This article is mentioned in:

Introduction

Medicinal plants have proven to be a rich source of bioactive compounds for therapeutic agents, and currently 75% of prescribed drugs worldwide are derived from plant sources (1). Feroniellin A (FERO), a novel furanocoumarin is isolated from the roots of Feroniella lucida. The chemical structure of furanocoumarins consists of a furan ring fused with coumarin, which is present in many plants. Coumarins possess anticoagulant, antimicrobial, antioxidant, anti-inflammatory, anti-allergic and anticancer properties (2). Some furanocoumarin derivatives isolated from plants show anticancer activity by functioning as topoisomerase I inhibitors (3), efflux transport inhibitors or drug metabolism inhibitors (4,5). However, the specific molecular mechanisms by which FERO shows anticancer activity have not been revealed.

Multidrug resistance (MDR) is a major problem in cancer therapy and is often the result of overexpression of the drug efflux protein P-glycoprotein (P-gp). P-gp is a 170-kDa protein that belongs to the ATP-binding cassette superfamily of membrane transporter proteins (6,7). P-gp is an energy-dependent drug efflux pump that maintains intracellular drug concentrations below cytotoxic levels, thereby decreasing the cytotoxic effects of a variety of chemotherapeutic agents (6–9). P-gp also plays a role in inhibition of drug accumulation and caspase activation in MDR tumors (10,11). Of special note, recent lines of evidence have shown that NF-κB- or SIRT1-mediated regulation of P-gp plays a critical role in anticancer drug resistance (12,13).

Autophagy, an ancient system necessary to maintain homeostasis in eukaryotic cells, degrades long-lived cytoplasmic proteins and organelles and provides nutrients during starvation or stress conditions (14) through programmed processing involving the sequential activity of autophagy-related gene (ATG) products. As autophagy is necessary for cellular homeostasis, it is involved in biological processes including development, aging and degeneration (15). However, aberrant regulation of autophagy is related to many diseases, including cancer and neurodegenerative disorders (16). As a specific example, the first report connecting autophagy to cancer showed that allelic loss of the essential autophagy gene Beclin-1 (BECN1) is prevalent in human breast, ovarian, and prostate cancers (17), and that Becn1+/− mice develop mammary gland hyperplasia and lymphomas as well as lung and liver tumors (18). Subsequent studies demonstrated that ATG5−/− and ATG7−/− livers give rise to adenomas (19). These lines of evidence suggest that autophagy acts as a tumor suppressor in cancer development. Contrary to this, many other reports have shown that autophagy exerts a pro-survival function in tumor cells (20–22). Additional studies have demonstrated that the inhibition of autophagy by pharmaceutical drugs sensitized cells to apoptotic cell death, and that combination therapies using autophagy inhibitors plus chemotherapy led to faster tumor cell death than did chemotherapy alone (23). These findings indicate that pro-survival autophagy may represent a major hindrance to successful cancer therapy.

The present study was initiated to screen small molecules derived from plants grown in Thailand in order to reverse MDR in A549RT-eto cells. We have identified that FERO induces autophagy, which is necessary for FERO-induced apoptosis in A549RT-eto cells. Therefore, we propose that FERO represents a powerful candidate for the treatment of multidrug-resistant lung cancer.

Materials and methods

Cell cultures

A549RT-eto cells were developed and kindly provided by the Laboratory of Biochemistry, Chulabhorn Research Institute, Thailand and have been described elsewhere (24). A549RT-eto cells were cultured in RPMI-1640 medium (Gibco, Grand Island, NY, USA) supplemented with 10% FBS and 1% penicillin, and streptomycin (Gibco) at 37°C in a humidified atmosphere of 5% CO2 in air.

Antibodies and reagents

For immunoblotting, antibodies against ATG5, LC3B (D11), mTOR, phospho-mTOR (Ser2448) and cleaved PARP (Asp214) were acquired from Cell Signaling Biotechnology (Beverly, MA, USA). Anti-P-gp (Calbiochem; San Diego, CA, USA) and β-actin (C4), BECN1 (H300), BID (FL-195), caspase-9 p35 (H-170), NF-κB p65 (F-6), P53 (DO-1), SIRT1 (H-300) and Sp1 (1C6) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Rapamycin, Z-VAD and BAY11-7082 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Feroniellin A was isolated from the roots of Feroniella lucida and was kindly provided by the Natural Products Research Unit, Department of Chemistry, Faculty of Science, Chulalongkorn University, Thailand and have been described elsewhere (25).

