New insights into the roles of the FOXO3 and P27Kip1 genes in signaling pathways

Abstract Background: The forkhead box O3 (FOXO3) and p27Kip1 are two important genes in breast cancer progression. In the present study we analyzed the effect of simultaneous FOXO3 silencing and p27Kip1 activation on breast cancer cell survival and the potential targets of these changes in cancer molecular pathways. Materials and methods: The present study involved the cloning of FOXO3a shRNA and p27Kip1 genes under the control of the bidirectional survivin promoter to down- and up-regulate FOXO3 and p27Kip1 genes, respectively. After transfection of the recombinant expression vector into the breast cancer cell line, the inhibition of cell growth was assessed by MTS and flow cytometry assays. Following the extraction of total mRNA and protein, the expression of target genes was evaluated by qPCR and Western blotting in both treated and untreated cell lines. Then, the downstream protein responses were examined by 2 D electrophoresis. The differentially expressed proteins were also identified by mass spectrometry. Results: Rates of cell proliferation were significantly inhibited in the transfected cell line 72 h post-transfection. Proteomic profiling of the cell line resulted in the identification of seven novel protein markers in breast cancer responsive to these changes in expression of FOXO3 and p27Kip1. The changes in expression of these markers suggested that certain signaling pathways contribute to the development of breast cancer. Conclusion: Simultaneous silencing of FOXO3 and activation of p27Kip1 in MDA-MB-231 cells caused alterations in the expression level of several genes involved in apoptosis, cell proliferation, cell cycle control, tissue invasion, drug resistance, and metastasis. It seems that the identified genes might serve as useful biomarkers for breast cancer.


Introduction
Breast cancer is an important global health concern and the second leading cause of cancer-related mortality among women (1). Omics studies have revealed several pathways and genes related to breast cancer (2), such as p27Kip1 and forkhead box O3 (FOXO3). The p27Kip1 protein interacts with cyclin E and cyclin-dependent kinase 2 (CDK2). In addition, the p27Kip1 gene arrests the progression of the cell cycle and leads to apoptosis, tumor suppression, and cell adhesion (3). Poor prognosis in patients with colorectal, gastric, pulmonary, and breast cancers is associated with a low expression level of the p27Kip1 gene (4).
The FOXO proteins, a subgroup of forkhead transcription factors, are involved in apoptosis, differentiation, stress response, cell cycle control, and cell metabolism (5). Overall, the FOXO3 gene is involved in many signaling pathways and makes major contributions to several processes, including cancer development (6). Therefore, the identification of FOXO3 inhibitors presents a promising strategy for future anticancer drugs (7). Proteomic analysis is generally applied in breast cancer serving two ends, namely the identification of novel molecular markers for breast tumor profiling and determination of the intracellular signaling pathways contributing to the growth of breast cancer cells (8). To our knowledge, there is no information on the effect of simultaneously decreasing FOXO3 gene and increasing p27Kip1 gene expression, respectively, to inhibit breast cancer. With this background in mind, the present study aimed to analyze the effect of the FOXO3 and p27Kip1 genes on the control of the breast cancer MDA-MB-231 cell line, using a bidirectional survivin promoter. Moreover, the changes in the protein profile of the transfected cell line were investigated in order to find cancer molecular markers.

Cell line, plasmid constructs, and transfection
The MDA-MB-231 cell line adopted in this study was supplied by the Pasteur Institute (Tehran, Iran). Cells were cultured in a high-glucose Dulbecco's modified Eagle's medium (ThermoFisher scientific, Waltham, MA, USA), supplemented with penicillin-streptomycin (100 units/mL) and fetal bovine serum (10% FBS; Gibco BI102-100) in a damp atmosphere (5% CO 2 ) at 37 C. The cells were transfected with a plasmid construct designed in a bidirectional promoter (dual targeting of FOXO3a shRNA and simultaneous induction of p27Kip1 gene).
In addition, an empty plasmid of pcDNA3.1 þ expression vector with EGFP was used as the control (Figure 1). The vectors were purchased from BioMatic, Toronto, Ontario, Canada. To verify the identity of the inserted sequences, all plasmids were examined through direct sequence analysis. Twenty-four hours before transfecting DNAs into the cells, some of the cells were applied in the growth medium without antibiotics. Transfection was performed as recommended by the manufacturer using Lipofectamine 2000 (Invitrogen, Waltham, MA, USA).

