Background
Ovarian cancer remains the most lethal gynecologic tumor in the world. Serous cystadenoma is the most common histologic subtype of ovarian cancer. Although aggressive cytoreductive surgery followed by adjuvant chemotherapy (e.g., cisplatin and paclitaxel) has been shown to induce a clinical response in the majority of ovarian cancer patients, most women will eventually relapse and die due to the development of chemotherapy-resistant disease [
1]. For this reason, there is a strong need novel treatment options in this patient population. Peptide therapeutics represents an emerging field of anti-cancer agents that are easily obtained from either natural resources or are designed based on target protein structure. Moreover, past reports indicate that therapeutic peptides typically exhibit little toxicity in normal host cells [
2]. Recently, Su L. Y. et al. [
3] reported that the anti-cancer bioactive peptide (ACBP), purified from goat spleens that were immunized with human gastric cancer extracts, significantly inhibited gastric cancer cell growth in vitro and gastric tumor growth in vivo, indicating that therapeutic peptides may represent a powerful anti-cancer tool.
Biologically active peptides have also been described in anti- ovarian cancer cells, including LfcinB (bovine lactoferricin), a peptide originally derived from bovine lactoferrin, which has been shown significantly to inhibit the in vitro growth and in vivo tumor development of the ovarian cancer cell line SKOV3 [
4]. LfcinB is derived in bovine lactoferrin (sequence of 17–41 residues), having 25 amino acids (FKCRRWQWRMKKLGAPSITCVRRAF). According to the literature, the active site of LfcinB residues in residues 4 to 9 of the amino acid sequence [
5,
6]. However, clinical use of LfcinB is most likely limited due to high toxicity, poor stability and the propensity of the molecule to undergo structural changes under different environments that affect the biological activity of LfcinB. Although the anti-cancer effects of LfcinB have been generally accepted, the underlying mechanisms of LfcinB function remain unclear.
The hexapeptide (Pro-Gly-Pro-Ile-Pro-Asn, PGPIPN, 63–68 residues of bovine ß-casein), also known as immune hexapeptide or immunomodulating peptide, was isolated from hydrolysate of bovine β casein and has been shown to elicit an immune response in cancer cells [
7‐
10]. In line with this finding, our research group recently showed that PGPIPN inhibited the growth of SKOV3 cells in vitro and promoted apoptosis and decreased tumor growth in a xenograft ovarian cancer model [
11]. However, the anti-cancer effects of PGPIPN are significantly lower than those of classical anti-cancer drugs, such as paclitaxel or 5-fluorouracil (5-FU) [
11].
Based on structure-function studies of LfcinB and PGPIPN, we performed molecular modification of the two biological peptides derived from milk proteins. Using web- (
http://swissmodel.expasy.org) and software-based (Accelery Insight II 2005.1LBIOVIA, San Diego, USA) tools, we designed an anti-cancer fusion peptide (ACFP) that connects regions of LfcinB and PGPIPN by a flexible link arm (GGGGS). Importantly, fusion of the two peptides led to a molecule with superior anti-cancer function and increased overall structural stability and anti-enzymatic hydrolysis. ACFP consists of a disulfide bond chain, which is likely to lead to cancer cell membrane localization. The theoretical molecular weight of ACFP was calculated to be 3960 (our experimental determination is 4020), which is smaller than the immunogenic molecules and does not stimulate immunogenicity.
In this study, we investigated the anti-viabilities of ACFP in primary ovarian cancer cell and normal ovarian epithelial cells in vitro. Using cDNA chip, we observed that ACFP treatment led to significant changes in gene expressions and cancer-associated signaling pathways in primary ovarian cancer cells. ACFP-specific effects on bcl-xl, bax, akt, caspase-3, CDC25C and cyclinB1 gene expressions were confirmed using real time PCR and western blot analysis. Overall, this study investigates the molecular mechanisms that underlie the anti-viabilities of ACFP on anti-ovarian cancer and provides the experimental basis for developing ACFP as a new therapeutic agent in ovarian cancer.
