w09, a novel autophagy enhancer, induces autophagy-dependent cell apoptosis via activation of the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 pathway

ABSTRACT

The EGFR (epidermal growth factor receptor) signaling pathway is frequently deregulated in many malignancies. Therefore, targeting the EGFR pathway is regarded as a promising strategy for anticancer drug discovery. Herein, we identified a 2-amino-nicotinonitrile compound (w09) as a novel autophagy enhancer, which potently induced macroautophagy/autophagy and consequent apoptosis in gastric cancer cells. Mechanistic studies revealed that EGFR-mediated activation of the RAS-RAF1-MAP2K-MAPK1/3 signaling pathway played a critical role in w09-induced autophagy and apoptosis of gastric cancer cells. Inhibition of the MAPK1/3 pathway with U0126 or blockade of autophagy by specific chemical inhibitors markedly attenuated the effect of w09-mediated growth inhibition and caspase-dependent apoptosis. Furthermore, these conclusions were supported by knockdown of ATG5 or knockout of ATG5 and/or ATG7. Notably, w09 increased the expression of SQSTM1 by transcription, and knockout of SQSTM1 or deleting the LC3-interaction region domain of SQSTM1, significantly inhibited w09-induced PARP1 cleavage, suggesting the central role played by SQSTM1 in w09-induced apoptosis. In addition, in vivo administration of w09 effectively inhibited tumor growth of SGC-7901 xenografts. Hence, our findings not only suggested that activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway may play a critical role in w09-induced autophagy and apoptosis, but also imply that induction of autophagic cancer cell death through activation of the EGFR pathway may be a potential therapeutic strategy for EGFR-disregulated gastric tumors.

Introduction

Gastric cancer is the third leading cause of cancer mortality worldwide. Each year, almost one million new cases of gastric cancer are diagnosed and more than 700,000 people die of the disease.^1,2 The high mortality rate of gastric cancer patients is associated with the absence of significant symptoms in the stages of gastric cancer. Although surgical resection is a potentially curative approach for localized cases of gastric cancer, most cases of gastric cancer are diagnosed at an advanced, incurable stage that is refractory to traditional chemotherapeutic agents.^1,2 Therefore, development of targeted therapeutic agents that are highly efficacious to advanced gastric cancers is warranted.

EGFR (epidermal growth factor receptor) is a 132 kDa transmembrane receptor tyrosine kinase.^3-5 Overexpression of EGFR has been observed in the variety of solid tumors.^4-6 Emerging evidence indicates that overexpression of EGFR is also associated with advanced stages of gastric cancer.^7-9 EGFR is a receptor for EGF (epidermal growth factor) and related growth factors including TGFA (transforming growth factor α). Binding of EGFR by its ligands activates the receptor tyrosine kinase (RTK) at the cell surface, which activates mitogen-activated kinases (MAPKs) and many other signaling pathways to regulate gene transcription in the nucleus.^2,10 It has been shown that EGFR-initiated signaling pathways play critical roles in regulating cancer cell proliferation, differentiation, migration, and survival.^10-12 Recently, a growing number of studies using MAP2K/MEK (mitogen-activated protein kinase kinase) inhibitors (PD98059 or U0126), and ectopic expression of dominant negative or constitutively active forms of RAS, RAF1/RAF, MAP2K or MAPK1/ERK2-MAPK3/ERK2 indicate that activation of the RAS-RAF1-MAP2K-MAPK1/3 signaling pathway is involved in the induction of apoptosis and autophagic cell death in many cancer cell lines.^13-20 Therefore, targeting the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 signaling pathway is regarded as a promising strategy for anticancer drug discovery.

Macroautophagy/autophagy is an evolutionarily conserved and highly regulated eukaryotic cellular degradation pathway for maintaining cell homeostasis via sequestering and targeting bulk intracellular components for lysosomal degradation.^21-23 Autophagy plays a crucial role in human physiology and defense against human diseases such as microbe infections, cancer, aging, immunity, and neurodegenerative disorders (Alzheimer, Parkinson and Huntington diseases).^21-27 When autophagy is inhibited or downregulated, cellular functions are often severely compromised, leading to elevated levels of cell death with apoptotic or nonapoptotic features, which is related to the occurrence of many human diseases such as cancer development, aging, and neurodegenerative disorders.^6,28-29 Recently, accumulating evidence suggests that induction of autophagy can limit tumorigenesis in the long-term via promoting the degradation of the accumulated damaged proteins and organelles.^6,25,27 Furthermore, under certain circumstances, excessive or persistent autophagy not only promotes autophagic cell death, but also enhances the sensitivity of anticancer therapeutics via induction of autophagic cell death in various types of cancer.^22-27 Currently, targeting the autophagic pathway is regarded as a promising new strategy for cancer drug discovery. One of the best-studied autophagy inducers, rapamycin and its analogs has been approved by the US FAD for the treatment of multiple tumors.^30,31 However, the application of these drugs in clinical use is limited by their adverse side effects such as mammalian target of rapamycin inhibitor-associated stomatitis.³²

To discover novel autophagy inducers, we established a high throughput drug-screening model using HeLa cells constitutively expressing mCherry-LC3B and screened 15,000 compounds from the China Pharmaceutical University-Chinese National Compound Library (CPU-CNCL). Herein, we identified one 2-amino-nicotinonitrile compound (w09, China patent No. CN:201410244411:A) as a novel autophagy enhancer that induced the formation of large cytoplasmic vacuoles and autophagy. This novel autophagy enhancer induced autophagy-dependent apoptosis of gastric cancer cell via activation of the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 pathway. Moreover, in vivo administration of w09 effectively inhibited the growth of SGC-7901 xenografts. Therefore, our findings strongly support the notion that w09-triggered autophagy induces gastric cancer cell apoptosis and that w09 or its more potent and selective derivatives might be promising candidates for targeted therapy of advanced gastric cancers.

Results

w09 induces autophagy in human gastric cancer cells

To identify novel compounds with autophagy-induced activity, each compound was evaluated separately with fluorescence using high context image analysis (Acumen eX3). Chloroquine (CQ) and rapamycin served as positive controls and dimethyl sulfoxide (DMSO) served as the negative control. After quantitative analysis of the number and density of mCherry-LC3 dots caused by each compound in HeLa cells expressing mCherry-LC3B, the inducing-autophagy activity of positive compounds was further confirmed by fluorescence microscopy and laser scanning confocal microscopy. In a high-throughput screening of 15,000 small molecular compounds, we identified a series of 2-amino-nicotinonitrile compounds that induced the formation of large cytoplasmic vacuoles and autophagy. w09 (Fig. 1A), the most active derivative of the initial “hit” compounds, was selected for further mechanistic studies. Firstly, we observed the morphological change in w09-treated cells under a light microscope. As shown in Fig. 1B and D, w09 treatment resulted in apparent vacuolization in the cytoplasm of SGC-7901 cells in a dose and time-dependent manner, which occurred as early as 2 h and reached the maximum at 6 h after w09 treatment. The w09-induced cytoplasmic vacuolization is also dose-dependent. Five µM of w09 treatment resulted in only small-size vacuolization in part of cells, while treatment with 10 µM or higher concentrations of w09 led to massive cytoplasmic vacuolization in all treated cells (Fig. 1B).

Figure 1.

w09 induces autophagy in gastric cancer cells. (A) The structure of compound w09. (B) SGC-7901 cells were treated with the indicated concentrations of w09 or vehicle (DMSO) for 12 h, representative images of SGC-7901 cells are shown as photographed with a phase contrast microscope (× 200); scale bar: 20 µm. (C) The expression of SQSTM1 and conversion of LC3B-I to LC3B-II were evaluated by immunoblotting. (D) SGC-7901 cells were exposed to w09 (10 μM) for the indicated time, and representative images of SGC-7901 cells are shown (× 400); scale bar: 10 μm. (E) Conversion of LC3B-I to LC3B-II was analyzed by immunoblotting. (F) SGC-7901 cells were treated with 10 μM w09 for 6 h. Representative microscopy images of SGC-7901 cells were obtained by transmission electron microscopy. The black arrow indicates autophagic vacuoles containing cytoplasmic context. Scale bars: 0.5 μm. (G) Representative confocal fluorescence images of SGC-7901 cells stably expressing mCherry-LC3B, treated with w09 or vehicle for 6 h. The difference of mCherry-LC3 puncta in cells treated with w09 or vehicle was quantified with ImageJ and statistical analysis was conducted using the Student t test. **P<0.01.

To investigate whether cytoplasmic vacuolization induced by w09 is caused by autophagy, transmission electron microscopy (TEM), as a gold standard technique for identification of autophagy,^33,34 was used to analyze the ultrastructural morphological changes. After treatment with w09 for 6 h, we observed typical ultrastructural features of autophagosomes, i.e., double-membrane vacuoles with visible cytoplasmic contents in SGC-7901 cancer cells, whereas in control cells treated with vehicle alone, fewer of these features were observed (Fig. 1F). To further confirm this observation, we examined the effect of w09 in SGC-7901 cells stably expressing mCherry-LC3B, which is an effective way to detect autophagosomes by monitoring the distribution of mCherry-LC3B from a diffuse pattern to the accumulation of puncta in the cytoplasm. As shown in Fig. 1G, w09 treatment markedly increased the accumulation of fluorescent LC3 dots in cells, suggesting an effect of w09-induced autophagosome formation in these cells.

