Epithelial -to-Mesenchymal Transition Activates PERK–eIF2a ... · constitutively activate the PERK–eIF2α axis of the unfolded protein response (UPR). Protein kinase RNA-like
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
702 | CANCER DISCOVERY�JUNE 2014 www.aacrjournals.org
RESEARCH ARTICLE
Epithelial -to-Mesenchymal Transition Activates PERK–eIF2a and Sensitizes Cells to Endoplasmic Reticulum Stress Yu-xiong Feng 1 , Ethan S. Sokol 1 , 2 , Catherine A. Del Vecchio 1 , Sandhya Sanduja 1 , Jasper H.L. Claessen 1 , Theresa A. Proia 1 , Dexter X. Jin 1 , 2 , Ferenc Reinhardt 1 , Hidde L. Ploegh 1 , 2 , Qiu Wang 6 , and Piyush B. Gupta 1 , 2 , 3 , 4 , 5
Authors’ Affi liations: 1 Whitehead Institute for Biomedical Research; 2 Department of Biology, Massachusetts Institute of Technology; 3 Koch Institute for Integrative Cancer Research; 4 Harvard Stem Cell Institute; 5 Broad Institute, Cambridge, Massachusetts; and 6 Department of Chemis-try, Duke University, Durham, North Carolina
Note: Supplementary data for this article are available at Cancer Discovery Online (http://cancerdiscovery.aacrjournals.org/).
Corresponding Author: Piyush B. Gupta, Whitehead Institute for Biomedi-cal Research, 9 Cambridge Center, Cambridge, MA 02142. Phone: 617-258-7778; Fax: 617-258-7226; E-mail: [email protected]
pounds tested, this screen identifi ed a few structurally related
small molecules (Cmp302, Cmp308, and Dev4) with EMT-
selective toxicity ( Fig. 1A ). These compounds exhibited between
20-fold and >100-fold selective toxicity toward nontumorigenic
(HMLE) and tumorigenic (HMLER) human mammary epithe-
lial cells induced through an EMT by inhibition of E-cadherin
(shEcad) or overexpression of Twist (Supplementary Fig. S1A
and S1B). Treatment of GFP-EMT and DsRed–non-EMT cell
cocultures with Cmp302, Cmp308, or Dev4 selectively depleted
GFP-EMT cells from the cocultures, further confi rming the
selective toxicity of these compounds. In contrast, two common
chemotherapy drugs, paclitaxel and doxorubicin, caused enrich-
ment of GFP-EMT cells within cocultures ( Supplementary
Figure 1. Small molecules with EMT-selective toxicity induce ER stress. A, schematic illustration of large-scale chemical screen. B, microarray analysis was performed on HMLE_shGFP and HMLE_Twist cells treated with or without Cmp302. GSEA was performed with four gene sets (see Methods for details) on the microarray dataset where Cmp302-induced genes in HMLE_Twist cells were ordered from largest to smallest. (continued on following page)
0.8
0.6
0.4
0.2
0
Rank in ordered Cmp302 dataset
Enrichm
ent
score
Rank in ordered Cmp302 dataset
Enrichm
ent
score
0.8
0.6
0.4
0.2
0
Rank in ordered Cmp302 dataset
Gene set doxorubicin, P = 0.058
Rank in ordered Cmp302 dataset
Enrichm
ent
score 0.3
0.2
0.1
–0.1
0
–0.2Enrichm
ent
score 0.3
0.2
0.1
–0.1
0
A Compound library (~315,000)
EMT cellsNon-EMT cells
EMT-selective compounds
Cl
O
O
NH
NNH
Cl O
O
O
NH
NNH
Cl
Cl
O
O
NH
NH
N
Cl
Dev4Cmp308Cmp302
B
Gene set hypoxia, P = 0.19
Gene set tunicamycin, P < 0.001 Gene set thapsigargin, P < 0.001
Figure 1. (Continued) C, expression of UPR pathway components in HMLER_shGFP and HMLER_Twist cells that were treated with increasing con-centrations of Dev2, Cmp302, Dev4, or DMSO solvent for 6 hours. Western blot analysis for phospho-eIF2α (p-eIF2α), total eIF2α, CHOP, and β-tubulin. RT-PCR analysis of XBP1 and XBP1 splice variant and GAPDH transcripts. D, ATF6 activation of HMLER_shGFP and HMLER_Twist cells in response to increasing concentrations of Dev2, Cmp302, Dev4, or DMSO solvent for 6 hours. ATF6 activation was measured by an ATF6 reporter assay. Reporter activity for each cell line was normalized relative to DMSO treatment. E, qPCR analysis for BiP expression in HMLE_shGFP, HMLE_Twist, and HMLE_shEcad cells treated with 4 μmol/L of Dev2, Cmp302, Dev4, or DMSO solvent for 30 hours. BiP expression was normalized relative to GAPDH for each sample. *, P < 0.05; **, P < 0.01.
