IL-24 Is ... - Cell Press

ARTICLE

doi:10.1016/j.ymthe.2003.11.014

The Tumor Suppressor Activity of MDA-7/IL-24 Is Mediated by Intracellular Protein Expression in NSCLC Cells Kerry A. Sieger,1,* Abner M. Mhashilkar,1,* Alexis Stewart,1 R. Bryan Sutton,2 Randall W. Strube,3 Si Yi Chen,3 Abujiang Pataer,4 Stephen G. Swisher,4 Elizabeth A. Grimm,5 Rajagopal Ramesh,4 and Sunil Chada1,5,y 1

Introgen Therapeutics, Inc., Houston, TX 77030, USA Department of Physiology & Biophysics, The University of Texas Medical Branch, Galveston, TX 77555, USA 3 Department of Cell and Gene Therapy, Baylor College of Medicine, Houston, TX 77030, USA 4 Department of Thoracic and Cardiovascular Surgery, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030, USA 5 Department of Bioimmunotherapy, The University of Texas M. D. Anderson Cancer Center, Houston, TX 77030, USA 2

*These authors contributed equally to this article. y

To whom correspondence and reprint requests should be addressed. Fax: (713) 797-1349. E-mail: [email protected].

mda-7/IL-24 (HGMW-approved symbol IL24) is a tumor suppressor gene whose expression is lost during tumor progression. Gene transfer using adenoviral mda-7/IL-24 (Ad-mda7) exhibits minimal toxicity on normal cells while inducing potent apoptosis in a variety of cancer cell lines. Ad-mda7transduced cells express high levels of MDA-7 protein intracellularly and also secrete a soluble form of MDA-7 protein. In this study, we sought to determine whether the intracellular or secreted MDA7 protein was responsible for anti-tumor activity in H1299 lung tumor cells. Ad-mda7 transduction of lung tumor cells increased expression of stress-related proteins, including BiP, GADD34, PP2A, caspases 7 and 12, and XBP-1, consistent with activation of the UPR pathway, a key sensor of endoplasmic reticulum (ER)-mediated stress. Blocking secretion of MDA-7 did not inhibit apoptosis, demostrating that intracellular MDA-7 was responsible for cytotoxicity. Consistent with this result, when applied directly to lung cancer cells, soluble MDA-7 protein exhibited minimal cytotoxic effect. We then generated mda-7 expression constructs using vectors that target the expressed protein to various subcellular compartments, including cytoplasm, nucleus, and ER. Only full-length and ER-targeted MDA-7 elicited cell death in tumor cells. Thus in lung cancer cells, Ad-mda7 activates the UPR stress pathway and induces apoptosis via intracellular MDA-7 expression in the secretory pathway. Key Words: mda-7, IL-24, apoptosis, ER, UPR, stress, cytokine, caspase, targeting, secretion, adenovirus, cancer gene therapy

INTRODUCTION Melanoma differentiation-associated gene-7 (mda-7; HGMW-approved symbol IL24) is a tumor suppressor gene that was identified using a subtractive hybridization approach from human melanoma cells induced to differentiate with IFN-h and mezerein [1]. mda-7 mRNA was expressed in normal melanocytes and early stages of melanoma, but was lost during melanoma progression [1]. Transfection of mda-7 expression plasmid into various tumor cell lines resulted in suppression of growth [2]. Based on these observations it was suggested that mda-7 is a novel tumor suppressor gene whose expression must be inhibited for tumor progression. The tumor suppressor

MOLECULAR THERAPY Vol. 9, No. 3, March 2004 Copyright B The American Society of Gene Therapy 1525-0016/$30.00

activities of mda-7 gene transfer have now been well established. Numerous studies have documented the growth suppression effected by elevated expression of MDA-7 protein in cancer cells from lung, breast, melanoma, prostate, mesothelioma, glioma, etc. (reviewed in [3,4]). MDA-7 overexpression in normal cells, however, lacks the cytotoxic effects seen in cancer cells [3,5,6]. We have recently demonstrated that adenoviral mda-7/IL-24 (Ad-mda7) causes growth suppression, cell cycle block, and apoptosis in non-small-cell lung cancer cells in vitro and in vivo using xenograft models [7 – 9]. The mda-7 gene is located on chromosome 1 at 1q32 and is contained within a cytokine cluster that encodes IL-10, IL-19, and IL-20 [5,10,11]. Based upon its chromosomal localization

355

ARTICLE

in the IL-10 family cluster and its limited amino acid identity to IL-10, mda-7 has now been designated IL24, and recent data indicate that this molecule functions as a pro-Th1 cytokine in human PBMC [12]. Further studies have demonstrated expression of mda-7 mRNA and MDA7 protein in immune cells [11,12]. Thus mda-7 is an unusual pro-Th1 cytokine with proapoptotic activity. A number of studies have evaluated molecules involved in mediating the apoptotic response of tumor cells to Admda7. These studies have used cell lines from different tumor types and have shown regulation of a variety of diverse cellular mediators. For example, analysis of molecules involved in the effector phase of the apoptotic pathway has shown that MDA-7 can increase expression of p53, BAK, and BAX; activate caspases 3, 8, and 9; and increase mitochondrial cytochrome c release in lung and breast cell lines [5 – 9,13]. Analyses of signaling pathways have revealed Ad-mda7 regulation of iNOS and MAPK in melanoma [14,15] and JUN kinase, h-catenin, PI3K, and PKR in lung and breast tumor cells [9,16,17]. Thus, different signaling pathways appear to be responsible for induction of apoptosis in different tumor types. This concept was recently validated by demonstrating that MDA-7mediated killing of breast cancer cells was blocked by inhibitors of MAPK or MEKK pathways, whereas blocking these pathways had minimal effects against lung tumor cells [17]. Recent studies have demonstrated that Ad-mda7 transduction causes high levels of MDA-7 protein to be released from transduced cells [5]. The role of this soluble human MDA-7 in the apoptotic process has not been explored. This study was designed to investigate whether the potent proapoptotic activity observed with Ad-mda7 is due to the intracellular or the secreted MDA-7 protein. We show that Ad-mda7 transduction of lung cancer cells results in activation of the unfolded protein response (UPR), a classic mediator of ER (endoplasmic reticulum) stress [18]. Inhibition of secretion of MDA-7 protein did not block apoptotic activity of Ad-mda7. We demonstrate that MDA-7 expression in the ER elicits potent cell death in tumor cells and thus it appears necessary for MDA-7 to enter the secretory pathway for apoptosis induction in receptor-negative lung cancer cells.

