Transglutaminase 2 Promotes Both Caspase-dependent and Caspase-independent Apoptotic Cell Death via the Calpain/Bax Protein Signaling Pathway*

Background: The function of transglutaminase 2 (TG2) in dying cells is ambiguous. Results: TG2 activity facilitates simultaneous caspase-dependent and apoptosis-inducing factor-mediated apoptotic cell death via a calpain/Bax signaling pathway. Conclusion: We demonstrated a new mechanism of apoptotic cell death induced by photodynamic therapy (PDT). Significance: Understanding the function of TG2 as a therapeutic target will contribute to improved cancer treatments using PDT. Transglutaminase 2 (TG2) is a versatile protein that is implicated in significant biological processes, including cell death and degenerative diseases. A possible role of TG2 in the apoptotic death of cancer cells induced by photodynamic therapy (PDT) was suggested recently; however, the mechanism by which TG2 regulates apoptotic responses to PDT remains to be elucidated. In this study, we investigated the key signaling pathways stimulated during apoptotic cell death following PDT and whether inhibition of TG2 activation using pharmacological approaches and siRNAs affects the signaling pathways. PDT caused the release of both cytochrome c and apoptosis-inducing factor (AIF) by damaging mitochondria, which resulted in caspase-dependent and caspase-independent apoptotic cell death, respectively. Released AIF translocated to the nucleus and, synergistically with the caspase-dependent pathway, led to apoptotic cell death. Both the caspase cascade and the activation of AIF following PDT were mediated by TG2 activation. In addition, PDT-activated calpain was responsible for the sequential events of Bax translocation, the collapse of ΔΨm, caspase-3 activation, and AIF translocation, all of which were provoked by TG2 activation. Together, these results demonstrate that PDT with a chlorin-based photosensitizer targets TG2 by activating calpain-induced Bax translocation, which induces apoptotic cell death through both caspase-dependent and AIF-mediated pathways. Moreover, these results indicate that TG2 may be a possible therapeutic target for PDT treatment of cancer.

Transglutaminase 2 (TG2), 2 the most ubiquitous member of the transglutaminase family, catalyzes post-translational pro-tein cross-linking with the formation of ⑀-(␥-glutamyl)lysine or (␥-glutamyl)polyamine bonds. This transamidating activity of TG2 regulates a variety of biological responses, including cell death, cell adhesion and proliferation, cytoskeletal reorganization, and cell migration and invasion (1)(2)(3). Additionally, TG2 is implicated in neurodegenerative diseases, celiac disease, ischemic stroke and cardiovascular diseases, cataractogenesis, diabetes, and cancer (4 -8). In addition to its fundamental transamidating activity, TG2 possesses various other enzymatic activities. TG2 retains GTPase activity like other G-proteins, a feature distinguishing TG2 from the other transglutaminases, and GTP-bound TG2 mediates the activation of phospholipase C␦1 by ␣ 1 -adrenergic receptor signaling (9). TG2 may also function as a protein-disulfide isomerase, protein kinase, DNA nuclease, or adaptor protein in subcellular position-and stimulus-dependent manners (3,4). Therefore, the role of TG2 in physiological and pathological processes is complex and can vary due to its multifunctional characteristics.
Although several studies have demonstrated the crucial role of TG2 in apoptotic cells, the mechanisms by which TG2 triggers apoptosis are not well understood. TG2 promotes apoptosis of cardiomyocytes by reducing phospholipase C␦1 and active PKC levels during oxidative stress (10). TG2 also mediates calphostin C-induced apoptosis via a mechanism involving activation of the c-Jun N-terminal kinase and translocation of Bax to the mitochondria in NIH3T3 fibroblasts and breast cancer cells (11). These events ultimately stimulate a caspase-dependent pathway. TG2 can induce apoptosis-inducing factor (AIF)-mediated caspase-independent apoptosis in pancreatic cancer cells treated with the calcium ionophore A23187 (12). However, the signaling pathway by which TG2 regulates the action of AIF remains to be explored. Thus, TG2 can activate the AIF-mediated pathway as an alternative mechanism when caspases are unavailable during the cell death process. However, the signaling mechanisms involving TG2 with regard to caspase-dependent and caspase-independent apoptosis are not clearly understood.
