Overexpression of POU3F2 promotes radioresistance in triple-negative breast cancer via Akt pathway activation

POU3F2 is associated with malignant behaviors and poor prognosis in cancer. However, the function and mechanism of POU3F2 in breast cancer remain to be elucidated. Our study aimed to explore the role of POU3F2 in triple-negative breast cancer and radiotherapy. POU3F2 expression was examined by RT-PCR and Western blot. The proliferation of cancer cells was measured by MTT assay. Migration of cancer cells was determined by Transwell assay and wound healing assay. To determine which protein interacts with POU3F2, Co-IP was performed. Survival analysis was performed based on the online database GEPIA. DNA damage after radiation was examined by Comet Assay. Radiosensitivity was evaluated with clonogenic survival assays. A tumor xenograft model was established with MDA-MB-231 breast cancer cells in BALB/c nude mice to explore the effect of POU3F2 in vivo. We found that the expression of POU3F2 was significantly elevated in breast cancer cells, especially in TNBC, and higher POU3F2 expression was related to poor prognosis of patients with breast cancer. Functional assays revealed that POU3F2 promoted proliferation, migration, and invasion of triple-negative breast cancer (TNBC) cells in vitro and in vivo. In addition, the knockdown of POU3F2 decreased the radioresistance of TNBC cells in vitro. Furthermore, POU3F2 could enhance the activation of the Akt pathway by interacting with ARNT2, thereby promoting proliferation and radioresistance in TNBC cells. Our results provide evidence that high expression of POU3F2 promotes radioresistance in triple-negative breast cancer via Akt pathway activation by interacting with ARNT2.


Introduction
Breast cancer is the most prevalent cancer of females around the world, with nearly 2.3 million new cases per year, and accounts for 11.7% of all cancer cases [1]. Triple-negative breast cancer (TNBC), one of the worst subtypes accounting for 15-25% of all breast cancer, lacks expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2), therefore, patients with TNBC does not respond to endocrine therapies or HER2-targeting therapies [2]. Though immune checkpoint inhibitors have benefited TNBC patients, a lower response rate and inevitable immune checkpoint inhibitor (ICI)-related complications also limit the clinical applications. Currently, radiotherapy (RT) is still commonly utilized and plays a pivotal role in the treatment of TNBC. Adjuvant RT decreased the 10-year recurrence rate of TNBC from 35% to 19%, resulting in significant improvement in 15-year survival rates for patients with TNBC [3]. However, compared with other breast cancer subtypes, TNBC is more resistant to RT, which often leads to disease relapse and metastatic dissemination after RT. Therefore, there is a pressing need to identify underlying molecular mechanisms to overcome radiotherapy resistance in TNBC and potential biomarkers for early recognizing patients who are likely to develop radioresistance.
POU Class 3 Homeobox 2 (POU3F2), also named BRN2, is a member of the POU3 family of transcription factors that bind to an octameric DNA sequence [4]. POU3F2 forms a transcriptional regulatory complex by cooperating with itself (via homo-dimerization) and other proteins, including TATA-binding protein (TBP), the transcriptional coactivator, p300, and Sox-10 in melanocytic regulation [5]. POU3F2 plays a significant role in cellular differentiation [6], could induce neuroendocrine-specific transcription factors, and is associated with cell viability [7]. Recently, it has been reported that POU3F2 is involved in the pathogenesis and treatment resistance of small cell lung cancer (SCLC) [8,9], prostate cancer [10,11], melanoma [12], and glioma [13]. Cook and Sturm reported that upregulated POU3F2 expression in melanoma was related to increased tumorigenesis and cancer cell growth. Sakaeda et al. observed that POU3F2 was highly expressed in small lung cancer cells when compared with non-small cell lung cancer (NSCLC) [8]. However, the unique role of POU3F2 on TNBC initiation, and progression, and whether it mediates radioresistance in TNBC remains unknown.
In the present study, we illuminated the carcinogenic role of POU3F2 by examining the expression of POU3F2 in breast cancer cell lines and human tumor tissue samples and further demonstrated that POU3F2 could interact with ANRT2 and activate Akt signaling, resulting in increased proliferation and radioresistance of TNBC cells.

Proliferation assay
The proliferation of cancer cells was measured by MTT assay. Briefly, cells were plated into 96-well plates (2 × 10 3 cells/well) in 200 μl of culture medium. After incubation for 24 h, 48 h, 72 h, and 96 h, the medium was removed, and 20 μl of MTT dye solution (Sigma-Aldrich, USA) was added and incubated for 2 h at 37 °C. The cells were dissolved in DMSO, and the optical density (OD) was measured at 490 nm in a microplate spectrophotometer (Thermo Scientific, USA).