RESULTS Ad-mda7 Induces Apoptosis and Secretion of MDA-7 Protein The mda-7 coding region possesses an unusually long leader sequence (48 amino acids). Analysis of the predicted primary amino acid sequence indicates that the MDA-7 protein contains a prototypic signal sequence, which is likely responsible for directing secretion of the protein (Fig. 1A). The protein also possesses three canonical N-glycosylation sites. Immunohistochemical analyses of H1299 cells treated with Ad-mda7 demonstrate

356

doi:10.1016/j.ymthe.2003.11.014

MDA-7 expression in secretory granules [5]. MDA-7 expression colocalizes with BiP, a marker for the secretory apparatus (data not shown). Inspection of the translated protein product demonstrates a strongly hydrophobic region at the N-terminus [12], and analysis of the primary amino acid sequence predicts cleavage of MDA-7 at amino acid 48, resulting in the secretion of the remaining (mature) protein product of amino acids 49 – 206. We transduced H1299 NSCLC tumor cells with Admda7 or Ad-luc and analyzed them for cytotoxicity. H1299 cells were killed in a temporal- and dose-dependent manner by Ad-mda7 (Fig. 1B). The control Ad-luc vector had no effect on cell viability, whereas even low doses (1000 vp/cell; 40 pfu/cell) of Ad-mda7 caused cell death within 24 h. Treatment of H1299 cells with Admda7 vector causes high levels of intracellular MDA-7 protein as well as release of MDA-7 protein (>100 ng/ml) into the supernatant (Fig. 1C). The soluble MDA-7 is of higher molecular weight than the intracellular form. We have sequenced the MDA-7 protein released from Admda7-transduced cells and have found that the secreted protein starts at amino acid 49, as predicted. To demonstrate that the MDA-7 protein in the supernatant is derived from an active secretion process rather than resulting from cell lysis, we probed the Western blot shown in Fig. 1C with an antibody against h-actin. Immunoreactive h-actin was not detected in the supernatant samples, whereas strong signals were obtained from cell lysates. Thus it is very unlikely that the MDA7 protein is released from cells by a passive mechanism. Ad-mda7 Activates Stress Proteins Associated with UPR Previous studies have demonstrated that Ad-mda7 transduction of tumor cells results in activation of molecules involved in apoptosis and death signaling pathways (p53, BAX, BAK, TRAIL, c-JUN, JNK, caspases, p38, PI3K, and hcatenin) [5,7 – 9,15 – 17]. Microarray analyses performed on Ad-mda7-treated H1299 cells indicated activation of a number of additional stress-related genes, especially those involved in regulation of protein folding and secretion (Table 1). We noted strong regulation, at the transcriptional level, of PP2A, HSP 70-5 (BiP), HSJ1, and TRA1. We then evaluated protein expression of various stress-related molecules (BiP, GADD34, PP2A) after Admda7 transduction of H1299 cells (Fig. 2A). Western blot analyses demonstrated consistent increases in steadystate levels of proteins for BiP, GADD34, and PP2A after MDA-7 expression. These proteins are implicated in activation of the mammalian stress response known as UPR [19]. We therefore analyzed additional members of the UPR pathway: caspases 7 and 12 and XBP-1. As shown in Fig. 2B, expression of these UPR-associated proteins is upregulated after Ad-mda7 transduction, suggesting that UPR activation was the mechanism by which MDA-7 kills cancer cells. We also evaluated expression of PERK, another protein characteristic of UPR activation, but did not

MOLECULAR THERAPY Vol. 9, No. 3, March 2004 Copyright B The American Society of Gene Therapy

ARTICLE

doi:10.1016/j.ymthe.2003.11.014

FIG. 1. MDA-7 kills lung tumor cells and is secreted. (A) Schematic of the MDA-7 protein. The large signal sequence (amino acids 1 – 48), hydrophobic region (stippled), putative cleavage site (arrow), and potential N-glycosylation sites (*) are indicated. (B) Ad-mda7 potently kills lung tumor cells. H1299 cells were treated with increasing doses (vp/cell) of Ad-mda7 or Ad-luc and analyzed at the indicated times for cell killing. Ad-mda7 caused a dose- and time-dependent loss of viability compared to Ad-luc. Results are shown as means + SD of triplicate samples. (C) MDA-7 is secreted from H1299 cells. H1299 cells were transduced with 1000 vp/cell of Ad-mda7 or Ad-luc (middle lane, L), and cell lysates (left; days 1 – 4) and supernatants (right; days 1 – 4) were analyzed by Western blot using anti-MDA-7 antibody. The intracellular MDA-7 protein is shown by open arrow and secreted MDA-7 indicated by filled arrow.

find detectable levels of PERK in H1299 cells. Previous studies have indicated that PERK is expressed in secretory cells, such as h-islet cells [20], and thus the lack of PERK expression in NSCLC H1299 cells is not surprising. We have previously shown that Ad-mda7 transduction of tumor cells causes activation of mitochondrial caspases 3 and 9 [7]. Fig. 2C shows that Ad-mda7 transduction of

H1299 cells results in activation (via cleavage) of caspase 7, an important caspase mediating UPR apoptotic responses. MDA-7 Is Heavily Glycosylated The higher molecular weight of the secreted MDA-7 protein, as seen in Fig. 1C, is suggestive of glycosylation

TABLE 1: MDA-7 regulates stress genes involved in protein folding Gene

Function

hsp 70-5 hsp 90 tra 1 hsj1 hsp 60 hsp 70-1 protein kinase inhibitor p58 pp2a

BiP/grp 78 ATP-regulated molecular chaperone ER heat-shock protein (grp 94/gp 96) Heat-shock protein Chaperonin Chaperone Overexpression leads to tumor formation in mice Protein phosphatase 2A; Ser/Thr phosphatase

mRNA changea z12.8 # 2.0 z8.4 z7.7 z1.4 z1.4 # 2.7 z2.8

Protein changeb z5 z 0.05), although both drugs potently blocked endostatin secretion and caused intracellular accumulation of endostatin protein (data not shown). Thus blocking MDA-7 secretion does not abrogate cell killing and apoptosis induction, whereas blocking endostatin secretion has no effect on cell viability. These results suggested that it was the intracellular and not the secreted form of MDA-7 that was primarily responsible for eliciting cell death in H1299 cells. We further tested this hypothesis by adding secreted MDA-7 protein to naı¨ve lung tumor cells and monitoring cell death. Initial studies adding supernatant from Ad-mda7transduced cells did indicate cell death; however, careful analysis demonstrated that the cell death was caused by residual Ad-mda7 vector in the culture supernatant. When we treated culture supernatants to minimize Admda7 contamination, negligible cell death was induced by MDA-7-containing supernatants. To eliminate any confounding aspect of Ad vector toxicity, we repeated this study using MDA-7-containing supernatant obtained from stably expressing 293-mda7 cells (Fig. 4B). Insignificant levels of cell death were seen (50% cell death. When we added anti-MDA-7 neutralizing antibody [3] to Ad-mda7-infected cultures, no inhibition of cell killing was observed (data not shown), further demonstrating that the primary killing activity from Ad-mda7-treated cultures was due to intracellular protein. MDA-7 protein made in insect cells using baculovirus vectors or bacterially expressed MDA-7 protein failed to induce significant killing in H1299 cells (Fig. 4B). An additional control included treating H1299 cells with MDA-7 protein and then transducing with Ad-luc to evaluate any interaction between Ad vector signaling and MDA-7—only background killing was observed, further indicating that the cytotoxicity in Ad-mda7-treated cultures was due to intracellular MDA-7 protein. Subcellular Targeting of MDA-7 We wished to investigate MDA-7 was being released into the cytosol or nucleus during supraphysiologic expression in Ad-mda7-infected cells and whether this was