To further confirm whether w09 does induce autophagy in gastric cancer cells, we determined the expression of 2 forms of MAP1LC3B/LC3B (microtubule associated protein 1 light chain 3 β). LC3B, an important autophagy-related protein, exists as 2 forms, LC3B-I and the phosphatidylethanolamine conjugate (LC3B-II) in cells. In the process of autophagy, the cytosolic form LC3B-I is converted to the phagophore (the autophagosome precursor) membrane-bound LC3B-II.³³ This characteristic conversion of endogenous LC3B-I to LC3B-II is an autophagic biomarker and can be used to monitor autophagic activity. Therefore, we measured the expression levels of LC3B-II by immunoblotting. As shown in Fig. 1C and E, w09 treatment significantly augmented the expression levels of LC3B-II in a dose- and time-dependent manner in SGC-7901 cells. Interestingly, w09 treatment also resulted in significant increases in the expression levels of SQSTM1 (Fig. 1C), which is a selective substrate of autophagy.³⁵ Moreover, we observed that w09 exposure also caused a dose-dependent conversion of LC3B-II and cytoplasmic vacuolization in other gastric cancer cells, HGC-27, BGC-823 and AGS (Fig. S1A to C).

w09 induces autophagic flux in human gastric cancer cells

To examine the effect of w09 treatment on autophagic flux, we first used 3-MA, an early-stage autophagy inhibitor to inhibit the formation of autophagosomes and explore the effect of 3-MA on w09-induced autophagy.³⁶ As shown in Fig. 2A (left panel), pretreatment with 10 mM 3-MA for 1 h not only markedly reduced w09-induced cytoplasmic vacuoles, but also significantly inhibited the conversion of LC3B-I to LC3B-II (Fig. 2A, right panel). We then pretreated the cells with 2 lysosomal inhibitors, chloroquine (CQ) or bafilomycin A₁ (BafA1),^33,37 which blocks the later steps of autophagy,¹⁸ to prevent autolysosome formation and the autolysosomal degradation of LC3B-II. Treatment with CQ (20 μM) or BafA1 (10 μM) completely eliminated w09-induced cytoplasmic vacuolization (Fig. 2B) and markedly increased the accumulation of LC3B-II in cells compared with cells treated with w09 alone (Fig. 2C and D). To further verify the effect of w09 on autophagic flux, we examined the protein expression levels of SQSTM1, which is a common readout of autophagic activity. Similar to the enhanced LC3B-II turnover, combination with CQ or BafA1 resulted in a marked accumulation of SQSTM1 compared with w09, CQ or BafA1 treatment alone, suggesting that autophagy inhibitors blocked w09-triggered autophagic flux in gastric cancer cells (Fig. 2B and C).

Figure 2.

Blockage of autophagy prevents w09-induced autophagy in gastric cancer cells. (A) SGC-7901 cells were pretreated with 3-MA (10 mM) for 1 h, followed by treatment with w09 (10 μM) or vehicle (DMSO) for an additional 6 h. Representative microscopy images of SGC-7901 cells were obtained by phase contrast microscopy (× 200); scale bar 20 μm. The expression of autophagy-related proteins, LCB3-I, LC3B-II and SQSTM1 was assessed by immunoblotting. Quantification represents the relative protein levels of the cells normalized to ACTB (mean ± SD from 3 independent experiments (***P<0.001). (B) SGC-7901 cells were pretreated with CQ (20 μM) or BafA1 (1 or 10 nM) for 1 h, followed by treatment with w09 (10 μM) and vehicle (DMSO) for an additional 6 h and representative microscopy images of SGC-7901 cells were obtained by phase contrast microscopy (× 200); scale bar: 20 μm. (C) and (D) The expression of autophagy-related proteins, LC3B-I, LC3B-II and SQSTM1 of SGC-7901 cells treated as described in (B) was assessed by immunoblotting. Quantification represents the relative protein levels of the cells normalized to ACTB (Mean ± SD from 3 independent experiments (***P<0.001). (E) and (F) SGC-7901 cells were treated with w09 (10 μM) or CQ (20 μM) for 0 to 12 h and the expression of SQSTM1 protein was analyzed by western blot. SQSTM1 fold-increase levels compared with “0 h” are shown.

To further distinguish the increases of SQSTM1 is attributed to altered autophagic flux or their increased expression we analyzed the time-effect of w09 on SQSTM1 expression with the autophagy inhibitor CQ as positive control by western blot. As shown in Fig. 2E, in extending the drug treatment time-frame experiment, we found that the accumulation of SQSTM1 in CQ-treated SGC-7901 cells was rapidly increased in a short time (about 1 h) and then its protein level was maintained at a stable plateau for 12 h. However, the SQSTM1 protein level of w09-treated SGC-7901 cells was gradually increased in a time-dependent manner during 12 h treatment (Fig. 2F). These results suggested that w09 increases the expression of SQSTM1, while CQ only induces the accumulation of existing SQSTM1. After extended treatment of 48 h, CQ could continuously promote the accumulation of SQSTM1 (Fig. S2A), while the w09-induced SQSTM1 expression was slightly decreased (Fig. S2B). All these results indicated that SQSTM1 might be not a good indicator for analysis of w09-induced autophagy flux.

To further confirm the effect of w09 on autophagic flux, an SGC-7901-derived cell line or a HeLa-derived cell line stably expressing the mCherry-LC3B reporter was used to assay autophagic flux (i.e., the dynamics of the autophagic process).³⁸ As shown in Fig. 3A and 3B, 10 μM of w09 treatment caused markedly the accumulation of mCherry-LC3B dots in HeLa or SGC-7901 cells. An autophagy inhibitor, 3-methyladenine (3-MA), significantly blocked the formation of mCherry-LC3B puncta. Moreover, pretreatment with lysosome inhibitors, CQ or BafA1, markedly increased the accumulation of fluorescent LC3 dots (Fig. 3C and D).

Figure 3.

w09 induces autophagic flux in gastric cancer cells. HeLa (A) or SGC-7901 (B) cells stably expressing mCherry-LC3 were treated with the indicated drugs for 6 h, and representative fluorescent images (× 200) were photographed by laser scaning confocal microscopy (Olympus FV1000, Tokyo, Japan); scale bar: 20 μm. The difference of mCherry-LC3 dots in HeLa (C) and SGC-7901 (D) cells was quantified with ImageJ and at least 20 cells of each treatment were used to calculate statistical significance using the Student t test (**p<0.01) as described previously.³⁹

To further confirm whether or not w09 can induce autophagy flux, SGC-7901 cells transiently transfected with mCherry-EGFP-LC3B plasmid were further analyzed under a laser scanning confocal microscope. As shown in Fig. S3A, w09 markedly increased the number of red puncta in SGC-7901 cells stably expressing mCherry-EGFP-LC3B. In addition, w09-induced autophagic flux was further confirmed in SGC-7901 cells transiently transfected with a mTagRFP-mWasabi-LC3B reporter, which is more sensitive than mRFP-EGFP-LC3B for monitoring autolysosomes in autophagic flux.³⁹ As shown in Fig. S3B, compared with control cells, w09 markedly increased the number of red puncta (GFP⁻ RFP⁺) in SGC-7901 cells expressing mTagRFP-mWasabi-LC3B, suggesting that w09 promotes the formation of autolysosomes. The results were similar with those from rapamycin-treated cells. Blocking the fusion of autophagosomes and lysosomes with CQ, the number of yellow puncta (GFP⁺ RFP⁺) was significantly increased in w09-treated cells, while there were almost no red puncta (GFP⁻ RFP⁺) in the CQ-treated group (Fig. S3B), further implying that CQ blocks w09-induced autophagic flux. Collectively, all these data suggested that w09 can promote autophagy flux in gastric cancer cells.

w09 induces autophagy through activation of the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 signaling pathway in gastric cancer cells

The signaling pathways mediated by MAPKs (MAPK1/3, MAPK11/p38β-MAPK12/p38gamma-MAPK13/p38delta-MAPK14/p38α, and MAPK8/JNK1-MAPK9/JNK2-MAPK10/JNK3) have been implicated in the modulation of autophagy.^10-13 To investigate the molecular mechanisms of this compound to induce autophagy, we first determined whether w09 could induce gastric cancer cell autophagy through the activation of the MAPK1/3 pathway. We pretreated SGC-7901 cells with a MAP2K inhibitor U0126 for 1 h followed by 10 μM w09 for 12 h. Interestingly, pretreatment with U0126 (10 μM) could significantly prevent the w09-induced cytoplasmic vacuoles in SGC-7901 cells (Fig. 4A). Furthermore, pretreatment with U0126 markedly abrogated the effect of w09 on phosphorylation of MAPK1/3 (Fig. 4B). To further explore whether the involvement of MAPK1/3 signaling pathway in w09-induced autophagy, we next examined the effects of w09 on the accumulation of LC3B-II in the presence or absence of U0126 using immunoblotting analysis. As expected, pretreatment with U0126 markedly reduced the conversion of LC3B-II compared with w09 treatment alone (Fig. 4C and D), suggesting that w09-induced accumulation of LC3B-II through activation of MAPK1/3 signaling pathway. Together, these findings suggest that w09 could induce gastric cancer cell autophagy through activation of the MAPK1/3 signaling pathway.

