CAAT/Enhancer Binding Protein Homologous Protein–Dependent Death Receptor 5 Induction Is a Major Component of SHetA2-Induced Apoptosis in Lung Cancer Cells

Introduction

Retinoids, including natural retinoic acid and its synthetic derivatives, are a group of promising anticancer agents endowed with both therapeutic and chemopreventive potential ( 1). Treatment of acute promyelocytic leukemia with all-trans retinoic acid is a successful example ( 2). However, the exploitation of the full potential of retinoids as chemopreventive and/or therapeutic drugs, particularly against solid tumors, has been hampered largely by their local and systemic toxicity and side effects, which are often associated with their ability to activate nuclear retinoid receptors ( 1, 3). Thus, efforts have been made for the past decades to develop novel retinoid-like compounds or atypical retinoids, which retain anticancer activity of retinoids with minimal toxicity and side effects ( 3). SHetA2 (see structure in Fig. 1A ) is such a compound because it did not activate retinoic acid receptor and retinoid X receptor receptors in receptor cotransfection assays and reporter cell lines ( 4), rescue or induce embryonic malformations in a RALDH knockout mouse model ( 5), or cause skin irritancy in a murine topical irritancy model ( 6). Thus, SHetA2 functions independently of retinoid receptors ( 7, 8). Accordingly, SHetA2 lacks conventional retinoid toxicities, such as skin irritation or teratogenecity, when tested in animal models ( 5, 6). However, SHetA2 has the ability to induce differentiation or reverse the cancerous phenotype ( 9). Moreover, SHetA2 induces apoptosis in cancer cells while sparing normal cells ( 7, 10). Importantly, SHetA2 effectively inhibits tumor growth in vivo without evidence of toxicity ( 6). Because of these encouraging results, SHetA2 was chosen for evaluation in the National Cancer Institute's Rapid Access to Intervention Development program (Application 196, compound NSC 726189) and is now in the Rapid Access to Preventive Intervention Development program, showing a potential as a cancer chemopreventive and therapeutic agent.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

SHetA2 (A) decreases the survival (B) and induces apoptosis (C) of human NSCLC cells and inhibits the growth of lung cancer xenografts (D). A, chemical structure of SHetA2. B, the indicted NSCLC cell lines were seeded in 96-well cell culture plates and treated the next day with the given concentrations of SHetA2. After 3 d, cell number was estimated using the SRB assay. Cell survival was expressed as the percentage of control (DMSO-treated) cells. Points, means of four replicate determinations; bars, SDs. C, the indicated NSCLC cell lines were treated with the given concentrations of SHetA2 for 48 h. Cell death including apoptosis and necrosis from these cell lines were then determined by Annexin V staining. D, mice carrying A549 xenografts were treated with vehicle control or SHetA2 by oral gavage (daily) for 8 d. Tumor sizes were measured once every 2 d. Points, mean; bars, SE (n = 6). *, P < 0.05 compared with vehicle control using Student's t test.

It is well known that there are two major apoptotic pathways: the extrinsic apoptotic pathway involving signals transduced through death receptors (DR) and the intrinsic apoptotic pathway relying on signals from the mitochondria. Both pathways involve the activation of a set of caspases, which in turn cleave cellular substrates and result in the characteristic morphologic and biochemical changes constituting the process of apoptosis ( 11, 12). The extrinsic pathway is characterized by the trimerization of cell surface DRs and activation of caspase-8, whereas the intrinsic pathway involves in the disruption of mitochondrial membranes, the release of cytochrome c from the mitochondria, and the activation of caspase-9. Through caspase-8–mediated cleavage or truncation of Bid, the extrinsic DR apoptotic pathway is linked to the intrinsic mitochondrial apoptotic pathway ( 11, 12).

The death ligand tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) and its receptors have recently attracted much attention because TRAIL preferentially induces apoptosis in transformed or malignant cells while sparing most normal cells, demonstrating potential as a tumor-selective apoptosis-inducing cytokine for cancer treatment ( 13). Currently, both TRAIL and agonistic antibodies against DR4 or DR5 are being tested in cancer clinical trials ( 14). Importantly, certain cancer therapeutic agents sensitize various types of cancer cells to TRAIL-induced or agonistic anti-DR4 or DR5 antibody–induced apoptosis ( 13, 15). Thus, these agents are useful in combination with TRAIL or an agonistic anti-DR4 or DR5 antibody to augment induction of apoptosis. It is well known that TRAIL binds to its receptors: DR4 (also called TRAIL-R1) and DR5 (also named Apo2, TRAIL-R2, or Killer/DR5) to activate the extrinsic apoptotic pathway ( 13).

