The red wine component ellagic acid induces autophagy and exhibits anti‐lung cancer activity in vitro and in vivo

Abstract Red wine consists of a large amount of compounds such as resveratrol, which exhibits chemopreventive and therapeutic effects against several types of cancers by targeting cancer driver molecules. In this study, we tested the anti‐lung cancer activity of 11 red wine components and reported that a natural polyphenol compound ellagic acid (EA) inhibited lung cancer cell proliferation at an efficacy approximately equal to that of resveratrol. EA markedly increased the expression of the autophagosomal marker LC3‐II as well as inactivation of the mechanistic target of rapamycin signalling pathway. EA elevated autophagy‐associated cell death by down‐regulating the expression of cancerous inhibitor of protein phosphatase 2A (CIP2A), and CIP2A overexpression attenuated EA‐induced autophagy of lung cancer cells. Treating tumour‐bearing mice with EA resulted in significant inhibition of tumour growth with suppression of CIP2A levels and increased autophagy. In addition, EA potentiated the inhibitory effects of the natural compound celastrol on lung cancer cells in vitro and in vivo by enhancing autophagy and down‐regulating CIP2A. These findings indicate that EA may be a promising chemotherapeutic agent for lung cancer, and that the combination of EA and celastrol may have applicability for the treatment of this disease.

wine compounds also have anti-cancer activity. 8 Ellagic acid (EA), a natural polyphenol compound found in grapes, strawberries, and nuts, 9 has higher concentration in red wine than resveratrol and reaches 15 mg/L in some Spain red wine. 10 Recent studies show that EA has antioxidant and preventive effects in several types of cancers. [11][12][13][14] EA exhibits in vitro and in vivo anticancer activities by arresting the cell cycle, inhibiting tumour cell proliferation, inducing cell apoptosis, ameliorating inflammation, and inhibiting angiogenesis. [15][16][17][18] EA suppresses Akt, Shh, and Notch pathways 19,20 and down-regulates the levels of 17β-estradiol-metabolizing enzymes. 21 However, the effects and underlying molecular mechanisms of EA in lung cancer remain to be investigated.
Cancerous inhibitor of protein phosphatase 2A (CIP2A) or KIAA1524/ p90 22,23 is an oncoprotein which promotes tumourigenesis through facilitation of the biological functions of cancer drivers (c-Myc, Akt, etc.) and inhibition of PP2A activity. 24 CIP2A controls autophagy and cell growth through mTORC1 activation, 25 synergizes with RAS, 26,27 and crosstalks with Wnt/β-catenin signal pathway 28 to promote cell proliferation and cancer progression. 23,29 It also mediates interleukin 10 (IL-10)-induced tumour aggressiveness. 30 CIP2A is overexpressed at a high frequency in most types of human cancers, inversely correlated with disease outcome of the patients, 24 and is associated with doxorubicin resistance. 31 CIP2A is recognized as a "druggable" target, and compounds including celastrol, 32 ethoxysanguinarine, 33 and rabdocoetsin B 29 inhibit CIP2A and induce apoptosis of lung cancer cells. CIP2A is targeted by some kinase inhibitors [34][35][36] and proteasome inhibitor 37 to exert anti-cancer activity.
Therefore, identifying novel CIP2A inhibitors may provide optimal drug candidates for clinical testing.
In this study, we tested the effects of 11 red wine polyphenolic compounds on lung cancer cells and found that EA had the strongest anticancer effects that significantly suppressed tumour growth in vivo. Moreover, the mechanistic studies showed that EA induces autophagy by targeting CIP2A in lung cancer cells.

| Immunofluorescence analysis
Cells (1.5 × 10 5 per well) were seeded on coverslips with 1% gelatin in six-well culture plates and grown to 60% confluence. After treatment, cells were washed three times with PBS and fixed in 4% paraformaldehyde for 15 minutes at room temperature.

| Enzyme-linked immunosorbent assay
Mouse blood samples were obtained before cervical dislocation.
Serum concentration of creatine (Cr), aspartate transaminase (ALT), and alanine transaminase (AST) was determined using a commercially available enzyme-linked immunosorbent assay kit (Shanghai Gene-Pharma, Shanghai, China). The absorbance was read at 450 nm using an automated Microplate Reader (Bio-Tek, Winooski, VT, USA).