Immunoblotting

Cells were harvested and lysed with lysis buffer [150 mM NaCl, 1% NP-40, 50 mM Tris-HCl (pH 7.5)] containing 0.1 mM Na2VO3, 1 mM NaF and protease inhibitors (Sigma). For immunoblotting, proteins from whole cell lysates were resolved by 10 or 12% SDS-polyacrylamide gel electrophoresis (PAGE) and then transferred to nitrocellulose membranes. Primary antibodies were used at 1:1,000 or 1:2,000 dilutions, and secondary antibodies conjugated with horseradish peroxidase were used at 1:2,000 dilutions in 5% non-fat dry milk. After the final washing, nitrocellulose membranes were exposed using enhanced chemiluminescence assay and visualized on a LAS 4000 mini (Fuji, Tokyo, Japan).

Nuclear NF-κB pull-down assay

A549RT-eto cells (1×106 cells/ml) were incubated with FERO or DMSO as a control for 12 h and nuclear extracts were prepared. Cells were pelleted and resuspended in 0.4 ml hypotonic lysis buffer [20 mM HEPES (pH 7.9), 10 mM KCl, 1 mM EDTA, 0.2% Triton X-100, and 1 mM Na2VO3 plus protease inhibitors] and kept on ice for 20 min. After centrifugation at 14 000 g for 5 min at 4°C, the nuclear pellet was extracted with 0.1 ml hypertonic lysis buffer on ice for a further 20 min. After centrifugation at 14,000 g for 5 min at 4°C, the supernatants were diluted to 100 mM NaCl and incubated with 25 μl of agarose beads conjugated to a consensus NF-κB binding oligonucleotide (Santa Cruz Biotech) for 1 h at 4°C. After 3 washes, sample buffer was added and boiled for 5 min. The binding NF-κB (p65) protein to the oligonucleotide conjugated with agarose was detected by immunoblotting using an anti-p65 NF-κB Ab (Santa Cruz Biotech).

Small interfering RNA (siRNA) transfection

Cells were trypsinized and incubated overnight to achieve 60–70% confluency before siRNA transfection. Beclin-1 human siRNA (pre-made at Bioneer, Daejeon, Korea; 100 nM; sense: 5′-UGG AAU GGA AUG AGA UUA A(dTdT)-3′; antisense: 5′-UUA AUC UCA UUC CAU UCC A(dTdT)-3′ or negative control siRNA (Bioneer); 100 nM; sense: 5′-CCU ACG CCA CCA UUU CGU (dTdT)-3′; antisense: 5′-ACG AAA UUG GUG GCG UAG G (dTdT)-3′ were mixed with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). The cells were incubated with the transfection mixture for 24 h and then rinsed with RPMI-1640 medium containing 10% fetal bovine serum. The cells were incubated for 48 h before harvest.

Reverse-transcription polymerase chain reaction (RT-PCR)

Total RNA was extracted from cells using the RNeasy mini kit (Qiagen, Valencia, CA, USA) in accordance with the manufacturer’s instructions. Three micrograms of total RNA was converted to cDNA using Superscript II reverse transcriptase (Invitrogen), and PCR was performed using the specific primers: human MDR1, sense: 5′-CCC ATC ATT GCA ATA GCA GG-3′ and antisense: 5′-GTT CAA ACT TCT GCT CCT GA-3′; MRP2, sense: 5′-ACA GAG GCT GGT GGC AAC C-3′ and antisense: 5′-ACC ATT ACC TTG TCA CTG TCC-3′; BCRP, sense: 5′-GAT CAC AGT CTT CAA GGA GAT C-3′ and antisense: 5′-CAG TCC CAG TAC GAC TGT GAC A-3′ were used. The cDNAs from each sample were diluted, and PCR was run at an optimized cycle number. β-actin mRNA was measured as an internal standard. After amplification, the products were subjected to electrophoresis on 2.0% agarose gels and detected by ethidium bromide staining.