Both DR4 and DR5 expressions are modulated by some anticancer agents ( 16), including certain synthetic retinoids ( 17– 23). As a result, these agents very often can sensitize cancer cells to TRAIL-induced or agonistic DR4 or DR5 antibody–induced apoptosis ( 17, 20, 21). Regulation of DR4 or DR5 occurs through p53-dependent mechanisms ( 24– 26) and p53-independent mechanisms [e.g., nuclear factor-κB, activator protein 1, and CAAT/enhancer-binding protein homologous protein (CHOP); refs. 23, 27– 30]. It has been recently shown that CHOP, also known as growth arrest and DNA damage gene 153 (GADD153), directly regulates DR5 expression through a CHOP binding site in the 5-flanking region of the DR5 gene ( 30, 31). Thus, certain drugs induce DR5 expression through CHOP-dependent transactivation of the DR5 gene ( 30– 34).

The mechanism of SHetA2-induced apoptosis has been shown to occur through the intrinsic mitochondrial pathway associated with loss of mitochondrial membrane integrity, generation of reactive oxygen species, release of cytochrome c from the mitochondria, and activation of caspase-3 in head and neck cancer cell lines ( 7). Likewise, SHetA2 induces apoptosis in human ovarian cancer cells though targeting the mitochondria associated with alterations in the balance of Bcl-2 proteins, independent of generation of reactive oxygen species ( 10). It seems that the mechanisms of SHetA2 may vary in different types of cancer cells. The present study investigated the effects of SHetA2 on induction of apoptosis and examined the underlying mechanism in human non–small cell lung cancer (NSCLC) cells, which are typically insensitive to most of the conventional retinoids ( 35). Given that other synthetic atypical retinoids, such as CD437 and N-(4-hydroxyphenyl)retinamide (4HPR), modulate the expression of TRAIL DRs and TRAIL-induced apoptosis independent of the retinoid receptors ( 17– 23), we focused on studying modulation of the extrinsic TRAIL DR-mediated apoptotic pathway in SHetA2-induced apoptosis.

Materials and Methods

Reagents. SHetA2 was described previously ( 6) and dissolved in DMSO at a concentration of 10 mmol/L, and aliquots were stored at −80°C. Stock solution was diluted to the appropriate concentrations with growth medium immediately before use. Human recombinant TRAIL was purchased from PeproTech, Inc.

Cell lines and cell culture. The human NSCLC cell lines used in this study were described previously ( 35). These cell lines were grown in monolayer culture in RPMI 1640 supplemented with glutamine and 5% fetal bovine serum at 37°C in a humidified atmosphere consisting of 5% CO₂ and 95% air.

Cell growth assay. Cells were cultured in 96-well cell culture plates and treated the next day with the agents indicated. Viable cell number was estimated using the sulforhodamine B assay, as previously described ( 35). Combination index (CI) for drug interaction (e.g., synergy) was calculated using the CompuSyn software (ComboSyn, Inc.).

Apoptosis assays. Apoptosis was detected either by analysis of caspase activation using Western blot analysis as described below or by Annexin V staining using Annexin V–phycoerythrin (PE) apoptosis detection kit (BD Bioscience) following the manufacturer's instructions and analyzed by flow cytometry using the FACScan (Becton Dickinson). In the Annexin V assay, the percentage of positive cells for Annexin V (PE) staining only in the bottom right quadrant and for both Annexin V and DNA (7-AAD) staining in the top right quadrant represents the early and late apoptotic populations, respectively. The percentage of positive cells for DNA staining only in the top left quadrant represents the necrotic population.

Western blot analysis. Preparation of whole-cell protein lysates and Western blot analysis were described previously ( 36). Mouse anti–caspase-3 monoclonal antibody was purchased from Imgenex. Rabbit polyclonal antibodies against caspase-8, caspase-9, Bid, and poly(ADP-ribose) polymerase (PARP), respectively, were purchased from Cell Signaling Technology. Mouse anti–caspase-10 monoclonal antibody was purchased from MBL International. Rabbit polyclonal anti-DR5 antibody was purchased from ProSci, Inc. Mouse monoclonal anti-DR4 antibody (B-N28) was purchased from Diaclone. Mouse anti-TRAIL monoclonal antibody was purchased from Imgenex. Mouse monoclonal anti-CHOP antibody (B-3) was purchased from Santa Cruz Biotechnology. Rabbit anti–β-actin polyclonal antibody and mouse anti-tubulin monoclonal antibody was purchased from Sigma Chemicals.

Detection of cell surface TRAIL receptors. The procedure for direct antibody staining and subsequent flow cytometric analysis of cell surface proteins was described previously ( 33). The mean fluorescent intensity (MFI) that represents antigenic density on a per cell basis was used to represent TRAIL receptor expression level. PE-conjugated mouse anti-human DR5 (DJR2-4), anti-human DR4 (DJR1), anti-human DcR1 (DJR3), and anti-human DcR2 (DJR4-1) monoclonal antibodies and PE mouse IgG1 isotype control (MOPC-21/P3) were purchased from eBioscience.

