ROS‑mediated hypomethylation of PRDX5 promotes STAT3 binding and activates the Nrf2 signaling pathway in NSCLC.

Deoxyribonucleic acid (DNA) epigenetic modification has been linked to specific sequences of CpG islands and plays roles in the progression of lung cancer. In this study, it was found that peroxiredoxin‑5 (PRDX5) was highly expressed in non‑small cell lung cancer (NSCLC) tissues; however, its specific regulatory mechanisms and functions in NSCLC remain unknown. The present study therefore explored the regulatory mechanism of PRDX5 under conditions of oxidative stress (OS) in NSCLC. The results revealed that 79 of 121 NSCLC patients exhibited demethylation in the PRDX5 promoter region, which was related to the tumor, node and metastasis (TNM) stage (P=0.027). PRDX5 messenger ribonucleic acid (mRNA) expression positively correlated with the demethylation status of the promoter region. The results of bisulfite sequencing polymerase chain reaction (BSP) revealed lower demethylation frequencies in H1299 cells treated with 0 µM H2O2, but maximum demethylation following treatment with 100 µM H2O2. Using chromatin immunoprecipitation (ChIP) and luciferase detection assays, the effective binding of STAT3 to the transcriptional binding sites of the PRDX5 promoter region was confirmed (2 sites confirmed: Site 1, ‑444 to ‑434 bp; and site 4, ‑1,417 to ‑1,407 bp). STAT3 knockdown significantly decreased the protein expression of PRDX5, while the overexpression of STAT3 significantly increased the protein levels of PRDX5. When PRDX5 was overexpressed in lung cancer cells under conditions of OS, the levels of the epithelial‑mesenchymal transition (EMT) biomarkers, E‑cadherin and vimentin, were significantly decreased and increased, respectively. By contrast, PRDX5 knockdown resulted in significantly increased E‑cadherin and decreased vimentin protein expression levels. Ultimately, when PRDX5‑small interfering RNA (siRNA) or pcDNA3.1‑PRDX5 expression vector were constructed and transfected into H1299 cells pre‑treated with 100 µM H2O2, the nuclear factor (erythroid‑derived 2)‑like 2 (Nrf2) signaling pathway was inhibited or activated. All these results suggested that the reactive oxygen species (ROS)‑mediated hypomethylation of PRDX5 enhanced STAT3 binding affinity with the promoter region, and resulted in the promotion of cell migration and invasion, as well as in the activation of the Nrf2 signaling pathway in NSCLC. The demethylation status of the PRDX5 promoter may thus be used as an epigenetic biomarker in NSCLC. STAT3/PRDX5 signaling may also prove to be a potential strategy for the treatment of this type of cancer.

Introduction

Currently, deaths from non-small cell lung cancer (NSCLC) account for 85% of lung cancer-related deaths worldwide (1). The disease is associated with a high degree of malignancy, and early and extensive metastasis, followed by a poor clinical prognosis (2). As lung cancers are the most lethal tumors in North America, their diagnosis and treatment are attracting increasing attention. Therefore, research on the molecular mechanisms of NSCLC is crucial for the investigation of the mechanisms responsible for the development of this tumor and for the identification of treatment strategies. In recent years, deoxyribonucleic acid (DNA) methylation, as one of the most common epigenetic modifications, has been linked to specific sequences of the CpG islands and plays an important role in the progression of lung cancer (3-6). Generally, DNA methylation results in the inactivation of gene expression (7,8), while demethylation in the promoter region activates gene transcription levels (9,10).

Reactive oxygen species (ROS), which are normally produced in and eliminated from all types of cells, exert physiological and pathological effects (11). In a tumor micro-environment (TME) of ischemia-hypoxia, ROS levels are commonly higher than in normal environments, which is crucial for the study of tumorigenesis and treatments (12,13). Peroxiredoxins (PRDXs) as a class of antioxidant enzymes that include 6 members, PRDX 1-6, which play a role in regulating cell proliferation, differentiation and apoptosis by modulating ROS (14). Recent studies have indicated that PRDXs participate in tumor progression, and upregulated levels of PRDXs have been suggested to be responsible for tumor prognosis and drug resistance (15-17). In the present study, it was found that PRDX5 was highly expressed in NSCLC tissues, while of note, there were obvious methyl islands in the promoter region. Therefore, it was hypothesized that the hypomethylation of the PRDX5 gene promoter region may be related to the pathogenesis of NSCLC.

