Fatty acid synthase (FASN), a novel signature for visceral adipose tissues, regulates proliferation, migration and predicts prognosis of clear cell renal cell carcinoma

Background: Growing evidence has indicated obesity one of the important etiological indicators of clear cell renal cell carcinoma (ccRCC). This study aims to investigated FASN mRNA expression in anthropometric adipose tissue, and elucidate prognostic value and potential functions in ccRCC patients. Methods: Transcriptional expression profiles was obtained from 380 paired ccRCC and adjacent normal tissues from Fudan University Shanghai Cancer Center (FUSCC) and 533 ccRCC samples from the Cancer Genome Atlas (TCGA) cohort. The visceral adipose tissue (VAT) and the subcutaneous adipose tissue (SAT) at the level of the umbilicus were measured using magnetic resonance imaging. The Kaplan-Meier method and log-rank test, with cooperation from Cox regression analysis, were used in survival analysis. Subsequently, we transiently transfected FASN plasmid into A498 and 786O cells, and investigated the role of FASN in ccRCC cell proliferation, apoptosis and migration in vitro. Related hub gens, functional annotations and significant signal pathways were predicted using integrated bioinformatics. Results: FASN mRNA expression was significantly higher in tumor than normal tissues in 913 ccRCC patients from FUSCC and TCGA cohorts. In addition, increased FASN mRNA expression was significantly relevant to advanced T stage ( p <0.001), N stage ( p =0.019), and AJCC stage ( p =0.002). Pearson’s correlation coefficient indicated that FASN amplification positively correlated with VAT% ( r =0.772, p <0.001). VAT% was significantly correlated with poor PFS and OS, with hazard ratios of 2.066 (1.113-7.261, p =0.028) and 2.773 (1.168-8.974, p =0.023) in FUSCC. Meanwhile, ccRCC patients with elevated FASN expression significantly responded for poor PFS and OS, with hazard ratios of 1.529 (1.135-2.061, p =0.005) and 1.450 (1.030-2.041, After FASN-RNAi2 and FASN-overexpression (OE) groups, we investigated FASN mRNA and protein expression level in A498 and 786O cells, and found significant differential expression levels in RNAi1, RNAi2 and OE groups compared with A498 and 786O cells. (C-D) After A498 and 786O cells were transfected for 5 days, the A450 OD values revealed that cells proliferation ability was significantly suppressed in RNAi1 and RNAi2 groups, and significantly promoted in OE groups compared with normal control. (E-F) We performed cells apoptosis detection assay in 786O cells and NC, RNAi1, RNAi2, OE groups. It suggested an increase of apoptosis cell percentage in RNAi1 and RNAi2 groups compared with 786O cells, whereas FASN overexpression showed similar apoptosis cell percentage compared with 786O cells. (G-H) To explore the role of FASN-mediated migration ability of ccRCC cells, we performed Transwell migration assay, and found that the inhibition of FASN markedly restrained migrated cell numbers. Meanwhile, the overexpression of FASN significantly promotes migrated A498 and 786O cell counts. (I) GSEA analysis also suggested that FASN, together with related hub genes, significantly involved in epithelial mesenchymal transition process After with FASN plasmid in FASN-RNAi1, FASN-RNAi2 and FASN-overexpression (OE) groups, we investigated FASN mRNA and protein expression level in A498 and 786O cells, and found significant differential expression levels in RNAi1, RNAi2 and OE groups compared with A498 and 786O cells. (C-D) After A498 and 786O cells were transfected for 5 days, the A450 OD values revealed that cells proliferation ability was significantly suppressed in RNAi1 and RNAi2 groups, and significantly promoted in OE groups compared with normal control. (E-F) We performed cells apoptosis detection assay in 786O cells and NC, RNAi1, RNAi2, OE groups. It suggested an increase of apoptosis cell percentage in RNAi1 and RNAi2 groups compared with 786O cells, whereas FASN overexpression showed similar apoptosis cell percentage compared with 786O cells. (G-H) To explore the role of FASN-mediated migration ability of ccRCC cells, we performed Transwell migration assay, and found that the inhibition of FASN markedly restrained migrated cell numbers. Meanwhile, the overexpression of FASN significantly promotes migrated A498 and 786O cell counts. (I) GSEA analysis also suggested that FASN, together with related hub genes, significantly involved in epithelial mesenchymal transition process

