E2F1 Regulates Adipocyte Differentiation and Adipogenesis by Activating ICAT.

Wnt/β-catenin is a crucial repressor of adipogenesis. We have shown that E2 promoter binding factor 1 (E2F1) suppresses Wnt/β-catenin activity through transactivation of β-catenin interacting protein 1 (CTNNBIP1), also known as inhibitor of β-catenin and TCF4 (ICAT) in human colorectal cancers. However, it remains unknown whether ICAT is required for E2F1 to promote differentiation by inhibiting β-catenin activity in pre-adipocytes. In the present study, we found that 1-methyl-3-isobutylxanthine, dexamethasone, and insulin (MDI)-induced differentiation and lipid accumulation in 3T3-L1 pre-adipocytes was reversed by activation of β-catenin triggered by CHIR99021, a GSK3β inhibitor. Intriguingly, we observed a reduced protein level of E2F1 and ICAT at a later stage of pre-adipocytes differentiation. Importantly, overexpression of ICAT in 3T3-L1 pre-adipocytes markedly promote the adipogenesis and partially reversed the inhibitory effect of CHIR99021 on MDI-induced adipogenesis and lipid accumulation by regulating adipogenic regulators and Wnt/β-catenin targets. Moreover, pre-adipocytes differentiation induced by MDI were markedly inhibited in siE2F1 or siICAT transfected 3T3-L1 cells. Gene silencing of ICAT in the E2F1 overexpressed adipocytes also inhibited the adipogenesis. These data indicated that E2F1 is a metabolic regulator with an ability to promote pre-adipocyte differentiation by activating ICAT, therefore represses Wnt/β-catenin activity in 3T3-L1 cells. We also demonstrated that ICAT overexpression did not affect oleic acid-induced lipid accumulation at the surface of Hela and HepG2 cells. In conclusion, we show that E2F1 is a critical regulator with an ability to promote differentiation and adipogenesis by activating ICAT in pre-adipocytes.


Introduction
The alarming increase in the incidence of obesity among adults and children worldwide has prompted extensive researches on molecular mechanisms responsible for the synthesis and catabolism of triglycerides (TG) in white adipose tissues [1]. A compelling evidence shows that an excessive accumulation of fat mass in obese subjects is associated with an increase in adipocyte volume (hypertrophy), number (hyperplasia), or a combination of both (hypertrophy-hyperplasia) [2]. It is

Western Blotting Analysis
Cells were harvested for the analysis of protein abundance by Western blot as previously described [26]. Cells were harvested and lysed on ice for 30 min in RIPA lysis buffer containing 50 mM Tris-HCl (pH 7.4), 150 mmol/L NaCl, 1% NP-40, 0.1% SDS, 1.0 mmol/L Phenylmethanesulfonyl fluoride (PMSF), 1.0 mmol/L Na 3 VO 4 , 1.0 mmol/L NaF, and protease inhibitor tablet (Roche, Indian apolis, IN, USA), followed by sonication for three times with 10 s/ time. The whole-cell lysates were centrifuged at 12,000 rpm for 10 min to collect the supernatant. The protein concentration of the supernatant was determined using the Pierce BCA protein Assay Kit (Huaxing Biotech, Beijing, China) with bovine serum albumin as standard. Equal amounts of proteins were separated using SDS-page gels and transferred to polyvinylidene difluoride (PVDF) membranes (Millipore, MA, USA). The membranes were blocked in 5% fat-free milk in Tris-buffered saline with Tween 20 (TBST) for 1 h at room temperature, and then were incubated with indicated primary antibodies overnight at 4 • C. After incubation with horseradish peroxidase (HRP)-conjugated secondary antibody for 1 h, the chemiluminescent signal was detected using Super-Enhanced Chemiluminescence Kit (Huaxing Biotech, Beijing, China).

Lipid Droplets' Staining
Lipid droplets in cells were stained by oil red O or BODIPY493/503. Cells were washed with PBS and fixed with 4% paraformaldehyde for 1 h on ice, followed by washing with 60% isopropanol. Then stained with oil red O working solutions containing 6 mL stock solution (5 g/L in isopropanol) and 4 mL double-distilled H 2 O or BODIPY493/503 (stock concentration 1 mg/mL and working solution 1:1000 dilution) for 15 min at room temperature followed by washing for three times with PBS and viewed with a microscope. BODIPY493/503-stained lipid droplets were viewed through a fluorescence microscope. To quantify intracellular lipids, the oil red O-stained lipid droplets were dissolved with 100% isopropanol for 10 min. The absorbance of extracted dye was then measured at 520 nm.

