Justicidin AInduced Autophagy Flux Enhances ...

ORIGINAL RESEARCH ARTICLE

Journal of

Justicidin A-Induced Autophagy Flux Enhances Apoptosis of Human Colorectal Cancer Cells via Class III PI3K and Atg5 Pathway

Cellular Physiology

SHEN-JEU WON,1 CHENG-HSIN YEN,2,3 HSIAO-SHENG LIU,1,4 SHAN-YING WU,1 SHENG-HUI LAN,1 YA-FEN JIANG-SHIEH,5 CHUN-NAN LIN,6 AND CHUN-LI SU3* 1

Department of Microbiology and Immunology, College of Medicine, National Cheng Kung University, Tainan, Taiwan

2

Department of Nursing, Chang Jung Christian University, Tainan, Taiwan

3

Department of Human Development and Family Studies, National Taiwan Normal University, Taipei, Taiwan

4

Center of Infectious Disease and Signaling Research Center, College of Medicine, National Cheng Kung University, Tainan, Taiwan

5

Department of Anatomy, College of Medicine, National Cheng Kung University, Tainan, Taiwan

6

School of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan

Our previous reports showed that justicidin A (JA), a novel and pure arylnaphthalide lignan isolated from Justicia procumbens, induces apoptosis of human colorectal cancer cells and hepatocellular carcinoma cells, leading to the suppression of both tumor cell growth in NOD-SCID mice. Here, we reveal that JA induces autophagy in human colorectal cancer HT-29 cells by conversion of autophagic marker LC3-I to LC3-II. Furthermore, LC3 puncta and autophagic vesicle formation, and SQSTM1/p62 suppression were observed. Administration of autophagy inhibitor (bafilomycin A1 and chloroquine) and transfection of a tandem fluorescent-tagged LC3 (mRFP-GFP) reporter plasmid (ptfLC3) demonstrated that JA induces autophagy flux in HT-29 cells. Expression of LC3, SQSTM1, Beclin 1, and nuclear DNA double-strand breaks (representing apoptosis) were also detected in the tumor tissue of HT-29 cells transplanted into NOD-SCID mice orally administrated with JA. In addition, the expression of autophagy signaling pathway-related molecules p-PDK1, p-mTOR, p-p70S6k/p-RPS6KB2 was decreased, whereas that of class III PI3K, Beclin 1, Atg5-Atg12, and mitochondrial BNIP3 was increased in response to JA. Pre-treatment of the cells with class III PI3K inhibitor 3-methyladenine or Atg5 shRNA attenuated JA-induced LC3-II expression and LC3 puncta formation, indicating the involvement of class III PI3K and Atg5. A novel mechanism was demonstrated in the anticancer compound JA; pre-treatment with 3-methyladenine or Atg5 shRNA blocked JA-induced suppression in cell growth and colony formation, respectively, via inhibition of apoptosis. In contrast, administration of apoptosis inhibitor Z-VAD did not affect JA-induced autophagy. Our data suggest the chemotherapeutic potential of JA for treatment of human colorectal cancer. J. Cell. Physiol. 230: 930–946, 2015. © 2014 Wiley Periodicals, Inc.

Abbreviations: 3-MA, 3-methylaldenine; Atg, autophagy-related gene; AVOs, acidic vesicular organelles; BAF, bafilomycin A1; BNIP3, Bcl-2/adenovirus E1B 19 kDa-interacting protein 3; CQ, chloroquine; Cyto, cytosol; Dim, dimer; DMEM, Dulbecco’s modified Eagle medium; DMSO, dimethyl sulfoxide; GFP, green fluorescent protein; Hoech, Hoechst 33258; JA, justicidin A; LC3, microtubule-associated protein 1 light chain; Mito, mitochondrial; Mon, monomer; mTOR, mammalian target of rapamycin; MTT, 3-[4,5-Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; PBMC, peripheral blood mononuclear cells; PDK1, phosphoinositide-dependent kinase 1; PI, propidium iodide; PI3K, phosphatidylinositol-3-kinase; ptfLC3, a tandem fluorescent-tagged LC3 (mRFP-GFP) reporter plasmid; S.E.M., standard errors of the means; shRNA, short hairpin RNA; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick-end-labeling; ULK, uncoordinated 51-like kinase. Conflict of interest: None. Shen-Jeu Won and Cheng-Hsin Yen contributed equally to this work. © 2 0 1 4 W I L E Y P E R I O D I C A L S , I N C .

Contract grant sponsor: National Science Council; Contract grant numbers: NSC 96–2313-B-309–001-MY2, 98–2313-B-003–002-MY3, 101–2313-B-003–002-MY3. Contract grant sponsor: The Ministry of Economic Affairs; Contract grant number: 100-EC-17-A-17-S1–152. Contract grant sponsor: National Taiwan Normal University; Contract grant number: 99-D, 100NTNU-D-06. *Correspondence to: Chun-Li Su, Department of Human Development and Family Studies, National Taiwan Normal University, No. 162, Sec. 1, He-ping East Road, Taipei 10610, Taiwan. E-mail: [email protected] Manuscript Received: 2 April 2014 Manuscript Accepted: 5 September 2014 Accepted manuscript online in Wiley Online Library (wileyonlinelibrary.com): 12 September 2014. DOI: 10.1002/jcp.24825

930

JA-INDUCED AUTOPHAGY PROMOTES APOPTOSIS

Macroautophagy (hereafter referred to as autophagy) has been defined as programmed cell death type II, which is involved in the degradation and recycling of proteins and intracellular components in response to starvation and stress (Kondo et al., 2005). At the beginning of autophagy, phagophores interact with microtubule-associated protein 1 light chain (LC3). After sequestering of intracellular organelles such as Golgi apparatus and mitochondria, phagophores become autophagosomes. Autophagosomes then fuse with lysosomes and become autolysosomes (also known as degradative autophagic vacuoles). Anticancer therapies such as chemicals, irradiation, and hyperthermia have been reported to induce autophagy and result in autophagic cell death of cancer cells (Yang et al., 2011). Justicidin A (JA) is a new, pure (> 98.9% purity), and structurally defined arylnaphthalide lignan which our research group has managed to isolate from the whole plant of Justicia procumbens (Day et al., 2002), one of the most popular traditional Chinese medicines in China for inhibiting inflammation and helping digestion (Wang et al., 2011). It has also been used as an herbal remedy in Taiwan for fever, cancer, and pain due to pharyngo-laryngeal swelling (Fukamiya and Lee, 1986). Lignans are a class of phytoestrogens which are polyphenols widely found in edible plants. Recently, the health benefits of consuming lignan-rich foods, including their anticancer effects, have been discussed (Saarinen et al., 2007; Virk-Baker et al., 2010). Our previous findings demonstrated that JA induced apoptosis (programmed cell death type I) of human colorectal cancer cells (HT-29 and HCT 116 [Lee et al., 2005]) and human hepatocellular carcinoma cells (Hep 3B and Hep G2 [Su et al., 2006]) via mitochondrion- and caspase-related pathways at very low concentrations. It is noteworthy that diverse cancer cells were much more sensitive to JA compared to normal human peripheral blood mononuclear cells (PBMC). The 50% growth inhibition (IC50) of JA for PBMC was at least 200-fold and 500-fold higher than those for cancer HT-29 and Hep 3B cells, respectively. Our in vivo study further indicates that oral administration of JA significantly suppressed the growth of HT-29 and Hep 3B cells transplanted into NOD-SCID mice (Lee et al., 2005; Su et al., 2006). No significant changes in liver or spleen weight, or illnesses were observed. Colorectal cancer is the third most common malignancy and the third leading cause of cancer death in USA (Jemal et al., 2010). Although the combination of 5-fluorouracilbased chemotherapy with platinum agent oxaliplatin or the topoisomerase I inhibitor irinotecan (CPT-11) has improved overall survival rates, drug resistance remains a problem for treatment of these cancers (Hector and Prehn, 2009). Recent reports suggest that chemotherapy- or radiation-induced autophagy facilitates the resistance of cancer cells, and inhibition of autophagy enhances apoptosis of colorectal cancer cells (Carew et al., 2010; Chen et al., 2011). Since there is crosstalk between apoptosis and autophagy (Liu et al., 2011b), the present study was performed to determine if induction of autophagy is involved in the JA-induced anticancer mechanism. Our data indicate that JA induces autophagy in human colorectal cancer HT-29 cells in vitro and in vivo. Inhibition of autophagy significantly suppresses apoptosis of the cells. In addition, JA-induced autophagy promotes cell death and decreases colony formation of HT29 cells. Materials and Methods Materials JA was purified from the whole plants of Justicia procumbens which were collected from Chu-Shan, Nantou County, Taiwan. The airdried plants (5 kg) were chipped and extracted with MeOH at room temperature. The MeOH extract (150 g) was chromatoJOURNAL OF CELLULAR PHYSIOLOGY

