Tumor Biol. (2014) 35:10051–10056 DOI 10.1007/s13277-014-2272-7

RESEARCH ARTICLE

Anticancer bioactive peptide (ACBP) inhibits gastric cancer cells by upregulating growth arrest and DNA damage-inducible gene 45A (GADD45A) Li-Ya Su & Hong-Yi Xin & Yong-Lei Liu & Jia-Ling Zhang & Hong-Wu Xin & Xiu-Lan Su

Received: 30 March 2014 / Accepted: 23 June 2014 / Published online: 12 July 2014 # International Society of Oncology and BioMarkers (ISOBM) 2014

Abstract Recently, we reported that anticancer bioactive peptide (ACBP), purified from goat spleens immunized with human gastric cancer extracts, significantly inhibited gastric cancer cells in vitro and gastric tumors in vivo via repressing cell growth and promoting apoptosis, making it a promising potential biological anticancer drug. However, it is not known what genes are functionally required for the ACBP effects. Here, we first found that two tumor suppressor genes, cyclindependent kinase inhibitor 2B (CDKN2B) and growth arrest and DNA damage-inducible alpha (GADD45A), were upregulated significantly in the cells with ACBP treatment by microarray screening and the findings were validated by real-time RT-PCR. Next, GADD45A mRNA and protein expressions were downregulated in the gastric cancer cells by lentivirus-mediated RNAi; then, cell viability, cell cycle, and apoptosis were assayed by MTT and flow cytometry. Interestingly, our results indicated that cell viability was not dependent on GADD45A without ACBP treatment; however, cell sensitivity to ACBP was significantly decreased in ACBPtreated gastric cancer cells with GADD45A downregulation.

Li-Ya Su and Hong-Yi Xin contributed equally to this paper. L.
Therefore, we demonstrate that GADD45A was functionally required for ACBP to inhibit gastric cancer cells, suggesting that GADD45A may become a biomarker for ACBP sensitivity. Our findings have significant implications on the molecular mechanism understanding, biomarker development, and anticancer drug development of ACBP. Keywords Anticancer bioactive peptide . Gastric cancer . Cell viability . Apoptosis . Cell cycle . CDKN2B . GADD45A

Introduction Cancer is the second leading cause of death in China, and the mortality of gastric cancer ranks the first in the malignant tumor mortality [1]. The morbidity and mortality of gastric cancer ranked the second in the world. Cancer has become a kind of disease that greatly threatens human life. Five methods for cancer treatment were put forward by WHO: operation, radiotherapy, chemotherapy, biological therapy, and Chinese medicine [1, 2]. But, the effects of these treatments are limited, and it is very important to find more effective treatment with less side effects. In a human body, there are hundreds of different kinds of bioactive peptides necessary to complete complex physiological activities. They play key roles in signal transduction, cell differentiation, and individual development and are the natural resources for drug screening [3]. In recent years, the research and development on bioactive peptides are very active, and there are a variety of therapeutic products available. Because of their small molecular weight and high stability, peptides have good biological activity for the treatment of cancer or its adjuvant therapy. Anticancer bioactive peptide (ACBP) is a kind of low molecular substances purified from organs of goats immunized with tumor extracts in our laboratory [4, 5]. ACBP has a

10052

molecular weight of less than 8 KD and won the national invention patent (ZL96122236) in China. It was found that ACBP has good inhibitory effects on the growth of cancer cells in vitro and some solid tumors in vivo. It suppressed a broad range of cancers, including breast cancer, gallbladder cancer, gastric cancer, and so on [6–9]. Our previous reports showed that ACBP inhibited cell proliferation, induced cell cycle arrest and apoptosis, and promoted tissue regeneration of livers and spleens injured due to cancer [10, 11]. However, the molecular mechanism of the anticancer effects is not clear. In this study, to find the molecular mechanism, microarray was performed to identify the differences of gastric cancer cell with or without ACBP treatment. It was found that cyclin-dependent kinase inhibitor 2B (CDKN2B) and growth arrest and DNA damage-inducible gene 45A (GADD45A) were increased significantly in MGC-803 gastric cancer cells with ACBP. To investigate the role of CDKN2B and GADD45A in gastric cancer, the two genes were knocked down using lentivirus-mediated RNAi. The results showed that GADD45A is functionally required for the anticancer effects of ACBP and may become a biomarker of ACBP sensitivity.

