PFKP Accelerates Malignant Features in Breast Cancer

Masahiro Shibata (  m-shibata@med.nagoya-u.ac.jp ) Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin https://orcid.org/0000-00031028-2796 Takahiro Inaishi Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Takahiro Ichikawa Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Mitsuro Kanda Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Masamichi Hayashi Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Ikumi Soeda Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Dai Takeuchi Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Yuko Takano Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Nobuyuki Tsunoda Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Yasuhiro Kodera Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin Toyone Kikumori Nagoya University Hospital: Nagoya Daigaku Igakubu Fuzoku Byoin


Introduction
Breast cancer (BC) is the most common malignant tumor among women throughout the world [1]. The development of adjuvant therapy improved the prognosis of patients with BC. Indeed, the 5-year overall survival (OS) rate of BC patients without metastasis currently exceeds 80% [2]. However, 20-30% of patients with BC develop metastases after primary tumor treatment [3]. Patients with recurrent BC are classi ed according to the immunohistochemical detection of conventional target molecules such as the estrogen receptor (ER), progesterone receptor (PgR), and human epidermal growth factor receptor 2 (HER2). Although various drugs have been developed and are available for the treatment of patients with recurrent BC, they are still insu cient to cure and only 5% of those patients achieve long-term disease control [4]. From this point of view, development of new biomarkers or therapeutic target molecules for the purpose of improving the prognosis of BC patients is demanded.
Phosphofructokinase (PFK), which catalyzes the formation of fructose 1,6-bisphosphate and adenosine diphosphate from fructose 6-phosphate and adenosine triphosphate, is one of the key regulating enzymes in the glycolytic pathway [5]. PFK is a complex tetrameric enzyme that has three isoforms: liver (PFKL), muscle (PFKM), and platelet (PFKP) [6]. The activity of PFK is regulated by quantitative and isozymic changes secondary to altered gene expression during neoplastic transformation [7]. Among the three isoforms, the expression and regulatory mechanisms of PFKP have been studied in several malignancies, including brain tumor, renal cancer, and bladder cancer, in which the increased expression of PFKP has been associated with the progression of cancer cells [8][9][10].
In BC cells, hypoxia inducible factor 1 subunit alpha, a major transcriptional regulator of the cellular response to hypoxia, and kruppel-like factor 4, a transcription factor that regulates the expression of several genes involved in cell cycle regulation and differentiation, activate the transcription of PFKP and enhance glycolytic metabolism [7,11]. However, whether PFKP promotes malignant features in BC has not been evaluated. This study aimed to investigate the functional roles of PFKP in BC cells and the signi cance of PFKP expression in patients with BC. Mitsubishi Chemical Medicine Corporation, Tokyo, Japan) and cultured in RPMI-1640 (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (FBS) and incubated in an atmosphere of 5% carbon dioxide at 37°C [12].

Materials And Methods
We also collected primary BC and non-cancerous specimens from 167 patients histologically diagnosed with BC after undergoing surgery at Nagoya University Hospital from March 2002 to May 2007. Surveillance data for more than ve years after surgery for all 167 patients were available. All specimens were immediately resected to a diameter of approximately 1.5 mm and stored at -80°C. Non-cancerous specimens were resected ≥3 cm from the edge of the tumor [13]. All specimens were histologically diagnosed as BC and classi ed using the Union for International Cancer Control (UICC) staging system (8th edition). Postoperative adjuvant therapy was determined on the basis of the patient's condition, pathological features, cancer subtype, and physicians' discretion [13].

PCR array analysis
To determine the correlation between the expression levels of PFKP and 84 cancer-related genes in BC cell lines, PCR array analysis was conducted using RT 2  Control siRNA Fluorescein-labeled (Cosmo Bio Co. Ltd., Tokyo, Japan) served as control nontargeting siRNA, named "siControl". BC cells were seeded in antibiotic-free RPMI-1640 supplemented with 10% FBS; 24 h after seeding, cells were transfected the corresponding siRNAs in the presence LipoTrust EX Oligo (Hokkaido System Science). After transfection, cells were cultured in antibiotic-free RPMI-1640 with 10% FBS for 72 h. Knockdown e ciency was determined using qRT-PCR.

Simple Western analyses
Simple Western analyses was performed using the WES machine (ProteinSimple, San Jose, CA, USA) according to the manufacturer's recommendations. Cells were incubated in RIPA lysis buffer, and the lysates were stored at -30°C. Protein concentration was measured using a BCA protein assay Kit (Thermo Fisher Scienti c, Waltham, MA, USA). Protein samples, biotin ladder, primary antibody, secondary antibody, blocking reagent, chemiluminescent substrate, and wash buffer were prepared and dispensed into the assay plate. Then, the assay plate was loaded into the instrument, and the protein was separated into individual capillaries. Protein separation and detection was performed automatically on individual capillaries. Anti-PFKP antibody (1:50 dilution) (Cell Signaling Technology, Beverly, MA, USA) and anti-beta actin antibody (1:50 dilution) (Abcam, Cambridge, UK) were used as primary antibodies.
Streptavidin HRP and anti-mouse or anti-rabbit secondary antibodies (ProteinSimple) were selected according to the corresponding primary antibody [14,15].

