This article is mentioned in:

Introduction

In eukaryotes, autophagy is a physiological response for the cell to undergo intracellular protein degradation and organelle turnover. It is one of the intrinsic cellular properties for maintaining cellular energy homeostasis under nutrition deficiency and stress. Recent progress has further extended the investigation of autophagy to tumorigenesis and cancer therapy. However, whether it represents a procancer or anticancer mechanism is far beyond our understanding (1). Autophagy reflects a complicated cellular process in hepatocellular carcinoma (HCC), a common cancer in Chinese and Asian populations (2). Deficiency/attenuation of autophagic function or downregulation of autophagy-related genes (ATGs), key regulator genes in autophagy, were significantly associated with the occurrence or poor prognosis of HCC (3–5). However, autophagy appears to mediate chemotherapy resistance as noted in in vitro experiments. For example, inhibition of autophagy was found to facilitate the killing of HCC cells by chemical drugs (6–14). This is of importance in clinical implications, as multidrug resistance (MDR) is one of the detrimental characteristics of HCC (2). An imminent task in cancer research is to elucidate the role of autophagy in HCC cells under variable microenvironments before we can address whether the process of autophagy may become a potential target for HCC therapy.

The high rate of proliferation and other adverse factors inherent in cancer cells usually result in an overload of endoplasmic reticulum (ER), leading to accumulation of misfolded and/or unfolded proteins in ER, a condition referred to as ‘ER stress’. An unfolded protein response (UPR) is subsequently evoked to alleviate this stress by activating a group of signal transduction pathways and the transcription of genes. Tunicamycin (TM) is an antibiotic, which blocks the formation of N-acetylglucosamine-lipid intermediates, thereby preventing glycosylation and maturation of proteins (15). TM is widely accepted as an ER stress stimulus since it induces the accumulation of unfolded proteins in the ER lumen. In HCC, a number of environmental factors such as hypoxia, viral infection, chemicals or radiation stimulation can trigger ER stress (16–21). Moreover, ER stress is involved in several signal pathways related to hepatocellular proliferation, survival and apoptosis (22–24).

ER stress and autophagy in HCC often share the same stimuli (10,18,21,25,26). Yet, whether or not ER stress itself triggers autophagy remains unknown, and the role of ER stress and autophagy in HCC cell survival and death is still unsolved. The aim of this study was to investigate autophagic responses in the human HCC cell line HepG2 under ER stress stimulation and its consequent effect on cell survival and death.

Materials and methods

Reagents

TM from Streptomyces sp. (cat. no. T7765), 3-methyladenine (3-MA; cat. no. M9281), chloroquine diphosphate salt (CQ; cat. no. C6628) and rapamycin (2.5 mg/ml in DMSO, cat. no. R8781) were purchased from Sigma-Aldrich (USA). To prepare stock solutions, 3-MA and CQ were dissolved in sterile ultrapure water, while TM was dissolved in DMSO with a final concentration of DMSO in the culture medium no more than 1/1,000 (v/v). Earle’s balanced salt solution (EBSS) and Dulbecco’s modified Eagle’s medium (DMEM) were obtained from Invitrogen (USA); fetal bovine serum (FBS) was obtained from HyClone. The Cell Counting Kit-8 (CCK-8) was purchased from Dojindo Laboratories (Japan). The primary antibodies used for western blot analysis were anti-LC3 antibody (cat. no. PM036; MBL Co., Ltd.); anti-Beclin 1 antibody (cat. no. ab16998), anti-GRP78 BiP antibody (cat. no. ab53068; both from Abcam, Inc., USA) and anti-caspase-3 antibody (cat. no. 9662; Cell Signaling Technology).

Cell culture

The HCC cell line HepG2 was obtained from the Cell Bank of Shanghai Institute for Biological Science, Chinese Academy of Sciences. Cells were routinely cultured in DMEM supplemented with 10% (v/v) FBS and maintained in a humidified incubator with 5% CO2 at 37°C.