Signal transducers and activators of transcription (STATs) are a family of transcription factors (TFs) that can be activated by different cytokines and act as carriers during interaction between cytokines and receptors, maintaining the intracellular transmission of signals. Different STATs have their own specific functions; e.g., STAT4 and STAT1 induce T-helper 1 cell (Th1) differentiation, whole STAT6 mediates Th2 cell differentiation (18,19). STAT3, as the first-discovered member of the STAT family, was first defined as an acute response protein participating in various physiological or pathological processes; it is widely expressed in the human body and can be activated by a number of types of cytokines or by various stressors (20). Recent studies have demonstrated that the abnormal expression of STAT3 is also found in a variety of tumors (21-23). However, the mechanisms through which STAT3 promotes tumor progression remain unclear.

The present study aimed to explore the methylation status of the PRDX5 gene promoter region in NSCLC. It was deter-mined that the ROS-mediated hypomethylation of PRDX5 promoted STAT3 binding. Furthermore, the results revealed that the upregulation of PRDX5 mediated by STAT3 activated the nuclear factor (erythroid-derived 2)-like 2 (Nrf2) signaling pathway.

Materials and methods

Patients and tissue samples

NSCLC and adjacent non-cancerous tissues were obtained from 121 patients with NSCLC who underwent surgical resection at the Affiliated Hospital of Nantong University (Nantong, China) between January, 2006 and January, 2011. Detailed clinicopathological parameters of the patients are provided in Table I. The study protocol was approved by the Ethics Committee of Affiliated Hospital of Nantong University. Written informed consent was obtained from patients prior to collecting samples.

Table I

Clinicopathological characteristics and PRDX5 methylation in 121 patients with NSCLC.

Clinicopathological parameters	No. of patients	PRDX5 methylation		P-value
		Methylated (n=42)	Unmethylated (n=79)
Age (years)				0.879
<60	53	18	35
≥60	68	24	44
Sex				0.830
Male	56	20	36
Female	65	22	43
Clinical TNM stage
I-II	73	31	42	0.027a
III-IV	48	11	37
Lymph node involvement				0.438
Negative	49	19	30
Positive	72	23	49
Differentiation				0.830
Well and moderate	65	22	43
Poor	56	20	36

Indicates a statistically significant difference (P<0.05). PRDX5, peroxiredoxin-5; NSCLC, non-small cell lung cancer.

Cell lines and cell culture

The lung cancer cell lines, A549 (SCSP-503), H1299 (SCSP-589) and H157 (ATCC® CRL5802™), and the normal bronchial epithelial cell (EC) line, 16HBE (ATCC® PCS-300-040™), were purchased from the Shanghai Cell Bank of the Chinese Academy of Sciences or the American Type Culture Collection (ATCC). Following cell line authentication using short tandem repeat (STR) profiling, the cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc.), 50 U/ml penicillin and 50 µg/ml streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.); they were then incubated in a humidified atmosphere at 37°C and 5% CO2.

Establishment of model of oxidative stress (OS) induced by H2O2

H2O2 (Sigma-Aldrich; Merck KGaA) was used to the establishment of a model of OS. In brief, the lung cancer cell lines, A549, H1299 and H157, were pre-treated with various concentrations of H2O2 (0, 50, 100 or 200 µM) for 30 min of stimulation.

Methylation-specific and bisulfite sequencing polymerase chain reaction (MSP and BSP)

