). The potential significance of
TRAIL as an anticancer agent has been supported by studies in animal
models showing selective toxicity to human tumor xenografts but not
normal tissues (4,5). Induction of apoptosis by TRAIL is mediated by
its interaction with two death domain-containing receptors,
TRAILAbbreviations: ATF, activation of transcription factor; eIF2a, eukaryotic
initiation factor a; ER, endoplasmic reticulum; HUVEC, human umbilical vein
endothelial cell; IRE1a, inositol-requiring transmembrane kinase and
endonuclease 1a; mRNA, messenger RNA; PCR, polymerase chain reaction; PERK,
protein kinase-like endoplasmic reticulum kinase; siRNA, small interfering
RNA; TG, thapsigargin; TNF, tumor necrosis factor; TRAIL, tumor necrosis
factor-related apoptosis-inducing ligand; XBP1, X-box-binding protein 1;
CHOP, CCAAT/enhancer-binding protein-homologous protein; FADD,
Fasassociated death domain.
yThese authors contributed equally to this work.
R1 and -R2. This in turn orchestrates the assembly of the
deathinducing signaling complex that contains adapter components such
as Fas-associated death domain (FADD) that activates initiator
caspases, caspase-8 and -10, leading eventually to activation of effector
caspases such as caspase-3 and apoptosis (13).
We have shown previously that sensitivity of cultured melanoma
cells to TRAIL-induced apoptosis is correlated with the levels of
cellsurface expression of TRAIL death receptors, in particular,
TRAILR2 (6,7). Subsequent studies demonstrated that fresh melanoma
isolates are relatively resistant to TRAIL-induced apoptosis due to low
levels of TRAIL-death receptor expression (8). Moreover, melanoma
cells selected for TRAIL resistance by prolonged exposure to TRAIL
express substantially reduced levels of TRAIL-R2 on their surface
(9,10). Studies on melanoma tissue sections revealed that reduced
TRAIL-R2 expression is associated with disease progression and
a poor prognosis (11). Taken together, these studies indicate that
clinical potential of TRAIL in treatment of melanoma may be limited
unless given with agents that increase the cell-surface expression of
TRAIL death receptors, in particular, TRAIL-R2.
The cellular response to endoplasmic reticulum (ER) stress, the
unfolded protein response (UPR), is currently known to consist of
three distinct yet coordinated signaling pathways initiated,
respectively, by inositol-requiring transmembrane kinase and endonuclease
1a (IRE1a), activation of transcription factor (ATF)-6 and protein
kinase-like endoplasmic reticulum kinase (PERK) (1214). As an
adaptive response, the UPR is activated to alleviate the stress
condition imposed on the ER and is orchestrated by transcriptional
activation of multiple genes mediated by IRE1a and ATF6, a general
decrease in translation initiation and selective translation of specific
messenger RNAs (mRNAs) mediated by PERK (1214). However, if
the stress on ER remains unresolved, prolonged activation of the UPR
can lead to apoptosis (1214). Thapsigargin (TG), a sesquiterpene
lactone that induces ER stress by depletion of Ca2 within the ER
through inhibiting ER Ca2 ATPases (15), has been reported to induce
apoptosis via a TRAIL-R2-dependent apoptotic pathway (16). In
addition, TG was shown to enhance TRAIL-induced apoptosis in a
number of human cancer cells via up-regulation of TRAIL-R2 (1719).
Although the transcription factor CHOP is believed to be involved
(16), a potential role of the UPR in regulation of TRAIL-R2 by TG
has not been fully studied. Moreover, the effect of TG-induced ER
stress on the expression of TRAIL death receptors in melanoma cells
We show in this study that TG selectively up-regulated cell-surface
expression of TRAIL-R2 and enhanced TRAIL-induced apoptosis in
human melanoma cells. However, apoptosis of melanoma cells
induced by TG alone did not appear dependent on the
TRAIL-R2mediated apoptotic pathway. Up-regulation of TRAIL-R2 expression
on the cell surface was associated with enhanced TRAIL-R2 gene
transcription and elevated TRAIL-R2 total protein levels. The
IRE1aand ATF6-mediated signaling pathways of the UPR, along with the
transcription factor CHOP, appeared to play key roles in up-regulation
of TRAIL-R2 by TG in melanoma cells.
