Akbu-LAAO exhibits potent anti-tumor activity to HepG2 cells partially through produced H2O2 via TGF-β signal pathway

Previously, we characterized the biological properties of Akbu-LAAO, a novel L-amino acid oxidase from Agkistrodon blomhoffii ussurensis snake venom (SV). Current work investigated its in vitro anti-tumor activity and underlying mechanism on HepG2 cells. Akbu-LAAO inhibited HepG2 growth time and dose-dependently with an IC50 of ~38.82 μg/mL. It could induce the apoptosis of HepG2 cells. Akbu-LAAO exhibited cytotoxicity by inhibiting growth and inducing apoptosis of HepG2 as it showed no effect on its cell cycle. The inhibition of Akbu-LAAO to HepG2 growth partially relied on enzymatic-released H2O2 as catalase only partially antagonized this effect. cDNA microarray results indicated TGF-β signaling pathway was linked to the cytotoxicity of Akbu-LAAO on HepG2. TGF-β pathway related molecules CYR61, p53, GDF15, TOB1, BTG2, BMP2, BMP6, SMAD9, JUN, JUNB, LOX, CCND1, CDK6, GADD45A, CDKN1A were deregulated in HepG2 following Akbu-LAAO stimulation. The presence of catalase only slightly restored the mRNA changes induced by Akbu-LAAO for differentially expressed genes. Meanwhile, LDN-193189, a TGF-β pathway inhibitor reduced Akbu-LAAO cytotoxicity on HepG2. Collectively, we reported, for the first time, SV-LAAO showed anti-tumor cell activity via TGF-β pathway. It provides new insight of SV-LAAO exhibiting anti-tumor effect via a novel signaling pathway.

of Akbu-LAAO or exogenous H 2 O 2 for 24 h. Typical morphological characteristics of apoptosis such as nuclear size reduction, cell pyknosis and chromatin condensation were more easily observed in Akbu-LAAO-treated HepG2 than control HepG2 cells (Indicated by arrow, Fig. 5). Hoechst assay showed 0.1 and 0.2 mg/mL of catalase could not induce observable HepG2 apoptosis, which ensures no influence of catalase on the flow cytometry assay. Akbu-LAAO induced the in vitro apoptosis of HepG2 cell in a dose-dependent manner. The apoptotic rates of HepG2 cells flowing Akbu-LAAO administration with the dosages of 0, 20, 38.82 and 60 μ g/mL for 24 h were measured as ~3.54%, 7.61%, 10.85% and 23.36% (Fig. 6), respectively. The apoptotic rates of HepG2 cells following the treatments of 38.82 μ g/mL Akbu-LAAO + 0.1 mg/mL catalase and 38.82 μ g/mL Akbu-LAAO + 0.2 mg/ mL catalase for 24 h were ~6.19% and 5.59% (Fig. 6). However, the apoptotic rate of HepG2 cells only decreased ~42.95% following the treatment of 38.82 μ g/mL Akbu-LAAO + 0.1 mg/mL catalase compared to the HepG2 cells treated with 38.82 μ g/mL Akbu-LAAO, which was still 74.86% higher than that of the control HepG2 cells without Akbu-LAAO treatment (Fig. 6). Catalase scavenging only partially reduces Akbu-LAAO-inducible apoptosis of HepG2 cells. In sharp contrast, the apoptotic rates of HepG2 in the presences of 0, 0.1, 0.21, 0.4 mM of H 2 O 2 , 0.21 mM H 2 O 2 + 0.1 mg/mL catalase and 0.21 mM H 2 O 2 + 0.2 mg/mL catalase for 24 h were ~2.26%, 4.36%, 7.14%, 7.55%, 2.23% and 2.34%, respectively. Interestingly, no differences were measured for the apoptosis between HepG2 cells without H 2 O 2 treatment and with 0.21 mM H 2 O 2 + 0.2 mg/mL catalase treatment (Fig. 6). In another word, catalase scavenging completely restores H 2 O 2 -inducible apoptosis on HepG2 cells. Conclusively, all the above results indicated the cytotoxicity and apoptosis induction of Akbu-LAAO on HepG2 cells are linked to but not solely contributed to the produced H 2 O 2 .