The CpG island online prediction software (http://www.urogene.org/index.html) was used to predict the CpG islands of the PRDX5 gene promoter. MSP was used to measure the methylation status of CpG islands in the PRDX5 gene promoter region. The methylated (M) band indicated that CpG sites were methylated, while the unmethylated (U) band indicated unmethylated status. The patients with NSCLC were divided into the PRDX5-methylated and -unmethylated group (U/M ≥1, unmethylated; U/M <1, methylated). The demethylation ratio was calculated as U/(M + U). BSP was used to verify the methylation status of these islands. Extracted DNA samples isolated from NSCLC tissues and cells were modified with bisulfite reagents (Zymo Research) as per the manufacturer's instructions, which specified a change from unmethylated cytosine to thymine. The MSP primer pairs for PRDX5 were as follows: PRDX5-MSP-M forward, 5′-GGG GTT GAA TTT TAT AGG GTA GAT AC-3′ and reverse, 5′-GAC CTA ACG AAA ATT TAT ACG ACG A-3′; PRDX5-MSP-U forward, 5′-GGG GTT GAA TTT TAT AGG GTA GAT AT-3′ and reverse, 5′-AAC CTA ACA AAA ATT TAT ACA ACA AC-3′. For BSP, bisulfite-treated DNA was amplified by PCR using the following primers: PRDX5-BSP forward, 5′-GGG TTG AAT TTT ATA GGG TAG ATA-3′ and reverse, 5′-CTA CTT ACC CAC AAT CTA CTA AAC TC-3′. PCR products were purified using the Wizard SV Gel and PCR Clean-up System and then cloned into a pGEM-T Easy Vector System (both from Promega Corporation). A total of 8 colonies were randomly selected for the extraction of plasmid DNA using a Promega Spin Mini kit (Promega Corporation) and the DNA was sequenced on an ABI 3130 Genetic Analyzer (Applied Biosystems; Thermo Fisher Scientific, Inc.).

Reverse transcription-quantitative PCR (qRT-PCR)

RT-qPCR was used to determine PRDX5 mRNA expression. Total RNA was extracted from tissues and cells using TRIzol reagent (Takara Bio, Inc.) as per manufacturer's instructions. First-strand complementary deoxyribonucleic acid (cDNA) was then synthesized using a PrimeScript RT Reagent kit and SYBR-Green I (Takara Bio, Inc.) was used for the qPCR analysis of the cDNA as per the manufacturer's instructions. The PCR thermocycling conditions were as follows: 95°C for 5 min, (95°C for 15 sec, 58°C for 45 sec, and 72°C for 60 sec) 40 cycles, and 72°C for 1 min. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as the internal control. The primer sequences for PRDX5and GAPDH were as follows: PRDX5 forward, 5′-CCA ATC AAG ACA CAC CTG CC-3′ and reverse, 5′-TCT TGA GAC GTC GAT TCC CA-3′; GAPDHforward, 5′-GAA GGT GAA GGT CGG AGT C-3′ and reverse, 5′-GAA GAT GGT GAT GGG ATT TC-3′. Quantities were calculated using the comparative 2−ΔΔCq method (24). All steps were performed in triplicate.

Western blot analysis

Tissues and cells were lysed using radioimmunoprecipitation assay (RIPA) buffer containing protease inhibitors (Promega Corporation) to obtain total protein. The protein concentrations were then determined using a bicinchoninic acid (BCA) Protein Assay kit (Bio-Rad Laboratories, Inc.). Protein samples (containing 30 µg protein) were separated on 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels and then transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked with 5% skimmed milk for 1 h at room temperature and then incubated overnight with primary antibodies diluted to 1:1,000 at 4°C. The following day, the membranes were further incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for 1 h (1:2,000; cat no. 7056, Cell Signaling Technology, Inc.). GAPDH (1:1,000; cat no. A00227-1, Boster Biological Technology, Ltd.) was used as an internal reference. The immunoreactive proteins were detected using an enhanced chemiluminescence kit (Cell Signaling Technology, Inc.). The blots were then scanned on an Odyssey Fc Imaging System (Li-COR Biosciences) and the grayscale value was used for statistical analysis. The antibodies against PRDX5(1:1,000; cat no. ab180587), STAT3(1:1,000; cat no. ab76315), E-cadherin (1:1,000; cat no. ab40772), β-actin (1:1,000; cat no. ab8226), vimentin (1:1,000; cat no. ab92547), Nrf2(1:1,000; cat no. ab89443) and reduced nicotinamide adenine dinucleotide phosphate (NADPH) dehydrogenase [quinone] 1 (NQO1) (1:1,000; cat no. ab80588) were all from Abcam.

Construction and transfection of plasmids and small interfering RNA (siRNA)

The overexpression plasmids (pcDNA3.1-STAT3 and PRDX5) and the control (pcDNA3.1-vector), siRNAs against STAT3 and PRDX5 [si-STAT3(CGT CAT TAG CAG AAT CTC ATT) and si-PRDX5(GGA ATC GAC GTC TCA AGA GGT), respectively] and corresponding negative-control (NC) siRNAs (si-NC, GCA GAT AGG TAG GCG TTA T) were all designed and synthesized by Guangzhou RiboBio Co., Ltd. The process of transient cell transfection was conducted using standard methods as per manufacturer's instructions. Briefly, cells were maintained in medium with fetal bovine serum (FBS, 10%) until the confluence reached 70-80%, and all the oligonucleotides (RNA and DNA) were then transfected into the cells using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.). following 48 h of transfection, the transfection efficiency was detected by western blot analysis.