Materials and methods
Human melanoma cell lines Mel-RM, MM200, IgR3, Mel-CV, Me4405,
SkMel-28 and Mel-FH have been described previously and were cultured in
Dulbeccos modified Eagles medium containing 5% fetal calf serum
(Commonwealth Serum Laboratories, Melbourne, Australia) (6,20). Melanocytes
were kindly provided by Dr P.Parsons (Queensland Institute of Medical
Research, Brisbane, Australia) and cultured in medium supplied by Clonetics
(Edward Keller, Victoria, Australia). Human umbilical vein endothelial cells
(HUVECs) were kindly supplied by D.Clark (Transplantation Unit, John
Hunter Hospital, Australia) and were cultured as described elsewhere (21).
Human embryonic fibroblasts (FLOW 2000) were cultured in Dulbeccos
modified Eagles medium containing 5% fetal calf serum as described
Generation of TRAIL-selected cells
Generation of TRAIL-selected Mel-RM and MM200 cells was performed as
described previously (9,10). The resulting TRAIL-resistant cell lines were
designated as Mel-RM.S and MM200.S, respectively.
Antibodies, recombinant proteins and other reagents
TG was purchased from Sigma Chemical Co. (Castle Hill, Australia). It was
dissolved in dimethyl sulfoxide to make up a stock solution of 1 mM.
Recombinant human TRAIL and the TRAIL-R2/Fc chimera were supplied by
Immunex (Seattle, WA). The mouse mAbs against TRAIL-R1, -R2, -R3 and
-R4, Fas, TNF-R1 and -R2 were also supplied by Immunex. The cell-permeable
pan caspase inhibitor Z-Val-Ala-Asp(OMe)-CH2F (z-VAD-fmk) and the
caspase-8-specific inhibitor Z-lle-Glu(Ome)-Thr-Asp(Ome)-CH2F (z-IETD-fmk)
were purchased from Calbiochem (La Jolla, CA). The rabbit polyclonal
antibodies against caspase-3 and -8 were from Stressgen (Victoria, British
Columbia, Canada). The mouse mAbs against PARP and Bid were from PharMingen
(Bioclone, Marrickville, Australia). The rabbit mAbs against Bip1, eukaryotic
initiation factor2a (eIF2a), phosphorylated eIF2a, X-box-binding protein 1 (XBP1),
IRE1a, ATF6, PERK and CHOP were purchased from Santa Cruz
Biotechnology (Santa Cruz, CA). Isotype control antibodies used were the ID4.5 (mouse
IgG2a) mAb against Salmonella typhi supplied by Dr L.Ashman (Institute for
Medical and Veterinary Science, Adelaide, Australia), the 107.3 mouse IgG1
mAb purchased from PharMingen and rabbit IgG from Sigma Chemical Co.
Immunostaining on intact and permeabilized cells was carried out as described
previously (6,20). Analysis was carried out using a Becton Dickinson
(Mountain View, CA) FACScan flow cytometer.
Quantitation of apoptotic cells by measurement of subG1 DNA content using
the propidium iodide method was carried out as described elsewhere (6,20).
Mitochondrial membrane potential
Melanoma cells were seeded at 1 105 cells per well in 24-well plates and
allowed to reach exponential growth for 24 h before treatment. JC-1 staining
was performed according to the manufactures instructions (Molecular Probes,
Eugene, OR) as described previously (9,23).
Western blot analysis
Western blot analysis was carried out as described previously (9,23). Labeled
bands were detected by Immun-StarTM HRP chemiluminescent kit, and images
were captured and the intensity of the bands was quantitated with the Bio-Rad
VersaDocTM image system (Bio-Rad, Regents Park, New South Wales, Australia).
Detection of XBP1 mRNA splicing
The method used for detection of unspliced and spliced XBP1 mRNAs was as
described previously (24). Briefly, reverse transcriptionpolymerase chain
reaction (PCR) products of XBP1 mRNA were obtained from total RNA
extracted using primers 5#-cggtgcgcggtgcgtagtctgga-3# (sense) and
5#tgaggggctgagaggtgcttcct-3# (anti-sense). Because a 26 bp fragment containing
an Apa-LI site is spliced upon activation of XBP-1 mRNA, the reverse
transcriptionPCR products were digested with Apa-LI to distinguish the active
spliced form from the inactive unspliced form. Subsequent electrophoresis
revealed the inactive form as two cleaved fragments and the active form as
a non-cleaved fragment.