Cell viability assay
The viability of the cells (2-5 Â 10 3 /well) was tested in a 96well plate in a final volume of 200 lL/well, incubated in a CO 2 incubator (5% CO 2 ) at 37 C overnight. Then, 20 lL/well of MTS reagent (Promega, Fitchburg, WI, USA) were added to each well and incubated for 4 h at 37 C. The plate was shaken briefly, and then the absorbance of the treated and untreated cells was measured at 490 nm 24 h, 48 h, and 72 h after transfection, using a plate reader (BioTek, Winooski, VT, USA).

Quantitative real-time polymerase chain reaction assay
The extraction of total RNA was accomplished using TRIzol solution (Invitrogen). Total RNA (2 mg) was used for complementary DNA synthesis as recommended by the manufacturer (Thermo Scientific, Vilnius, Lithuania). Quantitative realtime reverse transcriptase polymerase chain reaction (qPCR) assay was performed using ABI 7300 RT-PCR systems (Applied Biosystems, Foster City, CA, USA) to determine the relative expression of the p27Kip1 and FOXO3a genes ( Table  1). The threshold cycle (Ct) value was estimated, and the REST 2009 software was used to assess the amplification plots. In addition, the comparative Ct (2 ÀDDCt ) method was applied to analyze the data (8).

Apoptosis assay
To identify apoptotic cells, the FITC Annexin V Detection Kit with propidium iodide (PI) (BioLegend, San Diego, CA, USA) was exclusively developed. After seeding the cells (2 Â 10 5 cells/mL) in tissue culture plates, they were incubated overnight, and then treated after 24 h, 48 h, and 72 h. In the next stage, the cells were washed with a cell-staining buffer, and then re-suspended in the annexin V binding buffer at a concentration of 0.25-1.0 Â 10 7 cells/mL. Subsequently, the PI solution (10 lL) and FITC annexin V (5 lL) were added. The cells were vortexed and incubated at 25 C for 15 min in darkness. Then, the annexin V binding buffer (400 lL) was added to the tubes. Finally, a BD Accuri TM C6 cytometer (Seattle, WA, USA) was used to perform flow cytometry.

Protein extraction and immunoblotting
A protein lysis buffer (100 mM Tris-HCl, 150 mM NaCl, 5% SDS, 1 mM dithiothreitol, 5 mM EDTA, pH of 7.4, 5% sodium deoxycholate, and 10% glycerol) was used for lysing the harvested cells collected via trypsinization. A protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO,USA) mixture was added to 1 mM of PMSF. The total protein in the resultant supernatant was used for Western blotting analysis. After separating the total protein using polyacrylamide gel electrophoresis, it was transferred to a nitrocellulose membrane, which was blocked with bovine serum albumin (5% w/v bovine serum albumin; Sigma-Aldrich) in TBST. Furthermore, the primary antibody was used against FOXO3a and p27Kip1 antibodies (Santa Cruz, Santa Cruz, CA, USA). Visualization of the bands was accomplished by means of the Enhanced Chemiluminescence Detection Kit (Parstous, Tehran, Tehran, Iran) following the manufacturer's protocol. In addition, the National Institutes of Health ImageJ program (Bethesda, MD, USA) was used to determine the band density.

Two-dimensional gel electrophoresis
Cell pellets were resuspended by adding an adequate amount of 2 D lysis buffer, containing 2 M thiourea, 1% dithiothreitol, 4% CHAPS, 7 M urea, 0.001% bromophenol blue, and 0.5% immobilized pH gradient (IPG) buffer (pH of 3-10; GE Healthcare, USA), followed by extracting the total protein. In addition, a 2 D Quant Kit (GE Healthcare) was used to measure the protein level with bovine serum albumin as the standard. To this end, after rehydrating the IPG strips (7 cm 3-10 NL, Immobilone Dry Strip; GE Healthcare) at a pH gradient of 3-10, 300 mg of each protein sample was loaded on the IPG strip. The 2 D electrophoresis was carried out based on our previously published method (9). For staining the gels, a colloidal Coomassie Brilliant Blue dye was used as described by Candiano et al. (10). A scanner (GE Healthcare Life Sciences, Chicago, IL, USA) was also employed to scan the stained gels. The analysis of the protein spots in the gel images was performed by the ImageMaster 2 D Platinum 6.0 (GE Healthcare Life Sciences). Finally, the intensities of the treated and untreated gels were compared, and their ratios were determined.