Cell proliferation and viability analysis

Cell proliferation assays were performed using CCK-8. Re-suspended cells were seeded onto 96-well plates at a concentration of 104 cells/well, and incubated for 24 h to allow adherence. The cells were then exposed to TM for the indicated concentrations and time points. For the inhibition assay, a stock solution of the autophagic inhibitors (3-MA or CQ) or an equal volume of PBS was added into the culture medium 1 h prior to TM application. The final concentrations of 3-MA and CQ were 5 mmol/l and 5 μg/ml, respectively. After incubation, cell medium in each well was substituted with 100 μl pre-prepared WST®-8 solution (dilution of 1:10 in fresh DMEM); the plate was then incubated for an additional 1 h at 37°C. Absorbance at 450 nm for WST-8 formazan was measured using the Elx800 absorbance microplate reader (BioTek Instruments, Inc). Triplicate measurements were made for each treatment subgroup in one plate, and the average optical density (OD) of the three wells was calculated. The average OD450nm of another two cell-free wells reserved in each plate was calculated as the background value. The cell viability was calculated according to the formula: % viability = (OD450nm treated cells - OD450nm background) / (OD450nm control cells - OD450nm background) × 100. At least three independent experiments were performed to generate the mean data for each intervention.

Transmission electron microscopy (TEM)

Electron microscopy is a traditional and reliable method to observe autophagic compartments (27,28). We performed TEM to demonstrate autophagosome formation in HepG2 cells after ER stress stimulation. Briefly, HepG2 cells were grown in 100-mm-diameter dishes. Following treatment with DMEM (control), EBSS, 1 μg/ml rapymycin or 2.5 μg/ml TM for the indicated time periods, respectively, cells were collected and centrifuged at 3,000 rpm. Cell pellets were primarily fixed in a solution with 2.5% glutaraldehyde (v/v) overnight, and then in 1% osmium tetraoxide (v/v) for 1 h for secondary fixation. After dehydration in a series of concentrations of ethanol, the cells were finally embedded in Epon 812. Ultrathin (70 nm) sections were cut on an NOVA ultramicrotome (LKB, Sweden), stained with uranyl acetate (saturated aqueous solution) and lead acetate, and then examined under a transmission electron microscope (JEM-1230; Jeol Ltd., Japan).

Western blot analysis

Briefly, cells were washed with pre-cooled PBS twice and lysed on ice for 30 min in RIPA lysis buffer containing 50 mmol/l Tris-HCl (pH 7.4), 150 mmol/l NaCl, 1 mmol/l phenylmethylsulfonyl fluoride (PMSF), 1 mmol/l ethylene diamine tetraacetic acid (EDTA), 0.1% sodium dodecyl sulfate (SDS), 1% Triton X-100 and 1% sodium deoxycholate. Lysates were centrifuged at 12,000 rpm for 10 min at 4°C. The supernatant was transferred into a new tube; the protein concentration was determined using BCA protein assay. Lysates were incubated with 2X Laemmli sample buffer (Bio-Rad, Hercules, CA, USA) and heated for 10 min at 95°C. The proteins from each sample were separated by sodium dodecyl-sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred onto polyvinylidene difluoride (PVDF) membranes (Millipore, Bedford, MA, USA). Membranes were blocked for 2 h at room temperature in PBS containing 5% defatted milk powder and washed with Tris-buffered saline/Tween-20 (TBST) three times, then incubated overnight at 4°C with the diluted primary antibody (anti-LC3, 1:1,000; anti-GRP78, 1:2,000; anti-Beclin 1, 1:1,000; anti-β-actin, 1:500) in PBST. After washing in PBST, the membranes were then incubated for 1 h at room temperature with the secondary antibody (goat anti-mouse-HRP, 1:10,000). The immunoreactive proteins were detected using the enhanced chemiluminescence (ECL) kit (Pierce) and the chemiluminescent signals were captured by ImageQuant™ LAS-4000 Mini Imager (Fuji, Japan). For quantitative analysis, the integrated density of each band was obtained using ImageJ software (US National Institutes of Health).

Data analysis

At least three independent experiments were performed for the numerical variables unless otherwise stated. Data are expressed as the means ± standard deviation in each group. Student’s t-test and one-way ANOVA were used to examine the differences between two or multiple groups, respectively. Statistical tests were performed using SPSS 15.0 software. A probability-value <0.05 was considered to indicate a statistically significant result.

