Flotillin-2 promotes cell proliferation via activating the c-Myc/BCAT1 axis by suppressing miR-33b-5p in nasopharyngeal carcinoma

Previously, we elucidated the function of flotilin-2 (FLOT2) and branched-chain amino acid transaminase 1(BCAT1) in nasopharyngeal carcinoma (NPC). However, the relationship between FLOT2 and BCAT1 in promoting NPC progression remains unknown. Here, we observed that FLOT2 upregulated BCAT1 expression in NPC cells. Ectopic expression of BCAT1 significantly antagonized the inhibitory effects on NPC cell proliferation induced by FLOT2 depletion. Consequently, BCAT1 knockdown markedly inhibited the pro-proliferative effects of FLOT2 overexpression in NPC cells. FLOT2 expression was positively correlated with BCAT1 expression in NPC tissues and was inversely correlated with the prognosis of NPC patients. Mechanistically, FLOT2 maintains the expression level of c-Myc, a positive transcription factor of BCAT1, and subsequently promote BCAT1 transcription. FLOT2 inhibited miR-33b-5p in NPC cells and attenuated its inhibitory effects on c-Myc. Further, experimental validation of the function of the FLOT2/miR-33b-5p/c-Myc/BCAT1 axis in regulating NPC cell proliferation was performed. Our results revealed that FLOT2 promotes NPC cell proliferation by suppressing miR-33b-5p, to maintain proper levels of c-Myc, and upregulate BCAT1trancription. Therefore, the FLOT2/miR-33b-5p/c-Myc/BCAT1 axis is a potential therapeutic target for NPC.


INTRODUCTION
Nasopharyngeal carcinoma (NPC) is one of the most common head and neck cancer, that is highly prevalent in southern China and Southeast Asia [1,2]. Radiation therapy is the primary treatment option, with a high cure rate in early-stage NPC [1,3]. Unfortunately, most NPC patients are diagnosed at advanced stages, characterized by peripheral invasion and lymph node and distant metastasis. Besides, advanced stage NPC patients lack distinguishable symptoms and early diagnostic markers.
Radiotherapy and chemotherapy show synergistic effects on killing NPC cells, however, the treatment outcome of advanced-stage NPC patients is far from satisfactory indicating poor prognosis [1][2][3][4]. Thus, to reveal the molecular mechanism involved in the development and progression of NPC, and subsequently identify potential prognostic markers of NPC is of great significance.
FLOT2 is a lipid raft scaffold protein that is highly ordered membrane subdomains and plays a critical role in maintaining the homeostasis of lipid rafts and BCAT1, a cytosolic branched-chain aminotransferase, catalyzes the conversion of branched-chain amino acids, including leucine, isoleucine, and valine, into their corresponding branched-chain a-keto acids (BCAAs) by transferring an amino group onto a-ketoglutarate to generate glutamate [14]. Numerous studies reveal that activating BCAT1 enhance the tumor-driving andpromoting roles of BCAAs metabolism in gliomas and myeloid leukemia [15][16][17][18]. Moreover, BCAT1 upregulation and its oncogenic role have been observed in other tumors such as hepatocellular, gastric, breast, ovarian, endometrial, prostate, and esophageal squamous cell carcinomas [14,[19][20][21][22][23][24]. Our previous study also demonstrated that the transcription factor c-Myc upregulates BCAT1 expression to promote tumor initiation and progression in NPC. Besides, Microarray analysis indicated that FLOT2 depletion significantly downregulated BCAT1 expression in NPC cells, suggesting that BCAT1 may be involved in the protumor roles of FLOT2 in NPC.
Therefore, in the present study, we investigated the FLOT2 regulation of BCAT1 in NPC. The results demonstrated that FLOT2 activated BCAT1 through c-Myc regulation of BCAT1 transcription, thereby promoting NPC proliferation. miR-33b-5p, a direct negative regulator of c-Myc, was inhibited by FLOT2, subsequently suppressing c-Myc mRNA degradation, and promoting BCAT1 expression. Moreover, FLOT2, c-Myc, and BCAT1 were positively correlated in NPC tissues, and FLOT2/c-Myc/BCAT1 axis was positively correlated with NPC progression and negatively associated with the prognosis of NPC patients. Therefore, our findings suggest that FLOT2/miR-33b-5p/ c-Myc/BCAT1 axis is a potential therapeutic target in NPC.

