Prognostic Worth of Nrf2/BACH1/HO-1 Protein Expression in the Development of Breast Cancer

Breast cancer is the principal cause of cancer deaths in females worldwide. In 2012, around 1.7 million new cases were diagnosed worldwide, and it was estimated to rise to about 2.1 million in 2018, attributing 1 in 4 cancer cases among females [1]. The pathogenesis of breast cancer is not entirely clear as our comprehension of the various biological and molecular processes involved in the development of breast cancer is still incomplete.

Nuclear factor erythroid 2-related factor (Nrf2), a member of basic leucine-zipper subfamily transcription factors, plays an important function in the adaptive response to oxidative stress by directing several transcriptional activities. Nrf2 forms a complex with its repressor Kelch-like ECH-associated protein 1 (KEAP1) and cullin 3 (CUL3) ubiquitin ligase, which subjects Nrf2 to degradation in proteasomes under homeostatic conditions. However, under stress state, Nrf2 is separated from KEAP1 and transported into the nucleus to form heterodimers with small Maf proteins which then attach to antioxidant response element (ARE) of the desired genes regulating its expression [2]. Nrf2 has been regarded as a tumor suppressor owing to its cytoprotective role against reactive oxygen species (ROS) and electrophilic stressors. Upregulated Nrf2 in cancer aids malignant cells to cope with increased levels of ROS and evade apoptosis via activation of metabolic and cytoprotective genes that promote cell growth [3]. Overexpression of Nrf2 plays a critical role in inducing cancer cell growth, proliferation, and survival and decreasing the sensitivity of the cells to chemo/radiotherapeutic agents [4]. Report indicates that upregulated Nrf2 expression results in lower overall survival and disease-free survival in breast cancer patients [5]. The pro-oncogenic role of Nrf2 in breast cancer cells and its anti-oncogenic capacity in healthy cells relies on metabolic adaptation, cell proliferation, and induction of Nrf2 [6, 7].

BACH1, a cap “n” collar protein transcriptional factor, is part of the basic leucine-zipper superfamily. BACH1 binds to Maf recognition elements to form heterodimers that mediate transcription in gene promoter regions. Expression of BACH1 in human tissues has been observed and analyzed. Upregulated MALAT1 and BACH1 demonstrate shorter overall survival and disease-free survival in triple-negative breast cancer [8]. Modulation of high mobility group A2 and BACH1 promotes the proliferation and migration of breast cancer cells and inhibits apoptosis [9].

Heme oxygenase-1 (HO-1) a rate-limiting enzyme in heme catabolism is inducible as feedback to many stress stimuli including heat shock, hypoxia, heme, ROS, and nitric oxide. When heme concentrations are low, BACH1 links directly to the HO-1 promoter and silences it; however, at higher heme levels, BACH1 is relieved, followed by overexpression of HO-1 [10]. The induction of Nrf2 into the nucleus exports nuclear BACH1 to upregulate HO-1 expression to inhibit apoptosis. Thus, Nrf2 and BACH1 are the main transcriptional factors involved in modulating HO-1 complexes. Notwithstanding the cytoprotective ability of heme oxygenase, it also plays a critical role in carcinogenesis. Inhibition of HO-1 abrogates dipeptidyl peptidase-4 inhibitor-induced upregulation of Nrf2 to prevent breast cancer metastasis [11]. Silencing of HO-1 suppresses breast cancer metastasis [12]. Modulation of Bach-1/Nrf2 can reduce HO-1 to promote upregulated apoptosis and decreased proliferation of breast cancer cells and vice versa [13].

Despite these observations, the clinical relevance of Nrf2, BACH1, and HO-1 in breast cancer remains understudied and unclear. This study investigated the expression of Nrf2, BACH1, and HO-1 protein in conjunction; the results suggest that the expression of these proteins could be a biomarker for the stage, progression, and prognosis of cancer.

The study was approved by the Institutional Review Board of Jiamusi University, China, with ethical clearance number SYXK 2016-014. Breast cancer and noncancerous (control) tissues were obtained from 114 patients who underwent surgery at the First People’s Hospital affiliated to Jiamusi University from 2015 to 2017. This was a cross-sectional study in which patients were selected based on the following inclusion criteria: complete medical history, no preoperative cancer treatment, histopathologic report with regional lymph node metastasis, and absence of distant metastasis. Patients with history of other cancers were excluded from the study. 114 breast cancer tissues and 30 adjacent noncancerous (control) tissues used for the study were collected from 30 of the patients as paired specimens. Pathological parameters such as age, tumor stage, tumor grade, tumor node metastasis (TNM) stage, ER, PR, HER2, and histologic type obtained from the pathology report of each patient were used for the analyses. The patient’s average age was 50 years.

