The Fto Gene Regulates the Proliferation and Differentiation of Pre-Adipocytes in Vitro

The highly regulated differentiation and proliferation of pre-adipocytes play a key role in the initiation of obesity. Fat mass and obesity associated (FTO) is a novel gene strongly associated with the risk of obesity. A deficiency of FTO may cause growth retardation in addition to fat mass and adipocyte size reduction in vivo. To investigate the potential role of Fto gene on the proliferation and differentiation of pre-adipocytes, we generated Fto-knockdown and overexpressed 3T3-L1 cells. Using numerous proliferation assays our results suggest that Fto knockdown leads to suppression of proliferation, lower mitochondrial membrane potential, less cellular ATP, and decreased and smaller intracellular lipid droplets compared with controls (p < 0.05). Western blot analysis demonstrated that Fto knockdown can significantly suppress peroxisome proliferator-activated receptor gamma (PPARγ) and glucose transporter type 4 (GLUT4) expression and inhibit Akt phosphorylation. By contrast, overexpression of Fto had the opposing effect on proliferation, mitochondrial membrane potential, ATP generation, in vitro differentiation, Akt phosphorylation, and PPARγ and GLUT4 expression. Moreover, we demonstrated that Wortmannin, a phosphoinositide 3-kinase (PI3K) inhibitor, could inhibit phospho-Akt in Fto overexpressed 3T3-L1 cells. Taken together, the results suggest that Fto regulates the proliferation and differentiation of 3T3-L1 cells via multiple mechanisms, including PPARγ and PI3K/Akt signaling.


Introduction
Obesity increasingly is a pressing public health concern with a rapidly surging prevalence in both developed and developing countries [1][2][3]. According to the World Health Organization, in 2014 more than 1.9 billion adults were overweight, 600 million of whom were obese [4]. In addition to psychosocial ramifications of obesity, this condition causes a multitude of health complications [1][2][3]. It has been widely reported that the majority of obese individuals have multiple comorbidities including inflammation, hypertension, coronary heart disease and stroke, type 2 diabetes mellitus, insulin resistance, gall bladder disease, nonalcoholic fatty liver disease, and cancers, among others. Together, these result in the poor health of those affected and may result in accelerated aging and death [3,5].
Unfortunately, the existing interventions for obesity control, including diet modification and exercise-based managements, seem largely ineffective [5], in part related to our inadequate and incomplete understanding of mechanisms of pathogenesis. It has been well accepted that adipose tissues play an integral role in lipid and glucose metabolism; therefore precise regulation of differentiation and proliferation of pre-adipocytes are thought to play a key role in the initiation of obesity [6][7][8][9][10][11]. Recently, the molecular mechanisms and signaling pathways regulating pre-adipocytes differentiation and proliferation have been extensively studied; however, our understanding of these processes remains incomplete [10,12].
The fat mass and obesity associated gene (Fto) is an important genes that is strongly associated with the risk for obesity [13,14]. In vivo experiments have demonstrated that Fto-deficiency causes growth retardation in addition to fat mass and adipocyte size reduction [15]. Furthermore, overexpression or dysregulation of Fto alleles is associated with increased risk of obesity while inhibition of Fto may have a protective-effect against obesity in animal models [15][16][17]. Recently, several studies have highlighted the importance of Fto gene during adipogenesis [18][19][20]. Zhao et al. [18] suggested Fto can modulate SRSF2-RNA binding and demethylation of N6-methyladenosine, which together impact RNA splicing and critically regulate adipogenesis. Tews et al. [19] hypothesized that Fto deficiency may lead to the induction of white adipose tissue (WAT) browning and in turn cause mitochondrial uncoupling and increase energy expenditure. Taken together, this suggests an FTO-associated susceptibility to obesity. Merkestein et al. [20] reported that FTO promotes adipogenesis via mitotic clonal expansion. However, the specific mechanisms and signaling pathways by which FTO impacts adipogenesis and obesity remains incompletely characterized.
In the present study, the pre-adipose cell line 3T3-L1 was utilized as an in vitro model, and the biological roles and potential mechanisms of Fto on the proliferation and differentiation of pre-adipocytes were evaluated via classical gain/loss-of-function experiments.
Used for this study are 3T3-L1 cells within 15 passages.
Using the genomic sequence of the mouse Fto gene (GeneBank ID: 26383), an Fto recombinant plasmid construction was constructed as previously described [21]. Three Fto siRNAs and a negative control siRNA (NC siRNA) were designed and synthesized (Invitrogen, Carlsbad, CA, USA). The Fto inhibition efficiency by siRNA was evaluated by qRT-PCR and Western blot assay. The siRNA #2 elicited the most effective suppression of Fto and was selected for subsequent studies. The sequences for Fto-siRNA #2 is as follows (5' to 3'): The siRNA and plasmid transfections were performed using Lipofectamine®2000 (Thermofisher scientific, Waltham, MA, USA) according to the manufacturers' instructions. Briefly, 3T3-L1 cells (2 ¢ 10 5 ) were seeded onto 6-well plates and incubated overnight. Then 30nM siRNAs and NC control, as well as 0.5 µg plasmid DNA were applied to each well together with recommended volume of transfection reagent. Cells were harvested at indicated time points following transfection.