359

ARTICLE

doi:10.1016/j.ymthe.2003.11.014

FIG. 4. Inhibition of secretion does not abrogate Ad-mda7 cytotoxicity. (A) H1299 cells were treated with Ad-mda7 (M), Ad-luc (L), or Ad-endostain (E) in the presence or absence of brefeldin A (BFA or B) or tunicamycin (TUN or T) and cell viability was assessed. Data are shown as means + SD of triplicate samples. (B) Intracellular MDA-7 kills H1299 cells. Supernatant from 293mda7 cells (M) was applied to H1299 cells at 10 and 100 ng/ml and cell viability monitored after 3 days. r.MDA-7 protein prepared from Escherichia coli (EM; 20 – 100 ng/ml) and from baculovirus cultures (BacM; 100 ng/ml) was evaluated for cell toxicity. Ad-mda7 and Ad-luc (2000 vp/cell) were included as controls. Ad-mda7 induced cell killing, which was not inhibited by addition of neutralizing anti-MDA-7 antibody (data not shown). r.MDA-7 supernatant did not induce cell death. Data are shown as means + SD of triplicate samples.

responsible for inducing death. To investigate the effects of subcellular localization of MDA-7 protein on cell viability, we constructed expression vectors designed to target MDA-7 expression to different subcellular compartments. In constructing these retargeting vectors, the secretion signal sequence in the mda-7 cDNA was first deleted and an ATG inserted to generate an initiator codon. As shown in Fig. 5, the nuclear targeting vector contains three nuclear localization signals, the ER targeting vector contains an ER signal sequence and ER retention signal, and the cytoplasmic targeting vector contains no targeting signals, allowing the default expression of proteins in the cytoplasm. The full-length (FL) plasmid uses the cytoplasmic-targeting vector backbone but contains full-length mda-7 cDNA. All proteins expressed by these plasmids also contain a myc tag (2 kDa) at the C-terminus. We transiently transfected the vectors into H1299 cells and analyzed targeted MDA-7 protein expression by immunocytochemistry. As shown in Fig. 6A, each vector successfully targets MDA-7 protein to the intended subcellular compartment. The MDA-7 protein expressed from

360

the full-length plasmid can be seen to be in secretory granules within the cell, similar to the results observed after Ad-mda7 transduction (Fig. 6A). We confirmed the

FIG. 5. Subcellular targeting vectors for MDA-7 protein expression. Schematic of plasmid vectors used to retarget MDA-7 protein to subcellular compartments. The MDA-7 signal sequence (aa 1 – 48), extracellular protein (aa 49 – 206; 00processed protein00 ), myc tag, nuclear localization signal (NLS), and endoplasmic reticulum targeting sequences (ER signal peptide and ER retention signal) are indicated. The predicted sizes of the proteins are indicated. FL, full length; Cyto, cytoplasmic; Nuc, nuclear; ER, endoplasmic reticulum-targeted MDA-7.


doi:10.1016/j.ymthe.2003.11.014

ARTICLE

FIG. 6. Subcellular expression and activity of MDA-7. (A) Subcellular localization of targeted MDA-7 proteins. H1299 cells were transiently transfected with the indicated plasmid or Ad vectors and 48 h later, cells were fixed and immunostained for MDA-7 expression using anti-MDA-7 antibody. The retargeted proteins demonstrate appropriate subcellular localization. (B) Protein expression and secretion by retargeted MDA-7 proteins. H1299 cells were transiently transfected with the indicated plasmid or Ad vector and 48 h later, cell lysates and supernatants were analyzed for MDA-7 expression by Western blot. The top shows results from supernatants (filled arrow) and the bottom indicates MDA-7 expression from cell lysates (open arrows). MDA-7 is secreted only from full-length- and Admda7-treated cells. Arrows indicate size of MDA-7 proteins produced by Ad-mda7-treated H1299 cells. Note FL MDA-7 bands are increased due to myc tag. (C) H1299 cells were transiently transfected with targeting plasmids and analyzed for cell viability using the Live/Dead assay. Green fluorescence indicates viable cells and red fluorescence indicates dead cells (original magnification 100). Only FL and ER-targeted plasmids induce cell death. Ad-mda7 and Ad-luc (2000 vp/cell) are shown as controls (original magnification 40).


361

ARTICLE

precise subcellular localization of the targeted proteins by comparison with expression patterns of molecules known to reside in these compartments. For example, nucleartargeted MDA-7 colocalized with Hoechst staining of nuclei and ER-targeted MDA-7 costained with BiP/grp78 (data not shown). We also compared subcellular localization to control GFP vectors targeted to the cytoplasm, nucleus, and ER. The targeted GFP expression patterns were identical to the targeted MDA-7 staining patterns. Fig. 6A shows that each vector successfully promotes the expression of MDA-7 protein within the correct cellular compartment, while only the full-length mda-7 cDNA, which includes the N-terminal secretion signal, permits secretion of MDA-7 protein into the medium (Fig. 6B). As predicted, the ER-targeted MDA-7 protein was not secreted as it contained an ER-retention signal. Addition of the myc tag did not adversely affect MDA7 protein stability, as a control full-length mda-7 expression plasmid (without a myc tag) expressed comparable

doi:10.1016/j.ymthe.2003.11.014

steady-state levels of MDA-7 protein. The myc tag did not appear to interfere with protein processing or secretion, as the full-length myc-tagged protein showed two intracellular MDA-7 bands similar to the full-length MDA-7 protein expressed by full-length plasmid or Ad-mda7 (Fig. 6B): note that the FL bands are larger due to the myc tag. The myc-tagged MDA-7 appeared to be secreted and glycosylated similar to native MDA-7 from Ad-mda7treated cells (Fig. 6B). Subcellular Localization of MDA-7 Determines Cytotoxicity We used Hoechst staining of nuclei to evaluate cytotoxic effects of retargeted MDA-7 expression. Nuclear- or cytoplasmic-targeted MDA-7 expression had no effect on nuclear morphology. Cells containing FL and ER-localized MDA-7 protein, however, had disrupted nuclear morphology indicative of apoptosis. To determine the anti-tumor effects of targeted MDA-7 protein expression,