Measurement of TG Content
TG contents in adipocytes were measured using an assay kit as described previously [24]. Briefly, cells were washed and lysed in provided lysis buffer, and then, the TG assay reagents were added, according to the manufacturer's instructions. The optical density of the solution was measured at 510 nm using a spectrophotometer plate reader. TG contents were calculated from a standard curve for each assay, and data are normalized to total cellular protein contents.

siRNA-Mediated Knockdown
For E2F1 or ICAT knockdown in 3T3-L1 pre-adipocytes, 100 pmol of each siRNA oligonucleotides (Genepharma, Shanghai, China) were transfected into cells plated in 6-well dishes using Lipofectamine Cells 2020, 9, 1024 5 of 15 RNAiMAX Reagent as described by instructions. Cells transfected with non-targeting siRNA (NC) were used as control. The pre-adipocytes stably overexpressed E2F1 were transfected with 100 pmol siICAT or siNC. Cells were harvested to detect the protein expression or processed for designated assays 48 h post-transfection.

Statistical Analysis
Data were expressed as means ± SEM and were statistically analyzed by GraphPad PRISM5 (GraphPad Software, CA, USA). Both normality and homogeneity tests were examined before the statistical analysis in our study. Differences between two groups were assessed using the unpaired two-tailed Student t test. Data sets that involved more than two groups were assessed using ANOVA, followed by Tukey post-hoc test. In the figures, data with different superscript letters are significantly different at p < 0.05. A value according to the post hoc ANOVA statistical analyses. The results were considered statistically significant when p < 0.05.

MDI-Induced Differentiation in 3T3-L1 Cells Was Associated with Increased Protein Levels of E2F1 and ICAT at Day 3 of Differentiation
In consistency with the previous study [19], 3T3-L1 pre-adipocytes were successfully differentiated into adipocytes by MDI medium with the appearance of marked multiple vesicles and lipid accumulation as shown by oil red O and BODIPY493/503 staining ( Figure 1A, upper lane). The representative micrographs of cells during differentiation showed that accumulation of the lipid droplets was observed at day 3 ( Figure 1A, lower lane) and differentiated into mature adipocytes with 7-day MDI induction. The time course study showed that transcriptional ( Figure S1A) and protein levels of PPARγ and C/EBPα ( Figure 1B), two critical adipogenic regulators, were significantly enhanced (p < 0.05). Both the mRNA level ( Figure S1B) and protein abundance of β-catenin, as well as these of c-MYC and CCND1 ( Figure 1C), two classic downstream targets of Wnt/β-catenin signaling, were dramatically downregulated (p < 0.05) in differentiated cells, as compared with un-differentiated cells. In agreement with the phenotype changes, mRNA level of fatty acid binding protein (AP2), a well-known adipocyte marker, was upregulated (p < 0.05) ( Figure S1A, lower panel). Of interest, protein levels of E2F1 and ICAT were significantly increased (p < 0.05) at day 3 of differentiation and were reduced to an undetectable level at the later stages of adipocyte differentiation ( Figure 1C). These results showed that MDI-induced differentiation in 3T3-L1 cells was associated with an increased protein level of E2F1/ICAT at day 3 of differentiation.

