ros-induced autophagy in cancer cells assists in · pdf fileros-induced autophagy in cancer...

34
Citation Garg AD, Dudek AM, Ferreira GB, Verfaillie T, Vandenabeele P, Krysko DV, Mathieu C, Agostinis P, 2013 ROS-induced autophagy in cancer cells assists in evasion from determinants of immunogenic cell death Autophagy. 2013 Sep;9 Archived version Author manuscript: the content is identical to the content of the published paper, but without the final typesetting by the publisher Published version https://www.landesbioscience.com/journals/autophagy/2012AUTO0449R3.pdf Journal homepage https://www.landesbioscience.com/journals/autophagy. Author contact your email [email protected] your phone number + 32 (0)16 37 75 36 IR https://lirias.kuleuven.be/handle/123456789/415137 (article begins on next page)

Upload: truongbao

Post on 12-Feb-2018

220 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Citation Garg AD, Dudek AM, Ferreira GB, Verfaillie T, Vandenabeele P,

Krysko DV, Mathieu C, Agostinis P, 2013

ROS-induced autophagy in cancer cells assists in

evasion from determinants of immunogenic cell death

Autophagy. 2013 Sep;9

Archived version Author manuscript: the content is identical to the content of the published

paper, but without the final typesetting by the publisher

Published version https://www.landesbioscience.com/journals/autophagy/2012AUTO0449R3.pdf

Journal homepage https://www.landesbioscience.com/journals/autophagy.

Author contact your email [email protected]

your phone number + 32 (0)16 37 75 36

IR https://lirias.kuleuven.be/handle/123456789/415137

(article begins on next page)

Page 2: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

ROS-induced autophagy in cancer cells assists in evasion from

determinants of immunogenic cell death

Abhishek D. Garg1, Aleksandra M. Dudek1,5, Gabriela B. Ferreira2,5, Tom Verfaillie1, Peter

Vandenabeele3,4, Dmitri V. Krysko3,4, Chantal Mathieu2 and Patrizia Agostinis1,*

1Cell Death Research & Therapy Unit, Department of Cellular and Molecular Medicine, University of

Leuven (KU Leuven), Leuven, Belgium;

2Laboratory for Experimental Medicine and Endocrinology (LEGENDO), University of Leuven

(KULeuven), Leuven, Belgium;

3Molecular Signaling and Cell Death Unit, Department for Molecular Biomedical Research, VIB, Ghent, Belgium;

4Department of Biomedical Molecular Biology, Ghent University, Ghent, Belgium;

5These authors share second coauthorship;

*Corresponding author:

Patrizia Agostinis; Address: Department of Cellular and Molecular Medicine, Faculty of Medicine,

University of Leuven (KULeuven), Campus Gasthuisberg O&N1, Herestraat 49, Box 802, B-3000 Leuven,

Belgium; Tel.: +32 16 330650; Email: [email protected]

Keywords: autophagy, calreticulin, ATP, immunogenic cell death, T cells, cancer, photodynamic therapy

(PDT), dendritic cells.

Abbreviations: ATF4, activating transcription factor 4; ATP2A2, ATPase, Ca++ transporting, cardiac

muscle, slow twitch 2; BAK1, BCL2-antagonist/killer 1; BAX, BCL2-associated X protein; CALR,

1

Page 3: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

calreticulin; CD, cluster of differentiation; CNTR(s), control(s); DC(s), dendritic cell(s); Ecto, surface

exposed; EIF2A, eukaryotic translation initiation factor 2A; EIF2AK3, eukaryotic translation initiation

factor 2-alpha kinase 3; ER, endoplasmic reticulum; ERN1, endoplasmic reticulum to nucleus signaling 1;

GPX4, glutathione peroxidase; HLA-DR, major histocompatibility complex class II; HMGB1, high

mobility group box 1; HSP(s), heat shock protein(s); Hyp-PDT, hypericin-based photodynamic therapy;

ICD, immunogenic cell death; IFN, interferon; IL, interleukin; L-Hist, L-Histidine; LPS,

lipopolysaccharide; PDT, photodynamic therapy; Phox-ER stress, photo-oxidative endoplasmic reticulum

stress; PIK3CA, phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha; ROS, reactive

oxygen species; XBP1, X-box binding protein 1;

Abstract

Calreticulin surface exposure (ecto-CALR), ATP secretion, maturation of dendritic cells (DCs) and

stimulation of T cells are prerequisites for anticancer therapy-induced immunogenic cell death (ICD).

Recent evidence suggests that chemotherapy-induced autophagy may positively regulate ICD by favoring

ATP secretion. We have recently shown that reactive oxygen species (ROS)-based endoplasmic reticulum

(ER) stress triggered by Hypericin-mediated photodynamic therapy (Hyp-PDT) induces bona fide ICD.

However, whether Hyp-PDT-induced autophagy regulates ICD was not explored. Here we show that, in

contrast to the expectations, reducing autophagy (by ATG5 knockdown) in cancer cells did not alter ATP

secretion after Hyp-PDT. Autophagy-attenuated cancer cells displayed enhanced ecto-CALR induction

following Hyp-PDT, which strongly correlated with their inability to clear oxidatively damaged proteins.

Furthermore, autophagy-attenuation in Hyp-PDT-treated cancer cells increased their ability to induce DC

maturation, IL6 production and proliferation of CD4+ or CD8+ T cells, which was accompanied by IFNG

production. Thus, our study unravels a role for ROS-induced autophagy in weakening functional

interaction between dying cancer cells and the immune system thereby helping in evasion from ICD pre-

requisites or determinants.

2

Page 4: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Introduction

Most anticancer modalities kill cancer cells through ‘physiological’ apoptosis,1, 2 which in immunological

terms is tolerogenic in nature.2-5 However, it has recently emerged that certain modalities can kill the

cancer cells through an apoptotic cell death subroutine that is actively immunostimulatory, termed as

immunogenic apoptosis or cell death.6-8 Cancer cells undergoing immunogenic cell death have the ability

to stimulate and activate the innate immune cells like dendritic cells, which in turn “cross-prime” the

adaptive immune system for the antigens of the dying cancer cells, thereby leading to effective activation

of antitumor immunity.6-9 In molecular terms ICD is characterized by the rapid (1 to 4 h post-treatment)

surface exposure of calreticulin,7, 9, 10 early secretion of ATP9, 11-13 (both in the absence of plasma

membrane permeabilization) as well as late apoptotic (passive) release of proteins (i.e., following plasma

membrane permeabilization) like heat shock proteins (HSPs)9, 14 and high mobility group box 1;15 all of

which are potent immunogenic signals.2, 16-18 In immunological terms ICD is characterized by increased

maturation or stimulation of DCs accompanied by IL1B production, which is required for the subsequent

stimulation and proliferation of antitumorigenic, IFNG-producing, CD4+ or CD8+ T cells;11, 13 all of which

together are indispensible for ICD’s ability to induce potent antitumor immunity.2, 19

We recently characterized that hypericin-based photodynamic therapy19-21 causes induction of bona

fide immunogenic cancer cell death.9, 10, 17 Hypericin is an endoplasmic reticulum-associated

photosensitiser,14 which when activated by light, evokes the production of reactive oxygen species, photo-

oxidative (phox) stress-mediated loss-of-function of ATP2A2 (ATPase, Ca++ transporting, cardiac muscle,

slow twitch 2, also called sarcoplasmic/endoplasmic reticulum calcium ATPase 2), disruption of ER-Ca2+

homeostasis and induction of phox-ER stress.9, 22, 23 Phox-ER stress is characterized by the activation of the

EIF2AK3-EIF2A-ATF4 branch (EIF2AK3: eukaryotic translation initiation factor 2-alpha kinase 3, also

called PKR-like ER kinase; EIF2A: eukaryotic translation initiation factor 2A; and ATF4: activating

transcription factor 4) and ERN1-XBP1 branch (ERN1: endoplasmic reticulum to nucleus signaling 1, also

called inositol-requiring enzyme 1; XBP1: X-box binding protein 1) of the unfolded protein response;24 and

ultimately culminates into BAX (BCL2-associated X protein) and BAK1 (BCL2-antagonist/killer 1)-

dependent mitochondrial apoptosis.22, 23, 25 Hyp-PDT-induced ICD8, 16, 17 consists of the concomitant

preapoptotic emission (30 to 60 min post-PDT, before phosphatidylserine externalization)9, 17 of ATP and

CALR through an overlapping molecular trail primarily governed by ROS production, the ER stress-

associated kinase EIF2AK3, ER-to-Golgi anterograde transport and class I phosphoinositide-3-kinase

(phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha/PIK3CA) activity.8, 9, 17 This is

3

Page 5: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

followed by late apoptotic (passive) release of chaperones like HSPA1A/HSP70 and HSP90AA1/HSP90.9

Cancer cells undergoing Hyp-PDT-induced ICD cause efficient in vitro maturation of DCs (characterized

by surface upregulation of CD86 and HLA-DR i.e. major histocompatibility complex, class II, maturation

markers, amongst others),9, 26 accompanied by production of IL1B in the absence of IL10 and induction of

potent antitumor immunity, in vivo.9

In terms of stimulation of the main antigen presenting cell of the innate immune system, i.e. DCs;3,

7, 16 while cancer cell-associated macroautophagy (hereafter referred to as autophagy) is important for the

attraction of DCs,13 it has not been shown (to the best of our knowledge) how this affects the stimulation

and maturation of DCs. Similarly, the knowledge on how cancer cell-associated autophagy modulates the

clonal expansion or proliferation of T cells seldom exists. Rapid and extensive proliferation of T cells

following encounter with ‘danger’ or antigenic stimulation (along with proper costimulation)27 is one of the

keystone functions of adaptive immunity.28 This property is important for antitumor immunity29 or tumor

immunogenicity30 such that inhibition of T cell proliferation causes tumoral immune resistance.31

Interestingly, IFNG-production from T cells, which is crucial for the immunogenicity of cancer cells,11, 13

has been shown recently to be strongly reduced when autophagy was ablated in cancer cells undergoing

ICD.13 Moreover, autophagy also plays an important role in optimal ATP release from cancer cells

undergoing chemotherapy-induced ICD.13, 32

We have previously observed that Hyp-PDT-mediated phox-ER stress9, 22 in cancer cells causes

accumulation of oxidatively damaged proteins (characterized by the presence of protein carbonyls) along

with stimulation of the autophagic flux.33, 34 Autophagy contributes to the clearance of these oxidatively

damaged proteins,33 thereby mitigating the damage caused by Hyp-PDT-induced ROS at the ER.33, 34 This

is interesting, considering that the surface mobilization of CALR, which is a ROS-dependent process,9, 35

has been linked to the accumulation of damaged proteins.35-38 Since autophagy activated by Hyp-PDT

counteracts the accumulation of oxidatively damaged proteins,33 this raises the interesting possibility that

autophagy may affect Hyp-PDT-induced ecto-CALR. Moreover, it has been proposed recently that

autophagy might be crucial for Hyp-PDT-elicited ATP secretion as well,16 an avenue that needs further

attention. To this end, we wondered how autophagy in Hyp-PDT-treated cancer cells modulates various

characteristics of ICD like emission of (active) immunogenic signals e.g. ecto-CALR and secreted ATP,

maturation or stimulation of interacting DCs and stimulation or proliferation of CD4+ or CD8+ T cells.