graphed on a Si gel column, and JA (0.2 g) was eluted with cyclohexane-EtOAc (7:3) (Day et al., 2002). Most of the chemicals were obtained from Sigma (St. Louis, MO) unless otherwise indicated. Z-VAD was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). A tandem fluorescent-tagged LC3 (mRFP-GFP) reporter plasmid (ptfLC3) was a gift from Dr. T. Yoshimori (Department of Genetics, Osaka University Graduate School of Medicine, Japan). Cell culture Human colorectal cancer HT-29 cells (American Type Culture Collection, ATCC, Rockville, MD) were cultured in complete DMEM (Hyclone, Logan, UT) in an incubator at 37°C with a humidified atmosphere of 5% CO2. JA dissolved in dimethyl sulfoxide (DMSO) at a concentration of 1 mM was stored at 20°C. Before experiments, the JA stock was diluted to the final desired concentrations with complete DMEM. Control cells were cultured in medium containing an equal amount of DMSO (this concentration was tested and found to be nontoxic to the cells) without JA. Animal experiment and immunohistochemistry of tumor sections NOD-SCID mice at 6–7 weeks of age obtained from the Animal Center of National Cheng Kung University (NCKU, Tainan, Taiwan) were housed at the Animal Center in a temperaturecontrolled and air-conditioned environment with a 10/14 h light/ dark cycle. Food and water were provided ad libitum. All animal experiments were approved by the Animal Research Committee of NCKU and were performed under the guidelines of the National Research Council, Taiwan. HT-29 cells were implanted subcutaneously to the flank of the mice at day 0. At day 4, the treatment group of mice received oral administration of JA (6.2 mg/mouse/day) for 56 consecutive days, and the control group received vehicle (0.05% DMSO) on the same schedule (Lee et al., 2005). Tumor sections (5 mm) were cut using Cryotome0620 (Thermo Shandon, Waltham, MA) then subjected to apoptosis and autophagy analysis (Su et al., 2011). Terminal deoxynucleotidyl transferase-mediated dUTP nickend-labeling (TUNEL) method (ApoAlert DNA Fragmentation assay Kit, Clontech, Palo Alto, CA) was performed to determine DNA double-strand breaks in apoptotic nuclei according to the manufacturer's protocol, and the incorporated fluoresceindUTP at the free 3'-hydroxyl ends of fragmented DNA was detected. LC3-II puncta in autophagic cells was determined by incubating with rabbit anti-LC3 antibody (Abgent, San Diego, CA) overnight at 4°C followed by incubation with goat antirabbit Alexa Fluor568-conjugated secondary antibody (Molecular Probes, Inc., Eugene, OR; 1:450) in blocking solution for 2 h at room temperature. After washing, the nuclei were visualized by incubating with Hoechst 33258 (0.05 mg/ml in PBS) for 10 min at room temperature. The signals were examined using a fluorescence microscope (Leica DMRBE microscope). Determination of autophagy in cultured cells by confocal microscopy HT-29 cells were fixed with methanol at 20°C for 20 min and then permeabilized with 0.1% Triton X-100 for 30 min, as previously described (Liu et al., 2011a). After washing with PBS and blocking with 2% skim milk, the cells were stained with rabbit anti-LC3 antibody (Abgent) and incubated at 4°C overnight. After washing, the cells were then stained with Fluorescein (FITC)-conjugated affiniPure goat-anti-rabbit IgG (Jackson ImmunoResearch, West Grove, PA) to stain the LC3-II puncta. Nuclei were visualized by incubating the cells with

931

932

WON ET AL.

propidium iodide (PI; 5 mg/ml) for 30 min. The signal was examined immediately using a confocal laser scanning microscope system (Leica TCS SP2).

Fractionation of cellular proteins

HT-29 cells were fixed with 2.5% glutaraldehyde in 0.1 mM cacodylate buffer containing 4% sucrose, 1 mM MgCl2, and 1 mM CaCl2, then post-fixed in 1% osmium tetroxide, as described previously (Lee et al., 2008). Following dehydration with ethanol, the cells were embedded in Epon-Araldite mixture (Electron Microscopy Sciences, Hatfield, PA). Ultrathin sections obtained using an ultramicrotome (Reichert-Jung, Heidelberg, Germany) were stained with uranyl acetate and lead citrate. The images were examined with a Hitachi 7000 transmission electron microscope (Hitachi, Tokyo, Japan).

As previously described (Lee et al., 2005), whole cells were lysed with lysis buffer. After centrifugation at 15,000g for 10 min, the resulting supernatants were used as the total cell lysates for immunoblotting. For preparation of mitochondrial and cytosol fractions, cells were harvested by centrifugation at 800 g for 10 min. The pellets were resuspended in TSE buffer and broken with 10 strokes of a Teflon pestle in a Dounce homogenizer (Glas-Col, Terre Haute, IN). The homogenates were centrifuged at 750 g for 30 min. The supernatants were centrifuged at 12,000 g again for 30 min. The lysed solutions were centrifuged again at 100,000 g for 1 h. The resulting supernatants were used for cytosolic fractions. The obtained pellets were incubated with lysis buffer. The lysed solutions were used as mitochondrial fractions for immunoblotting.

Flow cytometric analysis

Western blot analysis

The percentage of HT-29 cells that had undergone apoptosis and autophagy was determined by flow cytometry (Papandreou et al., 2008). For apoptosis analysis, cells washed with PBS were stained with PI (50 mg/ml) and/or annexin V-FITC (0.5 mg/ml; Biovision, Milpitas, CA) (Yang et al., 2007). For cell cycle analysis, cells were stained with PI and the PI-stained cells were sorted by flow cytometry based on their DNA contents. Apoptotic signaling cascades ultimately lead to fragmentation of DNA (Igney and Krammer, 2002) followed by disintegration of the nucleus and by budding of the cells as a whole to form a membrane-bounded apoptotic body (Kerr et al., 1994; Wickman et al., 2012). The apoptotic cells with reduced DNA content (< 2N) produce a peak at the sub-G1 position (Lee et al., 2005). For quadrant dot blot analysis, cells were stained with both PI (50 mg/ml) and annexin V-FITC. During apoptosis, annexin V-FITC stained the exposed phosphatidylserine on the cell surface prior to the loss of the ability to exclude cationic dye PI (Wlodkowic et al., 2012), whereas necrotic cells lost membrane integrity early on and were quickly stained with PI. Therefore, the percentages of PI-positive, double positive, and annexin V-positive cells in the quadrant represent the proportion of necrotic, late apoptotic, and early apoptotic cells, respectively. The double negative cells are the normal cells. For the autophagy analysis, cells stained with acridine orange (1.5 mg/ml) for 15 min were sorted in an immunocytometry system (FACS200, Becton Dickinson, Lexington, KY). Autophagy requires formation of autophagosomes, which therefore fuse with endosomes/lysosomes to form mature acidified autolysosomes (Klionsky et al., 2007), which can be quantified by staining cells with acridine orange and examining the increase in red fluorescence by flow cytometry (Aoki et al., 2007). Data were analyzed using WinMDI 2.8 (Windows Multiple Document Interface Flow Cytometry Application; Scripps Research Institute, San Diego, CA).