Akbu-LAAO does not affect the cell cycle of HepG2. The effect of Akbu-LAAO on HepG2 cell cycle was measured by flow cytometry assay. As shown in Fig. 7, there were no distribution differences of cell populations Akbu-LAAO acts on HepG2 cells via TGF-β pathway. We screened the differentially expressed genes in HepG2 cells in responding to Akbu-LAAO treatment using Affymetrix Genechip (Human Transcriptome Array 2.0). A total of 254 mRNAs were identified as up-or down-regulated over 1.5-fold in Akbu-LAAO-treated HepG2 cells in comparison with control HepG2 cells. Among these targeting genes, we focused on the molecules involved in TGF-β signal pathway. Gene-gene interaction network analysis indicated the genes CYR61, p53, GDF15, TOB1, BTG2, BMP2, BMP6, SMAD9, JUN, JUNB, LOX, CCND1, CDK6, GADD45A and CDKN1A in TGF-β signal pathway were apparently differentially expressed following Akbu-LAAO treatment (Table 1), which implicates Akbu-LAAO might exert anti-tumor activity to HepG2 cells via TGF-β pathway.
The expression level changes of above 15 genes in HepG2 cells in responding to Akbu-LAAO were further validated by qRT-PCR analysis. In consistent with the microarray results, qRT-PCR data revealed the same trend for the level changes of these genes in HepG2 cells following Akbu-LAAO treatment (Fig. 8, Table 1). We also checked the effect of catalase on the expression levels of these genes. 0.2 mg/mL catalase could only slightly restore the level changes of mRNAs in HepG2 induced by Akbu-LAAO (Fig. 8, Table 1), which again implicated the presence of other action factor for endowing Akbu-LAAO the cytotoxicity to HepG2 except for H 2 O 2 production. It is necessary to confirm Akbu-LAAO might exert anti-tumor activity to HepG2 through TGF-β pathway. The treatment of LDN-193189, a TGF-β pathway inhibitor, could decrease the cytotoxicity of Akbu -LAAO (38.82 μ g/mL) to HepG2 cells by ~52% in 24 h by MTT assay (Fig. 9A). In addition, morphological changes of HepG2 induced by Akbu-LAAO could be restored to certain extent in the presence of LDN-193189 (Fig. 9B). Taken together, current work concluded Akbu-LAAO exhibits potent anti-tumor activity to HepG2 cells partially through produced H 2 O 2 and via TGF-β signal pathway.
Previously, we purified an LAAO (Akbu-LAAO) from Agkistrodon blomhoffii ussurensis snake venom and characterized its biological properties 16 . Current work demonstrated that Akbu-LAAO inhibited the in vitro proliferation ( Fig. 1) and induced the apoptosis of HepG2 cells (Figs 5 and 6). It showed cytotoxicity toward HepG2 cells with an IC 50 of 38.82 μ g/mL (~0.3 μ M, Fig. 1) that was ~700 folds lower than that of exogenous H 2 O 2 (~0.21 mM, Fig. 2C). H 2 O 2 is commonly believed to play an important role in the anti-tumor activities of SV-LAAOs. LAAOs could bind directly to cell surface. The enzymatic-released H 2 O 2 accumulates at the localized area to a relative higher concentration to trigger cell apoptosis 18,23 . The anti-tumor activities of SV-LAAO were reported to be inhibited by catalase and other H 2 O 2 scavengers 8,9,18 . Current work indicated that the anti-tumor activity of Akbu-LAAO toward HepG2 differed from exogenous H 2 O 2 . It inhibited the in vitro proliferation of HepG2 partially through enzymatic-released H 2 O 2 ( Fig. 2B) compared with exogenous H 2 O 2 (Fig. 2D). So, except for H 2 O 2 action mechanism, Akbu-LAAO might act on HepG2 through unknown path.
We also found the difference of HepG2 apoptosis induced by Akbu-LAAO and exogenous H 2 O 2 by Hoechst staining and flow cytometry assays (Figs 5 and 6). Comparing to the Akbu-LAAO-treated HepG2 cells (Fig. 5A), the addition of catalase significantly inhibited the apoptosis of HepG2 cells induced by exogenous H 2 O 2 (Fig. 5B). 0.1 mg/mL catalase decreased the apoptotic rate of HepG2 cells treated with 38.82 μ g/mL Akbu-LAAO by ~42.95% that was 74.86% higher than the cells receiving no Akbu-LAAO treatment (Fig. 6). In the sharp contrast, in the presence of 0.1 mg/mL catalase, the apoptosis rate of HepG2 receiving 0.21 mM H 2 O 2 stimulation was 2.23% comparable to 2.26% of HepG2 cells without H 2 O 2 stimulation (Fig. 6). The cytotoxicity of Akbu-LAAO to HepG2 cells is partially linked to enzymatic-released H 2 O 2 . Akbu-LAAO shows no effect on the cell cycle of HepG2. B. atrox LAAO was reported to arrest HL-60 cells at G 0 / G 1 phase by delaying its progression to S and G2/M phases 19 . A. acutus LAAO (ACTX-6) could markedly increase cell accumulation at sub-G1 phase 24 . Interestingly, our results showed Akbu-LAAO did not affect the cell cycle of HepG2 cells (Fig. 7). Akbu-LAAO exhibits anti-tumor activity on HepG2 cells by inhibiting cell proliferation and inducing cell apoptosis without disturbing cell cycle.