Western Blot Assay
As previously described [22], cells were collected, washed in ice-cold PBS and lysed in 50 µL of lysis buffer with protease inhibitor cocktail (Roche Life Science, Indianapolis, IN, USA). The protein concentration was determined using the BSA assay, and 50 µg total protein for each sample was separated by SDS-polyacrylamide gel electrophoresis. The protein was transferred onto polyvinylidene fluoride (PVDF, Millipore, Billerica, MA, USA) microporous membranes, incubated with primary antibody overnight at 4 ¥ C, and subsequently incubated with the corresponding secondary antibodies.

Cell Viability Assay
As previously described, MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, a tetrazole) was used to evaluate cell viability [22]. In brief, cells were seeded at a density of 5 ¢ 10 4 cells/well in 24-well plates. Following various treatments, 5 mg/mL MTT (Sigma-Aldrich, MO, USA) was added at indicated time points. After 4 h of incubation, the supernatant was discarded, and DMSO was added to dissolve the formazan generated. Absorbance at 490 nm was measured using multimode microplate reader (BioTek, VT, USA). Cell viability was expressed as the percentage of viable cells with different treatments relative to cell viability detected in the control cells.

Flow Cytometry Assay
Based on flow cytometry assay, JC-1 (Beyotime Biotechnology, Haimen, China) staining was used to measure the mitochondrial membrane potential (∆ψm) of 3T3-L1 cells according to the manufacturer's directions. Briefly, after indicated treatments, cells were collected from 6-well plates and incubated with 1 mL JC-1 working solution for 20 min at 37 ¥ C. Cells treated with 10 µmol/L carbonyl cyanide m-chlorophenylhydrazone (CCCP) were used as negative control. The cells were then rinsed twice with JC-1 staining buffer, and the fluorescence intensity of both mitochondrial JC-1 monomers (λ ex 514 nm, λ ex 529 nm) and aggregates (λ ex 585 nm, λ em 590 nm) were detected using

Live Cell Fluorescent Imaging of Mitochondria
MitoTracker Red CM-H 2 X Ros (Thermofisher Scientific, Waltham, MA, USA) was applied to stain mitochondria of living cells. Cells were cultured in 35 mm glass bottom culture dishes (Nest Scientific, Rahway, NJ, USA) and incubated with 200 ng/mL MitoTracker Red CM-H 2 X Ros at 37 ¥ C for 30 min.

ATP Assay/Determination of ATP Content of Cells
According to ATP Assay Kit (Beyotime, BioTECH, Haimen, China), 3T3-L1 cells were lysed using commercial lysis buffer for 5 min on ice. Following centrifugation at 12,000 ¢ g for 5 min, reconstituted luciferase substrate was added to the supernatant and incubated for 5 min in the dark. Per 6-well plate, 50 µL of each sample was mixed with 100 µL ATP detection working dilution. The luminance (RLU) was measured using a multimode microplate reader. Standard curves were generated, and the protein concentration of each sample was determined using the Bradford Protein assay (Thermofisher Scientific, Waltham, MA, USA). Total ATP levels were expressed as nmol/mg protein.

Statistical Analysis
All data were presented as mean¨standard deviation, and all experiments were performed in triplicate. The statistical significance of differences was evaluated with the student's t-test or one-way analysis of variance (ANOVA) using SPSS Statistics (Version 19.0, IBM, Armonk, NY, USA). p < 0.05 was considered significant.