Hematoxylin and eosin-stained slides were used to study histopathologic features obtained from pathology reports of patients to confirm the diagnosis. 4 μm formalin fixed paraffin-embedded tissue mounted on a slide was deparaffinized in 100% xylene (I and II) and rehydrated in ethanol 100% (I, II), 95%, 90%, 80%, 70% for 15 min and 5 min, respectively. After rinsing with phosphate-buffered saline, mounted sections were heated for 40 min in citrate solution and allowed to cool. Subsequently, the sections were treated with 5% bovine serum albumin for 30 min at 37°C and then incubated with either Nrf2 or HO-1 (1:100 Santa Cruz, CA, USA) or BACH1 (1:50 Santa Cruz, CA, USA) primary antibody overnight at 4°C. Following overnight incubation, sections were incubated with biotinylated goat anti-rabbit secondary antibody and washed three times with phosphate-buffered saline. 3,3ʹ-diaminobenzidine was applied to the mounted section for 10 min, washed, and stained with Mayer’s hematoxylin. After rinsing, it was then dehydrated in ethanol, cleared in xylene, and observed under the microscope. Brown reaction and deep blue-purple color indicated positive and negative staining, respectively. The immunoreactivity recorded was an average score observed by two pathologists by using the semiquantitative method classified as follows: (0), negative, no staining; (1+), weak staining, ≤10% stained cells; (2+), moderate staining, 11−50% stained cells; (3+), strong staining, >50% stained cells. The sum of the scores for both intensity and proportion were used as a measure of the expression.

Minced tumor tissues lysed in 500 µL cell lysis buffer for 30 min at 4°C were centrifuged at 12,000 g for 15 min and then run on SDS-PAGE initially at 60 V for 15 min and upregulated to 110 V to complete the process in 2 h. Proteins were transferred to PVDF membranes after which the membranes were blocked and incubated overnight with either Nrf2 or BACH1 or HO-1 or β-actin primary antibody. The subsequent day, it was rinsed thrice for 5 min per wash using Tris-buffered saline with Tween 20. Additionally, the membrane was incubated with HRP-labeled goat anti-rabbit IgG for an hour at room temperature and rinsed thrice with Tris-buffered saline with Tween 20. The protein reactive band was intensified with a chemiluminescence kit, exposed to X-ray film and images captured with LabWork 3.0 (UVP Inc., Upland, CA, USA).

Statistical analysis was performed using SPSS version 21.0 (SPSS Inc., Chicago, IL, USA). χ² test was used to determine the correlation between Nrf2, BACH1, and HO-1 expression and the clinicopathologic features. The Mantel-Haenszel-Cochran test was used for calculating the relationship between Nrf2, HO-1 expression and tumor grade, BACH1 and tumor stage. p < 0.05 was deemed as significant.

To investigate the effect of Nrf2, BACH1, and HO-1 in breast cancer, immunohistochemistry was carried out to demonstrate their expression in 114 cancerous tissues. Protein expression was recorded as (A) negative staining (0); (B) weak staining (1+); (C) moderate staining (2+); (D) strong staining (3+) as shown in (Fig. 1).

Expression of Nrf2 protein: 2 cases of negative staining (0), 28 cases of weak staining (1+), 26 cases of moderate staining (2+), and 58 cases of strong staining (3+). Based on immunostaining results, 84 (73.7%) tissues of patients expressed high Nrf2, and 30 (26.3%) expressed low Nrf2 (Table 1).

Expression of BACH1 protein: 8 cases of negative staining (0), 62 cases of weak staining (1+), 16 cases of moderate staining (2+), and 28 cases of strong staining (3+). Immunostaining results revealed that 44 (38.6%) tissues of patients expressed high BACH1 and 70 (61.4%) expressed low BACH1 (Table 2).

Expression of HO-1 protein: 3 specimens of negative staining (0), 35 specimens of weak staining (1+), 19 specimens of moderate staining (2+) and 57 specimens of strong staining (3+). Immunostaining results showed that 76 (66.7%) tissues of patients expressed high HO-1 and 38 (33.3%) expressed low HO-1 (Table 3).

To determine the expression of Nrf2, BACH1, and HO-1, 114 breast cancer and 30 control tissues were evaluated by immunohistochemistry. We found Nrf2 and HO-1 were highly expressed in the nucleus with low cytoplasmic staining, while BACH1 was expressed in the low levels in the cytoplasm of cancerous tissues as compared to control tissues (Fig. 2).