Luciferase reporter assay

A549RT-eto cells were transfected with hMDR1-luciferase or control pGL3 empty luciferase vector. To normalize transfection efficiency, a pGK-β-gal vector that expresses galactosidase from a phosphoglucokinase promoter was included in the transfection mixture. FERO was added to the transfected cells at 12 h before harvest. At 48 h post-transfection, cells were washed with cold PBS and lysed in lysis solution [25 mM Tris (pH 7.8), 2 mM EDTA, 2 mM DTT, 10% glycerol and 1% Triton X-100]. Luciferase activity was measured with a luminometer using a luciferase kit (Promega, Madison, WI, USA).

Cell viability assay

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay was performed to measure cell survival as described previously (26). Dye solution containing tetrazolium was added to cells in a 96-well plate format and incubated for 2 h. The absorbance of the formazan produced by living cells was measured at 570 nm. The relative percentage of cell survival was calculated by the mean absorbance of the treated cells (ODT) and the mean absorbance of control cells (ODC) following the formula: % Cell survival = (ODT / ODC).

Statistical analysis

Data are presented as a mean ± standard deviation (SD). Student’s t-test was used for statistical analysis, with significance defined as a p-value <0.05.

Results

Feroniellin A induces apoptosis in human A549 lung cells resistant to etoposide

Previous observations have revealed that human etoposide resistant A549 lung cancer cells (A549RT-eto) exhibit upregulation of Stat1 and HDAC4, leading to elevated levels of P-glycoprotein (P-gp) encoded by multidrug resistance (MDR) 1 (unpublished data). Thus, we performed a screen for small molecules derived from medicinal plants in Thailand that reverse MDR. We found that a small compound known as Feroniellin A (FERO; molecular weight = 388) (molecular structure shown in Fig. 1A) that efficiently induced death in A549RT-eto cells in a dose- and time-dependent manner (Fig. 1B). We found that parental A549 cells showed more cytotoxicity toward FERO as well (data not shown). Since it has been reported that NF-κB and SIRT1 are involved in the regulation of MDR through P-gp protein levels (27), we sought to examine protein levels of NF-κB, SIRT1 and P-gp in A549RT-eto cells treated with 0.05 and 0.5 mM FERO 12 h post-treatment. Our findings show that FERO treatment reduces expression levels of NF-κB, SIRT1 and P-gp and enhances expression levels of P53, a representative tumor suppressor protein (Fig. 1C). In addition, we detected reduced pre-caspase-9 and pre-Bid levels in the same lysates of A549RT-eto cells, indicating induction of intrinsic apoptosis (Fig. 1C).

Figure 1. Chemical structure of Feroniellin A (FERO) and its effect on cell viability and protein levels in A549RT-eto cells. (A) The chemical structure and molecular weight of Feroniellin A (FERO). (B) A549RT-eto cells were treated with FERO (0, 0.05, 0.5 and 1 mM) for 12 h and were treated with FERO (1 mM) at 0, 6 and 12 h. After treatment, the morphological changes of cells and cell viability were observed under a light microscope. (C) A549RT-eto cells were treated with FERO (0, 0.05 and 0.5 mM) for 12 h. Cell lysates from A549RT-eto cells treated with FERO were prepared and separated by 12% SDS-PAGE. The expression of pro-Bid and pro-caspase-9 proteins was detected by immunoblotting to confirm apoptosis. Protein levels of NF-κB, SIRT1, P53, and P-gp were examined by immunoblotting using the corresponding antibodies.

Feroniellin A reduces MDR1 transcript levels and promoter activity in A549RT-eto cells