Protein contents in whole cell lysates, as well as mitochondrial and cytosol fractions were determined by a protein assay kit (Bio-Rad Laboratories, Hercules, CA). Proteins were resolved using 8–12% SDS-PAGE and subsequently transferred to polyvinylidene fluoride membranes (Millipore Corp., Billerica, MA). Western blot analysis was performed (Su et al., 2007) using antibodies obtained from the following sources: anti-PDK1, anti-class I phosphatidylinositol-3-kinase (PI3K), and anti-Rack 1 antibodies were obtained from BD Transduction Laboratories (Texarkana, TX); anti-p-PDK1, anti-mTOR, anti-p-mTOR, anti-p70S6k, and anti-p-p70S6k antibodies were obtained from Cell Signaling Technology (Danvers, MA); anti-Beclin 1, antiAtg5, anti-LC3, and anti Bcl-2/adenovirus E1B 19 kDa-interacting protein 3-(BNIP3) antibodies were obtained from Abcam (Cambridge Science Park, Cambridge, UK); anti-class III PI3K antibody was obtained from Abgent. Anti-caspase 3, anti-class I p-PI3K p85 a, and anti-SQSTM1/p62 primary antibodies and goat anti-mouse conjugated horseradish peroxidase secondary antibody were obtained from Santa Cruz Biotechnology; Peroxidase-conjugated AffiniPure goat anti-rabbit antibody was obtained from Jackson ImmunoResearch.

Transmission electron microscopy

Transfection of short hairpin RNA (shRNA) autophagy-related gene (Atg) 5 and ptfLC3 plasmid As previouslydescribed (Chang et al., 2010), Atg5 was knocked down by lentiviral transduction to express stable shRNA against human Atg5. Lentiviral transduction to express stable shRNA against green fluorescent protein (GFP; an Atg5-unrelated shRNA) was used as a negative control (Byun et al., 2009). HT-29 cells were incubated with the viral particles overnight. Protein expression of Atg5 was determined by Western blot. Autophagic cells were examined by confocal microscopy. As previously described (Wu et al., 2011), plasmid ptfLC3 was transfected into HT-29 cells using LipofectamineTM 2000 (Invitrogen, Carlsbad, CA) in DMEM without supplementation of FBS, according to the manufacturer's instructions. JOURNAL OF CELLULAR PHYSIOLOGY

Cell growth assay Growth of cells was examined using a modified colorimetric 3-[4,5Dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) assay (Lee et al., 2005). Cells were seeded in 96-well/plates. After treatment, DMEM was removed to avoid color interference of the MTT assay. MTT at a final concentration of 0.5 mg/ml in PBS was added to each well. The MTT was taken up and converted to a purple formazan. After 4 h of incubation, 100 ml of 10% sodium dodecylsulphate in 0.01 N HCl was added to each well to dissolve the formazan. The absorbance of each well was determined at 590 nm in an ELISA Reader (MRX II, Thermo Fisher Scientific Inc., Waltham, MA). Colony formation assay Two layers of agar were made in 6-well plates (Lee et al., 2005). The base layer was 1 ml 0.6% solid basal agar containing complete DMEM. The upper layer was made by mixing cells (6 104 cells/ well) in 0.1 ml with 0.9 ml 0.33% agar containing complete DMEM in the presence or absence of JA at 37°C. Colonies with a diameter greater than 1 mm were counted 14 days later. Statistical analyses The results were expressed as means standard errors of the means (S.E.M.). The data were analyzed by t-test and one-way ANOVA. Differences among groups were analyzed by Duncan's


multiple range test (SPSS software, version 14.0). A P-value of < 0.05 was considered significant. Results JA induced autophagy in vitro and in vivo

LC3 is involved in the formation of phagophores, in which LC3-I is converted to a lipidated form of LC3-II leading to its translocation from the cytosol to the phagophores (Xiao, 2007). Since LC3-II stays with autophagosomes until fusion with lysosomes is completed, LC3-II has been used as a marker of autophagy (Kondo and Kondo, 2006). In this study, time(Fig. 1A) and dosage-related (Fig. 1B) increases of LC3-II expression were observed in JA-treated HT-29 cells. JA also increased the mRNA expression of LC3B in HT-29 cells (Fig. S1). In autophagic cells, LC3-II is associated with the membrane of phagophores or autophagosomes, appearing as bright puncta under a fluorescent microscope (Liang et al., 2006). As shown in Figure 1C and D, a time- and dosage-related increase in the intensity of LC3 puncta was observed. The antibody used in the figures exhibited better affinity to LC3-II (Fig. S2). Electron microscopic investigation is still the most reliable method for monitoring autophagic morphology (Mizushima, 2004). As shown in Figure 1E, far more double-membrane vesicles containing subcellular materials, representing formation of autophagosomes, were seen in the JA-treated cells compared with the non-treated control cells. During the process of autophagy, autophagosomes fuse with lysosomes to form mature acidified autolysosomes (Klionsky et al., 2007). To further confirm the induction of autophagy by JA, the formation of the acidic vesicular organelles (AVOs) representing autolysosomes was quantified. As shown in Figure 2A, JA showed a time- and dosage-dependent increase of bright red fluorescence intensity. Although the control cells reached confluence before 72 h, the confluence status did not affect the JA-induced autophagy (Fig. S3). SQSTM1, a major interaction partner for LC3, is degraded in autolysosomes (Pankiv et al., 2007). As shown in Figure 2B, JA increased SQSTM1 in a timerelated manner between 3 and 48 h. A dramatic decrease of SQSTM1 at 72 h (Fig. 2B) suggested that JA induced formation of autolysosomes. Treatment of the cells with various concentrations of JA for 24 h resulted in a significant increase in SQSTM1 (Fig. 2C). To further demonstrate the autophagic activity (flux) induced by JA, bafilomycin A1(BAF) was used to prevent the fusion of autophagosomes and lysosomes (Eskelinen and Saftig, 2009). It is noteworthy that administration of BAF caused accumulation of LC3-II and SQSTM1 (Fig. 2D) and attenuated the formation of JA-induced AVOs (Fig. 2E). Chloroquine (CQ) is another autophagy inhibitor, which increases intralysosomal pH and impairs autophagic protein degradation (Lotze et al., 2012). As shown in Figure 2F, addition of CQ caused accumulation of LC3-II and SQSTM1. The increase in LC3-I may be due to the upregulation of LC3 mRNA (Kimura et al., 2007a) and LC3 protein (Huang et al., 2011) in response to CQ. To confirm that JA indeed induced the autophagy flux, a tandem fluorescent-tagged LC3 (mRFPGFP) reporter plasmid (ptfLC3) was transfected into HT-29 cells (Wu et al., 2011). The GFP-LC3 loses fluorescence due to acidic and degradative conditions in autolysosomes, while mRFP is resistant. Therefore, a GFP and mRFP signal was detected before fusion with lysosomes occurred (merged image exhibits yellow fluorescence), but only the mRFP signal was detected after fusion with lysosomes (merged image exhibits red fluorescence) (Kimura et al., 2007b). JA induced red puncta in the merged image, which indicates formation of autolysosomes by fluorescence microscopy (Fig. 2G) and by confocal microscopy (Fig. 2H). It is noteworthy that combination treatment of JA with CQ changed the puncta color to yellow, representing accumulation of autophagosomes JOURNAL OF CELLULAR PHYSIOLOGY

resulting from blocked JA-induced autophagy. Taken together, these findings demonstrated that JA treatment induced autophagy flux in HT-29 cells. Our previous results showed that oral administration of JA significantly suppressed the growth of HT-29 cells transplanted into NOD-SCID mice (Fig. S4) (Lee et al., 2005). To determine if the decrease in tumor size was due to the induction of autophagy in vivo, LC3 expression in tumor was investigated. The antibody used in the experiment has better affinity to LC3-II (Fig. S2). As shown in Figure 2I, only tumor sections obtained from mice that received JA showed significant red fluorescence compared with that in the control mice, indicating the induction of autophagy by JA. Our previous results indicate that JA induced apoptosis of HT-29 cells (Lee et al., 2005). The condensed chromatin in JA-treated samples may have been due to the induction of apoptosis. In summary, JA induced autophagy in vivo, which is consistent with the result that showed JA induced autophagy in vitro. JA inhibited mTOR but activated class III PI3K, Atg5, and BNIP3 pathways in HT-29 cells