FIG. 7. MDA-7 must be in secretory pathway to induce apoptosis. (A) PC-3 cells were transfected with the indicated plasmids mixed with pSV2neo, and colony formation assays were performed. After 14 days the numbers of colonies were assessed and compared to control pSV2neo transfection. Data indicate means + SD of triplicate samples. Similar results were seen with H1299, A549, 293, and MCF-7 cells. (B) H1299 cells were transiently transfected with the indicated plasmids and 3 days later assessed for cell viability using a trypan blue assay. Data indicate means + SD of triplicate samples. Only FL and ER-targeted MDA-7 elicit cell killing. (C) H1299 cells were transiently transfected with the indicated plasmids and 48 h later assessed for apoptosis by Hoechst staining. Data indicate means + SD of triplicate samples. (D) H1299 cells were transiently transfected with the indicated plasmids or Ad-mda7 or Ad-luc and 48 h later, lysates were immunoblotted and probed with anti-BiP and anti-a-tubulin antibodies. BiP is induced by full-length and ER-targeted MDA-7.

362


doi:10.1016/j.ymthe.2003.11.014

we evaluated cell viability, apoptosis induction, and colony formation activity in a panel of tumor cell lines. We assessed cell viability using a fluorescence-based assay, which uses calcein staining to identify viable cells (green). When cells die, they become membrane permeant and take up EtBr dye and stain red. Transient transfection of plasmid constructs into H1299 lung cancer cells demonstrated high level expression of targeted MDA-7 proteins (Fig. 6B). No cytotoxicity was observed in mock or cytoplasmic- or nuclear-targeted MDA-7expressing cells. However FL and ER-targeted MDA-7 showed significant cytotoxicity (red cells; P < 0.01 compared to mock or cytoplasmic- or nuclear-targeted constructs)—see Fig. 6C. The targeted mda-7 expression constructs incorporate a neomycin resistance marker and were stably transfected into various tumor cell lines and assessed for colony formation (Fig. 7A). Neither nuclear nor cytoplasmic mda-7 expression constructs had a significant effect ( P > 0.05) on the formation of stable transfectant colonies. Full-length, secreted MDA-7 and ER-targeted MDA-7 caused a reduction in colony formation, indicating the lethality of MDA-7 in these environments. ER-targeted MDA-7 caused a greater reduction in colony formation than FL MDA-7 ( P < 0.01). We observed similar results in H1299 and A549 NSCLC, 293 HEK, and MCF-7 breast cancer cells (data not shown). We then evaluated transiently transfected cells for cell viability and apoptosis induction. H1299 cells were transiently transfected with the MDA-7 targeting plasmids and control GFP targeting plasmids. Only cells expressing FL or ER MDA-7 showed loss of viability and significant cell killing ( P < 0.01; Fig. 7B). We evaluated the mechanism underlying this cytotoxicity by assessing cells for apoptosis. Cells expressing nuclear or cytoplasmic MDA-7 did not exhibit significantly increased apoptosis compared to mock-transfected cells ( P > 0.05; Fig. 7C). Both full-length and ER-targeted MDA-7 showed significantly higher levels of apoptosis than mock, nuc, or cyto vectors ( P < 0.01; Fig. 7C). Thus expression of MDA-7 in the ER or the secretory pathway (FL) results in significant apoptosis induction. The GFP targeted plasmids did not cause cell killing (Fig. 7B) or apoptosis. We then evaluated BiP expression as a marker of UPR in cells transiently transfected with the MDA-7 or GFP targeting plasmids. Only ER-targeted MDA-7 and full-length MDA-7 were able to induce BiP expression in transfected cells. ER-GFP did show some BiP induction, but ER-targeted MDA-7 produced significantly higher BiP induction (Fig. 7D). ER-targeted MDA-7 caused greater BiP induction than FL MDA-7 and was comparable to Ad-mda7.

DISCUSSION Melanoma differentiation-associated gene-7 (mda-7) is a tumor suppressor gene whose expression is lost as tumors


ARTICLE

progress [3,14]. MDA-7 expression in cancer cell lines via adenoviral vector delivery results in suppression of cellular proliferation (Fig. 1B), G2/M cell cycle arrest, and activation of the apoptotic cascade [5,7]. We have demonstrated that MDA-7 protein is actively secreted from transduced cells (Figs. 1 and 3). The goal of this study was to evaluate the role of the secreted MDA-7 protein in apoptosis induction in human NSCLC cells. Microarray analyses identified a number of mRNA species regulated by Ad-mda7 in H1299 cells (Table 1) and supported a role for activation of stress-related proteins prior to apoptosis induction by Ad-mda7. As shown in Table 1, a number of heat-shock-related chaperones are up-regulated by Ad-mda7, including BiP, TRA1, and HSJ1. Note that the major heat shock proteins, HSP 60-1 and HSP 70-1, were not significantly modulated, and HSP 90 was down-regulated, indicating that Ad-mda7 does not induce a global heat shock response, but appears to elicit selective induction of specific stress genes. MDA-7 expression in H1299 NSCLC cells results in activation of various stress proteins, including BiP, GADD34, PP2A, XBP-1, and caspases 7 and 12 (Fig. 2). This spectrum of protein expression is consistent with activation of the UPR stress pathway [23]. UPR is an ER-to-nucleus signal transduction pathway that regulates a variety of target genes responsible for maintaining cellular homeostasis. Thus activation of UPR signaling occurs prior to apoptosis induction by Ad-mda7. The UPR system appears to have evolved as a means for dealing with physiologic stresses; transient stresses such as glucose deprivation or perturbation of calcium or redox homeostasis can result in transient inhibition of protein translation and growth arrest. However, prolonged UPR activation leads to activation of death-related signaling pathways and, ultimately, to apoptotic cell death [23,24]. UPR is mediated by activation of two related signaling cascades. One cascade prevents the assembly of 80S ribosomal initiation complexes, resulting in an inhibition of protein translation [25]. This survival pathway includes the transcription factor XBP-1, which up-regulates expression of ER chaperones [26], the predominant one being BiP/GRP78, an ER luminal chaperone that binds unfolded proteins [27]. A second cascade leads to activation of the ER-localized caspase 12, resulting in initiation of the apoptotic pathway [28]. The interactions between members of molecular stress and death/survival pathways are complex; however, substantial strides have been made in recent years toward defining an ‘‘integrated stress response" [24,29,31]. Regulation of translational initiation by eIF2a appears to be a nodal point for control in a variety of diverse stress pathways [23,30]. Phosphorylation of the a subunit of eIF2 prevents formation of the eIF2/GTP/Met-tRNA complex and thereby inhibits global protein synthesis. MDA-7 expression in H1299 cells results in induction of BiP mRNA (Table 1) and increased BiP and XBP-1

