HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 20/6/756 05-4305fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Buytaert, E. Articles by Agostinis, P. Search for Related Content PubMed PubMed Citation Articles by Buytaert, E. Articles by Agostinis, P.

Role of endoplasmic reticulum depletion and multidomain proapoptotic BAX and BAK proteins in shaping cell death after hypericin-mediated photodynamic therapy

Esther Buytaert^*, Geert Callewaert^*, Nico Hendrickx^*, Luca Scorrano, Dieter Hartmann^*, Ludwig Missiaen^*, Jackie R. Vandenheede^*, Ingeborg Heirman, Johan Grooten and Patrizia Agostinis^*,1

^* Department of Molecular and Cell Biology, Faculty of Medicine, Catholic University of Leuven, Leuven, Belgium;
Department for Molecular Biomedical Research, Unit of Molecular Immunology, Flanders Interuniversity Institute for Biotechnology and Ghent University, Ghent, Belgium; and
Dulbecco-Telethon Institute, Venetian Institute of Molecular Medicine, Padova, Italy

¹Correspondence: Division of Biochemistry, Campus Gasthuisberg, Herestraat 49, Leuven B-3000, Belgium. E-mail: patricia.agostinis{at}med.kuleuven.be

SPECIFIC AIMS

Photodynamic therapy (PDT) is an established and useful modality for the treatment of cancer and the naturally occurring photosensitizer hypericin is a promising drug in PDT. Although PDT utilizes mainly the mitochondrial pathway of caspase activation, evidence is emerging that caspase-independent pathways contribute to cell killing. This study aims to provide a better understanding of the commitment event and the cell death modality after hypericin-PDT.

PRINCIPAL FINDINGS

1. A rapid [Ca²⁺]_cyt rise precedes the onset of intrinsic apoptosis in photodamaged cells
Colocalization studies revealed that hypericin associated predominantly with the endoplasmic reticulum (ER) independently of cell type. Because the ER is the main cellular store of Ca²⁺, the fluorescent Ca²⁺ indicator Indo-1 was used to evaluate the effects of hypericin photoactivation (e.g., PDT) on cytosolic Ca²⁺ levels. Exposure of cells to PDT provoked a rise in [Ca²⁺]_cyt within minutes, as a result of intracellular Ca²⁺ stores depletion. This initial [Ca²⁺]_cyt elevation was followed by manifestation of several parameters of intrinsic apoptosis, including mitochondrial BAX translocation, cytochrome c release, caspase-3 activation, and phosphatidylserine exposure.

2. PDT-mediated loss of native SERCA2 protein causes ER-Ca²⁺ depletion and cell death
Changes in steady-state intracellular Ca²⁺ levels in response to PDT were monitored by targeting aequorin Ca²⁺ photoreporter to the ER lumen (erAEQ), cytosol (cytAEQ), and mitochondria (mtAEQ) of HeLa cells. In permeabilized cells exposed to 200 nM free [Ca²⁺], steady-state [Ca²⁺]_ER was drastically reduced after PDT (Fig. 1 A, left panel). As a consequence of PDT-induced deregulation of ER-Ca²⁺ homeostasis, histamine-dependent [Ca²⁺]_cyt and [Ca²⁺]_mit transients, typically observed in untreated cells, were drastically blunted in PDT treated cells (Fig. 1A , middle and right panels).

View larger version (32K): [in this window] [in a new window]   Figure 1. Disturbed Ca2+ homeostasis and SERCA2 photodamage by hypericin-PDT. A) Representative recording of [Ca2+]ER, [Ca2+]cyt, and [Ca2+]mit in HeLa cells transfected with different aequorin constructs (erAEQ, cytAEQ, and mtAEQ, respectively). Cells were transfected for 24 h prior to incubation with 125 nM hypericin for 16 h, followed by PDT treatment (control=untreated). erAEQ transfected cells were permeabilized with digitonin prior to addition of 200 nM free Ca2+ as indicated. ER-Ca2+ release was triggered by addition of InsP3 (IP3) where indicated. Ca2+ signals in mtAEQ and cytAEQ transfected cells were measured in modified KRB containing 1 mM CaCl2 and histamine was added when indicated. Traces represent the average of 4–8 measurements from a representative experiment. Each experiment was repeated at least twice. B) HeLa cells, HeLa Neo/GPx4, and MEFs loaded with 125 nM, 65 nM, and 200 nM hypericin, respectively, were harvested at indicated time points (0 h=immediately) after irradiation (4 J/cm2 for HeLa cells and 2.7 J/cm2 for MEFs). Cells were pretreated for 30 min with 25 mM L-histidine (L-hist) prior to irradiation where indicated. Cell lysates were analyzed by Western blot for SERCA2. Equal protein levels were confirmed with a specific antitubulin antibody. C) Representative recording of [Ca2+]ER in untreated (control) and PDT-treated HeLa cells transfected with erAEQ (A). L-histidine was added as described in B.