Figure 4.

w09 induces autophagy via activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway. (A) SGC-7901 cells were pretreated with 10 μM U0126 for 1 h, followed by coincubation with w09 (10 μM) or vehicle (DMSO) for 12 h. Representative microscopy images of SGC-7901 cells are shown (× 200); scale bar: 50 μm. (B) The expression of the autophagy-associated protein and autopahagosome marker LC3B-II was analyzed by western blot, and its relative levels, normalized to ACTB (mean ± S.D) from 3 independent experiments (**P<0.01) were quantified (C). (D) The phosphorylation levels of the MAP2K-MAPK1/3 signaling pathway in cells treated with w09 were detected by immunoblot analysis. (E) SGC-7901 cells were pretreated with sorafenib (20 μM) for 1 h, followed by coincubation with w09 (10 μM) or vehicle (DMSO). The phosphorylation levels of the MAP2K-MAPK1/3 signaling pathway were assessed by immunoblot analysis, and (F) the expression levels of LC3B-I/LC3B-II were also detected by immunoblot analysis. The quantification of the LC3B-II/ACTB ratio is presented (***P<0.001). (G) SGC-7901 cells were pretreated with gefitinib (20 μM) for 1 h, followed by coincubation with w09 (10 μM) or vehicle (DMSO), the phosphorylation levels of the MAP2K-MAPK1/3 signaling pathway of SGC-7901 cells were assessed by immunoblot analysis, and (H) the expression levels of LC3B-I:LC3B-II were also detected by immunoblot analysis. The quantification of LC3B-II/ACTB ratio is presented (**P<0.01). After SGC-7901 cells were treated with various concentrations of w09 for 2 h (I), or inoculated with 10 μM w09 for different time intervals as indicated (J), the phosphorylation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway was detected by western blot.

The above findings led us to investigate whether the upstream signaling of the MAPK1/3 pathway was also involved in w09-induced autophagy. As shown in Fig. S4, pretreatment with sorafenib, which is the first oral multikinase inhibitor that targets RAF kinases, significantly prevented w09-induced cytoplasmic vacuolation. Furthermore, sorafenib strongly blocked the effect of w09 on the MAPK1/3 pathway activation (Fig. 4E) and drastically reduced LC3B-II accumulation (Fig. 4F). In addition, gefitinib, a specific EGFR inhibitor,⁴⁰ not only significantly inhibited the activation effect of w09 on the MAPK1/3 pathway (Fig. 4G), but also markedly suppressed the formation of w09-mediated cytoplasmic vacuolation (Fig. S4) and increased the accumulation of LC3B-II (Fig. 4H). Since previous studies have demonstrated that overexpression of EGFR is directly related with advanced stages of gastric cancer and unfavorable prognosis,^7-9 we further investigated the effect of w09 on activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway. As shown in Fig. 4I and J, w09 treatment resulted in a rapid activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway in a dose-dependent and time-dependent manner in SGC-7901 cells. Taken together, these results suggested that activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway does play an important role in w09-induced autophagy.

EGF receptor plays a critical role in w09-induced autophagy

To further confirm whether w09-induced autophagy depends on the EGF receptor, we investigated the effect of EGF on w09-induced autophagy. Firstly, the MAPK1/3 phosphorylation effect promoted by EGF was confirmed (Fig. 5A). As shown in Fig. 5B and C, pretreatment with EGF not only synergistically promoted w09-induced cytoplasmic vacuoles but also markedly increased the conversion of LC3B-II, suggesting that the EGF Receptor is involved in w09-induced autophagy. We further determined whether w09 could induce autophagy in an egfr-null cell line (CHO-K1), which lacks EGFR (Fig. 5D).⁴¹ CHO-K1 cells were treated with different concentrations of w09 for 6 h and the expression of LC3B was analyzed by immunoblotting assay. As shown in Fig. 5E and F, cytoplasmic vacuoles and the accumulation of LC3B-II were seldom observed in the egfr-null cells treated with w09. However, in the Egfr-transfected CHO-K1 cells, w09 treatment not only induced cytoplasmic vacuoles (Fig. 5E), but also markedly increased the accumulation of LC3B-II (Fig. 5G), suggesting that EGFR plays a critical role in w09-induced autophagy.

Figure 5.

EGFR plays a critical role in the regulation of w09-induced autophagy. (A) SGC-7901 cells were exposed to w09 or EGF for 6 h and the activation effect of w09 or EGF on the MAPK1/3 signaling pathway was assessed by western blot. (B) SGC-7901 cells were pretreated with EGF for 2 h, followed by incubation with w09 for 6 h and representative microscopy images of SGC-7901 cells treated as described is shown (× 200); scale bar: 20 μm. (C) SGC-7901 cells were treated with w09 alone or combined with EGF for 6 h. The conversion of LC3B-I to LC3B-II was assessed by western blot and relative fold-increase of LC3B-II is presented. (D) Expression of EGFR in the egfr-null CHO-K1 and CHO-K1 cells transfected with an EGFR-expression plasmid was examined by immunoblot analysis. (E) egfr-null CHO-K1 cells and Egfr-transfected CHO-K1 cells were treated with the indicated concentrations of w09 for 6 h. Representative microscopy images of SGC-7901 cells were obtained (× 200; scale bar: 20 μm) and the expression of LC3B-II was determined by immunoblot analysis (F).

w09 induces caspase-dependent apoptosis in gastric cancer cells

To ascertain the cytotoxic effect of w09, we first evaluated the antiproliferative effect of w09 on 4 human gastric cancer cell lines, SGC-7901, BGC-823, AGS, and HGC-27 using MTT assay. As shown in Table 1, the half-maximal inhibitory concentration (IC₅₀) values of w09 on these cell lines were 10 to 20 µM for 48 h treatments. In addition, because of the sensitive difference of w09 in HGC-27 and SGC-7901 cells, these 2 cell lines were chosen for most subsequent studies. As shown in Fig. 6A and S5A, exposure to w09 at concentrations ranging from 0 to 40 µM dose-dependently inhibited SGC-7901 and HGC-27 gastric cancer cell growth. Additional long-term clonogenic formation assays also showed that w09 treatment resulted in a dose-dependent reduction of clonogenic formation in SGC-7901 cell lines and HGC-27 cell lines (Fig. 6B and S5B).

Table 1.

IC₅₀ values of w09 in human gastric cancer cells.

Cell lines	SGC-7901	HGC-27	AGS	BGC-823
IC50 (µM)	16.13	12.70	17.58	26.20

IC₅₀ values for each cell lines were calculated by using GraphPad software. One representative result from triplicate was shown.

Figure 6.

w09 inhibits gastric cancer cell growth through the CASP8-dependent apoptosis pathway. (A) SGC-7901 cells were treated with various concentrations of w09 for 72 h. Cell viability was assessed by MTT assay. (B) SGC-7901 cells were treated with various concentrations of w09 for 3 d, followed by incubation with new medium without w09 for an additional 9 d. Cell colonies were stained with crystal violet and quantification of the colonies (cell number >20) normalized to control is presented (mean ± SD for 3 independent experiments, *P<0.05, and ***P<0.001). (C) SGC-7901 cells were treated with w09 or vehicle (DMSO) for 48 h and apoptosis was analyzed by FACS using an ANXA5-EGFP and PI cell apoptosis kit. Representative results from 3 independent experiments are presented. (D) Cells were treated as in (C), total protein was prepared and the expression of the apoptosis-related proteins was analyzed by western blot using antibodies against total or cleaved CASP9, CASP8, CASP3 and PARP1. (E) SGC-7901 cells were pretreated with a pan-caspase inhibitor, Z-VAD-fmk (10 μM) for 1 h, followed by incubation with 20 μM of w09 or vehicle (DMSO) for 2 more d and then cells were cultured in absence of w09 for an additional 10 d. Cell colonies were stained with crystal violet, and quantification of the colonies (cell number >50) normalized to control is presented (mean ± SD for 3 independent experiments; **P<0.01). (F) SGC-7901 cells were preincubated with Z-VAD-fmk (10 μM) for 1 h, followed by treatment with the indicated concentrations of w09 for an additional 48 h. Whole-cell lysates were prepared for western blot analysis of cleaved-CASP3 and quantification of the cleaved CASP3 levels is shown (mean ± SD from 3 independent experiments; **P<0.01 and ***P<0.001).