Activation of Wnt/β-catenin Signaling by GSK3β Inhibitor Blocked Adipogenesis
To further explore a functional role of Wnt/β-catenin signaling on differentiation, 3T3-L1 cells were incubated with MDI to induce differentiation in the presence of CHIR99021 (0, 0.5, 1.0, 2.0, 3.0, or 4.0 µM), a GSK3β inhibitor, which has been reported to activate the canonical Wnt/β-catenin pathway in 3T3-L1 pre-adipocytes [22]. Adipogenesis was assessed at day 7 and we found that CHIR99021 blocked 3T3-L1 differentiation in a dose-dependent manner, as assessed by oil red O and BODIPY493/503 staining (Figure 2A,B). Quantification of lipid accumulation ( Figure 2C) and intracellular TG ( Figure 2D) indicated that differentiation of pre-adipocytes was significantly inhibited by the presence of 1.0 to 4.0 µM CHIR99021 in the media. CHIR99021-activated Wnt/β-catenin signaling was validated by an increased protein level of β-catenin in the nucleus, as well as upregulated proteins abundance of CCND1 and c-MYC (p < 0.05) ( Figure 2E,F) in 3T3-L1 adipocytes. In agreement with phenotypes observed and activation of Wnt/β-catenin signaling, 3T3-L1 adipocytes incubated with 3.0 and 4.0 µM CHIR99021 led to significantly reduced protein abundance of PPARγ and C/EBPα (p < 0.05) ( Figure 2E,F). In our study, 3.0 µM of CHIR99021 was chosen for the further research, considering a markedly repressing effect on pre-adipocyte differentiation.
binding protein (AP2), a well-known adipocyte marker, was upregulated (p < 0.05) ( Figure S1A, lower panel). Of interest, protein levels of E2F1 and ICAT were significantly increased (p < 0.05) at day 3 of differentiation and were reduced to an undetectable level at the later stages of adipocyte differentiation ( Figure 1C). These results showed that MDI-induced differentiation in 3T3-L1 cells was associated with an increased protein level of E2F1/ICAT at day 3 of differentiation. Values are means ± SEMs, n = 3 independent experiments. Means without a common letter differ, p < 0.05. C/EBPα, CCAAT-enhancer binding protein α; E2F1, E2 promoter binding factor 1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ICAT, Inhibitor of β-catenin and TCF4; PPARγ, peroxisome proliferator activated receptor γ.

Activation of Wnt/β-catenin Signaling by GSK3β Inhibitor Blocked Adipogenesis
To further explore a functional role of Wnt/β-catenin signaling on differentiation, 3T3-L1 cells were incubated with MDI to induce differentiation in the presence of CHIR99021 (0, 0.5, 1.0, 2.0, 3.0, or 4.0 μM), a GSK3β inhibitor, which has been reported to activate the canonical Wnt/β-catenin pathway in 3T3-L1 pre-adipocytes [22]. Adipogenesis was assessed at day 7 and we found that CHIR99021 blocked 3T3-L1 differentiation in a dose-dependent manner, as assessed by oil red O and BODIPY493/503 staining (Figure 2A and 2B). Quantification of lipid accumulation ( Figure 2C) and intracellular TG ( Figure 2D) indicated that differentiation of pre-adipocytes was significantly inhibited by the presence of 1.0 to 4.0 μM CHIR99021 in the media. CHIR99021-activated Wnt/βcatenin signaling was validated by an increased protein level of β-catenin in the nucleus, as well as upregulated proteins abundance of CCND1 and c-MYC (p < 0.05) ( Figure 2E and 2F) in 3T3-L1 adipocytes. In agreement with phenotypes observed and activation of Wnt/β-catenin signaling, 3T3-L1 adipocytes incubated with 3.0 and 4.0 μM CHIR99021 led to significantly reduced protein abundance of PPARγ and C/EBPα (p < 0.05) ( Figure 2E and 2F). In our study, 3.0 μM of CHIR99021 was chosen for the further research, considering a markedly repressing effect on pre-adipocyte differentiation.  Values are means ± SEMs, n = 3 independent experiments. Means without a common letter differ, p < 0.05. C/EBPα, CCAAT-enhancer binding protein α; GSKi, CHIR99021; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PPARγ, peroxisome proliferator activated receptor γ.