Tumor Biol. (2014) 35:10051–10056

the manufacturer’s instructions. MGC803 cells were seeded in each well of a six-well plate and transduced by shRNAexpressing lentivirus. The medium was removed and replaced with fresh complete medium next day. RNA isolation and real-time RT-PCR Total RNA, following the manufacturer’s instructions, was isolated from the cells using Trizol reagent (Invitrogen). Briefly, the cells were lysed in TRIzol and then mixed with chloroform. The lysate was centrifuged to separate RNA, DNA and protein, total RNA recovered, precipitated with isopropanol, washed in 75 % ethanol to remove impurities before dissolving in water. After that, 2 μg of RNA was taken and treated with DNase to remove contaminating DNA prior to the reverse transcription to cDNA using PrimeScriptTM RTPCR Kit (Takara). To measure mRNA expression, real-time RT-PCR was performed using ABI-Prism Sequence Detection System. The relative expression levels were calculated by comparing Ct values of the samples with those of the reference, all data normalized to the internal control β-actin. Western blotting

Materials and methods Cell culture MGC803 cells were purchased from ATCC (Manassas, VA, USA). RKO cells were purchased from Genechem Co., Ltd. (Shanghai, China). The cells were maintained in RPMI1640 (Gibco, USA) supplemented with 10 % fatal bovine serum (Gibco, USA) in a humidified incubator at 37 °C and 5 % CO2. Microarray assay Affymetrix U133 plus 2.0 microarray chip was used to investigate the expression changes of cell apoptosis and cell cyclerelated genes before and after administration of ACBP Affymetrix microarray scanner, and GCOS1.2 software was used to read and process the data. RNAi lentiviral vector construction, lentivirus production, and transduction The RNAi target gene sequences were chosen with siRNA converter software (www.ambion.com); four sequences were designed for each gene. RKO cells were used as foreign target cells for the detection of the interference efficiency. Doublestranded oligonucleotides targeting the endogenous GADD45A and CDKN2B gene were generated and cloned into PGCSIL-PUR. The lentivirus was produced according to

Anti-P15INK4B antibody was purchased from Cell Signaling. Anti-GADD45A antibody was ordered from ABCAM. AntiGAPDH antibody and secondary antibodies conjugated with HRP were purchased from SANTA CRUZ. The cells were scraped from the dishes, cellular protein extracts prepared by homogenization in an ice-cold lysis buffer and their lysates obtained by centrifugation at 12,000 × g for 20 min, and the total protein concentration determined using Lowry method. Equal amounts of protein, separated by SDS-PAGE, were electrophoretically transferred to a PVDF membrane at 320 mA for 2 h at a low temperature, and the membrane was blocked with 5 % fat-free milk with 0.05 % Tween 20 in PBS. Subsequently, the membrane was probed with the primary antibodies. The blots were washed in PBST and then incubated in anti-mouse IgG or anti-rabbit IgG secondary antibody for about 3 h at RT. Washed in PBST (×3, 10 min each wash), the proteins were finally visualized using ECL based on the manufacturer’s instructions. MTT assay MTT assay was employed to detect the growth of gastric cancer cells, and the growth curve was delineated. Logarithmic phase cells were collected, and the concentration of the cell suspension was adjusted to 5,000 cells per well (the edge wells of the plate are filled with aseptic PBS buffer). The cells were incubated at 37 °C, 5 % CO2 until the cells cover the bottom of the well (a flat bottom 96-well plate), and then, the cells were cultured. Twenty microliters of the MTT solution

