E3 Ubiquitin Ligase E6AP Negatively Regulates Adipogenesis by Downregulating Proadipogenic Factor C/EBPalpha

CCAAT/Enhancer Binding Protein Alpha (C/EBPα) is a key transcription factor involved in the adipocyte differentiation. Here for the first time we demonstrate that E6AP, an E3 ubiquitin ligase inhibits adipocyte differentiation in 3T3-L1 cells as revealed by reduced lipid staining with oil red. Knock down of E6AP in mouse 3T3L1 preadipocytes is sufficient to convert them to adipocytes independent of external hormonal induction. C/EBPα protein level is drastically increased in E6AP deficient 3T3L1 preadipocytes while inverse is observed when wild type E6AP is over expressed. We show that transient transfection of wild type E6AP downregulates C/EBPα protein expression in a dose dependent manner while catalytically inactive E6AP-C843A rather stabilizes it. In addition, wild type E6AP inhibits expression of proadipogenic genes while E6AP-C843A enhances them. More importantly, overexpression of E6AP-C843A in mesenchymal progenitor cells promotes accumulation of lipid droplets while there is drastically reduced lipid droplet formation when E6AP is over expressed. Taken together, our finding suggests that E6AP may negatively control adipogenesis by inhibiting C/EBPα expression by targeting it to ubiquitin-proteasome pathway for degradation.


Introduction
Adipose tissue metabolism is crucial for modulation of metabolic processes and whole body insulin sensitivity homeostasis [1]. The trend of increased obesity and diabetes observed over the past few years has made it essential to get new insights into the understanding of adipocyte biology and adipogenesis per se for novel possibilities of preventive and effective treatment [2]. During adipocyte differentiation, fibroblast-like preadipocytes differentiate into lipid-loaded adipocytes, a highly insulin-sensitive cell type. These preadipocytes differentiate under hormonal induction and there are arrays of transcription factors which execute this process of differentiation [3,4]. Peroxisome proliferator-activated receptor gamma (PPARc) [5] and members of the CCAAT/enhancerbinding protein (C/EBP) family of proteins are key factors involved in adipogenesis [6]. C/EBP family of proteins regulate proliferation and differentiation of large number of cell types and were among first to be implicated in adipocyte differentiation [7,8,9]. C/EBPa is the founding member of the C/EBP family and is expressed predominantly in post-mitotic cells. Overexpression of C/EBPa in preadipocytes was shown to arrest 3T3-L1 preadipocytes in G0/G1 [10], providing the first evidence that C/ EBPa can enforce cell-cycle exit. Moreover, Ectopic expression of C/EBPa in mouse fibroblast cell lines has been shown to promote adipogenic differentiation [11]. During adipogenesis C/EBPa expression is induced and it activates genes specifically expressed in differentiated fat cells [12]. C/EBPb and C/EBPd are induced early during the adipogenesis, while C/EBPa is induced at a later stage. C/EBPb and C/EBPd increase the PPARc gene expression by binding to its promoter. PPARc subsequently activates the expression of multiple genes involved in lipogenesis and adipogenesis. PPARc also activates the C/EBPa gene. Notably, C/ EBPa can bind to the PPARc promoter and enhance the expression of PPARc, thereby creating a proadipogenic feedforward loop (13).
E6AP, a 100 kDa cellular protein belongs to a class of functionally related E3-ubiquitin-protein ligases defined by the domain homologous to the carboxy terminus (HECT) domain [13]. E3 ligases are known to ubiquitinate and degrade various transcription factors [14] and play major role in the regulation of cell cycle and various other cellular functions. E6AP has been reported to play role in the ubiquitination of various proteins like p53, p27, PML-RARa which play important role as tumor suppressors and cell proliferation inhibitors [15,16,17]. In recent years some of ubiquitin ligases have been implicated in the modulation of adipogenesis. Fbwx7 degrades C/EBPa and negatively regulates adipogenesis [18] while Skp2 has been shown to mediate p27 degradation during S/G2 phase progression of adipocyte hyperplasia [19]. In a recent report, using mass spectrometry based proteomics approach we identified E6AP as a target of tamoxifen. In addition, previously, we and others have also shown that C/EBPa can be ubiquitinated and degraded via proteasomal pathway [20]. However, to our knowledge, except Fbwx7 no other E3 ligase for C/EBPa ubiquitination is known.
We hypothesized that E6AP may target C/EBPa for degradation. C/EBPa is a transcription factor involved in the regulation of granulopoiesis, in fact, in a recent study we have shown that E6AP promotes C/EBPa ubiquitination leading to its proteasome mediated degradation and thereby inhibits granulopoiesis in blood cells (data not shown). As C/EBPa plays an important role in adipocyte differentiation we further explored its role during adipogenesis in the presence and absence of E6AP. In the present study, we show that E6AP targets C/EBPa for ubiquitin mediated degradation in 3T3L1 cells and thus inhibits adipogenesis. In contrary, E6AP knock down in 3T3L1 cells by siE6AP and over expression of catalytically inactive E6AP-C843A induces adipogenesis. Taken together, our data suggests that E6AP may negatively regulate adipogenesis by targeting C/EBPa for degradation.