In addition to the MDR1 gene, other genes such as multidrug resistance-associated protein (MRP) 2 and breast cancer resistance protein (BCRP) are known to be involved in drug resistance (28,29). Therefore, we examined transcript levels of MDR1, MRP2 and BCRP in A549RT-eto cells during FERO treatment. RNA was isolated and cDNA generated from A549RT-eto cells treated with 0.05 and 0.5 mM FERO for 12 h and genes were amplified using an optimized number of cycles. We found a significant reduction in MDR1 transcript levels with slight reductions in MRP2 levels following FERO treatment in a dose-dependent manner (Fig. 2A). However, transcript levels of BCRP were not reduced by FERO treatment (Fig. 2B). In addition to transcript levels, we also examined MDR1 promoter activity in A549RT-eto cells during FERO treatment using an MDR1-promoter lucif-erase reporter vector (12). FERO treatment induced a dramatic decrease of MDR1-mediated luciferase activity in A549RT-eto cells (Fig. 2B), indicating that FERO reduces MDR1 promoter activity. Since it has been reported that the transcription factor NF-κB is involved in the regulation of P-gp (30,31), we next examined the nuclear localization of NF-κB and its DNA binding activity during FERO treatment. We found that FERO treatment inhibits translocation of NF-κB from the cytoplasm to the nucleus, which results in lowered levels of NF-κB at its binding site (Fig. 2C). On the basis of these results, we suggest that FERO inhibits MDR1 transcription, resulting in a decrease in P-gp protein levels, which leads to sensitization to apoptosis in A549RT-eto cells.

Figure 2. FERO suppresses MDR1 transcription and promoter activity and NF-κB transcriptional activity. (A) Total RNAs of A549RT-eto cells following FERO treatment (0, 0.05 and 0.5 mM) for 12 h were isolated and subjected to RT-PCR. Transcripts of MDR1, MRP2 and BCRP were examined after optimization of PCR conditions. Relative mRNA ratios of each MDR-related gene are described relative to levels of β-actin following measurements of band intensities using Multi Gauge Version 2.1 (Fuji). (B) A549RT-eto cells were transfected with a MDR1-luciferase vector (1.6 μg) or pGL3 vector (1.6 μg) as a control and were harvested at 24 h post-transfection. When luciferase activity was measured, transfection efficiency was normalized to β-galactosidase reporter vector pGK-β-gal (1.6 μg). Results shown are the average of triplicate experiments. Bar indicates standard deviation. Black bars indicate MDR1-luciferase vector and white bars indicate pGL3 vector (*p<0.05, FERO 0 vs. FERO 0.05; **p<0.01, FERO 0 vs. FERO 0.5). (C) NF-κB oligonucleotides conjugated to agarose were added to nuclear extracts from A549RT-eto cells treated with FERO (0, 0.05 and 0.5 mM) and the precipitation mixtures were isolated after centrifugation. The binding of NF-κB (p65) to the oligonucleotide was detected by immunoblotting using anti-NF-κB (p65) antibody.

Inhibition of NF-κB accelerates FERO-induced apoptosis of A549RT-eto cells through downregulation of P-gp

Because we observed enhanced NF-κB protein levels and activity in A549RT-eto cells, we next explored whether NF-κB is involved in resistance to etoposide in A549 cells through upregulation of P-gp. A549RT-eto cells were treated with 10 μM BAY11-7082 (BAY), an inhibitor of NF-κB, and examined for cellular viability using the MTT assay. BAY treatment alone did not influence inhibition of cell growth in A549RT-eto cells, while 0.05 mM FERO treatment inhibited cell growth by ∼30% at 24 h post-treatment compared to DMSO controls (Fig. 3A). Furthermore, we found that a combined treatment of FERO and BAY accelerated FERO-mediated apoptosis in A549RT-eto cells (Fig. 3A), which was confirmed by observations of cleaved PARP and pre-Bid (Fig. 3B). Moreover, when we examined protein levels of NF-κB, SIRT1 and P-gp after treatment with BAY alone, FERO alone, or BAY plus FERO, we found that BAY treatment alone decreased not only expression levels of NF-κB but also P-gp, but did not influence protein levels of SIRT1 in the cells (Fig. 3B). FERO treatment alone was shown to drastically diminish protein levels of NF-κB, SIRT1 and P-gp in A549RT-eto cells (Fig. 1C). We also observed that the combined treatment resulted in more significantly reduced protein levels of NF-κB, SIRT1 and P-gp (Fig. 3B). These results suggest that NF-κB is involved in MDR in A549 cells by upregulation of P-gp, resulting in resistance to etoposide. To examine whether activation of caspase caused by FERO treatment is involved in apoptotic cell death in A549RT-eto cells, a pan-caspase inhibitor, Z-VAD, was used. FERO treatment alone at 0.5 mM resulted in a significant induction of apoptotic cell death at 24 h post-treatment, while the combined treatment with FERO and Z-VAD blocked FERO-induced apoptotic cell death (Fig. 4A), which was confirmed by a lack of cleavage of PARP and pre-Bid proteins (Fig. 4B). These results indicate that activation of caspases is involved in FERO-induced apoptotic cell death in A549RT-eto cells. However, Z-VAD treatment did not block downregulation of P-gp expression induced by FERO, indicating that a decrease of P-gp expression is necessary but not sufficient for apoptotic cell death.