We next evaluated the possibility that these profound perturbations in the ER-Ca²⁺ homeostasis were provoked by direct photodamage to the sarco(endo)plasmic-reticulum Ca²⁺-ATPase (SERCA) pump, whose activity is essential to maintain the physiological level of resting [Ca²⁺]_ER. A drastic drop in protein levels of native 100 kDa SERCA2 as observed on Western blots, occurred immediately after exposure of the cells to PDT (Fig. 1B ). This initial event was largely prevented by ¹O₂ quencher L-histidine or overexpression of glutathione peroxidase (GPx4), which removes ¹O₂-induced phospholipid hydroperoxides in cellular membranes (Fig. 1B ), while it was unaffected by inhibitors of caspases or of the proteasome. Increasing either pharmacologically or genetically the cellular antioxidant capacity not only preserved native SERCA2 protein levels after PDT, but also restored ER-Ca²⁺ refilling after exposure of permeabilized HeLa cells to 200 nM free Ca²⁺ (Fig. 1C ) and protected all cells against photokilling, thus establishing a functional link between SERCA2 photodamage, loss of ER-Ca²⁺ homeostasis, and cell death.

3. BAX and BAK are strictly necessary for mitochondrial apoptosis but are dispensable for overall PDT-induced cell death
To explore the link between ER-Ca²⁺ depletion and the BAX/BAK gateway at the mitochondria, we used Bax^–/–Bak^–/– double-knockout (DKO) murine embryonic fibroblasts (MEFs). DKO cells have been reported to display a reduced resting [Ca²⁺]_ER, a secondarily decreased Ca²⁺ uptake by mitochondria, and to be resistant to a wide range of apoptosis inducers.

SERCA2 protein levels dropped drastically after PDT in both (wild-type) WT and in DKO MEFs. However, while WT cells manifested apoptotic hallmarks of cell death, DKO cells showed negligible nuclear fragmentation, absence of caspase-3 activation, and PARP cleavage, as well as insensitivity to caspase inhibitors. These apoptotic features were completely restored in DKO cells reexpressing BAX targeted at the mitochondria (DKO-mtBAX), indicating that BAX is the required gateway at the mitochondria for effector caspase activation after hypericin-PDT. However, PDT-treated DKO cells readily exhibited permeabilization of the plasma membrane and suffered a significant time dependent loss of viability, indicating that photokilling was proceeding through a caspase-independent pathway. These observations suggest that while SERCA photodamage (and consequent ER-Ca²⁺ depletion) is sufficient to commit cells to death, BAX/BAK are necessary to induce apoptosis in PDT-treated cells.

4. In the absence of BAX and BAK, photokilling is executed through an autophagic cell death pathway
We next analyzed at the morphological, ultrastructural, and biochemical level which type of caspase-independent death pathway was initiated by PDT in DKO cells. PDT-treated WT and DKO-mtBAX cells exhibited cellular shrinkage and membrane blebbing characteristic of apoptotic cell death, whereas DKO cells and, more prominently, DKO cells overexpressing SERCA2 (DKO-SERCA) readily showed extensive cytosolic vacuolization, which was prevented by the inhibitor of phosphatidylinositol 3-kinase (PI3K), wortmannin (WO). In the presence of WO, PDT-treated DKO cells appeared rounded yet they did not show the typical morphology of apoptotic cells. Since class III PI3K has been found to be necessary for sequestration of cytoplasmic material in autophagy, these observations suggested the induction of an autophagic pathway.