| Animals
The animal studies were approved by the Institutional Review Board of the Institute of Zoology, Chinese Academy of Sciences.
Female BALB/C nude mice (5-6 weeks old) were purchased from Vital River Laboratory (Beijing, China) and treated in accordance with the guidelines of the Animal Use and Care Committee of the Institute of Zoology, Chinese Academy of Sciences. HOP62 cells (1 × 10 6 ) were injected subcutaneously into the right rear flank of each mouse. After 4 days, the mice were randomized into five groups and treated with EA (0, 40, 80 mg/kg) via oral gavage or celastrol (1 mg/kg) via intraperitoneal injection in corn oil once every 2 days for 22 days. Tumour size was measured every alternate day with an electronic caliper and calculated with the following formula: volume (mm 3 ) = 1/2 (width) 2 × length. After 22 days of treatment, the mice were always killed by cervical dislocation.
Tumour tissues were excised, homogenized, and lysed for Western blot analysis.

| Statistical analysis
All experiments were repeated at least three times and the data are presented as the mean ± SD. Differences between data groups were evaluated for significance using the Student's t test of unpaired data (two-tailed). For animal studies, the data are presented as the mean ± SEM. The tumour volume was analysed with one-way ANOVA and then Tukey's test after ANOVA to identify the pairs with significant differences using the software SPSS 16.0 for Windows (Chicago, IL, USA). P < 0.05 was considered statistically significant. Interestingly, EA and resveratrol showed significant inhibition rate in the cells (Figure 1).  The cells were then lysed and the lysates were subjected to Western blot to assess the changes in the cytosolic form (LC3-I) and lipidated form (LC3-II) of the LC3 autophagic activity marker. 38 Notably, EA induced accumulation of LC3-II in HOP62 and H1975 cells in a dose-dependent manner ( Figure 3A). To confirm the presence of LC3-II accumulation in the autophagic process, the formation of green fluorescent protein (GFP)-tagged LC3-positive autophagosome was examined. We found that EA induced a significant increase in GFP-LC3 puncta in the nucleus and mitochondria ( Figure 3B). In addition, EA treatment reduced protein levels of p62/SQSTM1, a well-established receptor for autophagic cargo and substrate for autophagic degradation, and increased the levels of ATG5, an essential molecule for the induction of autophagy ( Figure 3C). The results of MTT assays showed that the cell viability inhibition of EA was attenuated by autophagy inhibitor 3-MA in both HOP62 and H1975 cells ( Figure 3D). Moreover, lysosomal acidification inhibitor Bafilomycin A1 promoted LC3-II accumulation in EA-treated lung cancer cells ( Figure 3E). In addition, we showed that knockdown of ATG5 by siRNA blocked the formation of LC3-II in EA-treated HOP62 cells ( Figure 3F). These results indicated that EA induced autophagy and enhancement of the autophagic flux.

| Inhibitory effects of EA on lung cancer cells
The mTOR is an Akt signalling protein and a critical regulator of cellular metabolism, growth, and proliferation, with p70 ribosomal protein S6 kinase (p70S6K) functioning as an important effector. 39 Recent studies have reported that the Akt-mTOR pathway plays a key role in autophagy process. 40 We showed that EA down-regulated phosphorylated Akt (p-Akt) and inhibited mTOR phosphorylation as well as p70S6K in both HOP62 and H1975 cells ( Figure 3G).

| EA induces autophagy through downregulation of CIP2A
To examine how autophagy is induced by EA, the expression of CIP2A was detected by Western blot. We found that CIP2A in lung  Figure 4C). The results were confirmed by immunofluorescence staining with an anti-LC3 antibody in CIP2Aexpressing cells coincubated with or without EA ( Figure 4D).  Collectively, these results suggest that EA induced autophagy to inhibit lung cancer cell proliferation by down-regulating CIP2A.
We examined the mechanism underlying CIP2A down-regulation in cells treated with EA. The proteasome inhibitor PS341 (100 nM), lysosomal protease inhibitor chloroquine (80 μM), and protein synthesis inhibitor cycloheximide (CHX, 50 μg/mL) were used to treat cells in the absence or presence of EA. The results showed that these inhibitors failed to prevent CIP2A down-regulation ( Figure 4E).
Instead, qPCR showed that treatment with EA decreased CIP2A mRNA in HOP62 cells, indicating that CIP2A down-regulation triggered by EA is regulated at the transcriptional level ( Figure 4F).

