Association Between NOX4 And Nrf2 Genes in Non-Small-Cell Lung Carcinoma: A Case-Control Study

Background: Epithelial malignancy in lung cancer, which is initiated with myofibroblast differentiation and remodeling, promotes hypoxia and intracellular ROS generation most affected by the prototypical enzyme, NADPH oxidase 4 (NOX4). In addition, nuclear factor erythroid 2-related factor 2 (Nrf2) acts as a critical transcription factor by stimulating antioxidant proteins as redox homeostasis regulators. The aim of this study was to investigate a possible correlation between lung tissue NOX4 and Nrf2 genes (NOX4 and Nrf2) mRNA expression and bronchoalveolar lavage fluid (BALF) protein expression in non-small-cell lung carcinoma (NSCLC) patients. Methods: Samples from 25 patients with various NSCLC types and stages and 20 healthy controls were collected. NOX4 and Nrf2 mRNA were measured by qRT-PCR, and protein by western blot analysis. Results: NOX4 mRNA and protein expression was significantly up-regulated in NSCLC patients’ lung tissues and BALFs (p= 0.03 and 0.01, respectively). In addition, by adjusting for age, sex, and NSCLC types and stages, a significant and positive correlation was observed between NOX4 and Nrf2 mRNA expression (r= 0.927, p= 0.001). This was also true when not adjusted as above (r= 0.944, p< 0.001). Conclusions: NOX4 mRNA and protein expression is significantly up-regulated in NSCLC patients’ lung tissues and BALFs, and NOX4 and Nrf2 mRNA expression is positively correlated in NSCLC tissues.


Introduction
Lung cancer, a major health problem affecting millions of people worldwide, is the leading cause of cancer-related mortality in humans (1). It is a complex and heterogeneous malignancy which, based on tumor histology, is classified into two main types; (I) non-small-cell, and (II) small cell, lung cancer (NSCLC and SCLC, respectively). NSCLC is the most common type with approximately 85% of all cases that contribute to metastatic lung cancers (2). Common NSCLC subtypes include squamous cell carcinoma, large cell carcinoma, and adenocarcinoma. Identifying the biological pathways and genetic abnormalities involved in lung cancer is essential for effective treatment (3). Hypoxia is a critical situation that influences the activity of important carcinogenic pathways. Hypoxia, tumor invasion, and metastasis contribute to the generation of TGF-β1, induced by the epithelial-mesenchymal transition (EMT) that is developed by losing cell-to-cell contact and transforming to mesenchymal properties. TGF-β1 is a multifunctional cytokine that induces its biological effects through both canonical and non-canonical pathways (4,5). Growing evidence suggests that the Smadindependent, non-canonical pathway, mediates p38 mitogen-activated protein kinase (P38 MAPK) and the phosphatidylinositol 3kinase/Akt/mammalian target of rapamycin (PI3/AKT/mTOR) signaling pathway, and is thought to promote cancer progression through EMT involved in migration, invasion, and metastasis (5)(6)(7). İn this process, the hypoxiainduced EMT is involved so that TGF-β produces reactive oxygen species (ROS) mediated by NOX4 upregulation (5). ROS play an important role in the early stage of hypoxiainduced EMT in various cancers, including lung cancer (8,9). ROS contribute to cancer development by modulating transcription and signal transduction pathways, and increasing replication errors and genomic instabilities (10). NADPH oxidase (NOX) complexes are major sources of endogenous non-mitochondrial ROS production in normal and cancerous cells. NOX4, a member of the NOX family, is highly expressed and constitutively activated in airway and alveolar epithelial cells (11). NOX4 is activated in response to pulmonary disorders including ischemia and hypoxia (12) and, in relation to TGF-β, produces ROS that are regulated via various transcription factors (TFs) including E2F, Nrf2, hypoxia-inducible factor 1alpha (HIF-1a), NF-κB, and STAT3 through a mechanism that involves binding to the NOX4 DNA promoter site (8,13). One of these TFs is a basic leucine zipper called nuclear factor erythroid 2-related factor 2 (Nrf2), which, under special conditions, such as hyperoxia and oxidative stress, binds to antioxidant response elements on the NOX4 promoter and regulates its expression (14).
Recent evidence suggested that Nrf2 can upregulate SOD and GST, which can deplete ROS and lipid peroxidation (15). On the other hand, PI3K/AKT mediated Nrf2 up-regulation produces redox conditions, promoting cancer cell migration and survival (16). For example, using various NSCLC cell lines, Wu et al. reported that NOX4-derived H2O2 upregulates Nrf2 activity and expression (17). Here, we aimed to analyze lung tissue for Nrf2 and NOX4 mRNA and BALF for Nrf2 and NOX4 proteins and their possible associations in patients who underwent lung bronchoscopies.