Chromatin immunoprecipitation (ChIP) assay

In order to explore the transcription factors that may be involved in the regulation of PRDX5 gene expression, the online software Jaspar (http://jaspar.genereg.net) was used to predict the transcription factors that may bind to the promoter region. ChIP assay was then performed using a ChIP kit (magnetic beads; Cell Signaling Technology, Inc.) as per the manufacturer's instructions. Briefly, DNA-protein complexes were cross-linked with 1% formaldehyde, and 1% SDS Lysis Buffer was then added followed by sonication. The used antibody to immunoprecipitated anti-STAT3(1:20, ab76315; Abcam) or normal mouse immunoglobulin G (IgG, 1:200, cat. no. 554002, BD Biosciences) was added at 4°C for 12 h. DNA was then purified out of the antibody-protein-DNA complex and used for PCR. Specific ChIP primers for detailed sequences of the PRDX5 promoter were as follows: Site 1 forward, TAT TGG ATA GCC AGG AGA ACC and reverse, GGA ACC TCC TGC TGA GAC G (131 bp); site 2 forward, ATG TCG CCG CAC AAA CT and reverse, CCC ACA AAC ACG AGA AGT TCC (147 bp); site 3 forward, GAA ACC GCT TTT GGT TTT AAA C and reverse, CCA ACC CTT GAC CCA ATG AC (94 bp); site 4 forward, CTC AGG GGG TAG GAG AGC A and reverse, GGT TTA AAA CCA AAA GCG GT (182 bp); and site 5 forward, CTC TCC TCC CCC TCC TAG GG and reverse, TGG CCT CCA TCC CCT TCC (188 bp). The following PCR conditions were used: 95°C for 5 min, 95°C for 30 sec, 55-60°C for 30 sec, and 72°C for 30 sec for a total of 30 cycles, and then 72°C for 10 min. The acquired products were observed using agarose gel electrophoresis.

Luciferase reporter assay (LRA)

The sequences (760 and 640) from upstream to the start of the PRDX5gene were cloned into pGL3 luciferase reporter vector (Promega Corporation), and were transfected into the cells with Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) together with internal reference vectors, phRL-TK. Luciferase activities were detected using a Dual Luciferase Assay System (Promega Corporation) at 48 h following transfection in triplicate, and Renilla luciferase activity served as the internal control.

DNA methylation in vitro

PGL3-640 and PGL3-760 were treated with DNA SssI methylase (New England Biolabs) for 4 h at 37°C; and these plasmids were incubated similarly but without SssI methylase (unmethylated control). The plasmids were then further purified with a PCR product clean-up kit (Axygen). The unmethylated or methylated activities of PGL3-640 and PGL3-760 were measured as per the above methods.

Cell migration and invasion assay

Cell migration and invasion were detected using Transwell chambers (Corning, Inc.) coated with or without Matrigel (no. 356234; BD Biosciences). A total of 2×104 cells in 100 µl were added to the chambers with serum-free DMEM, while the lower chambers were filled with culture medium containing 10% FBS in a humidified 5% CO2 atmosphere at 37°C. After 48 h, the invading or migrating cells were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet (Sigma-Aldrich; Merck KGaA) for 20 min at room temperature. Cells were counted under a micro-scope (Leica DM2500, Leica Microsystems, Inc.) at ×200 magnification.

Statistical analysis

SPSS software version 17.0 (IBM Corp.) was used for statistical analysis. All data are expressed as the means ± standard deviation (SD) in triplicate. A paired or independent sample t-test was used to analyze differences between 2 groups. For the comparison of multiple groups, differences were analyzed by one-way analysis of variance (ANOVA). When ANOVA detected significant differences, the data of the variables of each experimental group were compared with that of the control group using a Dunnett's post hoc test. An χ2 test was used to examine the association between the promoter methylation of PRDX5 and the patient clinicopathological parameters. Correlation analysis was performed using Spearman's correlation coefficient (SCC). A value of P<0.05 was considered to indicate a statistically significant difference.