Immunohistochemical studies revealed that 74% of the cancerous samples had high Nrf2 protein expression and 26% of them had low Nrf2 protein expression. Regarding the non-cancer samples, 43% had high expression of Nrf2 protein and 57% had low expression of Nrf2 protein (Table 1). Expression of Nrf2 protein was associated significantly with tumor grade (p < 0.05). However, there were no significant differences in the expression of Nrf2 in terms of age, tumor stage, TNM, ER, PR, HER2, and histologic type. Furthermore, expression of Nrf2 protein in both cancerous and non-cancer tissues also showed a significant correlation (p < 0.05) (Table 1).

39% of the cancer tissues had high expression of BACH1 protein, while 61% had low expression of BACH1 protein. With regards to the non-cancer samples, 80% had high expression of BACH1 protein and 20% of them had low expression BACH1 protein. BACH1 protein expression was not significantly correlated with age, tumor grade, TNM, ER, PR, HER2, and histologic type. However, the expression of BACH1 protein was significantly associated with tumor stage (p < 0.05). Moreover, there was also a significant difference observed between the cancerous and noncancerous tissues (p < 0.05) (Table 2).

67% of the cancer tissues had high expression of HO-1 protein, while 33% had low expression of HO-1 protein. However, 17% of the noncancerous tissues had high expression of HO-1 protein and 83% had low HO-1 protein expression (Table 3). Expression of HO-1 protein was significantly associated with tumor grade (p < 0.05). However, no significant correlation was found between the expression of HO-1 protein and age, and the expression of HO-1 was not significantly influenced by tumor stage, TNM, ER, PR, HER2, and histologic type. Furthermore, HO-1 protein expression in both cancerous and non-cancer tissues showed a significant correlation (p < 0.05) (Table 3).

Cochran-Mantel-Haenszel statistics was used to work out the correlation between the expression of Nrf2 protein and tumor grade. The odds that poorly differentiated tissues will express high Nrf2 are greater than that of well/moderately differentiated tissues. Moreover, BACH1 and tumor stage also demonstrated that the odds that tissues of T3–T4 stage will have low BACH1 expression are greater than that of T1–T2 stage. Finally, HO-1 and tumor grade showed that the odds that well/moderately differentiated tissues will express high HO-1 are greater than that of poorly differentiated tissues (p < 0.05) (Table 4).

To further demonstrate the protein expression of Nrf2, BACH1, and HO-1, Western blot analysis was carried out in both cancerous and noncancerous tissues. As indicated in (Fig. 3), the cancer tissues revealed high Nrf2, low BACH1, and high HO-1 compared to controls; this is consistent with the results of immunohistochemical analyses.

This study demonstrates the adverse effect of expression of Nrf2, BACH1, and HO-1 on the clinical outcome of patients with breast cancer. Nrf2, a transcription factor, controls the expression of different antioxidant and cytoprotective genes modulating oxidative and electrophilic stress to cellular response. The modulation of nuclear Nrf2/BACH1 via CXCR3-B/CXCL4 signals can upregulate HO-1 expression to inhibit apoptosis [13]. This study found high expression of Nrf2 in 74% of breast cancer tissues, while it was positively expressed in 43% of the control tissues. Researchers have previously shown that the expression of Nrf2 is detected frequently in lung cancers with an expression rate of 74% and 77% [14, 15] which is consistent to our findings. Moreover, we observed that the level of Nrf2 expression correlated with tumor grade and that its expression in poorly differentiated tumors is higher than that of well/moderately differentiated tumors consistent with study [16]. Further assessment revealed that Nrf2 expression was more concentrated in the nucleus of breast cancer tissues which is consistent with this study [16].

Expression of nuclear Nrf2 protein plays a critical function in the growth and development of breast cancer. Therefore, functional Nrf2 activity in human breast carcinoma is reflected by nuclear Nrf2 immunoreactivity, and its relatively wide distribution indicates the relevance of activated Nrf2 signaling pathway in breast cancer. Aberrant Nrf2 expression is associated with increased resistance to therapy in breast cancer, implying that Nrf2 genes are downregulated in breast cancer after starvation, which is associated with increased ROS levels [7, 17]. Hyperactivation of Nrf2 has been reported to upregulate glucose‐6‐phosphate dehydrogenase/HIF‐1α/Notch1 signaling to promote migration and metastasis of breast cancer cells [18]. Recently, high expression of Nrf2 was shown to downregulate GSK-3β to enhance breast cancer [19]. Nrf2 siRNA reversed the tamoxifen resistance in tamoxifen-resistant breast cancer with upregulated production of Nrf2-dependent antioxidant proteins [20]. Thus, compared to Nrf2-negative subjects, residual breast cancer in Nrf2-positive breast cancer after surgical treatment might be able to proliferate rapidly and metastasize despite adjuvant therapy and result in high resistance and poor prognosis of breast cancer.