EMT Activates PERK and Sensitizes Cells to ER Stress RESEARCH ARTICLE
Figure 2. EMT sensitizes cells to chemicals that perturb ER function. A, Western blot analysis of HMLER_shGFP and HMLER_Twist cells treated with vehicle (DMSO) or increasing concentrations of tunicamycin (Tm), thapsigargin (Tg), DTT, or A23187 for 4 hours and probed for phospho-eIF2α (p-eIF2α), CHOP, and β-tubulin (loading control). RT-PCR analyses of XBP1 and XBP1 splice variant and GAPDH transcripts. B, qPCR analysis of BiP expression in HMLER_shGFP and HMLER_Twist cells treated with increasing concentrations of thapsigargin or DTT. BiP expression was normalized relative to GAPDH for each sample. C, qPCR analysis of GADD34 expression in HMLER_shGFP and HMLER_Twist cells treated with increasing concentrations of thapsigargin or DTT. GADD34 expression was normalized relative to GAPDH for each sample. D, dose–response curves of HMLER_shGFP (blue circle), HMLER_shEcad (red square), and HMLER_Twist (black diamond) cells treated with various concentrations of tunicamycin, thapsigargin, DTT, or A23187 for 3 days. Cell survival was determined using an ATP-based luminescence assay. E, representative fl uorescence images and quantifi cation of dsRed-labeled non-EMT (HMLER_shGFP) and GFP-labeled EMT (HMLER_Twist) cells mixed in 1:1 ratio and treated with 5 nmol/L thapsigargin, 10 nmol/L thapsigargin, or solvent control for 5 days. Scale bar, 50 μm. F, dose–response curves of four luminal breast cancer cells, MCF7, MDA361, T47D, and ZR-75-3 (blue curves), and six basal-B lines, BT549, 4T1, Hs578T, MDA231, MDA436, and MDA157 (red curves) treated with various concentrations of tunicamycin, thapsigargin, DTT, or A23187 for 3 days. Cell survival was determined using an ATP-based luminescence assay. *, P < 0.05; **, P < 0.01.
708 | CANCER DISCOVERY�JUNE 2014 www.aacrjournals.org
Feng et al.RESEARCH ARTICLE
Figure 3. Cells that undergo an EMT are highly secretory. A, expression of 12 genes encoding ECM proteins in EMT cells (HMLE_Gsc, HMLE_shEcad, HMLE_Snail, HMLE_TGFβ, and HMLE_Twist) relative to control HMLE epithelial cells. B, confocal microscopy and quantifi cation of Sec16-GFP localization to ERES in EMT (HMLE_shEcad) and control (HMLE_Ctrl) cells. Data were presented as mean +/− SD. Nuclei counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Scale bar, 10 μm. C, autoradiograph showing 35 S-methionine/cysteine–labeled secreted proteins in EMT (HMLE_shEcad, HMLE_Twist) and control (HMLE_shGFP) cells. Secreted proteins were harvested at the indicated time points. Quantifi cation of signal in each lane is provided in arbitrary units. D, representa-tive electron microscopy images and quantifi cation of ER branching in EMT (HMLE_Twist) and non-EMT (HMLE_shGFP) cells. Arrows, examples of ER branch points in HMLE_Twist cells; scale bar, 500 nm. E, expression of genes encoding secreted ECM proteins in a panel of luminal ( N = 13; blue) and basal-B ( N = 9; red) human breast cancer lines. These data were derived from GSE16795 ( 24 ). F, autoradiograph showing 35 S-methionine/cysteine–labeled secreted proteins in luminal and basal-B human breast cancer lines. Quantifi cation of signal in each lane is provided in arbitrary units. *, P < 0.05.