Tumor Biol. (2014) 35:10051–10056

was added to each well (5 mg/ml, 0.5 % MTT), and the cells were continued to culture for 4 h. After the incubation, the supernatant was discarded, and 150 μl dimethyl sulfoxide was added to each well, and the culture plate was shaked at low speed for 10 min until crystal dissolved completely. The ELISA reader was used to measure the absorbance at 570 nm. Determination of apoptosis by flow cytometry For apoptosis assay, the Annexin V staining was quantified by flow cytometry. The cells were plated in a six-well plate, transfected with the indicated plasmid, or transduced with shRNA-expressing lentivirus; at 24 h later, the complete growth medium was changed to growth medium without serum. At another 24 h later, the cells were collected, washed in cold PBS twice, and resuspended in 1× binding buffer at a concentration of 1×106 cells/ml. After that, the cells in 100-μl solution were transfered to a 5-ml culture tube, with 5-μl Annexin V-FITC and 5 μl PI (BD Biosciences) added, and gently vortexed and incubated for 15 min at RT in the dark. And finally, 400-μl 1× binding buffer was added to each tube to be analyzed by flow cytometry within 1 h. Cell cycle analysis In 2-ml culture medium, 2×105 cells/well (six-well plate) were seeded and cultured for 24 h before collection. The cells were stabilized with 75 % ethanol for 24 h, dyed with PI, and analyzed with ModFit of flow cytometry.

10053

assays of cell proliferation, gene expression, cell cycle, and apoptosis, error bars representing ± SE.

Results CDKN2B and GADD45A are upregulated in gastric cancer cells by ACBP ACBP induces cell cycle arrest and apoptosis in vivo and in vitro, but the mechanism is not clear. In order to clarify its mechanism, MGC803 gastric cancer cells were exposed to ACBP (25 μg/ml) for 48 h, and gene expression was analyzed by microarray. Different gene expression patterns in cell cycle and apoptosis were found by cluster analysis in the cells with ACBP treatment compared with the control cells. It was very interesting that CDKN2B and GADD45A were significantly upregulated (Fig. 1a). Next, effect on the expression of these two genes was validated by real-time RT-PCR. The cells were treated with no peptide control, peptide (the extraction of the peptide without immune induction), or ACBP for 48 h (Fig. 1b). The results showed that treatments of peptide and ACBP both upregulated CDKN2B and GADD45A expressions in the gastric cancer cells, and the increase was significantly greater in the cell with ACBP treatment (Fig. 1b). Effective knockdown of CDKN2B and GADD45A by lentivirus-mediated RNAi

Each experiment was repeated at least three times, Student’s t tests performed to determine the statistical significance for the

The above data suggested that CDKN2B and GADD45A might play important roles in the anticancer effects in MGC803 gastric cancer cells. In order to investigate the role of the two genes in gastric cancer cells, shRNAs targeting the two genes were designed and cloned in the lentivirus vector.

Fig. 1 CDKN2B and GADD45A are upregulated in MGC803 gastric cancer cells. a Gene expression profile in the MGC803 gastric cancer cells. MGC803 cells were treated with ACBP (25 μg/ml) for 48 h, and total RNA was isolated for microarray assay. The cells without ACBP treatment was the control. b, c MGC803 cells were treated with the

peptide without immune induction or ACBP (25 μg/ml) for 48 h, and total RNA was isolated for real-time RT-PCR. The cells without peptide are the control. CDKN2B (* vs control, p<0.05; # vs peptide, p<0.05) and GADD45A (* vs control, p<0.01; # vs peptide, p<0.01) were significantly upregulated

Statistical analysis

10054

MGC803 cells were transducted with lentiviruses containing different sequences of RNAi, and CDKN2B and GADD45A expressions were analyzed by real-time RT-PCR and Western blot. The results showed that mRNAs of the two genes were knocked down with each of the four target shRNAs and knockdown 3 (KD3) was the most efffective shRNA for CDKN2B (Fig. 2a) and knockdown 2 (KD2) for GADD45A (Fig. 2b) in RKO cells. The results were confirmed (Fig. 2c–d) in MGC803 cells. Our data showed that CDKN2B mRNA was knocked down significantly, but the CDKN2B protein is

