Introduction

Colorectal cancer is the third most commonly diagnosed cancer worldwide (Obrand and Gordon 1997), and 40–50% of patients who undergo potentially curative surgery will ultimately relapse and die of metastatic disease (Jemal et al. 2011). Considering the high morbidity, treatment options that provide little efficacy, and the substantial growth of cancer incidence every year, more effective treatments are desperately needed. Over the past few decades, phytochemicals have exhibited diverse pharmacological properties including anti-tumorigenic, anti-inflammatory, anti-oxidative and anti-bacterial activities (Satoh 2014) with little toxicity, so phytochemicals show promise as potential anti-cancer drugs.

The Actinidiaceae family, which includes more than 54 species such as Actinidia chinensis, Actinidia deliciosa, and Actinidia eriantha, is widely distributed in most regions of China and has traditionally been used to treat a variety of diseases including hepatitis, edema, rheumatic arthritis, dysentery, lymphoid tuberculosis gastric and breast cancer (Liang et al. 2007). Several types of compounds in the Actinidiaceae family have been reported in previous literature, such as triterpenoids, flavones, anthraquinones and polysaccharides (Lai and Xu 2007; Fu et al. 2010; Zhou et al. 2010). The root of A. deliciosa, a representative herb in the Actinidiaceae family (Lai and Xu 2007), possesses numerous, beneficial pharmacological activities (Zhong et al. 2003; Bai and Qiu 2006; Liang et al. 2007). In our study, three known ursane-type triterpenoids isolated from the roots of A. eriantha, another herb in Actinidiaceae family, were characterized by nuclear magnetic resonance and mass spectrometry. Among them, 2α,3α,24-trihydroxyurs-12-en-28-oic acid (TEOA) is naturally the most abundant. However, the pharmacological mechanisms of TEOA have not been investigated.

Various natural plant products exerted antitumor capabilities via induction of autophagy (Zhang et al. 2012). Autophagy, which can be activated by classical endoplasmic reticulum (ER) stress inducers in mammalian cells, is important for a number of cellular functions to maintain cellular homeostasis (Ding et al. 2007). Although autophagy is generally considered to be a non-selective process, accumulating evidence shows that it can also selectively degrade specific targets (Mizushima et al. 2008). Selective degradation of mitochondria (mitophagy) (Kim et al. 2007; Kanki and Klionsky 2008; Kirkin et al. 2009) can occur in a specific developmental process, such as the maturation of erythroid cells (Schweers et al. 2007; Tolkovsky 2009), or pathological mitochondrial damage (Sandoval et al. 2008).

The ER is the central intracellular organelle in the secretory pathway. It is responsible for protein translocation, folding, and post-translational modifications that allow further transport of proteins to the Golgi and ultimately to vesicles for secretion or display on the plasma membrane (Sano and Reed 2013). To restore ER homeostasis, following misfolding of proteins, cells activate the unfolded protein response (UPR), a signaling pathway governed by three major ER stress sensors: PKR-like ER kinase (PERK), inositol requiring enzyme 1 (IRE1) and activating transcription factor 6 (ATF6). Together, these signaling pathways coordinate a temporal shut down in protein translation and induce a complex program of gene transcription to restore the folding capacity of the ER (Verfaillie et al. 2012). Recent studies have shown that intense or persistent ER stress may induce cell death through an autophagic mechanism (Lugea et al. 2011).

While a number of contributing events have been identified, the exact molecular mechanism of mitophagy has not been clearly elucidated. Early studies in yeast, suggest that Parkin, an E3 ubiquitin ligase, whose loss-of-function mutations are responsible for the most common recessive disorder, Parkinson’ disease (Zhang and Ney 2009), can be selectively recruited to depolarized mitochondria to promote their degradation via autophagy (Kitada et al. 1998; Geisler et al. 2010). In addition, PINK1, a mitochondrial-located kinase and component of the autophagy machinery, was shown to be expressed and imported into all mitochondria and then rapidly degraded by proteolysis (Narendra et al. 2010), however, when the mitochondria become damaged, PINK1 accumulates, resulting in the recruitment of the Parkin to facilitate mitochondrial protein ubiquitination and the subsequent fusion of mitochondria with lysosomes (Youle and Narendra 2011). Moreover, the multi-signaling adaptor molecule p62/SQSTM1 can directly interact with ubiquitin and LC3, promoting autophagic degradation of ubiquitinated proteins, peroxisome and bacteria (Durcan and Fon 2015). These studies suggest that Parkin, ubiquitn, and p62 play functional roles in preparing mitochondria for mitophagy (Kirkin et al. 2009).