Gene silencing using small interfering RNAs. Silencing of caspase-8, DR5, and CHOP were achieved by transfecting small interfering RNA (siRNA) using RNAifect transfection reagent (Qiagen) following the manufacturer's instructions. Control (i.e., nonsilencing), caspase-8, DR5, and CHOP siRNAs were described previously ( 33, 37). These siRNAs were synthesized from Qiagen. Cells were plated in six-well cell culture plates and transfected the next day with the given siRNAs. After 24 h, the cells were trypsinized and replated in new plates and, on the second day, treated with SHetA2, as indicated. Gene silencing effects were evaluated by Western blot, as described above, after indicated times of treatment.

Construction of DR5 reporter plasmid, transient transfection, and luciferase activity assay. pGL3-DR5(−552) containing a wild-type CHOP binding site and pGL3-DR5(−552)CHOPm, in which the CHOP binding site was mutated, were generously provided by H.G. Wang (University of South Florida College of Medicine; ref. 30). The pGL3-DR5(−420) and pGL3-DR5(−240) reporter construct were described previously ( 33). The reporter constructs containing −1,400 and −810 bp 5′-flanking regions of the DR5 gene upstream of the translation start site, respectively, were amplified by PCR using the plasmid containing a 5′-flanking region of DR5 gene provided by Dr. G.S. Wu (Wayne State University School of Medicine) as a template. These amplified fragments were then subcloned into pGL3-basic reporter vector (Promega) through KpnI and BglII restriction sites. In the PCR amplification, the same reverse primer 5′-CTTAAGATCTGGCGGTAGGGAACGCTCTTATAGTC-3′ was used to make these constructs. The upstream primers were 5′-CTTAGGTACCGCAATAAATCTTGCTACTGC-3′ (for −1,400) and 5′-CTTAGGTACCAGCTACATGGGAGGCTGAGG-3′ (for −810), respectively. These constructs were named pGL3-DR5(−1,400) and pGL3-DR5(−810), respectively. The plasmid transfection and luciferase assay were the same, as described previously ( 33).

Lung cancer xenograft and treatments. Animal experiments were approved by the Institutional Animal Care and Use Committee of Emory University. Four-week-old to 6-wk-old female athymic (nu/nu) mice (∼20 g of body weight) were ordered from Taconic and housed under pathogen-free conditions in microisolator cages with laboratory chow and water ad libitum. The A549 cells (5 × 10⁶) in serum-free medium were injected s.c. into the flank region of nude mice. When tumors reached certain size ranges (50–100 mm³), the mice were randomized into two groups (n = 6 per group) according to tumor volumes and body weights for the following treatments: vehicle control and SHetA2 dissolved in sesame oil (oral gavage; daily). Tumor volumes were measured using caliper measurements once every 2 d and calculated with the formula V = (length × width²) / 6.

Results

SHetA2 inhibits the growth of human NSCLC cells in vitro and in vivo and induces apoptosis. We began our study by examining the effects of SHetA2 on the growth of a panel of human NSCLC cell lines, which are typically resistant to the conventional retinoids ( 35). As presented in Fig. 1B, SHetA2 exhibited concentration-dependent effects on decreasing the number of six tested NSCLC cell lines with comparable activity, indicating that SHetA2 effectively inhibits the growth of human NSCLC cells. The IC₅₀s, which are the concentrations required for inhibiting cell growth by 50%, were ∼2 μmol/L for all these cell lines after a 3-day exposure. Moreover, we showed that SHetA2 exerted concentration-dependent effects on increasing apoptotic cell death of the six NSCLC cell lines after a prolonged treatment (e.g., 48 h). Among these cell lines, H1299 cells were relatively less sensitive to SHetA2-induced apoptosis, because SHetA2 at 10 μmol/L induced <20% apoptotic cells ( Fig. 1C). In addition to induction of apoptosis, SHetA2 induced either G₁ (e.g., H1299) or G₂-M (e.g., H460, A549, and H157) cell cycle arrest (see Supplementary Fig. S1). Together, these results suggest that SHetA2 inhibits the growth of NSCLC cells through both apoptosis induction and growth arrest.

Using a A549 xenograft model in nude mice, we found that SHetA2 significantly inhibited the growth of A549 xenografts (P < 0.05; Fig. 1D), indicating that SHetA2 also effectively inhibits the growth of human lung tumors in vivo. We noted that there was a reduction of mouse body weight (10–15%) during the treatment, suggesting that the current treatment schedule with the maximal tolerated dose of 60 mg/kg may have some toxicity.