Fig. 2 Knocking down of CDKN2B and GADD45A in MGC803 cells. a Selection of shRNA targeting CDKN2B in RKO cells. Four sequences targeting CDKN2B were designed, cloned into lentivial vector. Lentiviruses were produced in 293 T cells. RKO cells were infected with the lenviruses; total RNA was extracted for real-time RT-PCR (* vs control, p<0.01; # vs NC, p<0.01;). Control parent cells, NC lentivirus negative control, KD1-4 knockdown sequence 1-4 targeting CDKN2B. b Selection of shRNA targeting GADD45A in RKO cells. Four sequences targeting GADD45A were designed, cloned into lentivial vector. Lentiviruses were produced in 293 T cells. RKO cells were infected with the

Tumor Biol. (2014) 35:10051–10056

not knocked down (data not shown). GADD45A mRNA and protein were both knocked down effectively (Fig. 2b, d, e). GADD45A alone does not change gastric cancer cell viability but is required for ACBP sensitivity To investigate the role of GADD45A in ACBP inhibition of cell proliferation, the gastric cancer cells were treated by ACBP, and cell proliferation was assayed by MTT method. The data showed that cell proliferation was not changed by

lenviruses; total RNAs were extracted for real-time RT-PCR (* vs control, p<0.01; # vs NC, p<0.01). Control parent cells, NC lentivirus negative control, KD1-4 knockdown sequence 1-4 targeting GADD45A. c CDKN2B mRNA in MGC803 cells was analyzed by real-time RT-PCR (* vs NC, p<0.01). d GADD45A mRNA in MGC803 cells was analyzed by real-time RT-PCR (* vs NC, p<0.01). e GADD45A protein decreased in the cells with the lentivirus containing GADD45A shRNA. MGC803 cells were infected with lentivirus with GADD45A shRNA, and total protein was extracted for Western blot (M protein marker, Con control, NC negative control, KD knocking down)

Tumor Biol. (2014) 35:10051–10056

GADD45A downregulation without ACBP treatment, but cell proliferation was increased by GADD45A knockdown when the cells were treated with ACBP (Fig. 3). These results demonstrated that GADD45A is required for ACBP sensitivity. The cell proliferation data suggested that GADD45A control the cancer cell growth negatively when cells were treated with ACBP. To study whether GADD5A effect on ACBP sensitivity is associated with cell cycle and apoptosis, GADD5A was knocked down in gastic cancer cells with/without ACBP treatment (see “Materials and methods”), and cell cycle and apoptosis were analyzed by flow cytometry. The results showed that cell cycle and apoptosis changes measured in our experiments were not consistently associated with GADD5A effect on ACBP sensitivity (data not shown).

Discussion Our previous work shows that ACBP inhibited a wide range of cancers in vitro and in vivo and promoted regeneration of cancer-induced liver and spleen damages. ACBP was found to inhibit cancer cell proliferation and induce cell cycle arrest and apoptosis. Those discoveries suggest that ACBP is a promising potential biological anticancer drug. However, lack of the molecular mechanisms and biomarkers of ACBP anticancer effects rendered its further development as a cancer therapeutic drug. In this study, microarray gene chip was used to analyze gene expression profiles of gastric cancer cells treated

Fig. 3 GADD45A alone does not change gastric cancer cell viability but is required for ACBP sensitivity. MGC803 cells with GADD45A knocking down were exposed to ACBP for 24 h, and cell prolieration was assayed by MTT method