In this study, we investigated the potential pharmacological mechanism of TEOA in SW620 colon cancer cells, demonstrating the potential of this natural compound to serve as a chemopreventive agent.

Materials and methods

Materials

LC3-II, Ubiquitin, Caspase-9, PARP, β-Actin, PINK1, and Parkin antibodies were purchased from Cell Signaling Technology, Beverly, MA. GRP78, CHOP/GADD153, PERK, p-PERK, p62, elF2α, and p-elF2α antibodies were purchased from Santa Cruz Biotechnology, Santa Cruz, CA. The second antibodies goat anti-rabbit IgG and goat anti-mouse IgG were purchased from Biosharp Biological science and technology, Shanghai, China. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) was purchased from Solarbio, Beijing, China and N-acetyl-l-cysteime (NAC) was purchased from Sigma-Aldrich, St. Louis, Mo. 3-Methyladenine (3-MA) and Dimethyl sulfoxide (DMSO) were purchased from Gibco, Grand Island, New York, USA. All cell media were purchased from Gino Biomedical Technology Co., Ltd. Hangzhou, China and fetal bovine serum (FBS) was from Gemini, CA, USA. The DNA content quantitation assay kit, MDC assay kit and reactive oxygen species (ROS) assay kit were obtained from KeyGEN Nanjing, China. BCA protein concentration determination kit, Annexin V-FITC apoptosis detection kit and ECL system were obtained from Beyotime, Shanghai, China).

Plant material

The roots of A. eriantha were provided by Lishui People’s Hospital, Zhejiang province. A voucher specimen was deposited in the Department of Biomedical Engineering, Zhejiang University, Hangzhou, China.

Extraction and isolation of TEOA

The dried roots of A. eriantha (2.0 kg) were soaked in 2.5 times of dichloromethane (v/v) for 30 min, then extracted twice with ultrasound for 1 h each time (2 × 1 h). The combined extracts were poured through a 200 mesh sieve and evaporated to no dichloromethane in vacuum. After removal of the solvent, the residue (40.6 g) was chromatographed over silica gel (100–200 mesh) column, and eluted with CH2Cl2–MeOH to obtain seven portions (P1-7). Portion 5 (3.8 g) was further carried out by silica gel chromatography (100–200 mesh), and eluted with a gradient CH2Cl2–MeOH (95:5) to obtain six portions (P5.1–P5.6). Fractions were evaporated, and the portions 5.3–5.5 were combined after inspection of the analytical TLC (CH3Cl–MeOH 7:3) to give a residue of 3 g. The residue (3 g) was subjected to Sephadex LH-20 column with CH2Cl2–MeOH (1:1) to yield TEOA (1.48 g).

Cell lines and cell culture

Human colon cancer cell line SW620, human gastric cancer cell line BGC-823, human liver cancer cell line HepG-2, human hung cancer cell line A549, and human prostate cancer cell line PC-3 were purchased from Type Culture collection of the Chinese Academy of Sciences (Shanghai, China). Cells were cultured in DMEM or RPMI-1640 medium supplemented with 10% heat in-activated FBS and 100 IU/mL of both penicillin and streptomycin. Cells were grown in incubators at 37 °C and in a humid atmosphere of 95% air/5% CO2. All experiments were conducted after cells were cultured for 24 h. TEOA was dissolved in DMSO and diluted with 5% FBS culture medium to a final DMSO concentration of 0.08% in each well.

Cell viability analysis

The cytotoxicity of TEOA in cancer cells was determined by MTT assay in six replicates. Cells were seeded in 96-well plates at 5000 cells/well (SW620, BGC-823, HepG-2, A549, and PC-3) in a final volume of 100 µL. After 24 h incubation, cancer cells were treated with TEOA (0–60 µM) for 24,36 or 48 h at 37 °C, then MTT was added at a final concentration of 2.5 mg/mL and incubated for 2 h. The medium was carefully removed from each well, and 200 µL DMSO was added to each well. The plates were gently shaken for 3 min and the optical density was measured at 570 nm with a microplate reader. The experiments were repeated three times and the cell viability was calculated. Relative cell proliferation inhibition rate (IR) = (1 − average A570 of the experimental group/average A570 of the control group) × 100%.