Study subjects, tissue, and lavage sampling
Patients with suspected cases of lung carcinoma who attended the five lung and respiratory disease offices in Tabriz, Iran, between Sept 2016 and Feb 2018 were subjected to diagnostic tests.
Patients with tuberculosis or hemoptysis, those with prior radiotherapy or chemotherapy, and those unwilling to participate were excluded from the study. Patient medical histories were reviewed and interpreted independently by both a pulmonary pathologist and a lung disease specialist. Those with NSCLC who agreed to participate were included in the study and their disease stages were determined.
Forty-five subjects underwent bronchoscopic needle biopsy and histopathological diagnosis at Imam Reza and Shahid Madani hospitals following previously described methods (18). The collected tissue samples were suspended immediately in RNAlater (Qiagen, Cat. No. 76106), and the BALF specimens were collected in 50 ml sterilized tubes and stored at -80 °C. The study protocol was approved by the Ethics Committee of the Tabriz University of Medical Sciences (IR.TBZMED.REC.1396.371) and all specimens were obtained from contributors with informed and signed consent.

RNA extraction and real-time PCR
Total RNA was extracted from tissue samples using the RNeasy Mini Kit (Qiagen, Cat. No. 74104) according to the manufacturer's instructions, and its quantity and quality were assessed on a Thermo Scientific™ NanoDrop™ One Spectrophotometer and by 1.5% agarose gel electrophoresis, respectively. Also, for cDNA synthesis a Thermo Scientific™ K1691 kit was used following the manufacturer's instruction. The specific primers (Table 1) for target genes NOX4 and Nrf2 were designed using Primer3, Gene Runner version 6.5.52, and NCBI-BLAST programs, and for internal control,

Western-blot analysis
For immunoblotting, the BALF samples were lysed using the lysing buffer and the lysates were centrifuged at 13300 g. The total protein concentration in the supernatants was measured by Bradford assay. The proteins were separated by SDS-PAGE and transferred to PVDF membranes. The membranes were blocked with 2% low-fat milk in Trisbuffered saline and then incubated for 16-18 h with commercially available β-Actin (C4: sc-47778), NOX4 (3H2G11: sc-517188), and Nrf2 (A-10: sc-365949) primary antibodies. The membrane was washed in washing buffer, incubated with the appropriate HRP-conjugated secondary antibody (m-IgGκ BP-HRP: sc-516102), and bands were visualized by enhanced chemiluminescence.
Densitometry was performed using NIH Image/Image J software. All antibodies were purchased from Santa Cruz Biotechnology Inc.

Statistics analysis
Kolmogorov-Smirnov's with kurtosis and skewness indices tests were applied to assess normality. Unpaired t-test was performed to compare expression of the target genes between the case and control groups. Results were expressed as means±SDs, and a twotailed p< 0.05 indicated statistical significance. Additionally, the Pearson product-moment correlation test was used to test the correlation between the target genes' expression. Also, a partial correlation test was used to analyze the association of NOX4 and Nrf2 expression controlling for age, sex, and NSCLC types and stages. All analyses were performed using SPSS software, version 16.