Photodynamic therapy (PDT) can induce not only routine caspase-dependent apoptosis but also caspase-independent apoptosis, depending on the photosensitizer and the cell context. PDT generally causes apoptosis by activating a caspase cascade, a result of the release of cytochrome c through the mitochondrial permeability transition (13)(14)(15), which is controlled by Bcl-2 family members. On the other hand, by damaging mitochondria, PDT induces the release of AIF into the cytosol and its subsequent nuclear translocation, leading to apoptosis in a caspase-independent manner (16). Similar results were observed in cells loaded with the phthalocyanine derivative 2,3,9,10,16,17,23,24-octakis(3-aminopropyloxy) phthalocyaninato zinc II following PDT. PDT damages lysosomes as a primary target and, in turn, induces Bid activation, mitochondrial dysfunction, and nuclear translocation of AIF, resulting in caspase-independent apoptosis (17). Furthermore, studies have indicated that the caspase-dependent pathway and the AIF-mediated pathway are simultaneously involved in the apoptotic death of cancer cells following PDT (18,19). However, the mechanisms underlying the simultaneous induction of the two pathways and the mechanism by which AIF is regulated during PDT remain to be elucidated.
We recently presented a possible role of TG2 in apoptotic cell death following PDT with a chlorin-based photosensitizer localizing in the endoplasmic reticulum, mitochondria, and lysosomes. The activity of TG2 may mediate the apoptotic death of cancer cells through a mechanism coupled with increases in intracellular reactive oxygen species (ROS) and Ca 2ϩ levels (20). However, the signaling mechanism by which TG2 regulates apoptotic cell death during PDT is not clearly understood. In this study, we elucidated the apoptotic signaling pathways mediated by TG2 activation in human cancer cells during PDT with a photosensitizer. TG2 promoted apoptotic cell death in response to PDT through a signaling cascade involving the calpain/Bax signaling pathway due to increased transamidating activity, but not due to increased TG2 expression levels. The resulting activation of calpain was sufficient to induce the collapse of the mitochondrial membrane potential (⌬⌿ m ) through the translocation of Bax to mitochondria, which ultimately induced both caspase-dependent and caspase-independent death pathways.
Transfection with siRNAs-Cells were transfected with siRNAs following the procedures of Yi et al. (21). Briefly, TG2specific siRNA duplexes (5Ј-AAGAGCGAGAUGAUCUG-GAAC-3Ј) targeting the coding sequence of human TG2 mRNA, AIF-specific siRNAs (ON-TARGETplus SMARTpool), and control siRNA (ON-TARGETplus non-targeting pool) were obtained from Dharmacon (Lafayette, CO). Cells were transfected with siRNAs using siLentFect lipid reagent (Bio-Rad) according to the manufacturer's instructions. Twenty hours after transfection, the media were replaced with fresh culture medium, and the transfected cells were further incubated for 24 h prior to experimentation.
In Situ Transglutaminase Activity Assay-In situ transglutaminase activity was determined by a confocal microscopic assay following the procedures of Yoo et al. (20). Briefly, cells were incubated with 1 mM 5-(biotinamido)pentylamine for 1 h at 37°C, and biotinylated proteins were probed with FITC-conjugated streptavidin. Single-cell fluorescence intensities were determined by confocal microscopy (Olympus FluoView FV300) in ϳ30 cells from three separate experiments. Relative TG2 activity was determined by comparing the fluorescence intensities of irradiated cells with those of non-irradiated control cells (-fold differences).