Protein identification via liquid chromatography-mass spectrometry
After cutting the protein spots into 1-mm 3 pieces, they were removed from the stained gel manually and added to tubes. The pieces were destained with an equal solution of ammonium bicarbonate and acetonitrile. Then, they were digested with trypsin (Promega, USA) at 37 C for a minimum of 16 h. Trifluoroacetic acid (TFA) 10% was also added to prevent digestion. Extraction of the gel pieces was accomplished using an equal solution of 0.1% TFA and acetonitrile (50 lL).
Supernatants from the pooled samples were dried, using Speed-Vac centrifugation (Thermo Scientific, Waltham, MA, USA). Then, 2% acetonitrile (10 lL) and 0.1% TFA were used for dissolving the samples. The Q-TRAP nLC-MS/MS system (Applied Biosystems) was used for all analyses, as described (11). Mass data were collected and analyzed using the Analyst software (Applied Biosystems). For finding the SwissProt-Trembl subdatabase, the MS/MS lists were searched, and Mascot 3 (version 2.2., Matrix Science, Boston, MA, USA) was used to blast the data lists against databases.

Data analysis
Data were analyzed in Prism, version 6.0 (GraphPad Software Inc, San Diego, CA, USA) using two-way ANOVA and Student's t test. The data were also presented as mean-± standard error of mean of at least three independent analyses. A p values less than 0.05 was considered statistically significant.

Combinatorial effects of induced p27Kip1 and shFOXO3a on cell viability
The p27Kip1 mRNA expression level was significantly higher, and FOXO3a mRNA expression significantly lower, in treated cells 72 h post-transfection in comparison with that found in controls (P < 0.05). Analysis of the qPCR results showed a 0.41-fold change in FOXO3a gene expression (down-regulation) and more than a three-fold change in p27Kip1 gene expression (up-regulation) 72 h after transfection ( Figure 2).
The expression of construct in MDA-MB-231 significantly reduced the growth of the cell line (P < 0.05) (Figure 4).
The effect of the construct on the rate of cell proliferation was not statistically significant 24 h post-transfection. However, it was significantly inhibited at 72 h after treatment (P < 0.05; Figure 5).

Flow cytometry
Apoptotic cells included both early and late apoptosis (Annexin Vþ/PI-and Annexin Vþ/PIþ, respectively), while Annexin V and PI were negative for viable cells. In the controls, the majority of the cells (99.9%) were viable and nonapoptotic at baseline. However, early apoptotic cells showed a significant increase after treatment, compared to the untreated cells after 72 h (from 5.0% to 36.4%; Annexin Vþ/ PI-; Figures 6(A, B)).

Effect of overexpression p27Kip1 and FOXO3a shRNA on downstream proteins
To identify the downstream responsive proteins associated with the changes in p27Kip1 and shFOXO3a expression at 72 h after transfection, the treated and untreated samples were compared by 2 D gel electrophoresis. There was differential expression of 10 protein spots in the breast cancer cell line (Figure 7). Out of the 10 proteins, 3 and 7 proteins showed up-regulated and down-regulated expression, respectively (Figure 7). Table 2 presents the mass spectrometry (MS) information of the appeared spots. Succinate dehydrogenase (SDH; T1/C1), albumin (ALB; T3/C3), and serine/threonine-protein kinase NLK (NLK; T4/C4) were significantly up-regulated, while tubulin alpha (TUBA1C; T2/C2),   tubulin beta chain (TUBB; T10/C10), b-actin (ACTB) with two isoforms (T8/C8 and T9/C9), and vimentin (VIM) with three isoforms (T5/C5, T6/C6, and T7/C7) were significantly downregulated (Figure 7). The theoretical and experimental masses and the isoelectric points of these differentially expressed proteins were also nearly the same, verifying the identified MS results (Table 2). Gene ontology (GO) analysis of 10 differentially expressed proteins consisted of proteinprotein interactions using the GeneMANIA (https://genemania.org) and KEGG (https://www.genome.jp/kegg/kegg2.html) databases (Table 3). Based on the result of two-way ANOVA, after the transfection of genes in MDA-MB-231 cells for 72 h, FOXO3 gene expression showed the greatest reduction (P < 0.05), while P27Kip1 gene showed the highest increase, compared to the control (Figure 7). The 2 D gel   Figure 7) that had the highest scores were selected and confirmed to be involved in breast cancer.