HMLE_Ctrl
HMLE_shEcad
*
0
HM
LE_C
trl
HM
LE_s
hEca
d
50
100
150
200
ER
ES
HMLE_shGFP HMLE_shEcad HMLE_Twist
0 1 3 5 0 1 3 5 0 1 3 5
250150
100
75
50
37
25
Hour
2468
2468
2468
2468
DCN NID1 SPOCK
COL1A2 FBN1 FBLN1
FBLN5 POSTN COL6A1
COL3A1 COL1A1 COL5A2Log
2 [fo
ld c
hange] in
EM
T v
s. contr
ol sta
te
Gsc
shE
cad
Snail
TG
Fβ
Tw
ist
Gsc
shE
cad
Snail
TG
Fβ
Tw
ist
Gsc
shE
cad
Snail
TG
Fβ
Tw
ist
Rela
tive e
xpre
ssio
n
COL4A3
COL10A
1
COL13A
1
COL15A
1
COL1A1
COL1A2
COL4A1
COL5A1
POSTN
COL6A1
FN1FBN1
0.1
1
10
100
1,000
10,000
LuminalBasal-B
Unchanged
in EMT
Upregulated
in EMT
HMLE_Ctrl HMLE_Twist
Unbranched cells
1–4 branch points
4–10 branch points
>10 branch points
HMLE_Ctrl HMLE_Twist
A B C
D E F Luminal Basal-B
MC
F7
T4
7D
BT
47
4
ZR
-75-3
Hs5
78
T
BT
549
MD
A157
SU
M159
MD
A231
4T
1
250
150
100
75
50
37
25
KD
KD
non-EMT cells had ER membranes with one or more branch
points ( Fig. 3D ). Because professional secretory cells (e.g., pan-
creatic β cells) often display a highly developed ER network
( 34 ), this further suggested that as part of their function, EMT
cells also have an increased demand for protein secretion.
To determine whether increased ECM secretion is a general
feature of EMT, we examined the expression of ECM genes iden-
tifi ed to be upregulated upon EMT ( Fig. 3A ) in basal-B and lumi-
nal breast cancer lines ( 35 ). Basal-B cancer lines ( n = 9) expressed
many EMT ECM genes—including FN1 , COL1A1 , COL1A2 ,
of PAI1 and FN1 also strongly abrogated UPR induction in
response to either Dev4 or thapsigargin treatment ( Fig. 4C
and D , HMLE_shEcad and HMLE_Twist cells, respectively;
Supplementary Fig. S4C). These fi ndings indicated that ECM
secretion was required for EMT cells to migrate, while also
increasing their sensitivity to ER perturbations.
EMT Increases Dependence on the ER Chaperone BiP Nascent polypeptides en route to secretion are folded by
critical chaperone proteins that reside within the ER. Because
EMT cells are more secretory and therefore have a higher ER
load, we hypothesized that they might also be more sensitive
to reductions in chaperone proteins. To test this, we used
shRNAs to inhibit the key ER chaperone BiP ( 37 ) in a cell
line model in which EMT could be induced within 3 days by
addition of 4-hydroxytamoxifen (4-OHT; HMLE_ER_Twist;
ref. 3 ). Using two different shRNAs, a 65% to 75% reduction
in BiP had negligible effects on the viability of this line in the
uninduced (non-EMT) condition ( Fig. 5A and B ). However,
induction of EMT in these shBiP lines caused signifi cant
reduction of cell growth (8-fold less in mesenchymal vs.
epithelial cells), and the surviving cells were clustered in
epithelial islands ( Fig. 5B ). This indicated that the reduced
BiP levels, although suffi cient for the needs of epithelial cells,
were not suffi cient for cells to survive EMT. Inhibition of BiP
also differentially affected the viability of basal-B (EMT-like)
and luminal (non-EMT-like) breast cancer lines. Although
BiP inhibition only modestly affected the viability of two
luminal lines (MCF7, T47D), it caused signifi cant death in
two basal lines (MDA-231 and BT549) together with CHOP
upregulation ( Fig. 5C and D ), suggesting that ER stress was
more readily induced in BiP-defi cient EMT cells. This was
confi rmed by examining UPR signaling, which revealed that
the UPR was activated upon BiP inhibition in the basal-B
cancer cells, but not the luminal cancer cells (Supplementary
Fig. S5).