Fto Gene Affects the Proliferation of 3T3-L1 Cells
The expression level of mouse Fto was confirmed by real time PCR and Western blot assay. Our results demonstrate that Fto siRNA transfection effectively suppressed Fto at both mRNA and protein level ( Figure 1A The effect of Fto expression on the proliferation of pre-adipose cells was evaluated by MTT cell viability assay and using an EdU proliferation assay. As shown in Figure 2, when Fto expression was suppressed, the viability of 3T3-L1 cells was significantly reduced as compared to NC siRNA transfected cells (Figure 2A) (p < 0.05). By contrast, Fto recombinant plasmid increased the cell viability relative to the control transfected group (p < 0.05) ( Figure 2B).
Moreover, cell EdU proliferation assay was performed 24 h after adipogenic induction. Our results indicate that pre-adipocytes with Fto-inhibition were less proliferative 24 h following adipogenesis induction as compared to NC control. However, Fto overexpression, induced by recombinant plasmid transient transfection, increased the proliferation of 3T3-L1 cells, as early as 24 h after adipogenesis induction ( Figure 2C,D). Our data were consistent with the study by Merkestein et al. [20], which demonstrated that FTO acts early in adipogenesis to enhance adipocyte number. Taken together, these results demonstrate that Fto may regulate the proliferation of pre-adipocytes as well as the cell proliferation during early adipogenesis in vitro. The effect of Fto expression on the proliferation of pre-adipose cells was evaluated by MTT cell viability assay and using an EdU proliferation assay. As shown in Figure 2, when Fto expression was suppressed, the viability of 3T3-L1 cells was significantly reduced as compared to NC siRNA transfected cells (Figure 2A) (p < 0.05). By contrast, Fto recombinant plasmid increased the cell viability relative to the control transfected group (p < 0.05) ( Figure 2B).
Moreover, cell EdU proliferation assay was performed 24 h after adipogenic induction. Our results indicate that pre-adipocytes with Fto-inhibition were less proliferative 24 h following adipogenesis induction as compared to NC control. However, Fto overexpression, induced by recombinant plasmid transient transfection, increased the proliferation of 3T3-L1 cells, as early as 24 h after adipogenesis induction ( Figure 2C,D). Our data were consistent with the study by Merkestein et al. [20], which demonstrated that FTO acts early in adipogenesis to enhance adipocyte number. Taken together, these results demonstrate that Fto may regulate the proliferation of pre-adipocytes as well as the cell proliferation during early adipogenesis in vitro.

Fto Gene Influences the Mitochondrial Membrane Potential and ATP Levels of Pre-adipocytes
The cellular metabolism, mitochondrial membrane potential (MMP) and cellular ATP levels of 3T3-L1 cells were determined as previously described [23]. As shown in Figure 3, 24 h following transfection, 3T3-L1 cells with different Fto gene expression levels exhibited different MMP levels.