Observation of cell morphology

SW620, BGC-823, HepG-2, A549 and PC-3 cells were seeded at 2.5 × 105 cells per well in 6-well plates., incubated for 24 h, and treated with vehicle or 30 µM TEOA for another 36 h. Then the cells were stained with crystal violet, and the cellular morphology was observed under a microscope and photographed.

Clonogenic assay

SW620 cells were cultured overnight in a 6-well plate (1 × 104 cells per well) and then treated with TEOA for 15 days to form colonies. Then culture medium was removed and the cell monolayer was fixed with solvent contained methanol and glacial acetic acid (3:1) for 15 min and stained with crystal violet (0.2%) for 30 min to visualize cell colonies. Colonies were counted and normalized to the colonies amount of the control group. Each sample was repeated in triplicate.

Analysis of apoptosis

106 cells were seeded in plates and cultured for 24 h, followed by treatment with medium containing different concentrations of TEOA (0, 30, 40 and 50 µM) for 36 h. Cells were washed twice with cold PBS, and 5 µL of Annexin V-FITC (fluorescein isothiocyanate) and 10 µL of propidium iodide (PI) were added to each sample, respectively. After a 15-min incubation at room temperature in the dark, cells were analyzed using flow cytometry (Becton–Dickinson) equipped with an argon laser (488 nm). The percentage of apoptotic cells was calculated using the internal software system of the FACS can (Becton Dicknson).

Visualization of autophagic vacuoles

Cancer cells were seeded into 6-well plates (2 × 105 cells per well) and cultured for 24 h, followed by treated with TEOA (0, 30, 40 and 50 µM) for 48 h, then the cells were stained with 0.05 mM MDC at 37 °C for 1 h. After washed three times using 1× wash buffer, the cells were covered with collection buffer, and the changes of cellular fluorescence were observed under a fluorescence microscope and photographed (Zeiss Axioplan).

Measurement of the intracellular ROS levels

The ROS levels were measured using fluoroprobe DCF-DA. Cancer cells were seeded into a 6-well plate (5 × 105 cells per well) and cultured 24 h, then the cells were treated with TEOA (0, 30, 40 and 50 µM) or pre-treated with NAC (5 mM) for 2 h and then co-treated with TEOA at various concentrations for 36 h. After treatment, the cells were incubated with 1 µM DCF-DA in the dark at 37 °C for 30 min. The cells were then washed three times with PBS. The fluorescence intensity was measured by flow cytometry.

Western blotting assay

Cancer cells were seeded into a 6-well plate (7 × 105 cells per well) and cultured 24 h. Then the cells were treated with TEOA (0, 30, 40 and 50 µM) or pre-treated with 3-MA (5 mM) for 2 h, and then co-treated with TEOA at various concentrations for 48 h. After treatment, cells were collected from plates by scraping and centrifugation. The cells were washed twice with ice-cold PBS buffer (pH 7.2) and lysed with RIPA buffer for 30 min, which contains 15 mM Tris–HCl, 50 mM NaCl, 5 mM ethyleneglycolteraacetic acid (EGTA), 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM Na3CO4, 10 μg/mL aprotinin, and 10 μg/mL leupeptin, pH 7.4. The lysates were centrifuged at 4 °C at 12,000 g for 15 min. Protein concentrations in the supernatants were determined by using a BCA protein concentration determination kit. Equal amounts of proteins were separated by 7.5% SDS–polyacrylamide gel and then transferred to a polyvinylidene fluoride (PVDF) membrane. The membrane was blocked with 5% non-fat milk in 0.1% Tween-20 in TBS (TBST) for 2 h and incubated overnight with the primary antibodies (1:500–1:1000) at 4 °C. After washing thrice with TBST, the membrane was incubated for 2 h at room temperature with the appropriate secondary detection antibodies (1:5000). The immunoblots were visualized with an ECL system.