Western Blot Analysis-To obtain cell lysates, cells were incubated for 30 min in lysis buffer containing 50 mM Tris-HCl (pH 7.5), 1% Triton X-100, 10% glycerol, 150 mM NaCl, 1 mM EDTA, 0.1 mM PMSF, 10 g/ml aprotinin, and 10 g/ml leupeptin. Cells were scraped and centrifuged at 20,000 ϫ g for 10 min at 4°C. Cytosolic and mitochondrial extracts were prepared using the Thermo Scientific Mitochondria Isolation Kit for Cultured Cells (Pierce) according to the manufacturer's instructions.
Cell lysates and fractionated extracts were separated by SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blocked with 5% nonfat dried milk in 20 mM Tris (pH 7.6), 138 mM NaCl, and 0.1% Tween 20 for 1 h at room temperature. Proteins were probed with antibodies against the target proteins. Protein bands were visualized by exposure to x-ray film and a chemiluminescent substrate (Pierce).

TG2/Calpain/Bax-dependent Apoptotic Cell Death
Determination of ⌬⌿ m -⌬⌿ m was monitored by confocal microscopy with the potential-sensitive JC-1 cationic dye (Molecular Probes, Eugene, OR), which exhibits potential-dependent accumulation in mitochondria. Mitochondrial depolarization (loss of ⌬⌿ m ) is indicated by an increase in the ratio of the fluorescence intensity of the monomeric form (green) to that of the J-aggregates (red). To monitor ⌬⌿ m , DH-II-24loaded cells were irradiated with the built-in halogen lamp and incubated with 2 g/ml JC-1 dye for 30 min at 37°C. Fluorescence intensities of the cells were determined at the single-cell level using FluoView FV300 software with a confocal microscope. Data are expressed as relative fluorescence intensities and the ratio of the fluorescence intensity of monomers to that of J-aggregates.
Immunostaining and Confocal Microscopy-Apoptotic cells were identified by confocal microscopy with the BD Pharmingen FITC Caspase-3, Active Form, mAb Apoptosis Kit I according to the manufacturer's instructions. For quantification of apoptosis, at least 300 cells were selected from three independent experiments, and the percentages of apoptotic cells were calculated.
To analyze AIF translocation, cells were fixed with 3.7% formaldehyde in PBS for 30 min, permeabilized with 0.2% Triton X-100 in PBS for 30 min, and incubated with blocking solution (2% BSA in TBS containing 0.1% Triton X-100) for 30 min. Cells were incubated with anti-AIF antibody in blocking solution, probed with FITC-conjugated anti-mouse IgG, and further incubated with 10 g/ml propidium iodide. Stained samples were observed by confocal microscopy. Approximately 20 cells were randomly selected from each of three independent experiments, and the fluorescence intensities of the cells were measured by processing only their FITC images at the single-cell level using FluoView FV300 software. AIF translocation was expressed as the ratio of the AIF fluorescence intensity of the nucleus to that of the whole cell.
Calpain Activity Assay-The in vitro activity of calpain (Promega, Madison, WI) was measured according to the manufacturer's instructions. Briefly, cell extracts were obtained using lysis buffer containing 50 mM HEPES (pH 7.5), 1% Triton X-100, 150 mM NaCl, and 1 mM EDTA and incubated for 15 min with equal volumes of Calpain-Glo TM reagent, including succinyl-LLVY-aminoluciferin (a succinyl proluminescent calpain substrate), and 2 mM CaCl 2 (final concentration of 1 mM). The luminescence of each sample was recorded in relative light units by a microplate luminescence reader (Molecular Devices, Sunnyvale, CA). Calpain activity was determined by comparing the luminescence intensities of irradiated cells with those of non-irradiated control cells (-fold differences).
Statistical Analysis-Data are expressed as means Ϯ S.D. Statistical significance was analyzed using one-way analysis of variance (Origin 6.1 software, OriginLab Corp., Northampton, MA). p values of Ͻ0.05 were considered statistically significant.