Total RNA was isolated with SV total RNA isolation system (Promega).
Reverse transcription PCR was carried out using Moloney murine leukemia virus
(MMLV) transcriptase and Oligo d(T) and the resulting cDNA products were
used as templates for real-time PCR assays. Real-time PCR was performed
using the ABI Prism 7700 sequence detection system (Applied Biosystems,
Foster City, CA). For TRAIL-R2, 25 ll mixture was used for reaction, which
contains 5 ll cDNA sample (0.51 lg/ll), 300 nM forward primers for
TRAIL-R2 (CGCTGCACCAGGTGTGATT), 300 nM reverse primers for
TRAIL-R2 (GTGCCGGCTTCGCACTGACA), 200 nM probes for
TRAILR2 (6FAM-CCCTGCACCACGACCAGAAACACAG-TAMRA) and 9 mM
MgCI2. After incubation at 50 C for 2 min followed by 95 C for 10 min,
the reaction was carried out for 45 cycles of the following: 95 C for 15 s
and 60.6 C for 45 s. For TRAIL-R1, assay-on-demand for TRAIL-R1 (Assay
ID: Hs01560092-g1) was used according to the manufacturers protocol
(Applied Biosystems). Analysis of cDNA for b-actin was included as a control.
The threshold cycle value (Ct) was normalized against b-actin cycle numbers.
The relative abundance of mRNA expression of a control sample was
arbitrarily designated as 1, and the values of the relative abundance of mRNA of
other samples were calculated accordingly.
Small interfering RNA
Melanoma cells were seeded at 3.5 104 cells per well in 24-well plates and
allowed to reach 50% confluence on the day of transfection. The small
interfering RNA (siRNA) constructs used were obtained as the siGENOME
SMARTpool reagents (Dharmacon, Lafayette, CO), the siGENOME
SMARTpool IRE1a (M-004951-01-0010), the siGENOME SMARTpool ATF6
(M-009917-01-0010) and the siGENOME SMARTpool CHOP
(M-00481901-0010). The non-targeting siRNA control, SiConTRol non-targeting SiRNA
pool (D-001206-13-20) was also obtained from Dharmacon. Cells were
transfected with 50100 nM siRNA diluted in Opti-Eagles Minimal Essential
Medium (MEM) (Invitrogen, Carlsbad, CA) in no-antibody Dulbeccos
modified Eagles medium with 5% fetal calf serum using Lipofectamine reagent
(Invitrogen) according to the manufacturers transfection protocol. Efficiency
of siRNA was measured by western blot analysis 24 h after transfection.
TG sensitizes melanoma cells to TRAIL-induced apoptosis
We studied if TG induces apoptosis of melanoma cells by treating
a panel of melanoma cell lines with the compound at 1 lM for 48 h.
As shown in Figure 1A, TG induced only minimal to moderate
levels of apoptosis in all the lines with the exception of Me1007, in
which .80% of the cells were found to undergo apoptosis. Figure 1B
shows that both Mel-RM and MM200 cells remained relatively
resistant to TG-induced apoptosis with no more than 20% of the cells
undergoing apoptosis even when treated with the compound at 8 lM.
To examine if TG sensitizes melanoma cells to TRAIL-induced
apoptosis, we treated Mel-RM and MM200 cells with the
combination of TG (1 lM) and TRAIL (200 ng/ml) for varying periods up to
48 h. Figure 1C shows that the levels of TRAIL-induced apoptosis
increased in both cell lines in the presence of TG, which was observed
as early as 16 h and reached a peak at 36 h after treatment. As shown
in Figure 1D and E, sensitization of Mel-RM and MM200 cells to
TRAIL-induced apoptosis by TG was associated with enhanced
TRAIL-induced activation of caspase-8 and Bid, reduction in
mitochondrial membrane potential, activation of caspase-3 and cleavage
of its substrate PARP.