Cell Culture and Expression Plasmids
Mouse preadipocyte cell line 3T3L1 was cultured in DMEM supplemented with 10% FBS and antibiotics. Mouse mesenchymal progenitor cells were isolated from bone marrow of C57Bl/6J mice and were cultured in DMEM supplemented with 10% FBS and antibiotics. Cells were incubated in 5% CO2 humidified chamber.

Plasmids and siRNA
Expression plasmids for pcDNA3.1-E6AP [15] and pCAG-HA-E6AP-C843A [21] were kind gifts from Nihar Jana [15] and Ikuo Shoji [21] respectively. The siE6AP and scrambled siRNA were purchased from Dharmacon. E6AP-C843A is a catalytically inactive form of E6AP where active site cysteine residue is substituted with alanine (C843A). This cysteine residue present in the catalytic domain transfers ubiquitin directly to the substrate via ubiquitin-enzyme cascade leading to their degradation.

Oil Red O Staining
Fully confluent 3T3L1 cells were stimulated with 0.5 mM methylisobutylxanthine, 1 uM dexamethasone and 10 ug/ml insulin (MDI) in DMEM supplemented with 10% FBS to induce differentiation. After 48 h, culture medium was replaced with DMEM supplemented with 10% FBS and 10 ug/ml insulin for additional 48 h. Cells were then fed every other day with DMEM containing 10% FBS. After cells were fully differentiated, Oil red O staining was performed by fixing the cells in 10% formalin. Oil red O stain was prepared in water and was filtered with 0.22 u filter. After staining cells with Oil red O for 10 minutes, cells were washed thrice with water and were air dried. Cells were photographed under light microscope (Leica).

In vivo Ubiquitination Assay
It was performed as previously described [23]. Briefly, 3T3-L1 preadipocytes pre-treated with MDI for 48 h were transiently transfected with expression plasmids for E6AP, E6AP-C843A and His-ubiquitin as indicated. Post 48 h transfection, whole cell extracts (WCEs) were prepared and endogenous C/EBPa was coimmunoprecipitated using 3 ug of C/EBPa antibody and protein G Agarose beads (Millipore). The co-immunoprecipitates were then resolved on 8% SDS-PAGE and immunoblotted with anti-His antibody. Same blot was stripped and reprobed with anti-C/ EBPa antibody to assess and confirm the expression and immunoprecipitation of endogenous C/EBPa.

Quantitative PCR (Real Time) Analysis
3T3L1 cells were transfected with siE6AP and siRNA control, 48 h post transfection, cells were treated with MDI until ten days and subsequently RNA was isolated using Trizol reagent as previously described [24]. Further, RNA was retro-transcribed in to cDNA and subsequently used for quantitative PCR analysis on Roche Light Cycler 480 using SYBR green master mix from Applied Biosystems. Statistical analysis was performed using DD CT method. Primers used for Real time PCR are listed in table 1.