PERK–eIF2a–ATF4 Signaling Is Activated upon EMT and Promotes Malignancy
Before their differentiation, progenitors of secretory cells
activate UPR pathways in anticipation of an increased ER
load ( 38, 39 ); this UPR activation is not a response to ER
stress but rather a means of preventing it. Because EMT cells
are also highly secretory, we examined whether, in the absence
of ER stressors, they also activate one or more UPR pathways.
Compared with non-EMT cells, EMT cells had reduced PERK
protein mobility suggestive of its phosphorylation, increased
eIF2α phosphorylation ( Fig. 6A ), and increased expression of
the downstream gene GADD34 ( Fig. 6B ). In contrast, IRE1
signaling was not increased in EMT or non-EMT cells in the
absence of exogenous ER stressors (Supplementary Fig. S6A).
To confi rm that PERK was in fact phosphorylated in EMT cells,
we also performed phosphatase treatments and immunofl uo-
rescence with a phospho-specifi c PERK antibody. Treatment
of lysates with lambda phosphatase before Western blotting
Figure 4. ECM secretion upon EMT sensitizes cells to ER stress and promotes migration. A, Western blot analysis showing stable shRNA-mediated knockdown of FN1 and PAI1, individually and in combination, in HMLE_shEcad and HMLE_Twist cells. Two distinct hairpins were applied per gene. DK-1 refers to double knockdown of FN1 and PAI1 by hairpins shFN1-1 and shPAI1-1, and DK-2 refers to double knockdown of FN1 and PAI1 by hairpins shFN1-2 and shPAI1-2. B, migratory ability of HMLE_shEcad_shLuc and HMLE_shEcad_DK-1 cells, HMLE_Twist_shLuc, and HMLE_Twist_DK-1 cells was measured using an in vitro wound-healing assay. Representative images and quantifi cations at 0 hours and 7 or 8 hours postwounding are shown. C, expression of UPR pathway components in HMLE_shEcad_shLuc and HMLE_shEcad_DK-1 cells treated with increasing concentrations of Dev4, thapsigargin, or DMSO solvent for 6 hours. Western blot analysis is shown for phospho-eIF2α (p-eIF2α), total eIF2α, and β-tubulin. RT-PCR analysis is shown for unspliced XBP1 , spliced XBP1 , and GAPDH transcripts. D, similar analysis of C was applied to HMLE_Twist_shLuc and HMLE_Twist_DK-2 cells. **, P < 0.01.
710 | CANCER DISCOVERY�JUNE 2014 www.aacrjournals.org
Feng et al.RESEARCH ARTICLE
abolished the reduced PERK mobility present in EMT cells
under basal conditions; as a control, phosphatase treatment
also abolished the reduced PERK mobility caused by thapsi-
gargin in both EMT and non-EMT cells ( Fig. 6C ). Immunofl uo-
rescence with a phosphorylation-specifi c antibody also showed
that PERK was constitutively activated upon EMT, but not in
non-EMT cells ( Fig. 6D ). Consistent with these fi ndings, cells
induced through an EMT by TGFβ treatment also activated
PERK but not IRE1 signaling (Supplementary Fig. S6B).
Because there are several kinases upstream of eIF2α, we
next examined whether its phosphorylation in EMT cells
was dependent on PERK. Suppression of PERK activity with
a specifi c chemical inhibitor strongly decreased both PERK
and eIF2α phosphorylation in EMT cells (Supplementary Fig.
S6C). Similarly, PERK inhibition by shRNA also decreased
eIF2α phosphorylation in two basal-B breast cancer lines
( Fig. 6E ). Collectively, these observations established that
the PERK–eIF2α–ATF4 branch of the UPR is selectively and
constitutively induced by cells that have undergone an EMT.