FLOT2 positively regulates BCAT1 expression in NPC cells
Previously, we used cDNA microarray (GSE67456) to explore the effects of FLOT2 knockdown on the transcription profile in NPC cells [13]. The results showed that BCAT1 was downregulated in 5-8F-shFLOT2 cells. Thus, the expression of BCAT1 was further validated in NPC cells by altering the expression of FLOT2. qPCR and western blot analysis ( Figure 1A, 1B) revealed that FLOT2 knockdown significantly decreased BCAT1 expression, while FLOT2 overexpression remarkably increased BCAT1 expression in 5-8F and 6-10B cells, respectively. FLOT2 expression levels in NPC cells were also determined by altering the expression of BCAT1 alteration. The results showed that both BCAT1 knockdown and overexpression had no effects on FLOT2 expression (Supplementary Figure 1).

FLOT2 and BCAT1 are positively correlated in NPC and predicts poor prognosis of NPC patients
The correlation between FLOT2 and BCAT1 was explored, as well as their clinical and prognostic value in NPC cells. IHC staining showed that FLOT2 expression had a positive correlation with BCAT1 expression in NPC cells (r = 0.57) ( Figure 1C, 1D). Moreover, FLOT2 and BCAT1 expression were positively correlated with several pathological variables, including T, N, M, and TNM stages (Table 1). Besides, FLOT2 and BCAT1 expression levels were negatively correlated with the OS of NPC patients ( Figure 1E). Furthermore, univariate Cox proportional hazards regression analysis revealed that primary T stage, TNM stage, and FLOT2 and BCAT1 expression levels were significantly related to the OS of NPC patients (Table 2). Multivariate Cox proportional hazards regression analysis demonstrated that FLOT2 and BCAT1 were independent predictors of reduced OS of NPC patients (Table 2). Therefore, these results indicate that the FLOT2 and BCAT1 axis play some important roles in NPC progression and poor prognosis in patients with NPC.

FLOT2 promotes cell proliferation both in vitro and in vivo via BCAT1 in NPC
Next, we explored whether BCAT1 mediated FLOT2 function in NPC. Firstly, western blot analysis successfully demonstrated ectopic expression of BCAT1 in 5-8F-shFLOT2 cells and BCAT1 depletion in 6-10B-FLOT2 cells (Figure 2A). In vitro and in vivo AGING experiments determined the effects of BCAT1 alterations on cell proliferation. CCK-8, plate clone formation, and EdU incorporation assays revealed that BCAT1 restoration significantly rescued the inhibitory effects of FLOT2 knockdown on 5-8F cell proliferation ( Figure 2B, left arm; 2C upper arm; 2D left arm). Depletion of BCAT1, antagonized the pro-proliferation effects of FLOT2 overexpression on 6-10B cells ( Figure 2B, right arm; 2C low arm; 2D right arm). Consequently, ectopic expression of BCAT1 remarkably removed the inhibition of FLOT2 silencing on subcutaneous growth of 5-8F cells ( Figure 2E-2G, left arm); while BCAT1 knockdown significantly counteracted the activation of FLOT2 overexpression on subcutaneous growth of 6-10B cells ( Figure 2E-2G, right arm). Therefore, these results indicated that FLOT2 promoted the proliferation of NPC cells via positive regulation of BCAT1 both in vitro and in vivo.
The pro-proliferation effect of BCAT1 depends on its enzymatic activity in NPC BCAT1 function depends on its enzymatic activity [17]. Therefore, BCAT1 activity was inhibited with gabapentin (GABA), a γ-aminobutyric acid analogue and a competitive inhibitor of BCAT1, to determine its effects on NPC cell proliferation. Firstly, NPC cell viability was measured using CCK-8 assay, at different gabapentin concentrations. As shown in Figure 3A, GABA inhibited the growth of 5-8F cells expressing high levels of BCAT1, in a dose-dependent manner, but exerted little influence on the growth of 6-10B cells expressing low levels of BCAT1. We further explored the inhibitory effects of GABA on NPC cell proliferation. Consistently, the inhibitory effects of GABA on NPC cell proliferation were further supported by results of the plate clone formation ( Figure 3B) and EdU incorporation assays ( Figure 3C). Besides, GABA did not influence the protein expression level of BACT1 and FLOT2 ( Figure  3D). However, CCK-8 ( Figure 3E), plate clone formation ( Figure 3F), and EdU incorporation assays ( Figure 3G) showed that GABA treatment significantly antagonized the rescue effects of BCAT1 on 5-8F-shFLOT2 cell proliferation. Taken together, these results suggested that the pro-proliferation role of BCAT1 highly depends on its enzymatic activity in NPC.