and overexpression of FASN significantly regulates cells proliferation and migration abilities, and inhibition of FASN displayed higher apoptotic rate in ccRCC cells.
Conclusion: In conclusion, this study demonstrated that FASN mRNA expression is positively related with aggressive proliferation, migration progression and predicts prognosis of ccRCC. In addition, we first reveal that elevated FASN mRNA expression is significantly correlated with abdominal obesity distribution, especially VAT%, which also is a significant predictor of poor prognosis in patients with ccRCC.

Background
Renal cell carcinoma (RCC) is a highly malignant tumors originating from the urinary tubular epithelial system of the renal parenchyma, and incidence of RCC is increasing at a rate of 2% a year, especially in developed country [1]. Pathologically, RCC can be divided into four subtypes, and the vast majority of which is clear cell renal cell carcinoma (ccRCC), accounting for approximately 70% of cases [2]. ccRCC portends highly aggression and poor prognosis even in patients who were diagnosed in early stage and treated with nephrectomy [3,4]. Contemporarily, many indicators of poor prognosis contributing to RCC have been confirmed, including obesity and physical inactivity.
Fatty acid synthase (FASN) is one of the important enzymes involved in lipid metabolism.
FASN uses NADPH (reduced nicotinamide adenine dinucleotide phosphate) as a reductant, acetyl-CoA as a primer and malonyl-CoA as a substrate to synthesize long-chain fatty acids [5]. But in well-nourished individuals, most of the fatty acids are provided by exogenous fatty acid from diet, which makes the role of FASN less important [6]. FASN is low expressed in most normal cells and tissues, except for cycling endometrium and lactating breast in adults [7]. Compared with normal cells, all esterified fatty acids in most tumor cells are synthesized from scratch, and the expression of FASN in tumor cells is significantly increased, which is related to tumor aggression and poor prognosis in several cancer types. [5,8,9]. Notably, Hakimi et al. indicated that the upregulated expression of FASN was also significantly associated with poor prognosis [10]. Therefore, FASN may cause the tumorigenesis of ccRCC based on its lipid metabolism pathway.
Several studies indicated that obesity, defined as body mass index (BMI) greater than 30 kg/m², has been identified as a risk factor for ccRCC [11][12][13]. Whether ccRCC patients who received nephrectomy or metastatic ccRCC patients who had been treated with targeted therapy, obese patients survive longer than those with normal weight (BMI 18·5-24·9 kg/m², according to WHO's BMI categories) [14,15]. However, there is a counterintuitive association between BMI and prognosis [16]. Although BMI is an internationally accepted convenient and effective weight indicator, recently, many researchers have controversies on the accuracy of BMI in indicating body fat distribution and metabolic risks [17][18][19]. So, it is essential to present a new measurement to indicate the distribution of adipose and its prognostic value in ccRCC [20,21].
In 2010, Ibrahim MM indicated the anatomical, cellular, molecular difference of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT), and described the heterogenous nature of obesity [22]. Unlike SAT (present in subcutaneous areas), VAT present around abdominal viscera in mesentery and omentum, and is associated with inflammation, type II diabetes and several fat-related diseases [22,23]. Recent study also showed that ccRCC patients with higher Fuhrman grade contribute to significant correlation with elevated VAT [24]. How to more precisely indicate the distribution of adipose and which parts of adipose in body will affect the disease through which metabolic pathway has become the focus in recent study, especially in ccRCC.
To define the prognostic implications of FASN mRNA expression in ccRCC patients and the correlation of FASN mRNA and adipose, we recruited 533 ccRCC samples from TCGA cohort, and 380 patients who have received radical nephrectomy in our institution.