Overexpression of ICAT Reversed the Effect of GSK3β Inhibitor on Cell Differentiation and Adipogenesis in 3T3-L1 Pre-Adipocytes
To investigate a functional role of ICAT on adipogenesis, 3T3-L1 pre-adipocytes were transfected with ICAT expression vector using a lentivirus-mediated transfection method. The transfection efficiency was confirmed at 48 h post-infection ( Figure 3A) and almost all pre-adipocytes were GFP positive after 7-day selection by puromycin ( Figure 3B). As expected, lentivirus infection resulted in marked upregulation of ICAT at both mRNA ( Figure S2A) and protein levels ( Figure 3C) (p < 0.05). To identify a repressing effect of ICAT on Wnt/β-catenin activity in 3T3-L1 pre-adipocytes, the pre-adipocytes stably overexpressed ICAT or the empty vector were treated with or without CHIR99021, a GSK3β inhibitor. As shown, Western blot analysis showed that CHIR99021 treatment led to remarkable upregulation of β-catenin and its downstream targets, including c-MYC and CCND1, as well as significantly downregulation of protein abundances for PPARγ and C/EBPα (p < 0.05) ( Figure 3D,E). These effects of CHIR99021 were partial revered by ICAT overexpression. In agreement with protein expression, we found that MDI-induced differentiation was markedly inhibited by CHIR99021 treatment, as shown by oil red O and BODIPY493/503 staining ( Figure 4A,B), as well as quantification of lipid accumulation ( Figure 4C) and intracellular TG (p < 0.05) ( Figure 4D). Interestingly, a repressing effect of GSK3β inhibitor CHIR99021 on differentiation, TG accumulation was partially reversed by ICAT overexpression in 3T3-L1 cells ( Figure 4). Although ICAT overexpression had a modest repressive effect on the protein level of β-catenin, the protein levels of c-MYC and CCND1, two downstream targets of β-catenin, were remarkable downregulated, while those of PPARγ and C/EBPα ( Figure 3D,E), as well as CHIR99021-induced repressing effect on differentiation were significantly reversed by ICAT (Figure 4). This finding indicates that ICAT repressed the β-catenin-TCF4-mediated transactivation instead of directly downregulating the β-catenin expression in the adipocytes in our study.

Overexpression of ICAT Reversed the Effect of GSK3β Inhibitor on Cell Differentiation and Adipogenesis in 3T3-L1 Pre-Adipocytes
To investigate a functional role of ICAT on adipogenesis, 3T3-L1 pre-adipocytes were transfected with ICAT expression vector using a lentivirus-mediated transfection method. The transfection efficiency was confirmed at 48 h post-infection ( Figure 3A) and almost all pre-adipocytes were GFP positive after 7-day selection by puromycin ( Figure 3B). As expected, lentivirus infection resulted in marked upregulation of ICAT at both mRNA ( Figure S2A) and protein levels ( Figure 3C) (p < 0.05). To identify a repressing effect of ICAT on Wnt/β-catenin activity in 3T3-L1 pre-adipocytes, the pre-adipocytes stably overexpressed ICAT or the empty vector were treated with or without CHIR99021, a GSK3β inhibitor. As shown, Western blot analysis showed that CHIR99021 treatment led to remarkable upregulation of β-catenin and its downstream targets, including c-MYC and CCND1, as well as significantly downregulation of protein abundances for PPARγ and C/EBPα (p < 0.05) ( Figure 3D and 3E). These effects of CHIR99021 were partial revered by ICAT overexpression. In agreement with protein expression, we found that MDI-induced differentiation was markedly inhibited by CHIR99021 treatment, as shown by oil red O and BODIPY493/503 staining ( Figure 4A and 4B), as well as quantification of lipid accumulation ( Figure 4C) and intracellular TG (p < 0.05) ( Figure 4D). Interestingly, a repressing effect of GSK3β inhibitor CHIR99021 on differentiation, TG accumulation was partially reversed by ICAT overexpression in 3T3-L1 cells ( Figure 4). Although ICAT overexpression had a modest repressive effect on the protein level of β-catenin, the protein levels of c-MYC and CCND1, two downstream targets of β-catenin, were remarkable downregulated, while those of PPARγ and C/EBPα ( Figure 3D and 3E), as well as CHIR99021-induced repressing effect on differentiation were significantly reversed by ICAT (Figure 4). This finding indicates that ICAT repressed the β-catenin-TCF4-mediated transactivation instead of directly downregulating the β-catenin expression in the adipocytes in our study.