RESULTS
Role of TG2 in Stimulation of Caspase-dependent Apoptotic Pathway by DH-II-24-mediated PDT-Because PDT-induced apoptotic death of AGS cells may be prevented by inhibiting caspase activation (20), we asked whether DH-II-24-mediated PDT induces caspase-dependent apoptotic cell death. The contribution of well characterized apoptotic factors to the cytotoxic response to PDT was assessed by Western blotting (Fig.  1A). PDT resulted in a release of cytochrome c from the mitochondria to the cytosol in a time-dependent manner. This event was accompanied by procaspase-9 cleavage into active caspase-9. In addition, the activation of caspase-3 and the

TG2/Calpain/Bax-dependent Apoptotic Cell Death
cleaved product of poly(ADP-ribose) polymerase were detected in a similar time-dependent manner. However, TG2 expression did not significantly change during this period (Fig. 1A). Using a FITC-conjugated active caspase-3 monoclonal antibody apoptosis kit, we further demonstrated that PDT stimulated caspase-3 in a time-dependent manner, with maximal stimulation at 12 h (Fig. 1B). These results suggest that DH-II-24-mediated PDT stimulates caspase-dependent apoptotic cell death in AGS cells.
The potential role of TG2 in the PDT-induced activation of caspase-3 was examined using various inhibitors and TG2 siRNA. Because intracellular ROS and Ca 2ϩ are required for the activation of TG2 by DH-II-24-mediated PDT in AGS cells (20), we used a ROS scavenger, Trolox, and a Ca 2ϩ chelator, BAPTA-AM. The PDT activation of caspase-3 was completely abrogated by Trolox and was significantly inhibited by BAPTA-AM (Fig. 1C). Consistent with a previous report (22), caspase-3 activation was prevented by cystamine. As expected, the broad-spectrum caspase inhibitor Z-VAD-fmk completely inhibited caspase-3 activation. However, the inhibitor had no effect on the activation of TG2 by PDT (data not shown). In addition, the activation of caspase-3 by PDT was suppressed by TG2 siRNA in a dose-dependent manner, but transfection with TG2 siRNA did not affect the basal caspase-3 activity (Fig. 1D). Control siRNA had also no effect on the activation of caspase-3 by PDT. Transfection with TG2 siRNA suppressed the expression of TG2 in a dose-dependent manner (Fig. 1E). Therefore, TG2 is required for caspase-dependent apoptotic cell death induced by DH-II-24-mediated PDT.
Role of TG2 in AIF-mediated Apoptotic Cell Death Induced by DH-II-24-mediated PDT-Although apoptotic cell death caused by DH-II-24-mediated PDT is caspase-dependent, it may also involve other apoptotic factors, such as AIF, because cell shrinkage was observed in a fraction of cells treated with Z-VAD-fmk (data not shown). To verify this possibility, we investigated whether PDT could induce AIF release from the mitochondria into the cytosol. PDT resulted in AIF release into the cytosol in a time-dependent manner, but AIF expression levels were unaffected during this period ( Fig. 2A). In addition, nuclear translocation of AIF was confirmed by confocal microscopy (Fig. 2B).
We examined the possible role of TG2 in PDT-induced nuclear translocation of AIF by analyzing the effects of various inhibitors on the ratio of the AIF fluorescence intensity in the nucleus to that in the whole cell (Fig. 2C). Nuclear translocation of AIF was inhibited by Trolox, BAPTA-AM, and cystamine, but not by Z-VAD-fmk. Additionally, AIF nuclear translocation was prevented by TG2 siRNA in a dose-dependent manner (Fig.  2D). Thus, TG2 is involved in the nuclear translocation of AIF by DH-II-24-mediated PDT.