Transwell assay and wound healing assay
For the transwell assay, cells (1 × 10 5 cells/μl) were resuspended in 200 μl of RPMI-1640 medium or DMEM and seeded in the upper chambers of transwell inserts (8 μm pore size, 24-well plates, Corning, USA). The lower chambers were filled with 750 μl of RPMI-1640 medium or DMEM containing 10% FBS. After incubation, the migrated cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet at room temperature. The migrated cells were imaged with a microscope. For the wound healing assay, cells (5 × 10 5 /well) were seeded into 6-well plates. When the cell reached 100% confluence, sterile 200 µl pipette tips were used to scratch a straight line across the middle of the well. Detached cells were removed by PBS wash and the plates were incubated. A picture was taken to record the width at 0 h, and after incubation, for 12 h, the width of the scratch was measured and photographed.

Co-immunoprecipitation assays (Co-IP)
Cells were harvested and washed with PBS three times. The clarified cell lysates were incubated for 16 h with the appropriate antibody at 4 °C. A total of 30 μl of 50% slurry of protein A-and G-Sepharose (Amersham, UK) was added. After incubation for 2 h at 4 °C, the mixture was pelleted and washed several times with Co-IP buffer. Each sample was analyzed by SDS-PAGE and immunoblotting with the appropriate antibodies.

Bioinformational analysis
The data on breast cancer patients were downloaded from TCGA [14], a public database. Clinical information on ER, PR, as well as HER2 status, was obtained from the clinical biospecimen core resource text files and only patients who haven't accepted any treatment with complete data were enrolled for analysis. Transcripts per million (TPM) was used as the measure of gene expression. The TPM dataset was log-transformed and then normalized based on quantiles. The differential POU3F2 expression analysis between tissues of TNBC and other breast cancer subtypes was conducted using the Wilcoxon test. Survival analysis was performed using the data from the TCGA database by GEPIA.

Comet assay
Reagents and glue boards were prepared with the manufacturer's protocol. Cells were harvested after 48 h in each condition and resuspended in 1% low-melting-point agarose (at 37 °C). The agarose was allowed to gel at 4 °C and the coverslip was removed. Cells were lysed overnight in Alkaline Lysis (A1 lysis solution) solution supplemented with 1% Triton. After lysis of the cells (typically 1-2 h at 4 °C), the slides were washed in distilled water to remove all salts and immersed in an electrophoresis solution.

Cells transfection and lentivirus transduction
Cells were seeded in 6-well plates and grown to 70-90% confluency at transfection. Opti-MEM (Gibco) was diluted according to the manufacturer's protocol. Then DNA was diluted into Opti-MEM medium and the DNA was added to the Lipofectamine3000 Reagent (Gibco). After incubation for 5 min, the DNA-liquid complexes were added to each well. Transfected cells were analyzed for gene expression 2 or 3 days later. SiRNA was purchased from Invitrogen. POU3F2 and ARNT2 plasmids were purchased from Fulen Gen (China). For lentivirus packaging, transduction, and screening of stable cell lines, the whole procedure was conducted as previously detailed [15].

Clonogenic survival assays
Cells were seeded in plates and were then exposed to radiation at the indicated doses (Varian2300EX, USA). After incubation for 14 days, the cells were fixed and stained with 4% paraformaldehyde and 1% crystal violet. Colonies with more than 50 cells were counted by microscopy. Survival curves were generated using the multitarget single-hit model.

Animal study
Athymic nude mice (6-week-old female) were purchased from Guangdong Medical Laboratory Animal Center. MDA-MB-231 cells stably transduced with lentivirus or control were subcutaneously injected into the left mammary fat pads of 6-week-old female nude mice (n = 6 mice per group). All mice were fed in a standard SPF-rearing center and experiments were approved by the Institutional Animal Care and Use Committee of Nanfang Hospital and obeyed the rules required by the Guide for the Care and Use of Laboratory Animals.

Statistic
Data are expressed as mean ± standard deviation. Statistical comparisons were conducted using Student's t-test and oneway analysis of variance (ANOVA) in SPSS software (version 26.0), followed by the LSD test. All experiments were performed in triplicate. Gene expression difference was analyzed in R software (version 3.6.4). A p-value of < 0.05 was considered to indicate a statistically significant difference.