Results

ER stress triggers accumulation of autophagic compartments in HepG2 cells

After exposure to TM (2.5 μg/ml) for 24 h, TEM revealed that there was a greater number of autophagic compartments accumulated in the TM-treated cells (Fig. 1A) when compared with that in the non-treated cells (Fig. 1B); the latter also showed rare autophagic compartments. Autophagosomes were recognized as double membrane vacuolar structures containing cytoplasmic contents (Fig. 1A and C). Other types of autophagic compartments such as autolysosomes also appeared in the form of membrane vacuolar structures containing high-density materials (Fig. 1A). Two known autophagy inducers, starvation medium EBSS (Fig. 1D) and rapamycin (Fig. 1E), acting via amino acid deprivation and mTOR inhibition, respectively, were used as the positive controls, and accumulation of autophagic compartments in response to these interventions clearly indicated autophagy.

Figure 1 Autophagic compartments in HepG2 cells. Ultrastructure observed under transmission electron microscopy. Ly, lysosome; arrows and triangles indicate autophagosomes and autolysomes, respectively. (A) TM (2.5 μg/ml) incubation for 24 h; ×10,000. (B) Control group; ×10,000. (C) Magnified autophagosome from panel A; ×30,000. (D) EBSS incubation for 6 h; ×10,000. (E) Rapamycin (1 μg/ml) incubation for 24 h; ×10,000.

ER stress induces LC3 conversion and enhances autophagic flux in HepG2 cells

Western blot analysis was used to detect several key proteins involved in the process of autophagic flux and ER stress. GRP78, a resident protein of ER, was used as a molecular marker of ER stress. The microtubule-associated protein 1-light chain 3 (MAP1-LC3, LC3), also known as Atg8 in yeast, is cleaved by the Atg4 homologue to form LC3-I (29). Upon induction of autophagy, LC3-I conjugates to phosphatidylethanolamine (PE) to become LC3-II which anchors to the autophagosomal membrane throughout the process of autophagosomal maturation until degradation by lysosomes (30). Although LC3-II is a reliable marker of autophagosomes (31), the process of conjugation of LC3-I to PE to form LC3-II is more indicative of the autophagic reaction. Thus, the conversion of LC3-I to LC3-II is considered to be a surrogate of autophagy induction. Both proteins can be detected by protein electrophoresis and immunoblotting and the LC3 ratio (calculated by scanned intensities of LC3-II/LC3-I in each group) is a measure of autophagic LC3 conversion (32).

Initially, we performed two positive controls for autophagy detection: the two well-known autophagy inducers EBSS (incubated for 6 h) and rapamycin (1 μg/ml for 24 h). Both induced overexpression of LC3-II and an increase in the LC3 ratio (Fig. 2A and B). Consequently, decreased expression of LC3-I and increased expression of LC3-II were observed in HepG2 cells after ER stress stimulation (Fig. 2C). Accordingly, the LC3 ratio (LC3-II/I) was elevated in the cells treated with 2.5 and 5 μg/ml TM when compared to the ratio in the unstressed cells (P<0.05). This indicates an increase in conversion of LC3-I to LC3-II after ER stress (Fig. 2D). 3-MA, a PI3K class III inhibitor (also named Vps34) suppressed the LC3 conversion which suggests the involvement of PI3K ClassIII/Vps34 in the observed autophagy by ER stress. In parallel, we ascertained the occurrence of ER stress in HepG2 cells through evidence of GRP78 overexpression after TM stimulation (Fig. 2C). Furthermore, we also demonstrated a time-dependent increase in autophagy induction in the ER-stressed HepG2 cells by treating the cells with TM for 3, 6, 9, 12 and 24 h (Fig. 2E and F). Another autophagy-related protein, Beclin 1, appeared to be unaffected either by the ER stressor or co-existing 3-MA (Fig. 2C).

Figure 2 LC3 conversion assay in ER-stressed HepG2 cells by western blotting. (A) Positive controls used in the autophagic LC3 conversion. Cells were treated with DMEM, EBSS for 6 h or rapamycin (RAPA) (1 μg/ml) for 24 h, respectively; β-actin was used as the loading control. (B) Quantitative analysis of data from A. Integrated density of LC3-I and LC3-II bands expressed as arbitrary units measured by ImageJ software; LC3-II/LC3-I ratio for each group was calculated; *P<0.05, comparison between the indicated two groups. (C) Cells were treated with indicated concentrations of TM for 12 h in the absence or presence of 3-MA (5 mmol/l). LC3 together with GRP78 and Beclin 1 were detected. (D) Quantitative analysis of LC3 conversion data from C; *P<0.05, comparison between the indicated two groups. (E) Cells were treated with 2.5 μg/ml TM for different time durations. (F) Quantitative analysis of LC3 conversion data from E, *P<0.05, comparison with the 0 h group.