Discussion
There are studies regarding the independent effect of P27Kip1 and FOXO3a genes on the inhibition of breast cancer; however, they have reported a poor prognosis (12,13). The present study involved the design of a bidirectional expression vector suppressing FOXO3a and inducing p27Kip1 expression. Our results showed that the synergistic effects of p27Kip1 and shFOXO3a induced 36.4% apoptosis in cells. However, the overexpression of p27Kip1 alone has been reported to induce 12.73% apoptosis (12). Recently, proteomic methods have been used as a strategy for the detection of protein biomarkers (14). Expression of seven proteins changed after FOXO3a suppression and p27Kip1 induction (Figure 8). These proteins are all involved in apoptosis, metastasis, drug resistance, proliferation, and cell cycle control.
Three up-regulated proteins included SDHA, albumin (with cell source), and NLK, while four down-regulated proteins included TUBA1C, TUBB, ACTB with two isoforms, and VIM with three isoforms. The SDH is identified as a mitochondrial enzyme, involved in the Krebs cycle and electron transfer chain, and contributes to tumor suppression. This enzyme has four subunits, namely SDHA, SDHB, SDHC, and SDHD (15). Research over the past decades has shown an inverse relationship between albumin level and severity/risk of the progression of different cancers, including pulmonary, gastric, ovarian, and breast cancers. There are various mechanisms for the anticancer effects of albumin, such as DNA stabilization and duplication, calcium preservation and regulation, and antioxidant properties. In a study performed by Xiao-Jing Du et al., the growth of human cancer cells, such as breast cancer cells, was inhibited by high concentrations of albumin (16). The main cause of the reduction of the albumin level, especially in cancer patients, is the inhibition of albumin-regulating gene by tumor necrosis factor which results in mRNA expression reduction of approximately 90% in the liver (17). NLK, which belongs to the mitogen-activated protein kinase family, is a vital regulator of various cancer types. In a study conducted by Lv et al., NLK inhibited non-small-cell lung cancer through the Wnt signaling pathway (18). In addition, Huang et al. showed that NLK inhibits cellular proliferation in breast cancer and induces apoptosis. Similar studies confirm that NLK is a tumor suppressor gene in ovarian cancer (19). The theoretical molecular weight of NLK is about 58 kDa, while its experimental molecular weight is about 35 kDa, which can be attributed to the degradation of this protein. Actin and its proteins have a major functional and structural role in cell motility, endocytosis, cell adhesion and division, and maintenance of cell morphology. The expression of ACTB increases in a metastatic state (20). Accordingly, Chunmei Guo et al. reported that ACTB   Physical interaction SERPINH, MAGOH, EIFA3,   SPRM4, KRT18, UPP1,  ITGB4, BASP1, PPL,  URGDP, XPNPEP3  CFL1, CFL2, PFN1, CCT8,  TCP1, CCT3, CCT4,  RUVBL1, MYH11   TUBA4B, TUBB6, TUBB,  TUBA4A, TUBB4A,  TUBB4B, TUBB3,  TUBB2A, TUBB6,  TUBB2B, GAPDH,   THBS1, CLIP1, TCP11L1,  TUBB2A, YBX1,  TUBA1A, TUBA1B,  TUBA1C, TUBA1A,  TUBA4B, PNKD,  TNPO2, DNAJA4   SLC2A14, CPNE1, RAB14,  LRRC8A, HIGD1A,  SDHAF2, SDHAF2,  SDHC, SDHB, UBC   MYB, RANGAP1, ZHX3,  MAP3K7, TAB1, HIPK2,  LEF1, TP53, TCF7L2   F2, HP, ITIH1, AMBP,  AHSG, APOA2, RBP4,  FCGRT   Co-expression   SERPINH, TTN   CFL1, PFN1   TUBB6, TUBB, TUBA4A,  TUBB4B, TUBB3,  TUBB2A, TUBB6,  TUBB2B, GAPDH,  TUBA1A, TUBA1B,  TBCB, PFDN1, STMN1   TUBAB, TUBB2A, TUBG2,  YBX1, TUBA1B,  TUBA1C, TUBA4A,  TUBB3, TUBB4B,  TUBG1   RAB14 increased in many cancers of the liver, kidneys, colon, stomach, pancreas, esophagus, lung, breast, prostate, and ovaries (21). These tumors are associated with increased drug resistance and metastasis, especially in MDA-MB-231 cells (21). Tubulins consist of microtubules, with two alpha and beta subunits (approximate weight of 50 kDa). They play a large role in cellular activities, such as cellular mobility. Dysfunction of the microtubule network causes disturbance in apoptosis and differentiation (20). In addition, VIM is one of the most filamentous proteins, expressed in various types of cancers, such as prostate, gastrointestinal, central nervous system, pulmonary, and breast cancers. In a study carried out by Satelli, it was shown that the increased invasive/migratory potential of cancer cells is related to VIM overexpression (22). The VIM is generally considered as an indicator of poor prognosis (22).
In conclusion, the simultaneous silencing of FOXO3 and activation of p27Kip1 in MDA-MB-231 cells caused alterations in the expression level of several genes which are involved in cell apoptosis, cell proliferation, cell cycle control, tissue invasion, drug resistance, and metastasis. Therefore, it seems that these genes may serve as useful biomarkers for breast cancer.