In vivo study

Female athymic nude mice (Foxnlnu, 5–7 weeks of age, HuaFuKang Biotechnology Co., Ltd, Beijing, China) were group-housed under constant photoperiod (12 h light:12 h dark) with ad libitum access to sterilized food and water. All experimental protocols were approved by the Review Committee for the Use of Animal Subjects of Changshu Realistic Technology Co., Ltd (Suzhou, China). Xenograft tumors were established by subcutaneous of 3 × 106 SW620 cells in a total volume of 0.15 mL of PBS (BD Biosciences, 354234). To assess the effect of TEOA on SW620 tumor growth, mice with established tumors (mean starting tumor volume, 52.7 ± 9.1 mm3) were randomly divided into the following five groups (6 mice per group) for oral gavage treatment once a day for three weeks: control group (0.5% CMC-Na, 1 mL); positive treatment group (20 mg/kg CTX); and TEOA groups (40, 80, 120 mg/kg TEOA). All the TEOA and CTX powders were suspended in 0.5% carboxymethylcellulose sodium (CMC-Na). Then the weight of the mice, tumor, liver and spleen of the drug-treated mice was measured after the last delivery. Tumor volumes were calculated from caliper measurements using a standard formula (volume = width2 × length × 0.5). The research was conducted in accordance with the Declaration of Helsinki and with the Guide for Care and Use of Laboratory Animals as adopted and promulgated by the United National Institutes of Health.

Statistical analysis

The experiments were repeated at least three times. Data are expressed as the mean ± SD. Differences in the results for two groups were evaluated by the Student’s t test. p < 0.05 was considered to be statistically significant.

Results

Identification of TEOA

TEOA was isolated from the root of A. eriantha as a white powder with a melting point of 211–212 °C. The molecular formula of TEOA was determined as C30H48O5 on based on its HRESIMS, which showed a [M-H]− quasi-molecular ion peak at m/z 487.3435 (Calc. 488.3502). The 1H-NMR spectrum of TEOA showed six angular methyls: δ1.67 (s, 3H), 1.11 (s, 3H), 1.00 (s, 3H), 0.98 (d, J = 8.6 Hz, 3H), 0.95 (d, J = 6.3 Hz, 3H), 0.92 (s, 3H), indicating the presence of a triterpenoid. δ 4.58 (s, 1H), 4.44 (d, J = 10.8 Hz, 1H) indicated the two hydroxyls (Supplementary Material Fig. 1). The 13C NMR spectrum of TEOA showed 30 carbons including six methyls signals: 24.4 (C-27), 24.3 (C-30), 19.4 (C-29), 18.1 (C-26), 17.9 (C-25), and 17.8 (C-24). A signal due to C-13 at δ 139.8 supported the conclusion that it had an ursane skeleton (Supplementary Material Fig. 1). From these data and through comparison with the literature (Sakakibara and Kaiya 1983; Kojima and Ogura 1989), the structure of TEOA was confirmed as 2α,3α,24-thrihydroxyurs-12-en-28-oicacid and the chemical structure of TEOA is shown in Fig. 1a.

Fig. 1
figure 1

TEOA reduces cancer cell viability and induces morphological changes in cancer cells. a The chemical structure of TEOA. b The growth inhibitory effect of TEOA on various cancer cells. Cells were seeded at 5000 cells per well in 96-well plates, incubated for 24 h, and treated with vehicle or TEOA (0, 20, 30, 40, 50 or 60 μM) for 24, 48 or 72 h; The relative cell viability was analyzed by performing the MTT assay. c The cells were treated with 30 μM TEOA for 48 h, stained with crystal violet, and photographed under a microscope (×40 magnification). d The death-inducing effects of TEOA on SW620 cells were determined by counting the colonies formed. Data are presented as mean ± SD or standard error of three independent experiments, *p < 0.05, **p < 0.01

Full size image

TEOA decreases cell viability and alters morphology of cancer cells

To explore the anti-cancer activity of TEOA, the cytotoxic effects of TEOA (0–60 µM) on five cancer cell lines (SW620, BGC-823, HepG-2, A549, and PC-3) were determined using MTT assay. Results showed that TEOA significantly reduced cancer cell viability in a does- and time-dependent manner (Fig. 1b). Among the examined cancer cell lines, SW620 cells were found to be the most sensitive to TEOA treatment.