Isolation of Mesenchymal Progenitor Cells
Mesenchymal stem cells were isolated using established protocol [25]. This study was carried out in strict accordance with CPCSEA/IAEC (Registration No.: 34/1999 dated 11-3-99) guide lines. The study involving use of C57Bl/6J mice was approved by the Institutional Animal Ethics Committee at Central Drug Research Institute. Approval reference no. for this study was IAEC/2012/47, dated 16.05.2012. Briefly, bone marrow aspirates were flushed out from tibia and femur, mixed with MACS buffer and loaded on histopaque-1077 (Sigma) gradient to harvest cells from the interphase. Isolated cells were washed with MACS buffer (1X PBS, 2 mM EDTA and 2% FBS) and seeded for expansion in   Post 48 h transfection, WCEs were prepared and resolved on 10% SDS-PAGE followed by immunoblotting with C/EBPa, E6AP and b actin antibodies; lysates of 293T alone and transfected with C/EBPa were used as positive and negative control, note that there is no endogenous expression of C/EBPa in 293T. E6AP promotes C/EBPa degradation through ubiquitinproteasome pathway: (c) 3T3-L1 preadipocytes pre-treated with MDI for 48 h were transiently transfected with expression plasmids for E6AP, E6AP-C843A and His-ubiquitin. Post 48 h Transfection, WCEs were prepared and C/EBPa was co-immunoprecipitated. IgG was used as a control. Coimmunoprecipitates were resolved on 8% SDS-PAGE and blots were probed with anti-His and anti-C/EBPa antibody respectively. E6AP inhibits C/ EBPa transactivation potential to activate PPRE-Luc: (d) E6AP mediated downregulation of C/EBPa curtails its transactivation potential: 3T3-L1 preadipocytess were transiently transfected with pPPRE-luc reporter and expression plasmids for C/EBPa, E6AP and E6AP-C843A. 24

E6AP Negatively Regulates Adipogenesis
In recent years some of E3 ligases have been shown to be involved in the regulation of adipogenesis [15,16,19]. In the present study we have investigated if E6AP, an E3 ubiquitin ligase also modulates adipogenesis. To address this, we transiently transfected 3T3-L1 preadipocytes with wild type E6AP and a dominant negative mutant E6AP-C843A [21]. Over expression of E6AP potentially inhibited adipogenesis even in the presence of adipogenic factors (MDI: 0.5 mM Methylisobutylxanthine, 1 uM Dexamethasone and 10 ug/ml Insulin), whereas no inhibition was seen with E6AP-C843A; Intensity of staining correlated with extent of adipogenesis induction (Fig. 1).
This result suggests that E6AP has inhibitory effect on the differentiation of 3T3-L1 cells even in the presence of hormonal induction. To further explore the role of E6AP on the adipogenesis we studied its effect on an important proadipogenic factor C/ EBPa as in a recent study we have shown that E6AP targets C/ EBPa for ubiquitin mediated degradation [26].

E6AP Downregulates Proadipogenic Factor C/EBPa
C/EBPa is a key transcription factor involved in the regulation of adipogenesis. We and others have previously shown that C/ EBPa can be ubiquitinated and degraded via proteasome mediated pathway [20,27]. Since our data suggest that E6AP, an E3 ubiquitin ligase negatively regulates adipogenesis, we sought to assess if E6AP modulates stability of C/EBPa protein.
We sought to assess C/EBPa protein expression in 3T3-L1 preadipocytes by transiently transfecting increasing amounts of E6AP or E6AP-C843A (0.5 ug-4.0 ug). Post 48 h transfection, cell lysates were prepared and resolved on 10% SDS-PAGE (Note that cells were induced with MDI for just 2 days for inducing expression of C/EBPa prior to transfection). Immunoblot with C/ EBPa antibody shows that E6AP over expression led to substantial decrease in endogenous C/EBPa protein expression while it was  . 2a, b). Notably, C/EBPa protein expression was restored in MG132 treated cells (in 4.0 ug E6AP transfected 3T3-L1 cells) which indicates that C/EBPa protein stability seems to be regulated via proteasomal pathway.
To further consolidate the notion that E6AP destabilizes C/ EBPa protein expression by promoting its ubiquitination and subsequent protasome mediated degradation, we performed in vivo ubiquitination assay. For this, 3T3L1 preadipocytes prior induced with MDI for 48 h were co-transfected with expression plasmids for E6AP, E6AP-C843A and His-Ubiquitin. Post 48 h transfection, WCEs were prepared and C/EBPa was co-immunoprecipitated with anti-C/EBPa antibody. The co-precipitates were resolved on 8% SDS-PAGE and immunoblotted with anti-His antibody. Heavy C/EBPa ubiquitination as a ladder was observed in E6AP and His-Ubiquitin co-transfected cells while almost no ubiquitination was seen in cells co-transfected with E6AP-C843A and His-Ubiquitin suggesting E6AP indeed promotes ubiquitination of C/EBPa leading to its degradation via proteasome pathway. Same blot was stripped and probed with C/EBPa antibody to confirm the presence of C/EBPa in the immunoprecipitates ( fig. 2c).