Methods
Reagents
The ACFP peptide was provided by Shanghai Sangon Biological Engineering Technology, and the purity was confirmed by RP-HPLC to be > 99.5 %. Trizol Kit was purchased from Invitrogen, USA. Reverse Transcription System was purchased from Promega, USA. Eight joint tubes for PCR were purchased from ABI, USA. SYBR Green I Premix Ex Taq was purchased from Takara Biotechnology (Dalian) Co., Ltd, China. Mouse monoclonal antibodies of Bcl-xl, Bax, Akt, Caspase-3, CDC25C, CyclinB1 and β-Actin were purchased from Santa Cruz Biotechnology, Inc. The horseradish peroxidase conjugated secondary antibody (goat anti-mouse IgG) and Super Signal West Pico Trial Kit (ECL chromogenic reagent kit) were purchased from Pierce, USA; BCA Kit for protein quantitative assay was purchased from Shanghai Sangon Biological Engineering Technology.
Cell cultures
Fresh primary ovarian tumor tissues that were assessed and classified as serous ovarian adenocarcinoma (III-IV grade) according to WHO criteria were collected from 53 ovarian cancer patients who underwent initial debulking surgery at the first affiliated hospital of Anhui Medical University between June 2012 and June 2014. All patients had not received adjuvant therapies, such as chemotherapy or radiotherapy, prior to surgery. For comparison, normal ovarian glandular epithelium cells (NOGECs) were cultured from fresh primary normal ovarian tissues harvested from 7 patients with uterine fibromas, which were confirmed as negative for any neoplastic disease by pathological examination. Prior to tissue deposition, all patients signed written consent forms confirming their donation of tissue for research purposes according to the Declaration of Helsinki. This study was approved by the Anhui Medical University Review Board. Tumor or normal ovarian tissues were cut into small 1.0 mm3 pieces, rinsed two times in phosphate buffered saline (PBS) and digested with 0.25 % trypsin in a sterile centrifuge tube at 37 °C for 30 min. To obtain a single cell suspension cell, digested tissues were filtered with a 100 μm cell strainer. Cells were collected by centrifugation at 1000 rpm for five minutes, and the cell pellet was re-suspended in Dulbecco’s modified eagle medium (DMEM) supplemented with 10 % fetal bovine serum (FBS). Cells were subsequently cultured in DMEM containing 0.1 mg/L epidermal growth factor (EGF), 0.1 mg/L insulin-like growth factor (IGF) and 0.1 mg/L beta-estradiol with 10 % FBS in 5 % CO2 at 37 °C. Once cells reached 70 to 80 % confluence, cell culture medium was drained from the flask, and cells were digested with 0.25 % collagenase II until approximately 1/3 of the cells fell to the bottom of the dish by eye using a microscope. Due to their initial shedding, most fibroblasts were eliminated by collagenase digestion. The remaining cells were cultured in 5 % CO2 at 37 °C. To determine purity, cells were analyzed using immunofluorescence of cytokeratin 7 (for ovarian cancer cells) or cytokeratin 19 (for normal ovarian epithelium cells).
Cell viability assay
Primary ovarian cancer cells were seeded into 96-well plates in sextuplicate at a starting density of 5 × 103 cells/well and incubated with ACFP at the concentrations: 0 (as control), 5 × 10−6, 5 × 10−5, 5 × 10−4, 5 × 10−3, 5 × 10−2, 5 × 10−1 and 5 g/L for 24, 48 and 72 h, respectively. Cells treated with paclitaxel at 5 × 10−4 g/L were included in the same plate as a positive control. Cell viability was later measured using the WST-1 (water-soluble tetrazolium 1) cell viability and cytotoxicity assay kit (Beyotime, Haimen, China) according to the manufacturer’s instructions. The percent viability of cells was calculated using the formula to calculate the cell viability ratio (VR): VR (%) = (the experimental group A450nm value/control group A450nm value) × 100 %. To assess general toxicity of ACFP, viability of normal ovarian cells treated with ACFP was assayed using the same procedure. The effect of ACFP was compared with LfcinB and PGPIPN (the products of Shanghai Sangon Biological Engineering Technology, China). Each experiment was performed in two independent sets.