DKO cells were further examined by transmission electron microscopy. In PDT-treated DKO cells, compartimentalization of the cytoplasm by double-layered isolation membranes surrounding organelles and formation of autophagosomes characteristic of macroautophagy were clearly evident. Kinetic experiments in DKO cells expressing GFP fused to light-chain 3 (LC3), visualized the redistribution of GFP-LC3 from a diffuse to a vacuolar/punctuate pattern after PDT, indicating recruitment of cytosolic LC3 (LC3 I) to autophagosomal membranes upon processing (LC3 II), a biochemical hallmark of autophagy. In DKO and DKO-SERCA, WO specifically blocked accumulation of the electrophoretically faster migrating LC3 II (16 kDa) form induced by PDT and markedly lowered cell death in these cells. Conversely, this autophagy antagonist did not significantly affect apoptotically dying WT and DKO-mtBAX cells. In DKO cells, WO did not prevent SERCA2 photodamage, indicating that this initial event is not the result of a proteolytic pathway activated by autophagy, thus positioning the protective action of PI3K inhibition in DKO cells downstream of ER-Ca²⁺ depletion.

5. PDT-induced autophagic cell death is not mediated by caspase inhibition
To evaluate whether autophagy-induced cell death and the caspase activation cascade are two separate routes of cell demise or two signaling pathways that are subjected to a mutual control, we studied whether blocking DEVD-directed caspase activity in PDT-treated WT cells would increase the level of autophagy. In PDT-treated WT cells, zDEVD-fmk abolished caspase-3 activity and reduced PDT-induced mitochondrial apoptosis, consistent with previous findings, but did not instigate autophagy. Inhibition of autophagy by WO did not alter caspase-3 activation in WT cells or its lethal effect. In PDT-treated DKO cells, zDEVD-fmk did not affect autophagic cell death and WO did not induce caspase-3 processing, indicating again the strict requirement of multidomain BAX and BAK for induction of mitochondrial apoptosis.

These observations suggest that PDT-induced autophagy in MEFs is not merely a consequence of the inhibition of effector caspases and that caspase-mediated apoptosis is not affected by inhibitors of autophagy.

CONCLUSIONS AND SIGNIFICANCE

Our study provides novel evidence indicating that the "point of no return" in the cell death program by hypericin-PDT is upstream of the BAX/BAK gateway at the mitochondria and consists of irreversible oxidative damage to the SERCA2 pump, a key ER-Ca²⁺ regulator, with consequent loss of Ca²⁺ homeostasis within the cell. In the presence of BAX and BAK, cell death occurs by apoptosis, while in the absence of BAX and BAK, autophagy leads to the execution of the death program (Fig. 2 ). Irrespective of the possibility that PDT-induced autophagy in Bax^–/–Bak^–/– double-knockout cells might be initiated as an attempt to remove an irreversibly damaged organelle (e.g., the ER) and to preserve cells from a metabolic collapse, this study reveals that its persistence contributes to cell killing.

View larger version (21K): [in this window] [in a new window]   Figure 2. Schematic overview of the main cell death pathways induced by hypericin-PDT. Upon light activation, ER-associated hypericin mediates oxidative damage to the SERCA2 pump through photogeneration of singlet oxygen (1O2), causing irreversible depletion of ER-Ca2+ stores. Downstream of the ER, the presence of proapoptotic BAX/BAK proteins promotes mitochondrial outer membrane permeabilization (MOMP) leading to apoptotic cell death. However, PDT-induced emptying of ER-Ca2+ stores can activate a WO-inhibitable autophagic pathway, which is revealed by the absence of multidomain BAX/BAK proteins and occurs without detection of MOMP. Sustained progression of the autophagic pathway leads to nonapoptotic death of Bax–/–Bak–/– DKO cells.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-4305fje;

This article has been cited by other articles:

				J.-W. Chang, H. Choi, H.-J. Kim, D.-G. Jo, Y.-J. Jeon, J.-Y. Noh, W. J. Park, and Y.-K. Jung Neuronal vulnerability of CLN3 deletion to calcium-induced cytotoxicity is mediated by calsenilin Hum. Mol. Genet., February 1, 2007; 16(3): 317 - 326. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 20/6/756 05-4305fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Buytaert, E. Articles by Agostinis, P. Search for Related Content PubMed PubMed Citation Articles by Buytaert, E. Articles by Agostinis, P.