To investigate the mechanism of cell death induced by w09, we next examined the w09-induced apoptotic effect on SGC-7901 and HGC-27 cells. As shown in Fig. S6A, classical apoptotic cells with nuclear condensation and fragmentation induced by 40 μM w09 treatment of 48 h can be visualized in SGC-7901 cells by propidium iodide (PI) staining. To confirm this observation, we performed flow cytometry analysis of SGC-7901 (treatment of 48 h) and HGC-27 (treatment of 24 h) cells treated with w09 using ANXA5-EGFP and PI double-staining to make a quantitative evaluation of apoptosis. As shown in Fig. 6C and S5C, w09 treatment induced apoptosis of SGC-7901 (treatment of 48 h) or HGC-27 (treatment of 24 h) cells in a dose-dependent manner, respectively. Moreover, in most cases, the process of apoptosis was accompanied by the activation of the caspase cascade. Therefore, we next examined the effect of w09 on the cleavage of some caspase-related proteins, such as PARP1 (poly[ADP-ribose] polymerase 1), CASP9 (caspase 9), CASP8 and CASP3 by immunoblotting. As shown in Fig. 6D and S5D, w09 treatment increased the cleavage of PARP1, CASP3 and CASP8 in a dose-dependent manner in SGC-7901 and HGC-27 gastric cancer cell lines, respectively. In addition, we also observed w09 time-dependently increased the cleavage of CASP3 and PARP1 (Fig. S6B) in SGC-7901 cells. In contrast, the cleavage of CASP9 was not observed (Fig. 6D). Moreover, this observation was further confirmed by evidence that w09 treatment had no effect in SGC-7901 cells on the cleavage of BID (Fig. S6C), which is a substrate of CASP8 after death receptor activation,⁴² and CYCS (cytochrome c, somatic) release from mitochondria has not been observed in w09-treated SGC-7901 cells (Fig. S6C). These findings suggested w09 might directly induce apoptosis of gastric cancer cells through an active CASP8-mediated mitochondria-independent apoptosis pathway.

To further confirm the apoptotic effect of w09 on gastric cancer cells, we examined whether a pan-caspase inhibitor, Z-VAD-fmk, can abrogate the apoptosis induced by w09. As shown in Fig. 6E and S5E, pretreatment of cells with Z-VAD-fmk markedly reduced the w09-induced antiproliferative effect on SGC-7901 and HGC-27 cells. Moreover, pretreatment with Z-VAD-fmk markedly attenuated the activation of CASP3 and inhibited the cleavage of PARP1 induced by w09 (Fig. 6F and S5F), further indicating the caspase-dependent pathway is involved in w09-induced apoptosis. Collectively, these results clearly indicated that w09 induced CASP8 dependent apoptosis of gastric cancer cells.

Autophagy is involved in w09-induced cell death in gastric cancer cells

Because our studies have demonstrated thus far that w09 treatment induced autophagy in a few hours and caspase-dependent apoptosis in gastric cancer cell lines, these findings led us to investigate the causal relationship between w09-induced autophagy and apoptosis. To this end, we first blocked w09-induced autophagy with CQ or U0126, respectively, and then investigated the cytotoxic effect of w09 on SGC-7901 cells. As shown in Fig. 7A, compared with the w09 treatment alone, the pretreatment with CQ or U0126 profoundly attenuated the cytotoxic effect of w09 on SGC-7901 gastric cancer cells. In accordance with the antiproliferative activity data, results of clonogenic formation assays also confirmed that the pharmacological inhibition of w09-induced autophagy with CQ or U0126 markedly attenuated cell growth inhibition induced by w09 in the SGC-7901 cell line (Fig. 7B and C). Flow cytometric analysis also confirmed that pretreatment with CQ or U0126 markedly attenuated w09-induced cell apoptosis (Fig. 7D). Furthermore, western blotting analysis of the CASP3 cleavage fragment has also confirmed that CQ and U0126 markedly decreased CASP3 activation induced by w09 (Fig. 7E and F), suggesting that w09-induced cell apoptosis depends on activation of the RAS-RAF1-MAP2K-MAPK1/3 signaling pathway. Taken together, these data strongly suggest that w09 induced autophagy-dependent apoptosis of gastric cancer cells.

Figure 7.

Blocking w09-induced autophagy significantly attenuates w09-mediated apoptosis in gastric cancer cells. (A) SGC-7901 cells were pretreated with CQ (20 μM) or U0126 (10 μM) for 1 h, followed by coincubation with various concentrations of w09 for an additional 72 h. Cell viability was assessed by MTT assay (**P<0.01 vs control). (B) SGC-7901 cells were pretreated with CQ (20 μM) or U0126 (10 μM) for 1 h, followed by coincubation with 20 μM w09 for 48 h and then cells were cultured in the absence of w09 for an additional 10 d. Cell colonies were stained with crystal violet and (C) quantification of the colonies (cell number >50) normalized to control in (B) is presented (mean ± SD for 3 independent experiments; **P<0.01). (D) SGC-7901 cells were pretreated with CQ (20 μM) or U0126 (10 μM) for 1 h, followed by coincubation with 20 μM w09 for 48 h. Cell apoptosis analysis was performed by FACS using ANXA5-EGFP and PI cell apoptosis kit. Quantification of apoptosis in cells treated as described in (D) is presented (mean ± SD for duplicate experiments; **P<0.01). ((E)and F) Total proteins of SGC-7901 cells treated as described in (D) were extracted and the expression of cleaved CASP3 was detected by western blot analysis, and quantification of cleaved CASP3 is presented as mean ± SD from 3 independent experiments (*** P<0.001).

ATG5 (autophagy-related 5) is an essential protein for autophagy.³³ To further confirm whether w09 induces autophagy-associated apoptosis in the gastric cancer cells, we tested the effect of siRNA-silencing of ATG5 on w09-induced autophagy and apoptosis in SGC-7901 cells. Knockdown of ATG5 by siRNA significantly prevented the effect of w09-induced cell growth inhibition of SGC-7901 cells (Fig. 8A). Moreover, silencing ATG5 significantly suppressed the accumulation of LC3B-II induced by w09 in SGC-7901 cells (Fig. 8B). To further verify the association of w09-induced autophagy and apoptosis, we examined the effect of siRNA silencing of ATG5 on w09-induced apoptosis in SGC-7901 cells. As shown in Fig. 8C, downregulation of ATG5 significantly suppressed the cleavage of CASP3 induced by w09. Moreover, knockout of ATG7 markedly rescued w09-induced cell death (Fig. S7A and S7B), prevented the conversion of LC3B-II induced by w09 (Fig. S7C and S7D), and inhibited w09-induced CASP3 activation in SGC-7901 cells (Fig. S7E and S7F). Furthermore, these above results were further confirmed in a double knockout of ATG7 and ATG5 in the SGC-7901 cell line. As shown in Fig. 8D, double knockout of ATG7 and ATG5 completely inhibited the conversion of LC3B-I to LC3B-II. Compared with the wild type of SGC-7901 cells, double knockout of ATG5 and ATG7 markedly rescued w09-induced cell death (Fig. 8E) and significantly prevented the cleavage of PARP1 (Fig. 8F and G). Collectively, these data strongly suggest that w09-induced autophagy might be a prerequisite to cell death, i.e. autophagy plays a proapoptotic role in w09-mediated cell apoptosis in gastric cancer cells.

Figure 8.

Knockdown or knockout of autophagy-related genes markedly prevents w09-induced autophagy and apoptosis in gastric cancer cells. (A) SGC-7901 cells were transfected with ATG5 or control (Scrambled) siRNA for 48 h and the expression of ATG5 was evaluated by western blot (upper panel). 48 h after transfection, SGC-7901 cells were treated with w09 for 48 h and cell viability was assessed by MTT assay (lower panel) (**P<0.01). (B) SGC-7901 cells were transfected with ATG5 or control (Scrambled) siRNA for 48 h and then treated with 10 μM w09 for 6 h, the conversion of LC3B-I to LC3B-II and the expression of SQSTM1 was evaluated by western blot. (C) SGC-7901 cells were transfected with ATG5 or control (Scrambled) siRNA for 48 h and then treated with 20 μM w09 for 24 h. Cleaved CASP3 in cells was analyzed by western blot. (D) A wild-type SGC-7901 cell line and a new SGC-7901 cell line with double knockout of ATG5 and ATG7 genes were treated with w09 for 6 h, the expression of ATG5 and ATG7, as well as the conversion of LC3B-I to LC3B-II were assessed by western blot. (E) Wild-type SGC-7901 cells or SGC-7901 cells with ATG5 ATG7 double-knockout were treated with w09 for 48 h. Cell viability was evaluated by MTT assay (**P<0.01) and (F) the cleavage of PARP1 was analyzed with western blot and quantified (G).