A summary of studies on the effect of TG on TRAIL-induced
apoptosis in a panel of melanoma cell lines and normal cells including
melanocytes, HUVECs and fibroblasts is shown in Figure 1F. As
reported before (6,7), TRAIL induced varying degrees of apoptosis
in the melanoma cell lines. Pretreatment with TG markedly enhanced
TRAIL-induced apoptosis in all but Me1007. TRAIL alone did not
induce apoptosis in melanocytes, HUVECs and fibroblasts, but
treatment with TG alone resulted in apoptosis in .30% of HUVECs
(Figure 1F). Notably, pretreatment with TG followed by the addition
of TRAIL led to marked apoptosis in all three types of normal cells
(Figure 1F). It was of interest that Me1007 did not express caspase-8
and Bid at either the protein or the mRNA level (7).
TG selectively up-regulates TRAIL-R2 in melanoma cells
To study if sensitization of melanoma cells to TRAIL-induced
apoptosis by TG results from changes in the cell-surface expression of
TRAIL receptors, we treated Mel-RM and MM200 cells with TG at
1 lM for varying intervals. As shown in Figure 2A and B, TG markedly
up-regulated cell-surface expression of TRAIL-R2 in both cell lines,
with a significant increase being detected at 16 h, and further increases
at 24 and 36 h after exposure to the compound. In contrast, it did not
induce any change in the expression of the other TNF receptor family
members, TRAIL-R1, -R3 and -R4, Fas and TNF-R1 and -R2 on the
cell surface or did it cause up-regulation of TRAIL-R2 on the surface
of melanocytes, fibroblasts and HUVECs (Figure 2A and C and data
not shown). Studies on the panel of melanoma cell lines revealed that
TG could up-regulate TRAIL-R2 on the cell surface in all but Me1007
We next studied if TG regulates TRAIL-R2 total protein levels by
measuring TRAIL-R2 expression in permeabilized Mel-RM and
MM200 cells in flow cytometry. Figure 2D shows that TG induced
a marked increase in the levels of the TRAIL-R2 total protein with
a similar kinetics to that of up-regulation of TRAIL-R2 on the cell
surface (Figure 2B). Up-regulation of the TRAIL-R2 total protein
levels by TG was confirmed by western blot analysis (Figure 2D).
To understand the mechanism by which TG up-regulates
TRAILR2 expression in melanoma cells, we quantitated TRAIL-R2 mRNA
expression in Mel-RM and MM200 cells before and after exposure to
the compound for varying intervals. As shown in Figure 2E, treatment
with TG up-regulated the levels of TRAIL-R2 mRNA in both cell
lines. In contrast, there was no alteration in the levels of TRAIL-R1
mRNA after exposure to TG.
Because a significant increase in TRAIL-R2 expression at either the
protein or the mRNA level was initially observed at 16 h and a
tendency of the increase retained for up to 36 h, after treatment with
TG, we hereafter arbitrarily designated 16 h as a relatively early stage
and 36 h a relatively late stage, in regard to TG-induced up-regulation
of TRAIL-R2 in melanoma cells.
Up-regulation of TRAIL-R2 is responsible for sensitization of
melanoma cells to TRAIL-induced apoptosis by TG
The role of up-regulation of TRAIL-R2 in sensitization of melanoma
cells to TRAIL-induced apoptosis by TG was studied by inhibition of
the interaction between TRAIL and TRAIL-R2 using a TRAIL-R2/Fc
chimeric protein. Figure 3A shows that the TRAIL-R2/Fc chimera
significantly inhibited TRAIL-induced apoptosis in both Mel-RM
and MM200 cells in the absence or presence of TG. In contrast, it
had no effect on the low levels of apoptosis induced by TG alone.
Similarly, the TRAIL-R2/Fc chimera did not inhibit TG-induced
apoptosis in Me1007 cells (data not shown). As shown in Figure
3B, TG-mediated sensitization of melanoma cells to TRAIL-induced
apoptosis was blocked by either the pan caspase inhibitor z-VAD-fmk
or the caspase-8-specific inhibitor z-IETD-fmk.
To confirm that up-regulation of TRAIL-R2 expression is
responsible for sensitization of melanoma cells to TRAIL-induced apoptosis
by TG, we transfected a TRAIL-R2-specific siRNA pool into
Mel-RM and MM200 cells. Figure 3C and D shows that inhibition of
TRAIL-R2 expression by the TRAIL-R2 siRNA pool markedly
blocked TRAIL-induced apoptosis in the absence or presence of
TG. Similar to the TRAIL-R2/Fc chimera, the TRAIL-R2 siRNA
had no effect on apoptosis induced by TG alone.