In addition, optic microscopy revealed morphological changes after treatment with 30 µM TEOA for 48 h. Extensive cytoplasmic vacuolation was observed in SW620, BGC-823, HepG-2, and A549 cells lines after TEOA treatment, whereas more detached and shrunken cells were observed in PC-3 cancer cells (Fig. 1c). This finding clearly indicates that TEOA could potently reduce the cell viability of various cancer cell lines. As SW620 cells were more sensitive to TEOA-induced growth inhibition and cytoplasmic vacuolation, they were used for further investigation.

Further, clonogenic assay was performed to determine the long-term growth inhibitory effect of TEOA. SW620 cells were treated with a series of concentrations of TEOA (0, 10, 20, 25, 30 and 35 µM) for 15 days, and the colony formation was measured. Results showed that the number of surviving colonies was significantly reduced by TEOA in a dose-dependent manner (Fig. 1d).

TEOA induces apoptosis in SW620 cells

Annexin V-FITC/PI staining assay was used to quantify the percentage of apoptotic cells in the total cell population. Flow cytometry analysis results showed that TEOA induced both early and late apoptosis in a concentration-dependent in SW620 cells. As shown in Fig. 2a, after a 48 h treatment, 0, 30, 40, and 50 µM of TEOA slightly enhanced the percentage of apoptotic cells from 6.94% (control) to 11.03, 12.21, and 13.59%, respectively. Next, we investigated the expression of apoptotic protein caspase-9 and its downstream molecule PARP, and the results showed that TEOA treatment activated caspase-9 and PARP (Fig. 2c–e). As showed in Fig. 2c, TEOA over 40 µM significantly increased the cleavage of caspase-9, while below 40 µM had minor effect to caspase-9 (Fig. 2c, e). The results of cleaved PARP were similar to the cleavage of caspase-9, but even treatment with 30 µM TEOA induced dramatic increasing in the cleavage of PARP (p < 0.05, Fig. 2d). Taken together, our data suggest that TEOA could reduce cell viability through induction of apoptosis in human colon cancer cells.

Fig. 2
figure 2

TEOA induces apoptosis in SW620 cells. a Dose-dependent induction of apoptosis by TEOA analyzed using annexin V/PI staining of SW620 cells at 24 h. b Histograms from cell apoptosis were shown for analyzed cells. c Effect of different doses of TEOA on apoptotic markers in SW620 cell after 48 h treatment. The cleavage of PARP and caspase 9 by various concentrations of TEOA in SW620 cells were determined by western blotting. β-actin was used as a loading control. d, e The gray analysis of cleaved caspase-9 and PARP. Data are presented as mean ± SD or standard error of 3 independent experiments, *p < 0.05, **p < 0.01

Full size image

TEOA induces autophagic cell death

The cytotoxicity of some chemo-therapeutic drugs is mediated by the activation of autophagy rather than the apoptotic cell death. To further investigate whether the TEOA-induced cytoplasmic vacuolation-mediated death was autophagic death, cells were stained with MDC following TEOA treatment. The number of MDC-labeled mature autophagosomes was increased in a dose-dependent manner in TEOA-treated SW620 cells (Fig. 3a). LC3-II, a hallmark of autophagy, was upregulated in SW620 cells after TEOA treatment. The expression level of another autophagic marker, p62, was also increased. As expected, the expressions of LC3-II and p62 were significantly decreased by treatment with the autophagy inhibitor 3-methyladenine (3-MA) (Fig. 3b).