SHetA2 induces caspase-8–dependent apoptosis. To show the mechanism by which SHetA2 induces apoptosis, we determined the effects of SHetA2 on the activation of intracellular caspase cascades and its dependence on caspase activation. Procaspase-10 was not detected in H157 and H1299 cells or was not altered in A549 cells. The levels of procaspase-9 and Bid were decreased in a dose-dependent manner in both H157 and A549 cell lines, but not in H1299 cells. Consistently, cleavage of procaspase-8, procaspase-3, and PARP was detected in SHetA2-treated H157 and A549 cell lines, but cleavage of these proteins was not detected or only minimally detected in SHetA2-treated H1299 cells ( Fig. 2A ). These results indicate that both caspase-8–mediated and caspase-9–mediated caspase cascades are activated during SHetA2-induced apoptosis. Because caspase-8 activation can result in activation of the caspase-9–mediated cascade, we questioned whether this was the case in SHetA-induced apoptosis. To this end, we inhibited caspase-8 activation by silencing caspase-8 expression using a caspase-8 siRNA. As shown in Fig. 2B, we detected pro-forms of caspase-8 and cleaved caspase-8 in control siRNA-transfected cells, but not in caspase-8 siRNA-transfected cells, indicating a successful silencing of caspase-8 expression. Correspondingly, we detected cleaved forms of caspase-9, caspase-3, and PARP in control siRNA-transfected cells, but not in caspase-8 siRNA-transfected cells. Moreover, the effect of SHetA2 on decreasing cell number was substantially attenuated in caspase-8 siRNA-transfected cells compared with that in control siRNA-transfected cells ( Fig. 2C). Thus, these results collectively show that SHetA2 induces caspase-8–dependent apoptosis. In addition, caspase-9 activation induced by SHetA2 is secondary to caspase-8 activation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Effects of SHetA2 on cleavage of caspases and their substrates (A) and effect of caspase-8 silencing on SHetA2-induced caspase cleavage (B) and decrease in cell survival (C). A, the indicated cell lines were exposed to the given concentrations of SHetA2 for 30 h. The cells were then harvested for preparation of whole-cell protein lysates and subsequent Western blot analysis for detecting cleavage of caspases and their substrates. B and C, A549 cells were cultured in a six-well plate and the next day transfected with control (Ctrl) or caspase-8 siRNA. Twenty-four hours after transfection, the cells were re-seeded in a six-well plate or in a 96-well plate. On the second day, the cells were treated with DMSO or 10 μmol/L SHetA2 (in the six-well plate; B) or with the indicated concentrations of SHetA2 (in the 96-well plate; C). After 48 h, the cells were subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis (B) or to an estimation of cell number using the SRB assay (C). C, columns, mean of four replicate determinations; bars, SD. Procasp, procaspase; CF, cleaved fragment.

SHetA2 up-regulates DR5 expression. Given that caspase-8 activation plays a critical role in mediating the extrinsic apoptotic pathway, we next asked whether SHetA2 activates the DR-mediated apoptotic pathway. Therefore, we examined the effects of SHetA2 on the expression of DR5, DR4, and their ligand TRAIL. As presented in Fig. 3A , SHetA2 at 5 or 10 μmol/L increased the levels of DR5 protein in all the tested cells lines. Under the same conditions, SHetA2 did not increase DR4 expression; interestingly, SHetA2 even decreased the levels of DR4 protein in some cell lines (e.g., A549 and H1792). We failed to detect the basal levels of TRAIL or increase in TRAIL expression in H460, H157, H1299, and H1792 cells. In cell lines that did express TRAIL (e.g., A549 and Calu-1), SHetA2 either did not alter TRAIL levels (e.g., A549) or decreased TRAIL expression (e.g., Calu-1). A time course analysis of DR5 expression in cells exposed to SHetA2 showed that DR5 up-regulation occurred 4 hours post–SHetA2 treatment and was sustained up to 24 hours ( Fig. 3B). Given that functional DR5 and other TRAIL receptors are located on the cell surface, we further analyzed cell surface TRAIL receptor distributions in cells exposed to SHetA2. All three tested cell lines (i.e., H157, A549, and H1299) expressed basal levels of DR5 and DR4. Upon SHetA2 treatment, surface DR5 but DR4 levels were further increased. The MFIs of DR5 were 25 and 43 in DMSO-treated and SHetA2-treated H157 cells, respectively; 26 and 45, respectively, in A549 cells; and 11 and 27, respectively, in H1299 cells ( Fig. 3C). In H1299 cells, we found that SHetA2 actually decreased the levels of cell surface DR4. The MFIs of DR4 in H1299 cells were 20 in DMSO-treated cells and 12 in SHetA2-treated cells ( Fig. 3C). Cell surface DcR1 and DcR2 were not or only minimally detected in these three cell lines and not increased by SHetA2 ( Fig. 3C). Collectively, these results indicate that SHetA2 primarily up-regulates DR5 expression, including cell surface DR5 levels in human NSCLC cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Effects of SHetA2 on the expression of DR5, DR4, and TRAIL (A and B) and cell surface distributions of TRAIL receptors (C) in human NSCLC cell lines. A and B, the indicated cell lines were treated with the given concentrations of SHetA2 for 16 h (A) or with 5 μmol/L SHetA2 for the given time as indicated (B). The cells were then subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis. P, positive control lysates prepared from human HaCaT keratinocytes. C, the indicated cell lines were treated with 10 μmol/L SHetA2 for 16 h and then harvested for analysis of cell surface TRAIL receptors by immunofluorescent staining and subsequent flow cytometry. The filled gray peaks represented cells stained with a matched control PE-conjugated IgG isotype antibody. The open peaks were cells stained with PE-conjugated antibody against an individual TRAIL receptor.