Increased class III PI3K and decreased class I PI3K activities are involved in autophagy (Kondo et al., 2005). Class I PI3K downstream molecule mTOR is also decreased during the progress (Yang et al., 2011). As shown in Figure 3A, phosphorylation level of PDK1, an upstream regulator of mTOR (Zhang et al., 2011), was downregulated. Consistently, the phosphorylation levels of mTOR and its downstream molecule p70S6k (Liu et al., 2010) were significantly suppressed (Fig. 3B). Furthermore, the total protein and phosphorylation of class I PI3K expression, an upstream regulator of PDK1 (Liu et al., 2011b), remained unchanged (Fig. 3C). A positive and negative control for PDK1, mTOR, and class I PI3K were shown (Fig. 3D). The above data indicate that JA-induced autophagy occurs via downregulation of the PDK1/mTOR/ p70S6k signaling pathway. It has been reported that the class III PI3K-Beclin 1 complex is essential for the recruitment of Atg products during autophagosome formation (Xiao, 2007; Yang et al., 2011). The covalent linkage of Atg12 to Atg5 leads to the formation of Atg12-Atg5-Atg16 complex, which further initiates elongation and curvature of phagophores by recruiting and converting LC3-I to LC3-II (Xiao, 2007). As shown in Figure 3E, an increase in the expression of class III PI3K was observed in JA-treated cells. A dramatic elevation of Beclin 1 expression was also seen. In Figure 3F, a large increase in the level of Atg5-Atg12 complex was detected as early as 3 h. Furthermore, Atg5 alone was increased at 12 h (Fig. 3F). These data suggest that JA-induced autophagy takes place via activation of class III PI3K/Beclin 1 and formation of Atg5 complex. BNIP3 is a BH3-only Bcl-2 family protein. Overexpression of BNIP3 induces various types of cell death (Burton and Gibson, 2009). Residues of serine172 and histidine173 form hydrogen bonds to yield BNIP3 homodimers which act as proton channels in the outer mitochondrial membrane (Bocharov et al., 2007). Homo-dimer of BNIP3 induces autophagic cell death by insertion into the mitochondrial outer membrane where it is involved in the fusion of autophagosomes and lysosomes as well as the conversion of LC3-I to LC3-II (Burton and Gibson, 2009). In JA-treated cells, the total BNIP3 protein levels of both monomer and dimer forms remained unchanged under the experimental conditions (Fig. 3G). However, the significant decrease in cytosol BNIP3 (Fig. 3G) and the significant increase in the mitochondrial BNIP3 (Fig. 3H) suggest movement of the BNIP3 from the cytosol to the mitochondria. Taken together, JA was shown to induce autophagy through suppression of mTOR expression and activation of class III PI3K/Beclin 1, Atg5, and BNIP3 signaling pathways in HT-29 cells.

933

934

WON ET AL.

Fig. 1. JA induced LC3-II expression in HT-29 cells. (A) Time-related increase in LC3 protein expression. (B) Dosage-related increase in LC3 protein expression. After treatment, whole cell lysates were subjected to immunoblotting. Anti-LC3 antibody served as a probe. Rack 1 served as a loading control. The intensity of each LC3-II protein expression band of at least three independent experiments was quantified by densitometry normalizing to that of Rack 1, with control level arbitrarily set to one. (C) Time-related increase in LC3 puncta pattern. (D) Dosage-related increase in LC3 puncta pattern. After treatment, HT-29 cells were stained with anti-LC3 (green). DNA morphologies were visualized by incubating the cells with PI (red). The images were taken by confocal microscopy. The cells with green fluorescence of at least three independent experiments were quantified as a percentage of total cell number per field. (E) Ultrastructure of JA-treated cells. The autophagic vesicles with double-membrane (the lower panel) were examined by electron microscopy. The arrows indicate autolysosomes/ autophagosomes. Data are presented as means S.E.M. *denotes a significant difference from the control group, P < 0.05. Results are representative of three independent experiments.

JOURNAL OF CELLULAR PHYSIOLOGY

Fig. 2. JA induced formation of autophagic vesicles in HT-29 cells and in animals. (A) Quantification of AVOs. After treatment, cells were stained with acridine orange. Formation of AVOs was determined by green and red fluorescence using flow cytometry. Accumulation of AVOs results in increased red light emission. The percentages in the figure indicate the proportion of cells (upper two quadrants) with AVOs staining of at least three independent experiments. (B) Time-related expression of SQSTM1 protein. (C) Dose-related expression of SQSTM1 protein. After treatment, whole cell lysates were subjected to immunoblotting. Anti-SQSTM1 antibody served as a probe. Rack 1 served as a loading control. (D) BAF caused accumulation of LC3-II and SQSTM1. After pretreated with BAF (1 mM) for 2 h, HT-29 cells were incubated with or without JA (1 mM) for 48 and 72 h. Whole cell lysates were subjected to immunoblotting. Anti-LC3 and anti-SQSTM1 antibodies served as probes. Rack 1 served as a loading control. (E) BAF inhibited JA-induced AVOs formation. After treatment, cells were stained with acridine orange. The percentages in the figure indicate the proportion of cells (upper two quadrants) with AVOs staining of at least three independent experiments. (F) CQ caused accumulation of JA-induced LC3-II and SQSTM1. HT-29 cells were treated with JA for 120 h. CQ was added at 48 h after JA induction. Whole cell lysates were subjected to immunoblotting. Anti-SQSTM1 and anti-LC3 antibodies served as probes. b-actin served as a loading control. (G) JA induced autophagic flux using fluorescence microscopy. (H) JA induced autophagic flux using confocal microscopy. After transfection with plasmid ptfLC3 for 24 h, cells were treated with JA for 120 h. CQ was added at 48 h after JA induction. DNA morphologies were visualized by incubating the cells with Hoechst 33258 (Hoech, blue). White arrow indicates the spot of autophagosomes (red RFP and green GFP), and yellow arrow indicates the spot of autolysosomes (red RFP only). The green and red spots of at least three independent experiments were quantified. At least 50 cells were counted for each experiment. (I) JA-induced autophagy of HT29 cells implanted into NOD-SCID mice. Sections of tumors were stained with Hoechst 33258 (Hoech, blue) and anti-LC3 antibody (red) to respectively visualize cell nuclei and autophagic puncta pattern of LC3 imaging by fluorescence microscopy. Data are presented as means S.E.M. *denotes a significant difference from the control group, P < 0.05. denotes a significant difference from the cells treated with JA alone, P < 0.05. Results are representative of three independent experiments.


936

WON ET AL.

Fig. 2.

(Continued)

Class III PI3K inhibitor or Atg5 shRNA blocked the JA-induced autophagy, enhancing cell growth and colony formation

Flow cytometric analysis revealed that class III PI3K inhibitor 3-MA (Kondo et al., 2005) reversed the formation of JAinduced AVOs (Fig. 4A). Western blot analysis further confirmed that 3-MA attenuated the JA-induced LC3-II expression in HT-29 cells (Fig. 4B). Suppression in LC3 puncta formation by 3-MA in JA-treated HT-29 cells was consistently observed under confocal microscopy (Fig. 4C). To demonstrate JA-induced autophagy, Atg5 shRNA was transfected into HT-29 cells to knockdown Atg5 protein expression (Fig. 4D). Immunoblotting (Fig. 4E) and confocal microscopic analysis (Fig. 4F) indicated that Atg5 shRNA totally blocked JAinduced LC3-II expression and LC3 puncta in HT-29 cells, respectively. These findings indicate that class III PI3K and JOURNAL OF CELLULAR PHYSIOLOGY

Atg5 are required for JA-induced autophagy and further suggest the involvement of class III PI3K in the recruitment of Atg5 to form phagophores and conversion of LC3-I to LC3-II. To elucidate the role of JA-induced autophagy in colorectal cancers, growth inhibition was examined. Anchoragedependent growth (Guo et al., 2005) was evaluated using MTT assay in the presence and absence of 3-MA. As shown in Figure 4G, addition of 3-MA blocked the JA-induced cell death, suggesting that the autophagy induced by JA contributes to the death of colorectal cancer cells. The effect of JA-induced autophagy on anchorage-independent growth (Lawson et al., 2012) was also evaluated. As shown in Figure 4H, JA alone inhibited the formation of tumor colony and combination with Atg5 shRNA restored the number of colonies. The observations described above demonstrate that JA-induced autophagy leads to inhibition in both anchoragedependent and -independent cancer cell growth.