To determine whether PDT induces AIF-mediated apoptotic cell death, AGS cells were transfected with AIF siRNA. Transfection with AIF siRNA strongly suppressed the expression of AIF in a dose-dependent manner (Fig. 3A). However, AIF siRNA had no inhibitory effects on TG2 activation by PDT, whereas Trolox, BAPTA-AM, and cystamine consistently inhibited TG2 activation (Fig. 3B). We studied the effects of AIF siRNA on apoptotic cell death induced by PDT. Transfection

TG2/Calpain/Bax-dependent Apoptotic Cell Death
with AIF siRNA significantly inhibited the decrease in cell viability caused by PDT in a dose-dependent manner (Fig. 3C). Interestingly, the combination of Z-VAD-fmk and AIF siRNA enhanced the inhibition of decreased cell viability induced by PDT (Fig. 3D). Thus, DH-II-24-mediated PDT induces apoptotic cell death through both caspase-dependent and AIF-dependent pathways involving TG2 activation. (20), and PDT induced the release of cytochrome c (Fig. 1A) and AIF ( Fig. 2A) from the mitochondria into the cytosol, suggesting that PDT may damage mitochondria. Thus, we investigated whether a loss of ⌬⌿ m is associated with PDT-induced apoptotic cell death. The loss of ⌬⌿ m increased with time following PDT; the amount of green fluorescent monomers increased, whereas J-aggregates with red fluorescence mostly disappeared in a time-dependent manner (Fig. 4A).

Loss of ⌬⌿ m Induced by DH-II-24-mediated PDT-DH-II-24 accumulates in the mitochondria of AGS cells
We evaluated a possible role for TG2 in the PDT-induced loss of ⌬⌿ m by calculating the ratio of monomers to J-aggregates. The loss of ⌬⌿ m was inhibited by Trolox, BAPTA-AM, and cystamine, but not by Z-VAD-fmk (Fig. 4B), indicating a possible role of TG2. TG2 siRNA also prevented the loss of ⌬⌿ m in a dose-dependent manner (Fig. 4C). These results demonstrate that DH-II-24-mediated PDT induces apoptotic cell death through a mitochondria-mediated intrinsic pathway that is regulated by TG2.
Calpain Is Involved in Apoptotic Cell Death Induced by DH-II-24-mediated PDT-Calpain is involved in AIF release from mitochondria (23,24), and PDT induced AIF-dependent apoptotic death of AGS cells (Fig. 3). Thus, we examined whether calpain is activated following DH-II-24-mediated PDT. PDT activated calpain in a time-dependent manner, with maximal activation at 3 h (Fig. 5A). As expected, a selective calpain inhibitor, PD150606, inhibited the activation of calpain, but not TG2, by PDT (Fig. 5B). The activation of calpain by PDT was significantly prevented by Trolox, BAPTA-AM, and TG2 siRNA, but not by Z-VAD-fmk (Fig. 5C), indicating that TG2 functions as an upstream signal of calpain.
To determine whether calpain is involved in TG2-dependent apoptotic responses to PDT, we analyzed the effects of PD150606 on the activation of caspase-3 and the translocation of AIF into the nucleus following PDT. Treatment with PD150606 effectively abrogated both caspase-3 activation and AIF translocation in a dose-dependent manner (Fig. 5D). Therefore, calpain is involved in the TG2-mediated death mechanism induced by DH-II-24-mediated PDT. The role of calpain was further supported by data indicating that PD150606 significantly inhibited the loss of ⌬⌿ m (Fig. 5E).

TG2/Calpain/Bax-dependent Apoptotic Cell Death
Additionally, the decrease in cell viability induced by PDT was sufficiently blocked in a dose-dependent manner by PD150606 (Fig. 5F). Thus, calpain participates in TG2-dependent apoptotic cell death through caspase-and AIF-dependent pathways in response to DH-II-24-mediated PDT.
DH-II-24-mediated PDT Induces Translocation of Bax-We asked whether Bax is involved in apoptotic cell death induced by DH-II-24-mediated PDT. PDT induced the translocation of Bax from the cytosol to mitochondria (Fig. 6A). Bax translocation was significantly blocked by Trolox and TG2 siRNA. In addition, the translocation of Bax was inhibited by the calpain inhibitor PD150606. These results demonstrate that TG2 and calpain are involved in PDT-induced translocation of Bax to the mitochondria.