Upregulation of SQSTM1 plays a critical role in w09-induced cell apoptosis in gastric cancer cells

SQSTM1 is a multifunctional protein that serves as a signaling hub that shuttles ubiquitinated proteins to the lysosome during autophagy.⁴³ In the present study, we observed that SQSTM1, an autophagic substrate, is markedly upregulated at protein levels in w09-treated gastric cancer cells (Fig. 1C). Furthermore, different from CQ, w09 can dose and time-dependently increase the expression of SQSTM1 (Fig. 1C and 2F). To investigate the mechanism of upregulation of SQSTM1, we first measured SQSTM1 mRNA levels in cells treated with w09 by performing RT-PCR and quantitative qPCR analyses. As shown in Fig. 9A, w09 treatment dose-dependently increased the SQSTM1 mRNA levels, suggesting that the increases of SQSTM1 protein levels induced by w09 are attributed to SQSTM1 mRNA upregulation. To examine whether w09 can transactivate SQSTM1 transcription, we constructed a reporter plasmid containing −1781 to +46 base pairs of the human SQSTM1 promoter fused to Luciferase (called pGL3-SQSTM1 (−1781/+46) as described previously (Fig. 9B).⁴⁴ As shown in Fig. 9C, w09 markedly increased the transcription of reporter gene. A series of 5′ truncations of the SQSTM1 promoter demonstrated that the region between nucleosides −1457 and −1781 is required for w09 activation of SQSTM1 transcription (Fig. 9D). This enhancer was further mapped by introduction of 3 single-nucleotide point mutations, at −1298, −1300, and −1302, as described previously.⁴⁵ These mutations completely abolished transactivation of SQSTM1 by w09 (Fig. 9C), confirming that the sequence 5′-TGCTGAGTCAC-3′ between nucleotides −1305 and −1295 is responsible for w09-mediated induction of SQSTM1 transcription.

Figure 9.

SQSTM1 plays an important role in w09-mediated cell apoptosis in gastric cancer cells. (A), w09 promotes SQSTM1 mRNA transcript in a dose-dependent manner. SGC-7901 cells were treated with w09 for 12 h with the indicated dose of w09 and the relative level of SQSTM1 mRNA of cells in the absence or presence of w09 was analyzed by real-time PCR. ((B)and C), Reporter gene assays were performed using wild-type (−1781/+46) or the indicated deleted or mutated SQSTM1 promoter constructs as described previously.^45,47 SGC-7901 cells were transfected with an empty vector (pGL3-Basic) or the indicated SQSTM1 promoter constructs. Cells were harvested 24 h after transfection and the relative promoter activities are expressed as the ratio between measured luciferase and renilla plasmid activities. The data shown are the mean activities obtained in one experiment preformed in triplicate (**P<0.01). A knockout of SQSTM1 in the SGC-7901 cell line (D) and a SGC-7901 cell line lacking the SQSTM1 LIR domain (E) were constructed and the autophagic effect of w09 on these 2 new cell lines was confirmed by western blot. All these cell lines were treated with w09 for 6 h and the autophagic markers LC3 and SQSTM1 were analyzed with western blot. (F) Compared with wild type, knockout of SQSTM1 or lack of the SQSTM1 LIR domain significantly rescued cell death-induced by w09 and (G) significantly inhibited the cleavage of PARP1 induced by w09. All these cell lines were treated with w09 for 48 h. The inhibitory effect of w09 was analyzed with MTT assay and the apoptotic effect of w09 was examined with western blot. Relative fold-increase of cleaved PARP1 compared with the negative control (DMSO) is presented.

Furthermore, SQSTM1 is a mediator of crosstalk between autophagy and apoptosis. We investigated whether SQSTM1 is involved in w09-induced apoptosis. To this end, we constructed one SQSTM1 knockout SGC-7901 cell line using CRISPR-Cas9 (Fig. 9D) and one SGC-7901 cell line stably expressing lack of the SQSTM1 LC3-interacting region (LIR) domain (Fig. 9E) using a retrovirus infection system. As shown in Fig. 9F, knockout of SQSTM1 or deleting the SQSTM1 LIR domain significantly rescued the w09-mediated inhibition effect on SGC-7901 cells. Furthermore, compared with the w09-treated wild type of SGC-7901 cells, knockout of SQSTM1 or deleting the SQSTM1 LIR domain markedly reduced the cleavage of w09-induced PARP1. Collectively, these results suggested that SQSTM1 might play a critical role in w09-induced apoptosis.

w09 inhibits the growth of gastric cancer in a mouse xenograft model in vivo

To evaluate the antitumor activity of w09 in vivo, human gastric cancer xenografts were established by subcutaneous injection of approximately 5 × 10⁶ SGC-7901 cells into the left dorsal area of nude mice. When the tumor volume reached about 100 mm³ in size, the tumor-bearing mice were randomized into vehicle control and treatment groups (6 animals each) and were given a daily intragastric administration of either at 0 mg/kg (vehicle control group) or 40 mg/kg w09 (treatment groups) for 14 consecutive d. Taxel as a positive control, was administrated via intraperitoneal injection at a dose of 2 mg/kg/3 d. As depicted in Fig. 10A, w09 effectively inhibited the growth of tumors in vivo. Moreover, the changes in body weight of w09-treated mice were similar to those of the vehicle-treated mice, suggesting no significant toxicity of w09 at a dose of 40 mg/kg (Fig. 10B).

Figure 10.

w09 inhibits tumor growth in an SGC-7901 xenograft model. (A) Relative tumor volume in vehicle control mice, w09-treated mice and taxel-treated mice. Error bars represent means ± SD **P<0.01 indicates significant difference in relative tumor volume compared with control, w09 or taxel. (B) Relative weight changes of mice during 21 d of exposure. Error bars represent means ± SD **P<0.01 indicates significant difference in relative body weight growth of the indicated treatment group. (C) Paraffin-embedded tumor tissue sections were stained with hematoxylin and eosin staining and an antibody against cleaved CASP3 for evaluating apoptosis was used. Representative photos are presented (× 200); scale bar: 20 μm.

To investigate whether w09 treatment results in induction of apoptosis of SGC-7901 cells in vivo, tumor tissues were sectioned, and stained with hematoxylin and eosin, and analyzed by immunohistochemistry with an antibody against cleaved CASP3. As shown in Fig. 10C, w09 treatment caused a significant increase in immunoreactivity for cleaved CASP3 compared with control, suggesting the mechanism of in vivo antitumor of w09 is associated with induction of apoptosis.

Discussion

One of the most important physiological roles of autophagy is to maintain cellular homeostasis via removal of long-lived proteins and damaged organelles and thus promoting cell survival.^21,46 However, extensive or sustained activation of autophagy can result in cell death, which is now designated as type II programmed cell death.^12,47 Therefore, autophagy-promoting cell survival or induction of cell death is highly context dependent. In this study, we demonstrated for the first time that w09 (Fig. 1A) induced autophagy and subsequent apoptosis in gastric cancer cells. Furthermore, we found that EGFR-mediated activation of the RAS-RAF1-MAP2K-MAPK1/3 signaling pathway played an essential role in w09-induced autophagy in gastric cancer cells. Interestingly, our data showed that w09 induced autophagy-related caspase-dependent apoptosis of gastric cancer cells. Therefore, w09 is a novel autophagy inducer with the mechanism of action distinct from other known autophagy inducers.

Autophagy and apoptosis are 2 evolutionarily conserved processes that regulate cell fate in response to various stresses.⁴⁸ The functional relationship between these 2 processes is very complicated. In some cellular settings, it can serve as a cell survival mechanism, suppressing apoptosis, and in others, it can result in cell death.^48-50 The induction of autophagy is generally considered an adaptive and cytoprotective mechanism for the recycling of nutrients and the removal of cytotoxic materials.^51,52 However, recent emerging evidence has suggested that autophagy is also implicated in the induction of caspase-dependent and caspase-independent cell death.^48-52 For example, Young et al.⁵³ have reported that activation of CASP8 is a critical step in the autophagy-dependent initiation of apoptosis in response to SKI-I, a pan-SK inhibitor, and bortezomib, a proteasome inhibitor. In this study, we found that w09 induced autophagy and apoptosis in gastric cancer cells. Moreover, w09-induced autophagy and apoptosis of gastric cancer cells have the following characteristics: at the conditions of low concentrations and short-term treatment, w09 mainly induces cell autophagy in gastric cancer cells, while at the conditions of high concentrations and long-term treatment, w09 not only triggers autophagy, but also induces apoptosis in gastric cancer cells. In addition, we further investigate the relationship between w09-induced cell apoptosis and autophagy in gastric cancer. First, blockade of autophagy with autophagy inhibitors such as CQ, which inhibits the fusion of autophagosome and lysosome, completely inhibited w09-induced autophagy and markedly attenuated the cytotoxic and cell apoptosis effect of w09 on the gastric cancer cell. Similar results were also observed from experiments using U0126 to block w09-induced autophagy. Thus, these results suggested that activation of autophagy plays an important role in the subsequent induction of apoptotic cell death in the w09-treated gastric cancer cells. Second, blocking autophagy with CQ or U0126 partially, but not completely, prevented w09-induced cell death in the gastric cancer cell. Moreover, inhibition of apoptosis with a pan-caspase inhibitor, Z-VAD-fmk, did not completely block the w09-mediated cell death effect, suggesting that other signaling pathways including autophagic cell death might be involved in cell death induced by w09. Third, it has been demonstrated that sustained MAPK1/3 activation can promote either intrinsic or extrinsic apoptotic pathways by induction of mitochondrial CYCS release or CASP8 activation, permanent cell cycle arrest or autophagic vacuolization.²² In this study, we also revealed that w09 promoted cell apoptosis of gastric cancer cell through CASP8-dependent extrinsic apoptotic pathway. Moreover, our results indicated that sustained MAPK1/3 phosphorylation promoted by w09 is positively correlated with the treatment time and concentration of w09 for triggering apoptosis of gastric cancer cells. This result is further supported by our ongoing work that the basal level of MAPK1/3 phosphorylation of different types of tumor cells plays a critical role in determining the activity of w09-induced autophagy and apoptosis (data not shown). Therefore, these results suggested that sustained MAPK1/3 activation might be also directly implicated in the w09-induced apoptosis of gastric cancer cells; instead of indirectly activating apoptosis through the induction of autophagy.