Fig. 2. (Continued)

JA-induced autophagy promoted apoptosis whereas JA-induced apoptosis did not result in autophagy

To determine if JA can induce autophagy and apoptosis simultaneously in vivo, sections of tumors from mice were JOURNAL OF CELLULAR PHYSIOLOGY

examined. As shown in Figure 5A, treatment of JA induced green fluorescence at the nuclei by TUNEL staining, representing nuclear DNA double-strand breaks, a hallmark of apoptosis (Nagata, 2000). On the same section of tumor, JA also significantly induced LC3 red fluorescence, indicating the

937

938

WON ET AL.

Fig. 3. JA induced expression of proteins in autophagic signaling transduction in HT-29 cells. (A) Protein expression of p-PDK1. After treatment, whole cell lysates were subjected to immunoblotting. Anti-p-PDK1 and anti-PDK1 served as probes. (B) Protein expression of pmTOR and p-p70S6k. Whole cell lysates were subjected to immunoblotting. Anti-p-mTOR, anti-mTOR, anti-p-p70S6k, and anti-p70S6k served as probes. (C) Protein expression of class I p-PI3K. Whole cell lysates were subjected to immunoblotting. Anti-class I p-PI3K and anticlass I PI3K served as probes. (D) Protein expression of p-mTOR, class I p-PI3K, p-PDK1, SQSTM1, and LC3-II in response to rapamycin (a positive control for induction of autophagy) and 3-MA (a negative control for induction of autophagy). HT-29 cells were treated with JA (1 mM), rapamycin (100 nM), or 3-MA (10 mM) for 48 h. Whole cell lysates were subjected to immunoblotting. Anti-p-mTOR, anti-class I pPI3K, anti-p-PDK1, anti-SQSTM1, and anti-LC3 antibodies served as probes. (E) Protein expression of class III PI3K and Beclin 1. Whole cell lysates were subjected to immunoblotting. Anti-class III PI3K and anti-Beclin 1 served as probes. (F) Protein expression of total Atg5 and Atg5-Atg12 complex. Whole cell lysates were subjected to immunoblotting. Anti-Atg5 antibody served as a probe to detect Atg5 alone and covalent linkage of Atg5-Atg12 complex. (G) Expression of total and cytosol BNIP3. After treatment, whole cell lysates and cytosol fractions were subjected to immunoblotting. Anti-BNIP3 served as a probe to detect BNIP3 monomer and covalent linkage of BNIP3 dimer. (H) Protein expression of mitochondrial BNIP3. After treatment, mitochondrial fractions were subjected to immunoblotting. Anti-BNIP3 served as a probe to detect BNIP3 monomer and covalent linkage of BNIP3 dimer. Rack 1 served as a loading control. The intensity of each protein expression band of at least three independent experiments was quantified by densitometry normalizing to that of Rack 1, with control level arbitrarily set to one. Data are presented as means S.E.M. *denotes a significant difference from the group at time 0, P < 0.05. Results are representative of three independent experiments. dim, dimer; mon, monomer; Cyto, cytosol; Mito, mitochondrial.



Fig. 3. (Continued)

formation of phagophores/autophagosomes. It is noteworthy that JA-induced autophagy was observed in the presence and absence of apoptosis whereas JA-induced apoptosis was observed mainly in combination with autophagy because almost every nucleus with green TUNEL staining was surrounded with red fluorescence and several other blue nuclei were also surrounded with red puncta (Fig. 5B). These findings revealed that JA induces both apoptosis and autophagy in HT-29 cells transplanted into NOD-SCID mice. Consistent with the in vitro results (Figs. 2B, C and 3E), formation of JOURNAL OF CELLULAR PHYSIOLOGY

SQSTM1 puncta in some areas and an increase in Beclin 1 expression were observed in tumor sections (Fig. 5C). To characterize the relationship between JA-induced apoptosis and JA-induced autophagy, HT-29 cells were cultured with JA in the presence and absence of autophagy inhibitors (3-MA or Atg5 shRNA) or apoptosis inhibitor (Z-VAD). In agreement with our previous findings (Lee et al., 2005), JA induced early and late apoptosis in a time- and dosage-dependent manner (Fig. 5D). Since cell cycle analysis (PI staining) had been used previously to determine JA-induced apoptosis in a time- and

939

Fig. 4. Effect of autophagic inhibitors in JA-treated HT-29 cells. (A) 3-MA inhibited JA-induced AVOs formation by flow cytometry. HT-29 cells were pretreated with or without 3-MA (10 mM) for 2 h prior to the addition of JA for 24 h. The percentages in the figure indicate the proportion of cells (upper two quadrants) with AVOs staining. (B) 3-MA inhibited JA-induced expression of LC3-II by immunoblotting. HT-29 cells were pretreated with or without 3-MA (10 mM) for 2 h prior to the addition of JA for 24 h. Whole cell lysates were subjected to immunoblotting. Anti-LC3 antibody served as a probe. Rack 1 served as a loading control. The intensity of each protein expression band of at least three independent experiments was quantified by densitometry normalizing to that of Rack 1, with control level arbitrarily set to one. (C) 3-MA inhibited JA-induced expression of LC3 puncta pattern was detected by confocal microscopy using anti-LC3 antibody. HT-29 cells were pretreated with or without 3-MA (10 mM) for 2 h prior to the addition of JA for 24 h. After treatment, cells were stained with anti-LC3 (green). DNA morphologies were visualized by incubating the cells with PI (red). The images were taken by confocal microscopy. The cells with green fluorescence were quantified as a percentage of total cell number per field. (D) Atg5 shRNA inhibited JA-induced protein expression of Atg5 was detected by immunoblotting using anti-Atg5 antibody. Cells were infected with lenti-virus shRNA Atg5 or shRNA GFP (a negative control) for 16 h. The cells were then treated with or without JA for 24 h. Whole cell lysates were subjected to immunoblotting. Anti-Atg5 antibody served as a probe to detect Atg5 alone and covalent linkage of Atg5-Atg12 complex. Rack 1 served as a loading control. The intensity of each protein expression band of at least three independent experiments was quantified by densitometry normalizing to that of Rack 1, with control level arbitrarily set to one. (E) Atg5 shRNA inhibited JA-induced protein expression of LC3-II was detected by immunoblotting using anti-LC3 antibody. Cells were infected with lenti-virus shRNA Atg5 or shRNA GFP (a negative control) for 16 h. The cells were then treated with or without JA for 24 h. After treatment, whole cell lysates were subjected to immunoblotting. Anti-LC3 antibody served as a probe. Rack 1 served as a loading control. The intensity of each protein expression band of at least three independent experiments was quantified by densitometry normalizing to that of Rack 1, with control level arbitrarily set to one. (F) The Atg5 shRNA inhibited JAinduced LC3 puncta pattern was detected by confocal microscopy. Cells were infected with lenti-virus shRNA Atg5 or shRNA GFP (a negative control) for 16 h. The cells were then treated with or without JA for 24 h. After treatment, cells were stained with Hoechst 33258 (Hoech, blue), antiLC3 antibody (red), and GFP (green) to respectively visualize cell nuclei, autophagic puncta pattern of LC3, and GFP in cytosol. The cells with red fluorescence were quantified as a percentage of total cell number per field. (G) JA-induced cell growth inhibition was attenuated by 3-MA. Cells were pretreated with or without 3-MA (10 mM) for 2 h prior to the addition of JA (1 mM) for 24 h. MTT was added to determine the growth of the cells. (H) Atg5 shRNA reversed JA-induced suppression in colony formation. Cells were infected with lenti-virus shRNA Atg5 or shRNA GFP (a negative control) for 16 h. The cells cultured in agarose were then treated with or without JA for 24 h. After 14 days, the colonies were counted. Data are presented as means S.E.M. *denotes a significant difference from the vehicle group, P < 0.05. denotes a significant difference from the cells treated with JA alone, P < 0.05. Results are representative of three independent experiments.