SHetA2 cooperates with TRAIL to augment induction of apoptosis. If SHetA2-induced DR5 is functional, we speculated that inclusion of exogenous recombinant TRAIL in SHetA2 treatment would result in enhanced apoptosis induction. To test this hypothesis, we treated four NSCLC cell lines (i.e., H1299, Calu-1, A549, and H157) with SHetA2 alone, TRAIL alone, or both drugs combined and then assessed cell survival and apoptosis. As presented in Fig. 4A , the combination of SHetA2 at concentrations of 2.5 to 10 μmol/L with either dose of TRAIL (20 or 40 ng/mL) was much more effective in decreasing tumor cell number than either single agent alone. For example, in Calu-1 cells, both SHetA2 alone at 5 μmol/L and TRAIL (20 ng/mL) alone decreased cell number by ∼30%, but the combination of the two agents reduced cell number by >80%, which is greater than the sum of the effects of each agent alone. The CIs for these combinations in all four cell lines were smaller than 1 ( Fig. 4B), indicating that the combination of SHetA2 and TRAIL synergistically decreases cell survival. Moreover, we used Annexing V staining to detect apoptosis in three NSCLC cell lines (A549, H157, and H460) exposed to the combination of SHetA2 and TRAIL. During a 24-hour treatment, the SHetA2 and TRAIL combination was much more effective than each of the single agents alone in increasing apoptosis ( Fig. 4C). For example, in H157 cells, 10 μmol/L SHetA2 alone and 20 ng/mL TRAIL alone induced ∼14% and 17% apoptosis, respectively; however, their combination caused ∼54% of cells to undergo apoptosis ( Fig. 4C). By Western blotting, we also detected the strongest bands of cleaved caspases and PARP or the most reduction of procaspases and PARP in cells exposed to the combination of SHetA2 and TRAIL in comparison with those cells treated with SHetA2 or TRAIL alone ( Fig. 4D). Collectively, these results clearly indicate that SHetA2 synergizes with TRAIL to augment induction of apoptosis in human NSCLC cells.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

SHetA2 combined with TRAIL synergistically decreases cell survival (A and B) and induces apoptosis (C and D) in human NSCLC cells. A and B, the indicated cell lines were treated with the given concentrations of SHetA2 alone, TRAIL alone, and their individual combinations as indicated. After 24 h, cell number was estimated using SRB assay for calculation of cell survival (A) and CI (B) with CompuSyn software. Columns, mean of four replicate determinations; bars, SDs. A CI smaller than 1 indicates synergy. C, the indicated cell lines were treated with 10 ng/mL (H460) or 20 ng/mL (A549 and H157) TRAIL alone, 10 μmol/L SHetA2 alone, and their respective combinations as indicated. After 24 h, the cells were subjected to measurement of apoptosis using Annexin V staining. The percentage of positive cells in the top right and bottom right quadrants were added to yield the total of apoptotic cells. D, A549 cells were treated with 10 μmol/L SHetA2 alone, 20 ng/mL TRAIL alone, and their combination. After the indicated times, the cells were subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis.

Induction of DR5 expression is required for SHetA2-induced apoptosis and augmentation of TRAIL-induced apoptosis. To show the role of DR5 up-regulation in SHetA2-induced apoptosis and enhancement of TRAIL-induced apoptosis, we blocked DR5 induction through siRNA-mediated silencing of DR5 expression and examined its effect on the ability of SHetA2 to trigger apoptosis and to enhance TRAIL-induced apoptosis. As shown in Fig. 5A , in both H157 and A549 cells, DR5 siRNA transfection dramatically decreased basal levels of DR5 expression and, more importantly, abolished SHetA2-induced DR5 expression as detected by Western blot analysis ( Fig. 5A). As a result, the cleavage of casapse-8, caspase-3, and PARP were substantially inhibited in cells transfected with DR5 siRNA, compared with control siRNA-transfected cells ( Fig. 5A). Accordingly, SHetA2 had attenuated effects on decreasing cell number in DR5 siRNA-transfected cells compared with control siRNA-transfected cells ( Fig. 5B). Consistently, we detected less apoptotic cells (∼25%) in DR5 siRNA-transfected cells than in cells transfected with control siRNA (∼45%; Fig. 5C). Collectively, these results indicate that DR5 induction is required for SHetA2-induced apoptosis.