An important consideration in autophagy detection was whether or not the observed accumulation of autophagosomes (or its surrogate, LC3-II) was indicative of an authentic autophagy induction or simply the blockage of autophagosomal degradation in the lysosomes. Thus, performing an ‘autophagic flux’ assay by using a lysosomal inhibitor was necessary to distinguish between the aforementioned two possibilities (31). We performed ‘LC3 turnover assay’ to demonstrate autophagic flux enhancement in the ER-stressed HepG2 cells. Briefly, we pretreated the cultured cells with lysosomal inhibitor CQ for 1 h before TM, then compared changes in LC3 expression with those in the absence of CQ. The results showed a marked increase in LC3-II accumulation after lysosomal inhibition. As shown in Fig. 3, the upregulated LC3-II was lysosomal-dependent. This indicated an authentic induction of autophagy in the HepG2 cells by TM, not a defect in autophagosomal degradation.

Figure 3 Effect of CQ on LC3 turnover analysis in HepG2 cells after ER stress. (A) Cells were treated with TM (2.5 or 5 μg/ml) for 24 h in the presence or absence of CQ (5 μg/ml). (B) Quantitative analysis of intensities of bands in A; band intensities of LC3-II expressed as arbitrary units. *P<0.05, comparison between the indicated two groups. (C) Cells were treated with 2.5 μg/ml TM for different durations in the presence or absence of CQ (5 μg/ml). (D) Quantitative analysis of LC3-II intensity data from C expressed as arbitrary units. *P<0.05, comparison between the indicated two groups.

Role of autophagy in the maintenance of viability of ER-stressed HepG2 cells

ER-stress associated cell death was demonstrated in HepG2 cells by the cleavage and activation of caspase-3 after treatment with 1.25, 2.5 or 5 μg/ml TM for 48 h (Fig. 4A). Furthermore, results of the CCK-8 test also showed a decrease in HepG2 cell viability after ER stress stimulation, and the cell viability was well correlated with the dose and duration of TM incubation (Fig. 4B). We confirmed induction of autophagy in the ER-stressed HepG2 cells. To further ascertain the role of the observed autophagy in ER-stress related cell death and survival, we introduced two autophagy inhibitors which blocked the autophagic pathway at different sites; 3-MA, an inhibitor of phosphatidyl inositol 3-kinase, blocks initiation of autophagic vesicles and CQ, a lysosomal inhibitor, affects the downstream process of autophagy. In the presence of 3-MA (5 or 10 mmol/l at final concentrations), 48 h of TM stimulation induced a more significant decrease in cell viability than that without 3-MA co-incubation, and 10 mmol/l 3-MA exerted a greater degree of cell viability suppression than 5 mmol/l 3-MA (Fig. 5A). The data described above showed that inhibition of autophagy facilitated ER stress-associated cell death, i.e., enhancement of autophagy induced by ER stress contributed to cell viability maintenance. However, CQ did not appear to exert an effect on the viability of ER-stressed HepG2 cells (Fig. 5B), although lysosome inhibition was considered to a certain degree as a dysfunction of autophagy.

Figure 4 ER stress-induced HepG2 cell death. (A) Caspase-3 cleavage in HepG2 after treatment with TM (1.25, 2.5 and 5 μg/ml) for 48 h, as detected by western blotting. (B) Changes in HepG2 viability after incubation with TM (1.25, 2.5 and 5 μg/ml) for the indicated time durations. At least three independent experiments were performed; data are presented as means ± SD.

Figure 5 Influence of autophagy inhibitors (3-MA or CQ) on cell viability of ER-stressed HepG2 cells. Data are presented as means ± SD. Cells were incubated with different concentrations of TM for 48 h in the presence or absence of (A) 3-MA (5 or 10 mmol/l) or (B) CQ (5 or 10 μg/ml) and cell viability was measured by CCK-8. *P<0.05, compared to ‘0 μg/ml TM and 0 mmol/l 3-MA’ subgroup. #P<0.05, compared to ‘0 mmol/l 3-MA subgroup’ for each group of different TM dosage.