The role of the IRE1a- and ATF6-mediated signaling pathways of the
UPR in up-regulation of TRAIL-R2 by TG in melanoma cells
We next studied if TG induces the UPR in melanoma cells. Figure 4A
and B shows that, similar to our findings in a separate study with
another ER stress inducer tunicamycin (25), exposure of melanoma
cells to TG resulted in changes consistent with activation of the UPR
including up-regulation of the ER chaperone protein Bip/GRP78,
phosphorylation of the translation initiator eIF2a, increases in
XBP1 mRNA and protein levels and XBP1 mRNA splicing, increases
in the expression of the UPR transducers, IRE1a and ATF6 and
appearance of cleaved form of ATF6 (1214,24).
To elucidate if any of the known UPR-signaling pathways play a role
in up-regulation of TRAIL-R2 by TG in melanoma cells, we silenced
IRE1a and ATF6 by specific siRNA pools in Mel-RM and MM200 cells,
respectively. Figure 4C shows that siRNA knockdown of IRE1a
markedly inhibited its basal expression and its up-regulation by TG.
Similarly, inhibition of ATF6 by siRNA reduced its expression levels as
either the native p90 form or as the cleaved p50 form in cells before
and after treatment with TG. As shown in Figure 4D and E, siRNA
knockdown of either IRE1a or ATF6 had no effect on up-regulation of
TRAIL-R2 measured at 16 h after treatment with TG. In contrast, by 36 h,
the levels of TRAIL-R2 expression in cells transfected with either the
IRE1a or ATF6 siRNA were markedly reduced in comparison with
those transfected with the control siRNA, with the mean fluorescence
intensities being approximately the same as those in parental cells.
CHOP is involved in up-regulation of TRAIL-R2 by TG in
We studied the expression of the transcription factor CHOP, which is
a UPR effector primarily downstream of PERK and has been shown to
be responsible for up-regulation of TRAIL-R2 induced by TG (12
14,1618), in Mel-RM and MM200 cells before and after treatment
with TG. As shown in Figure 5A, surprisingly, CHOP was
constitutively expressed at relatively high levels in both cell lines, but was not
up-regulated by TG. Instead, treatment with TG resulted in a marked
decrease in CHOP expression.
To clarify a potential role of CHOP in up-regulation of TRAIL-R2
by TG, we inhibited CHOP by transiently transfecting a CHOP siRNA
pool into Mel-RM and MM200 cells. Figure 5B shows that siRNA
knockdown of CHOP inhibited its expression in both cell lines. As
shown in Figure 5C and D, the CHOP siRNA markedly inhibited the
increase in TRAIL-R2 at 16 h after treatment with TG. However, by
36 h, the inhibitory effect of CHOP on up-regulation of TRAIL-R2 in
both cell lines was attenuated, with the levels of the TRAIL-R2 in
cells transfected with the CHOP siRNA being only moderately lower
than those in cells transfected with the control siRNA.
TG reversed resistance of TRAIL-selected melanoma cells to
TRAILinduced apoptosis by restoring TRAIL-R2 expression
We have previously reported that melanoma cells selected for TRAIL
resistance expressed low levels of TRAIL-R2 (810). To study if TG
may up-regulate TRAIL-R2 and enhance TRAIL-induced apoptosis in
these cells, we treated Mel-RM.S and MM200.S cells with TG for 24 h.
As shown in Figure 6A and B, treatment with TG markedly increased
the levels of TRAIL-R2 on the cell surface and the TRAIL-R2 total
protein levels in both Mel-RM.S and MM200.S cells. Figure 6C
shows that pretreatment with TG followed by the addition of TRAIL
resulted in a marked increase in the percentages of apoptotic cells.