Results

PRDX5 upregulation in NSCLC tissues is associated with CpG island demethylation in the promoter region

First, to determine whether PRDX5 upregulation in NSCLC tissues was associated with the demethylation of CpG islands in the promoter region, the CpG island online prediction software (http://www.urogene.org/index.html) was searched and it was found that 3 CpG islands existed near the transcription start site (Fig. 1A). Since the promoter region upstream of TSS may have more transcription factors binding to start gene transcription, and the length of the first CpG island is longer, the 1st one was selected to design primers. Subsequently, 121 pairs of NSCLC tissues and adjacent non-cancerous tissues were analyzed to determine the methylation status of CpG islands in the PRDX5 promoter region by MSP. The results revealed that 79 of the 121 patients with NSCLC exhibited PRDX5 promoter demethylation, and that the demethylation was associated with tumor, node and metastasis (TNM) stage (P=0.027), but not with age, sex, lymph nodes or differentiation (Table I). A total of 8 representative cases of MSP results are presented in Fig. 1B. All the 121 tumor tissues exhibited a significantly higher demethylation ratio in the PRDX5promoter region compared with that in adjacent non-cancerous tissues (P<0.05; Fig. 1C). Additionally, the results of RT-qPCR revealed that PRDX5 mRNA expression was upregulated in NSCLC tissues compared with adjacent tissues (P<0.05, Fig. 1C). As shown in Fig. 1D, PRDX5 mRNA expression positively correlated with the demethylation status of the promoter region.

Figure 1

Demethylation status of the CpG islands in the PRDX5 gene promoter in NSCLC. (A) Online prediction of CpG islands around the PRDX5 gene promoter region. The light blue areas represent the potential CpG islands. (B) Representative methylation analysis results by MSP in NSCLC tissues and adjacent normal lung tissues. M, methylation; U, unmethylation. (C) Relative expression of PRDX5 mRNA significantly increased in tumor tissues than that in normal ones and demethylation ratio of PRDX5 gene promoter showed significantly higher in tumor tissues than that in the adjacent normal ones. *P<0.05, vs. PRDX5 mRNA or demethylated ratio in the respective tumor or normal group. (D) Correlation analysis between the demethylation ratio by MSP and PRDX5 mRNA expression by qPCR. PRDX5, peroxiredoxin-5; NSCLC, non-small cell lung cancer; MSP, Methylation-specific PCR.

Hypomethylation of specific CpG sites within the PRDX5 promoter region promotes transcriptional activities under conditions of OS

To obtain further details of the demethylation status of specific CpG sites within the PRDX5 promoter region under conditions of OS, an in vitro model of ROS was induced using H2O2 (Sigma-Aldrich; Merck KGaA). A 193 bp length of PCR products (-760 to -568 bp) was analyzed following sodium bisulfite treatment as part of the BSP method. The lung cancer cell lines, A549, H1299 and H157, were pre-treated with various concentrations of H2O2 (0, 50, 100 or 200 µM). Following 30 min of stimulation, the results of RT-qPCR for relative PRDX5 mRNA expression revealed the significant upregulation of PRDX5 expression in the 100 µM H2O2 group in the 3 cell lines (Fig. 2A, upper panel); similar results were obtained for protein expression in the H1299 cells (Fig. 2A, bottom panel). The H1299 cells that had been treated with 100 µM H2O2 were then selected for BSP. As shown in Fig. 2B, the sequencing region contained 13 CpG sites from -760 to -568 bp. The results of BSP revealed lower demethylation frequencies in H1299 cells treated with 0 µM H2O2, but maximum demethylation in those treated with 100 µM H2O2 (Fig. 2C).

Figure 2

Hypomethylation of specific CpG sites within the PRDX5 promoter promotes transcriptional activities under oxidative stress. (A) Relative PRDX5 mRNA levels (upper panel) in different cell lines and protein expression in H1299 cells (lower panel) that were pre-treated with various concentrations of H2O2 (0, 50, 100 and 200 µM). *,#,&P<0.05, vs. the respective cell line in the untreated groups. (B) Schematic BSP regions (from -760 to -568 bp) showing locations of the 13 CpG sites in the PRDX5 gene promoter region. (C) Bisulfite sequencing analysis of the demethylation status of the PRDX5 gene promoter under different conditions of oxidative stress. Each oval indicates clones from different cell lines. Vertical row represents the number of clones sent for bisulfite sequencing and horizontal row represents CpG numbers. Black circles represent the methylated clones and white circles indicate an unmethylated status. (D) Treatment with SssI methylase in 2 luciferase constructs in vitro. *P<0.05, vs. SssI methylase (-) group. Independent data results were repeated 3 times. PRDX5, peroxiredoxin-5; BSP, bisulfite sequencing PCR.