To confirm the role of TG2 in Bax translocation, we expressed Bax-EGFP and Cox8-DsRed in AGS cells and analyzed the effect of TG2 siRNA on the PDT-induced translocation of Bax by confocal microscopy. Cox8-DsRed was used as a mitochondrial marker (25). Bax-EGFP was co-localized with Cox8-DsRed following PDT, whereas its translocation was inhibited by TG2 siRNA (Fig. 6B). Considering the role of TG2 in calpain activation and AIF translocation following PDT (Figs. 2 and 5), these results demonstrate that DH-II-24-mediated PDT induces Bax translocation from the cytosol to mitochondria via activation of TG2 and calpain. Bax translocation to mitochondria results in mitochondrial permeability transition, which leads to a disruption in ⌬⌿ m and subsequently induces a mitochondria-dependent death pathway that includes caspases  APRIL
Apoptotic Cell Death Mediated by TG2-dependent Mechanism in Human Bladder Carcinoma Cells-To further explore a broad relevance of the preceding results, we investigated the apoptotic signaling pathway induced by DH-II-24-mediated PDT in human bladder carcinoma J82 cells. Initially, we exam-ined the effect of PDT on cell viability. As shown in Fig. 7A, PDT decreased the viability of J82 cells in a post-irradiation time-dependent manner, and both DH-II-24 alone and light alone had no effect on cell viability (inset).
Next, we investigated the effect of various inhibitors and TG2 siRNA on TG2 activation by DH-II-24-mediated PDT. The activation of TG2 by PDT was inhibited by Trolox, BAPTA-AM, and TG2 siRNA, but not by Z-VAD-fmk and PD150606 (Fig. 7B). In contrast, transfection with TG2 siRNA prevented  the activation of calpain by PDT (Fig. 7C), indicating that TG2 functions as an upstream signal of calpain. As expected, PD150606 inhibited calpain activation. Moreover, TG2 siRNA and PD150606 effectively abrogated both caspase-3 activation and AIF translocation (Fig. 7D) and the decrease in cell viability (Fig. 7E) by PDT. PDT-induced cell death was also prevented by AIF siRNA and Z-VAD-fmk (Fig. 7, E and F). Especially, a combination of Z-VAD-fmk and AIF siRNA enhanced the inhibition of decreased cell viability by PDT, as it did in AGS cells. Thus, DH-II-24-mediated PDT induces apoptotic cell death through a TG2-dependent mechanism that leads to both the caspase-and AIF-dependent pathways.

DISCUSSION
PDT is a therapeutic modality based on photochemical reactions that require selective uptake of a light-activatable photosensitizer by tumor cells, followed by irradiation with light of an appropriate wavelength to activate the photosensitizer (30). Activation of a photosensitizer catalyzes the localized produc-tion of ROS that lead to cell death (13,31). Although PDT usually induces apoptosis through the caspase cascade, PDT can stimulate the caspase-independent death pathway (13). PDT induces the release of cytochrome c and AIF (17)(18)(19), which can cause caspase-dependent and caspase-independent apoptotic cell death, respectively. Several studies have shown that calpain promotes the release of AIF (23,24), which may result from the translocation of Bax into mitochondria (29). TG2 can induce apoptosis in an AIF-dependent manner in pancreatic cancer cells (12). Moreover, TG2 can activate calpain, triggering ionomycin-induced death of erythrocytes (32). However, the methodical signaling pathway by which these proteins signal apoptosis remains unclear.
TG2 is a multifunctional protein implicated in a number of cellular events, including apoptosis and cell adhesion, survival, migration, and invasion (1-3). TG2 overexpression and activation have been linked to induction of apoptosis (3,11,12,33). Recently, we reported a possible role of TG2 in the PDT-induced apoptotic death of cancer cells (20). However, the detailed mechanism of this apoptotic cell death has not been clearly elucidated. This study clearly demonstrated the essential role of TG2 in apoptotic cell death mechanisms induced by DH-II-24-mediated PDT in human cancer cells. PDT induced apoptotic death of cancer cells through the activation of both caspase-dependent and caspase-independent pathways and required TG2 activation. This process used a TG2-dependent mechanism that involved calpain-mediated translocation of Bax to mitochondria, which resulted in a loss of ⌬⌿ m and the release of both cytochrome c and AIF (Fig. 8).
