PFKFB4 Overexpression Facilitates Proliferation by Promoting the G1/S Transition and Is Associated with a Poor Prognosis in Triple-Negative Breast Cancer

Background 6-Phosphofructo-2-kinase/fructose-2,6-biphosphate-4 (PFKFB4) is a key factor that plays an important role in tumorigenesis. However, its role in triple-negative breast cancer (TNBC) progression needs to be further validated. We investigated whether PFKFB4 is directly involved in the oncogenic signaling networks of TNBC. Methods First, we assessed the expression level of PFKFB4 in tumor tissue specimens by immunohistochemistry and evaluated its prognostic value. Next, the effect of PFKFB4 on TNBC cell growth and associated mechanisms were investigated. Finally, the results were further verified in vivo. Results We found that PFKFB4 overexpression was associated with an unfavorable prognosis in TNBC patients. PFKFB4 was overexpressed in TNBC cell lines in hypoxic environments, and its overexpression promoted tumor progression in vitro and in vivo. Further analyses demonstrated that the possible mechanism might be that PFKFB4 overexpression facilitates TNBC progression by enhancing the G1/S phase transition by increasing the protein level of CDK6 and phosphorylation of Rb. Conclusions These data suggest that PFKFB4 plays significant roles in the tumorigenesis and development of TNBC.


Introduction
Breast cancer is a global disease and one of the main causes of female morbidity and mortality [1,2]. Approximately 15% of breast cancers are defined as triple-negative breast cancer (TNBC), which lacks expression of the estrogen receptor (ER) and progesterone receptor (PR) and lacks overexpression of human epidermal growth factor 2 (HER2) [3,4]. TNBC is one of the most aggressive breast cancer subtypes, and patients with TNBC have a worse clinical outcome than those with other breast cancer subtypes [4]. The development of metastasis in TNBC represents a highly complex and poorly understood process. Further research on TNBC is urgently required to advance our treatment approaches.
The most common feature of all cancer cells is the production of large amounts of lactate and pyruvate, which is due to enhanced glycolysis despite the presence of oxygen. This phenomenon was first described by Warburg [5]. Hypoxia and hypoxia-inducible factors (HIFs) are the master regulators of metastasis in solid tumors, which are usually exposed to a hypoxic state due to their indefinite growth and nutrient deficiency [6]. HIF-1 is a heterodimer with the HIF-1β subunit constitutively expressed and the HIF-1α subunit regulated in an oxygen-dependent manner. HIF-1α is known to be hyperactivated in TNBC [7,8], while the mechanism and the target genes involved in the process by which HIF-1α regulates growth and metastasis in TNBC remain to be elucidated.
Hypoxia increases tumor glycolysis, angiogenesis, and cancer cell stemness, as well as invasion and metastasis. The activation of genes that increase the availability of oxygen, especially genes involved in glycolysis for maintaining cellular energy, is important in adaptations to hypoxia [9][10][11]. Glucose metabolism is regulated by fructose-2,6-bisphosphate, an allosteric activator of 6-phosphofructo-1-kinase [12,13]. A single family of bifunctional 6-phosphofructo-2kinase/fructose-2,6-biphosphate (PFK-2/FBPase-2 or PFKFB) enzymes is responsible for maintaining the cellular levels of fructose-2,6-bisphosphate [14], which has both kinase and phosphatase activities [15,16]. Within a few years of the initial discovery of fructose-2,6-bisphosphate, multiple tissue-specific mammalian PFKFB isoenzymes were identified in several organs [17]. PFKFB isoenzymes are encoded by four different genes (PFKFB1-4) in cells and are considered key factors in many malignant conditions [18][19][20]. These genes encode proteins that differ not only in their tissue distribution but also in their function [14,21]. PFKFB4 is located on chromosome 3 (bands p21-p22) and encodes an isoenzyme of PFKFB that was originally found in the testes [15]. Minchenko et al. suggested that PFKFB4 was induced by hypoxia in various cancer cell lines [19]. The overexpression of PFKFB4 in lung [22], breast, colon [18], and stomach cancer tissues [23] has been reported. Researchers have demonstrated the function of PFKFB4 in tumor growth by showing that silencing the gene inhibits the growth of human lung adenocarcinoma xenografts in athymic mice [24]. Independent studies have demonstrated that PFKFB4 is required for cancer cell survival [25,26] but not for normal cell survival. This aberrant expression indicates that this protein may play a key role in tumor development. Hence, these studies demonstrated that PFKFB4 might be a useful molecular marker and potential target for the development of cancer therapeutics.
In the present work, to investigate the role of PFKFB4 in the progression of TNBC, we detected the expression of PFKFB4 in tissue specimens from TNBC patients by  Figure 1: The expression of PFKFB4 in tumor tissue specimens and its prognostic value in TNBC patients. (a) Immunohistochemical analyses of PFKFB4 expression were performed with TNBC samples (scale bar: 100 μm). The protein staining was mainly distributed in the cytoplasm. PFKFB4 overexpression was associated with relatively poor OS (b) and DFS (c, * p < 0:05) by survival curve analysis. 2 Disease Markers immunohistochemistry (IHC). Next, we studied the effects of PFKFB4 on cell proliferation and tumor growth and associated mechanisms in vitro and in vivo. 2.2. Production of Tissue Microarrays (TMA), Immunohistochemistry, and Assessment. TMAs were constructed by extracting 2 mm diameter cores of histologically confirmed representative cancer cell areas from each original paraffin block according to a previously reported procedure [27]. Immunohistochemistry was performed on 4 mm sections of formalin-fixed, paraffin-embedded tissue samples that were dewaxed and rehydrated, and endogenous peroxidase activity was blocked with 0.3% H 2 O 2 in methanol. The slides were boiled in 10 mM sodium citrate (0.05% Tween, pH 6.0) at high power for 10 min and medium power for 20 min in a microwave for antigen retrieval. A primary anti-PFKFB4 antibody (ab137785, 1 : 400, Abcam, Cambridge, MA, USA) was used and incubated overnight at 4°C. Then, a goat anti-rabbit secondary antibody was used at room temperature for 30 min. Finally, hematoxylin was used to counterstain nuclei. A negative control (NC) was obtained by omitting the primary antibody. The specific intensities of staining were scored as follows: 0 = none; 1 = weak; 2 = moderate; and 3 = strong. The proportion scores were as follows: 0 = none; 1 = 1-10%; 2 = 11-50%; 3 = 51-80%; and 4 = >81% [28]. The scores for the percentage and intensity were multiplied to calculate an overall score, and the receiver operating characteristic (ROC) curve for the overall scores for PFKFB4 expression was plotted to select an appropriate cut-off score.