4

Page 6: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

In the present study we found that, in contrast to the expectations, Hyp-PDT-elicited ATP secretion

was not regulated by autophagy. Similarly, the immune cells-based production of IL1B and IFNG in

response to cancer cells undergoing Hyp-PDT-induced ICD was not compromised due to cancer cell-

specific ablation of autophagy. On the other hand, ecto-CALR induction was regulated by autophagy in a

manner that strongly correlated with accumulation of oxidatively damaged proteins. Within the context of

cancer cells, autophagy acted as a suppressor of ecto-CALR induction following ICD triggered by Hyp-

PDT. Last but not least, knocking-down autophagy in Hyp-PDT treated cancer cells increased their ability

to induce DC maturation accompanied by IL6 production and increased the proliferation of CD4+ or CD8+

T cells; thus suggesting that ROS-based stimulation of autophagy may help (but not act as a sole mediator)

in evading ICD-based antitumor immunity.

Results

Glutathione peroxidase-4 overexpression or L-histidine pretreatment reduces Hyp-PDT-induced

CALR surface exposure and ATP secretion

Hyp-PDT was found to induce CALR surface exposure (ecto-CALR, exhibiting the characteristic “patchy”

pattern of surface distribution)10 (Fig. S1A) and ATP secretion (Fig. S1B) in the treated T24 cancer cells,

as early as 1 h after PDT treatment, under plasma membrane non-permeabilizing conditions (Fig. S1C),

which is in line with our previous reports.9, 10 Hyp-PDT-treated cancer cells secreted ATP in a dose-

dependent manner (Fig. S1B and Fig. S2A) and, in accordance with our earlier observations,9 this process

was paralleled by an increase in the intracellular ATP content in these early stages following Hyp-PDT

treatment (Fig. S2B). This early (1 h post-PDT), preapoptotic increase in intracellular ATP is probably a

result of increased ER-mitochondria coupling39 observed after Hyp-PDT.9, 22 Moreover, we also detected

late (24 h post-PDT) release of immunogenic signals like extracellular CALR, HSP90AA1 and HSPA1A in

the conditioned media (Fig. S2C). Curiously in this setup we did not observe, as strong a release of

extracellular HMGB1, as observed for the aforementioned immunogenic signals (Fig. S2C).

To understand the impact of the ‘oxidative module’ on emission of ecto-CALR and secreted ATP

following Hyp-PDT we decided to investigate their emission in T24 cancer cells following L-histidine (L-

Hist) pretreatment. L-histidine is a 1O2 quencher, which (as reported earlier by us) reduces the ROS-

dependent effects of Hyp-PDT.9, 23 Cell surface biotinylation followed by immunoblot analysis of the

5

Page 7: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

isolated plasma membrane proteins following treatment with Hyp-PDT showed that indeed, as expected,

pre-treatment with L-histidine reduced both ecto-CALR (Fig. 1A) and secreted ATP (Fig. 1B) in the

treated cancer cells. To further substantiate these observations, we decided to investigate the emission of

these immunogenic signals in HeLa human cancer cells overexpressing the antioxidant enzyme,

glutathione peroxidase 4 (GPX4). GPX4 is a selenoenzyme that removes phospholipid hydroperoxides

from cellular membranes,40 which we have shown previously to confer protection from Hyp-PDT-induced

apoptosis upon overexpression.23, 34 This analysis revealed that following Hyp-PDT and also in the basal

untreated conditions, cancer cells overexpressing GPX4 exhibited reduced levels of ecto-CALR as

compared to their wild-type counterparts (Fig. 1C). Similarly, GPX4 overexpression in cancer cells also

reduced the overall ATP secretion following Hyp-PDT (Fig. 1D). In order to rule out cancer cell type-to-

cell type and species-associated differences in the effects of GPX4 overexpression we repeated the above

analysis in murine L929 cancer cells overexpressing GPX4.34 In line with the results obtained in HeLa

cells, overexpression of GPX4 in L929 cells also suppressed the induction of ecto-CALR (both in Hyp-

PDT treated as well as basal conditions; Fig. 1E) and secretion of ATP (Fig. 1F). Interestingly, we

observed that the levels of ATP secretion following Hyp-PDT treatment were cancer cell type-dependent –

a point that has been more systematically analyzed in the next section.

Taken together with our previous observations,9, 22, 23 these results indicate that ROS signaling

generated by Hyp-PDT contributes to CALR surface exposure and regulates ATP secretion by the cancer

cells.

ATP secretion induced by Hyp-PDT is not autophagy-dependent

In our previous reports we had observed that Hyp-PDT-mediated activation of functional autophagy

mitigates the oxidative damage caused by this treatment to proteins and organelles.33, 34 Since, ecto-CALR

and secreted ATP were a consequence of Hyp-PDT mediated oxidative stress,9 we wondered whether

autophagy might affect their emission. To investigate this, we knocked down ATG5 protein (ATG5KD) via

RNA interference in two human cancer cell lines; the T24 human bladder cancer cells (Fig. 2A) and the

A375m human melanoma cells (Fig. 2B). This approach resulted in approximately 60% silencing of this

key proautophagic protein (Fig. 2A and 2B). In line with our previous reports,33, 34 ATG5-competent

cancer cells exhibited early (1 h post-PDT) activation of autophagy following Hyp-PDT; as revealed by

progressive conversion of LC3-I into its cleaved and lipidated form LC3-II by western blot (Fig. S3A and

6

Page 8: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

S3B) and the time-dependent accumulation of LC3-positive puncta, by immunofluorescence analysis (Fig.

S4).41 Consistent with our previous studies,33 partial suppression of ATG5 expression reduced autophagy

stimulation following Hyp-PDT in cancer cells as revealed by decreased LC3-conversion and formation of

LC3 puncta (Fig. S3 and S4). Moreover, as shown previously,33 attenuation of autophagy by the partial

reduction of ATG5 expression, also decreased the overall survival of the cancer cells after Hyp-PDT

(thereby substantiating its prosurvival role during Hyp-PDT),33 an effect that was more pronounced at

milder Hyp-PDT doses (Fig. S5).

Next, we analyzed the role of autophagy in ATP secretion following Hyp-PDT treatment.

Consistent with the previous studies published by us9 as well as others,42 this analysis revealed that the

levels of ATP secretion are cancer cell type-dependent; indeed, at the same recovery time (1 h post-PDT)

and Hyp-PDT doses [equal doses as measured by survival—(Fig. S5) high PDT dose for T24 cells (36.36%

± 2.7 s.d. survival) vs. higher dose for A375m cells (35.78% ± 6.71 s.d. survival), statistically non-

significant at P<0.05]; A375m cells secreted more ATP than T24 cells, even if the starting basal

extracellular ATP concentrations were very different for these two cell lines (Fig. 2C). However, the

similar levels of ATG5KD obtained in both cell lines (Fig. 2A and 2B), did not significantly alter the

corresponding levels of ATP secreted by these cancer cells following Hyp-PDT (Fig. 2D and 2E). This

effect was not due to a residual ATG5 expression observed upon the siRNA/shRNA-based knockdown

(Fig. 2A and 2B), since even the genetic deficiency of atg5 (in murine embryonic fibroblasts or MEFs i.e.

atg5-/- MEFs), which renders these cells autophagy-incompetent,33, 41 failed to alter the levels of ATP

secreted following Hyp-PDT treatment (Fig. 2F).

Thus, from the above observations we can conclude that, following Hyp-PDT autophagy is

dispensable for optimal ATP secretion.

Ecto-CALR induced by Hyp-PDT correlates with the accumulation of oxidatively damaged proteins

which is modulated by autophagy

We next analyzed the surface exposure of CALR following Hyp-PDT in autophagy-competent or

autophagy-compromised (ATG5KD) cancer cells. Hyp-PDT induced rapid ecto-CALR in the autophagy-

competent T24 and A375m cancer cells (Fig. 3A and 3B). Interestingly, ATG5KD cancer cells exhibited a

7

Page 9: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

further increase in ecto-CALR induction following Hyp-PDT (Fig. 3A and 3B). These observations reveal

that autophagy has the ability to downregulate ecto-CALR levels following Hyp-PDT.

It has been reported that ecto-CALR induction tends to correlate or associate with processes that

cause accumulation of damaged proteins.9, 37, 38, 43-45 On the other hand, autophagy is a catabolic process

that tends to remove damaged proteins like oxidized proteins and/or damaged organelles and reduce the

effects of stress on a cell.33, 41, 46 Since carbonylation of amino acid chains is a good indicator of the level of

irreversible oxidative damage of proteins and since such proteins are marked for degradation/removal;33, 47

we next investigated the pattern of early accumulation of carbonylated or oxidatively damaged proteins in

cells responding to Hyp-PDT. Hyp-PDT treatment resulted in an early (1 h post-PDT) accumulation of

carbonylated proteins in the T24 and A375m cancer cells (Fig. 3C and 3D). As anticipated based on

former studies41, 46 and our previous report,33 knockdown of ATG5 resulted in an increased accumulation

of carbonylated proteins in the cancer cells in response to both high (Fig. 3C and 3D) and medium (Fig.