POU3F2 expression is upregulated in breast cancer and related to poor prognosis in breast cancer patients
The expression of POU3F2 in human normal mammary epithelial cells, including MCF-10A and 76N-F2V, and five human breast cancer cell lines was assessed by qPCR and Western blot. POU3F2 was upregulated at the mRNA ( Fig. 1A) and protein levels (Fig. 1B) in breast cancer cell lines, particularly in the TNBC cell lines (BT549 and MDA-MB-231). In addition, we collected 30 breast cancer samples and 30 adjacent normal samples and evaluated the expression of POU3F2. We found that POU3F2 expression was increased more than tenfold tumor tissue compared with normal adjacent breast tissue on mRNA level (Fig. 1C). Consistent with the gene expression, the expression of POU3F2 protein levels were significantly higher in tumor samples than in normal breast tissue (Fig. 1D).
To compare POU3F2 expression between TNBC and biomarker-positive breast carcinoma, we made bioinformatical analysis and found that POU3F2 expression in the former was higher than the latter (Fig. 1E) (p = 0.0106). We used GEPIA [16] to plot survival curves, and found that high expression of POU3F2 was related to the poor clinical outcome (p = 0.016) (Fig. 1F). These data showed that expression of POU3F2 was not only increased in breast cancer cell lines and breast tumor specimens but that high expression of POU3F2 was associated with poor overall survival.

POU3F2 promotes the proliferation, migration, and invasion of TNBC cells
To investigate the biological functions of POU3F2 in TNBC cells, we transiently knocked down POU3F2 expression using siRNA (Fig. S1A, B), and overexpressed POU3F2 by transfecting cells with a POU3F2 overexpression plasmid (Fig. S1C, D). Knockdown of POU3F2 inhibited the proliferation of MDA-MB-231 and BT549 cells ( Fig. 2A, B). POU3F2 overexpression enhanced proliferation in 76N-F2V and MCF-10A cells (Fig. 2C, D). To investigate the effects of short-term knockdown of POU3F2 on cell migration and invasion, we performed wound healing assays and transwell assays. In wound healing assays, we observed slower migration in MDA-MB-231 and BT495 cells with POU3F2 knockdown, when compared with control groups (Fig. 2E, F). Similarly, in transwell assays, the number of MDA-MB-231 and BT495 cells with POU3F2 knockdown in the lower chamber was fewer than those of the control groups (Fig. 2G, H). These data demonstrated that POU3F2 facilitated the proliferation, migration, and invasion of TNBC cells. Next, we investigated whether overexpression of POU3F2 in MDA-MB-231 cells could enhance tumor growth in vivo. We subcutaneously inoculated nude mice with MDA-MB-231 cells infected with POU3F2-overexpressing lentivirus or control lentivirus. In line with our results in vitro, the POU3F2 overexpressing MDA-MB-231 tumors exhibited increased tumor growth, compared with the control group (Fig. 2I,  J). There was also a significant increase in tumor weight in the POU3F2 overexpression groups (Fig. 2K). These results suggested that POU3F2 could promote proliferation, migration, and invasion in breast cancer cells and enhance tumor growth in vivo.

POU3F2 enhances the radioresistance of TNBC cells in vitro
TNBC often exhibits radioresistance, and transcription factors, such as STAT3 [17] and NF-κB [18], are involved in cancer radioresistance [19]. Currently, the role of POU3F2 in radioresistance is still unknown. To explore the effects of POU3F2 on the radioresistance of MDA-MB-231 and BT549 cell cells, we irradiated cells with varying x-ray doses and performed colony-forming assays. Compared with the control cells, MDA-MB-231 and BT549 cells with POU3F2 overexpression showed an increased surviving fraction, indicating increased radioresistance (Fig. 3A,  B). We performed a similar assay in which POU3F2 was knocked down in MDA-MB-231 and BT549 cells. POU3F2 knockdown reduced the surviving fraction, indicating a decrease in radioresistance (Fig. 3C, D). These results suggested that POU3F2 affects the sensitivity to radiotherapy in TNBC cells. Because ionizing radiation kills cancer cells mainly by damaging DNA [20], we examined whether shortterm knockdown of POU3F2 in MDA-MB-231 and BT549 cells affected DNA damage. Comet assays were performed to assess DNA damage in these cells, and we observed that POU3F2 knockdown in MDA-MB-231 and BT549 cells resulted in more DNA damage than the control groups (Fig. 3E, F). When DNA gets damaged and double-strand breaks occur, cells initiate repair mechanisms. H2AX phosphorylation is a classical marker of DNA double-stranded breaks [21]. To further confirm DNA damage at the protein level, we evaluated γ-H2AX expression when these cells were exposed to radiation. Two hours after irradiation, we found that γ-H2AX expression was increased in MDA-MB-231 and BT495 cells with POU3F2 knockdown, compared to that of control groups (Fig. 3G, H). These results suggested that POU3F2 mediates the radioresistance of TNBC cells.