ICAT or E2F1 Knockdown Inhibited MDI-Induced Differentiation of Pre-Adipocytes
ICAT has been reported to be a direct target of E2F1 that is responsible for the suppression of βcatenin activity in cancer cells [17]. To test an involvement of E2F1/ICAT on cell differentiation and adipogenesis, 3T3-L1 cells were transfected with siRNA targeting E2F1 or ICAT. As shown, siRNAmediated knockdown of E2F1 in 3T3-L1 cells resulted in a marked decrease of E2F1 and ICAT at protein level (p < 0.05) ( Figure 5A), confirming the regulation of ICAT expression by the endogenous E2F1 in 3T3-L1 cells. ICAT siRNA transfection led to a reduced protein level of ICAT (p < 0.05) without affecting that of E2F1 ( Figure 5B), validating an efficiency of siRNA in 3T3-L1 pre-adipocytes. Then, cells transfected with non-targeting siRNA (NC), siCAT, or siE2F1 were incubated with MDI to evaluate whether ICAT or E2F1 is functionally required for adipogenesis in 3T3-L1 cells. We found that differentiation induced by MDI was markedly inhibited in siE2F1 or siICAT transfection in 3T3-L1 pre-adipocytes. In addition, MDI-induced increased TG contents ( Figure 5D) and lipid accumulation ( Figure 5E) were dramatically reduced (p < 0.05) by gene silencing of E2F1 or ICAT. Western blot analysis showed that protein levels of C/EBPα and PPARγ were reduced, while those of c-MYC and CCND1 were upregulated (p < 0.05) ( Figure 5C) by siE2F1 or siIACT transfection in 3T3-L1 cells. Compared with siE2F1 transfected cells, ICAT knockdown had a modest inhibitory effect on differentiation ( Figure 5D,E). The TG contents in the stable E2F1 overexpressed adipocytes were markedly reduced (p < 0.05) by gene silencing of ICAT ( Figure 5F), confirming the ICAT was required for E2F1 during differentiation.

ICAT or E2F1 Knockdown Inhibited MDI-Induced Differentiation of Pre-Adipocytes
ICAT has been reported to be a direct target of E2F1 that is responsible for the suppression of β-catenin activity in cancer cells [17]. To test an involvement of E2F1/ICAT on cell differentiation and adipogenesis, 3T3-L1 cells were transfected with siRNA targeting E2F1 or ICAT. As shown, siRNA-mediated knockdown of E2F1 in 3T3-L1 cells resulted in a marked decrease of E2F1 and ICAT at protein level (p < 0.05) ( Figure 5A), confirming the regulation of ICAT expression by the endogenous E2F1 in 3T3-L1 cells. ICAT siRNA transfection led to a reduced protein level of ICAT (p < 0.05) without affecting that of E2F1 ( Figure 5B), validating an efficiency of siRNA in 3T3-L1 pre-adipocytes. Then, cells transfected with non-targeting siRNA (NC), siCAT, or siE2F1 were incubated with MDI to evaluate whether ICAT or E2F1 is functionally required for adipogenesis in 3T3-L1 cells. We found that differentiation induced by MDI was markedly inhibited in siE2F1 or siICAT transfection in 3T3-L1 pre-adipocytes. In addition, MDI-induced increased TG contents ( Figure 5D) and lipid accumulation ( Figure 5E) were dramatically reduced (p < 0.05) by gene silencing of E2F1 or ICAT. Western blot analysis showed that protein levels of C/EBPα and PPARγ were reduced, while those of c-MYC and CCND1 were upregulated (p < 0.05) ( Figure 5C) by siE2F1 or siIACT transfection in 3T3-L1 cells. Compared with siE2F1 transfected cells, ICAT knockdown had a modest inhibitory effect on differentiation ( Figure 5D,E). The TG contents in the stable E2F1 overexpressed adipocytes were markedly reduced (p < 0.05) by gene silencing of ICAT ( Figure 5F), confirming the ICAT was required for E2F1 during differentiation.

. ICAT Did Not Affect Lipid Accumulation in Hela and HepG2 Cells
Hela and HepG2 cells can directly accumulate lipid droplets in the cells with the induction o ic acid [27][28][29]. Importantly, these characteristics of lipid droplets fusion and degradation in th la and HepG2 cells are shared by mature adipocytes, therefore, these two cell lines are used as ce e models in studies related to lipid droplets turnover during the lipogenesis and lipolysis processe -29]. To investigate an effect of ICAT on oleic acid-induced lipid accumulation, Hela and HepG ls were transfected with pIRES2-EGFP-ICAT expression vector or empty vector. The transfectio iciency of cells ( Figure 6A) and purity of cells stably expressed ICAT was confirmed ( Figure 6B e mRNA level of ICAT in cells after G418 selection was confirmed (Supplemental Figure S2B an C). The protein level ( Figure 6C) and in situ expression of ICAT ( Figure 6D) was validated b estern blot analysis and immunofluorescence staining in both cell lines. Oleic acid treatment led t Means without a common letter differ, p < 0.05. C/EBPα, CCAAT-enhancer binding protein α; E2F1, E2 promoter binding factor 1; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; ICAT, Inhibitor of β-catenin and TCF4; PPARγ, peroxisome proliferator activated receptor γ.