Depending on the context, UPR signaling can either pro-
mote survival or induce apoptosis in cells challenged with ER
stress ( 7 ). Inhibition of PERK in EMT cells with a chemical
inhibitor dramatically increased their sensitivity to thapsi-
gargin ( Fig. 6F ), indicating that activation of the PERK path-
way is adaptive and benefi cial for the survival of cancer cells
that have undergone an EMT. We next examined whether
PERK signaling also contributed to the malignant properties
of EMT cells. PERK inhibition strongly reduced the ability
of EMT cells to form tumorspheres ( Fig. 6G ) and migrate in
Transwell assays ( Fig. 6H ); at the same dose, the PERK inhibi-
tor minimally affected cell proliferation (Supplementary Fig.
S7A and S7B). Pretreatment of metastatic 4T1 cells with either
the PERK inhibitor or thapsigargin resulted in signifi cantly
Figure 5. EMT cells require higher BiP expression for their survival. A, left, BiP mRNA expression levels in HMLE_ER_Twist cells transduced with a control hairpin targeting LacZ or two different hairpins targeting BiP. Right, representative growth curves of control (LacZ) or BiP-inhibited (shBiP) HMLE_ER_Twist cells treated with or without 125 nmol/L 4-OHT to induce EMT. B, representative bright-fi eld images of the morphology of control (LacZ) or BiP-inhibited (shBiP) HMLE_ER_Twist cells treated with or without 4-OHT to induce EMT. Images were taken 4 days after 4-OHT treatment. C, left, quantifi cation of cell viability of non-EMT luminal breast cancer cell lines (MCF7 and T47D) and EMT Basal-B cell lines (MDA231,BT549) transduced with control or BiP-targeted hairpins. Right, RT-PCR expression of CHOP mRNA expression in cells of the left. D, representative bright-fi eld images of the morphology of cell lines in C, 4 days after hairpin transduction. *, P < 0.05; **, P < 0.01.
A
D
B
C
0
4
8
12
MCF7 T47D MDA-231 BT549
CH
OP
exp
ressio
n
0
7
14
0 3 6 9
Nu
mb
er
of
ce
lls (
Lo
g2)
Days
shLacZ
shBiP-1
shBiP-2
shLacZ_4OHT
shBiP-1_4OHT
shBiP-2_4OHT
0
0.4
0.8
1.2
HMLE_ER_Twist
BiP
exp
ressio
nshLacZ
shBiP-1
shBiP-2
0
0.4
0.8
1.2
MCF7 T47D MDA-231
****
****
**
**
*
*
BT549
Re
lative
ce
ll su
rviv
al
shLuc shBiP-1 shBiP-2 shLuc shBiP-1 shBiP-2
shLacZ shBiP-1 shBiP-2
DM
SO
4-O
HT
shLuc shBiP-1 shBiP-2
MC
F7
MD
A-2
31
diminished metastatic capacity, as assessed by lung tumor bur-
den 15 days after tail-vein injection ( Fig. 6I ). Collectively, these
fi ndings indicated that disruption of the PERK pathway sig-
nifi cantly compromises the malignant phenotype of EMT can-
cer cells and further increases their sensitivity to ER stressors.
EMT Correlates with PERK but Not IRE1 Signaling in Primary Human Tumors
We next examined the clinical relevance of the above fi nd-
ings by assessing primary human tumors. Primary cancer
cells (<3 passages) from breast tumors expressing EMT mark-
ers had elevated PERK and BiP expression, and increased
eIF2α phosphorylation, when compared with primary breast
cancer cells that did not express EMT markers ( Fig. 7A and B ).
Primary cancer cells expressing EMT markers were also more
sensitive to the ER stressor thapsigargin as indicated by UPR
pathway activation ( Fig. 7B ). Consistent with this, these cells
also exhibited signifi cantly reduced viability upon treatment
with ER stressors ( Fig. 7C and Supplementary Fig. S8).