S6A) Hyp-PDT doses. Thus, reducing the capacity of autophagy to clear oxidized or damaged proteins

following Hyp-PDT correlates with the increased surface exposure of CALR.

In previous studies33, 48 atg5-/- MEFs were found to be protected from oxidative stress and to display

a decreased accumulation of carbonylated/oxidized proteins as compared to their wild-type counterparts

(Atg5+/+ MEFs),33 due to the upregulation of compensatory mechanisms (like chaperone-mediated

autophagy)33, 48 in these untreated MEFs.33, 49 Indeed, early after Hyp-PDT (1 h) we observed that, while

Atg5+/+ MEFs exhibited the characteristic increase in carbonylated proteins yet atg5-/- MEFs displayed a

reduced accumulation of ROS-damaged (i.e. carbonylated) proteins (Fig. S6B). In order to further test

whether there was a correlation between ecto-CALR induction and accumulation of carbonylated proteins,

we decided to use this system for obtaining a proof-of-concept. Basically, if ecto-CALR induction

correlates with carbonylated protein accumulation than the atg5-/- MEFs would be expected to exhibit

reduced ecto-CALR. In line with this expectation, while Hyp-PDT efficiently induced ecto-CALR in

Atg5+/+ MEFs yet it failed to induce ecto-CALR in atg5-/- MEFs (Fig. S6C). Thus, autophagy-based

modulation of ecto-CALR following Hyp-PDT strongly correlates with the accumulation of oxidatively

damaged proteins.

8

Page 10: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Autophagy attenuation in Hyp-PDT treated cancer cells increases phenotypic maturation of

interacting DCs and IL6 production

Hyp-PDT induced ICD of cancer cells has been found to be associated with increased phenotypic and

functional maturation of dendritic cells interacting with them.9, 26 Increased maturation of DCs is one of the

important prerequisites for ICD.6, 7, 9 To this end, in view of the above results, we examined the impact of

autophagy attenuation in Hyp-PDT treated cancer cells on the maturation of interacting DCs.

Autophagy-competent or autophagy-compromised A375m cancer cells were treated with Hyp-PDT

and these stressed/dying cells (autophagy-competent A375m, 35.78% ± 6.71 s.d. survival; autophagy-

compromised A375m, 29.21% ± 3.45 s.d. survival; Fig. S5) were used to pulse human immature DCs (hu-

iDCs) followed by analysis of the maturation profile of the coincubated hu-iDCs. In comparison to the

untreated human cancer cells (CNTR), Hyp-PDT treated autophagy-competent cancer cells induced

significant phenotypic maturation of hu-iDCs, as indicated by increased positivity for HLA-DR and

costimulatory CD86 molecules (Fig. 4A). Lipopolysaccharide (LPS), a known pathogen-associated

molecular pattern and a positive control in this study, induced strong phenotypic maturation of hu-iDCs, as

expected (Fig. 4A), thereby proving their competence. Interestingly, the Hyp-PDT treated ATG5KD cancer

cells caused a statistically significant increase in phenotypic maturation of the hu-iDCs when compared to

their autophagy/ATG5-competent counterparts (Fig. 4A). This shows that, autophagy attenuation in cancer

cells treated with Hyp-PDT increases the phenotypic maturation of the interacting DCs.

Based on these interesting observations, we envisaged that it was crucial to gain further insights

into the effect of blockage of cancer cell-associated autophagy on the functional status of these

phenotypically mature DCs. To this end, we evaluated the conditioned media of A375m-DC cocultures for

the pattern of certain important cytokines like IL1B (a proinflammatory cytokine, found to be important for

ICD),11 IL10 (a known immunosuppressive cytokine)9, 50 and IL6 (an important proinflammatory

cytokine).43 On the one hand, in line with expectations, LPS strongly elicited the hu-iDCs to produce all of

these three cytokines (Fig. 4B-D, Table S1). The exposure of hu-iDCs to Hyp-PDT treated, autophagy-

competent cancer cells did not induce production of any IL10 (Fig. 4B, Table S1) but led to production of

considerable amounts of IL1B (Fig. 4C, Table S1) and IL6 (Fig. 4D, Table S1). On the other hand, while

the exposure of hu-iDCs to Hyp-PDT treated, autophagy-compromised (ATG5KD) cancer cells did not alter

IL10 (Fig. 4B, Table S1) and IL1B (Fig. 4C, Table S1) production patterns observed in comparison to

their autophagy-competent counterparts; yet this cancer cell-specific ablation of autophagy led to a

9

Page 11: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

comparatively higher production of IL6 cytokine (Fig. 4D, Table S1), not only in Hyp-PDT-treated

conditions but interestingly also in the untreated conditions. Overall, this shows that following Hyp-PDT

treatment and exposure to hu-iDCs, the autophagy-ablated cancer cells tend to create a relatively more

proinflammatory cytokine pattern when compared to the autophagy-competent cancer cells.

Last, considering that autophagy attenuation in Hyp-PDT treated cancer cells increased ecto-CALR

as well as phenotypic maturation of the interacting DCs; we wondered whether this increase in DC

phenotypic maturation was attributable to the increase in ecto-CALR. To verify this conjecture, we decided

to carry out hu-iDC phenotypic maturation analysis in the presence or absence of anti-CALR antibody; a

strategy that we have previously utilized to reduce CALR-dependent phagocytosis of Hyp-PDT treated

cancer cells.10 To this end, we found that while autophagy attenuation in Hyp-PDT treated cancer cells

increased the phenotypic maturation of hu-iDCs (Fig. 4E) as observed previously (Fig. 4A) yet antibody-

based blocking of CALR failed to significantly reduce hu-iDC phenotypic maturation in response to both

Hyp-PDT treated autophagy-competent as well as autophagy-compromised cancer cells (Fig. 4E). This

suggests that the increase in DC phenotypic maturation observed following interaction with autophagy-

compromised cancer cells treated with Hyp-PDT is probably not ecto-CALR dependent.

Autophagy attenuation in Hyp-PDT treated cancer cells increases CD4+ or CD8+ T cell expansion in

the presence of IFNG production

Beyond stimulation of the DCs, the subsequent stimulation of T cells is also crucial for antitumor

immunity.27, 29, 30 To this end, we decided to investigate the effect of autophagy-ablation in cancer cells

undergoing Hyp-PDT elicited ICD on T-cell immunity. Autophagy-competent or autophagy-ablated

A375m cancer cells were treated with Hyp-PDT and these were fed to the hu-iDCs for 24 h. Subsequently,

these pulsed DCs were exposed to T cells followed by analysis of activation based on their increased

proliferation. As compared to the untreated human cancer cells (CNTR), Hyp-PDT treated autophagy

competent cancer cells induced statistically significant increase in proliferation of both human CD4+ (Fig.

5A) and human CD8+ T cells (Fig. 5B), as indicated by the increased number of proliferating CD4+/CD3+

and CD8+/CD3+ double-positive T cells respectively (as measured by the eFluor®670 dye dilution). LPS-

treated hu-iDCs, as per expectations, induced strong increase in the proliferation of both CD4+ (Fig. 5A)

and CD8+ T cells (Fig. 5B). Remarkably however, autophagy-ablation (ATG5KD) in cancer cells in absence

of any treatment, was on its own robust enough to cause a significant increase in the proliferation of both

10

Page 12: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

CD4+ and CD8+ T cells, to a degree that even exceeded that observed for Hyp-PDT treated autophagy

competent cells (Fig. 5A and 5B). Interestingly, Hyp-PDT treatment of these ATG5KD cancer cells caused

a further, statistically significant, increase in the proliferation of both CD4+ (Fig. 5A) and CD8+ T cells

(Fig. 5B), when compared to the CD4+/CD8+ T cell proliferations observed in response to untreated,

ATG5KD counterparts (Fig. 5A and 5B). Thus, the combination of autophagy-ablation in cancer cells and

Hyp-PDT treatment was still superior to autophagy-ablation alone in causing CD4+ or CD8+ T cell

expansions (Fig. 5A and 5B). This clearly shows that autophagy attenuation in cancer cells (alone or in

combination with Hyp-PDT treatment) instigates superior stimulation of T cells, leading to a significantly

increased expansion of the CD4+ or CD8+ T cells. Overall these observations support a role for cancer cell-

associated autophagy in suppressing robust T cell proliferation and expansion.

On the other hand, it is important to note that a mere increase in T cell expansion in the absence of

the crucial antitumorigenic cytokine IFNG, may not sustain an antitumor effector function for these

proliferating T cells.51 The presence of IFNG has been associated with cytostatic/cytotoxic T cell function

and antitumor effects like detection/elimination of tumor cells, robust antitumor effects instigated by ICD

as well as enhancement of tumor cell immunogenicity.52 To this end, we evaluated the conditioned media

of A375m-DC-T cell cocultures for the production of IFNG. As expected, we observed that exposure of

LPS-treated hu-iDC to naive T cells (i.e. purified CD3+ T cells) elicited the production of IFNG (Fig. 5C).

Moreover, we found that the coincubation of Hyp-PDT treated autophagy-competent cancer cells along

with DCs and T cells, led to considerably more production of IFNG in comparison to cocultures consisting

of corresponding untreated cancer cells (CNTR) or untreated cancer cells in the presence of T cells only

(Non-treated) (Fig. 5C). Although Hyp-PDT treated autophagy-ablated (ATG5KD) cancer cells did not

increase IFNG production in these coculture settings to a significant level, they clearly did not repress

IFNG production by T cells (Fig. 5C). Intriguingly, this observation is in sharp contrast to the reported

suppressive effect of autophagy on INFG production following chemotherapy-induced ICD.13

In conclusion, these observations suggest a role for cancer cell-associated autophagy in suppressing

efficient T cell immunity by decreasing the expansion of interacting CD4+ or CD8+ T cells that are

producing IFNG to a certain extent following Hyp-PDT elicited ICD.