Materials and Methods
2.3. Cell Lines and Culture. The cell line used in this study was from the American Type Culture Collection and was identified by DNA (STR) profiling. MDA-MB-231 cells (a human breast cancer cell line, ER/PR/HER2 negative) were maintained in high-glucose (4.5 mg/ml) DMEM (Invitrogen, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA). Cells were cultured under normoxia in a humidified atmosphere containing 5% CO 2 at 37°C. For hypoxia, cells were exposed to an atmosphere composed of 5% CO 2 balanced with N 2 (<0.1% O 2 ) in a controlled-oxygen MIC-101 chamber (Billups Rothenberg, Inc., Del Mar, CA, USA), and the duration of exposure was 4-6 h.     2.8. Cell Cycle Analysis. Cells were harvested, washed with PBS 3 times, and then, fixed in cold 70% ethanol at 4°C overnight. The cells were centrifuged at 500 g and washed in PBS twice. The cells were stained with 500 μl propidium iodide (PI) solution (50 μg/ml) containing 100 μg/ml RNase A for 10 min in the dark and then analyzed by using a NovoCyte (ACEC Biosciences, Inc., USA).

Mouse Xenograft Model.
In vivo experiments were performed using female nude mice (4-to 5-week-old) purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The animals were kept in a temperature-and humiditycontrolled facility with a 12-h light/dark cycle with free access to food. Animal experiments were approved by the Sun Yatsen University Cancer Center ethics committee. To establish a breast cancer model, 1 × 10 5 MDA-MB-231 cells (0.1 ml) were injected subcutaneously (five mice in each group). Every 3-4 days, tumor size was measured, and body weight was recorded. On the 39th day postinjection, the mice were sacrificed, and the tumors and organs were carefully removed, weighed, and subjected to immunohistochemical staining for Ki67 (ZA-0502, ZSGB-Bio, China) and PFKFB4 (ab137785, Abcam, USA). Tumor volumes were calculated with the following formula: A × B 2 /2 (A: the longest diameter; B: the diameter perpendicular to A).
2.11. Statistical Analysis. Statistical tests were carried out using SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). Differences in mean values were evaluated using Student's t -test. Differences in frequencies were assessed with the chisquare test. Overall survival (OS) was calculated from the time of diagnosis to the day of death or the last date of follow-up. Disease-free survival (DFS) was defined as the duration between the time of disease diagnosis and the time of tumor relapse. Survival curves were plotted using the Kaplan-Meier method and compared with the log-rank test. Multivariate Cox proportional hazard models were used to define the potential prognostic significance of individual parameters. A ROC curve and the median were used to determine the cut-off value to distinguish high and low PFKFB4 expression. A two-sided p value less than 0.05 was considered significant.