Kaplan-Meier survival analysis using Kaplan-Meier Plotter
We used the website of the Kaplan-Meier Plotter (http://kmplot.com/analysis/index.php?p=background) to analyze relapse-free survival (RFS) and OS for patients with BC with respect to expression of PFPK by classifying its expression levels into the upper quartile and others [16].

Statistical analysis
Numeric variables between two groups were compared using the Mann-Whitney test. Spearman's rank correlation test was performed to evaluate the correlation between PFKP and cancer-related gene expression levels in the PCR array analysis. We analyzed the association between PFKP mRNA expression and clinicopathological factors using the χ 2 test. Disease-free survival (DFS) and OS were calculated using the Kaplan-Meier method, and survival curves were compared using the log-rank test. All statistical analyses were performed using JMP 15 software (SAS Institute Inc., Cary, NC, USA), and statistical signi cance was de ned as P < 0.05.

Results
PFKP mRNA expression levels in BC and non-cancerous cell lines and its association with cancer-related genes in BC cell lines PFKP mRNA expression levels in 13 BC cell lines and two non-cancerous cell lines are shown in Fig. 1a. ER, PgR, and HER2 statuses of the cell lines have been evaluated in previous studies [17,18]. PFKP mRNA expression levels in ER-negative cell lines were signi cantly higher than those in ER-positive BC cells (P = 0.003). In addition, PFKP in triple-negative cell lines showed higher mRNA expression levels than that in the other cell lines. (P = 0.038). Subsequent PCR array analysis showed that PFKP mRNA expression levels were positively correlated with those of several well-known oncogenes, such as transforming growth factor beta 1 (TGFB1) (correlation coe cient 0.758, P = 0.003) and MYC proto-oncogene (MYC) (correlation coe cient 0.648, P = 0.017) (Fig. 1b). The correlation between PFKP mRNA expression levels and those of 84 cancer-related genes is shown in Supplementary Table S1.

Effects of PFKP knockdown in various BC subtypes
Considering the results of PFKP mRNA expression levels, PFKP protein expression was evaluated in representative BC cell lines. Among these cell lines, MCF7 represents the ER-positive/HER2-negative subtype, SK-BR-3 represents the ER-negative/HER2-positive subtype, and MDA-MB-231 represents the triple-negative subtype. HCC1419, which expressed the lowest mRNA expression level, was used as negative control (Fig. 2a). PFKP knockdown was con rmed at both mRNA and protein expression levels ( Fig. 2b and 2c).
To determine the tumor-progressive roles of PFKP in BC cells, cell proliferation, invasiveness, and migration were evaluated in the knockdown cells. Compared with the untransfected and siControltransfected cells, proliferation was signi cantly inhibited in siPFKP-transfected MCF7 and SK-BR-3 cells during the entire study period (P < 0.05). Proliferation of MDA-MB-231 cells transfected with siPFKP resulted signi cantly inhibited on days 4 and 5 (P < 0.05; Fig. 3a). In the invasiveness assay, fewer siPFKP-than siControl-transfected or untransfected MCF7, SK-BR-3, and MDA-MB-231 cells passed the Matrigel (P < 0.001; Fig. 3b). Moreover, the migration ability of SK-BR-3 and MDA-MB-231 cells was inhibited after siPFKP transfection (P < 0.01; Fig. 3c). siPFKP-transfected MCF7 cells did not show enough proliferation to perform the migration assay, as shown in Fig. 3a.
We grouped the patients in the highest quartile of PFKP mRNA expression into a "High PFKP group" (n = 42) and the remaining patients into a "Others" (n = 125). The association between clinicopathological factors and PFKP expression is shown in Table 1. As expected, the high PFKP group included more patients with T2/T3/T4 (P = 0.023) and with more advanced UICC pathological stages (P = 0.001). In addition, the high PFKP group had more ER-negative and PgR-negative patients than the others (P = 0.004 and P < 0.001, respectively).
When prognosis was evaluated in our cohort, there were no signi cant differences in terms of DFS or OS between these two groups (Fig. 5a). Because the small sample size was concerned in our cohort, we subsequently investigated the impact of PFKP expression on prognosis using the Kaplan-Meier Plotter website. Similarly, when patients were assigned either to the upper quartile (High PFKP group) or to other quartiles (Others), the high PFKP group showed signi cantly worse RFS (n = 3951; P < 1E-16) and OS (n = 1402; P = 2.6E-06; Fig. 5b).