Finally, to determine which CpG sites were responsible for the demethylation-related activation of the PRDX5 gene under conditions of OS, two PRDX5 gene promoter regions (PGL3-640 and PGL3-760) were constructed; these were then treated with SssImethylase in vitro and transfected into H1299 cells that had been pre-treated with 100 µM H2O2 (Fig. 2D). Compared with the treated promoter constructs, the untreated constructs exhibited a significantly greater demethylation and promoter activity. No marked differences in the promoter activity of PGL3-640 or PGL3-760 were observed between the SssImethylase-treated and untreated groups. These results indicated that the region of the CpG sites from -640 to -568 bp may play as an important role in regulating PRDX5 gene transcription.

Promoted binding of STAT3 to the PRDX5 gene promoter region due to DNA demethylation

To explore related TFs that may be involved in the regulation of PRDX5 expression, the TFs that could potentially bind to the TF binding site near the promoter region were predicted using the website, http://jaspar.genereg.net/ (Fig. 3A). A total of 5 potential STAT3 binding sites were screened near the PRDX5 promoter region (site 1, -444 to -434; site 2, -502 to -492; site 3, -1,381 to -1,371; site 4, -1,417 to -1,407; site 5, -1,941 to -1,931 bp). Subsequently, the actual binding of STAT3 to the transcriptional binding sites of the DNA promoter region under conditions of OS was examined by ChIP assay. The results revealed that STAT3 could obviously bind to sites 1, 3 and 4, but not to sites 2 or 5 (Fig. 3B and C).

Figure 3

Promoted binding of STAT3 to the PRDX5 gene promoter region due to DNA demethylation. (A) The predicted potentially binding sequence and binding position of STAT3 binding to the transcription factor binding site near the promoter region of PRDX5. TSS, transcription start site. (B) ChIP assay showing the direct binding of STAT3 to the PRDX5 gene promoter in H1299 cells. The ChIP-enriched DNA fragments of the PRDX5 gene promoter using IgG and an anti-STAT3 antibody were amplified by PCR. Total input was used as a positive control. (C) Sequential mutation analysis identified the direct STAT3 binding sites in the PRDX5 gene promoter region by determination of relative luciferase activities. Data are presented as the means ± SD. *P<0.05, **P<0.01, vs. the PGL3-WT group. PRDX5,peroxiredoxin-5; STAT3, signal transducer and activator of transcription 3.

Finally, to further verify the effective binding sites indicated by the results of ChIP assay, mutant plasmids that were directed against each of sites 1, 3 and 4 were constructed. The results of luciferase detection revealed a significant decrease in the PGL3-MT1 and PGL3-MT4 regions (Fig. 3C). This indicated that sites 1 and 4 were the effective TF binding sites for PRDX5 gene transcription.

STAT3-regulated PRDX5 signaling affects the migration and invasion of lung cancer cells under conditions of OS

To further demonstrate that STAT3 was involved in regulating PRDX5 expression, STAT3 siRNA and the expression plasmid, pcDNA3.1-STAT3 were constructed. These were then transfected each into H1299 cells pre-treated with 100 µM H2O2. After 48 h, the results of western blot analysis revealed that STAT3 protein expression was significantly decreased in the STAT3 siRNA group compared with the scramble group, while protein expression levels were significantly increased in the group transfected with pcDNA3.1-STAT3 compared with the control vector group. In addition, STAT3 knockdown significantly decreased the protein expression of PRDX5, while the overexpression of STAT3 significantly increased the protein level of PRDX5 (Fig. 4A and B).