11

Page 13: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Discussion

In the current study, we found that autophagy induced in response to Hyp-PDT treatment in cancer cells

regulated ecto-CALR induction, maturation of interacting DCs (and IL6 production) and T cell

proliferation without significantly affecting the levels of secreted ATP. Moreover, the autophagy-based

modulation of ecto-CALR was found to correlate with the autophagy-based modulation of accumulation of

oxidized/carbonylated proteins following Hyp-PDT. Considering that ecto-CALR, secreted ATP,

maturation of interacting DCs and robust T cell immunity are some of the vital pre-requisites and

established molecular determinants of ICD,7, 9, 13, 16, 18 these results provide unanticipated new insights

about the role of autophagy in ICD.

Surface exposure of CALR during ICD depends upon an apical ‘oxidative module’8 such that

antioxidants have been shown to reduce it,7, 35 an observation that we had previously extended to Hyp-

PDT-induced ecto-CALR.9 In this study, we further substantiate these reports by showing that

overexpression of the antioxidant enzyme, GPX4, or pretreatment with 1O2 scavenger, L-histidine, also

reduces Hyp-PDT induced ecto-CALR. On the other hand, the importance of the ‘oxidative module’ was

still unresolved in the case of Hyp-PDT-induced ATP secretion. The observation of GPX4 overexpression

or L-histidine pretreatment reducing ATP secretion following Hyp-PDT suggests that the ‘oxidative

module’ is the most apical event able to modulate, at least in part, the emission of this immunogenic signal

as well. Interestingly, we also observed that GPX4 overexpression reduced not only Hyp-PDT induced

ecto-CALR but also basal ecto-CALR in untreated conditions. We have previously shown that on various

levels9, 10 an overlap (rather than independence) exists between the ecto-CALR trafficking pathways in

basal, untreated conditions and ICD-inducing conditions. For instance, we had reported that, absence or

ablation of EIF2AK3, PIK3CA and BAX and/or BAK1, reduced both basal as well as chemotherapy or

Hyp-PDT induced ecto-CALR.9 To this end, the above observations about the effect of stable GPX4

overexpression on ecto-CALR imply that the ‘oxidative module’ might be another level on which the basal

and Hyp-PDT-induced ecto-CALR trafficking pathways overlap. In case of basal, untreated conditions,

variable amounts of intracellular ROS is generated due to the artificial in vitro cell culture conditions,53 and

this probably is responsible for the observed basal ecto-CALR induction.

Autophagy promotes survival following Hyp-PDT treatment (especially at mild PDT doses) by

taking care of the damage (to proteins or organelles) caused by the oxidative stress exerted by this

anticancer treatment modality.33, 34 In view of our recent findings that Hyp-PDT was in fact capable of

12

Page 14: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

inducing bona fide ICD9, 17 and the recent findings by Michaud et al. that chemotherapy-induced ATP

secretion was autophagy-dependent;13, 32 it was imperative to ascertain the impact of autophagy on the

emission of ecto-CALR and secreted ATP after Hyp-PDT. However, contrary to the chemotherapy-induced

ATP secretion,13 we observed that Hyp-PDT-induced ATP secretion was independent of autophagy.

Autophagy attenuation (via ATG5KD in cancer cells) or deficiency (in atg5-/- MEFs) had no significant

effect on Hyp-PDT-induced ATP secretion. Taken altogether, these results suggest that the ability of

autophagy to regulate ATP secretion might vary from one immunogenic cancer cell death inducer to

another.19 This might be due to the overall difference in the pathways engaged for ATP secretion by the

respective agents. For instance, while in the case of Hyp-PDT, ATP is secreted very early (30 min to 1 h

post-treatment) via the ‘classical’ secretory pathway9, 17 yet in the case of chemotherapy, ATP is secreted

relatively late (18 h post-treatment) via an as-yet-less-characterized autophagy-dependent pathway.13, 54

On the other hand, interestingly, the presence of Hyp-PDT induced ecto-CALR varied depending

on whether autophagy was attenuated or absent in a manner that correlated with the accumulation of

oxidized/carbonylated proteins after Hyp-PDT. More specifically, while autophagy attenuation was found

to increase both ecto-CALR and accumulation of oxidized/carbonylated proteins, yet autophagy deficiency

(in atg5-/- MEFs), which correlated with a decreased amount of oxidized proteins (probably due to the

constitutive compensatory upregulation of chaperone-mediated autophagy)33, 48 was found to decrease both.

Thus, in the context of a predominantly ROS-based anticancer treatment, like PDT, these observations

indicate that autophagy might play a role in suppressing ecto-CALR, a major ‘eat me’ signal.

Whether there is a direct link between the mobilization of ecto-CALR from the ER and increased

amount of oxidized proteins, at this subcellular site, is a matter that needs to be investigated in the near

future. A direct link is plausible because accumulation of damaged or unfolded proteins has been known to

overwhelm the retention system for chaperones like CALR thereby causing its ‘escape’ towards the

surface.36-38, 43 Moreover, this correlation is further supported by the observations that while blocking

caspase signaling delays the execution of apoptosis23 without altering ecto-CALR induction9 following

Hyp-PDT-based phox-ER stress22 yet, interference with autophagy (via ATG5-silencing) increases

accumulation of oxidized proteins,33 ecto-CALR and cell death,33 likely by an amplification of ROS-

mediated ER stress (A. D. Garg and P. Agostinis, unpublished results). Last but not least, we recently

found that Hyp-PDT induced an initial reticulophagy in the ER-photodamaged cells, which we revealed by

the increase in the colocalization of CALR with the LC3-decorated autophagosomes.31 Thus all together,

13

Page 15: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

these observations suggest that stimulation of autophagy after phox-ER stress interferes with the trafficking

of CALR to the surface, by directly dampening the consequences of phox-ER stress.

Interestingly, autophagy attenuation in Hyp-PDT-treated cancer cells increased the phenotypic

maturation of interacting DCs. Moreover comparison of the cytokine profiles produced by the interaction

of DCs with autophagy-ablated cancer cells and the autophagy-competent cancer cells hinted at a small

overall shift towards a more proinflammatory profile (i.e., a significantly increased IL6 production) when

cancer cell-associated autophagy is ablated. Overall, this means that autophagy stimulated in ROS-

damaged cancer cells might play an important role in suppressing the stimulation of DCs. Of note, we

observed that the increased ecto-CALR caused by autophagy attenuation was not responsible for the

improved DC maturation, since CALR-blocking antibodies10 failed to reduce it. This raises the attractive

possibility that autophagy regulates (within the Hyp-PDT treated cancer cells) the “emission” of other

unknown DC maturation-inducing factors. Here, HMGB13 and its recently found association with

autophagy-based secretion (induced by epidermal growth factor receptor-targeted diphtheria toxin),55

makes it a possible candidate. However, the fact that HMGB1 is susceptible to ROS-based inactivation56-58

and that we did not detect a significant HMGB1 release from Hyp-PDT treated cancer cells in conditions

where the release of other immune-modulatory molecules like HSPs was strongly detected,9 makes

HMGB1 an unlikely candidate in this ICD paradigm.

DCs can provide the proper stimulatory and costimulatory signals required for the activation of

naive T cells.27 Such activation leads to increased expansion or proliferation of CD4+ or CD8+ T cells29

accompanied by IFNG production,27 processes that have been found to be important for antitumor

immunity, ICD and tumor immunogenicity.11, 13, 29, 30 Interestingly it was recently reported that in the case

of chemotherapy-induced ICD, blockage of autophagy can cause ablation of CD4+ or CD8+ T cell

immunity.13 In case of Hyp-PDT induced ICD instead, autophagy attenuation in cancer cells caused

superior stimulation of T cells, leading to a significantly increased expansion of the CD4+ or CD8+ T cells.

Moreover, mitigation of autophagy in cancer cells did not hamper the production of IFNG. This

observation is in direct contrast to previous observations where autophagy-ablation was shown to decrease

IFNG production in response to cancer cells treated with chemotherapeutic ICD inducers.13 These

observations further substantiate the possibility that in cancer cells treated with Hyp-PDT, autophagy might

be suppressing the emission or presence of as-yet-unknown factors that ultimately help in mediating

processes that are crucial for the efficacy of antitumor immunity. This creates a vital need to characterize in

14

Page 16: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

the near future, the identity of these antitumor immunity promoting factors that are suppressed by cancer

cell-associated autophagy.

Interestingly, we observed that the ablation of autophagy in untreated cancer cells (in the presence

of DCs), was on its own capable of significantly increasing IL6 production and the proliferation or

expansion of the CD4+ or CD8+ T cells (accompanied by IFNG production). In line with this, the ablation

of autophagy by deleting Rb1cc1 (RB1-inducible coiled-coil 1; an essential autophagy gene, also

sometimes called FAK family kinase-interacting protein 200 kDa) in a mouse model of breast cancer has

been reported to result in accentuation of antitumor immunity mediated by the increased number of IFNG+

CD8+ T cells.59 Moreover, in another recent study, the ablation of hypoxia-induced autophagy in cancer

cells was reported to result in accentuation of cytotoxic T cell responses thereby meaning that hypoxia-

induced autophagy can help evade T cells-based immunity.60 Overall our observations along with the

reported ones imply a role for cancer cell-associated autophagy in general evasion from cancer

immunosurvelliance and antitumor immunity; a conjecture that needs to be thoroughly investigated and

confirmed in future studies.

In conclusion, this study shows that in cancer cells treated with Hyp-PDT, ATP secretion is partly

ROS-dependent but autophagy-independent while ecto-CALR induction is downregulated by autophagy

(Fig. 5D). We had shown previously that after Hyp-PDT treatment of cancer cells, ecto-CALR and secreted

ATP follow overlapping molecular pathways to reach the extracellular space.9, 10, 17 The observations made

in this study imply that this overlapping molecular trail detaches on the level of autophagy. Moreover, we

found that cancer cell-associated autophagy after Hyp-PDT suppresses the maturation of the interacting

DCs, IL6 production and proliferation of CD4+ or CD8+ T cells (Fig. 5D). Considering that ecto-CALR,

maturation of interacting DCs and stimulation of T cells are important determinants of ICD in cancer

cells;6, 7, 9 our study suggests that cancer cell-associated autophagy in the case of Hyp-PDT treatment might

be helping in evasion from molecular and immunological determinants of ICD.