ICAT Did Not Affect Lipid Accumulation in Hela and HepG2 Cells
Hela and HepG2 cells can directly accumulate lipid droplets in the cells with the induction of oleic acid [27][28][29]. Importantly, these characteristics of lipid droplets fusion and degradation in the Hela and HepG2 cells are shared by mature adipocytes, therefore, these two cell lines are used as cell line models in studies related to lipid droplets turnover during the lipogenesis and lipolysis processes [27][28][29].
To investigate an effect of ICAT on oleic acid-induced lipid accumulation, Hela and HepG2 cells were transfected with pIRES2-EGFP-ICAT expression vector or empty vector. The transfection efficiency of cells ( Figure 6A) and purity of cells stably expressed ICAT was confirmed ( Figure 6B). The mRNA level of ICAT in cells after G418 selection was confirmed (Supplemental Figure S2B,C). The protein level ( Figure 6C) and in situ expression of ICAT ( Figure 6D) was validated by Western blot analysis and immunofluorescence staining in both cell lines. Oleic acid treatment led to increased TG contents as shown by oil red O staining in Hela and HepG2 cells ( Figure 6E). ICAT overexpression did not affect oleic acid-induced lipid accumulation in both cell lines ( Figure 6F). These data indicated that ICAT did not implicate in lipid accumulation at the surface of Hela and HepG2 cells.

Discussion
In the present study, we found a novel function of E2F1/ICAT on promoting pre-adipocytes differentiation and lipid accumulation in 3T3-L1 cells. Depletion of E2F1 or ICAT by siRNA led to reduced differentiation and decreased TG contents by inhibiting expression of C/EBPα and PPARγ in MDI-treated cells. Overexpression of ICAT in 3T3-L1 pre-adipocytes markedly promoted the adipogenesis induced by MDI. In addition, this promotive effect was also observed in CHIR99021treated cells, in which Wnt/β-catenin was activated by inhibiting GSK3β, a negative regulator of Wnt/β-catenin signaling, therefore repressing differentiation of pre-adipocytes. An inhibitory effect of CHIR99021 on adipogenesis was partially attenuated by lentivirus-mediated ICAT overexpression in 3T3-L1 cells. In addition, the adipogenesis in the stable E2F1 overexpressed adipocytes were markedly reduced by ICAT knockdown. This effect of E2F1/ICAT on differentiation was mainly mediated by regulating Wnt/β-catenin signaling. These findings revealed a crosstalk between E2F1/ICAT signaling and Wnt/β-catenin on pre-adipocyte differentiation (Figure 7).