E6AP Mediated Downregulation of C/EBPa Negatively Affects its Transactivation Potential
Since E6AP downregulates C/EBPa protein expression by promoting its degradation, we asked if this E6AP mediated downregulation of C/EBPa has any affect on C/EBPa transactivation potential. To answer this, we performed luciferase reporter assay on a minimal PPRE-Luc promoter containing PPRE response elements. Indicated amounts of reporter vector and expression plasmids for C/EBPa, E6AP and E6AP-C843A were transfected in 293T cells. Post 24 h transfection, luciferase activity was measured which showed that co-transfection of E6AP with C/ EBPa substantially inhibited C/EBPa transactivation capacity in a dose dependent manner ( fig. 2d). Further, MG132 and Lactacystin (LCN) treatment efficiently restored C/EBPa transactivation potential even in the presence of E6AP. Additionally, cotransfection of E6AP-C843A with C/EBPa did not inhibit transactivation potential of C/EBPa. This data indicates that catalytically active E6AP negatively modulates C/EBPa protein stability thereby hampers its functions.

siE6AP Promotes Adipogenesis
Since overexpression of E6AP led to decrease in C/EBPa protein expression while catalytically inactive E6AP stabilized it, we sought to assess if inhibition of E6AP via siRNA against E6AP   stabilizes C/EBPa protein expression. 3T3-L1 preadipocytes pretreated with MDI for 48h were transiently transfected with control siRNA and siE6AP; 48 h post transfections WCEs were prepared and resolved on 10% SDS-PAGE. Immunoblot with C/EBPa and E6AP antibody showed that siE6AP potentially reduced E6AP expression with concomitant increase in C/EBPa expression ( fig. 3a). This makes a very interesting observation and confirms our hypothesis that E6AP acts as negative regulator of adipogenesis probably by targeting C/EBPa.
In order to further corroborate the anti-adipogenic role of E6AP, we used siRNA against E6AP. 3T3-L1 cells were transfected either with control siRNA (Scrambled) or siE6AP and were further grown for 10 days in DMEM with or without MDI. Post 10 day transfection; cells were stained with Oil red O staining. As shown in fig. 3b; as compared to control siRNA, siE6AP transfected preadipocytes differentiated into adipocytes showing accumulation of large number of intracellular lipid droplets in response to MDI treatment. Notably, siE6AP transfection without MDI treatment also led to substantial intracellular lipid droplet accumulation which suggests that E6AP inhibition per se may induce adipogenesis.