Relationships between PFKFB4 Expression and Clinicopathological
Factors. There were 180 female TNBC patients who were defined as having stage I to IV disease on the basis of the American Joint Committee on Cancer TNM staging manual (8 th edition updates), with a mean age of 50 years (range of 17-78 years), enrolled in our study. The assessment of PFKFB4 is shown in Figure 1. The area under the ROC curve was 0.657 (p < 0:001, 95% confidence interval (CI) 0.574-0.740). An overall score of 5 maximized the Youden Index (sensitivity ð0:448Þ + specificity ð0:832Þ − 1 = 0:280), indicating 5 to be the optimal cut-off score. Thus, expression of the PFKFB4 protein with a score of 0-4 was designated "low or weak expression" and that with a score of 6-12 was designated "high or strong expression." The PFKFB4 protein was located in the cytoplasm (Figure 1(a)), and the high-expression rate of TNBC tissues was 26.7%. We further found important relationships between high levels of PFKFB4 expression in TNBC and disease relapse (p < 0:001) and death rates (p < 0:001) ( Table 1).

Prognostic Value of PFKFB4 Overexpression in TNBC
Patients. As demonstrated in Figure 1(b), a significant difference in 5-year overall survival (OS) was observed between the high PFKFB4 expression group Additionally, a statistically significant difference in 5-year diseasefree survival (DFS) was observed when patients were stratified by PFKFB4 expression (Figure 1(c), p < 0:001). In univariate and multivariate survival analyses, a high PFKFB4 protein level was an important prognostic factor for shortened OS (p < 0:001) and DFS (p < 0:001) (Tables 2 and 3). Thus, our findings indicate that PFKFB4 overexpression is significantly associated with the prognosis of TNBC.

PFKFB4 Overexpression
Facilitates the Proliferation of Breast Cancer Cells by Promoting the G1/S Phase Transition. To determine whether PFKFB4 is involved in the regulation of cell cycle progression, we saturated transfected MDA-MB-231 cells with PI, which stains the nuclear contents of a cell. Then, the cells were subjected to fluorescence-activated cell sorting (FACS). The findings showed that the percentage of cell number in S phase significantly increased in the 231 PFKFB4 group (25:8 ± 1:8%) (p < 0:01) compared with the 231 NC (16:1 ± 2:6%) and Vector groups (18:6 ± 1:7%) (Figures 2(g) and 2(h)). Furthermore, elevated levels of CDK6 and pRb Ser795, which are associated with the G1/S checkpoint, were observed in 231 PFKFB4 cells (Figure 2(i)). These combined results suggest that the PFKFB4 protein induces an increase in the frequency of cells in the S phase of the cell cycle in a TNBC cell line, resulting in the promotion of proliferation.