Fig. 3
figure 3

TEOA induces autophagic cell death in SW620 cells. a SW620 cells exposed to the indicated concentrations of TEOA for 24 h and autophagolysosomes were observed using MDC staining. Bars 10 μm. b SW620 cells were treated with TEOA (0, 30, 40 or 50 μM) and/or 3-MA (5 mM) for 48 h, and then were lysed to determinate the levels of LC3-II, p62 and β-actin. c, d The gray analysis of cleaved LC3-II and p62. e Viability of SW620 cells after treatment with TEOA (0–60 μM) with or without the autophagy inhibitor 3-MA for 48 h. Data are presented as mean ± SD or standard error of 3 independent experiments. f Apoptosis was measured through flow cytometry analysis of SW620 cells following pre-treatment with 5 mM 3-MA. Top Dose-dependent effect of TEOA evaluated by annexin V/PI staining of SW620 cells at 24 h. Bottom Histogram representing three independent experiments. Data are presented as mean ± SD or standard error of three independent experiments. *p < 0.05. g SW620 cells were treated with TEOA or combination of TEOA and 3-MA for 48 h and western blotting was performed to detect PARP cleavage. h The gray analysis of cleaved PARP. Data are presented as mean ± SD or standard error of 3 independent experiments, *p < 0.05, **p < 0.01

Full size image

To investigate whether the autophagy triggered by TEOA is pro-survival or pro-death, SW620 cells were treated with either TEOA or co-treated with TEOA and 3-MA (5 μM) for 24, 48, and 72 h. As shown in Fig. 3e, after pre-treatment with 3-MA for 2 h, cell viability was not affected, as compared with TEOA treated cells alone. As shown in Fig. 3f, pre-treatment with 3MA in TEOA-treated SW620 cells did not significantly increase the proportion of apoptotic cells. In addition, the expression levels of PARP were unchanged (Fig. 3g). Collectively, these data suggest that TEOA induced autophagic cell death not protective autophagy in SW620 cells.

ER stress mediates TEOA-induced autophagy in SW620 cells

We next investigated the relationship between autophagy and ER stress in TEOA-treated SW620 cells through exploring the expression of ER stress associated proteins. GRP78, a chaperone that binds unfolded proteins, is an important ER stress marker. As shown in Fig. 4a, the expressions of GRP78, p-PERK, p-elF2α and CHOP were increased in a dose-dependent manner, suggesting that ER stress may play a crucial role in TEOA-induced autophagy in SW620 cells. To further investigate the relationship between autophagy and ER stress, an autophagy inhibitor, 3-MA was introduced. The addition of 3-MA altered the levels of p-PERK, p-elF2α and CHOP (Fig. 4a) implicating the involvement of PERK/eIF2α/CHOP pathway in TEOA-induced autophagy.

Fig. 4
figure 4

TEOA induces autophagy by increasing ER stress. SW620 cells were treated with 0–50 μM TEOA and/or 3-MA for 48 h. a Western blot analysis examined the protein expression level of GRP78, p-PERK, PERK, p-elF2α and CHOP, and β-actin was applied for equal loading. Representative blots were shown. b–g The gray analysis of GRP78, CHOP, p-PERK, PERK, p-elF2α and elF2α. Values are expressed as mean ± SD, *p < 0.05, **p < 0.01 versus control

Full size image

TEOA induces mitophagy through ROS-mediated PINK1/Parkin pathway in SW620 cells

Damaged mitochondria are detrimental to cellular homeostasis. One mechanism for removal of damaged mitochondria involves the PINK1-Parkin pathway, which poly-ubiquitinates damaged mitochondria to promote mitophagy (Durcan and Fon 2015). As shown in Fig. 5b, TEOA increased the expression of PINK1 in a dose-dependent manner with concomitant increases in the abundance of Parkin, and led to the recruitment of ubiquitin. It has been reported that p62/SQSTM1 serves as an adaptor for the ubiquitinated mitochondria to be recognized by the autophagic machinery and recruitment of p62 to mitochondria is mediated by its ability to bind to ubiquitin (Youle and Narendra 2011). Western blot analysis revealed that TEOA increased the abundance of p62 in a dose-dependent manner (Fig. 5b).

Fig. 5
figure 5

Reactive oxygen species are involved in TEOA-induced mitophagy. a The ability of TEOA to induce ROS production in SW620 cells. H2O2 (100 μM) was used as positive control and NAC as negative control. b SW620 cells were pre-treated with NAC (5 mM) and subsequently treated with TEOA for 48 h; Levels of Parkin, PINK1, p62 and ubiquitin were detected by western blotting. c–f The gray analysis of Parkin, p62, PINK1 and ubiquitin. Values are expressed as mean ± SD, *p < 0.05, **p < 0.01 versus control

Full size image

To address whether ROS plays a role in the activation of mitophagy, we analyzed the expression of mitophagy-related proteins. Exposure of cells to TEOA resulted in a significant increase in intracellular ROS production, and the recruitment of ROS was inhibited by pretreatment with the ROS scavenger NAC (5 mM) (Fig. 4a). Additionally, NAC blocked TEOA-mediated upregulation of PINK1, Parkin and p62 (Fig. 5b), highlighting the role of ROS in TEOA-induced mitophagy.