Materials and Methods

Reagents

Hypericin was prepared, purified and stored as described previously.9, 23 Antibody against LC3

(microtubule-associated protein light-chain 3) and HSP90AA1 were purchased from Cell Signaling

15

Page 17: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Technology (3868 and 4874, respectively). Anti-calreticulin antibody was purchased from Stressgen (SPA-

600), unless mentioned otherwise. Anti-FAS (N-18, M-20 and C-20) and anti-HSPA1A antibodies were

purchased from Santa Cruz Biotech (sc-714, sc-716, sc-715, and sc-33575, respectively). Anti-ACTB/β-

actin antibody and L-histidine were purchased from Sigma (A5316 and 53319, respectively). Anti-ATG5

and anti-HMGB1 antibodies were purchased from Abcam (ab109490 and ab65003). Secondary antibodies

conjugated to horseradish peroxidase were purchased from Cell Signaling Technology (7076 and 7074).

Cell lines, LDH assay, MTS assay and induction of immunogenic cell death

HeLa, T24 and MEF cells were cultured at 37°C under 5% CO2 in DMEM (containing 0.11 g/L sodium

pyruvate and 4.5 g/L glucose) supplemented with 2 mM glutamine, 100 units/ml penicillin, 100 µg/L

streptomycin and 10% fetal bovine serum or FBS (referred to as normal culture medium hereafter). A375m

cells stably expressing ATG5-shRNA or CO-shRNA (CO = vector expressing scrambled shRNA) were

received from the lab of Prof. Peter Carmeliet (Vesalius Research Centre, University of Leuven, Belgium).

They were maintained in normal culture medium containing 4 mM glutamine and 3 mg/ml puromycin

(Invivogen, ant-pr-1). L929 cells were maintained in normal culture medium containing 0.5 mg/ml

geneticin and 43 nM NaSe. Here, the atg5-/- MEF cells were received from the lab of Prof. N. Mizushima

(Tokyo Medical and Dental University, Tokyo, Japan). LDH assay was carried out (wherever applicable)

for estimation of plasma membrane permeabilization by estimating the amounts of LDH released into the

serum-free conditioned media using LDH Cytotoxicity Detection Kit (Clontech, 630117). Percent survival

or cytotoxic effects were measured using the MTS assay (Promega, G3582), as per the manufacturer’s

instructions.

ICD was induced by hypericin-based PDT (Hyp-PDT). Here, T24 and A375m cells were incubated

with 150 nM Hypericin (for 16 h in normal culture medium), HeLa cells were incubated with 125 nM

Hypericin (for 16 h in normal culture medium) while MEF and L929 cells were incubated with 200 nM

Hypericin (for 2 h in serum-free culture medium). Irradiation was performed as described previously.9, 10

Four Hyp-PDT doses, based on the light fluence, were defined as low PDT dose (fluence - 0.54 J/cm2),

medium PDT dose (1.35 J/cm2), high PDT dose (2.16 J/cm2) and higher PDT dose (2.43 J/cm2). Cells

loaded with Hypericin were handled in either dark or subdued light (<1 µW/cm2). All the untreated or

control (CNTR) sample conditions had Hypericin in absence of irradiation (i.e. absence of PDT

phenomenon).

16

Page 18: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Transfection of siRNA

Cells were transfected by adding 500 µL serum-free culture media with 10 µL Dharmafect (Thermo Fisher

Scientific, T-2001) and 5 µL ATG5 or scrambled (SCR) siRNA (20 µM, Dharmacon, Thermo Fisher

Scientific; L-004374-00-0005) to culture dishes with 2 mL serum-free culture media (final concentration of

40 nM siRNA). 3 h following transfection, 2.5 mL culture media containing 5% FBS was added.

Experiments were carried out 48 h after transfection.

Analysis of human dendritic cell (DC) maturation and cytokines

Human immature dendritic cells (hu-iDC) were isolated and cultured as described previously61 followed by

DC maturation analysis as described previously.9 Briefly, respective A375m cells exposed to higher Hyp-

PDT treatment (and recovered 24 h post-PDT) were coincubated with hu-iDCs at a ratio of 1:20 (hu-

iDCs:A375m) for 24 h. Hu-iDCs stimulated with 100 ng/ml of Escherichia coli-derived lipopolysaccharide

(LPS) for 24 h (Sigma, L3012) were used as positive controls. In certain experiments, the A375m cells

were coated either with 1.5 µl/106 cells of isotype control (Chicken IgY; Abcam, ab37382) or with chicken

anti-CALR antibody (Abcam, ab14234) for 30 min at 37°C (as described previously10) before carrying out

coincubation with hu-iDCs as described above.

Following this, the cells were detached, washed and stained with FITC-conjugated anti-HLA-DR

(Invitrogen, MHLDR01) and APC-conjugated anti-CD86 (Invitrogen, MHCD8605) antibodies according

to the manufacturer’s instructions. Anti-human IgG1 (Beckton & Dickinson, 555786) were used as isotype

controls. After the antibody staining, the cells were washed and analyzed with an Attune Acoustic

Focusing Flow Cytometer (Applied Biosystems, 4469120AS, Life Technologies Europe B.V., Gent,

Belgium) or FACSCalibur flow cytometer (Beckton & Dickinson, 342975, BD Biosciences, San Jose,

California, USA). The conditioned coculture media derived from these hu-iDCs and A375m cell

coincubation experiments were collected and checked for immunoreactive levels of human IL10, IL1B and

IL6 cytokines (in quadruplicates, i.e., two intra-assay replicates each of two inter-assay replicates)

measured using the Fluorokine multiplex assay kits (R&D Systems, LUH217, LUH201 and LUH206,

respectively) as per the manufacturer’s instructions. Data analysis with 5-parametric-curve fitting was

performed using the Bio-Plex Manager software on a Bio-Plex 200 workstation (Bio-Rad, 171-000201).

17

Page 19: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Analysis of T cell proliferation and IFNG production

The hu-iDCs were isolated and cultured as described above. Thereafter, the respective A375m cells

exposed to higher Hyp-PDT treatment (and recovered at 24 h post-PDT) were coincubated with hu-iDCs at

a ratio of 1:1 for 24 h (in 96 well U-bottomed plates). Hu-iDCs stimulated with 100 ng/ml of Escherichia

coli-derived lipopolysaccharide (LPS) for 24 h (Sigma, L3012) were used as positive controls.

Simultaneously, purified CD3+ T-cells were prepared from peripheral blood mononuclear cells

(PBMCs) of healthy allogeneic donors by negative selection using the Pan T Cell Isolation Kit II according

to the manufacturer’s recommendations (Miltenyi Biotec, 130-095-130). Purity was 90 to 95%, as assessed

by flow cytometry. To assess cell proliferation in vitro, T cells were labeled with eFluor®670 Proliferation

Dye (eBioscience, 65-0840-85). Fluorescent dye-labeled lymphocytes were cultured in the presence of

DCs/A375m cells at a DC:T cell ratio of 1:50 in RPMI1640 supplemented with 10% FCS, 10 mM HEPES

and 1 mM glutamine, in a total volume of 200 µL in 96 well U-bottomed plates. Cells were cultured for 5

days before antibody staining and analysis by flow cytometry. IL2 (Peprotech, 200-02) was added at day 2

of coculture, at a 25 U/mL concentration. Antibodies (conjugated with fluorochromes) used to delineate

lymphocyte populations by flow cytometry included anti-CD3-eFluor®450, anti-CD4-FITC and anti-CD8-

PE-Cy7 (eBioscience, 48-0038, 11-0049 and 25-0049 respectively). Dead cells were excluded by using the

Fixable Live/Dead Yellow stain according to the manufacturer’s specifications (Invitrogen, L34959). Data

acquisition was performed on a GalliosTM flow cytometer (Beckman Coulter, A94299, Analis SA, Suarlee,

Belgium) and the KaluzaTM software (Beckma Coulter) was used for data analysis.

The conditioned coculture media derived from these DC/A375m/T cell coincubation experiments

were collected and checked for immunoreactive levels of human IFNG measured using the human IFNG

FlowCytomix Simplex kit (eBioscience, BMS8228FF). The samples were prepared according to the

manufacturer’s instructions and analyzed on a GalliosTM flow cytometer (Beckman Coulter, A94299) and

the FlowCytomixTM software (eBioscience) was used for data analysis.

18

Page 20: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

ATP assays

Following Hyp-PDT treatments, extracellular ATP was measured in the conditioned (serum-free) media

and intracellular ATP was determined after saponin-based lysis via an ATP assay mix (Sigma, FLAAM-

1VL) based on luciferin-luciferase conversion, following the manufacturer’s instructions. Bioluminescence

was assessed by optical top reading via a FlexStation 3 microplate reader (Molecular Devices Inc., 0310-

5627, Sunnyvale, California, USA).

Protein carbonylation detection assay

The detection of protein carbonyl derivates was carried out as described previously.62 Briefly, the cells

were either treated with Hyp-PDT or left untreated, followed by preparation of cell lysates. Thereafter, 50

µL of cell lysates corresponding to a total protein concentration of between 5 to 10 mg/ml were reacted

with an equal volume of 200 µM fluorescein-5-thiosemicarbazide (FTC) (Sigma, 46985-100MG-F)

overnight in the dark at room temperature. The following day, proteins were precipitated by adding 400 µL

of cold 20% trichloroacetic acid and incubating for 10 min on ice. This was followed by centrifugation,

discarding of supernatant and washing of the protein precipitate three times with acetone. Finally, the

acetone supernatant was carefully discarded and the precipitates were allowed to be air dried. The

desiccated precipitates were then solubilized with 50 µL of 6 M guanidine hydrochloride (GuHCl) and

diluted by adding 450 µL of 0.1 M NaH2PO4 (pH 7.0). Consequently, the protein concentrations in these

samples were measured via BCA assay. 50 µL of these samples were aliquoted in a transparent-bottom

black microtiter plate and the fluorescence was measured via a FlexStation 3 microplate reader (Molecular

Devices Inc., 0310-5627) with excitation at 485 nm and emission at 535 nm. The nanomoles of FTC-

reacted carbonyls were calculated using a standard curve that was generated from the readings of various

concentrations of FTC. These values were then divided by the respective protein precipitate concentration

values to derive the amount of protein carbonyls expressed as nmol/mg protein. Thereafter, a fold-change

of these values with respect to the corresponding controls was taken.