Figure 4

STAT3 mediates the regulation of PRDX5 expression. (A) STAT3 protein expression significantly increased in the cells when transfected with pcDNA3.1-STAT3 and significantly upregulated the PRDX5 protein expression level (*P<0.05, vs. control vector group). Data are presented as the means ± SD. (B) STAT3 protein expression significantly decreased in the cells when transfected with STAT3 siRNA compared with that of cells transfected with the scramble and significantly downregulated the PRDX5 protein expression level (*P<0.05, vs. scramble group). Data are presented as the means ± SD. (C) PRDX5 protein expression levels significantly increased or decreased when transfected with pcDNA3.1-PRDX5 or siRNA, respectively (*P<0.05, vs. control vector or scramble group). PRDX5, peroxiredoxin-5; STAT3, signal transducer and activator of transcription 3.

To further clarify the effects of PRDX5 on the migration and invasion of lung cancer cells under conditions of OS, PRDX5 siRNA and the corresponding overexpression plasmid pcDNA3.1-PRDX5 were constructed and transfected into H1299 cells pre-treated with 100 µM H2O2. The results indicated that the PRDX5 protein levels significantly decreased following transfection with PRDX5 siRNA (Fig. 4C). In addition, the cell migratory and invasive abilities were also significantly suppressed following the knockdown of PRDX5 (Fig. 5). By contrast, PRDX5 protein expression significantly increased following transfection with pcDNA3.1-PRDX5 (Fig. 4C), and the cell migratory and invasive abilities were also increased (Fig. 5). At the same time, when PRDX5 was overexpressed in lung cancer cells under conditions of OS, the levels of the epithelial-mesenchymal transition (EMT) biomarkers, E-cadherin and vimentin, were significantly decreased and increased, respectively (Fig. 6A). Conversely, PRDX5 knockdown resulted in significantly increased E-cadherin and decreased vimentin protein expression levels (Fig. 6B). The above-mentioned results indicated that PRDX5 may be regulated by STAT3, and that it affected the migratory and invasive abilities of lung cancer cells under conditions of OS, promoting the EMT phenotype.

Figure 5

Effects of PRDX5 on cell migration and invasion. (A) Upregulation of PRDX5 significantly promoted cell migration, while the downregulation of PRDX5 yielded opposite results (*P<0.05, vs. control vector or scramble group). Data are presented as the means ± SD. (B) Upregulation of PRDX5 significantly promoted cell invasion, while downregulation of PRDX5 yielded opposite results (*P<0.05, vs. control vector or scramble group). Magnification, ×200. Data are presented as the means ± SD. PRDX5, peroxiredoxin-5.

Figure 6

PRDX5 affects EMT phenotypes and the oxidative stress signaling pathway. (A) Overexpression of PRDX5 significantly downregulated E-cadherin expression, while it upregulated the vimentin protein expression level (*P<0.05, vs. control vector group). Data are presented as the means ± SD. (B) Knockdown of PRDX5 completely reversed the above-mentioned results (*P<0.05, vs. scramble group). Data are presented as the means ± SD. (C) Overexpression of PRDX5 significantly increased the protein expression level of Nrf2 and NQO1 (*P<0.05, vs. control vector group). Data are presented as means ± SD. (D) Knockdown of PRDX5 significantly downregulated the expression of Nrf2 and NQO1 protein (*P<0.05, vs. scramble group). Data are presented as the means ± SD. PRDX5, peroxiredoxin-5; Nrf2, nuclear factor (erythroid-derived 2)-like 2; NQO1, nicotinamide adenine dinucleotide phosphate (NADPH) dehydrogenase [quinone] 1.

PRDX5 activates the Nrf2 signaling pathway in lung cancer cells under conditions of OS

To verify that PRDX5 can activate Nrf2 signaling under conditions of OS, PRDX5 siRNA and pcDNA3.1-PRDX5 were transfected into H1299 cells pre-treated with 100 µM H2O2. The expression of key molecules of OS, Nrf2 and NQO1, was significantly upregulated when the cells were transfected with pcDNA3.1-PRDX5 (Fig. 6C). By contrast, PRDX5 knockdown resulted in significantly decreased protein expression levels of Nrf2 and NQO1 (Fig. 6D). These results indicated that PRDX5 may be involved in the activation of the Nrf2 signaling pathway in lung cancer cells under conditions of OS.