TEOA retards the growth of SW620 xenograft tumors in BALB/C nude mice

We next evaluated the in vivo effect of TEOA on SW620 xenograft tumor growth. As shown in Fig. 6a, both the tumor volumes and weights were significantly decreased by TEOA treatment (p < 0.05) as compared with the control group. After three weeks treatment, the tumor weights of control group and TEOA-treated groups (40, 80 and 120 mg/kg) were 0.95 ± 0.15, 0.60 ± 0.14, 0.49 ± 0.13 and 0.36 ± 0.09 g, respectively, and the tumor weight of the cyclophosphamide (CTX, the most commonly prescribed alkylating agent in clinical medicine) treated group was 0.33 ± 0.08 g (Fig. 6a). In addition, the weights of liver and spleen were unaffected by TEOA treatment compared with the control group and CTX-treated group (Fig. 6b, p < 0.05). In conclusion, these results suggest that intraperitoneally administered TEOA can significantly suppress SW620 tumor growth with low toxicity in xenograft mice.

Fig. 6
figure 6

TEOA inhibits SW620 tumor in vivo. TEOA treatment significantly inhibited tumor growth compared to the control group (*p < 0.05). a Body weight. b Tumors. c The tumor weight, liver weight and spleen weight of each group (*p < 0.05)

Full size image

Discussion

There has been a long history of using Chinese herbs as anti-cancer agents. It has been demonstrated that the application of natural products and their derivatives provides successful anti-cancer remedies. As agents isolated from natural sources, particularly plants, may have reduced toxicity, we sought to identify novel and effective anti-cancer agents from these sources. For decades, Chinese folk medicine has used the roots of A. eriantha for treatment of cancers, which have been shown to have anti-tumor and immunopotentiating activities (Lin et al. 1987). In our present study, we successfully purified TEOA from the roots of A. eriantha, and demonstrated that TEOA-mediated ER stress triggered autophagy and mitophagy in colon cancer.

Emerging evidence has indicated that ER stress is a potent inducer of autophagy (Høyer-Hansen and Jäättelä 2007; Kouroku et al. 2007). The ability of cells to respond to perturbations in ER functioning, includes increasing protein synthesis and protein misfolding, leading to the development of ER stress (Tabas and Ron 2011). One of the best characterized ER chaperone proteins is the glucose regulated protein (GRP78). GRP78 is an ER resident protein and its synthesis can be stimulated by a variety of environmental and physiological stress conditions that perturb ER function and homostasis (Lee 1987, 2001). Increasing evidence has indicated that autophagy can be activated by eukaryotic translation initiation factor 2 (eIF2α)-phosphorylation during starvation and viral infection (Tallóczy et al. 2002). There are four elF2α kinases (PKR, GCN2, HRI and PERK) that can be activated by viral infection, amino acid starvation, heme depletion and ER stress, respectively (Garcia et al. 2007). Recently, Kouroku et al. reported that expanded polyglutamine-induced ER stress activates autophagosome formation and LC3 conversion from LC3-I to LC3-II via the PERK-eIF2α pathway (Kouroku et al. 2007). In non-stressed cells, GRP78 binds to PERK, leading to an inactive state, however, upon ER stress, PERK is released from GRP78 (Bertolotti et al. 2000). PERK is the major protein responsible for attenuation of mRNA translation under ER stress, preventing influx of newly synthesized proteins into the already stressed ER compartment. This translational attenuation is mediated by phosphorylation of eIF2α (Cullinan and Diehl 2004). Phosphorylation of eIF2α is cytoprotective during ER stress, as cells are sensitized when this pathway is genetically ablated (Harding et al. 2000a, b; Scheuner et al. 2001) and protected when it is ectopically enforced (Jousse et al. 2003; Lu et al. 2004). PERK induced rapid and robust eIF2α phosphorylation, resulting in down-regulation of cyclin D1 (Brewer and Diehl 2000) and up-regulation of GADD34 and CHOP/GADD153 (Harding et al. 2000a, b), two proteins whose expression is induced by eIF2α phosphorylation. Our study demonstrated that TEOA induced PERK/eIF2α activation in SW620 cells (Fig. 7). Since autophagy and ER stress have been implicated in various human cancers, exploration of novel signaling pathways related with ER stress and autophagy may provide new therapeutic strategies for diseases.