E6AP Knockdown Mediated Induction of Adipogenesis is Marked by Increase in Expression of Proadipogenic Factors
siE6AP transfection in 3T3-L1 cells enhances lipid droplet formation both in the presence and absence of MDI treatment. This suggests E6AP to be a negative switch in the process of adipogenesis. We therefore asked if this enhanced lipid droplet formation (adipogenesis) is also marked with increase in the expression of adipogenic markers (C/EBPa, PPARc, SREBP1c and Leptin). PPARc and C/EBPa are known to function cooperatively to transactivate adipocyte genes and thereby bring about adipocyte differentiation [28]. Leptin is secreted by adipocytes and it may function in regulating body fat mass. Leptin levels are elevated in human obesity and animal models of obesity [29]. Role of SREBP1c in adipogenesis is suggested by its mRNA level increase during adipocyte differentiation where it further transactivates other lipogenic genes [8] [30]. We performed real time PCR for indicated adipogenic genes from mRNA isolated from different conditions as mentioned in fig. 4a. Expression of adipogenic markers (C/EBPa, PPARc, SREBP1c, Leptin and Adipsin) in siE6AP transfected cells treated with MDI compared to control correlated well with enhanced intracellular lipid droplet formation. This increase in the expression of proadipogenic factors in the presence siE6AP as compared to control siRNA treated with MDI indicates E6AP inhibition is required to drive adipogenesis.

Over Expression of Wild Type E6AP Down Regulates while Catalytically Inactive E6AP-C843A Up Regulates Proadipogenic Factors
E6AP inhibits C/EBPa protein expression leading to its functional inactivation ( fig. 3 and 4); moreover, proportion of lipid droplet formation varies in E6AP and E6AP-C843A transfected 3T3-L1 cells treated with MDI. In addition, we showed that siE6AP transfection enhanced mRNA expression levels of proadipogenic factors; therefore, we further assessed the expression of these adipogenic markers (C/EBPa, PPARc, SREBP1c and Leptin) in 3T3-L1 cells transfected with E6AP and E6AP-C843A. As shown in fig. 4b mRNA expression levels of these proadipogenic factors correlated well with proportion of lipid droplet formation in E6AP and E6AP-C843A transfected cells treated with MDI ( fig. 1). Taken together these data endorse our finding that E6AP negatively regulates adipogenesis.

E6AP Over Expression Inhibits while E6AP-C843A Promotes Adipocyte Differentiation of MDI Treated Mesenchymal Stem Cells Isolated from Murine Bone Marrow
In order to further corroborate our finding in a more physiological setting, we isolated mouse mesenchymal progenitor cells from bone marrow of C57Bl/6J mice. These mesenchymal progenitor cells can differentiate in adipogenic or osteogenic lineage depending upon the kind of hormonal inductions. Therefore, we separately over expressed both E6AP and E6AP-C843A in these progenitors cells and cultured in DMEM in the presence or absence of MDII (methylisobutylxanthine, dexamethasone, insulin and indomethacin). Post 10 days of transfection and culture in MDII, Oil red O staining was performed to assess the adipogenic potential of these transfected cells. Intracellular lipid droplet formation was observed in the mesenchymal progenitor cells transfected with E6AP-C843A mutant even in the absence of MDII induction. However, in E6AP transfected cells lipid droplet formation was inhibited even in the presence of MDII ( fig. 5). This further consolidates our finding that E6AP negatively regulates adipogenesis apparently by downregulating proadipogenic protein C/EBPa.