To assess whether these fi ndings extended to other tumor
types, we analyzed gene-expression microarray data from
patient tumors to test for associations between the expres-
sion of EMT, ECM, and UPR pathway genes (see Methods
for details). This analysis revealed that the expression of EMT
and ECM genes is strongly correlated across patient tumors
and could be observed in fi ve datasets spanning 792 breast,
colon, gastric, and lung tumors, as well as metastatic tumors
( Fig. 7D ). EMT and ATF4 genes were also strongly correlated
in their expression (mean corr. = 0.80), whereas a signifi cant
correlation was not observed between the expression of EMT
and XBP1 genes (mean corr. = −0.14; Fig. 7D ). These fi ndings
established that EMT is strongly associated with PERK but
not IRE1 signaling across a spectrum of tumor types.
EMT Activates PERK and Sensitizes Cells to ER Stress RESEARCH ARTICLE
Figure 6. PERK signaling is constitutively activated upon EMT and promotes malignancy. A, Western blot analysis of EMT (shEcad and Twist) or control (shGFP) HMLE cells, luminal (MCF7, T47D, BT474, and ZR-75-30) and basal-B human breast cancer cell lines (SUM159, MDA-MB-231, MDA-MB-157, Hs578T, and BT549) for UPR pathway components; β-tubulin is used as a loading control. B, expression of GADD34 in EMT (shEcad and Twist) or control (shGFP) HMLE cells (left), luminal (MCF7, T47D, BT474, and ZR-75-3) and basal-B (SUM159, MDA-MB-231, MDA-MB-157, Hs578T, and BT549) human breast cancer lines (right). C, cell lysates from control (HMLE_shGFP) and EMT (HMLE_shEcad) cells treated with or without thapsigargin were col-lected. The lysates were then treated with or without lambda phosphatase, and PERK protein expression was analyzed by Western blotting. D, non-EMT (HMLE_shGFP) and EMT (HMLE_Twist) cells were treated with DMSO control, 40 nmol/L thapsigargin for 2 hours, or 1 μmol/L PERK inhibitor (PERKi) for 24 hours. Phosphorylated PERK (p-PERK) protein was analyzed by immunofl uorescence staining. E, Western blotting for PERK and phospho-eIF2α expres-sion in cell lysates from the Hs578T and SUM159 lines infected with control (shCtrl) or PERK-specifi c (shPERK) hairpins. F, non-EMT (HMLE_shGFP) and EMT (HMLE_shEcad and HMLE_Twist) cells were cotreated with 1.5 nmol/L thapsigargin and 0, 0.5 or 1 μmol/L of PERK inhibitor for 6 days. Surviving cells were quantifi ed by cell counting. Data are represented as mean + SE from three replicates. G, HMLE_shGFP and HMLE_shEcad cells were pretreated with 1 μmol/L PERK inhibitor or vehicle control (DMSO) for 2 days before tumorsphere formation assay. The PERK inhibitor–pretreated cells were then cultured in tumorsphere- forming condition for another 4 days in the presence of 1 μmol/L PERK inhibitor, while the vehicle-pretreated cells were cultured in drug-free conditions for the same period of time. Representative bright-fi eld images and quantifi cation were shown. Scale bar, 50 μm. H, representative images of crystal violet staining of HMLE_shGFP and HMLE_shEcad cells pretreated with or without 1 μmol/L PERK inhibitor for 2 days following Transwell migra-tion assay. Cells that migrated within 8 hours following seeding into 8-μm pore inserts were visualized and quantifi ed. Scale bar, 50 μm. I, 4T1 cells were treated with DMSO control, 1 μmol/L PERK inhibitor for 3 days, or 2 nmol/L thapsigargin for 3 days followed by a 4-day recovery period in drug-free media. A total of 2 × 10 6 cells were injected into NOD/SCID mice via the tail vein, and lung tissues were harvested 15 days after injection. Representative images of mouse lung tissue stained with H&E are shown. Quantifi cation of metastasis is also shown (5 mice per condition). *, P < 0.05; **, P < 0.01.