363

ARTICLE

protein expression (Fig. 2). GADD34 is a member of a family of GADD genes that are induced by DNA damage, growth factor deprivation, and other forms of cell stress [32], and its induction mirrors that of other stress-response targets, such as CHOP (GADD154) and BiP [33]. GADD34 induction by ER stress is dependent on the activity of the eIF2a kinases, PERK and GCN2. GADD34 expression appears to be principally associated with the ER, and either GADD34 overexpression or physiologic stress can promote redistribution of PP1 to the ER. Thus GADD34 functions as a regulatory subunit for PP1 and the complex regulates eIF-2a activity [34]. In addition to its role in translational regulation, GADD34 overexpression is known to facilitate apoptosis in mammalian cells following DNA damage and other cell stresses. The phosphatase PP2A can modulate apoptosis at multiple levels by regulating activation of proapoptotic proteins such as BAD and inhibiting antiapoptotic BCL-2 family members in addition to serving as a substrate for caspases [35]. In this study we show that MDA-7 expression increases GADD34 protein and PP2A mRNA and protein levels in H1299 cells (Table 1 and Fig. 2A). In melanoma cells Admda7 transduction induces p38MAPK in concert with induction of GADD34, 45, and 153 [15]. MDA-7 does not have to be actively secreted for antitumor activity since inhibition of secretion did not abrogate MDA-7-mediated killing (Fig. 4A and Table 2). Direct transfer of supernatant from MDA-7-expressing cells to naı¨ve tumor cells did not result in bystander-mediated killing or apoptosis (Fig. 4B). Cell mixing studies (using cells expressing MDA-7) demonstrated that cell – cell contact did not enhance any bystander effect (data not shown). Thus high local concentrations of MDA-7 do not elicit killing of H1299 tumor cells, suggesting that MDA-7-receptor-mediated signaling is not involved in apoptosis of H1299 cells. We then evaluated the role of subcellular localization of MDA-7 in apoptosis induction. MDA-7 was successfully targeted to the cytosol, nucleus, and ER (Fig. 6); however, only ER and full-length (wild type, secreted) MDA-7 were cytotoxic (Fig. 7). This study clearly demonstrates that MDA-7 must be expressed in the secretory pathway to regulate growth control and activate apoptosis. Furthermore, targeting of MDA-7, but not GFP, to the ER recapitulated the growth inhibition and apoptosis induction observed with Ad-mda7 (Fig. 7). Therefore, we hypothesize that MDA-7 activates the UPR stress response and initiates apoptotic signals from within the secretory pathway that ultimately lead to mitochondrial cell death. Additional evidence implicating involvement of the ER in MDA-7 killing is the regulation of the inositol 1,4,5-trisphosphate receptor (IP3R) by Ad-mda7 in H1299 cells [17]. IP3R is an important intracellular calcium (Ca2+)-release channel on the endoplasmic reticulum and is up-regulated within 24 h of Ad-mda7 transduction. Overexpression of IP3R has been implicated in apoptosis

364

doi:10.1016/j.ymthe.2003.11.014

induction in cardiomyocytes. Calcium release by IP3R is modulated by PP1 and PP2A [44]. Recent studies have shown that MDA-7 expression in various tumor lines induces the release of ER calcium pools (manuscript in preparation). Three independent studies have now provided a link between ER stress, calcium signaling, and the apoptotic gateway proteins BAX and BAK in regulation of apoptotic signals from the ER to mitochondria [45 – 47]. Antiapoptotic proteins such as BCL-2 can negatively regulate the flow of Ca2+ from the ER/mitochondrial couple, whereas both BAX and BAK are found in the ER where they can control Ca2+ release. In H1299 cells, MDA-7 induces Ca2+ release, up-regulates apoptotic gateway proteins, and activates mitochondrial caspases. We hypothesize that translocation of BAX and/or BAK from mitochondria to the ER may be involved in MDA-7mediated apoptosis. Ad-mda7 treatment of H1299 cells induces the doublestranded RNA-activated kinase (PKR) and results in increased phosphorylation of both PKR and its target, eIF2a, with a concomitant decrease in protein synthesis [9]. PKR phosphorylates eIF2a in response to viral infection, dsRNA, and other stresses [36]. Other kinases, such as PERK, can also phosphorylate eIF2a in cells undergoing stress from protein misfolding in the ER [37,39]. p58 is an HSP 40 family member known to inhibit PKR by binding to its kinase domain. p58 is induced during UPR and attenuates PERK-mediated eIF2a phosphorylation during ER stress and negatively regulates translation of UPR target proteins BiP and CHOP [38,40]. We find downregulation of p58 mRNA in Ad-mda7-treated cells (Table 1), consistent with the observed up-regulation of PKR and p-PKR [9]. Supraphysiologic expression of MDA-7 induces GADD34, PP2A, PKR, p-PKR, and p-eIF-2a; we propose that this balance of signaling molecules facilitates the apoptosis induction UPR pathway rather than the growth arrest and accommodation pathway. Ad-mda7 transduction of H1299 lung tumor cells results in expression of high levels of MDA-7 protein intracellularly in addition to active secretion of MDA-7 (Figs. 1 and 3). We have previously demonstrated potent apoptosis induction in these cells by Ad-mda7 using both in vitro and in vivo xenograft models [5,7,8]. MDA-7 expression in H1299 cells results in activation of caspases 3, 8, and 9 and PARP and up-regulation of BAK, TRAIL, DR4, p-c-JUN, and JNK1 and release of cytochrome c from mitochondria [5,7 – 9,13,16,17]. This spectrum of apoptosis effectors demonstrates that MDA-7 induces mitochondrial apoptosis. Caspase 12 can activate caspase 9, which in turn activates caspase 3. Upon ER stress, procaspase 7 is activated and recruited to the ER membrane [23]. Thus ER stress can lead to overlapping and potentially redundant pathways for caspase activation [41,42]. We demonstrate here that the ER-related caspases 7 and 12 are activated by Ad-mda7 in H1299 cells and thus provide a link between