Discussion
In the present study, we found a novel function of E2F1/ICAT on promoting pre-adipocytes differentiation and lipid accumulation in 3T3-L1 cells. Depletion of E2F1 or ICAT by siRNA led to reduced differentiation and decreased TG contents by inhibiting expression of C/EBPα and PPARγ in MDI-treated cells. Overexpression of ICAT in 3T3-L1 pre-adipocytes markedly promoted the adipogenesis induced by MDI. In addition, this promotive effect was also observed in CHIR99021-treated cells, in which Wnt/β-catenin was activated by inhibiting GSK3β, a negative regulator of Wnt/β-catenin signaling, therefore repressing differentiation of pre-adipocytes. An inhibitory effect of CHIR99021 on adipogenesis was partially attenuated by lentivirus-mediated ICAT overexpression in 3T3-L1 cells. In addition, the adipogenesis in the stable E2F1 overexpressed adipocytes were markedly reduced by ICAT knockdown. This effect of E2F1/ICAT on differentiation was mainly mediated by regulating Wnt/β-catenin signaling. These findings revealed a crosstalk between E2F1/ICAT signaling and Wnt/β-catenin on pre-adipocyte differentiation (Figure 7).
Adipogenesis is a critical process implicated in the development of metabolic disease. Both clinical and experimental data have indicated that small molecules or compounds with the ability to interfere in the adipogenesis process might be a potentially therapeutic agent for obesity [30,31]. Considering that the number of fat cells in the body is relatively stable in adults [32], it is of great significance to reveal underlying mechanisms responsible for pre-adipocytes differentiation and adipogenesis. Wnt/β-catenin is a crucial repressor for adipogenesis, whose activity is inhibited during pre-adipocytes differentiation [22]. In response to extracellular stimulations, the canonical Wnt/β-catenin signaling is activated and leads to stabilization of β-catenin in the cytoplasm, accumulated β-catenin translocated to the nucleus, and interacts with T-cell factor/ lymphoid enhancer-binding factor (TCF/LEF) to trans-activate downstream targets, such as CCND1, c-MYC, PPARγ, and C/EBPα, therefore regulating cellular metabolism [33,34]. Consistently, it has been reported that some Chinese herbal medicines, such as curcumin, shikonin, and resveratrol, regulate metabolic homeostasis by regulating Wnt/β-catenin signaling pathway [35][36][37]. Adipogenesis is a critical process implicated in the development of metabolic disease. Both clinical and experimental data have indicated that small molecules or compounds with the ability to interfere in the adipogenesis process might be a potentially therapeutic agent for obesity [30,31]. Considering that the number of fat cells in the body is relatively stable in adults [32], it is of great significance to reveal underlying mechanisms responsible for pre-adipocytes differentiation and adipogenesis. Wnt/β-catenin is a crucial repressor for adipogenesis, whose activity is inhibited during pre-adipocytes differentiation [22]. In response to extracellular stimulations, the canonical Wnt/βcatenin signaling is activated and leads to stabilization of β-catenin in the cytoplasm, accumulated βcatenin translocated to the nucleus, and interacts with T-cell factor/ lymphoid enhancer-binding factor (TCF/LEF) to trans-activate downstream targets, such as CCND1, c-MYC, PPARγ, and C/EBPα, therefore regulating cellular metabolism [33,34]. Consistently, it has been reported that some Chinese herbal medicines, such as curcumin, shikonin, and resveratrol, regulate metabolic homeostasis by regulating Wnt/β-catenin signaling pathway [35][36][37].
ICAT has been identified as a direct transcriptional target of E2F1 [17], through which E2F1 interacts with Wnt/β-catenin and regulates proliferation in colon cancers [7,17,38,39]. In addition to a well-known function in tumorigenesis of humans, E2F1 has been described as a transcription factor participates in the development of multiple metabolic diseases, including obesity, diabetes, and fatty liver disease [10,12,14]. An in vitro study has shown that induction of E2F1 during the early phase of differentiation is critical for adipogenesis [7]. However, mechanistic insight responsible for this regulation remains largely unknown.
In the present study, we found that the protein levels of E2F1 were reduced at 24 h treatment, which was dramatically enhanced on the day 3 of differentiation, and then reduced to an undetectable level as the cell differentiated into mature adipocytes. This result was consistent with a ICAT has been identified as a direct transcriptional target of E2F1 [17], through which E2F1 interacts with Wnt/β-catenin and regulates proliferation in colon cancers [7,17,38,39]. In addition to a well-known function in tumorigenesis of humans, E2F1 has been described as a transcription factor participates in the development of multiple metabolic diseases, including obesity, diabetes, and fatty liver disease [10,12,14]. An in vitro study has shown that induction of E2F1 during the early phase of differentiation is critical for adipogenesis [7]. However, mechanistic insight responsible for this regulation remains largely unknown.