TGF-β pathway was associated with tumor cell proliferation, differentiation, apoptosis, tumor occurrence and development 25 . CYR61, p53, TOB1/BTG2 and CCND1/CDK6 are critical molecules in TGF-β signaling pathway. Based on the validated targeting genes differentially expressed in HepG2 cells following Akbu-LAAO treatment and summarized results from published literatures, we propose Akbu-LAAO acts on HepG2 mainly via the following detailed pathways.
(1) CYR61-p53-CDKN1A-CCND1/CDK6 pathway: As a tissue growth factor, the activation of CYR61 promoted tumor cell proliferation and inhibited apoptosis by inhibiting p53 expression 26,27 . In case of cell damage, the increased expression of p53 induced CDKN1A upregulation 28 . CCCND1 could form complex with CDK4 and CDK6 whose overexpression promoted cell cycle progression and cancer development 29,30 . The tumor suppression activity of CCCND1/CDK6 was negatively correlated with CDKN1A level 29,30 . Current work showed Akbu-LAAO treatment decreased CYR61 level, consequently might enhance mRNA levels of p53 and CDKN1A, finally suppressed the level of CCCND1/CDK6 (Table 1 and Fig. 10) to inhibit HepG2 proliferation and induce HepG2 apoptosis, which suggests its action to HepG2 cells via CYR61-p53-CDKN1A-CCND1/ CDK6 mechanism.

Conclusions
Akbu-LAAO significantly inhibits the in vitro proliferation and induces the apoptosis of HepG2 cells without interrupting its cell cycle. Akbu-LAAO administration can induce morphology and ultrastructure changes of HepG2 cells. The cytotoxicity of Akbu-LAAO towards HepG2 cells is partially linked to enzymatic-produced H 2 O 2 . Gene microarray, qRT-PCR and TGF-β pathway activity blocking assays prove that Akbu-LAAO exerts tumor suppression effect on HepG2 cells via TGF-β pathway. The current study suggests Akbu-LAAO as a potential anti-tumor drug and provides new clues to anti-tumor action mechanism for SV-LAAOs.

Methods
Materials. RPMI 1640 and pancreatin were from Gibco (USA). Fetal bovine serum (FBS) was from TransGen (China). Trizol TM reagent was from Life (USA). PrimeScript TM RT reagent kit with gDNA eraser was from TaKaRa (Japan). FastStart universal SYBR green master and BrdU assay ELISA kit were from Roche (Switzerland). Annexin    The influences of catalase or LDN-193189 (TGF-β signaling pathway inhibitor) on the cytotoxicities of Akbu-LAAO or exogenous H 2 O 2 to HepG2 was measured using MTT assay. HepG2 cells were pre-incubated with 0, 0.1, 0.2 mg/mL catalase or 10 μ M LDN-193189 and treated with 38.82 μ g/mL Akbu-LAAO or 0.21 mM H 2 O 2 for 24 h at 37 °C with 5% CO 2 . The rest steps were the same as described above.
Cell proliferation by BrdU assay. The influence of Akbu-LAAO or exogenous H 2 O 2 on the proliferation of HepG2 was determined by BrdU assay. Being treated with 0, 5, 10, 20, 40, 80 μ g/mL of Akbu-LAAO for 24 h at 37 °C with 5% CO 2 , HepG2 cells were incubated with 20 μ L BrdU labeling solution per well for 4 h, fixed and incubated with anti-BrdU mAb according to the manufacturer's instruction. Finally, the absorbance at 450 nm was measured using a microplate reader. Results were the averages from triplicate measurements.
Cell morphology assay. The effect of Akbu-LAAO or exogenous H 2 O 2 on HepG2 morphology was measured by an inverted light microscope. 7 × 10 3 cells/well were seeded into a 96-well plate and incubated at 37 °C, 5% CO 2 overnight. Then, the cells were administrated with 20, 38.82, 60 μ g/mL Akbu-LAAO or 0.1, 0.21, 0.4 mM exogenous H 2 O 2 in the presence or absence of catalase for 24 h. Cell morphology images were taken at the magnification of 100×.