SQSTM1 as one autophagic substrate directly interacts with LC3 through the LC3-interacting region (LIR) and is finally degraded in the autolysosomes.³³ Thus, SQSTM1 as an autophagic marker has been widely used to monitor autophagic flux. However, many of recent studies indicated that endogenous SQSTM1 as an indicator of autophagy flux should be cautious, as its level is regulated by several factors.³³ For example, upon starvation SQSTM1 is initially degraded by autophagy. However, SQSTM1 is restored back to basal levels during prolonged starvation in mouse embryonic fibroblasts and HepG2 cells, suggesting that the expression level of SQSTM1 does not always inversely correlate with autophagic activity.⁵⁴

Recently, Kim Jin-Hwan et al. also report that aberrant RAF1-MAP2K-MAPK1/3 activation is sufficient to induce signals that trigger an alteration in LC3 and SQSTM1 levels in a similar pattern indicating autophagy.⁵⁵ In this study, we have demonstrated that w09 can dose- and time-dependently increase the expression of SQSTM1 by regulating mRNA transcription. But in fact, our data tend to support the conclusion that w09 can induce autophagy flux in gastric cancer cells. Therefore, it is important to confirm that the SQSTM1 mRNA level has not changed if the SQSTM1 protein level is used as an indicator of autophagy flux. In addition, SQSTM1 as a mediator of crosstalk between autophagy and apoptosis plays critical roles in the life and death decisions of the cell.^35,43 In this study, we found that knockout of SQSTM1 or deleting the SQSTM1 LIR domain not only significantly rescued the w09-mediated inhibitory effect on SGC-7901 cells, but also markedly inhibited the w09-induced PARP1 cleavage, suggesting that SQSTM1 might play an important role in w09-induced apoptosis in gastric cancer cells. These findings also indicate that further investigation concerning the role of SQSTM1 in w09-induced apoptosis of gastric cancer cells is needed.

Although autophagy is critical for maintenance of cellular homeostasis, the molecular mechanisms of autophagy regulation remain largely unknown. It has been demonstrated that several signaling pathways are implicated in autophagy in mammalian cells, such as the GTPase GNAI3/G_i3, the class III phosphatidylinositol 3-kinase, and MTOR, PRKC/protein kinase C, PRKAC/cAMP-dependent protein kinase A, and MAPK11/12/13/14.¹⁸ Currently, based on the above signaling pathways whether or not depending on MTOR inhibition to induce autophagy, autophagy inducers are also classified into 2 categories: autophagy inducers depending on MTOR, including glucagon, sorafenib, curcumin, arsenic trioxide, resveratrol, CCI-779, RAD001, AP23576 or other similar compounds, and autophagy activators independent of the MTOR signaling pathway, including Lithium chloride, L-690, 330, Carbamazepine or Valproic acid sodium salt.^56-60 Our results showed that w09, as a novel inducer, maybe belong to the latter group. Notably, Tasdemir et al.⁵⁸ report that the variety of autophagy inducers, such as ABT737, lithium, rapamycin, tunicamycin or nutrient depletion, rarely induce autophagic vacuolization in more than 50% of the treated cells and stereotypically induce autophagy preferentially in the G₁ and S phases of the cell cycle. However, unlike these autophagy inducers, we observed that w09, at >10 µM, can induce autophagic vacuolization in various types of cancer cells, suggesting that the effects of w09 on cell autophagy might be different from these known autophagy inducers. Therefore, it will be important to investigate the underlined mechanism for the observed discrepancies in the future.

The EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 signaling pathway plays a crucial role in cell functions, such as proliferation, differentiation, cell cycle progression, or cell death.^62-65 It has been widely demonstrated that activation of MAPK1/3 is involved in the regulation of autophagy in many cell models in response to different stimuli, such as lindane,¹⁸ soyasaponins,¹⁷ aurintricarboxylic acid (ATA75),¹⁶ dihydrocapsaicin, cadmium,²⁰ and TNF.¹² Interestingly, recent investigations reveal that inhibition of EGFR tyrosine kinase can induce autophagy in cancer cells.^66-68 For example, Wei et al.⁶⁷ recently report that erlotinib, a small molecule that inhibits the tyrosine kinase activity of EGFR, can induce autophagy in cancer cells through blocking the BECLIN1/EGFR interaction. The results from Chen et al.⁶⁹ also show that targeting cathepsin S with cathepsin inhibitors or RNAi induces tumor cell autophagy via the activation of the EGFR-RAS-MAP2K-MAPK1/3 signaling pathway.

These studies clearly document the relationships between the EGFR tyrosine kinase activity and the induction of cell autophagy. However, our current knowledge on the role of EGFR-mediated activation of the RAS-RAF1-MAP2K-MAPK1/3 signaling pathway in autophagy induction is limited. Furthermore, few specific small molecules are reported to induce cell autophagy via activation of the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 signaling pathway. Here, our data clearly indicated that activation of the EGFR-mediated RAS-RAF1-MAP2K-MAPK1/3 signaling pathway plays a critical role in the regulation of autophagy induced by w09 in gastric cancer cells. Moreover, this conclusion was further supported by evidence that blocking either the upstream signaling pathway with gefitinib or the downstream signaling pathway with sorafenib, markedly attenuated the effect of w09-induced autophagy. More interestingly, in the egfr-null CHOK1 cells, w09 treatment could not induce cytoplasmic vacuoles and LC3B conversion, whereas in the Egfr-transfected CHOK1 cells, the same treatment could effectively induce cytoplasmic vacuoles and the conversion of LC3B-I to LC3B-II. Collectively, the above results suggest that EGFR plays a key role in w09-induced cell autophagy. However, we also found that inhibition of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway with pharmacological or genetic approaches did not completely block w09-induced autophagy and apoptosis, suggesting that other potential mechanisms of action of w09 need to be further investigated in the future.

In conclusion, the EGFR signaling pathway is highly deregulated in the majority of epithelial malignancies where overactivation of EGFR signaling results in enhanced cell survival, proliferation and resistance to anticancer therapeutics. Disregulated EGFR is also associated with the advanced stages of gastric cancer and an unfavorable prognosis.^7-9 Currently, targeting EGFR pathway with small molecules by inhibition of intracellular tyrosine kinase activity has been regarded as a promising therapeutic strategy in these cancers. However, these strategies are very limited in most of human tumors. Thus, inducing autophagic-related cancer cell death through activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway might be a promising strategy for EGFR-disregulated cancer treatment. In this present study, our results for the first time demonstrate that w09 as a novel autophagy enhancer effectively induces autophagic-related gastric cancer cell death through activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 pathway. Upregulation of SQSTM1 played an important role in w09-induced apoptosis. These findings, not only suggest that w09 could be further developed as a novel autophagy enhancer, but also suggest that w09 might be a potentially good tool to discover the molecular mechanisms of regulating autophagy by activation of the EGFR-RAS-RAF1-MAP2K-MAPK1/3 signaling pathway in cancer cells. A more thorough understanding of the molecular mechanisms by which w09 induces autophagic-related cell death will facilitate the development of the compound as a candidate of therapeutics to treat gastric cancer.