In the present study, we found that the protein levels of E2F1 were reduced at 24 h treatment, which was dramatically enhanced on the day 3 of differentiation, and then reduced to an undetectable level as the cell differentiated into mature adipocytes. This result was consistent with a previous study [14]. The shift of E2F1 during the differentiation is dependent on its function in cell proliferation and differentiation. Before induction of differentiation, 3T3-L1 pre-adipocytes was proliferated and reached confluency, therefore, protein level of E2F1 was higher as seen at day 0. After that, 3T3-L1 preadipocytes were growth-arrested and the proliferative effect of E2F1 is repressed by repressive regulator of cell cycle proliferation program, and leading to a reduced protein level at day 1. After being treated with MDI for 72 h, the activity of E2F1 was activated to facilitate the adipogenesis during differentiation as shown in Figure 1C. Of interest, the protein level of E2F1 was reduced to an undetectable level, indicating a requirement of E2F1 for initiation of differentiation, whose activity was not needed at the later stages of adipocyte differentiation [7,14]. In addition, we observed a similar expression profile on the protein level of ICAT, followed by the induction of PPARγ and C/EBPα, as well as decreased protein level of β-catenin, indicating a potential implication of E2F1/ICAT in adipogenesis.
Adipogenesis is associated with decreased activation of β-catenin signaling. To investigate a promotive effect of ICAT on adipogenesis by repressing β-catenin, 3T3-L1 pre-adipocytes transfected with ICAT or empty vector were treated with GSK3β inhibitor to upregulate β-catenin. As expected, GSK3β inhibitor treatment led to enhanced β-catenin and blocked cell differentiation. This effect was partially reversed by ICAT overexpression as shown by alterations in lipid droplets and TG contents. PPARγ and C/EBPα are critical regulators for cell differentiation, which were negatively regulated by CCND1 and c-MYC, respectively [40][41][42], through which β-catenin exert a regulatory effect on adipogenesis [43]. ICAT overexpression reversed GSK3β inhibitor-induced phenotype alteration as well as protein levels of adipogenic regulators, therefore exerting a promotive effect on differentiation. Importantly, overexpression of ICAT in 3T3-L1 pre-adipocytes also markedly promoted the adipogenesis compared to the control cells during differentiation. Moreover, MDI-induced differentiation was attenuated in siE2F1 or siICAT transfected cells and protein expressions of Wnt/β-catenin targets including c-MYC and CCND1 were increased, solidifying a promotive effect of endogenous E2F1/ICAT on differentiation by repressing Wnt/β-catenin signaling. It has been reported that ICAT inhibit the interaction of β-catenin with TCF4, leading to the repression of β-catenin-TCF4-mediated transactivation [18]. Our previous study about the TCF4-dependent luciferase reporter activity also confirmed that ICAT was a robust inhibitor of β-catenin/TCF4 activity [17]. It is worthwhile to note that ICAT overexpression reversed GSK3β inhibitor-induced phenotype alteration to promote differentiation and downregulated the c-MYC and CCND1 without affecting the accumulation of β-catenin induced by the GSKi in the nucleus, indicating that ICAT repressed the β-catenin-TCF4-mediated transactivation instead of directly downregulating its protein level. Furthermore, ICAT knockdown inhibited the adipogenesis in the stable E2F1 overexpressed adipocytes, confirming the ICAT was required for E2F1 during differentiation. These findings demonstrated that ICAT is a novel target that is responsible for E2F1 s regulation for pre-adipocyte differentiation by interacting with Wnt/β-catenin signaling.
Of note, we found that siE2F1 transfected cells had a higher inhibitory effect on MDI-induced differentiation, as compared with siICAT transfected cells. This result is due to the following reasons. First, besides ICAT as described herein, E2F1 can regulate pre-adipocytes differentiation by inducing other targets, such as Perp-1 [44,45] and PPAR [46], and contribute to lipid accumulation in 3T3-L1 cells [47]. Second, E2F1 can regulate cell cycle progress in multiple cells, knockdown of E2F1 might affect adipogenesis due to its interfering effect on cell proliferation as previously described [48,49]. Notably, ICAT overexpression did not affect oleic acid-induced lipid accumulation in cell membrane of Hela and HepG2 cells. It has been reported that Wnt/β-catenin signaling is not implicated in oleic acid-induced lipid accumulation [50,51], therefore, it is plausible that ICAT did not act in this model. The findings indicated that E2F1/ICAT mainly regulated initiation of pre-adipocytes differentiation at early stage and led to lipid accumulation, without affecting lipid content in mature adipocytes. These data indicated a specific effect of E2F1/ICAT on differentiation and lipid accumulation in pre-adipocytes. It has been reported that protein level E2F1 is markedly reduced in later stage of differentiation, but the reason is not known. Another novel finding of our study is we showed, for the first time, that E2F1/ICAT is critical for the maintenance of PPARγ and C/EBPα by blocking the c-MYC and CCND1, two negative regulators, such as PPARγ and C/EBPα, and contributing to the progress of differentiation.
It has been reported that E2F1 transcriptional activity is enhanced in obese subjects and non-alcoholic fatty liver disease [52]. Our data provided herein indicated that E2F1 might promote adipogenesis by ICAT-mediated inhibition on β-catenin. Of note, E2F1 is a transcriptional factor with multiple functions, including cell proliferation, differentiation, and apoptosis [6]. Depletion of E2F1 might affect adipogenesis as well as other biological processes, and lead to potentially deleterious effects in both humans and animals [53]. Considering that ICAT is required for E2F1 s activity on pre-adipocyte differentiation, ICAT might be a therapeutic target for metabolic diseases, especially the patients with enhanced protein level of E2F1 and ICAT.

Conflicts of Interest:
The authors declare no conflict of interest.