Sensitization of the TRAIL-selected melanoma cells to
TRAIL-induced apoptosis by TG was substantially inhibited by a recombinant
TRAIL-R2/Fc chimera (Figure 6D), indicating that the effect of TG
on TRAIL-induced apoptosis in the TRAIL-selected cells is
accounted for by the increase in TRAIL-R2 expression on the cell
The present study shows that TG, an inhibitor of ER Ca2 ATPases
(15), can potently enhance TRAIL-induced apoptosis in human
melanoma cells by selectively up-regulating TRAIL-R2 on the cell
surface. In contrast to previous reports in other cellular systems (1618),
TG by itself induced only minimal apoptosis in the majority of
melanoma cell lines, which did not appear to involve in the
TRAIL-R2mediated apoptotic pathway. We demonstrate, for the first time, that
the IRE1a- and ATF6-mediated signaling pathways of the UPR and
the transcription factor CHOP both contribute to up-regulation of
TRAIL-R2 by TG in melanoma cells.
Although TRAIL appears to be a promising candidate for cancer
therapy (15), our past studies indicated that fresh isolates of
melanoma and melanoma in tissue sections frequently had low TRAIL
death receptor expression and therefore may be unresponsive to
TRAIL (811). However, unlike studies in many other solid cancers,
in which TRAIL death receptors could be up-regulated by other
therapeutic drugs (2629), we have not found these to increase TRAIL
death receptor expression in melanoma. Agents tested have included
DNA-damaging agents, microtubulin-targeting agents, histone
deacetylase inhibitors and extracellular signal-regulated protein kinase
kinase (MEK) inhibitors (23) (data not shown). The ability of TG to
upregulate TRAIL-R2 in melanoma is therefore of particular interest.
Importantly, up-regulation of TRAIL-R2 by TG appeared to be highly
selective and did not up-regulate receptors for the other TNF receptor
family members, TRAIL-R1, -R3 and -R4, TNF-R1 and -R2 and Fas.
Moreover, TG did not up-regulate TRAIL-R2 expression in normal
cells, including melanocytes, fibroblasts and HUVECs. Our results
and those showing that TG can also increase TRAIL-R2 expression
in other solid cancer cells indicate that selective up-regulation of
TRAIL-R2 by TG would be an advantage for its potential clinical
Up-regulation of TRAIL-R2 by TG was associated with enhanced
apoptotic signaling induced by TRAIL. This was shown by increased
activation of caspase-8 and Bid, reduction in mitochondrial
membrane potential, activation of caspase-3 and cleavage of its substrate
PARP. Caspase-8 and -3 are the major initiator and effector caspase,
respectively, in TRAIL-induced apoptosis of melanoma cells,
whereas Bid is the essential mediator that links the death receptor
apoptotic pathway to the mitochondrial apoptotic pathway (3,7,30
32). The latter is known to play an important role in TRAIL-induced
apoptosis of melanoma (7,33). The finding that a TRAIL-R2/Fc
chimera, a caspase-8-specific inhibitor, or a TRAIL-R2 siRNA pool
efficiently blocked TRAIL-induced apoptosis in the presence of TG
indicates that enhanced apoptotic signaling was due to the increased
interactions between TRAIL and TRAIL-R2.
Although these results show that TG up-regulated TRAIL-R2 and
enhanced TRAIL-induced apoptosis of melanoma, our studies show
that direct induction of apoptosis by TG appeared to be largely
independent of the TRAIL-R2 apoptotic-signaling pathway, e.g. the
majority of melanoma cell lines were insensitive to TG-induced
apoptosis despite up-regulation of TRAIL-R2 by the compound.
Second, TG induced marked apoptotic cell death in a cell line that
expressed low levels of TRAIL-R2 even in the presence of TG and
contained no caspase-8 and Bid. Third, a TRAIL-R2/Fc chimera and/
or TRAIL-R2 siRNA did not inhibit TG-induced apoptosis in
melanoma cells. Besides TRAIL-R2, a number of potential regulators have
been identified to participate in ER stress-induced apoptosis, such as
the caspase-12 in mice and its human homolog, caspase-4, the
BH3only protein p53 up-regulated modulator of apoptosis (PUMA), and
activation of c-jun N-terminal kinase (3437). Irrespective of the
pathway involved in induction of apoptosis in melanoma cells by
ER stress, the present results show that it is inhibited in the majority
of melanoma cell lines.