Apoptosis assay
Apoptosis of primary ovarian cancer cell treated with ACFP was measured by flow cytometry (FCM) using FITC-conjugated Annexin-V and propidium iodide (PI) from Sigma. Cells were washed twice with cold PBS and resuspended in Annexin-V binding buffer (10 mM HEPES, 140 mM NaCl and 5 mM CaCl2) at a concentration of 1 × 106 cells/mL. A single cell suspension of 1 × 106 cells was prepared in a 5 mL culture tube according to the instrument manual and 5 μL Annexin-V-FITC at 10 μg/mL and 10 μL propidium iodide at 10 μg/mL was added. The tube was gently vortexed and incubated for 15 min at room temperature in the dark. Binding buffer (400 μL) was subsequently added to each tube and the cells were analyzed by flow cytometry (EPICSR XL-MCL, Beckman, USA) with EXPO32TM ADC software (Beckman, USA). Cells that stained positive for annexin V were counted as apoptotic.
cDNA microarrays in screening of differentially expressed genes
We utilized cDNA microarrays to observe the effect of ACFP on gene expression in primary ovarian cancer cells. Cells were treated with 0 (the vehicle group, as control), 5 × 106 g/L and 5 × 103 g/L ACFP, respectively, and cultured for 48 h at 37 °C in 5 % CO2. Cells were later digested and collected for total RNA extraction using Trizol. The RNA concentrations and purities were detected with a spectrophotometer. The experiments were performed in duplicate on a single total RNA preparation from the cells. Signal values were presented as the mean value of two replicate experiments. RNA samples were used to generate human whole genome expression profiling microarray (Yeli Bioscience Co., Ltd.; Shanghai, China). RNA was subsequently converted into digoxigenin-labeled complementary RNA and hybridized to a human genome microarray system (Human OneArray Microarray, from Phalanx Biotech Group, Taiwan). The chips were scanned by a GeeDom® LuxScan 10 K microarray scanner. LuxScan3.0 software was used to extract probe fluorescence signals and analyze images. Finally, the original data were processed by normalizing. The differential gene screening, cluster analysis and pathway analysis were conducted from microarray data by Shanghai Sensichip Infotech Co. Ltd. (China). Genes that displayed a signal value greater than 100 and a ratio of ACFP treatment vs control greater than 2 were defined as up-regulated, while genes with a signal value greater than 100 and a ratio of ACFP treatment vs control less than 0.5 were defined as down-regulated.
Real time PCR in measuring mRNA of bcl-xl, bax, akt, caspase-3, CDC25C and cyclinB1
An optimized RT-PCR protocol was employed to analyze the mRNA levels of
bcl-xl,
bax,
akt,
caspase-3,
CDC25C and
cyclinB1. Beta-
actin was used as a housekeeping gene. According to primer sequences of
bcl-xl,
bax,
akt, caspase-3,
CDC25C,
cyclinB1 and β-
actin genes retrieved from Primer-Bank, primers were designed with the Primer 5.0 software, which were synthesized by Shanghai Sangon Biological Engineering Technology. These primer sequences are as follows:
-
bcl-xl forward 5′-AGCTGGTGGTTGACTTTCTCTC-3′,
-
bcl-xl reverse 5′-CCTCAGTCCTGTTCTCTTCCAC-3′;
-
bax forward 5′-GGTTGTCGCCCTTTTCTACTTT-3′,
-
bax reverse 5′-GTGAGGAGGCTTGAGGAGTCT-3′;
-
akt forward 5′-CGGGGTAGGGAAGAAAACTATC-3′,
-
akt reverse 5′-TGACAGAGTGAGGGGACACA-3′;
-
caspase-3 forward 5′-GACTCTGGAATATCCCTGGACAACA-3′,
-
caspase-3 reverse 5′AGGTTTGCTGCATCGACATCTG-3′;
-
CDC25C forward 5′-GCTAACAAGTCACCAAAAGACA-3′,
-
CDC25C reverse 5′-TCCCTGAACCAATACAATCTC-3′;
-
cyclinB1 forward 5′-AGGTCCATCTCAGGTTCCACTT-3′,
-
cyclinB1 reverse 5′-GAGTAGGCGTTGTCCGTGAT-3′;
-
β-actin forward 5′-ATGTTTGAGACCTTCAACACCCC-3′,
-
β-actin reverse 5′-GCCATCTCTTGCTCGAAGTCCAG-3′.