FLOT2 upregulates BCAT1 and promotes NPC cell proliferation via c-Myc regulation
Our previous study confirmed that c-Myc promoted BCAT1 transcription in NPC cells [25]. Therefore, in the    AGING current study, we explored whether FLOT2 increased BCAT1 expression via c-Myc. Indeed, both mRNA and protein expression levels of c-Myc were downregulated in 5-8F-shFLOT2 cells and upregulated in 6-10B-FLOT2 cells ( Figure 4A), respectively. Moreover, ectopic expression of c-Myc restored BCAT1 expression which was decreased by siBCAT1 in 5-8F-shFLOT2 cells ( Figure 4B, left arm). c-Myc knockdown in 6-10B-FLOT2 downregulated BCAT1 expression levels, which was restored by ectopic expression of BCAT1 ( Figure  4B, right arm). Similarly, functional experimental results revealed that the proliferative capacity of 5-8F-shFLOT2 cells was restored by ectopic expression of c-Myc ( Figure  4C, upper arm; 4D, 4E, left arm), which was suppressed by BCAT1 knockdown. c-Myc knockdown inhibited 6-10B-FLOT2 cell proliferation, which was restored after ectopic expression of BCAT1 ( Figure 4C, low arm, 4D, 4E, right arm). Furthermore, in our previous study, we investigated the effects of BCAT1 and c-Myc depletion on AKT and NF-κB activity, which mediates FLOT2 oncogenic role in NPC [13]. Both BCAT1 and c-Myc knockdown significantly inhibited AKT and NF-κB activity, which was confirmed by a decreased in p-AKT and p-p65 expression (Supplementary Figure 2A) in NPC cells. Considering that inhibition of AKT and NF-κB activity by MK2206 and BAY 11-7082, respectively, exert no notable effects on the expression levels of BCAT1 and c-Myc (Supplementary Figure 2B), suggests that AKT and NF-κB could serve as the downstream signaling of c-Myc/BCAT1 in mediating the functions of FLOT2 in NPC. Therefore, these results demonstrated that FLOT2 upregulated BCAT1 and promoted NPC cell proliferation via positive regulation of c-Myc.