PDT induces intracellular ROS production and elevates intracellular Ca 2ϩ in various cell types. These messengers stimulate several cytotoxic factors (13,15). We showed previously that intracellular ROS are required for TG2 activation in fibroblasts (21,34) and also demonstrated a regulation mechanism of TG2 by intracellular Ca 2ϩ in AGS cells (35). Consistent with these findings, we demonstrated that TG2 activation by DH-II-24-mediated PDT was mediated by intracellular ROS and Ca 2ϩ . The ROS scavenger Trolox completely inhibited TG2 activation by PDT, and partial TG2 inhibition by the Ca 2ϩ chelator BAPTA-AM was also observed (20). Furthermore, apoptotic cell death induced by PDT was inhibited by Trolox and partially inhibited by BAPTA-AM (20). Thus, intracellular ROS function as major mediators in PDT-induced intracellular signaling that leads to TG2 activation and apoptotic cell death (Fig. 8).
What is a possible mechanism by which intracellular ROS activate the Ca 2ϩ -dependent enzyme TG2? First, direct activation of TG2 was investigated by measuring in vitro TG2 activity in the presence of exogenous H 2 O 2 . However, in vitro TG2 activity was not affected by exogenous H 2 O 2 (data not shown), indicating that ROS have no direct effect on TG2 activity in an extracellular environment. Another possible mechanism is that intracellular ROS activate TG2 by regulating other factors, including intracellular Ca 2ϩ levels. Intracellular ROS mediate the elevation of intracellular Ca 2ϩ levels in fibroblasts and skeletal muscle (36,37). Oxidative stress can activate TG2 by increasing intracellular Ca 2ϩ concentrations and by releasing TGF␤2 (38). We observed that Trolox partially suppressed PDT-induced increases in intracellular Ca 2ϩ levels, whereas FIGURE 8. Apoptotic cell death by a TG2/calpain/Bax-dependent mechanism involving both caspase-dependent and caspase-independent pathways. Solid arrows indicate apoptotic cell death signaling pathways induced by DH-II-24-mediated PDT as identified in this study. Dotted arrows represent signaling pathways demonstrated previously (13,15,18). Intracellular ROS (rather than intracellular Ca 2ϩ ) are major mediators of the cytotoxic response to DH-II-24-mediated PDT (20). DH-II-24-mediated PDT results in apoptotic cell death potentially through mitochondria-dependent death pathways that are triggered by TG2 activation. TG2 activation facilitates calpain-mediated Bax translocation to the mitochondria, where Bax may cause the collapse of the ⌬⌿ m . The loss of ⌬⌿ m allows for the release of cytochrome c (Cyto c) and AIF and subsequently induces apoptotic cell death through a synergic effect. MOMP, mitochondrial outer membrane permeabilization.

TG2/Calpain/Bax-dependent Apoptotic Cell Death
BAPTA-AM had no inhibitory effect on the production of ROS following PDT (20). Moreover, incubation with BAPTA-AM partially inhibited apoptotic cell death induced by PDT (20), indicating that intracellular Ca 2ϩ participates, in part, in the cellular responses to PDT. Thus, intracellular ROS are likely to activate TG2 by regulating the levels of intracellular Ca 2ϩ and many other factors. It remains necessary to elucidate the exact mechanism of TG2 activation by intracellular ROS during PDT.
In conclusion, we have demonstrated that, following DH-II-24-mediated PDT, TG2 plays a key role in the apoptotic death of cancer cells through a signaling cascade that includes calpain activation and Bax translocation, resulting in mitochondrial dysfunction. The mitochondrial dysfunction stimulates both caspase-dependent and AIF-mediated pathways. These findings contribute to the understanding of the complex cell death pathways induced by PDT and to the improvement of PDT treatment of cancers.