Discussion

Lung cancer is currently a malignancy with an unclear molecular mechanism, particularly as regards NSCLC (25). As NSCLC progresses, ROS levels are increased in cancer cells compared with in normal cells, mainly due to the abnormal metabolic level in tumors; this leads to the overproduction of ROS in ischemic hypoxic environments. As a peroxiredoxin family member, PRDX5 plays an important role in maintaining intracellular ROS or peroxide levels induced by cytokines (26). In the present study, H1299 cells were pre-treated with 100 µM H2O2 to establish the ROS model in vitro. Although this induced ROS and the upregulated endogenous expression of PRDX5, this does not have an impact on the results of subsequent experiments, as it is the result of the comparison between the two groups. As the results were based on the comparison between the 2 groups, and are from an epigenetic perspective, it was clarified that the upregulated expression of the PRDX5 gene mainly participated in the migration and invasion of lung cancer cells under conditions of OS and activated the Nrf2 signaling pathway.

DNA methylation is one of the most common means of epigenetic regulation. For example, it plays an important role in the modulation of cancers and inflammation or tissue-damaging pain (27-29). In the present study, it was found that PRDX5 was upregulated due to the demethylation of its promoter region, and at the same time, that different OS levels led to varying degrees of demethylation and PRDX5 mRNA expression. This demonstrated that OS promoted PRDX5 expression by demethylating the promoter region. There were 2 possibilities: The one is that the number of samples was still not sufficient to explain the problem. The other is that it was really no association between the two.

Using ChIP assay, it was verified that STAT3 functioned as a TF in binding to the PRDX5 gene promoter region. In addition, to the best of our knowledge, the present study also demonstrated for the first time that the ability of STAT3 to bind to this region was markedly enhanced in NSCLC under conditions of OS. This enhanced affinity promoted PRDX5 expression, while the mutation of the binding site between the PRDX5 gene promoter region and STAT3 resulted in a significantly decreased PRDX5expression. From an epigenetic perspective, these results may have been caused by the hypomethylation of the PRDX5gene promoter region, leading to enhanced affinity between binding sites. Choi et al (30) reported that the overexpression of PRDX5 suppressed the TGF-β induced upregulation of STAT3 phosphorylation. Perhaps under certain specific conditions (tumor or non-tumor conditions), PRDX5 and STAT3 may have a negative feedback regulation, forming a negative feedback regulation loop to perform specific functions. The present study did not notice this point; thus, the authors aim to continue to explore the association between the 2 genes in the future. In addition, it is hypothesized that the mutation of all 3 binding sites will not disrupt STAT3 binding completely, for the binding sites are only acquired by prediction and identification by experiments; it can also not be ruled out that there may be other binding sites.

EMT is an important biological process in the migration and invasion of malignant tumor cells derived from ECs (31). It plays an important role in embryonic development (32), chronic inflammation (33), cancer metastasis (34) and a variety of fibrotic diseases (35), the main characteristic of which is a decrease in the expression of cell adhesion molecules (CAMs), such as E-cadherin. The cytoskeleton of cytokeratin is transformed into vimentin, which has the morphological characteristics of mesenchymal cells. Ahn et al reported that PRDX5 promoted EMT in colon cancer (36). In contrast to this aforementioned study, the present study found that the hypomethylation of PRDX5 promoted STAT3 binding, and promoted cell migration, invasion and EMT progression, which manifested as E-cadherin downregulation and vimentin upregulation. By contrast, the lower expression of PRDX5 suppressed cell migration, invasion and EMT. These results indicated that PRDX5 may affect cell migration and invasion by activating EMT.

The Nrf2 antioxidant-responsive element (ARE) pathway is one of the most important endogenic anti-OS pathways to be discovered to date. There is evidence to indicate that when activated, this pathway can inhibit the degradation of Nrf2 protein mediated by ubiquitin, stabilize the concentration of Nrf2 protein in cytoplasm, and enhance the transcriptional activity of Nrf2 protein (37,38). The results of the present study revealed that PRDX5 overexpression under conditions of OS significantly increased the protein levels of Nrf2 and NQO1, which are key proteins of the Nrf2/ARE signaling pathway. These results may provide a strategy for the treatment of NSCLC.

In conclusion, the present study demonstrated that the ROS-mediated hypomethylation of PRDX5 enhanced STAT3 binding affinity with the PRDX5 gene promoter region, and promoted cell migration and invasion, as well as the activation of the Nrf2 signaling pathway in NSCLC. The demethylation status of the PRDX5 promoter may be used as an NSCLC epigenetic biomarker and STAT3/PRDX5 signaling may prove to be a potential target in the treatment of NSCLC.