| In vivo anti-lung cancer activity of EA
To evaluate the anti-lung cancer activity of EA and examine whether CIP2A is important for autophagy induction in vivo, nude mice were injected subcutaneously with HOP62 cells and treated with EA. The results demonstrated that EA significantly suppressed tumour growth, as reflected by a decrease in tumour volume ( Figure 5A).
The tumours grew more slowly in EA-treated mice compared to control mice, and tumour size dramatically decreased in a dose-dependent manner by EA ( Figure 5A and B). In addition, EA treatment did not lead to a reduction in body weight ( Figure 5C). Mice treated with EA had normal serum concentrations of ALT, Cr, and AST compared to control mice ( Figure 5D), indicating that EA treatment did not lead to liver or kidney toxicity. Moreover, Western blot analysis revealed that EA-treated mice showed a marked decrease in CIP2A levels and an increase in LC3 levels ( Figure 5E). Thus, EA treatment induced autophagy and down-regulation of CIP2A. We evaluated the therapeutic potential of combined use of EA and celastrol in mice-bearing HOP62 cells. To this end, nude mice were inoculated with HOP62 cells and subjected to EA (40 mg/kg), celastrol (1 mg/kg), or EA combined with celastrol treatment for 22 days. We found that combined treatment resulted in significant inhibitory effects on tumour growth compared with either monotherapy ( Figure 6D and E), without a reduction in body weight ( Figure 6F), or increase in serum concentrations of Cr, ALT, and AST ( Figure 6G). Moreover, combined use of EA and celastrol up-regulated LC3-II and down-regulated CIP2A compared to each monotherapy ( Figure 6H). These data indicate that the combination of EA and celastrol exerts synergistic anti-lung cancer effects both in vitro and in vivo.

| DISCUSSION
Polyphenols are believed to mediate many of the health benefits of moderate wine consumption. Some of these polyphenols have inhibitory effects and chemoprevention potentials in several types of cancers. 41 Here, we examined the potential anti-lung cancer effects of 11 red wine polyphenols and found that EA exhibited almost the same effect as resveratrol in lung cancer (Figure 1). EA was capable of inhibiting lung cancer cell proliferation and inducing autophagy, which was regulated by CIP2A oncoprotein. Previous studies showed that EA did not affect the viability of normal cells. 16,42 We showed that EA inhibited tumour growth in nude mice-bearing HOP62 cells without obvious side effects, as determined by the measurements of body weight as well as serum concentrations of Cr, ALT, and AST ( Figure 5). As EA monotreatment with a lower dosage only modestly inhibited tumour growth, we hypothesized that combining EA with other known anticancer agents may increase its therapeutic benefits.  The expression of CIP2A was tested by Western blot assay. Numbers under the bands are the relative expression values to Actin determined by densitometry analysis. (C, D) HOP62 cells were transfected with CIP2A and treated with EA for 24 hours; the cells were subjected to Western blot assays (C) and immunofluorescence analysis (D). Data are presented as the mean ± SD; P-values were determined with the t test. (E) HOP62 cells were treated indicated compounds for 24 hours, lysed, and the lysates were subjected to Western blot to test the expression of CIP2A. Numbers under the bands (LC3-II for LC3) are the relative expression values to Actin determined by densitometry analysis. (F) HOP62 cells were treated with EA for 24 hours, harvested, and the expression of CIP2A was detected qPCR. *P < 0.05; **P < 0.01; ***P < 0.001 The role of autophagy in cancer is diverse. Autophagy and autophagy-related processes promote different effects even in a single tumour, depending on the stage of disease progression, cell type, oncogenic drivers, and intensity of the activating signal. 43 The tumour inhibition role of autophagy in certain human cancers, including nonsmall cell lung cancer, is increasingly being reported. 44,45 Therefore, it is important to distinguish between the cytoprotective and cytotoxic effects of autophagy in tumourigenesis. In this study, we reported that EA induced autophagy in lung cancer cells (Figure 3), with decreases in p-Akt, p-mTOR, and p-P70S6K ( Figure 3E). CIP2A is an oncoprotein which is critical in controlling autophagy. 25 We found that CIP2A might mediate EAinduced autophagy in lung cancer cells because overexpression of CIP2A attenuated EA-induced autophagy (Figure 4). In accordance with this finding, EA dramatically suppressed tumour growth in a nude mouse model of lung cancer by activating autophagy and down-regulating CIP2A levels ( Figure 5). However, the decrease in CIP2A expression by EA treatment was not mediated by proteasomal degradation because proteasome inhibitor did not inhibit EAinduced CIP2A down-regulation ( Figure 4). EA down-regulated CIP2A at mRNA level (Figure 4), similar to another natural compound rabdocoetsin B which down-regulates CIP2A at mRNA level. 29 These results indicate that EA is a CIP2A inhibitor and an autophagy inducer in lung cancer.
Combination therapy has been advocated for more than 2500 years by prescriptions called formulae 46

in Traditional Chinese
Medicine to treat diseases. These formulae usually consist of several types of medicinal herbs to exert synergic efficacies with minimized side effects. 47 We showed that combination of EA and celastrol at

CONF LICT OF I NTEREST
No potential conflicts of interest were disclosed.