Figure 3. NF-κB inhibition enhances FERO-induced apoptosis by downregulation of P-gp. (A) A549RT-eto cells were treated with FERO (0.25 mM) alone, Bay11-7082 (BAY; 2 μM) alone, and FERO plus BAY for 24 h. Changes in cellular morphology were observed under a light microscope and cell growth was measured by MTT assay. Data were calculated as percent of relative cell viability and expressed as the average of three experiments (*p<0.01, Mock vs. FERO+BAY). (B) Cell lysates from the treated A549RT-eto cells were prepared and separated by 12% SDS-PAGE. The expression of pre-Bid and PARP protein was detected for apoptosis by immunoblotting, Protein levels of NF-κB, SIRT1 and P-gp were examined by immunoblotting using the corresponding antibodies.

Figure 4. Inhibition of caspase activity protects A549RT-eto cells from FERO-induced apoptosis. (A) A549RT-eto cells were treated with FERO (0.5 mM) alone, Z-VAD (20 μM) alone, or FERO plus Z-VAD for 24 h. Changes in cellular morphology were observed under a light microscope and cell growth was measured by MTT assay. Data were calculated as percent of relative cell viability and expressed as the mean of at least three experiments (*p<0.05, Mock vs. FERO; **p<0.05, FERO vs. FERO+Z-VAD). (B) Cell lysates from the treated A549RT-eto cells were prepared and separated by 12% SDS-PAGE. Expression of pro-Bid and PARP protein was detected by immunoblotting for apoptosis and protein levels of Beclin-1, LC3B, ATG5, mTOR and P-gp were compared by immunoblotting using the corresponding antibodies.

FERO also induces autophagy in A549RT-eto cells

Auto-phagy has a dual role in cancer, as a tumor suppressor and a pro-tumor cell survival mechanism (32,33). Therefore, we investigated whether FERO could induce autophagy, which is characterized by the formation of vacuoles, GFP-LC3 puncta, and LC3 I to LC3 II conversion. The presence of vacuoles was apparent in A549RT-eto cells treated with 0.5 mM FERO (Fig. 5A). There was also a clearly observable accumulation of GFP-LC3 II puncta in A549RT-eto cells treated with FERO, while no GFP-LC3 puncta were present in A549RT-eto cells treated with DMSO (Fig. 5B). Moreover, we observed LC3 I to LC3 II conversion in FERO-treated A549RT-eto cells in a dose-dependent manner (Fig. 5C). Lastly, we found induction of Beclin-1, required for the formation of autophagic vesicles (34), and reduction of phospho-mTOR levels, an inhibitor of autophagy, indicating that FERO treatment clearly induced autophagy in A549RT-eto cells (Fig. 5C).

Figure 5. Induction of autophagy in A549RT-eto cells by treatment with FERO. (A) Morphology of A549RT-eto cells was observed under a light microscope at 12 h following treatment with FERO (0.5 mM) (×40 magnification). (B) A549RT-eto cells were transfected with pEGFP-LC3B or pEGFP control vector and then treated with FERO (0.5 mM) or DMSO as a control at 24 h post-transfection. GFP puncta were then analyzed using a fluorescence microscope at 24 h post-treatment. (C) Cell lysates from A549RT-eto cells treated with FERO (0, 0.05 and 0.5 mM) were prepared at 12 h post-treatment and separated by 12% SDS-PAGE. LC3B I cleavage, Beclin-1, P-gp, phospho-mTOR and mTOR proteins were examined by immunoblotting using the corresponding antibodies.