Cell ultrastructure assay. Transmission electron microscopy (TEM) was used to characterize the ultrastructure alteration of HepG2 cells caused by Akbu-LAAO stimulation. Being incubated in 20, 38.82, 60 μ g/mL Akbu-LAAO for 24 h, the HepG2 cells were collected, washed with pre-cooled PBS, and fixed in 2.5% (v/v) glutaraldehyde at 4 °C for 24 h. The slices were stained with 2% osmic acid for 2 h at RT, washed with PBS twice, dehydrated in a graded series of ethanol 50%, 70%, 80%, 90%, 100%, 100%, substituted with propylene and embedded in epoxy resin for 6 h. Ultrathin sections (60-90 nm) were observed and photographed on a JEOL JEM-1200EX system (Japan) operated at 100 KV. Images were digitally acquired from 5 fields randomly selected for each condition.

Flow cytometry assay. Flow cytometry assay was performed to investigate the apoptosis induction of
Akbu-LAAO to HepG2 cells. 1 × 10 6 cells were seeded into a 6-cm dish and incubated at 37 °C with 5% CO 2 overnight. Following the treatments of 20, 38.82, 60 μ g/mL Akbu-LAAO or 0.1, 0.21, 0.4 mM exogenous H 2 O 2 with or without the presence of catalase at 37 °C, 5% CO 2 for 24 h, the corresponding HepG2 cells from each group were harvested with trypsin digestion, washed with PBS for 3 times and centrifuged at 1000 rpm for 5 min. The obtained cell pellets were resuspended in 500 μ L binding buffer, incubated in 5 μ L Annexin V-FITC (FITC-labeled Annexin V antibody) and 5 μ L PI in the dark for 30 min at RT, immediately subjected to flow cytometry and analyzed with Cell Quest software.
Cell cycle assay. Propidium iodide (PI) staining assay was used to analyze the influence of Akbu-LAAO on HepG2 cycle. 1 × 10 6 HepG2 cells were seeded into a 6-cm dish and incubated at 37 °C with 5% CO 2 overnight. The cells were then administrated with 20, 38.82, 60 μ g/mL of Akbu-LAAO at 37 °C, 5% CO 2 for 24 h, washed twice with PBS buffer and digested with trypsin (EDTA free). Cells were collected by centrifuging at 1000 rpm for 5 min and washed with PBS. The cell pellets were resuspended and fixed with ice-cold 70% ethanol at 4 °C overnight. Cell pellets were obtained by aspirating ethanol, washed with PBS, incubated with 100 μ L RNase A at 37 °C for 30 min and labeled in 400 μ L PI at 4 °C for 30 min in the dark. Cell cycle was analyzed using flow cytometry and analyzed using CellQuest software. cDNA microarray screening targeting genes of Akbu-LAAO in HepG2 cells. 1 × 10 7 HepG2 and 1 × 10 7 Akbu-LAAO-treated (38.82 μ g/mL Akbu-LAAO at 37 °C, 5% CO 2 for 24 h) HepG2 cells were harvested for total RNA extraction using Trizol TM reagent (Life Sciences). The RNA concentration and quality were assessed by NanoDrop 2000 spectrophotometer (Thermo) and 1.5% denaturing agarose gel electrophoresis. cDNA was synthesized using SuperScript II kit and purified by QIAGEN RNeasy Mini Kit. cRNA was created using a Genechip IVT Labeling Kit. The biotin-labeled fragmented cRNA (≤ 200 nt) was hybridized at 45 °C for 16 h to Affymetrix Genechip (Human Transcriptome Array 2.0). All the arrays were washed, imaged by 3000 7G Scanner and proceeded by Affymetrix Genechip Operating Software. Random variance model (RVM) t-test was performed to screen the differentially expressed genes from independent triplicate experiments.
Quantitative real-time PCR (qRT-PCR). qRT-PCR was performed to determine the level changes of targeted genes following Akbu-LAAO treatment. Total RNA was extracted from HepG2 cells using Trizol TM reagent. Reverse transcription was performed using PrimeScript TM RT Kit with gDNA Eraser. PCR was carried out on an Agilent M × 3005P real-time PCR machine. β -actin (ACTB) was used as the internal reference. PCR primers for targeted genes were listed in Table 2. The comparison of mRNA expression level was calculated using 2 −ΔΔCT method 46 .
Scientific RepoRts | 5:18215 | DOI: 10.1038/srep18215 Data processing and statistical analysis. SPSS 17.0 software was utilized for data analysis. Results are represented as mean ± SD of at least three independent experiments. The differences were assessed with student's t test. Values with *P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 were considered statistically significant differences.