Fto Gene Influences the Mitochondrial Membrane Potential and ATP Levels of Pre-adipocytes
The cellular metabolism, mitochondrial membrane potential (MMP) and cellular ATP levels of 3T3-L1 cells were determined as previously described [23]. As shown in Figure 3, 24 h following transfection, 3T3-L1 cells with different Fto gene expression levels exhibited different MMP levels. Specifically, 3T3-L1 cells with suppressed Fto expression demonstrated a~30% lower MMP level than NC siRNA transfected cells ( Figure 3A) while MMP levels in Fto-overexpressed cells were 40% higher than in vector transfected control cells ( Figure 3B). These results indicate that Fto is involved in the regulation on MMPs of adipocyte precursors.
investigate the effects of modulating Fto expression. The fluorescence dye MitoTracker Red CM-H2XRos is cell-permeable and sequesters in the mitochondria. This dye does not emit fluorescence unless it is oxidized; thus the observed fluorescence implies the presence of actively respiring mitochondria [24]. In the present study, decreased fluorescence was observed in Fto-deficient 3T3-L1 cells when compared with NC controls ( Figure 3E) while enhanced MitoTracker Red fluorescence was observed in the context of Fto-overexpression. Specifically, Fto-overexpression resulted in a twofold increase in the fluorescence intensity as compared to vector-transfected 3T3-L1 cells ( Figure 3E,F).
Cellular ATP content was also evaluated. As shown in Figure 3C,D, the cellular ATP level decreased in the Fto knockdown 3T3-L1 cells while Fto-overexpression enhanced luminance by up to two fold compared with mock transfection groups. The results were in agreement with MMP and Mitotracker staining results and together suggested that Fto might play a role in the regulation of mitochondrial activity of pre-adipocytes.  Live cell imaging of mitochondria using Mitotracker Red CM-H 2 X Ros staining was performed to investigate the effects of modulating Fto expression. The fluorescence dye MitoTracker Red CM-H 2 X Ros is cell-permeable and sequesters in the mitochondria. This dye does not emit fluorescence unless it is oxidized; thus the observed fluorescence implies the presence of actively respiring mitochondria [24]. In the present study, decreased fluorescence was observed in Fto-deficient 3T3-L1 cells when compared with NC controls ( Figure 3E) while enhanced MitoTracker Red fluorescence was observed in the context of Fto-overexpression. Specifically, Fto-overexpression resulted in a two-fold increase in the fluorescence intensity as compared to vector-transfected 3T3-L1 cells ( Figure 3E,F).
Cellular ATP content was also evaluated. As shown in Figure 3C,D, the cellular ATP level decreased in the Fto knockdown 3T3-L1 cells while Fto-overexpression enhanced luminance by up to two fold compared with mock transfection groups. The results were in agreement with MMP and Mitotracker staining results and together suggested that Fto might play a role in the regulation of mitochondrial activity of pre-adipocytes.

Fto Gene Regulated the Differentiation of Pre-adipocyte
Derived from mouse 3T3 cells, 3T3-L1 has the ability to differentiate into adipocytes. The effect of Fto on 3T3-L1 cells differentiation was detected using intracellular lipid droplets staining assay. As shown in Figure 4, following induction of differentiation, the lipid droplets in 3T3-L1 cells with Fto-suppression were irregular and smaller, and the amount of differentiated cells was significantly less than the amount observed in NC control group (p < 0.05). By contrast, Fto-overexpression led to more cellular lipid droplets and more differentiated cells as compared to the mock transfection group under typical culture conditions (p < 0.05).

Fto Gene Regulated the Differentiation of Pre-adipocyte
Derived from mouse 3T3 cells, 3T3-L1 has the ability to differentiate into adipocytes. The effect of Fto on 3T3-L1 cells differentiation was detected using intracellular lipid droplets staining assay. As shown in Figure 4, following induction of differentiation, the lipid droplets in 3T3-L1 cells with Ftosuppression were irregular and smaller, and the amount of differentiated cells was significantly less than the amount observed in NC control group (p < 0.05). By contrast, Fto-overexpression led to more cellular lipid droplets and more differentiated cells as compared to the mock transfection group under typical culture conditions (p < 0.05).

Fto Regulated the Differentiation and Proliferation of 3T3-L1 Cells via Multiple Mechanisms
To examine the contributions of Fto in regulating differentiation and proliferation, the expression of several lipid metabolism-related transcription factors and signaling pathways were detected in Fto knockdown or overexpressed 3T3-L1 cells. The results showed that knockdown of Fto decreased the expression of peroxisome proliferator-activated receptor γ (PPARγ) in addition to significantly suppressing Akt phosphorylation and the expression of Glucose transporter type 4 (GLUT4) (Figure 5A,C). No obvious effects on the expression of AMP-activated protein kinase (AMPK), adipose triglyceride lipase (ATGL), hormone-sensitive lipase (HSL) and cAMP response element-binding protein CREB were observed ( Figure 5A). Fto overexpression has the opposing effect as Fto-knockdown ( Figure 5B). We noticed increased HSL in 3T3-L1 cells with elevated Fto expression. Wortmannin (100 nmol/L, Cayman Chemical, Ann Arbor, MA, USA), which is a specific inhibitor of the phosphatidylinositol 3-kinase (PI3k)/Akt signaling cascade, was utilized to investigate the effect of Fto on the proliferation and differentiation. As shown in Figure 5D, phospho-Akt was markedly