Modulation of autophagy regulates FERO-induced apoptosis in A549RT-eto cells

Expression of Beclin-1 is known to be essential for double-membrane autophagosome formation required during the initial steps of autophagy (34). Therefore, we tested whether autophagy impairment via the suppression of Beclin-1 hinders or accelerates FERO-induced apoptosis. We first determined the optimal siRNA concentration for suppression of Beclin-1 expression, which was found to be 100 nM (data not shown). FERO treatment with control siRNA induced apoptosis by activation of caspase-3 or -7 due to cleavage of PARP in a dose-dependent manner (Fig. 6A). Concomitantly, FERO induced autophagy coincided with an upregulation of Beclin-1 and ATG5, increased LC3 I to LC3 II conversion, and a reduction of phospho-mTOR levels in A549RT-eto cells. However, suppression of Beclin-1 with siRNA inhibited FERO-induced apoptosis, indicating that autophagy is necessary for FERO-induced apoptosis. Interestingly, we found that suppression of Beclin-1 by siRNA inhibited the conversion of LC3 I to LC3 II but did not inhibit upregulation of ATG5 expression induced by FERO. These results indicate that conversion of LC3 I to LC3 II occurs later and ATG5 mediated processes take place earlier than Beclin-1 mediated autophagy following FERO treatment (Fig. 6B). In addition, similar to the observation that Z-VAD suppresses FERO-induced apoptosis but does not block downregulation of P-gp, we also found that suppression of Beclin-1 does not block FERO-mediated decrease of P-gp expression despite the inhibition of FERO-induced apoptosis. This result indicates that downregulation of P-gp expression is necessary but not sufficient for the progression of apoptosis.

Figure 6. Inhibition of autophagy by Beclin-1 siRNA transfection reduces FERO-induced apoptosis in A549RT-eto cells. (A) A549RT-eto cells were treated with FERO (0, 0.5 and 1 mM) for 24 h following transfection with control or Beclin-1 siRNAs (100 nM). Cell viability was observed under a light microscope and measured using the MTT assay. Data were calculated as percent of relative cell viability and expressed as the mean of three experiments (*p<0.05 control siRNA vs. Beclin-1 siRNA at FERO 0.5 mM; **p<0.05; control siRNA vs. Beclin-1 siRNA at FERO 1 mM). (B) Cell lysates from the treated A549RT-eto cells were prepared and separated by 12% SDS-PAGE. Cleavage of PARP protein was detected for apoptosis by immunoblotting. Protein levels of Beclin-1, ATG5, mTOR, P-mTOR and P-gp, and cleavage of LC3 I were examined by immunoblotting using the corresponding antibodies.

To verify that autophagy is necessary for FERO-induced apoptosis, we administered rapamycin (RAPA), an inhibitor of mammalian target of rapamycin (mTOR), in order to accelerate A549RT-eto cell autophagy in the presence of FERO (35,36). FERO induced apoptosis in A549RT-eto cells as well as dual administration of FERO and RAPA was observed to enhance apoptosis (Fig. 7A). Co-treatment of FERO and RAPA induced autophagy by upregulating Beclin-1 and ATG5 expression, and converting LC3 I to LC3 II (Fig. 7B). These results suggest that FERO-induced autophagy promotes apoptotic cell death in A549RT-eto cells.

Figure 7. Activation of autophagy by rapamycin increases FERO-induced apoptosis in A549RT-eto cells. (A) A549RT-eto cells were treated with FERO (0.25 mM) alone, rapamycin (RAPA; 10 nM) alone and FERO plus RAPA for 24 h. Cellular morphology was observed under light a microscope and cell growth was measured by MTT assay. Data were calculated as percent of relative cell viability and expressed as the mean of three experiments (*p<0.01, Mock vs. FERO; **p<0.05; FERO vs. FERO+RAPA). (B) Cell lysates from the treated A549RT-eto cells were prepared and separated by 12% SDS-PAGE. Expression levels of Beclin-1, ATG5, phospho-mTOR, mTOR and P-gp were examined by immunoblotting using the corresponding antibodies. Cleavage of PARP and LC3 I was detected by immunoblotting using the corresponding antibodies.