Fig. 7
figure 7

A proposed model of molecular interaction to delineate the mechanisms of TEOA’s action in SW620 cells

Full size image

PINK1 and Parkin have been implicated in the emerging pathway of selective autophagy for clearance of damaged mitochondria. PINK1 has already been suggested to control mitochondria translocation of Parkin from the cytosol (Lee 1987). The exact mechanism by which PINK1 senses depolarization is unclear, but increasing evidence suggests that full-length PINK1 accumulates in depolarized mitochondria, perhaps as a result of increased protein stability/decreased processing. PINK1 is a putative serine/threonine kinase, thus phosphorylation of Parkin by PINK1 via a kinase cascade may be the means of recruitment of Parkin to mitochondria (Gegg and Schapira 2011). More recently, many groups have dissected key mechanistic steps of this PINK1/Parkin dependent mitophagy and identified PINK1 as an essential upstream initiator of Parkin-directed mitophagy (Matsuda et al. 2010). When mitochondria become damaged and mitochondrial membrane potential declines, PINK1 rapidly accumulates on the outer membrane of mitochondria and subsequently facilitates the mitochondrial translocation of Parkin to modulate the process of mitophagy (Matsuda et al. 2010) and eventually causes mitochondrial mass decrease (Koh and Chung 2012).

The induction of mitophagy following Parkin recruitment is thought to involve the Parkin-mediated ubiquitination of mitochondrial substrates (Matsuda et al. 2010). Indeed, the E3 ubiquitin ligase activity of Parkin is increased following its translocation to the mitochondria. Furthermore, the ubiquitin-binding adaptor p62 (also known as sequestosome 1), which can both aggregate ubiquitinated proteins by polymerizing with other p62 molecules and recruit ubiquitinated cargo into autophagosomes by binding to LC3, accumulating in mitochondria following Parkin-mediated ubiquitylation (Ding et al. 2010; Geisler et al. 2010; Lee et al. 2010; Okatsu et al. 2010). The p62 protein is not only responsible for clumping of mitochondria through binding to Parkin-ubiquitylated mitochondrial substrates, but also essential for Parkin to promote mitophagy (Geisler et al. 2010; Lee et al. 2010). We determined that induction of mitophagy following TEOA treatment was illustrated by the increase expression of PINK1, Parkin, p62 and ubiquitin. PINK1 levels were significantly elevated after TEOA treatment, reflecting that mitopahgy progression is mediated through changes in mitochondrial proteins, such as PINK1 and Parkin (Narendra et al. 2010). It has been reported that ROS play a pivotal role in autophagy and mitochondrial dysfunction and may be an important upstream activator of mitophagy (Wang et al. 2012). To address the potential role of ROS in the regulation of TEOA-induced mitophagy, we investigated the relationships among TEOA-induced ROS, mitophagy and the PINK1/Parkin pathway. Consistent with the previous report that ROS regulates the localization of Parkin during oxidative stress in neurons (Joselin et al. 2012), we showed that pre-treatment with the ROS scavenger NAC blocked the TEOA-induced PINK1, Parkin, p62 and ubiquitin expression level changes. These data suggest that mid oxidative stress triggers mitophagy in mammalian cells, and are in agreement with reports that mitochondrial ROS appear to be a common mediator for mitophagy and a key signal to regulate the clearance of abnormal mitochondria in a PINK1-deficient cell line (Dagda et al. 2009). Overall, our findings revealed that TEOA-induced mitophagy via the PINK1/Parkin/ubiquitin/p62/LC3-II pathway and induction of ROS is an upstream activator of mitophagy (Fig. 7).

Meanwhile, we demonstrated that TEOA had preferential activity against the growth of SW620 xenografts without obvious toxicity. The precise mechanism requires further investigation.