Discussion
This study demonstrated that PFKP expression contributes to tumor progression by promoting cellular proliferation, invasiveness, and migration in various subtypes of BC cell lines. Furthermore, analysis of clinical samples showed that PFKP mRNA expression levels were higher in patients with advanced pathological stage, which supported our in vitro results.
The activity of glycolytic enzymes, such as hexokinase, PFK, and pyruvate kinase, is several folds higher in cancer cells than that in normal cells [19,20]. PFKP upregulation increases glycolytic ux and promotes tumor cell proliferation and tumor growth [9]. In hepatocellular carcinoma, PFKP is regulated by Tatactivating regulatory DNA-binding protein via microRNA 520 [21]. In glioblastoma, phosphorylation of PFKP S386 via AKT activation promotes aerobic glycolysis and tumor growth [8]. In addition to the transcription factors that directly upregulate PFKP, there is crosstalk between glycolysis and oncogenic signaling [22]. From these notions, we investigated the expression and functional roles of PFKP in BC.
Regarding PFKP mRNA expression levels in BC cell lines and non-cancerous cell lines, ER-negative BC cell lines had signi cantly higher PFKP mRNA expression levels than ER-positive BC cell lines. In addition, triple-negative BC cell lines expressed higher levels of PFKP mRNA than the other cell lines. Similarly, analysis of our clinical samples demonstrated that PFKP mRNA expression levels in ER-negative patients were signi cantly higher than those in ER-positive patients, and its expression levels in patients testing negative for PgR were also signi cantly higher than those found in patients with PgR-positive results. These results are consistent with a previous report showing that triple-negative BC is more dependent on glycolysis by upregulating several key glycolytic enzymes and transporters, including PFK and the glucose transporter [22]. To analyze the interactions between PFKP and several oncogenic signaling pathways, we investigated the correlation between the expression levels of cancer-related genes and those of PFKP in BC cell lines using a PCR array. Accordingly, TGFB1 and MYC were coordinately expressed with PFKP in BC cell lines. In fact, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase 3 (PFKFB3) promotes the synthesis of fructose 2,6-bisphosphate, which is the most potent allosteric stimulator of PFKP, and induced by TGFB1, it is involved in the activation of glycolysis observed in glioblastoma [23]. Myc suppresses the level of thioredoxin-interacting protein, which is a negative regulator of glucose uptake and glycolysis gene expression, and activates aerobic glycolysis in BC [24]. Although further mechanistic investigation is warranted, these results would provide important insights into understanding the involvement of PFKP in signaling pathways associated with BC progression.
In this study, PFKP inhibition suppressed cellular proliferation, invasiveness, and migration in various subtypes of BC cell lines, such as MCF7, SK-BR-3, and MDA-MB-231. Because PFKP protein expression in MDA-MB-231 cells was lower than that in MCF7 and SK-BR-3 cells, as shown in Fig. 2a, the impact of PFKP inhibition on cell proliferation in MDA-MB-231 cells was also lower than that in MCF7 and SK-BR-3 cells. In clinical samples, PFKP expression levels were higher in patients with larger tumor sizes, positive lymph node metastases, or more advanced stages. A previous study on PFK isoenzyme patterns in BC tissue revealed a positive correlation between increased pathological stages and the expression of PFKP [25], suggesting that PFKP is involved in promoting the malignant phenotype of BC regardless of the BC subtype. Regarding prognosis, although there was no signi cant difference in DFS or OS between the high PFKP group and others in our cohort, the analysis using the public database demonstrated that patients with high PFKP expression showed poorer RFS and OS. This discrepancy could be due to the small sample size in our cohort and the impact of adjuvant therapy. In summary, our results suggest that PFKP promotes malignant cellular features and contributes to a more advanced pathological stage, which leads to poor prognosis. Noticeably, there is no drug approved for BC that target glycolytic enzymes. These results suggest that PFKP could be a new therapeutic target molecule in BC.
This study had some limitations. First, the mechanism of PFKP expression involved in tumor progression has not been fully elucidated. Second, as noted above, due to the small number of patients in our study and use of adjuvant medication therapy such as endocrine therapy, chemotherapy, and molecular targeted therapy, the results of the prognostic analysis in our cohort data did not coincide with those using public databases. Finally, further in vivo studies are required to prove the potential therapeutic targets of PFKP.
In conclusion, this study showed the tumor-progressive roles of PFKP in various subtypes of BC cells expressing PFKP. These data support the possibility of PFKP as a therapeutic target in BC.