Discussion
C/EBPa is an important transcription factor involved in the regulation of various cellular processes including adipogenesis. E3 Ligases have attracted great interest because of their role in a variety of cellular processes by virtue of their ability to target regulatory proteins for degradation. In recent years, some of E3 ligases have also been implicated in the regulation of adipogenesis. Fbwx7 targets C/EBPa for phosphorylation dependent ubiquitin mediated degradation and thereby negatively regulates adipogenesis [18]. Expression of Fbxw7 is down-regulated early during adipogenesis, leading to the stabilization of C/EBPa, which in turn results in enhanced expression of proadipogenic target genes and enhanced adipogenesis. In addition, Skp2 has been shown to mediate p27 degradation during S/G2 phase progression of adipocyte hyperplasia [19]. HECT (homologous to the E6AP C terminus) ubiquitin ligases, like E6AP possess E3 ligase activity and are involved in diverse functions in eukaryotic cells, largely through their ability to ubiquitinate the target proteins thereby affecting their localization, interactions, and stability. Recently, HECT domain containing E3 ubiquitin Liagse WWP1 was implicated in lipid droplet turnover. WWP1 interacts with and ubiquitinates lipid droplet binding protein Spartin/SPG20. SPG20 in conjunction with TIP47 regulates lipid droplet number as well as size [31].
Here for the first time we show that a HECT domain containing E3 ubiquitin ligase E6AP inhibits adipogenesis by downregulating proadipogenic factor C/EBPa. Since C/EBPa is a master transcription factor that regulates number of genes required for adipogenic differentiation switch, proteins targeting this master regulator may be very crucial in regulating adipogenesis. E6AP seems to be one such crucial protein since it inhibits adipogenesis by downregulating C/EBPa protein expression ( fig. 1).
Our data is further substantiated by finding that siE6AP mediated knock down of E6AP is sufficient to induce adipocyte differentiation in 3T3-L1 cells even in the absence of hormonal treatment ( fig.2). Considerable knock down of E6AP in preadipocyte 3T3-L1 cells led to accumulation of C/EBPa protein and subsequent increased intracellular lipid droplet formation. This provides clear evidence regarding direct correlation between E6AP and C/EBPa activity because increase in lipid droplet formation (a direct measure of adipogenesis) is in fact induced by C/EBPa. This clearly demonstrates that E6AP is an inhibitory factor for the induction of adipogenesis. Fact that over expression of E6AP in 3T3-L1 cells dramatically destabilizes endogenous C/ EBPa protein and inhibits its transactivation capacity provides an explanation for inhibition of adipogenesis by E6AP. Note that, 3T3L1 cells are unique in a way that serum alone is not able to direct re entry of contact inhibited cells in cell cycle [32]. Several adipogenic agents (methylisobutylxanthine, dexamethasone and insulin) are required for induction of cell cycle in density dependent growth inhibited cells and thereby leading it to adipocytic differentiation [33]. Overexpression of catalytic domain mutant E6AP-C843A in mouse mesenchymal stem cells induces intracellular lipid droplet formation even in the absence of adipogenic factors which strongly indicates a role for E6AP in adipogenic differentiation. This is very interesting as mesenchymal progenitor cells are the pluripotent cells which can differentiate in varied lineages depending upon kind of hormonal inductions [34]. This apparently provides a clue that E6AP is an important regulatory switch in the adipocytic differentiation.
E6AP is encoded by UBE3A locus on chromosome 15. This chromosome 15 is important in various genetic disorders like Angelman syndrome (AS) [35] and Prader-Willi syndrome (PWS) [36]. In most of the Angelman syndrome cases there is deletion in maternal copy of 15q11-13 leading to mutation in UBE3A gene. Although, in PWS the exact role of UBE3A is not known but there is deletion in paternal copy of chromosome 15. However, one interesting phenotype reported both in AS and PWS patients is obesity. AS patients are more susceptible to gain weight [37,38], while obesity is one of the major problems in PWS patients leading to higher mortality. The exact genes involved in these phenotypes are not yet known, and in fact there may be some genes located on chromosome 15 unreported thus far which may control obesity. Nonetheless, our data showing negative regulation of adipogenesis by E6AP via degradation of C/EBPa may also provide an explanation for increased body weight and obesity in these patients. However, this further needs to be investigated in more detail in those patients.
Based on our data presented in this study we propose a hypothetical model ( fig. 6) where we show that E6AP negatively regulates adipogenesis. High expression of the E3 ubiquitin ligase E6AP in preadipocytes may promote degradation of proadipogenic transcription factor C/EBPa, thereby attenuating its transactivation potential and thus adipocyte differentiation. Expression of E6AP is down-regulated early during adipogenesis, leading to the stabilization of C/EBPa, which in turn binds to the promoters of target genes and effectively results in increased expression of proadipogenic genes and enhanced adipogenesis.
Furthermore, since E6AP inhibits adipogenesis by downregulating C/EBPa via targeting it for ubiquitination and subsequent proteasome mediated degradation; its potential as a target protein in the treatment of obesity and type-2 diabetes may further be explored. Also, it is very important to identify the signals and factors involved in the adipogenesis which may regulate E6AP expression. In addition, since C/EBPa plays crucial role in many other cellular processes, role of E6AP in regulating C/EBPa functions in those processes may also be investigated.