Materials and methods

Cell lines and reagents

SGC-7901 (Human, STCC, TChu 46), AGS (Human, ATCC, CRL-1739), HGC-27 (Human, STCC, TCHu 22), BGC-823 (Human, ATCC, CRL-2765), HeLa (Human, ATCC, CCL-2) and CHO-K1 (Hamster, ATCC, CCL-61) cells were obtained from the American Type Culture Collection (ATCC) and the Shanghai Type Culture Collection of Chinese Academy of Sciences (STCC) and cultured in Roswell Park Memorial Institute 1640 (RPMI-1640; HyClone, SH30809.01B: SGC-7901, AGS, HGC-27, HeLa) or Dulbecco modified Eagle medium (HyClone, SH30022.01B: BGC-823) or F12K (CHO-K1) supplemented with 10% (v/v) fetal bovine serum (Gibco, 10099–141), penicillin (100 units/mL) and streptomycin (100 µg/mL) in a 5% CO₂ incubator at 37°C. F12K medium was obtained from Poster (PYG0036). Chloroquine diphosphate salt (c6628) and bafilomycin A₁ (BafA1) (B1793) were purchased from Sigma-Aldrich. U0126-EtOH (S1102), ly294002 (S1105), wortmannin (S2758), sorafenib (S7939), and gefitinib (S0125) were purchased from Selleckchem. Z-VAD-fmk was obtained from Beyotime Institute of Biotechnology (C1202). Stock solutions (20 mM) of test compounds were prepared in DMSO and stored at −20°C, further diluted before use with the culture media to obtain an optimal range of concentrations for treatments. All other chemicals were obtained from Sigma-Aldrich. Primary antibodies against LC3B (3868), SQSTM1 (8025), phospho-MAPK1/ERK2/p42-MAPK3/ERK1/p44 (Thr202/Tyr204; 4370), CASP3/caspase 3 (9665), cleaved CASP3/caspase 3 (Asp175; 9664), CASP8/caspase 8 (9746), CASP9/caspase 9 (9502), cleaved CASP9/caspase 9 (7237), PARP1 (9542), cleaved PARP1 (Asp214; 5625), phospho-MAP2K1/MEK1-MAP2K2/MEK2 (Ser217/221; 9154), phospho-RAF1/c-RAF (Ser259; 9421), RAS/Ras (3339), and BID (2002) were purchased from Cell Signaling Technology. Primary antibodies against MAPK1/ERK2/p42-MAPK3/ERK1/p44 (AM076) and ACTB (AA128), were obtained from Beyotime Institute of Biotechnology. Cytochrome c, somatic (CYCS) (556433) was obtained from BD Biosciences. Horseradish peroxidase-conjugated species-specific secondary antibodies were obtained from Cell Signaling Technology (anti-rabbit IgG, 7074P2 and anti-mouse IgG, 7076P2).

Luciferase reporter plasmid constructs

The −1781/+46 SQSTM1 promoter was amplified by PCR using primers 5′-ATCCTAGAGCTAGCACGCTGACTCACTGCCGGC-3′ and 5′-GATGACTGAAGCTTAGCGAGCGGCGAGCTGGCG-3′ and recombined into the pGL3 luciferase reporter (Promega) using ClonExpress Entry One Step Cloning Kit (Vazyme, China). The −1475/+46 SQSTM1 promoter construct, the −1158/+46 SQSTM1 promoter construct and the −355/+46 SQSTM1 promoter construct were generated using primers 5′-ATCCTAGAGCTAGCAGGAGCTCCTTCTTGGG-3′ (−1475/+46), 5′-ATCCTAGAGCTAGCGAGGCCTGAACCCTCTC-3′ (−1158/+46), 5′-ATCCTAGAGCTAGCGGGTACCCCCAACTGAG-3′(−355/+46), and 5′-GATGACTGAAGCTTAGCGAGCGGCGAGCTGGCG-3′ as above described. The −1781/+46 or −1475/+46 SQSTM1 promoter construct with a mutated ARE (antioxidant response element, which is a DNA regulatory sequence in the SQSTM1 promoter) site at position −1300 were generated by using the QuikChange II Site-Directed Mutagenesis Kit (Agilent Technologies). All constructs were verified by DNA sequencing.

Reporter gene assay

Subconfluent SGC-7901 cells cultured in 96-well plates were transiently cotransfected with 40 ng of luciferase reporter plasmids and 5 ng of Renilla plasmid (Promega, E226A) to correct for variations in transfection efficiency. Four h post transfection, cells were treated with 10 μM w09 for 24 h and then were harvested. Luciferase activities were measured using the Dual Light luciferase kit (Promega, E1960). All assays were performed in triplicates and repeated at least 3 times.

Cell viability assay

Cells were seeded into 96-well flat-bottomed plates (Corning, 3799) at a density of 5,000 cells per well and were treated in triplicate with various concentrations of compounds for 48 to 72 h. Each well was supplemented with 20 μl MTT (5 mg/ml; Sigma, M5655) and incubated for 4 h. After removal of supernatants, 150 μl DMSO was added to each well to dissolve the formazan and the OD value of each well was measured with a microplate reader (TECAN Sirfire II, Switzerland) at 570 nm. Cell viabilities were normalized to the control.

Colony formation assay

SGC-7901 and HGC-27 cells were seeded in 6-well plates at a density of 500 cells per well and were treated with the indicated compounds or vehicle (DMSO) for the indicated time. The medium was replaced for every 2 d. After a total of 12 d of incubation, cell colonies were fixed with 4% paraformaldehyde and stained with crystal violet. Colonies defined as clusters of ≥ 20 cells. The colonies were counted, and normalized to the control.

Cell apoptosis assay

Apoptosis was determined by flow cytometry using an ANXA5-EGFP and PI Apoptosis Detection kit (Bestbio, BB16101) according to the manufacturer's directions. Briefly, 5 × 10⁵ cells were seeded in 6-well plates and then were exposed to various treatments of an additional 48 h. Cells were harvested by trypsinization and washed twice with cold phosphate-buffered saline (PBS; 0.27 g KH₂PO₄, 1.42 g Na₂HPO₄, 8 g NaCl, 0.2 g KCl, pH 7.2), and then resuspended in binding buffer. Aliquots of 10⁵ cells were stained with 5 μl of ANXA5-EGFP and 10 μl of PI. Cells were analyzed using a BD FACSCalibur flow cytometer (BD Biosciences, San Diego, CA, USA). Data were analyzed using FlowJo software. The apoptotic morphology of SGC-7901 cells caused by treatments was observed by fluorescence microscopy and the fluorescence intensity of apoptotic cells in each treatment was also quantified with a microplate reader (TECAN Sirfire II, Switzerland).

Transmission electron microscopy

Five × 10⁶ SGC-7901 cells were seeded in 100-mm dishes overnight and were treated with w09 or DMSO for the indicated times. After treatment, cells were harvested and washed twice with cold PBS (4ºC). Then cells were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH7.2) at 4ºC for 1 h and subsequently postfixed with 2% osmium tetroxide as described previously.⁷⁰ Then cells were dehydrated with sequential washes in ethanol and then embedded in Epon812 (SPI, 02660-AB). The samples were then sectioned with a Reichert-Jung ultracut E ultramicrotome 701704 (Leica, Germany) and stained with uranyl acetate and lead citrate. Finally, samples were observed with a transmission electron microscope (FEI Tecnai G² 12, Eindhoven, Netherlands).

Transfections

mCherry-LC3B-pcDNA3.1 was a gift from David Rubinsztein (Addgene plasmid #40827).³⁸ EGFR-GFP was a gift from Alexander Sorkin (Addgene plasmid # 32751).⁷¹ mTagRFP-mWasabi-LC3B was a generous gift from Prof. Lin Jian (Peking University).³⁹ Transfection was performed using Lipofectamine 2000 transfection reagent (Invitrogen, L2000) according to the manufacturer's protocol. After a 6-h incubation, the transfection medium was replaced with fresh medium.

RNA interference

The knockdown of ATG5 was performed in SGC-7901 cells using specific small interfering RNA. ATG5 siRNA oligos were purchased from GenePhrama (Shanghai, China). Nontargeting siRNA sequence is UUCUCCGAACGUGUCACGUTT. The target sequence of ATG5 is GCUAUAUCAGGAUGAGAUATT. Briefly, SGC-7901 cells with 30% confluence were transfected with 50 pmol of siRNA duplex and 10 μL of siRNA-Mate transfection reagents (GenePharma, G04001) according to the manufacturer's protocol for 48 h. The transfection efficiency was evaluated by a Fluorescein Conjugate-A siRNA (GenePharma, A07001). The knockdown efficiency in SGC-7901 cells was confirmed by western blot.

Knockout of ATG7 or SQSTM1 genes by CRISPR-Cas9

Targeting sequences for CRISPR interference were designed at CRISPR direct (http://crispr.mit.edu), provided by the Zhang Lab, MIT (Massachusetts Institute of Technology) 2015. The double-knockout ATG5 and ATG7 SGC-7901 cell line was generated by CRISPR-Cas9 targeting the following site: 5′-CACCGAAGCTGAACGAGTATCGGCGTTT-3′. The targeting site for knockout of SQSTM1 gene was 5′- CGCTACACAAGTCGTAGTCTGG −3′. Briefly, 2 ATG7 or SQSTM1 complementary oligonucleotides with Bpil restriction sites for guide RNAs (gRNAs) were synthesized and cloned into pU6gRNACas9puro vector (GenePharma, C051005) and confirmed by sequencing, named pU6gRNA/SQSTM1/Cas9puro and pU6gRNA/ATG7/Cas9puro, respectively. SGC-7901 cells were transfected with pU6gRNA/SQSTM1/Cas9puro or pU6gRNA/ATG7/Cas9puro plasmid using Lipofectamine 2000 according to the manufacturer's instructions. 48 h after transfection, cells were selected for 7 d in the presence of 2 μg/mL puromycin (Sigma, P7255). Single colonies were isolated and the gene knockout clones were confirmed by immunoblotting.