Fig. 4. (Continued)


941

Fig. 5. The interaction of autophagy and apoptosis. (A) JA-induced autophagy and apoptosis of HT-29 cells implanted into NOD-SCID mice. Sections of tumors were stained with Hoechst 33258 (Hoech, blue), fluorescein-dUTP (TUNEL assay, green), and anti-LC3 antibody (red) to respectively visualize cell nuclei, apoptotic nuclei, and autophagic puncta pattern of LC3 imaging by fluorescence microscopy. (B) JA induced autophagy without induction of apoptosis in several parts of tumor sections. Sections of tumors were stained with Hoechst 33258 (Hoech, blue), fluorescein-dUTP (TUNEL assay, green), and anti-LC3 antibody (red) to respectively visualize cell nuclei, apoptotic nuclei, and autophagic puncta pattern of LC3 imaging by fluorescence microscopy. (C) Expressions of SQSTM1 and Beclin 1 in HT-29 cells implanted into NOD-SCID mice. Sections of tumors were stained with anti-SQSTM1 (green) or anti-Beclin 1 antibody (green). Hoechst 33258 (Hoech, blue) was used to visualize cell nuclei. The images were taken by fluorescence microscopy. (D) JA induced early and late apoptosis of HT-29 cells. After treatment, cells were stained with annexin V-FITC (0.5 mg/ml) and PI (50 ng/ml) for 15 min before flow cytometry. The percentages in the upper left, upper right, and lower right quadrant indicate the proportion of necrotic, late apoptotic, and early apoptotic cells, respectively. The normal cells are located at the lower left quadrant. (E) 3-MA inhibited JA-induced apoptosis of HT-29 cells. Cells were pretreated with or without 3-MA (10 mM) for 2 h prior to the addition of JA (1 mM) for 24 h. After treatment, cells were stained with annexin V-FITC and PI before flow cytometry. The percentages in the upper left, upper right, and lower right quadrant indicate the proportion of necrotic, late apoptotic, and early apoptotic cells, respectively. The normal cells are located at the lower left quadrant. (F) shRNA Atg5 suppressed JA-induced apoptosis of HT-29 cells. Cells were infected with lenti-virus shRNA Atg5 or shRNA GFP (a negative control) for 16 h. The cells were then treated with or without JA for 24 h. After treatment, cells were stained with PI before flow cytometry. The percentages in the figure indicate the proportion of apoptotic cells at the sub-G1 phase. (G) General caspase inhibitor Z-VAD confirmed the suppression of JA-induced apoptosis. Cells were pretreated with or without Z-VAD (50 mM) for 3 h prior to the addition of JA for 24 h. After treatment, cells were stained with PI before flow cytometry. The percentages in the figure indicate the proportion of apoptotic cells. (H) Z-VAD did not change JA-induced protein expression of LC3-II in HT-29 cells. Cells were pretreated with or without Z-VAD (50 mM) for 3 h prior to the addition of JA (1 mM) for 24 h. After treatment, whole cell lysates were subjected to immunoblotting. Anti-caspase 3 and anti-LC3 antibodies served as probes. Rack 1 served as a loading control. (I) Z-VAD did not affect JA-induced formation of AVOs. Cells were pretreated with or without Z-VAD (50 mM) for 3 h prior to the addition of JA (1 mM) for 24 h. After treatment, cells were stained with acridine orange before flow cytometry. The percentages in the figure indicate the proportion of cells (upper two quadrants) with AVOs staining. Data are presented as means S.E.M. Results are representative of three independent experiments.



Fig. 5. (Continued)

dosage-related manner, annexin V/PI double staining method was carried out in the present study not only to confirm JAinduced apoptosis but also to further examine the stages of apoptosis in response to JA. Based on our observations of JAinduced apoptosis of HT-29 cells, the percentages in the upper right quadrant using annexin/PI staining method were found to be very similar to the percentages at the sub-G1 phase using cell cycle analysis. It is noteworthy that inhibition of autophagy by either 3-MA (Fig. 5E) or Atg5 shRNA (Fig. 5F) blocked the increase in JA-induced apoptotic cells, suggesting that JAinduced autophagy helps JA-induced apoptosis proceed. Our JOURNAL OF CELLULAR PHYSIOLOGY

previous discovery indicated that JA-induced apoptosis in HT29 cells was caspase-mediated (Lee et al., 2005). Administration of general caspase inhibitor Z-VAD prevented JAinduced apoptosis by flow cytometry (Fig. 5G) and decreased caspase-3 activation by Western blot (Fig. 5H, upper panel), confirming the ability of Z-VAD to inhibit apoptosis. Interestingly, combination treatment of Z-VAD and JA did not affect JA-induced expression of LC3-II by Western blot (Fig. 5H, lower panel) or JA-induced formation of AVOs by flow cytometry (Fig. 5I). The results suggest that the occurrence of JA-induced autophagy at an early time point (6 h;

943

944

WON ET AL.

Fig. 5.

(Continued)

Fig. 1A and C) promoted the process of JA-induced apoptosis, whereas JA-induced apoptosis at a later time point (24–72 h; Fig. 5D) did not change the development of autophagy. Discussion

The results of the present study revealed that a natural arylnaphthalide lignan JA significantly induced autophagy in human colorectal cancer HT-29 cells, as evidenced by the expression of LC3-II using Western blotting and confocal microscopy, the development of AVOs using flow cytometry, and manifestation of autophagic vesicles using electron microscopy in vitro, as well as LC3 puncta formation in tumors of mice that orally received JA using immunofluorescence microscopy. It is noteworthy that JA induced autophagy flux, as demonstrated by the decrease of JA-induced SQSTM1 and LC3-II protein expressions at the later time point and attenuation of JA-induced AVOs formation in the presence of BAF, an inhibitor of autophagosome and lysosome fusion, in vitro as well as a decrease in SQSTM1 staining in some areas and an increase in Beclin 1 staining of tumors of mice that received JA. Accumulation of autophagosomes in HT-29 cells transfected with ptfLC3 in the presence of JA and CQ confirmed that JA indeed induced autophagy flux. Two signaling pathways are involved in JA-induced autophagy. The first is activation of class III PI3K/Beclin 1, Atg5, and the BNIP3 pathway, and the second is inhibition of the PDK1/mTOR/ p70S6k related pathway. Addition of inhibitors further demonstrates that JA-induced autophagy enhanced apoptosis and therefore promoted cancer cell death and reduced tumor colony formation. In contrast, JA-induced apoptosis had no effect on the process of autophagy. JOURNAL OF CELLULAR PHYSIOLOGY

Autophagy might remove the proteins or organelles that are damaged by the cancer treatment and become protective against the treatment (protective autophagy) (Kondo et al., 2005). However, autophagy can also induce cancer cell death in apoptosis-deficient cells (Moretti et al., 2007). Mutations in apoptotic machinery have been detected in many cancer cells and result in resistance to chemotherapy. The PI3K signaling pathway is one of the most deregulated pathways in human cancers, including colorectal cancer (Zhang et al., 2011). Inhibitors of class I PI3K, AKT, or mTOR reduced colorectal tumor proliferation in preclinical studies that are currently underway for phase 1 and 2 clinical trials (Zhang et al., 2011). It is noteworthy that, unlike these anticancer drugs, JA-induced autophagy enhances the process of apoptosis (Fig. 5E and F) and leads to cell death (Fig. 4G) and reduced colony formation (Fig. 4H) of HT-29 cells, suggesting the novel therapeutic potential of JA in the treatment of colorectal cancer. In mammals, uncoordinated 51-like kinase (ULK) 1 localizes to the phagophores and recruits the class III PI3K-Beclin 1 complex and Atg12-Atg5-Atg16 complex to that location (Yang et al., 2011). During this process, mTOR causes the phosphorylation of Atg13 (a component of ULK1) and thus inhibits the formation of ULK1. Bcl-XL also negatively regulates this process by interacting with Beclin 1 (Liu et al., 2011b). In contrast, BNIP3 binds to Rheb (an activator of mTOR), inhibits mTOR activation, and therefore induces autophagic cell death (Burton and Gibson, 2009). Based on our results, we propose the following mechanism of JA-induced autophagy in colorectal cancer cells: JA increases the levels of class III PI3K, Beclin 1 (Fig. 3E), Atg12-Atg5, Atg5 (Fig. 3F), SQSTM1 (Fig. 2B and C), and LC3-II (Fig. 1A and B) to sustain the formation of autophagosomes (Fig. 1C and D). Furthermore, inhibition of