doi:10.1016/j.ymthe.2003.11.014

UPR-activated ER signaling and mitochondrial effector caspases. The intracellular MDA-7 killing pathway is not unique to human cells: Ad-mda7 does not kill normal rat smooth muscle cells, but efficiently kills the rat PAC1 cell line via an unidentified intracellular mechanism [48]. Further support for MDA-7-mediated induction of UPR is provided by studies using murine fibrosarcoma lines, which exhibit up-regulation of PERK and activation of caspase 12 after Ad-mda7 treatment. In these cells, Ad-luc transduction did not alter levels of PERK or caspase 12 (Ramesh et al., manuscript in preparation). Because MDA-7 must enter the secretory pathway to become cytotoxic in H1299 cells, and Ad-mda7 activates the UPR stress pathway, we hypothesize that the cytotoxicity mediated by Ad-mda7 in lung cancer cells is due to activation of the UPR, resulting in activation of death signals from the ER to mitochondria. The identification of these ER death-promoting signals is the subject of intense investigation. It is evident that apoptosis induction in lung cancer cells is independent of MDA-7 receptor engagement. We have recently found that lung tumor cells do not express receptors for MDA-7 (manuscript in preparation), whereas human endothelial cells express functional MDA-7 receptors that mediate anti-angiogenesis [43]. Others have suggested that high levels of recombinant MDA-7 produced in bacteria can induce death in glioma cells or growth arrest in renal carcinoma cells; however, no data on receptor utilization was reported [49,50]. Treatment of receptor-positive tumor cells with low concentrations of human MDA-7 results in ligand – receptor engagement, intracellular signaling, and ultimately tumor-specific cell death (manuscript in preparation). Therefore, Ad-mda7 transduction can induce apoptosis via UPR intracellular pathways as described here in MDA-7 receptor-negative H1299 cells; however, in receptor-positive cells, MDA-7 can kill via intracellular as well as extracellular (receptor engagement) pathways.

MATERIALS AND METHODS Cell lines and cell culture. All the tumor cell lines utilized were obtained from American Type Culture Collection (Rockville, MD, USA). The cell lines evaluated were lung (H1299, A549), prostate (PC3), MCF-7 (breast), and 293 HEK cells. The cells were cultured according to the supplier’s recommendation and used in the log phase of growth when viability was at least 90%. Human umbilical vein endothelial cells were obtained from Clonetics (Walkersville, MD, USA). The cells were free of mycoplasma and were used in the log phase of growth. Cells were harvested with 0.125% trypsin – 1.3 mM EDTA (GIBCO BRL, Life Technologies, Grand Island, NY, USA). Brefeldin A and tunicamycin were purchased from Sigma Chemicals (St. Louis, MO, USA). Plasmids. pShooter expression plasmids (Invitrogen, Carlsbad, CA, USA) direct subcloned proteins to the cytoplasm, nucleus, or ER. Each vector construct (containing an engineered C-terminal myc tag) was used to subclone mda-7 cDNA. The vector directing proteins to the cytoplasm contains a standard expression vector backbone, while the vectors directing proteins to the nucleus and ER contain signal sequences appropriate


ARTICLE

to those compartments. The nuclear-targeting vector has an N-terminal nuclear-targeting signal, while the ER-targeting vector has an N-terminal ER signal peptide sequence and a C-terminal ER retention sequence. mda7 was subcloned into these vectors using PCR to delete both the stop codon and the region encoding the first 48 amino acids, constituting the secretion signal, from full-length mda-7 cDNA. PCR was also used to provide restriction sites compatible with the Invitrogen targeting vectors and in-frame with the C-terminal myc tag contained in the vectors. The primer sequences used for cloning are as follows: forward PCR primer (with SalI site), TTTTTTGTCGACATGGCCCAGGGCCAAGAATTCC; reverse PCR primer (with NotI site), TTTTTTGCGGCCGCGAGCTTGTAAGAATTTCTGC. All plasmids were sequenced to confirm orientation and in-frame sequence. Recombinant adenovirus production and gene transfer. Construction and purification of replication-deficient human type 5 adenovirus carrying the mda-7 gene was described previously [5]. Cell lines were infected with Ad-mda7 or Ad-luc at indicated m.o.i. Cells were plated at 105 – 106 cells/well in a six-well plate format for protein expression, trypan-blue exclusion assay, or apoptosis assays. For transfection with pShooter vectors, 1 Ag of plasmid DNA was used to transfect cells using the Lipofectamine protocol as described [17]. Microarray analysis. The Micromax Human cDNA System I – Direct Kit (NEN Life Science Products, Inc.) containing 2400 human cDNAs was used. mRNA was isolated from H1299 cells treated with Ad-mda7 or Adluc (1000 vp/cell for 24 h) and analyzed according to the manufacturer’s instructions. Western blot analyses. Cell lysates (105 – 106 cells were suspended in 500 Al of Laemmli buffer with 5% 2-mercaptoethanol (2ME)) or supernatants (1:1 mixing with Laemmli buffer + 2ME) were analyzed by SDS gel electrophoresis and Western blot analysis. Expression of various proteins was determined by using the following primary antibodies: rabbit or mouse anti-human MDA-7 (Invitrogen Therapeutics, Inc.), goat antihuman BiP/GRP-78, rabbit anti-human GADD34, goat anti-human PP2A, goat anti-human PERK, rabbit anti-human XBP-1 (Santa Cruz Biotechnologies, Santa Cruz, CA, USA), and rabbit anti-human caspase 7 and caspase 9 (Cell Signaling Technology, Beverly, MA, USA). Antibodies against h-actin, and a-tubulin were purchased from Sigma. Following incubation with horseradish peroxidase-labeled secondary antibody (Amersham Biosciences, Piscataway, NJ, USA) the proteins were visualized on enhanced chemiluminescence film (Hyperfilm; Amersham Biosciences) by application of Amersham’s enhanced chemiluminescence Western blotting detection system. All Westerns were performed two to four times. Production of recombinant MDA-7 protein. Secreted MDA-7 protein was purified from the supernatants of MDA-7 baculovirus-infected cells. Briefly, Hi-Five cells (Invitrogen) were infected with MDA-7 baculovirus at an m.o.i. of 5, and supernatants were harvested after 72 h of growth. For purification, polyclonal rabbit anti-MDA-7 antibody was cross-linked to immobilized protein G using the Seize-X Kit (Endogen) and incubated with supernatant at 4jC for 4 h. After being washed, bound protein was eluted according to the manufacturer’s instructions, neutralized, and dialyzed against PBS. MDA-7 protein-containing fractions were verified by Western blot and quantified by ELISA. Bacterial MDA-7 was produced by cloning the human mda-7 cDNA (residues 55 – 206) into the p202 expression vector (kindly provided by Dr. Jim Sacchetini, TAMU). This plasmid expresses a His(6)/maltose-binding protein (MBP)/MDA-7 fusion protein after induction with IPTG. Recombinant His(6)-MBP – MDA-7 protein was partially purified with Ni – NTA agarose resin (Qiagen). The protein was eluted from the column and MDA-7 was released from the His(6)-MBP carrier protein by overnight cleavage with TEV protease. The MDA-7 protein was purified to homogeneity by anion-exchange chromatography. Immunofluorescence assays. Cells were seeded into multiwell chamber slides and transfected with plasmid DNA or transduced with Ad vectors.