Protein isolation and Western blot analysis
Proteins were extracted from A498 and 786O cells using RIPA lysis buffer (Beyotime Biotechnology Shanghai, China), and concentrated by the bicinchoninic acid protein assay kit (Beyotime Biotechnology, Shanghai, China). Samples were separated by electrophoresis on 6% or 10% SDS gel and then transferred to a methanol activated polyvinylidene fluoride (PVDF) membrane. Membranes were blocked with 5% bovine serum albumin (5% BSA) for 1 h at room temperature and then incubated with primary antibodies, anti-FASN (1:1000, ab22759, Abcam), and anti-beta-Actin primary antibody (1:3000, ab179467, Abcam) at 4 °C overnight. After washed for with TBST for three times, membranes were incubated with secondary antibody Goat Anti-Rabbit IgG conjugated with HRP (1:3000, ab205718, Abcam) at room temperature for 60 min. After three washes with TBST for 10 min each, the bands were visualized using ECL-plus™ western blotting chemiluminescence kits (BD Biosciences, New Jersey, USA).

Cell viability analysis
For viability assays, cells treated with shRNAs or inhibitors were seeded onto 96-well plates (2,000 cells/well). Next, 10 µL CCK8 solution (KeyGEN BioTECH, Nanjing, Chian) was added to each well, and cells were incubated at 37 °C for 2 h. The absorbance of each well at 450 nm was measured at 1, 2, 3, 4, and 5 days after seeding using an automatic microplate reader (TEAN, Swiss). Three replicate analyses were performed for each sample.
Cell apoptosis assays Apoptosis detection assay was performed using Annexin V-FITC Apoptosis Detection Kits (BD, USA) in accordance with the manufacturer's procedures. Briefly, A498 and 786O cells were obtained and triple washed with PBS, and then added 500 ul in 1 × binding buffer. 9 resuspended in each collection tube. After incubation for 15 min, cell apoptosis was analyzed using a FACS analyzer (BD, USA).

Transwell migration assay
After trypsinized and suspended in the medium, 1 × 10 5 A498 and 786O cells were seeded in medium with 10% FBS and placed in each Transwell chamber. The medium containing 20% fetal bovine serum was added in the lower 24-well plate chamber. After 24 hours, the bottom A498 and 786O cells were treated with 4% poly-oxymethylene for 15 minutes, deionized water, and 0.1% crystal violet for 30 minutes. Finally, the A498 and 786O migrating to the lower surface of Transwell chamber were counted using a microscope in six random fields.

Survival analysis
Chi-squared test was utilized to find out the association between different FASN mRNA expression sets and categorical clinicopathological data distribution. Pearson's correlation coefficient was utilized to determine association between VAT% and levels of FASN mRNA expression.
The primary endpoint was overall survival (OS), which was assessed from the date of receive radical nephrectomy to the date of death or the last follow-up. Progression-free survival (PFS) was the secondary endpoint and was defined as the length of time from the date of surgery to the date of progression, second-line treatment or death, whichever accrued first. Survival curves were established using the Kaplan-Meier method and analyzed by log-rank test with 95% confidence intervals (95% CI). To find independent predictors, the hazard ratio (HR) estimates and 95% CI were performed using univariate and multivariate Cox logistic regression model. As a supplement to survival, Cox logistic regression analysis was performed on 117 patients with available MRI scan to assess confounding covariates including A, P, SAT, VAT%, TNM stage, ISUP grade and FASN expression.

Protein-protein interaction (PPI) network construction
Search Tool for the Retrieval of Interacting Genes (STRING; http://strin g-db.org) (version 10.0) online database was utilized to detect PPI network of co-regulated hub genes and analyze the functional interactions between relative proteins. An interaction with a specificity score high than 0.4 was regarded as statistically significant.

Functional annotations
Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.ncifcrf.gov; version 6.8) online database were utilized to investigate the gene ontology (GO): BP (biological process), GO: MF (molecular function) and KEGG pathways analyses, then visualized in bubble chart. To predict potential hallmarks, gene set enrichment analysis (GSEA) was utilized to test significant genes using transcriptional sequences in TCGA database. A permutation test with 1000 times was used to identify the significantly changed pathways. Adj. p (the adjusted p values) less than 0.01 and FDR (false discovery rate) less than 0.25 were confirmed as significant related genes.