Construction of cell lines

SGC-7901 or HeLa cell line stably expressing mCherry-LC3B was constructed as follows. Briefly, SGC-7901 or HeLa cells were transfected with a plasmid encoding mCherry-LC3B (Addgene #40827, David Rubinsztein lab),³⁸ using Lipofectamine 2000. Forty-eight h after transfection, cells were selected for 14 d in the presence of 400 μg/mL G418 (Sigma, A1720). Single colonies were isolated and the expressing of mCherry-LC3B were confirmed by fluorecsence microscopy and by immunoblotting. One SGC-7901 cell line stably expressing an LIR-domain-deficient SQSTM1/p62 was constructed with a retroviral expression system. Briefly, total cellular RNA from SGC-7901 cells was extracted with TRIzol reagent (Invitrogen, 15596–026) and first-strand cDNA was transcribed with a SuperScript III DNA polymerase kit (Invitrogen, 18080–044) according to the manufacturer's instructions, The SQSTM1 cDNAs were amplified by PCR with a pair of specific primers (primer sequences are available upon request). The purified PCR fragments were cloned into a pEASY®-Blunt Cloning Vector (Transgen Biotech, CB101–01). The identity of each cDNA clone was verified by DNA sequencing. The SQSTM1-deficient LIR domain mutant was constructed by overlap extension PCR with specific primers (5'-CAGGAATTGATCCGCATGGCGTCGCTCACCGTG-3'; 5'- GTAGACGGGTCCACTTCTTTTCCTCCTGAACAGTTATC-3'; 5'-AAAGAAGTG GACCCGTCTACAGGTGAACTCCAGTCCCT-3'; 5'-GGGCAATTCCGGATCTCAC AACGGCGGGGGATG-3') and were cloned into a pQCXIN vector (Clontech, 631514) between the NotI and BamHI sites. The resulting plasmids were confirmed by sequencing. The pseudotyped retroviral particles were prepared according to the manufacturer's instructions as described previously.⁷² To establish SGC-7901-derived cell lines only stably expressing the SQSTM1-deficient LIR domain, SQSTM1 knockout SGC-7901 cells were seeded into 6-well plates overnight and infected with 2 ml of pseudotyped retroviruses for 24 h. 48 h after infection, cells were screened with media containing 400 μg/ml G418 for 15 d. A single colony was isolated and the postive clones were confirmed by immunoblotting.

Measurement of autophagy

SGC-7901 cells or HeLa cells stably expressing the mCherry-LC3B reporter were seeded into 12-well plates at a density of 10⁴ cells per well overnight and then were treated with the indicated compounds or vehicle (DMSO) for the indicated time. The cellular fluorescent changes were observed under a laser scanning confocal microscope (Olympus FV1000, Tokyo, Japan). SGC-7901 cells were transfected with plasmids encoding mTagRFP-mWasabi-LC3B or enconding mCherry-EGFP-LC3B and were treated with the indicated compounds for 6 h. The fluorescence of mTagRFP-mWasabi-LC3B or enconding mCherry-EGFP-LC3B was observed under a laser scanning confocal microscopy (Olympus FV1000, Tokyo, Japan).

Preparation of cytosolic and mitochondrial fractionation

Mitochondrial and cytosolic fractions were obtained using the mitochondrial and cytosolic fractionation kit according to the manufacturer's instructions (Beyotime, P0027). Briefly, SGC-7901 cells were treated with w09 or vehicle (DMSO) for 48 h, then were harvested and washed twice with cold PBS. Cell pellets were resuspended in cytosolic lysis buffer containing PMSF for 15 min and then homogenized using a glass tissue grinder. The homogenates were centrifuged at 600 g for 10 min at 4°C, and then the supernatants were further centrifuged at 10, 000 g for 20 min at 4°C. The pellet was considered as the “mitochondrial” fraction and the supernatant fractions as the “cytosolic” fractions.

Western blot analysis

For western blots, cells were washed 2 times with PBS and lysed with a NuPAGE LDS lysis buffer (Invitrogen, LC2676) to get total protein lysates. Protein samples were determined using a protein assay kit (Beyotime, P0006). Cell lysates were loaded onto SDS-PAGE gels for protein separation, and then transferred onto nitrocellulose (NC) membranes (0.45 µm; Millipore, ISEQ00010) using a tank transfer system (Bio-Rad, USA). The membranes were blocked for 1 h in 5% fat-free dry milk in Tris-buffered saline (BST; 20 mM Tris, 166 mM NaCl, 0.05% Tween 20 [Sigma, P7949], pH 7.5), then probed with primary antibodies overnight at 4ºC, and finally incubated with horseradish peroxidase-conjugated species-specific secondary antibodies at room temperature for 1 h. Bands were visualized using an enhanced chemiluminescence kit (Millipore, P90720) with a Molecular Imager Chemidoc XRS System (Bio-Rad, USA). Quantification relative to ACTB by densitometric analysis was conducted using Quality One software (Bio-Rad).

In vivo evaluation of antitumor activity

Six-wk-old female athymic nude mice (BALB/c, nu/nu) were purchased from the Shanghai SLAC Animal Center (Shanghai, China). All BALB/c athymic nude mice were maintained under specific pathogen-free conditions, provided with sterilized food and water, and housed in positive pressure isolators with 12 h light/dark cycles. Aliquots of SGC-7901 cells (5 × 10⁶ cells) were injected subcutaneously into the left dorsal area of the mice. When the tumor volume reached approximately 100 mm³ in size, the tumor-bearing mice were randomly divided into vehicle control and treatment groups (6 mouse/group) and were given a daily intragastric administration of either 0 (vehicle control group) or 40 mg/kg w09 for 14 successive d. Taxel as a positive control was administrated via intraperitoneal injection at a dose of 2 mg/kg/3 day. The volume of the tumor was calculated from the formula V = 1/2 (A × B²), where A indicates the length of tumor and B means width of tumor measured by calipers. After treatment, all the mice were killed and the tumors were excised from the body for further analysis. All the experiments were performed in accordance with the Guidelines for Animal Experimentation of China Pharmaceutical University (Nanjing, China) and the protocols had been approved by the Animal Ethics Committee of this institution.

Statistical analysis

Data of this study were obtained from at least 3 independent experiments performed in triplicate with the error bars denoting S.D and were compared using a 2-tailed Student t test or a one-way ANOVA test followed by Tukey's multiple comparison tests. A value of P<0.05 was considered statistically significant.

Abbreviations

3-MA

3-methyladenine

ACTB

actin β

ANXA5

Annexin V

ATG5

autophagy-related 5

ATG7

autophagy-related 7

BafA1

bafilomycin A₁

CASP3

caspase 3

CASP8

caspase 8

CASP9

caspase 9

CQ

chloroquine

CYCS

Cytochrome c, somatic

DMSO

dimethyl sulfoxide

EGF

epidermal growth factor

EGFR

epidermal growth factor receptor

GFP

green fluorescent protein

LC3MAP1LC3/LC3

microtubule associated protein 1 light chain 3

LC3B-I

the nonlipidated form of LC3B

LC3B-II

the lipidated form of LC3B

LIR

LC3-interacting region

MAP1LC3B/LC3B

(microtubule associated protein 1 light chain 3 β)

MAP2K/MEK

mitogen-activated protein kinase kinase

MAPK

mitogen-activated protein kinase

MAPK1/ERK2

mitogen-activated protein kinase 1

MTOR

mechanistic target of rapamycin

MTT

3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide

PARP1

poly(ADP-ribose) polymerase 1

Rapa

rapamycin

RAS

RAS type GTPase family

RAF1

rapidly accelerated fibrosarcoma

SQSTM1

sequestosome 1

TEM

transmission electron microscopy

Z-VAD-fmk

benzyloxycarbony (Cbz)-l-Val-Ala-Asp (OMe)-fluoromethylketone.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Dr. Ju-Tao Guo for critical reading of this manuscript and thank Dr. Zhen-Zhou Jiang for technological guidance and help in fluorescence microscopy and flow cytometry. We thank Dr. Bashir Alsiddig Yousef, Ariful Islam, Sooro Mopa Alina for modification of language.

Funding

This research was supported by the Program for New Century Excellent Talents in University (Program No. NCET-12–0975); the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry; Jiangsu Overseas Research& Training Program for University Prominent Young & Middle-aged Teachers and Presidents (for Pinghu Zhang); the Project Program of Jiangsu Key Laboratory of Drug Screening (JKLDS2015ZZ-8), the Students Innovation and Entrepreneurship Training Program (3054070008 and 201610316097).