365

ARTICLE

Untreated cells served as controls in these experiments. Forty-eight hours after transfection cells were washed in PBS, fixed with cold ethanol:acetic acid (95:5 vol/vol), and stained for MDA-7 using mouse anti-human MDA-7 antibody and Texas red-conjugated anti-mouse secondary antibody (Vector Laboratories, Burlingame, CA, USA). Slides were mounted using anti-fade mounting reagent (Vector Laboratories) and analyzed using a Nikon fluorescence microscope. Photographs were taken using a Nikon Digital camera DXM1200 system. All assays were performed at least three times. Cell viability assays. Cell viability was determined by trypan blue exclusion assay or Live/Dead assay. Cells were trypsinized and a small aliquot was suspended 1:1 volume with 0.1% trypan blue. Total cell numbers and cell viability counts were assessed using a hemocytometer under light microscopy. For Live/Dead assay (Molecular Probes, Eugene, OR, USA) cells were plated in chamber slides (Nunc) and then treated with pShooter plasmid vectors or with Ad vectors. Two to four days later the cells were treated with the probes calcein AM and ethidium homodimer (both at 5 AM final concentration in PBS), for 30 min, and then live (green) and dead (red) cells were counted via fluorescence microscopy. All assays were performed at least three times. FACS analyses and annexin V assay. Cells were analyzed for apoptosis using the ApoAlert Annexin V – FITC Kit (Clontech). Briefly, vectortransduced cells (105 – 106 cells total) were washed extensively in PBS and then incubated with annexin V – FITC reagent for 30 min on ice. Cells were then washed with PBS and processed for FACS analysis and fluorescence microscopy as above. Assays were performed two or three times. Statistical analysis. The statistical significance of the experimental results was calculated using Student’s t test. The level of significance was set at P < 0.05.

ACKNOWLEDGMENTS This work was supported by NCI Grants CA89778, CA88421, and CA097598 (S.C.); Texas Higher Education Coordinating Board ATP/ARP Grant 0036570078-2001 (R.R.); an Institutional Research Grant (R.R.); and a W. M. Keck Gene Therapy grant (R.R.). RECEIVED FOR PUBLICATION SEPTEMBER 4, 2003; ACCEPTED NOVEMBER 21, 2003.

REFERENCES 1. Jiang, H., Lin, J. J., Su, Z. Z., Goldstein, N. I., and Fisher, P. B. (1996). Subtraction hybridization identifies a novel melanoma differentiation associated gene, mda-7, modulated during human melanoma differentiation, growth and progression. Oncogene 11: 2477 – 2486. 2. Jiang, H., Su, Z. Z., Lin, J. J., Goldstein, N. I., Young, C. S., and Fisher, P. B. (1996). The melanoma differentiation associated gene mda-7 suppresses cancer cell growth. Proc. Natl. Acad. Sci. USA 93: 9160 – 9165. 3. Chada, S., et al. (2003). MDA-7/IL-24 is a unique cytokine-tumor suppressor in the IL10 family. Int. Immunopharmacol. (in press). 4. Fisher, P. B., et al. (2003). mda-7/IL-24, a novel cancer selective apoptosis inducing cytokine gene. Cancer Biol. Ther. 2: 24 – 38. 5. Mhashilkar, A. M., et al. (2001). Melanoma differentiation associated gene-7 (mda-7): a novel anti-tumor gene for cancer gene therapy. Mol. Med. 7: 271 – 282. 6. Lebedeva, I. V., Su, Z. Z., Sarkar, D., and Fisher, P. B. (2003). Restoring apoptosis as a strategy for cancer gene therapy: focus on p53 and mda-7. Semin. Cancer Biol. 13: 169 – 178. 7. Saeki, T., Mhashilkar, A., Chada, S., Branch, C., Roth, J. A., and Ramesh, R. (2000). Tumor-suppressive effects by adenovirus-mediated mda-7 gene transfer in non-small cell lung cancer cell in vitro. Gene Ther. 7: 2051 – 2057. 8. Saeki, T., et al. (2002). Inhibition of human lung cancer growth following adenovirusmediated mda-7 gene expression in vivo. Oncogene 21: 4558 – 4566. 9. Pataer, A., et al. (2002). Adenoviral transfer of the melanoma differentiation-associated gene 7 (mda7) induces apoptosis of lung cancer cells via up-regulation of the double-stranded RNA-dependent protein kinase (PKR). Cancer Res. 62: 2239 – 2243. 10. Blumberg, H., et al. (2001). Interleukin 20: discovery, receptor identification, and role in epidermal function. Cell 104: 9 – 19. 11. Huang, E. Y., et al. (2001). Genomic structure, chromosomal localization and expression