10055

with ACBP, and the data indicated that CDKN2B and GADD45A were upreguated significantly following ACBP treatment. To functionally test the role of CDKN2B and GADD45A, we effectively knocked down the mRNA expressions of CDKN2B and GADD45A and the protein levels of GADD45A by lentivirus-mediated RNAi. Further functional analysis showed that GADD45A alone did not decrease cell proliferation but increased drug sensitivity of cells exposed to ACBP. For the first time, we demonstrate that GADD45A was functionally required for ACBP to inhibit gastric cancer cells, suggesting that GADD45A may become a biomarker for ACBP sensitivity. These findings have significant implications on the molecular mechanism understanding, biomarker development, and anticancer therapeutic targeting of ACBP. During the past two decades, cancer genetics has shown that overexpression and hyperactivating mutations in growth signaling networks, coupled to loss of function of tumor suppressor proteins, drives oncogenic proliferation. Gene expression profiling of these complex and redundant mitogenic pathways to identify prognostic and predictive signatures and their therapeutic targeting has, however, proved challenging [12]. Our high-throughout microarray analysis found that the two cell cycle-related genes of CDKN2B and GADD45A were significantly upregulated in the gastric cancer cells with ACBP treatment compared with the control. The cell cycle machinery, which acts as an integration point for information transduced through upstream signaling networks, represents critical targets for diagnostic and therapeutic interventions [13]. Analysis of the DNA replication initiation machinery and mitotic engine proteins in human tissues is now leading to the identification of novel biomarkers for cancer detection and prognostication and is providing target validation for cell cycle-directed therapies. In our RNAi knockdown experiments, CDKN2B protein was not knocked down effectively; so, we focus in the role of GADD45A. We further demonstrate that GADD45A alone did not decrease cell proliferation but increased drug sensitivity of cells exposed to ACBP. GADD45 proteins, including Gadd45a, Gadd45b, and Gadd45g, have been implicated in stress signaling in response to physiological and environmental stress, including oncogenic stress, which can result in cell cycle arrest, DNA repair, cell survival, senescence, and apoptosis [14]. Gadd45 genes encode for small (18 kDa), evolutionarily conserved proteins, are highly acidic, and are localized within both the cell cytoplasm and nucleus [15]. The function of Gadd45 as a stress sensor is mediated via a complex interplay of physical interactions with other cellular proteins implicated in cell cycle regulation and the cell response to stress, notably PCNA, p21, cdc2/cyclinB1, and the p38 and JNK stress response kinases [16]. Altered expression of Gadd45 has been observed in multiple types of solid tumors as well as in hematopoietic malignancies. Using genetically engineered mouse models and bone marrow transplantation,

10056

evidence has been obtained indicating that Gadd45 proteins can function to either promote or suppress tumor development; this is dependent on the molecular nature of the activated oncogene and the cell type, via engagement of different signaling pathways [17]. The complex bidirectional functions of GADD proteins could be part of the reasons why the cell cycle and apoptosis changes were not consistently associated with GADD5A effect on ACBP sensitivity in our experiments. It is also speculated that there are other genes involved in the cell cycle and apoptosis induced by ACBP. There was a report that showed that ACBP induced p16, p21, p27, and bcl2, inhibited cyclin D1 and c-myc expression, and suppressed human gastric cancer growth [18]. Therefore, further research is needed to clarify the mechanism. In conclusion, in the present study, we find that ACBP treatment that upregulated GADD45A in gastric cancer cell and cell sensitivity to ACBP was decreased in ACBP-treated gastric cancer cells with GADD45A downregulation. Therefore, we demonstrate that GADD45A was functionally required for ACBP to inhibit gastric cancer cells, suggesting that GADD45A may become a biomarker for ACBP sensitivity. Our future research will try to test the role of GADD45A in other cell lines and tumors of gastric and other cancer types, as well as to understand the detailed mechanisms of the anticancer effects of GADD45A. Acknowledgment This research was supported by the National Natural Science Foundation (81160254).