Fto Regulated the Differentiation and Proliferation of 3T3-L1 Cells via Multiple Mechanisms
To examine the contributions of Fto in regulating differentiation and proliferation, the expression of several lipid metabolism-related transcription factors and signaling pathways were detected in Fto knockdown or overexpressed 3T3-L1 cells. The results showed that knockdown of Fto decreased the expression of peroxisome proliferator-activated receptor γ (PPARγ) in addition to significantly suppressing Akt phosphorylation and the expression of Glucose transporter type 4 (GLUT4) (Figure 5A,C). No obvious effects on the expression of AMP-activated protein kinase (AMPK), adipose triglyceride lipase (ATGL), hormone-sensitive lipase (HSL) and cAMP response element-binding protein CREB were observed ( Figure 5A). Fto overexpression has the opposing effect as Fto-knockdown ( Figure 5B). We noticed increased HSL in 3T3-L1 cells with elevated Fto expression. Wortmannin (100 nmol/L, Cayman Chemical, Ann Arbor, MA, USA), which is a specific inhibitor of the phosphatidylinositol 3-kinase (PI3k)/Akt signaling cascade, was utilized to investigate the effect of Fto on the proliferation and differentiation. As shown in Figure 5D, phospho-Akt was markedly reduced in Fto-overexpressing 3T3-L1 cells following application of Wortmannin, suggesting that Fto mediates the phosphorylation of Akt at serine 473. These results indicate that Fto might be a multi-functional regulator in pre-adipocytes proliferation and differentiation.

Discussion
White adipose tissues are distributed throughout the whole body, and the excessive accumulation of WAT is thought to account for obesity [2,8,9]. Pre-adipocytes are the precursors of