19

Page 21: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Biotinylation of cell surface proteins

Biotinylation of cell surface proteins was performed as described in our previous studies.9, 10 Time points

used for this analysis were confirmed to be associated with non-significant plasma membrane

permeabilization via LDH release assay as described previously.9, 10 For control samples (i.e. untreated

cells), one set of samples were exposed to the buffer containing biotin (denoted by ‘+BIO’) while the other

set was exposed to only the buffer as a negative control (denoted by ‘-BIO’). The capture of plasma

membrane proteins was confirmed by detecting FAS in the biotinylated protein fractions. In all the figures

(as applicable), ‘Plasma membrane proteins’ are proteins in the biotinylated surface fraction while

‘intracellular proteins’ are the unbound proteins that are non-biotinylated.

Immunoblotting

Preparation of cell lysates, determination of protein concentration and immunoblotting was done as

described.63 Wherever applicable, the bands of proteins on the blots were quantified through the ImageJ

software for determination of the relative integrated band density.

Fluorescence detection of cell surface CALR and intracellular LC3

1 h following Hyp-PDT treatment, T24 cells (on a glass slide) were incubated with 1 µM Sytox Green

(Invitrogen, S-34860) exclusion dye for 15 min. Here, T24 cells that were permeabilized with 0.1%

Saponin were included as positive control for cells with compromised plasma membrane. Here, the

principle was that, if a particular cell has a permeabilized plasma membrane then its nucleus would stain

green (due to Sytox Green) and in the presence of DAPI counter-stain the overall nucleus of this

permeabilized cell would show up as cyan. Thereafter, the cells were fixed (with 4% paraformaldehyde),

incubated with 0.1 M glycine in PBS and immunostained with rabbit anti-CALR antibody (1:25) (Cell

Signaling Technology, 2891S) (at 4°C for 1 h). This was followed by washing, incubation with the anti-

rabbit IgG antibody conjugated with Alexa fluor 680 (1:500) (Invitrogen, A-21109) and counterstaining

with DAPI (1 µg/ml for 10 min) (Invitrogen, D1306). Alternatively, at various time-points post-PDT,

A375m cells (on a glass slide) were fixed (with 4% paraformaldehyde), permeabilized with 0.1% Triton X-

100, incubated with 0.1 M glycine in PBS, blocked and immunostained with rabbit anti-LC3 antibody

20

Page 22: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

(1:200) (Cell Signaling Technology, 3868) (at room temperature for 1 h). This was followed by washing,

incubation with the anti-rabbit IgG antibody conjugated with Alexa fluor 488 (1:400) (Invitrogen, A-

11008) and counterstaining with DAPI (1 µg/ml for 10 min) (Invitrogen, D1306). Thereafter the respective

samples were mounted using the Prolong Gold antifade reagent (Invitrogen, P36930). Confocal

fluorescence images of these samples were obtained using a Nikon A1R confocal unit mounted on a

Ti2000 inverted microscope controlled by NIS elements acquisition software (Nikon Instruments Inc.,

Melville, NY, USA).

Statistical analysis

Data are presented either in fold changes, absolute reading units or in percentages with mean ± s.d. or mean

± s.e.m. indicated in figure legends. All statistical analyses were performed using either Prism software

(GraphPad Software, USA) or GraphPad QuickCalcs online software

(http://www.graphpad.com/quickcalcs/index.cfm). Student’s t-test was used for statistical analysis with

significance level set at P < 0.05. Wherever applicable (with respect to comparison between multiple data

sets), a one-way ANOVA analysis with Kruskal-Wallis test was performed.

Acknowledgements:

We thank Dr. Peter Carmeliet (Vesalius Research Center, VIB, Leuven, Belgium) for the A375m cells

stably expressing ATG5-shRNA or CO-shRNA and Dr. Johan Grooten (Ghent University, Belgium) for the

L929 cells stably expressing the empty vector (Neo) or the GPX4 overexpression vector. This work was

supported by grants from the Fund for Scientific Research Flanders (FWO-Vlaanderen; G.0661.09,

G.0728.10 and G.0584.12N) and KU Leuven (GOA/11/009) to P.A.; A.D.G is a postdoctoral fellow

supported by the BOF Postdoctoral Mandate (PDM) from KU Leuven (PDMK/12/146). G.B.F. is

supported by a postdoctoral fellowship (FWO) and C.M. is supported by a clinical fellowship (FWO). This

paper presents research results of the IAP7/32, funded by the Interuniversity Attraction Poles Programme,

initiated by the Belgian State, Science Policy Office. We thank Jan Piessens for the technical support. This

work is supported by grants from FWO-Vlaanderen (G.0728.10 and 31507110 to DVK). DVK is a

postdoctoral fellow paid by fellowship from FWO-Vlaanderen. We acknowledge 'The Hercules

Foundation' for the Nikon A1R confocal microscope (AKUL/09/037 to Dr. Wim Annaert). Some of the

21

Page 23: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

figures were produced using Servier Medical Art (www.servier.com), for which the authors would like to

acknowledge Servier.

References:

1. Garg AD, Bose M, Ahmed MI, Bonass WA, Wood SR. In vitro studies on erythrosine-based photodynamic therapy of malignant and pre-malignant oral epithelial cells. PLoS One 2012; 7:e34475. 2. Garg AD, Martin S, Golab J, Agostinis P. Danger signalling during cancer cell death: origins, plasticity and regulation. Cell Death Differ 2013; DOI: 10.1038/cdd.2013.48. 3. Garg AD, Nowis D, Golab J, Vandenabeele P, Krysko DV, Agostinis P. Immunogenic cell death, DAMPs and anticancer therapeutics: an emerging amalgamation. Biochim Biophys Acta 2010; 1805:53-71. 4. Green DR, Ferguson T, Zitvogel L, Kroemer G. Immunogenic and tolerogenic cell death. Nat Rev Immunol 2009; 9:353-63. 5. Zitvogel L, Kepp O, Kroemer G. Decoding cell death signals in inflammation and immunity. Cell 2010; 140:798-804. 6. Casares N, Pequignot MO, Tesniere A, Ghiringhelli F, Roux S, Chaput N, et al. Caspase-dependent immunogenicity of doxorubicin-induced tumor cell death. J Exp Med 2005; 202:1691-701. 7. Obeid M, Tesniere A, Ghiringhelli F, Fimia GM, Apetoh L, Perfettini JL, et al. Calreticulin exposure dictates the immunogenicity of cancer cell death. Nat Med 2007; 13:54-61. 8. Krysko DV, Garg AD, Kaczmarek A, Krysko O, Agostinis P, Vandenabeele P. Immunogenic cell death and DAMPs in cancer therapy. Nat Rev Cancer 2012; 12:860-75. 9. Garg AD, Krysko DV, Verfaillie T, Kaczmarek A, Ferreira GB, Marysael T, et al. A novel pathway combining calreticulin exposure and ATP secretion in immunogenic cancer cell death. EMBO J 2012; 31:1062-79. 10. Garg AD, Krysko DV, Vandenabeele P, Agostinis P. Hypericin-based photodynamic therapy induces surface exposure of damage-associated molecular patterns like HSP70 and calreticulin. Cancer Immunol Immunother 2012; 61:215-21. 11. Ghiringhelli F, Apetoh L, Tesniere A, Aymeric L, Ma Y, Ortiz C, et al. Activation of the NLRP3 inflammasome in dendritic cells induces IL-1beta-dependent adaptive immunity against tumors. Nat Med 2009; 15:1170-8. 12. Martin SJ, Reutelingsperger CP, McGahon AJ, Rader JA, van Schie RC, LaFace DM, et al. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. The Journal of experimental medicine 1995; 182:1545-56. 13. Michaud M, Martins I, Sukkurwala AQ, Adjemian S, Ma Y, Pellegatti P, et al. Autophagy-dependent anticancer immune responses induced by chemotherapeutic agents in mice. Science 2011; 334:1573-7. 14. Garg AD, Krysko DV, Vandenabeele P, Agostinis P. DAMPs and PDT-mediated photo-oxidative stress: exploring the unknown. Photochem Photobiol Sci 2011; 10:670-80. 15. Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, Criollo A, et al. Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med 2007; 13:1050-9. 16. Galluzzi L, Kepp O, Kroemer G. Enlightening the impact of immunogenic cell death in photodynamic cancer therapy. EMBO J 2012; 31:1055-7. 17. Garg AD, Krysko DV, Vandenabeele P, Agostinis P. The emergence of phox-ER stress induced immunogenic apoptosis. OncoImmunology 2012; 1:787-9. 18. Tesniere A, Panaretakis T, Kepp O, Apetoh L, Ghiringhelli F, Zitvogel L, et al. Molecular characteristics of immunogenic cancer cell death. Cell Death Differ 2008; 15:3-12. 19. Dudek AM, Garg AD, Krysko DV, De Ruysscher D, Agostinis P. Inducers of immunogenic cancer cell death. Cytokine Growth Factor Rev 2013; DOI: 10.1016/j.cytogfr.2013.01.005.