Primary ovarian cancer cells were harvested after ACFP treatment at different doses and times, respectively. The total RNAs in the primary ovarian cancer cells were extracted according to the Trizol kit manufacturer’s instructions, and the purity and concentration were determined by ultraviolet spectrophotometry. According to the RNA template and primers, cDNAs of specific genes were synthesized in system of reverse transcription reaction including 10 × 2 μl buffers, dNTPs (10 mM) 2 μl, AMV reverse transcriptase of 1 μl, 0.5 μl recombinant RNasin and total RNA 1 μl in final volume of 20 μl by adding RNase-free water. The reverse transcription reaction conditions were 42 °C 15 min and 95 °C 5 min. After the reaction, the reverse-transcribed cDNAs were diluted with RNase-free water to a final volume of 60 μL and preserved at 80 °C.
Real time PCR adopts TaKaRa SYBR Green as real time PCR Master Mix in ABI7500 fluorescent real-time PCR instrument. The reaction conditions were as follows: 95 °C × 30 s (1 cycle); 95 °C × 5 s, 60 °C × 34 s (40 cycles). At the end of PCR cycling steps, data for each sample were displayed as a melting curve. The specificity of the amplified products was confirmed using melting curve analysis. The ABI SDS software (Applied Biosystems) was used to determine a critical threshold (C
t), which was defined as the cycle number where the linear phase for each sample crossed the threshold level. The mRNAs of target gene expression were denoted by ΔC
t (ΔC
t = target gene C
t - β-
actin C
t value). Finally, the relative mRNA expression of all samples were calculated using the 2
-ΔΔC
t method [
12]. All reactions were performed in triplicate, and a mixture lacking a complementary DNA template (NTC) was used as the negative control.
Western blot for analysis of Bcl-xl, Bax, Akt, Caspase-3, CDC25C and CyclinB1 proteins
Proteins were isolated from primary ovarian cancer cells harvested after ACFP treatment, separated by SDS-PAGE and transferred to PVDF membrane using the standard protocol. After blocking with 5 % (w/v) dry skim milk, membranes were incubated with primary antibodies (mouse monoclonal Bcl-xl, Bax, Akt, Caspase-3, CDC25C, CyclinB1 and β-Actin antibodies, 1:1000 dilution) according to the manufacturer’s instructions and later incubated with a horseradish peroxidase conjugated secondary antibody (goat anti-mouse IgG, 1:8000 dilution). The proteins were detected with the enhanced chemiluminescence (ECL) system followed by exposure to X-ray film. The β-Actin was used as a loading control. Two independent experiments were performed. Digital images were captured by Gel DocTM gel documentation system (Bio-Rad, USA) and intensities were quantified using Quantity-One software version 4.62 (Bio-Rad, USA).
Statistical analysis
All data were expressed as the mean ± SD. The differences among groups were analyzed using the one-way ANOVA by SPSS 15.0 statistical software. The results were considered to be statistically significant when P < 0.05.
Discussion
Over the past 30 years, the slow improvement in the overall survival in high-grade serous ovarian cancer patients can be partly attributed to a lack of advancement in treatments beyond platinum-based combination chemotherapy [
13]. Clinical application of anti-tumor chemical drugs is often limited due to frequent toxicity, narrow spectrum of activity and acquired resistance [
14]. Thus, there is a need to explore or design new anti-tumor drugs with mechanisms of action that work, despite these obstacles. Among the newly developed anti-cancer drugs, bioactive peptides are one of the most promising drugs for the future of ovarian cancer therapy. Peptides are a novel class of anti-cancer agents that can be engineered to target cancer cells specifically with lower toxicity to normal tissues and offer new opportunities for cancer prevention and treatment [
15]. Bioactive peptides are specific protein fragments that may have a positive impact on health and represent an important source of new anti-carcinogenic and immunomodulatory agents. Exploration of bioactive peptides plays a significant role in the development of innovative and unconventional anti-cancer drugs [
16].