712 | CANCER DISCOVERY�JUNE 2014 www.aacrjournals.org
Feng et al.RESEARCH ARTICLE
DISCUSSION
Given the central role of EMT in tumor metastasis and ther-
apy resistance, there is a vital need to identify pathways and proc-
esses that modulate either the survival or malignancy of cancer
cells that have undergone EMT. In this study, we have assessed
the effects of EMT-selective small-molecule probes in the context
of global transcriptional profi ling. This revealed that EMT cells,
by virtue of their increased secretion of ECM, are highly sensitive
to ER stress. This fi nding is noteworthy because EMT cells are
resistant to a wide range of chemotherapies, and because the
secretory output of a cell has not previously been shown to infl u-
ence its sensitivity to chemicals that cause ER stress.
Our fi ndings are consistent with prior studies linking EMT
induction with ECM secretion. However, although the impor-
tance of ECM for tumor progression is well established ( 36 ),
our study is the fi rst to suggest that ECM secretion, while
promoting malignancy, also creates a key cellular vulnerabil-
ity. Thus, the acquisition of invasive and metastatic ability—
by virtue of increased ECM production—might invariably
lead to increased vulnerability to ER stress.
We have shown that EMT cells constitutively activate the
PERK branch of the UPR, which is required for them to invade,
metastasize, and form tumorspheres. The selective activation
by EMT cells of PERK–eIF2α–ATF4 signaling, but not the IRE1
branch of the UPR, raises the possibility that this branch may
Figure 7. EMT correlates with PERK but not IRE1 signaling in primary human tumors. A, two primary breast cancer cells (BT5104 and BT5094) were freshly collected from patient ascites, and cell lysates were collected and analyzed by Western blotting for differentiation markers. B, Western blot analysis of non-EMT–like BT5104 cells and EMT-like BT5094 cells treated with 0, 5, and 10 nmol/L of thapsigargin for 2 days for expression of PERK, BiP, phospho-eIF2α (p-eIF2α), and β-actin. C, dose–response curves of non-EMT–like BT5104 and EMT-like BT5094 cells treated with increasing concentrations of tunicamycin or thapsigargin for 3 days. Cell survival was determined using an ATP-based luminescence assay. Data are represented as mean +/− SD from three replicates. D, correlation analyses of expression of EMT genes and ECM genes, XBP1-targeted genes, or ATF4-targeted genes in breast cancers (GSE41998; n = 255), colon cancers (GSE37892; n = 130), gastric cancers (GSE26942; n = 90), lung cancers (GSE4573; n = 130), and metastatic cancers of various origins (GSE11360; n = 187).
–5 5 10 15 20–1
1
2
3
4
5
–5 0 5 10 15 20
2
4
6
8
10
–5 5 10 15 20
–2
2
4
6
–10 –5 5 10 15 20
–6
–4
–2
2
4
–10 –5 5 10 15 20
–5
5
10
15
–10 –5 5 10 15 20 10 20 30
–5 –5
5
10
15
5
10
15
5
10
15ρ = 0.90, P < 0.001
–10 10 20 30
–4
–2
2
4
6
–10 0 10 20 30
5
10
15
20
–10
ρ = 0.93, P < 0.001
–10 10 20 30
–5
5
10
–10 10 20 30
–5
5
10
15
–5 0 5 10 15 20 25
ρ = 0.87, P < 0.001 ρ = 0.92, P < 0.001 ρ = 0.84, P < 0.001
ρ = –0.01, P = 0.945ρ = 0.08, P = 0.690ρ = –0.18, P = 0.388ρ = –0.42, P = 0.02ρ = –0.36, P = 0.062
ρ = 0.80, P < 0.001 ρ = 0.87, P < 0.001 ρ = 0.88, P < 0.001 ρ = 0.68, P < 0.001 ρ = 0.75, P < 0.001
–5 5 10 15 20
–6
–4
–2
2
4
–5 5 10 15 20
–5
5
10
15
–5 5 10 15 20
–5
5
10
EMT vs. ECM
EMT vs. XBP1
EMT vs. ATF4
Breast cancer Colon cancer Gastric cancer Lung cancer Metastasis mix
2014;4:702-715. Published OnlineFirst April 4, 2014.Cancer Discovery Yu-xiong Feng, Ethan S. Sokol, Catherine A. Del Vecchio, et al. Sensitizes Cells to Endoplasmic Reticulum Stress