Discussion
White adipose tissues are distributed throughout the whole body, and the excessive accumulation of WAT is thought to account for obesity [2,8,9]. Pre-adipocytes are the precursors of the adipocytes, which consist of WAT, and regulation of the balance between proliferation and differentiation of these cells will determine the quantity and size of mature adipocytes [12]. As such, dysregulation of proliferation and differentiation in pre-adipocytes contributes to the onset of obesity.
Fisher et al. [15] previously demonstrated that mice lacking the Fto gene suffered from severe growth retardation. In the present study, we observed a significant inhibition of proliferation in Fto knockdown 3T3-L1 cells (Figure 2A,C), which is consistent with the in vivo results of Fisher et al. A newly published study by Jeffery et al. [25] indicated that increased proliferation of adipocyte precursors in visceral WAT during the early stage of high fat diet caused adipogenesis. Using mouse embryonic fibroblasts (MEF) derived from Fto-deleted or Fto-overexpressed mice, Merkestein et al. [20] found that FTO could enhance MEF proliferation as early as 48 h after adipogenic induction. In our study, by applying Fto siRNA or recombinant plasmid transfection, we demonstrated that Fto could regulate the proliferation of 3T3-L1 cell line in the precursor status, and even earlier during adipogenesis (24 h after adipogenic induction) (Figure 2A-C). These results confirmed the regulatory role of Fto in adipogenesis in vitro.
As a well-characterized cell model, fibroblast-like 3T3-L1 cells will differentiate into a signet ring appearance of adipocyte-like phenotype under appropriate stimulation, which represents the synthesis and accumulation of triglycerides [26]. Several studies have reported that Fto-overexpression can significantly increase the triglyceride content of the cell [17,27,28]. Merkestein et al. [20] recently reported that primary adipocytes and MEFs derived from Fto overexpression mice demonstrated increased potential for adipogenic differentiation while Fto knockout mice derivative MEFs exhibited reduced adipogenesis. In our study, following differentiation stimulation, we found Fto-overexpressing cells can differentiate into more mature adipocytes with more lipid droplets than vacant plasmid transfected cells ( Figure 4B). By contrast, less, irregular and smaller lipid droplets were observed after differentiation induction in Fto siRNA transfected pre-adipocytes ( Figure 4A). Our in vitro data were in agreement with the finding of Merkestein et al. [20] that FTO affects fat mass via regulating adipocyte differentiation in vivo.
Several studies have suggested that Fto expression level was strongly correlated with energy status. For example, overexpression of Fto led to fatty acid accumulation and increased food intake and thus resulted in obesity [17]. Tews et al. [19] recently reported that FTO is involved in WAT browning in human SGBS pre-adipocytes. Although they showed no difference in the rate of differentiation and triglyceride content between control and FTO-deficient SGBS cells, they determined that FTO deficiency could lead to mitochondrial uncoupling and increase energy expenditure [19]. In the present study, we found that Fto could regulate the MMP of mouse pre-adipocyte 3T3-L1 and affect ATP generation in 3T3-L1 cells. Our MTT assay results also indicated effects of Fto on the mitochondrial activity of live 3T3-L1 cells, since the conversion of MTT into formazan requires the functioning mitochondria. Based on these results, we hypothesize that FTO might affect the mitochondrial function in vitro. However, considering that Tews et al. did not show significantly varied citrate synthase activity, mitochondrial contents and structure in FTO deficient SGBS cells [20], we should carefully analyze the effects of Fto on mitochondrial function in our model system in future studies to fully characterize the implications of our findings. Nevertheless, our results clearly indicated that Fto could represent a functional regulator on the 3T3-L1 pre-adipocyte in the context of obesity.
There are numerous opinions regarding the formation of adipose tissue in the development and expansion of adipose tissue in obesity [25]. Jeffery et al. [25] recently suggested that the formation of adipocytes in obesity and development is controlled by distinct molecular mechanisms. Fat accumulation and morphology changes of adipocytes were thought to represent a consequence of specific genes expression, including several transcription factors, positive and negative signaling pathways effectors, and even non-coding RNAs, which were induced during the process of pre-adipocytes proliferation and differentiation [11,12]. In this study, we wanted to identify molecular mechanisms and signaling pathways that regulated pre-adipocyte proliferation and differentiation via Fto.
Our data demonstrate that Fto overexpression induced the expression of transcription activator PPARγ, while Fto deficiency impaired PPARγ expression in 3T3-L1 pre-adipocyte. PPARγ belongs to the ligand-activated nuclear receptor superfamily, which is preferentially expressed in adipocytes [29], and is vital in expression regulation of the gene networks involved in adipogenesis, lipid metabolism, inflammation and maintenance of metabolic homeostasis [29,30]. In 2013, Bravard et al. [31] hypothesized that FTO might participate in the increased expression of PPARγ2 during adipogenesis via demethylating its promoter region, but the subsequent study failed to verify their hypothesis. Further studies are required to clarify the mechanistic actions of FTO on PPARγ expression regulation during adipogenesis.
We found Fto may also be involved in Akt activity regulation in 3T3-L1 cells. The insulin-mediated activation of PI3K/Akt signaling pathway plays a crucial role in energy metabolism [32][33][34]. Activated Akt can induce glycogen synthesis by inhibiting GSK-3, promote protein synthesis via mTOR cascades, and facilitate cell survival through inhibiting pro-apoptotic factors [32,34]. The insulin receptor signaling is also important for fatty acid and cholesterol synthesis [35]. The activation of adipocyte precursors is thought to dependent on the PI3K-AKT2 pathway in multiple models of obesity, but Jeffery et al. [25] demonstrated that WAT-development does not require AKT2. However, using a specific PI3k inhibitor, we found that Akt signaling is involved in Fto regulated adipogenesis in vitro ( Figure 5). Moreover, in the present study, the regulation of Fto on Akt activity could in part interpret the variations of proliferation, energy homeostasis, and adipogenesis in 3T3-L1 cells with different Fto levels. Besides, Fto expression levels similarly affected GLUT4 expression, which also participates in insulin receptor signaling [36]. We were also able to conclude that FTO had a minor effect on HSL expression (following Fto-overexpression) and no effect on the expression of AMPK and ATGL.
In summary, our study demonstrates that Fto can affect the proliferation, energy homeostasis, and differentiation of 3T3-L1 pre-adipocyte. Fto might function as a regulator of PI3K Akt signaling and PPARγ expression. While our results have improved our understanding of the importance of Fto on obesity, several questions remain to be answered. Specifically, further studies are required to determine if the Fto-mediated effect on PPARγ is mediated through demethylation activity or via other mechanism. Similarly, the exact mechanisms through which Fto affects Akt phosphorylation requires attention. More detailed investigations using more sophisticated model system (primary adipocytes, embryonic fibroblasts, etc.) are extremely necessary to uncover the molecular mechanisms underlying Fto's role in adipogenesis and obesity.