POU3F2 promotes proliferation and radioresistance by interacting with aryl hydrocarbon receptor nuclear translocator 2
Previous studies have suggested that protein-protein interactions are important in mediating responses to radiotherapy [22]. We speculated that POU3F2 might interact with other proteins to mediate radioresistance in TNBC. We screened several proteins that could interact with POU3F2 by immunoprecipitation (IP) and mass spectrometry analysis. To further determine which protein interacts with POU3F2, we conducted Co-IP and found that POU3F2 binds to aryl hydrocarbon receptor nuclear translocator 2 (ARNT2, Fig. 4A). ARNT2 has been correlated with tumor progression, and has been reported to act as an anti-tumor gene [23,24]. To determine the role of ARNT2 role in TNBC cells, pAKT, survivin, and BCL-2 were evaluated by Western blot.
Overexpression of ARNT2 inhibited Akt phosphorylation and the activation of downstream signaling of Akt (Fig. 4B).
In addition, we observed Akt and its downstream signaling could be activated by POU3F2 overexpression (Fig. 4C).
To explore whether POU3F2 could mediate radioresistance by binding to ARNT2, we conducted the corresponding rescue experiments. Overexpression of ARNT2 reduced the increased proliferation capability (Fig. 4D, E) and an increased surviving fraction (Fig. 4F, G) in TNBC cells with POU3F2 overexpression. These results suggest that POU3F2 mediates radioresistance through the AKT pathway by interacting with ARNT2.

Discussion
TNBC is the most aggressive subtype of breast cancer. Compared with non-TNBC tumors, TNBC often presents with larger tumor size, increased tumor grade, and higher proliferation index when diagnosed [25]. Despite the initial response of TNBC to treatment, about 30% of patients will develop tumor relapse and 20% of patients will succumb to TNBC metastasis within 5 years [26]. It is urgent to illuminate the underlying molecular mechanism mediating radioresistance and identify potential targets for TNBC to improve radiosensitivity. In this study, we found that the expression of POU3F2 was closely related to the progression of TNBC, and it could be used as a biomarker of radioresistance and a potential target to develop novel therapeutic strategies for TNBCs. POU3F2 is a reprogramming transcription factor that regulates cellular differentiation, contributes to tumor progression, and induces stem-like cancer cells in different kinds of tumors, such as glioblastoma [27,28], melanoma [29] and small cell lung cancer [9]. Here, our results showed that POU3F2 not only promoted proliferation, migration, and invasion but also mediated radioresistance in TNBC. Previous studies had demonstrated that POU3F2 formed a transcriptional regulatory complex by cooperating with itself and other proteins, including TATA-binding protein (TBP), the transcriptional coactivator, p300, and Sox-10 in melanocytic regulation [5]. By Co-IP experiment, we found that POU3F2 could bind to ARNT2 protein which has been reported to act as an anti-tumor gene [23,24]. ARNT 2 is a transcription factor related to adaptive responses against cellular stress from xenobiotic substances. The expression level of ARNT2 mRNA was positively correlated with a favorable prognosis of breast cancer, and the presence of ARNT2 was significantly correlated with the smaller tumor size and improved 5-year survival rate after breast cancer diagnosis [30]. In the present study, our results provide evidence that low The Akt pathway contributes to many cellular functions in cancer, including survival, apoptosis, and proliferation [31]. It can be activated by irradiation and is involved in mediating radioresistance in many tumor types, including glioma [32], prostate cancer [33], and lung cancer [34]. Growing evidence revealed that the Akt serving as a key factor played a key role in prostate cancer progression and radiation resistance by activating pathway proteins or through mutations in the pathway [35]. Blocking PI3K/Akt signaling by inhibiting EGFR could improve radiosensitivity and affect tumor growth through a variety of mechanisms, including inhibition of neovascularization through VEGF. It has been reported that ARNT2 inhibits cancer cell proliferation by negatively regulating the Akt pathway [36]. In this research, our data showed that POU3F2 overexpression could increase the viability of TNBC cell lines by activating the Akt pathway upon irradiation. Meanwhile, ARNT2 overexpression in these cells reduced Akt activation. Thus, we speculated that POU3F2 might prevent the functions of ARNT2 by interacting directly with the ARNT2 protein and later inhibiting the activation of the Akt pathway. However, the specific binding sites of POU3F2 and ARNT2 are not defined, and additional investigation is needed to identify the specific binding sites.
In conclusion, POU3F2 was upregulated in TNBC cell lines, and higher expression of POU3F2 was prognostic of poor outcomes in patients with TNBC. Mechanistically, POU3F2 could enhance the activation of the Akt pathway by interacting with ARNT2, thereby promoting proliferation and radioresistance in TNBC cells. Our data provide novel insight into the radioresistance of TNBC and suggest that POU3F2 may be considered a novel biomarker for the prognosis of patients with TNBC. Targeting POU3F2 could also be a potential strategy to overcome radioresistance to improve survival outcomes for patients with TNBC. Nevertheless, in this study, survival analysis was performed using the TCGA database by the GEPIA web tool, and these online database resources did not identify a specific subtype of breast cancer, thus the patient survival data could not be analyzed by tumor subtype. Additionally, the treatment