DISCUSSION
Previously, we elucidated the pro-neoplastic role of FLOT2 and BCAT1 in NPC [13,25,31]. In this study, we explored the relationship between FLOT2 and BCAT1 and determined that FLOT2 upregulated BCAT1 expression, thereby promoting NPC cell proliferation. Mechanistically, we found that FLOT2 suppressed miR-33b-5p levels and target inhibition of miR-33b-5p on c-Myc, and upregulated BCAT1 expression. The results reveal that FLOT2/miR-33b-5p/c-Myc/BCAT1 is an oncogenic axis that is of potential clinical significance in NPC treatment.
Current studies have established the oncogenic roles of FLOT2 in cancers. Upregulation of FLOT2 promotes the proliferation, differentiation, migration, invasion, metastasis, and therapy resistance of cancer, and positively correlates with poor prognosis in patients with cancer [10,12,13,32]. The activation of signaling pathways, such as NF-κB, Raf/MEK/ERK1/2, TGF-β, and PI3K/AKT, drive process including suppression of apoptosis, cell cycle acceleration, and epithelial-mesenchymal transition (EMT), which accounts for mechanisms underlying the oncogenic function of FLOT2 [8,9,[11][12][13]33]. However, detailed mechanisms of how FLOT2 activates oncogenic signaling and other mechanisms mediating the functions of FLOT2, need to be further investigated. A recent study indicated that PLCD3 is directly associated with FLOT2 to promote the progression of NPC cells, however, the study failed to illustrate the specific role of PLCD3 in FLOT2's proneoplastic function in NPC [34]. In this study, we revealed that FLOT2 upregulated BCAT1 expression and activity, and in turn promoted the malignant progression of NPC, thus, presenting a new oncogenic mechanism of FLOT2. Besides, the results suggested that BCAA metabolism is involved in NPC development.
Dysregulation of BCAA metabolism contributes to tumorigenesis and progression in multiple cancers [14]. Besides, aberrant activation and upregulation of BCAT1, a critical enzyme in BCAA metabolism, serves as the main underlying molecular mechanism of reprogramming BCAA metabolism in cancer. BCAT1 overexpression enhances BCAA catabolism, which subsequently promotes the proliferation and stemness of cancerous cells in gliomas with wild-type IDH1, myeloid leukemia, and endometrial cancer [15][16][17][18]35]. Besides, BCAT1 overexpression is reported in other cancers, where BCAT1 functions as an oncogene, stimulates malignant progression and is negatively correlated with the prognosis of patients [14,19,20,[22][23][24]. Our previous study revealed that BCAT1 overexpression promotes cell proliferation, migration, and invasion in NPC [25,31]. Consistently, this study confirmed the oncogenic role of BCAT1 in NPC and further revealed that BCAT1 was an independent marker for poor prognosis of NPC patients.  [27,28]. The tumor-suppressive roles of miR-33b-5p have also been reported in other cancers including gastric cancer, colorectal cancer, and renal cell carcinoma [40][41][42]. The AGING current study also confirmed the anti-tumor functions of miR-33b-5p in NPC.
Our previous study confirmed that FLOT2 knockdown significantly decreased Snail level in NPC cells [13], suggesting that FLOT2 suppresses miR-33b-5p through the sustained expression of Snail1 in NPC, however, these findings will be validated through additional future experiments.
In conclusion, our study reveals that FLOT2 suppresses miR-33b-5p, subsequently attenuating the inhibitory effects of miR-33b-5p on c-Myc, leading to BCAT1 upregulation to promote NPC cell proliferation. These findings suggest that FLOT2/miR-33b-5p/c-Myc/BCAT1 axis is a potential therapeutic target in NPC.

Patients and tissue samples
103 cases of paraffin-embedded and poorly differentiated squamous NPC tissue samples were collected from primary patients without any therapy during Jan 2010 to Dec 2012 in Xiangya Hospital of Central South University. The clinical stage of the patients was classified according to the seventh edition of AJCC guidelines as our previous illustration [26]. The patients uniformly received radiochemotherapy and the clinicopathological features, prognostic conditions of patients were retrospectively collected. Overall survival (OS) was defined as the time from the initiation of primary therapy to the date of cancer-related death or when censured at the latest date if patients were still alive.

Western blot
Western blot was carried out as described previously by us [3,26]. Briefly, the total proteins were collected from RIPA cell lysis by centrifugation. The proteins were denatured and separated by SDS-PAGE and subsequently transmitted into 0.22μm PVDF membrane.