366

doi:10.1016/j.ymthe.2003.11.014

profile of a novel melanoma differentiation associated (mda-7) gene with cancer specific growth suppressing and apoptosis inducing properties. Oncogene 20: 7051 – 7063. 12. Caudell, E. G., et al. (2002). The protein product of the tumor suppressor gene, melanoma differentiation-associated gene 7, exhibits immunostimulatory activity and is designated IL-24. J. Immunol. 168: 6041 – 6046. 13. Pataer, A., Chada, S., Hunt, K. K., Roth, J. A., and Swisher, S. G. (2003). Adenoviral MDA-7 induces apoptosis in lung cancer cells through mitochondrial permeability transition independent cytochrome c release. J. Thoracic Cardiovasc. Surg. 126: 1325 – 1328. 14. Ekmekcioglu, S., et al. (2003). Negative association of melanoma differentiation associated gene (mda-7) and inducible nitric oxide synthase (iNOS) in human melanoma: MDA-7 regulates iNOS expression in melanoma cells. Mol. Cancer Ther. 2: 9 – 17. 15. Sarkar, D., et al. (2002). mda-7 (IL-24) mediates selective apoptosis in human melanoma cells by inducing the coordinated overexpression of the GADD family of genes by means of p38 MAPK. Proc. Natl. Acad. Sci. USA 99: 10054 – 10059. 16. Kawabe, S., Nishikawa, T., Roth, J. A., Chada, S., and Meyn, R. (2002). Adenovirus mediated mda-7 gene expression radiosensitizes non-small lung cancer cells. Mol. Ther. 6: 637 – 644. 17. Mhashilkar, A. M., et al. (2003). MDA-7 negatively regulates the beta-catenin and PI3K signaling pathways in breast and lung tumor cells. Mol. Ther. 8: 207 – 219. 18. Gething, M. J., and Sambrook, J. (1992). Protein folding in the cell. Nature 355: 33 – 45. 19. Bertolotti, A., Zhang, Y., Hendershot, L. M., Harding, H. P., and Ron, D. (2000). Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat. Cell Biol. 2: 326 – 332. 20. Zhang, P., et al. (2002). The PERK eukaryotic initiation factor 2 alpha kinase is required for the development of the skeletal system, postnatal growth, and the function and viability of the pancreas. Mol. Cell. Biol. 22: 3864 – 3874. 21. Yeo, T. K., Yeo, K. T., and Olden, K. (1989). Accumulation of unglycosylated liver secretory glycoproteins in the rough endoplasmic reticulum. Biochem. Biophys. Res. Commun. 160: 1421 – 1428. 22 Magner, J. A., and Papagiannes, E. (1988). Blockade by brefeldin A of intracellular transport of secretory proteins in mouse pituitary cells: effects on the biosynthesis of thyrotropin and free alpha-subunits. Endocrinology 122: 912 – 920. 23. Kaufman, R. J. (2002). Orchestrating the unfolded protein response in health and disease. J. Clin. Invest. 110: 1389 – 1398. 24. Ron, D. (2002). Translational control in the endoplasmic reticulum stress response. J. Clin. Invest. 110: 1383 – 1388. 25. Harding, H. P., Zhang, Y., and Ron, D. (1999). Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase. Nature 397: 271 – 274. 26. Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., and Mori, K. (2001). XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107: 881 – 891. 27. Yu, Z., Luo, H., Fu, W., and Mattson, M. P. (1999). The endoplasmic reticulum stressresponsive protein GRP78 protects neurons against excitotoxicity and apoptosis: suppression of oxidative stress and stabilization of calcium homeostasis. Exp. Neurol. 155: 302 – 314. 28. Nakagawa, T., et al. (2000). Caspase-12 mediates endoplasmic-reticulum-specific apoptosis and cytotoxicity by amyloid-beta. Nature 403: 98 – 103. 29. Harding, H. P., et al. (2003). An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Mol. Cell 11: 619 – 633. 30. Meric, F., and Hunt, K. K. (2002). Translation initiation in cancer: a novel target for therapy. Mol. Cancer Ther. 1: 971 – 979. 31. Kaufman, R. J. (1999). Stress signaling from the lumen of the endoplasmic reticulum: coordination of gene transcriptional and translational controls. Genes Dev. 13: 1211 – 1233. 32. Fornace, A. J., et al. (1989). Mammalian genes coordinately regulated by growth arrest signals and DNA-damaging agents. Mol. Cell. Biol. 9: 4196 – 4203. 33. Novoa, I., Zeng, H., Harding, H. P., and Ron, D. (2001). Feedback inhibition of the unfolded protein response by GADD34-mediated dephosphorylation of eIF2alpha. J. Cell Biol. 153: 1011 – 1022. 34. Brush, M. H., Weiser, D. C., and Shenolikar, S. (2003). Growth arrest and DNA damage-inducible protein GADD34 targets protein phosphatase 1 alpha to the endoplasmic reticulum and promotes dephosphorylation of the alpha subunit of eukaryotic translation initiation factor 2. Mol. Cell. Biol. 23: 1292 – 1303. 35. Van Hoof, C., and Goris, J. (2003). Phosphatases in apoptosis: to be or not to be, PP2A is in the heart of the question. Biochim. Biophys. Acta 1640: 97 – 104. 36. Samuel, C. E., Kuhen, K. L., George, C. X., Ortega, L. G., Rende-Fournier, R., and Tanaka, H. (1997). The PKR protein kinase—an interferon-inducible regulator of cell growth and differentiation. Int. J. Hematol. 65: 227 – 237. 37. Sood, R., Porter, A. C., Olsen, D., Cavener, D. R., and Wek, R. C. (2000). A mammalian homologue of GCN2 protein kinase important for translational control by phosphorylation of eukaryotic initiation factor-2alpha. Genetics 154: 787 – 801. 38. Zinszner, H., et al. (1998). CHOP is implicated in programmed cell death in response to impaired function of the endoplasmic reticulum. Genes Dev. 12: 982 – 995. 39. Harding, H. P., Zhang, Y., Bertolotti, A., Zeng, H., and Ron, D. (2000). Perk is essential for translational regulation and cell survival during the unfolded protein response. Mol. Cell 5: 897 – 904.


doi:10.1016/j.ymthe.2003.11.014

40. Yan, W., et al. (2002). Control of PERK eIF2alpha kinase activity by the endoplasmic reticulum stress-induced molecular chaperone P58IPK. Proc. Natl. Acad. Sci. USA 99: 15920 – 15925. 41. Nishitoh, H., et al. (2002). ASK1 is essential for endoplasmic reticulum stress-induced neuronal cell death triggered by expanded polyglutamine repeats. Genes Dev. 16: 1345 – 1355. 42. Urano, F., et al. (2000). Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 287: 664 – 666. 43. Ramesh, R., et al. (2003). Melanoma differentiation-associated gene 7/interleukin (IL)24 is a novel ligand that regulates angiogenesis via the IL-22 receptor. Cancer Res. 63: 5105 – 5113. 44. DeSouza, N., Reiken, S., Ondrias, K., Yang, Y. M., Matkovich, S., and Marks, A. R. (2002). Protein kinase A and two phosphatases are components of the inositol 1,4,5trisphosphate receptor macromolecular signaling complex. J. Biol. Chem. 277: 39397 – 39400.


ARTICLE

45. Nutt, L. K., et al. (2002). Bax and Bak promote apoptosis by modulating endoplasmic reticular and mitochondrial Ca2+ stores. J. Biol. Chem. 277: 9219 – 9225. 46. Scorrano, L., et al. (2003). BAX and BAK regulation of endoplasmic reticulum Ca2+: a control point for apoptosis. Science 300: 135 – 139. 47. Zong, W. X., et al. (2003). Bax and Bak can localize to the endoplasmic reticulum to initiate apoptosis. J. Cell Biol. 162: 59 – 69. 48. Chen, J., Chada, S., Mhashilkar, A. M., and Miano, J. M. (2003). MDA-7/IL-24 selectively inhibits vascular smooth muscle cell growth and migration. Mol. Ther. 8: 220 – 229. 49. Yacoub, A., et al. (2003). Melanoma differentiation-associated 7 (interleukin 24) inhibits growth and enhances radiosensitivity of glioma cells in vitro and in vivo. Clin. Cancer Res. 9: 3272 – 3281. 50. Yacoub, A., et al. (2003). MDA-7 (interleukin-24) inhibits the proliferation of renal carcinoma cells and interacts with free radicals to promote cell death and loss of reproductive capacity. Mol. Cancer Ther. 2: 623 – 632.

367