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Blockage of DR5 induction (A) attenuates the ability of SHetA2 to activate caspases (A), decrease cell survival (B), induce apoptosis (C), and augment TRAIL-induced apoptosis (D) in human NSCLC cells. A and B, the indicated cell lines were cultured in six-well plates and the next day transfected with control or DR5 siRNA. Twenty-four hours after the transfection, cells were re-seeded in six-well plates (A) or 96-well plates (B) and treated with 10 μmol/L SHetA2 (A) or the indicated concentrations of SHetA2 (B). After 48 h (30 h for A549 in A), the cells were subjected to preparation of whole-cell lysates and Western blot analysis (A) or to the SRB assay for calculation of cell survival (B). B, columns, mean of four replicate determinations; bars, SD. C and D, H157 (C) or A549 (D) cells were cultured in six-well plates and the next day transfected with control or DR5 siRNA. Forty-eight hours after the transfection, the cells were exposed to DMSO or 10 μmol/L SHetA2 for 48 h (C) or treated with DMSO control, 10 μmol/L SHetA2 alone, 20 ng/mL TRAIL alone, or SHetA2 combined with TRAIL for 24 h (D). The cells were then harvested for Annexin V assay to detect apoptosis. The percentage of positive cells in the top right and bottom right quadrants were added to yield the total of apoptotic cells.

In addition, we examined the effect of siRNA-mediated blockage of DR5 induction on cooperative induction of apoptosis by the combination of SHetA2 and TRAIL. As shown in Fig. 5D, the combination of SHetA2 and TRAIL induced ∼46% apoptosis in control siRNA-transfected cells but only 23% apoptosis in DR5 siRNA-transfected cells, indicating that DR5 induction is also important for cooperative induction of apoptosis by the SHetA2 and TRAIL combination.

SHetA2-induced DR5 up-regulation is CHOP dependent. To understand how SHetA2 up-regulates DR5 expression, we examined the effects of SHetA2 on the transactivation of reporter constructs with different lengths of DR5 5′-flanking regions to identify the region responsible for SHetA2-mediated DR5 transactivation. In a transient transfection and luciferase assay, SHetA2 minimally increased the luciferase activity of pGL3-DR5(−240) while drastically increasing the luciferase activity of pGL3-DR5(−420), pGL3-DR5(−810), and pGL3-DR5(−1400) ( Fig. 6A, left ), indicating that the region between −240 and −420 contains essential element(s) responsible for SHetA2-induced DR5 transactivation. There is a CHOP binding site in this region, which has been shown to be responsible for DR5 up-regulation by several anticancer agents ( 30– 32). Thus, we further compared the effects of SHetA2 on the transactivation of reporter constructs carrying wild-type and mutated CHOP binding sites, respectively. As presented in Fig. 6A (right), SHetA2 increased the luciferase activity of the constructs carrying the wild-type DR5 promoter region, but did not increase the luciferase activity of the construct carrying the DR5 promoter region with the mutated CHOP binding site. These results clearly indicate that the CHOP binding site in the DR5 promoter region is responsible for SHetA2-mediated DR5 transactivation.

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

Induction of CHOP-dependent DR5 expression (A and B) and ER stress (C) by SHetA2 and schematic summary of signaling pathways involved in SHetA2-induced apoptosis (D). A, SHetA2 increases CHOP-dependent DR5 promoter activity. The given reporter constructs with different lengths of 5′-flanking region of the DR5 gene (left) or the reporter constructs with wild-type and mutated CHOP binding sites (right) were cotransfected with pCH110 plasmid into A549 cells. After 24 h, the cells were treated with DMSO or 5 μmol/L SHetA2 for 16 h and then subjected to luciferase assay. Columns, mean of triplicate determinations; bars, SDs. B, SHetA2 exerts time-dependent effects on inducing CHOP expression, which is responsible for DR5 up-regulation by SHetA2. The indicated cell lines (top) were treated with 10 μmol/L SHetA2 for the given times as indicated and then subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis. A549 cells were transfected with control or CHOP siRNA (bottom). After 48 h, the cells were treated with the indicated concentrations of SHetA2 for 12 h and then subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis. C, the indicated cell lines were treated with 10 μmol/L SHetA2 for the given times as indicated and subjected to preparation of whole-cell protein lysates and subsequent Western blot analysis. SE, short exposure. D, schematic summary of SHetA2-induced apoptosis. CHOP-dependent up-regulation of DR5 is important for SHetA2 to regulate the extrinsic apoptotic pathway, including TRAIL-induced apoptosis. This effect may eventually cooperate with other mechanisms, e.g., effects on the intrinsic apoptotic pathway (dashed lines), to fully activate apoptosis of cancer cells.