22

Page 24: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

20. Agostinis P, Berg K, Cengel KA, Foster TH, Girotti AW, Gollnick SO, et al. Photodynamic therapy of cancer: An update. CA Cancer J Clin 2011; 61:250-81. 21. Garg AD, Nowis D, Golab J, Agostinis P. Photodynamic therapy: illuminating the road from cell death towards anti-tumour immunity. Apoptosis 2010; 15:1050-71. 22. Verfaillie T, Rubio N, Garg AD, Bultynck G, Rizzuto R, Decuypere JP, et al. PERK is required at the ER-mitochondrial contact sites to convey apoptosis after ROS-based ER stress. Cell Death Differ 2012; 19:1880-91. 23. Buytaert E, Callewaert G, Hendrickx N, Scorrano L, Hartmann D, Missiaen L, et al. Role of endoplasmic reticulum depletion and multidomain proapoptotic BAX and BAK proteins in shaping cell death after hypericin-mediated photodynamic therapy. FASEB J 2006; 20:756-8. 24. Verfaillie T, Garg AD, Agostinis P. Targeting ER stress induced apoptosis and inflammation in cancer. Cancer Lett 2010:DOI: 10.1016/j.canlet.2010.07.016. 25. Buytaert E, Matroule JY, Durinck S, Close P, Kocanova S, Vandenheede JR, et al. Molecular effectors and modulators of hypericin-mediated cell death in bladder cancer cells. Oncogene 2008; 27:1916-29. 26. Garg AD, Verfaillie T, Rubio N, Ferreira GB, Mathieu C, Agostinis P. Hypericin-PDT treatment of cancer cells leads to surface exposure/extracellular release of DAMPs and activates human immature dendritic cells. Belgian J Med Oncol 2010; 4:93-4. 27. Pearce EL. Metabolism in T cell activation and differentiation. Curr Opin Immunol 2010; 22:314-20. 28. Cantor JM, Ginsberg MH. CD98 at the crossroads of adaptive immunity and cancer. J Cell Sci 2012; 125:1373-82. 29. Ly LV, Sluijter M, Versluis M, Luyten GP, van Stipdonk MJ, van der Burg SH, et al. Peptide vaccination after T-cell transfer causes massive clonal expansion, tumor eradication, and manageable cytokine storm. Cancer Res 2010; 70:8339-46. 30. Ferradini L, Mackensen A, Genevee C, Bosq J, Duvillard P, Avril MF, et al. Analysis of T cell receptor variability in tumor-infiltrating lymphocytes from a human regressive melanoma. Evidence for in situ T cell clonal expansion. J Clin Invest 1993; 91:1183-90. 31. Uyttenhove C, Pilotte L, Theate I, Stroobant V, Colau D, Parmentier N, et al. Evidence for a tumoral immune resistance mechanism based on tryptophan degradation by indoleamine 2,3-dioxygenase. Nat Med 2003; 9:1269-74. 32. Maes H, Rubio N, Garg AD, Agostinis P. Autophagy: shaping the tumor microenvironment and therapeutic response. Trends in Molecular Medicine 2013; DOI: 10.1016/j.molmed.2013.04.005. 33. Dewaele M, Martinet W, Rubio N, Verfaillie T, de Witte PA, Piette J, et al. Autophagy pathways activated in response to PDT contribute to cell resistance against ROS damage. J Cell Mol Med 2011; 15:1402-14. 34. Rubio N, Coupienne I, Di Valentin E, Heirman I, Grooten J, Piette J, et al. Spatiotemporal autophagic degradation of oxidatively damaged organelles after photodynamic stress is amplified by mitochondrial reactive oxygen species. Autophagy 2012; 8:1312-24. 35. Panaretakis T, Kepp O, Brockmeier U, Tesniere A, Bjorklund AC, Chapman DC, et al. Mechanisms of pre-apoptotic calreticulin exposure in immunogenic cell death. EMBO J 2009; 28:578-90. 36. Johnson S, Michalak M, Opas M, Eggleton P. The ins and outs of calreticulin: from the ER lumen to the extracellular space. Trends Cell Biol 2001; 11:122-9. 37. Papp S, Opas M. Sub-Cellular Distribution of Calreticulin. In: Eggleton P, Michalak M, eds. Calreticulin. Georgetown, Tex.; New York, N.Y.: Landes Bioscience/Eurekah.com; Kluwer Academic/Plenum, 2003:38-48. 38. Xiao G, Chung TF, Pyun HY, Fine RE, Johnson RJ. KDEL proteins are found on the surface of NG108-15 cells. Brain Res Mol Brain Res 1999; 72:121-8. 39. Bravo R, Vicencio JM, Parra V, Troncoso R, Munoz JP, Bui M, et al. Increased ER-mitochondrial coupling promotes mitochondrial respiration and bioenergetics during early phases of ER stress. J Cell Sci 2011; 124:2143-52. 40. Kuhn H, Borchert A. Regulation of enzymatic lipid peroxidation: the interplay of peroxidizing and peroxide reducing enzymes. Free Radic Biol Med 2002; 33:154-72. 41. Klionsky DJ, Abdalla FC, Abeliovich H, Abraham RT, Acevedo-Arozena A, Adeli K, et al. Guidelines for the use and interpretation of assays for monitoring autophagy. Autophagy 2012; 8:445-544.

23

Page 25: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

42. Menger L, Vacchelli E, Adjemian S, Martins I, Ma Y, Shen S, et al. Cardiac glycosides exert anticancer effects by inducing immunogenic cell death. Sci Transl Med 2012; 4:143ra99. 43. Peters LR, Raghavan M. Endoplasmic reticulum calcium depletion impacts chaperone secretion, innate immunity, and phagocytic uptake of cells. J Immunol 2011; 187:919-31. 44. Tarr JM, Young PJ, Morse R, Shaw DJ, Haigh R, Petrov PG, et al. A mechanism of release of calreticulin from cells during apoptosis. J Mol Biol 2010; 401:799-812. 45. Tufi R, Panaretakis T, Bianchi K, Criollo A, Fazi B, Di Sano F, et al. Reduction of endoplasmic reticulum Ca2+ levels favors plasma membrane surface exposure of calreticulin. Cell Death Differ 2008; 15:274-82. 46. Yorimitsu T, Nair U, Yang Z, Klionsky DJ. Endoplasmic reticulum stress triggers autophagy. J Biol Chem 2006; 281:30299-304. 47. Nystrom T. Role of oxidative carbonylation in protein quality control and senescence. EMBO J 2005; 24:1311-7. 48. Kaushik S, Massey AC, Mizushima N, Cuervo AM. Constitutive activation of chaperone-mediated autophagy in cells with impaired macroautophagy. Mol Biol Cell 2008; 19:2179-92. 49. Pyo JO, Nah J, Kim HJ, Lee HJ, Heo J, Lee H, et al. Compensatory activation of ERK1/2 in Atg5-deficient mouse embryo fibroblasts suppresses oxidative stress-induced cell death. Autophagy 2008; 4:315-21. 50. Garg AD, Kaczmarek A, Krysko O, Vandenabeele P, Krysko DV, Agostinis P. ER stress-induced inflammation: does it aid or impede disease progression? Trends Mol Med 2012; 18:589-98. 51. Schroder K, Hertzog PJ, Ravasi T, Hume DA. Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol 2004; 75:163-89. 52. Zaidi MR, Merlino G. The two faces of interferon-gamma in cancer. Clin Cancer Res 2011; 17:6118-24. 53. Matarese G, La Cava A, Horvath TL. In vivo veritas, in vitro artificia. Trends Mol Med 2012; 18:439-42. 54. Martins I, Tesniere A, Kepp O, Michaud M, Schlemmer F, Senovilla L, et al. Chemotherapy induces ATP release from tumor cells. Cell Cycle 2009; 8:3723-8. 55. Thorburn J, Horita H, Redzic J, Hansen K, Frankel AE, Thorburn A. Autophagy regulates selective HMGB1 release in tumor cells that are destined to die. Cell Death Differ 2009; 16:175-83. 56. Kazama H, Ricci JE, Herndon JM, Hoppe G, Green DR, Ferguson TA. Induction of immunological tolerance by apoptotic cells requires caspase-dependent oxidation of high-mobility group box-1 protein. Immunity 2008; 29:21-32. 57. Venereau E, Casalgrandi M, Schiraldi M, Antoine DJ, Cattaneo A, De Marchis F, et al. Mutually exclusive redox forms of HMGB1 promote cell recruitment or proinflammatory cytokine release. J Exp Med 2012; 209:1519-28. 58. Kharbanda S, Garg AD. F1000 Recommendation of [Venereau E et al., J Exp Med 2012, 209(9):1519-28]. Faculty of 1000 2012; f1000.com/717953753#eval793459358. 59. Wei H, Wei S, Gan B, Peng X, Zou W, Guan JL. Suppression of autophagy by FIP200 deletion inhibits mammary tumorigenesis. Genes Dev 2011; 25:1510-27. 60. Noman MZ, Janji B, Kaminska B, Van Moer K, Pierson S, Przanowski P, et al. Blocking hypoxia-induced autophagy in tumors restores cytotoxic T-cell activity and promotes regression. Cancer Res 2011; 71:5976-86. 61. Ferreira GB, van Etten E, Lage K, Hansen DA, Moreau Y, Workman CT, et al. Proteome analysis demonstrates profound alterations in human dendritic cell nature by TX527, an analogue of vitamin D. Proteomics 2009; 9:3752-64. 62. Mohanty JG, Bhamidipaty S, Evans MK, Rifkind JM. A fluorimetric semi-microplate format assay of protein carbonyls in blood plasma. Anal Biochem 2010; 400:289-94. 63. Vantieghem A, Assefa Z, Vandenabeele P, Declercq W, Courtois S, Vandenheede JR, et al. Hypericin-induced photosensitization of HeLa cells leads to apoptosis or necrosis. Involvement of cytochrome c and procaspase-3 activation in the mechanism of apoptosis. FEBS Lett 1998; 440:19-24.