Results
The participants' general characteristics are shown in Table 2a. Of the 45 participants, 25 were diagnosed as NSCLC cases including 8 adenocarcinoma, 10 squamous, and 7 largecell cancerous patients, while 20 were diagnosed as noncancerous, and used as the control group (Table 2a). The frequency of each stage in the NSCLC-diagnosed patients is shown in Table 2b.
The quality and quantity of extracted RNA were confirmed by the A260/280 (1.8-2) and 230/260 (1.6-1.8) ratios, and the identification of sharp bands of the ribosomal RNAs on an agarose gel. NOX4 expression was significantly greater in NSCLC patient tissue samples than in controls (p= 0.03) (Fig. 1A). Although the Nrf2 expression was greater in NSCLC patients than in controls, this difference was not significant (p= 0.163) (Fig 1a).  Western blot and densitometry showed that the NOX4 protein level was significantly greater in the in the NSCLC patient BALF specimens than in controls (p= 0.01) (Figs. 1b  and 1c). Although the Nrf2 protein content was greater in NSCLC patients than in controls, the difference was not statistically significant. The Pearson product-moment correlation test without adjusting for covariates showed a significant positive correlation between Nrf2 and NOX4 expression (r= 0.944, p< 0.001) in NSCLC tissue samples. A partial correlation test adjusted for age, sex, and NSCLC types and stages also showed a significant positive correlation between Nrf2 and NOX4 expression (r= 0.927, p= 0.001), and also when not adjusted for those factors (r= 0.944, p< 0.001).

Discussion
In this study, we found significantly greater NOX4 expression and non-significantly greater Nrf2 expression in NSCLC patient lung tissue than in that of controls. Moreover, similar results were observed for their respective proteins in the BALFs. In addition, we found a significant correlation between Nrf2 and NOX4 expression in the NSCLC patient lung tissues.
In response to hypoxia and inflammatory conditions in various cancer cells, including NSCLC, reactive oxygen and nitrogen species (RONS) are induced by the main ROS producer, NOX4, and nitric oxide synthases (22). RONS have detrimental effects on proteins, DNA, and organelles. Indeed, NOX4induced oxidative stress promotes hypoxiainduced EMT through the driving force of TGFβ1 ( 22). Based on the physiological condition, concomitant with the accumulation of RONS, the enzymatic and non-enzymatic endogenous antioxidant systems are initiated to maintain cellular redox homeostasis. Recent evidence suggests that the redox-sensitive residues in regulatory proteins, including cysteine oxidation/S-sulfenylation/Sglutathionylation/S-nitrosylation and tyrosine nitration, activate intracellular signaling pathways and transcription factors that promote cell proliferation, migration, and survival (22). These oxidative/nitrosative modifications activate some intracellular transcription effectors, including Nrf2 as a master of the antioxidant response regulator in PI3K/Aktmediated signaling pathways (22). Recent studies indicate that Nrf2 exhibits a dual role in cancer progression (23 (17).
As discussed above, oxidative/nitrosative modifications are a major factor in cancer initiation and progression. İn this line, NOX4 and Nrf2 play key roles in oxidative stress regulation and the EMT process via various signaling pathways. Based on our findings, increased NOX4 and Nrf2 mRNA and protein, and also their significant positive correlation, indicated that they are significant factors in lung cancer proliferation and progression.
Some limitations should be addressed regarding the present study. Because of the difficulty of sampling via biopsy from lung and BALF, we had limited tissue and BALF specimens. for that reason, we were unable to measure some of the agents implicated in NOX4 and Nrf2 function, for exampleoxidative stress and TGF-β1. Moreover, our subject number was relatively small.
In conclusion, we found upregulated NOX4 and Nrf2 mRNA and protein in lung tissue and BALF of NSCLC patients relative to controls. İn addition, a strongly positive association was observed between NOX4 and Nrf2. Considering the implication of NOX4 and Nrf2 in key carcinogenic signaling pathways, our findings suggest that these biomarkers may be candidates as NSCLC diagnostic and treatment targets. Moreover, studying the molecular aspects of gene-gene interactions regarding Nrf2 and NOX4 could provide valuable information on the molecular regulation of the oxidative pathways in lung cancer pathobiology.