We next determined whether SHetA2 increased CHOP levels and, if so, whether CHOP is responsible for DR5 induction. By Western blot analysis, we detected a time-dependent DR5 induction accompanied by CHOP up-regulation in cells exposed to SHetA2, both of which occurred at 4 hours and were sustained up to 12-hour post–SHetA2 treatment ( Fig. 6B, top). Moreover, we blocked SHetA2-induced CHOP expression using CHOP siRNA to see if blockage of CHOP induction abrogated the ability of SHetA2 to up-regulate DR5 expression. Indeed, blockage of the effect of SHetA2 on CHOP induction also abrogated the ability of SHetA2 to up-regulate DR5 expression in cells transfected with CHOP siRNA ( Fig. 6B, bottom). Thus, these results clearly indicate that SHetA2-induced DR5 up-regulation is secondary to CHOP induction. Collectively, we conclude that SHetA2 induces a CHOP-dependent DR5 expression.

SHetA2 induces endoplasmic reticulum stress. It is well known that CHOP is one of the highest inducible genes during endoplasmic reticulum (ER) stress and a critical component of ER stress-induced apoptosis ( 38, 39). To determine whether SHetA2 induces ER stress, resulting in CHOP up-regulation, we further examined modulation of other featured proteins (e.g., Bip/GRP78, IRE1α, eIF2α, ATF4, and XBP1) of ER stress ( 39, 40) in cells exposed to SHetA2. As presented in Fig. 6C, SHetA2 increased the levels of Bip/GRP78, IRE1α, ATF4, and XBP1 in a time-dependent manner in both A549 and H157 cells, which occurred early at 2 hours and were sustained up to 12 hours. In addition, we found that SHetA2 weakly increased levels of p-eIF2α in these cell lines. However, these effects were transient because SHetA2 increased p-eIF2α levels only at early times (e.g., 2 hours). Collectively, these results suggest that SHetA2 induces ER stress in human NSCLC cells.

Discussion

By studying the effects of SHetA2 on the growth of human NSCLC cells, the present study has shown that SHetA2 effectively inhibits the growth of human NSCLC cells through induction of both apoptosis and cell cycle arrest. Importantly, SHetA2 significantly inhibits the growth of human lung cancer xenografts in nude mice. Thus, our results show the efficacy of SHetA2 for treatment of lung cancer and thereby warrant further evaluation of SHetA2 as a potential anticancer agent against lung cancer. We noted that, upon SHetA2 treatment, H1299 cells, which are the least sensitive to induction of apoptosis, primarily underwent G₁ arrest, whereas H460, A549 and H157 cells, which are relatively sensitive to induction of apoptosis, underwent G₂-M arrest. Whether G₁ arrest confers apoptosis resistance to SHetA2 in H1299 cells needs further investigation.

In this study, we found that SHetA2 increased cleavage of both caspase-8 and caspase-9. Importantly, silencing of caspase-8 using the caspase-8 siRNA abrogated SHetA2-induced cleavage of caspase-9, caspase-3, and PARP and attenuated the ability of SHetA2 to decrease the survival of lung cancer cells ( Fig. 2). Thus, we conclude that SHetA2 induces caspase-8–dependent apoptosis. Given that silencing of caspase-8 prevents caspase-9 from activation, it seems that caspase-9 activation is secondary to caspase-8 activation during SHetA2-induced apoptosis. It has been shown that SHetA2 activates the mitochondrial apoptotic pathway in other types of cancer cells ( 8, 10). In our study, we also observed the cleavage of Bid, a caspase-8 substrate that mediates caspase-8–dependent activation of mitochondrial apoptosis, in SHetA2-treated cells. Thus, it is likely that SHetA2 induces apoptosis through activation of caspase-8 followed by activation of the mitochondrial apoptotic pathway involving cleavage of caspase-9 and caspase-3.