p-mTOR (Fig. 3B), which may be induced by mitochondrial BNIP3 (Fig. 3H) and suppression of Bcl-XL, as demonstrated in our previous report (Lee et al., 2005), prevents the disruption of the class III PI3K-Beclin 1 complex, further promoting autophagosome formation. Administration of Atg5 shRNA (Fig. 4E and F) or class III PI3K inhibitor 3-MA (Fig. 4B and C) attenuated JA-induced LC3-II expression and LC3 puncta, which further confirms the critical roles of these two molecules in autophagosome formation. A recent study showed that increasing or silencing Atg5 expression sensitizes or leads to resistance of tumor cells to anticancer therapy, respectively (Yousefi et al., 2006). When cell death in induced, Atg5 plays a supportive role in the apoptosis process (Yousefi et al., 2006). Clinically, decreased Atg5 protein expression was detected in 23% of colorectal patients, indicating its role in autophagic/apoptotic cell death and gastrointestinal cancer pathogenesis (An et al., 2011). Beclin 1, which forms a complex with class III PI3K, is another important regulator of autophagy and apoptosis, and has been postulated to act as a haploinsufficient tumor suppressor (Ahn et al., 2007). Patients with increased Beclin 1 expression in their cancer cells exhibit longer survival (Li et al., 2009). The findings described above suggest that overexpression of Atg5 (Fig. 3F) and Beclin 1 (Fig. 3E) in JA-induced autophagy plays a beneficial role in colorectal cancer patients. Ku70, which plays a critical role in DNA double-strand break repair, sequesters pro-apoptotic Bax and therefore inhibits translocation of Bax from the cytosol to the mitochondria (Cohen et al., 2004). The increase of the mitochondrial Bax alters the balance between pro- and anti-apoptotic Bcl-2 family members in the cells and favors apoptosis. Our previous report (Lee et al., 2005) indicated that JA-induced apoptosis in HT-29 cells proceeds via inhibition of Ku70 and results in translocation of Bax to the mitochondria. We also observed that JA decreases mitochondrial Bcl-XL and prevents its binding to Bax and therefore favors Bax activation for apoptosis (Lee et al., 2005). Results from another study indicate that Bcl-XL interacts with Beclin 1 by the combination of the BH3-binding groove of Bcl-XL and a BH3 domain in Beclin 1 (Liu et al., 2011b). Based on these results, Bcl-XL may be an important connection in the crosstalk between the induction of apoptosis and autophagy of HT-29 cells in response to JA. Furthermore, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) and Fas-associated death domain protein (FADD), which function as a death ligand and an adaptor protein, respectively, recruit and activate caspase-8 to induce extrinsic apoptotic signaling (Tan et al., 2009; Liu et al., 2011b). Although HT-29 cells are TRAIL-resistant colon cancer cells (Nawrocki et al., 2007), interaction of Atg5-Atg-12 with FADD can induce autophagic cell death (Codogno and Meijer, 2006). Therefore, JA-induced formation of Atg5-Atg12 (Fig. 3F) may contribute to the promotion of apoptosis by JA-induced autophagy. In conclusion, our previous report demonstrated that oral administration of JA significantly suppresses the growth of HT29 cells implanted into NOD-SCID mice (Lee et al., 2005). In the present study, we further demonstrated that autophagy is involved in the reduction of tumor size. In JA-treated mice, JA appeared to show preferential inhibition of tumor growth and did not induce weight changes in liver and spleen, suggesting that JA may have a potential role in chemoprevention and chemotherapeutic treatments for human cancers. Literature Cited Ahn CH, Jeong EG, Lee JW, Kim MS, Kim SH, Kim SS, Yoo NJ, Lee SH. 2007. Expression of beclin-1, an autophagy-related protein, in gastric and colorectal cancers. APMIS 115:1344–1349. An CH, Kim MS, Yoo NJ, Park SW, Lee SH. 2011. Mutational and expressional analyses of ATG5, an autophagy-related gene, in gastrointestinal cancers. Pathol Res Pract 207: 433–437.