Milk is considered a nutritious food that consists of precursors of active peptides with biological and physiological properties. Bioactive peptides from milk proteins have been defined as specific protein fragments that have a positive impact on body functions or conditions and may ultimately influence health. The size of active milk peptides varies from 3 to 40 amino acid residues and many have been characterized as multi-functional proteins [
17]. A human milk peptidomics study conducted by Dallas D.C. et al. [
18] identified over 300 peptides by mass spectrometry analysis, of which the majority consisted primarily of peptides derived from β-casein and a large number of peptides that showed significant sequence overlap with peptides with known functions. Bovine milk proteins are currently the primary source of a range of biologically active peptides derived in milk, and among milk-born bioactive peptides, anti-cancer peptides have been exhibited a broad potential for clinical application in human trials and clinical studies. For example, treatment of biliary cancer patients with probiotic ingestion combined with radiotherapy led to significantly greater tumor regression and increased overall survival compared to radiotherapy treatment alone [
19].
Peptides possess many advantages for the development of anti-tumor medications, including high selectivity, high potency, a broad range of targets, and low toxicity. However, poor stability, low membrane permeability, and susceptibility to proteolytic digestion have limited their clinical use to date [
15]. To overcome these obstacles, modifications of peptide structure should be feasible and may lead to increased bioactivity. For example, development of flexible fusion peptides may lead to greater access into the cell and therefore more efficient disruption of targeted pathways [
20]. Currently, a number of modification strategies have been developed and have been successfully used to improve the efficiency of anti-cancer peptides.
In the present study, we developed an anti-cancer fusion peptide based on sequences from LfcinB from bovine lactoferrin and hexapeptide (PGPIPN) from bovine β casein. LfcinB was chosen because its anti-tumor activity has recently been established in several cell lines and
in vivo models, as demonstrated by membrane disruption and extensive hemorrhagic necrosis, respectively. Importantly, LfcinB peptides that harbor cationic residues within one sector of the helical structure were shown to be the most active in tumor cell lines, suggesting that specific structural regions are linked with bioactivity [
21]. Despite its anti-tumor activity, issues with LfcinB stability and susceptibility to proteases have prevented its clinical use. In contrast to LfcinB, the PGPIPN peptide is rich in proline residues, rendering the molecule resistant to proteolytic degradation [
21]. However, PGPIPN activity is lower than that of the traditional anti-cancer drugs, such as paclitaxel and cisplatin. Compared with its parent peptides, ACFP had many advantages. According to bioinformatics, the peptide has a disulfide bond and a stable α-helix in N-terminal, which is a relatively stable molecular. The peptide C-terminal containing three prolines can resist the hydrolysis of proteaseto [
21]. Our previous experiments also showed that the peptide was very stable (Additional file
2: Figure S2). ACFP contains 8 charged amino acids (three lysines, five arginines) and can easily dissolve in water. ACFP is slightly soluble in fat (octanol). According to our preliminary experiments, oil (octanol)/water partition coefficient of the peptide was −0.91 (pH 7). From the experimental results, the ACFP peptide is superior to LfcinB and PGPIPN with increased structural stability, anti-cancer activity, intracellular access and reduced toxicity in normal cells.
Genomic characterization of ovarian cancers has emphasized the role of gene mutation and/or altered gene expression in the initiation and progression of the disease. In this study, we show using cDNA microarray that ACFP can up-regulate the expression of certain genes (including
bax and
caspase-3), and down-regulate the expression of certain other genes (including
bcl-xl,
akt,
CDC25C and
cyclinB1). Building on these results, we hypothesize that screening a larger panel of gene expression profiles from human ovarian cancer cells will most likely help to elucidate how ACFP induces anti-tumor activity [
22]. Among the most differentially regulated genes, many are related to cell viability, cycle and apoptosis. However, we recognize that cDNA microarray results may include false positives. Therefore, we validated these results further using real-time PCR and western blot analysis and confirmed that apoptotic-related genes were regulated at the mRNA and protein levels in response to ACFP treatment [
23].