Statistical analysis
All statistical analyses and graphical plotting were performed with SPSS version 23.0 software or R software (version 3.3.2). All hypothetical tests were two-sided and p-values less than 0.05 were considered significant in all tests.

Results
In this study, research was conducted in three phases. First, significant differential FASN mRNA expressions and its novel implications in prognosis for 913 ccRCC patients from TCGA and FUSCC were assessed. Second, relationship among visceral adipose tissue, FASN expression, and prognosis for 117 ccRCC patients from FUSCC cohort was estimated.
Third, effective experiments of FASN in human ccRCC cells were performed in vitro.
Significant hub genes panel, functional annotations and signal pathways has been predicted using integrated bioinformatics methods.  Table 1. Chi-square test showed that baseline data were balanced on the distribution of categorical data. Increased FASN mRNA expression in ccRCC patients significantly correlated with advanced T stage (p < 0.001), N stage (p = 0.019), and AJCC stage (p = 0.002) in the FUSCC cohort.  Figure 2B) and OS (p <0.001; Figure 2C). in patients from FUSCC cohort, displayed in forest plots ( Figure 2D-E).
VAT% is a more accurate obesity indicator for ccRCC To verify the VAT% a better predictor using MRI scanning than BMI ( Figure 3A) Figure 3B). Additionally, we assessed expression (p=0.0146; Figure 3C).
Univariate Cox analysis indicated that SAT, VAT%, TNM stage, ISUP grade and FASN expression are independent parameters for prognosis of 117 ccRCC patients (Supplementary Table 2). In multivariate Cox regression analyses of PFS and OS in 117 ccRCC cases whose MRI scans were available from FUSCC cohort, high VAT% was significantly associated with poor PFS (HR=2.066, p=0.028; Figure 3D) and OS (HR=2.773, p=0.023; Figure 3E). However, BMI was not independent covariate affecting survival. In

Functional annotations and predicted signaling pathways in silico
A PPI network of FASN and its co-expression genes was established, including ACACA, ACACB, ACLY, AASDHPPT, ACSL1, ACSL3, CDC5L, MCAT, OLAH and SREBF1 ( Figure 5A). As shown in Figure 5B, bubble chart illustrated the functional enrichment of 11 related genes.
Significant genes involved in fatty acid biosynthetic process, fatty acid metabolic process, carboxylic acid, organic acid and lipid biosynthetic process, markedly participated in carboxylic acid binding, vitamin binding. After normalization of 11 hub genes transcriptional data, clustering analysis and heat map were displayed ( Figure 5C).

Functional annotations using ClueGO showed that changes in FASN biological processes
were closely related to fatty acid biosynthesis, fatty acid biosynthetic process and fatty acid synthase activity ( Figure 5D).

Significant genes and pathways obtained by GSEA
A total of 100 significant genes were obtained from GSEA, the genes with positive and negative correlation were plotted. The results illustrated that the most significant pathways including E2F targets, down-regulation of Kras signaling, estrogen response and G2M checkpoint ( Figure 6A-F). Besides, the heat map shows the transcriptional expression profiles of 100 most significant up-or down-regulated genes ( Figure 6G).

Discussion
In this study, we investigated whether FASN mRNA expression has a potentially implications, and its association with abdominal adipose distribution on ccRCC patients. To

Conclusion
In conclusion, this study demonstrated that FASN mRNA expression is positively related with aggressive proliferation, migration and predicts prognosis of ccRCC. In addition, we first reveal that elevated FASN mRNA expression is significantly correlated with abdominal obesity distribution, especially VAT%, which also is a significant predictor of poor prognosis in patients with ccRCC.

Consent for publication
Not applicable.

Availability of supporting data
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request or online database.

Competing interests
The authors declare no potential conflicts of interest.

Supplementary Files
This is a list of supplementary files associated with the primary manuscript. Click to download. supplementary.docx