The transcription factor CHOP is known to be induced by ER stress
and involved in ER-mediated apoptosis (1214,1618). However, we
show here that CHOP is constitutively expressed at relatively high levels
in melanoma cells and is decreased rather than increased by the ER
stress inducer TG. CHOP transcription has been shown to be primarily
regulated by the transcription factor ATF4, which was up-regulated as
a result of preferential translation of its mRNA upon phosphorylation/
inactivation of eIF2a by activation of PERK (1214). The high
expression of CHOP in melanoma cells may indicate that the UPR is
constitutively activated as reported in other cancer cells (3841). However,
phosphorylation of eIF2a was merely detectable in melanoma cells
before treatment with TG. This suggests that factors other than ATF4
may be responsible for the constitutively high levels of CHOP in the
melanoma cells. The mechanism involved in the decrease of CHOP
after exposure to TG is unknown, but it may be associated with reduced
stability of CHOP mRNA and protein in ER stress-adapted cells (42).
Despite its TG-induced reduction, CHOP appeared to play a critical
role at the initial phase of up-regulation of TRAIL-R2 induced by TG.
This was shown by knockdown experiments with CHOP siRNA,
which inhibited up-regulation of TRAIL-R2 at 16 h, but had minimal
effects by 36 h after treatment with TG. The TRAIL-R2 promoter is
known to contain a CHOP-binding site, which has been shown to play
a role in up-regulation of TRAIL-R2 upon activation of the UPR
(16,41). Inhibition of IRE1a or ATF6 by siRNA did not block the
initial increase in TRAIL-R2 induced by TG, suggesting that CHOP
acts independently of IRE1a or ATF6 in melanoma cells. The
minimal effect of CHOP on up-regulation of TRAIL-R2 at later stages of
the UPR may be due to the decrease in the CHOP expression levels
after exposure to TG.
In contrast to CHOP the siRNA knockdown studies showed that the
IRE1a- and ATF6-mediated signaling pathways of the UPR did not
appear to be involved in the initial phase of the increase in TRAIL-R2
but was responsible for the late stage (36 h) of the increase after
exposure to TG. Both IRE1a and ATF6 are ER membrane-localized
proteins that act as UPR transducers (1214). On activation of the
UPR, IRE1a displays its RNase activity that cleaves XBP1 mRNA
generating a splicing variant of XBP1 mRNA that encodes a potent
transcription factor. This in turn activates transcription of many UPR
target genes. ATF6 itself is a transcription factor that, on activation,
relocates to the Golgi where it is cleaved into the smaller active form
that activates transcription of UPR target genes. It is conceivable that
both the effector of IRE1a, XBP1 and ATF6 may act directly or
indirectly via other transcription factors to activate transcription of
TRAIL-R2 in melanoma cells (1214). No binding site for XBP1 or
ATF6 has so far been identified in TRAIL-R2 promoter region. The
only other transcription factor that is known to regulate TRAIL-R2 is
p53 (26,27), which does not appear to play a role in the increase of
TRAIL-R2 transcription induced by TG in that TRAIL-R2 was also
up-regulated in melanoma cells deficient in p53 or containing mutant
p53 (data not shown). Further studies are required to identify the
factors that are responsible for up-regulation of TRAIL-R2 by the
IRE1a- and ATF6-signaling pathways.
Our finding that TG could up-regulate TRAIL-R2 and re-sensitize
melanoma cells that had been selected for resistance to TRAIL would
appear to be of potential importance in the clinical use of TRAIL.
TRAIL-selected cells are known to mimic fresh melanoma isolates in
that the latter are also relatively resistant to TRAIL-induced apoptosis
due to low levels of TRAIL death receptor expression (810).
However, treatment with TG followed by the addition of TRAIL resulted
in increased toxicity against melanocytes, fibroblasts and HUVECs.
TRAIL-R2 was not up-regulated in the normal cells which indicates
that mechanisms other than regulation of TRAIL-R2 are involved in
modulating their sensitivity to TRAIL-induced apoptosis by the UPR.
Further studies are therefore needed to assess the safety of this
approach e.g. whether low doses of TG in combination with low
concentrations of TRAIL may be effective. Alternatively, whether other
agents togethering the ER Ca2 ATPase may have more selective
effects on melanoma compared with normal cells.
New South Wales State Cancer Council; the Melanoma and Skin
Cancer Research Institute Sydney; the Hunter Melanoma Foundation,
New South Wales; the National Health and Medical Research Council
(Project grant 351114), Australia.
X.D.Z. is a Cancer Institute New South Wales Fellow.
Conflict of Interest Statement: None declared.