Discussion

Presently, identification of effective chemotherapeutic agents in phytochemicals of great interest to cancer biologists, because of MDR in cancer cells, which are developed following prolonged treatment with anticancer drug. Feroniellin A (FERO) was reported to possess in vitro cytotoxicity in human KB carcinoma and HeLa carcinoma cells (25). However, the molecular mechanism by which this compound exerts cytotoxicity in these cancer cell lines is unknown. For the first time, we have provided a line of evidence showing that FERO reverses MDR1 activity by downregulation of NF-κB, which leads to enhanced apoptotic susceptibility in A549RT-eto cells. Consistent with our results, other studies have shown that metformin, mollugin, and puerarin, which are derived from plants also reverse MDR1 activity effectively through downregulation of NF-κB in human breast cancer cells resistant to adriamycin (30,37,38). Moreover, our study shows that inhibition of apoptosis with Z-VAD does not decrease P-gp expression, indicating that downregulation of P-gp is necessary but not sufficient for apoptosis. As SIRT1, β-catenin and HIF-1α have also been associated with MDR in many cancer models (39–41), we are currently investigating whether FERO affects the expression levels and activity of these proteins, rendering A549RT-eto cells susceptible to apoptosis.

Detailed reviews of autophagic progress can be found as described elsewhere (33,42). Briefly, autophagy is initiated by activation of the ATG1 complex, which includes ATG1/ATG13/ATG17, among other components. Next, autophagosome nucleation occurs, which requires class III phsophatidylinositol-3-kinase plus Beclin-1/ATG6 in addition to several other factors to recruit proteins and lipids involved in autophagosome formation. Vesicle elongation and completion are mediated by two-ubiquitin-like systems: ATG7 (E1-like) and ATG3 (E2-like), which are necessary for the lipid modification of LC3 (phosphatidylethanolamine; PE), which requires initial cleavage of LC3. Subsequently, the ATG12/ATG5/ATG16 complex mediates LC3-PE binding to the autophagosome membrane. Finally, the completed autophagosome fuses with lysosomes, where the autophago-some contents are degraded. We observed that suppression of Beclin-1 in the presence of FERO results in the inhibition of LC3 I to LC3 II conversion but does not inhibit ATG5 expression. Considering the sequential progress of autophagy, we assume that LC3 I to LC3 II conversion is very closely linked to Beclin-1 activity but that suppression of Beclin-1 itself does not affect FERO-induced ATG5 expression. Although we have observed that autophagy impairment such as Beclin-1 suppression with siRNA generated reactive oxygen species (ROS) (data not shown) as reported elsewhere (43,44), we also found that autophagy impairment with Beclin-1 suppression in fact protects cells from FERO-induced apoptosis. Based on this finding, we speculate that the amount of ROS generated by autophagy impairment does not influence aggravation of A49RT-eto cell survival. In addition, apoptosis inhibition with Z-VAD did not block FERO-induced autophagic progress such as upregulation of Beclin1 and ATG5, and converting LC3 I to LC3 II (Fig. 4B), indicating that autophagic progress is the upstream event of apoptosis.

Recently, autophagy inhibitors such as chloroquine (CQ) or hydroxychloroquine have been used in clinical trials because autophagy is believed to affect tumor survival (45–47). Therefore, it is possible that conventional chemotherapy with autophagy inhibitors may increase tumor cytotoxicity. In a mouse prostate cancer model, co-treatment with an Src family kinase inhibitor and CQ increased tumor cell numbers in vitro but reduced tumor cell growth in vivo (48). Conversely, since it has been suggested that autophagy also plays a role as a tumor suppressor, introduction of autophagy stimulants such as mTOR inhibitors with conventional chemotherapeutic drugs leads to accelerated tumor cell death compared to treatment with anti-cancer drugs alone (49,50). Since there are conflicting reports as to the positive and negative effects of autophagy on tumor cell death (32,33), the field needs to be cautious when attempting to manipulate autophagy to improve clinical outcomes. In conclusion, our study has demonstrated that FERO-induced autophagy is required for apoptotic progression and suggests that autophagy plays a role as a tumor suppressor in our model.

Acknowledgements

This study was supported by a grant from the World Class University Program (R31-2008-000-20004-0) through the National Research Foundation funded by the Korean government, and a grant from the Biomedical Fusion Technology Development Project of the Korea Research Institute of Bioscience and Biotechnology (KGM1231312). This study was also supported by the Office of the Higher Education Commission, Thailand, under the Strategic Scholarships Fellowships Frontier Research Networks (Specific for Southern region) for the Join PhD Program Thai Doctoral degree program; a CHE-SSR-PhD SW Scholarship to C.K.

References