Moreover, we also found that SHetA2 up-regulated DR5 expression, including cell surface DR5 levels without increasing the expression of DR4, DcR1, and DcR2, as well as their ligand TRAIL ( Fig. 3), suggesting that DR5 up-regulation may be important for SHetA2-induced caspase-8 activation and apoptosis. When combined with recombinant TRAIL, SHetA2 exerted augmented effects on induction of apoptosis ( Fig. 4), suggesting that SHetA2-induced DR5 is functional. Moreover, siRNA-mediated blockade of DR5 induction attenuated the ability of SHetA2 to activate caspases, including caspase-8, to decrease cell survival, to induce apoptosis, and to augment induction of apoptosis when combined with TRAIL ( Fig. 5). These results provide robust evidence for a critical role of DR5 up-regulation in mediating SHetA2-induced, caspase-8–dependent apoptosis, as well as cooperative induction of apoptosis by the SHetA2 and TRAIL combination. These findings indicate that SHetA2 is distinct from 4HPR, which induces caspase-8–dependent apoptosis independent of DRs ( 41). We noted that SHetA2 actually decreased the levels of DR4 and TRAIL in some cell lines. The underlying mechanism and their biological significance are currently unknown and need further investigation.

DR5 expression can be regulated through p53-dependent and p53-independent mechanisms ( 26, 42). In our study, SHetA2 induced DR5 expression in all of the tested human NSCLC cell lines ( Fig. 3), among which only H460 and A549 have wild-type p53 gene ( 22, 25). Thus, it is likely that SHetA2 induces DR5 expression independent of p53. Recently, CHOP has been shown to regulate DR5 expression through the CHOP binding site in the DR5 gene ( 30, 31), revealing an important DR5 regulation mechanism. In our study, SHetA2 increased DR5 promoter activity, suggesting that DR5 induction occurs at the transcriptional level. The deletion and mutation analysis of the DR5 5′-flanking region revealed that the region containing the CHOP binding site is essential for SHetA2-mediated DR5 transactivation ( Fig. 6). Moreover, SHetA2 induced a time-dependent CHOP expression, which was accompanied by the up-regulation of DR5 expression. Blockage of SHetA2-mediated CHOP induction by the CHOP siRNA accordingly inhibited DR5 up-regulation ( Fig. 6). Collectively, we conclude that SHetA2 induces DR5 expression through a CHOP-dependent mechanism. It has been suggested that p53 regulates CHOP expression ( 43). In our study, SHetA2 increased CHOP expression in both p53 wild-type cells (e.g., A549) and p53 mutant cells (e.g., H157) in a similar fashion to DR5 up-regulation. Thus, it seems that SHetA2 induces CHOP independent of p53.

It is well known that CHOP is a featured ER stress-regulated protein involved in ER stress-induced apoptosis ( 38). Thus, our finding on CHOP induction by SHetA2 suggests that SHetA2 triggers ER stress. Indeed, SHetA2 increased the levels of Bip, IRE1α, p-eIF2α, ATF4, and XBP1 ( Fig. 6), all of which are additional featured proteins accumulated or increased during ER stress ( 39, 40). Therefore, it seems that SHetA2 induces ER stress in NSCLC cells. Several studies have shown that 4HPR induces CHOP expression and ER stress, which are important for 4HPR-induced apoptosis ( 44– 47). A recent study actually showed that 4HPR induces CHOP-dependent DR5 expression ( 17). In this regard, it seems that SHetA2 and 4HPR share similarities. Our preliminary data show that cyclohexamide, an inhibitor of general protein synthesis that can decrease the overall protein burden, but not salubrinal, an inhibitor of eIF2α dephosphorylation that can protect cell from ER stress-induced apoptosis through temporary slow down of the global protein translation ( 48), abolished SHetA2-induced CHOP expression (see Supplementary Fig. S2). These results suggest that blockage of eIF2α-mediated protein translation is not sufficient to prevent CHOP induction by SHetA2, although eIF2α-signaling pathway is critical for CHOP induction in ER stress ( 38). Nevertheless, the relationship between ER stress and CHOP induction, as well as apoptosis by SHetA2, needs further investigation.

In practice, our findings that SHetA2 increases DR5 expression and enhances TRAIL-induced apoptosis have meaningful implications. Given the therapeutic potential of TRAIL and agonistic DR5 antibodies, both of which are being tested in phase I clinical trials, SHetA2 can be used in combination with TRAIL or an agonistic anti-DR5 antibody to achieve an enhanced cancer therapeutic effect through augmenting induction of apoptosis of human cancer cells.

In summary, our study has shown that the novel synthetic atypical retinoid compound SHetA2 effectively inhibits the growth of human NSCLC cells both in vitro and in vivo. For the first time, we have shown that SHetA2 triggers ER stress and induces CHOP-dependent DR5 expression, leading to caspase-8–dependent apoptosis. Given that SHetA2 also activates the intrinsic apoptotic pathways, as shown in other studies, we suggest that the intrinsic and extrinsic pathways may be simultaneously activated by an initiating stress induced by the drug or by cross-talk between the extrinsic and intrinsic apoptosis pathways ( Fig. 6D).

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.