References 1. Ren JS, Li Q, Guan P, Dai M, Yang L. Estimation and prediction for incidence, mortality and prevalence of common gastrointestinal tract cancers in China, in 2008. Zhonghua Liu Xing Bing Xue Za Zhi. 2012;33(10):1052–5. PubMed PMID: 23290850. 2. Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, et al. Cancer statistics, 2006. CA Cancer J Clin. 2006;56(2):106–30. PubMed PMID:16514137.

Tumor Biol. (2014) 35:10051–10056 3. Hou JF, Yan MR, Yan XH, Jiao TM, Rong YN, Su XL. Effect of anticancer bioactive peptide on leukemia mice. J Inn Mong Med Coll. 2004;26(1):3–6. 4. Yang ZY, Wang WL, Su XL. HPCE analysis of polypeptides isolated from goat spleens. Chin J Pharm Anal. 2005;10:1248–9. 5. Yang ZY, Zhang ZP, Su XL. MALDI-TOF-MS analysis of polypeptides isolated from goat spleens. Chin Pharm J. 2005;10:797. 6. Xu GH, Su XL, Shen JF, Bi, Xiaohui O. Control of anti-cancer bioactive peptide on the cell cycle of gastric cancer BGC-823. Chin J Clin Oncol. 2007;14:1361–4. 7. Zhao YY, Peng SD, Su XL. Effects of anti-cancer bioactive peptide on cell cycle in human nasopharyngeal carcinoma strain CNE. Chin J Otorhinolaryngol Head Neck Surg. 2006;41:607–11. 8. Jia SQ, Wang WL, Su XL. Inhibitory effect of anti-cancer bioactive peptide on proliferation of human breast cancer cell line nm231. Chin Med Biotechnol. 2007;2:270–5. 9. Su XL, Ying X, Cui HM, et al. And the mechanism of anti-cancer bioactive peptide on GBC-SD gallbladder cancer cells. Acta Anat Sin. 2007;38(Suppl):39–42. 10. Wang ZY, Yang CW, Oy XH, Su XL. Anti-cancer bioactive peptide-S induced apoptosis of hepatoma cells and cell cycle regulation. Chin J Lab Diagn. 2010;14:1–6. 11. Su X, Dong C, Zhang J, Su L, Wang X, Cui H, et al. Combination therapy of anti-cancer bioactive peptide with Cisplatin decreases chemotherapy dosing and toxicity to improve the quality of life in xenograft nude mice bearing human gastric cancer. Cell Biosci. 2014;4:7. 12. Williams GH, Stoeber K. The cell cycle and cancer. J Pathol. 2012;226:352–64. 13. Miller JP, Yeh N, Vidal A, Koff A. Interweaving the cell cycle machinery with cell differentiation. Cell Cycle. 2007;6:2932–8. 14. Yu KD, Di GH, Li WF, Rao NY, Fan L, Yuan WT, et al. Genetic contribution of GADD45A to susceptibility to sporadic and nonBRCA1/2 familial breast cancers: a systematic evaluation in Chinese populations. Breast Cancer Res Treat. 2010;121:157–67. 15. Gujuluva CN, Baek JH, Shin KH, Cherrick HM, Park NH. Effect of UV-irradiation on cell cycle, viability and the expression of p53, gadd153 and gadd45 genes in normal and HPV-immortalized human oral keratinocytes. Oncogene. 1994;9:1819–27. 16. Salvador JM, Brown-Clay JD, Fornace Jr AJ. Gadd45 in stress signaling, cell cycle control, and apoptosis. Adv Exp Med Biol. 2013;793:1–19. 17. Liebermann DA, Tront JS, Sha X, Mukherjee K, Mohamed-Hadley A, Hoffman B. Gadd45 stress sensors in malignancy and leukemia. Crit Rev Oncog. 2011;16:129–40. 18. Su L, Xu G, Shen J, Tuo Y, Zhang X, Jia S, et al. Anticancer bioactive peptide suppresses human gastric cancer growth through modulation of apoptosis and the cell cycle. Oncol Rep. 2010;23:3–9.