Discussion
In human cells, bifunctional PFKFB family members control the steady-state cytoplasmic levels of fructose-2,6-bispho-sphate, which activates the key enzyme (6-phosphofructo-1-kinase) in glycolysis [23]. PFKFB3 and PFKFB4 are the two primary isoenzymes overexpressed in various kinds of human cancers. PFKFB3 and PFKFB4 are widely involved in many biological processes, such as cell cycle regulation, autophagy, and apoptosis [29].
The results of this investigation indicate that the PFKFB4 protein is constitutively expressed in TNBC cells and that hypoxia significantly induces PFKFB4 expression in MDA-MB-231 cells. The regulation of expression appears to be related to a dependent mechanism that contains the activation of HIF-1 [30][31][32]. HIF-1 is a key factor that dominates the adaptation of cells to hypoxia and upregulates the expression of a series of genes involved in glycolysis. As shown in Figure 2(a), hypoxia increased the expression level of PFKFB4, and this phenomenon was correlated with an enhanced protein level of HIF-1α. These data suggest that hypoxic induction of PFKFB4 protein expression is mediated by HIF-1α. Since previous studies have demonstrated that the PFKFB4 gene is expressed in many tumor cell lines derived from different tissues [18,19,22], we provide consistent evidence that the testis isoform of the PFKFB protein is also expressed in TNBC cells.
Using multiple approaches, we demonstrated that (i) overexpression of PFKFB4 increased the proliferative ability of cancer cells; (ii) 231 cells expressing a high level of PFKFB4 exhibited increased resistance to cisplatin; and (iii) PFKFB4 expression at a high level promoted growth and invasion in vivo. Numerous studies have demonstrated the importance of PFKFB4 in cell malignancy [33,34], and depletion of PFKFB4 was shown to inhibit tumor growth in a xenograft model [25]. Taken together, these data provide possible evidence that a significant function of PFKFB4 is promoting the growth of TNBC cells. The current research is the first study to demonstrate that the expression of PFKFB4 is significantly associated with prognosis in TNBC patients. Our results suggest that PFKFB4, at the protein level, has strong predictive value and is sufficient to predict the risks of recurrence and progression in TNBC. PFKFB4 may be an important therapeutic target for the prevention of progression and needs to be further explored. Previous studies demonstrate that PFKFB4 is overexpressed in breast cancer [18], promotes breast cancer cell stemness [35] and metastasis [36], and drives a protein signature that correlates with poor survival in patients [37]. We propose that PFKFB4 has important functions in TNBC progression and can serve as a useful independent prognostic marker in the clinical setting.
Considering the central role of DNA metabolism in the evolution of TNBC, we tested the effects of PFKFB4 overexpression on cell cycle progression. This study found that PFKFB4 regulated the cell cycle by regulating the G1/S phase transition. Cyclin D1 interacts with CDK4/6 to phosphorylate Rb (pRb), causing E2F to dissociate from the Rb-E2F complex, which is essential for DNA replication. Therefore, phosphorylating Rb allows the release of S phasepromoting transcription factors and is indicative of cell proliferation. The increase in the pRb level is essential for the G1/S phase transition [38,39]. Palbociclib, an inhibitor of CDK4 and CDK6, was approved by the FDA in 2015 for 8 Disease Markers the treatment of breast cancer. Because of the key role of CDK6 in PFKFB4-overexpressing tumors, palbociclib administration is considered an effective strategy for this type of cancer. In addition, Ki67, a protein marker of cell proliferation, is expressed only in the G1, S, and G2 phases, not in the G0 phase, and is a prognostic factor in breast cancer [40]. More Ki67 protein expression was observed in the PFKFB4 overexpression nude mouse group than in the control mouse groups in this study. For all of the reasons mentioned above, we propose that PFKFB4 overexpression facilitates the proliferation of TNBC cells by enhancing the G1/S transition by increasing the CDK6 level and phosphorylating Rb. Although this is the first analysis to support the conclusion that the kinase activity of PFKFB4 is essential for TNBC progression, we discuss a specific mechanism that differs from the mechanisms identified in other studies. Previous studies identified that reduced lactate secretion and intracellular ATP levels were observed in malignant cells when PFKFB4 was silenced [26,33], and the induction of lipid synthesis, which is required for cancer cell growth, was inhibited, possibly by reducing NADPH availability [41]. Moreover, other studies found that PFKFB4 might control cell survival via Akt signaling and regulate caspase 3 or 7 activity and the levels of reactive oxygen species [33,34]. Minchenko et al. indicated that overexpression of PFKFB4 in malignant tumors correlated with enhanced expression of HIF-1a, glucose transporter 1 (Glut1), and vascular endothelial growth factor (VEGF), which are known HIF-1-dependent genes [18,42]. Furthermore, in breast cancer, it was discovered that the enzyme PFKFB4 activates the transcriptional coactivator SRC-3 to drive the occurrence and development of tumors [37].

Conclusions
In summary, we demonstrated that PFKFB4 was overexpressed in TNBC cell lines in hypoxic environments; the overexpression of PFKFB4 promoted cell proliferation, clone formation, and drug resistance in vitro and tumorigenicity in vivo; and high PFKFB4 expression was significantly associated with a poor prognosis in TNBC patients. Furthermore, PFKFB4 overexpression might facilitate TNBC progression by enhancing the G1/S transition by increasing the CDK6 level and phosphorylating Rb. We believed that inhibiting PFKFB4 could be an effective strategy for TNBC treatment and that PFKFB4 suppression in combination with a CDK4/6 inhibitor needs further exploration.

Data Availability
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Ethical Approval
The current study was approved by the Ethics Review Board of Sun Yat-Sen University Cancer Center. All activities were in accord with the 1964 Declaration of Helsinki.