The Bcl protein family plays an important role in regulating the apoptotic cell response. The Bcl proteins family is comprised of two classes: anti-apoptotic proteins (e.g., Bcl-xl) and pro-apoptotic proteins (e.g., Bax). Bcl-xl has been proposed as an oncogene based on its ability to block cell apoptosis and trigger tumorigenesis, while other proteins, such as Bak, are inhibitors of cell apoptosis [
24]. Importantly, other survival factors, such as Akt, can also suppress apoptosis in a transcription-independent manner by phosphorylating and inactivating components of the apoptotic machinery. Akt can prevent cell apoptosis by inhibiting and modifying Bcl-xl. Akt is associated with tumor growth through PI3K/Akt pathway by targeting IGF and/or IGF receptor [
25,
26]. Caspase-3 represents a core member of the apoptosis cascade pathway and is referred to as a death protease [
27]. Caspase-3 is the most important effector of cell apoptosis and is located downstream of executing proteases in the caspase cascade. The
CDC25C and
cyclinB1 genes are also related to cell apoptosis. One of the hallmarks of cancer is the lack of regulation in the cell cycle. The
CDC25C gene encodes for a tyrosine phosphatase protein that belongs to the Cdc25 phosphatase family and plays a key role in the regulation of cell division. CDC25C directs dephosphorylation of cyclin B-bound CDC2 (CDK1) and triggers entry into mitosis [
28]. Overexpression of cyclinB1 can lead to uncontrolled cell growth by deregulation binding and activation of cell cycle activating CDK kinases. Binding of Cdks can lead to phosphorylation of other substrates at inappropriate time and unregulated viability [
29].
In this study, ACFP can repress mRNA and protein expression of bcl-xl, cyclinB1, CDC25C and akt and enhance mRNA and protein expression of bax and caspase-3 by real time PCR and western blot analysis. The changes of bcl-xl, bax, akt, cyclinB1, CDC25C and caspase-3 expressions by real time PCR and western blot analysis were consistent with cDNA microarray results. Differential regulation of these pathways suggests that ACFP inhibits ovarian cancer cell growth by inhibiting anti-apoptotic protein Bcl-xl, Akt, CDC25C and cyclinB1 and promoting pro-apoptotic protein Bax and Caspase-3.
In the present study, we assessed the overall ACFP anti-cancer activity in primary ovarian cancer cells from ovarian tumor tissue of patients and analyzed ACFP effects on cancer-related gene expression and signal transduction pathways to determine whether transcription-related changes were related to ACFP-mediated anti-viabilities in primary ovarian cancer. However, we have not fully understood how the milk-derived fusion peptide bound the target cells. Fiedorowicz E, et al. reported that three bioactive peptides (opioid peptides) isolated from bovine caseins influenced the viability and cytokine secretion of human peripheral blood mononuclear cells by binding the μ-opioid receptor on the cytomembrane [
30]. In our previous studies, we observed that hexapeptide (PGPIPN) isolated from bovine β-casein reduced Bcl-2 expression and induced cell apoptosis by binding target molecules on SKOV3 [
11]. In the future, we plan to investigate whether ACFP binds cell specific receptor(s) in ovarian cell and, if so, determine the specific ACFP target molecule. We acknowledge that in our study, the number of ovarian tumor tissue samples (53 patients) and normal ovarian tissue samples (7 patients) was limited, emphasizing the need for future studies to extend on these findings. Together, our results provide a rationale for the future development of potent therapeutic peptides for the treatment of ovarian cancer.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
YQ and ZC conceived and designed the experiments. JZ, MZ, YT, JW, CW, FG and TL carried out the experiments. YQ, JZ and MZ analyzed the data, prepared the figures and wrote the manuscript. YQ and ZC reviewed the manuscript. All authors approved the final version of the manuscript.