Aoki H, Takada Y, Kondo S, Sawaya R, Aggarwal BB, Kondo Y. 2007. Evidence that curcumin suppresses the growth of malignant gliomas in vitro and in vivo through induction of autophagy: Role of Akt and extracellular signal-regulated kinase signaling pathways. Mol Pharmacol 72:29–39. Bocharov EV, Pustovalova YE, Pavlov KV, Volynsky PE, Goncharuk MV, Ermolyuk YS, Karpunin DV, Schulga AA, Kirpichnikov MP, Efremov RG, Maslennikov IV, Arseniev AS. 2007. Unique dimeric structure of BNip3 transmembrane domain suggests membrane permeabilization as a cell death trigger. J Biol Chem 282:16256–16266. Burton TR, Gibson SB. 2009. The role of Bcl-2 family member BNIP3 in cell death and disease: NIPping at the heels of cell death. Cell Death Differ 16:515–523. Byun JY, Yoon CH, An S, Park IC, Kang CM, Kim MJ, Lee SJ. 2009. The Rac1/MKK7/JNK pathway signals upregulation of Atg5 and subsequent autophagic cell death in response to oncogenic Ras. Carcinogenesis 30:1880–1888. Carew JS, Medina EC, Esquivel JA, 2nd, Mahalingam D, Swords R, Kelly K, Zhang H, Huang P, Mita AC, Mita MM, Giles FJ, Nawrocki ST. 2010. Autophagy inhibition enhances vorinostat-induced apoptosis via ubiquitinated protein accumulation. J Cell Mol Med 14:2448–2459. Chang YP, Tsai CC, Huang WC, Wang CY, Chen CL, Lin YS, Kai JI, Hsieh CY, Cheng YL, Choi PC, Chen SH, Chang SP, Liu HS, Lin CF. 2010. Autophagy facilitates IFN-gamma-induced Jak2-STAT1 activation and cellular inflammation. J Biol Chem 285:28715–28722. Chen S, Rehman SK, Zhang W, Wen A, Yao L, Zhang J. 2011. Autophagy is a therapeutic target in anticancer drug resistance. Biochim Biophys Acta 1806:220–229. Codogno P, Meijer AJ. 2006. Atg5: More than an autophagy factor. Nat Cell Biol 8:1045– 1047. Cohen HY, Lavu S, Bitterman KJ, Hekking B, Imahiyerobo TA, Miller C, Frye R, Ploegh H, Kessler BM, Sinclair DA. 2004. Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax-mediated apoptosis. Mol Cell 13:627–638. Day SH, Lin YC, Tsai ML, Tsao LT, Ko HH, Chung MI, Lee JC, Wang JP, Won SJ, Lin CN. 2002. Potent cytotoxic lignans from Justicia procumbens and their effects on nitric oxide and tumor necrosis factor-alpha production in mouse macrophages. J Nat Prod 65:379– 381. Eskelinen EL, Saftig P. 2009. Autophagy: A lysosomal degradation pathway with a central role in health and disease. Biochim Biophys Acta 1793:664–673. Fukamiya N, Lee KH. 1986. Antitumor agents, 81. Justicidin-A and diphyllin, two cytotoxic principles from Justicia procumbens. J Nat Prod 49:348–350. Guo XZ, Xu JH, Liu MP, Kleeff J, Ho CK, Ren LN, Li HY, Koninger J, Cui ZM, Wang D, Wu CY, Zhao JJ, Friess H. 2005. KAI1 inhibits anchorage-dependent and -independent pancreatic cancer cell growth. Oncol Rep 14:59–63. Hector S, Prehn JH. 2009. Apoptosis signaling proteins as prognostic biomarkers in colorectal cancer: A review. Biochim Biophys Acta 1795:117–129. Huang S, Yang ZJ, Yu C, Sinicrope FA. 2011. Inhibition of mTOR kinase by AZD8055 can antagonize chemotherapy-induced cell death through autophagy induction and downregulation of p62/sequestosome 1. J Biol Chem 286:40002–40012. Igney FH, Krammer PH. 2002. Death and anti-death: Tumour resistance to apoptosis. Nat Rev Cancer 2:277–288. Jemal A, Siegel R, Xu J, Ward E. 2010. Cancer statistics. CA Cancer J Clin 60:277–300. Kerr JF, Winterford CM, Harmon BV. 1994. Apoptosis. Its significance in cancer and cancer therapy. Cancer 73:2013–2026. Kimura N, Kumamoto T, Kawamura Y, Himeno T, Nakamura KI, Ueyama H, Arakawa R. 2007a. Expression of autophagy-associated genes in skeletal muscle: an experimental model of chloroquine-induced myopathy. Pathobiology 74:169–176. Kimura S, Noda T, Yoshimori T. 2007b. Dissection of the autophagosome maturation process by a novel reporter protein, tandem fluorescent-tagged LC3. Autophagy 3:452– 460. Klionsky DJ, Cuervo AM, Seglen PO. 2007. Methods for monitoring autophagy from yeast to human. Autophagy 3:181–206. Kondo Y, Kanzawa T, Sawaya R, Kondo S. 2005. The role of autophagy in cancer development and response to therapy. Nat Rev Cancer 5:726–734. Kondo Y, Kondo S. 2006. Autophagy and cancer therapy. Autophagy 2:85–90. Lawson EL, Mills DR, Brilliant KE, Hixson DC. 2012. The transmembrane domain of CEACAM1–4S is a determinant of anchorage independent growth and tumorigenicity. PLoS ONE 7:-e29606. Lee JC, Lee CH, Su CL, Huang CW, Liu HS, Lin CN, Won SJ. 2005. Justicidin A decreases the level of cytosolic Ku70 leading to apoptosis in human colorectal cancer cells. Carcinogenesis 26:1716–1730. Lee YR, Lei HY, Liu MT, Wang JR, Chen SH, Jiang-Shieh YF, Lin YS, Yeh TM, Liu CC, Liu HS. 2008. Autophagic machinery activated by dengue virus enhances virus replication. Virology 374:240–248. Li BX, Li CY, Peng RQ, Wu XJ, Wang HY, Wan DS, Zhu XF, Zhang XS. 2009. The expression of beclin 1 is associated with favorable prognosis in stage IIIB colon cancers. Autophagy 5:303–306. Liang C, Feng P, Ku B, Dotan I, Canaani D, Oh BH, Jung JU. 2006. Autophagic and tumour suppressor activity of a novel Beclin1-binding protein UVRAG. Nat Cell Biol 8:688–699. Liu B, Cheng Y, Liu Q, Bao JK, Yang JM. 2010. Autophagic pathways as new targets for cancer drug development. Acta Pharmacol Sin 31:1154–1164. Liu HS, Ke CS, Cheng HC, Huang CY, Su CL. 2011a. Curcumin-induced mitotic spindle defect and cell cycle arrest in human bladder cancer cells occurs partly through inhibition of Aurora A. Mol Pharmacol 80:638–646. Liu JJ, Lin M, Yu JY, Liu B, Bao JK. 2011b. Targeting apoptotic and autophagic pathways for cancer therapeutics. Cancer Lett 300:105–114. Lotze MT, Buchser WJ, Liang X. 2012. Blocking the interleukin 2 (IL2)-induced systemic autophagic syndrome promotes profound antitumor effects and limits toxicity. Autophagy 8:1264–1266. Mizushima N. 2004. Methods for monitoring autophagy. Int J Biochem Cell Biol 36:2491– 2502. Moretti L, Yang ES, Kim KW, Lu B. 2007. Autophagy signaling in cancer and its potential as novel target to improve anticancer therapy. Drug Resist Updat 10:135–143. Nagata S. 2000. Apoptotic DNA fragmentation. Exp Cell Res 256:12–18. Nawrocki ST, Carew JS, Douglas L, Cleveland JL, Humphreys R, Houghton JA. 2007. Histone deacetylase inhibitors enhance lexatumumab-induced apoptosis via a p21Cip1-dependent decrease in survivin levels. Cancer Res 67:6987–6994. Pankiv S, Clausen TH, Lamark T, Brech A, Bruun JA, Outzen H, Overvatn A, Bjorkoy G, Johansen T. 2007. P62/SQSTM1 binds directly to Atg8/LC3 to facilitate degradation of ubiquitinated protein aggregates by autophagy. J Biol Chem 282:24131–24145. Papandreou I, Lim AL, Laderoute K, Denko NC. 2008. Hypoxia signals autophagy in tumor cells via AMPK activity, independent of HIF-1, BNIP3, and BNIP3L. Cell Death Differ 15:1572–1581.

945

946

WON ET AL.

Saarinen NM, Warri A, Airio M, Smeds A, Makela S. 2007. Role of dietary lignans in the reduction of breast cancer risk. Mol Nutr Food Res 51:857–866. Su CL, Chen FN, Won SJ. 2011. Involvement of apoptosis and autophagy in reducing mouse hepatoma ML-1 cell growth in inbred BALB/c mice by bacterial fermented soybean products. Food Chem Toxicol 49:17–24. Su CL, Cheng CC, Lin MT, Yeh HC, Lee MC, Lee JC, Won SJ. 2007. Staphylococcal enterotoxin C1-induced pyrogenic cytokine production in human peripheral blood mononuclear cells is mediated by NADPH oxidase and nuclear factor-kB. FEBS J 274:3633–3645. Su CL, Huang LL, Huang LM, Lee JC, Lin CN, Won SJ. 2006. Caspase-8 acts as a key upstream executor of mitochondria during justicidin A-induced apoptosis in human hepatoma cells. FEBS Lett 580:3185–3191. Tan ML, Ooi JP, Ismail N, Moad AI, Muhammad TS. 2009. Programmed cell death pathways and current antitumor targets. Pharm Res 26:1547–1560. Virk-Baker MK, Nagy TR, Barnes S. 2010. Role of phytoestrogens in cancer therapy. Planta Med 76:1132–1142. Wang L, Pan J, Yang M, Wu J, Yang J. 2011. Chromatographic fingerprint analysis and simultaneous determination of eight lignans in Justicia procumbens and its compound preparation by HPLC-DAD. J Sep Sci 34:667–674.


Wickman G, Julian L, Olson MF. 2012. How apoptotic cells aid in the removal of their own cold dead bodies. Cell Death Differ 19:735–742. Wlodkowic D, Skommer J, Darzynkiewicz Z. 2012. Cytometry of apoptosis. Historical perspective and new advances. Exp Oncol 34:255–262. Wu SY, Lan SH, Cheng DE, Chen WK, Shen CH, Lee YR, Zuchini R, Liu HS. 2011. Ras-related tumorigenesis is suppressed by BNIP3-mediated autophagy through inhibition of cell proliferation. Neoplasia 13:1171–1182. Xiao G. 2007. Autophagy and NF-kB: Fight for fate. Cytokine Growth Factor Rev 18: 233–243. Yang ZJ, Chee CE, Huang S, Sinicrope FA. 2011. The role of autophagy in cancer: Therapeutic implications. Mol Cancer Ther 10:1533–1541. Yang D, Yaguchi T, Yamamoto H, Nishizaki T. 2007. Intracellularly transported adenosine induces apoptosis in HuH-7 human hepatoma cells by downregulating c-FLIP expression causing caspase-3/-8 activation. Biochem Pharmacol 73:1665–1675. Yousefi S, Perozzo R, Schmid I, Ziemiecki A, Schaffner T, Scapozza L, Brunner T, Simon HU. 2006. Calpain-mediated cleavage of Atg5 switches autophagy to apoptosis. Nat Cell Biol 8:1124–1132. Zhang J, Roberts TM, Shivdasani RA. 2011. Targeting PI3K signaling as a therapeutic approach for colorectal cancer. Gastroenterology 141:50–61.