RNA isolation and quantitative real-time polymerase chain reaction(qPCR)
The RNA extraction and qPCR assay were carried out according to our previous reports [8,12]. Briefly, total AGING RNAs were extracted from the indicated NPC cells with Trizol reagent (Thermo Fischer Scientific

CCK-8
Cell proliferation was analyzed using a CCK-8 kit (Beyotime, Shanghai, China) as our previous description [3,26]. The assay was performed three times in triplicate.

Plate clone formation
Plate colony formation assay was carried out according to our previous description [26]. The assay was applied three times in triplicate.

5-ethynyl-2′-deoxyuridine (EdU) incorporation assay
EdU incorporation assay was performed to detect cell proliferation as described previously by us [26]. The assay was performed three times in triplicate.

Immunohistochemistry (IHC)
The level of FLOT2 and BCAT1 in NPC tissues were detected with SP Rabbit & Mouse HRP Kit (ZSGB-BIO, Beijing, China) as previously described by us [26]. Simply, after antigen retrieval, removal of endogenous peroxidase activity, and blocking, tissue sections were incubated with rabbit anti-FLOT2 antibody (D225408, dilution 1:50, BBI, Shanghai, China) and rabbit anti-BCAT1 antibody (D121976, dilution 1:200, BBI, Shanghai, China) overnight at 4° C. Next, the sections were incubated with biotinylated secondary antibody followed by avidin-biotin peroxidase complex. Finally, tissue sections were stained with 3′, 3′-diaminobenzidine (DAB, ZSGB-BIO, Beijing, China) and counterstained with hematoxylin. Immunohistochemical staining was independently evaluated by two pathologists. The protein expressions were scored based on the staining intensity and area. The percentage of stained cells was categorized as no staining = 0, < 30% of stained cells = 1, 30~60% = 2, and > 60% = 3. The evaluation of staining intensity was as follow: absent staining as 0, weak as 1, moderate as 2, and strong as 3. The staining score (ranging from 0-6) for each tissue was calculated by adding the area score and the intensity score. A combined staining score of ≤3 was considered to be low expression, and > 3 was considered to be high expression.

In vivo proliferation assay
In vivo proliferation assay was analyzed as our previous illustration [26,48]. Briefly, female nude Balb/c mice, 4 weeks old, were purchased from the Laboratory Animal Center of Central South University (Changsha, China) and were maintained under specific pathogen-free conditions. For tumor formation experiment, 2×10 6 cells were subcutaneously injected into the flanks of mice (n=5 mice each). The length and width of xenografts were detected with caliper every 4 days. Tumor volume (in mm3) was calculated by using the modified ellipse formula (volume = length × width2/2). 20 days later, the mice were killed by cervical dislocation, and their tumors were excised and weighted.

Ethics statement
The usage of human tissues was approved by the Ethics Committee of Xiangya Hospital, Central South University. Considering only archived tumor specimens were enrolled in this study, the ethics committee waived the need for consent, and the patient records/information were analyzed anonymously. The animal experimental procedures were carried out following the Guide for the Care and Use of Laboratory Animals of Xiangya Hospital, Central South University, with the approval of the Institutional Animal Ethics Committee.

Statistical analysis
All experiments were independently repeated at least 3 times. Statistical analyses and charts were conducted by using IBM SPSS statistical software package 26 (IBM, NY, USA) and GraphPad Prism 8 (GraphPad, CA, USA). Student's t test or Chi-square test were used to compare statistical difference between two groups. Kaplan-Meier survival analysis was used to compare AGING NPC patient survival by the log rank test. Cox proportional hazards regression analyses were used to analyze the effect of clinical variables on patient survival. The Spearman rank correlation coefficient was used to determine the correlation between the two variables. All statistical tests were two-sided, and the P values <0.05 were considered to be statistically significant.

AUTHOR CONTRIBUTIONS
The study was conceptualized and supervised by Wei