As BACH1 is a transcriptional suppressor which negatively modulates various genes that play key roles in cell cycle progression, apoptosis, and oxidative stress response [21], it may be considered a potential tumor suppressor. Our study showed low expression of BACH1 in 61.4% of breast cancer tissues and even lower expression (20%) in control tissues similar to a previous report [22]. Moreover, this current study revealed that low expression of BACH1 correlates significantly with tumor stage. Conversely, high expression of BACH1 has been reported to promote growth and proliferation of breast cancer [8, 23]. Overexpression of long noncoding RNA (SNHG5) has been shown to enhance breast cancer growth and glycolysis by upregulating the expression of BACH1 through targeting miR-299 [23].

HO-1 has been reported to be highly upregulated in tumor tissues and facilitates tumor proliferation, metastasis, and reduced sensitivity to chemotherapy [11, 12, 24]. In the current study, out of 114 breast cancer tissues, high expression of HO-1 was observed in 76 (67%), whereas much lower expression of HO-1 was found in control tissues (5 of 30, 17%) consistent with previous studies [24, 25]. Moreover, the results demonstrated that upregulated HO-1 expression in breast cancer was associated with tumor grade which is also in agreement with a previous study [26]. Furthermore, the results showed that well/moderately differentiated tumors express high HO-1 compared to poorly differentiated tumors. This could be due to the fact that under low or normal heme conditions, BACH1 binds to ARE and represses the expression of HO-1. However, when heme level is elevated, BACH1 binds to heme allowing Nrf2 to bind to ARE and transcriptionally regulate HO-1 expression to break down the heme [10]. Breakdown of heme results in the production of antioxidants and antiapoptotic molecules as by-products with the latter preventing apoptosis and increasing tumor cell proliferation. Therefore, invasive and metastatic tumors go along with upregulated expression of HO-1. However, thorough evaluation of HO-1 expressive pattern in breast cancer revealed an important and interesting phenomenon, in contrast to the rate of expression and intensity of HO-1 in the cytoplasm, nuclear HO-1 expression was higher which is similar to a previous study [27]. The translocation of HO-1 from the cytoplasm to the nucleus has been suggested to be an important factor associated with the protective effects of HO-1 expression in tumors, conferring some mechanisms of tumor proliferation, angiogenesis, and drug resistance [28]. The above results at least partially support our observation that nuclear expression of HO-1 might be substantially associated with malignancy of breast cancer; we suggest that it is critically relevant to examine the expression pattern of HO-1 in tumors when assessing the roles of HO-1 in cancers. HO-1 knockout has been demonstrated to promote cisplatin-induced apoptosis to abolish proliferation and migration of breast cancer [29]. In triple-negative breast cancer patients, upregulated HO-1 expression has been reported to be significantly associated with poor disease-free survival, overall survival, and lower pathological complete response rate [30].

The correlation of Nrf2 and HO-1 with tumor grade and BACH1 with tumor stage suggests that Nrf2, HO-1, and BACH1 could serve as potential biomarkers for cancer stage, progression and prognosis as well as targets for therapy. Furthermore, in vivo and in vitro molecular studies are needed to evaluate their potential application in diagnosis and treatment. Therapies that can suppress Nrf2 and HO-1 and upregulate BACH1 might hold promise for the treatment of breast cancer.

The study was approved by the Institutional Review Board of Jiamusi University, China, with ethical clearance number SYXK 2016-014.

The authors have no potential conflicts of interest to declare.

This research received funding from Natural Scientific Foundation of HeiLongJiang Province (H2016089).

Precious Barnes: design, data analysis, and manuscript writing and editing; Elvis Agbo: literature review, data analysis, and manuscript writing and editing; Jianjie Wang: conception of the study, study design, funding acquisition, data collection and analysis, and manuscript preparation and review; Benjamin Amoani, Yeboah Kwaku Opoku, and Perditer Okyere: literature review, data interpretation, and manuscript writing and editing; and Roland Osei Saahene: conceptualization, design, data collection and analysis, laboratory investigation, project administration, manuscript writing, review, and editing.

All data generated or analyzed in this study have been included in this research article.