24

Page 26: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

Figure Legends:

Figure 1. GPX4 overexpression and L-histidine suppress Hyp-PDT induced ecto-CALR and secreted ATP

in cancer cells. (A, B) T24 cells were preincubated with 25 mM of L-histidine for 30 min followed by

treatment with indicated doses of Hyp-PDT (medium – 1.35 J/cm2, high – 2.16 J/cm2) or left untreated

(CNTR) and recovered 1 h post-PDT. On one hand, the surface proteins were biotinylated and

immunoblotted (A) while on the other, the resulting conditioned media was analyzed for the presence of

ATP (B). In (B), data are presented as relative light unit (RLU) values (3 independent experimental

determinations with duplicate determinations in each; mean ± s.e.m.; *P<0.05 as indicated by bars). (C, D)

HeLa cells stably expressing either empty vector (Neo) or overexpressing GPX4 were treated with the

indicated doses of Hyp-PDT (or with medium dose wherever not mentioned) or left untreated (CNTR). The

cells were recovered 1 h post-PDT and in one case the surface proteins were biotinylated followed by

immunoblotting (C) while on the other, the resulting conditioned media was analyzed for the presence of

ATP (D). In D, data are presented as relative light unit (RLU) values (3 independent experimental

determinations with duplicate determinations in each; mean ± s.e.m.; *P<0.05 as indicated by the bar). (E,

F) L929 cells stably expressing either empty vector (Neo) or overexpressing GPX4 were treated with

medium Hyp-PDT dose or left untreated (CNTR). The cells were recovered 1 h post-PDT and in one case

the surface proteins were biotinylated followed by immunoblotting (E) while on the other, the resulting

conditioned media was analyzed for the presence of ATP (F). In F, data are presented as relative light unit

(RLU) values (4 experimental determinations; mean ± s.e.m.; *P<0.05 as indicated by the bar). Here,

‘+BIO’ indicates controls exposed to buffer with biotin and ‘-BIO’ indicates controls exposed to buffer

without biotin.

Figure 2. ATG5 knockdown or knockout does not affect Hyp-PDT induced ATP secretion. (A, B) T24

cells were transfected with scrambled (SCR) siRNA or an ATG5-specific siRNA (ATG5KD) (A) and

A375m cells expressing control (CO = vector expressing scrambled shRNA) or an ATG5-specific shRNA

(ATG5KD) (B) were lysed followed by immunblotting. ATG5 and actin protein bands were quantified for

the integrated band density via Image J and their ratios calculated (T24 – A; A375m - B). Data are

presented as ATG5/actin ratio values normalized to those of SCR/CO samples multiplied by 100 (3

independent experimental determinations; mean ± s.e.m.). (C) T24 cells or A375m cells were treated with

Hyp-PDT doses of 2.16 J/cm2 or 2.43 J/cm2, respectively. They were recovered 1 h post-PDT and the

resulting conditioned media was analyzed for the presence of ATP. Data are presented as relative light unit

25

Page 27: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

(RLU) values (3 independent experimental determinations with duplicate determinations in each; mean ±

s.e.m.; statistical analysis was done using the Student’s t-test; *P<0.05 vs. CNTR, #P<0.05 as indicated by

the bars). (D-F) T24 cells (D) or A375m cells (E) transfected with/expressing respective siRNAs/shRNAs

(causing ATGKD as applicable) and wild-type MEF cells (Atg5+/+) or MEF cells lacking atg5 (atg5-/-) (F)

were treated with indicated Hyp-PDT doses (medium - 1.35 J/cm2 and high – 2.16 J/cm2) or the medium

dose (1.35 J/cm2) in case of MEFs; or left untreated (CNTR). They were recovered 1 h post-PDT and the

resulting conditioned media was analyzed for the presence of ATP. Data are presented as relative light unit

(RLU) values (3 independent experimental determinations with duplicate determinations in each; mean ±

s.d.; *P<0.05 vs. respective CNTRs, N.S. = not significant, as indicated by the bars).

Figure 3. ATG5 knockdown in cancer cells increases Hyp-PDT induced, ecto-CALR and accumulation of

oxidatively damaged proteins. (A, B) T24 cells transfected with scrambled (SCR) siRNA or an ATG5-

specific siRNA (ATG5KD) (A) and A375m cells expressing the control (CO = vector expressing scrambled

shRNA) or an ATG5-specific shRNA (ATG5KD) (B) were treated with the indicated Hyp-PDT doses

(medium – 1.35 J/cm2, high – 2.16 J/cm2, higher – 2.43 J/cm2) or left untreated (CNTR). The cells were

recovered 1 h post-PDT and the surface proteins were biotinylated followed by immunoblotting (A). Here

in A and B, ‘+BIO’ indicates controls exposed to buffer with biotin and ‘-BIO’ indicates controls exposed

to buffer without biotin. Ecto-CALR and FAS protein bands were quantified for the integrated band density

via Image J and their ratios calculated (T24 – A; A375m – B; 3 independent experimental determinations,

mean ± s.e.m.; *P<0.05 as indicated by the bars). (C, D) T24 cells (C) and A375m cells (D) transfected

with/expressing respective siRNA/shRNA constructs (as detailed in A, B) were treated with high Hyp-PDT

dose (2.16 J/cm2) or left untreated. They were recovered 1 h post-PDT and the levels of carbonylated

proteins were estimated. In (C and D), the carbonyl content was calculated as, nmols of carbonylated

proteins per mg of total proteins. Data are presented as carbonyl content values normalized to the

respective untreated controls (3 independent experimental determinations - mean ± s.e.m.; *P<0.05 as

indicated by the bars).

Figure 4. Hyp-PDT-induced autophagy in cancer cells suppresses maturation of the interacting dendritic

cells (DCs) in an ecto-CALR-independent manner. (A) A375m cancer cells expressing control CO-shRNA

(indicated by ‘-‘) (CO = vector expressing scrambled shRNA) or an ATG5-specific shRNA (ATG5KD;

indicated by ‘+’) were treated with Hyp-PDT (2.43 J/cm2) or left untreated (CNTR). Thereafter the

respective cells were recovered 24 h post-PDT and cocultured with human-immature DCs (hu-iDCs) for 24

26

Page 28: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

h. As a positive control, hu-iDCs were stimulated with LPS for 24 h. After coculturing, the cells were

immunostained for CD86/HLA-DR positivity and scored by FACS analysis. Data are presented as

percentage of CD86-positive and HLA-DR-positive cells (3 independent experimental determinations with

duplicate determinations in each; mean ± s.e.m.; *P<0.05, vs. respective CNTR; as indicated by bars). (B-

D) The A375m-hu-iDC coincubation conditioned media obtained during the experiments detailed in (A)

were collected followed by analysis for concentrations of IL10 (B), IL1B (C) and IL6 (D). Absolute

concentrations are depiction of one representative (four replicate determinations; mean ± s.e.m.; *P<0.05 or

N.S. = non-significant, as indicated by the bars) of 3 independent experiments. Data for hu-iDCs only is

not depicted since they produced no recordable levels of these three cytokines. (E) Respective A375m

cancer cells were treated as in (A), and recovered 24 h post-PDT, incubated with either chicken anti-CALR

or isotype IgY antibodies (Ab) and cocultured with hu-iDCs for 24 h. After coculturing, the cells were

immunostained for CD86/HLA-DR positivity and scored by FACS analysis. Data are presented as

percentage of CD86-positive and HLA-DR-positive cells (3 experimental determinations; mean ± s.e.m.;

*P<0.05 as indicated by the bars).

Figure 5. Hyp-PDT-induced autophagy in cancer cells reduces proliferation of CD4+ or CD8+ T cells. (A,

B) A375m cancer cells expressing control shRNA (indicated by ‘+‘ in CO-shRNA A375m line) (CO =

vector expressing scrambled shRNA) or an ATG5-specific shRNA (indicated by ‘+‘ in ATG5-shRNA

A375m line) were treated with Hyp-PDT (2.43 J/cm2) or left untreated (CNTR). Thereafter the respective

cells were recovered 24 h post-PDT and cocultured with human-immature DCs (hu-iDCs) for 24 h. As a

positive control, hu-iDCs were stimulated with LPS for 24 h. After that, the cells were further cocultured

with naïve human T cells for 5 days followed by immunostaining for CD3/CD4 positivity (A) or CD3/CD8

positivity (B); and scoring by FACS analysis. ‘T cells’ represents negative control for T cells proliferation.

Data are presented as percentage of respective proliferating T cells (4 experimental determinations, across

T cells from two separate healthy human donors; mean ± s.e.m.; statistical analysis was done using the

Student’s t-test; *P<0.05 as indicated by bars). (C) The A375m-hu-iDC-T cell coincubation conditioned

media obtained during the experiments detailed in (A, B) were collected followed by analysis for

concentration of IFNG. Hu-iDCs stimulated with LPS for 24 h represent positive controls whereas “T

cells” and “T cells + DCs” are the negative controls for IFNG production. Data are presented as fold

change in absolute concentrations (3 experimental determinations; mean ± s.e.m.; *P<0.05 or N.S. = non-

significant as determined by one-way ANOVA comparison between data sets as indicated by bars). (D)

Schematic overview of cancer cell-associated autophagy’s role in Hyp-PDT induced immunogenic cell

27

Page 29: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

death (ICD). Hyp-PDT treatment causes photo-oxidative (phox)-ER stress in the treated cancer cells which

is associated with increased surface expose of calreticulin (ecto-CALR), ATP secretion, DC maturation

characterized by increased CD86/HLA-DR positivity and production of IL6 as well as increased

proliferation of CD4+/CD8+ T cells. Here, we observed that autophagy induced following Hyp-PDT

treatment of these cancer cells suppresses various crucial determinants of ICD i.e. ecto-CALR induction

(without affecting ATP secretion), maturation of interacting DCs and IL6 production and proliferation of

CD4+ or CD8+ T cells.

28

Page 30: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

A T24 cells B T24 cells

C HeLa cells D HeLa cells

E L929 cells F L929 cells

Figure 1.

Page 31: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

a b c

A T24 cells B A375m cells

C D T24 cells

E A375m cells F MEF cells

Figure 2.

Page 32: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

A T24 cells

B A375m cellsB A375m cells

C T24 cells D A375m cells

Figure 3.

Page 33: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

A

B C DIL10 IL1B IL6

E

Figure 4.

Page 34: ROS-induced autophagy in cancer cells assists in · PDF fileROS-induced autophagy in cancer cells assists in evasion from ... cancer cells through an apoptotic cell death subroutine

A BCD4+ T cells CD8+ T cells

C

D

Figure 5.