Discussion

Induction of autophagy in HepG2 cells by TM

Although both UPR and autophagy have been interpreted as stress responses in eukaryotes, direct evidence linking ER stress to autophagy was first published in yeast (33), following by evidence in normal and transformed human cell lines (34–39). However, the so-called ‘increase in autophagy’ in most published studies (in fact, accumulation of autophagosomes) can be attributed not only to induction of autophagy but also blockage of the degradation in lysosomes (31). Few studies reportedly have performed lysosomal inhibition assay to verify the fluency of the whole autophagic flux. Thus, the purpose of the present study was to demonstrate the induction of autophagy and enhancement of autophagic flux in HepG2 cells following ER stress, by observing ultramicroscopic accumulation of autophagosomes, alterations in biochemical markers and functional enhancement of autophagic degradation in lysosomes.

We confirmed an authentic induction in autophagy and enhancement of autophagic flux in the ER-stressed HepG2 cells. However, in another study, ER stress-induced autophagic degradation of apoB in the HepG2 cell line was unable to be demonstrated (39). Yet, we did not refute our theory as the authors of that study also demonstrated enhancement of autophagic apoB degradation in two other hepatoma cell lines, and alternative degradation pathways for apoB may coexist in HepG2 cells; the propensity of apoB degradation varies in different cell lines.

ER stress signals to autophagy are still unresolved. PI3K class III/Vps34 is clearly involved in ER stress-induced autophagy in HepG2 cells, as 3-MA can suppress autophagy induction by TM in HepG2 cells. We also detected Beclin 1 expression in HepG2 cells. Decreased expression of autophagy protein Beclin 1 is associated with poor prognosis of HCC (5), and induction of autophagy in prostate cancer cells involves interaction between Bcl-2 and Beclin 1 at the ER (40). Yet, we failed to find a difference in its expression between ER-stressed and unstressed cells; possibly a distinct signaling pathway exists in ER stress triggered autophagic flux in HCC.

Role of ER stress-induced autophagy in the survival of HepG2 cells

ER stress leads to accumulation of unfolded protein, and autophagy is associated with the degradation of protein and organelles (41). Based on this finding, ER stress and related autophagy are cancer cell survival mechanisms in response to hypoxia, nutrition deficiency and stress. However, autophagic cell death (ACD, type II cell death) was proposed recently by the Nomenclature Committee on Cell Death (NCCD) as a new sub-routine of cell death (42), yet the role of autophagy in cell death has not yet been clarified (43–47). One study reported that ER stress-induced autophagy exerts differential effects on cell survival: pro-survival effects in cancer cells and pro-death effects in normal cells (34). Our results demonstrated a decrease in cell survival in the TM-treated HepG2 cells in the presence of the autophagic inhibitor 3-MA, indicating that a pro-survival effect of autophagy played a dominant role in ER-stressed HepG2 cells, similar to that in neuroblastoma cells (38). Autophagy inhibition induced by chemical drugs in HepG2 cells and other hepatocarcinoma cell lines also exhibited enhancement of cell death (48,49), but whether ER stress was involved in these mechanisms is unknown.

We did not observe an effect of the lysosomal acidification inhibitor, CQ, which blocks the downstream process of autophagy. This finding suggests that the linkage between autophagy signaling and death signals may be located upstream in the process of autophagy, as the targets of 3-MA were located at the initiation of autophagy. In fact, ER itself can function as a switching point in autophagy and death by interaction between Bcl-2 family proteins occurring on the ER membrane (50,51).

In conclusion, ER stress in HepG2 cells elicits enhancement of autophagic flux. This contributes to the maintenance of cell viability. Our results may help to eluciate the understanding of carcinogenesis and drug resistance of HCC.

Acknowledgements

We thank Professor Qi-Xing Zhu of Anhui Medical University for providing the experimental platform and encouragement to complete the study. This study was supported by grants from National Natural Science Foundation of China (no. 81071986 and no. 81272739), and by a key project of the Anhui Provincial Office of Education, China (no. KJ2011A154). C.W. gratefully acknowledges support from the Biotechnology and Biological Sciences Research Council (BBSRC) (BB/G015554/1; BB/I025379/1).

References