NR4A1 retards adipocyte differentiation or maturation via enhancing GATA2 and p53 expression

Abstract Nuclear receptor subfamily 4 group A member 1 (NR4A1) is an orphan nuclear receptor with diverse functions. It has been reported that NR4A1, as a transcriptional activator, is implicated in glucose and lipid metabolism. The aim of this study was to investigate the regulatory role of NR4A1 in adipogenesis and explore the underlying mechanisms. Quantitative real‐time PCR and Western blotting were used to analyse the expression of genes involved in synthesis and mobilization of fats in vivo and in vitro. Dual‐luciferase reporter assay was conducted to study the regulatory mechanisms of NR4A1. Our data from in vivo study confirmed that NR4A1 knockout (KO) mice fed with high‐fat diet were more prone to obesity, and gene expression levels of PPARγ and FAS were increased in KO mice compared to controls; our data from in vitro study showed that NR4A1 overexpression in 3T3‐L1 pre‐adipocytes inhibited adipogenesis. Moreover, NR4A1 enhanced GATA binding protein 2 (GATA2) expression, which in turn inhibited peroxisome proliferator‐activated receptor γ (PPARγ); NR4A1 inhibited sterol regulatory element binding transcription factor 1 (SREBP1) and its downstream gene fatty acid synthase (FAS) by up‐regulating p53. NR4A1 inhibits the differentiation and lipid accumulation of adipocytes by enhancing the expression of GATA2 and p53.

(FABP) and other fat synthesis-related genes are activated, 8,9 leading to the synthesis of lipids droplets. 10 On the other hand, lipolysis is the catabolic pathway of fatty acid (FA) cycle. Fatty acids are essential for energy production and the synthesis of most lipids involved in cell structure and cellular signalling. However, an oversupply of FA is highly detrimental. 11 Factors involved in lipolytic process include adipose triglyceride lipase (ATGL), hormone-sensitive lipase (HSL), lipoprotein lipase (LPL) and other molecules. To date, some enzymes have been reported to account for complete hydrolysis of triacylglycerol (TAG) molecules in cellular lipid stores: ATGL selectively performs the first and rate-limiting step hydrolysing TAGs to generate diacylglycerols (DAGs) and non-esterified fatty acids. HSL is a multifunctional enzyme that is capable of hydrolysing a variety of acylesters including TAG, DAG and monoacylglycerol (MAG). LPL is the rate-limiting enzyme for the hydrolysis of the triglyceride and very low-density lipoproteins. [12][13][14] The fate of obesity formation depends on the balance between the processes of adipogenesis or lipid accumulation and lipolysis.
There are many regulatory factors modulating adipogenesis and lipolysis. GATA2, GATA3, delta like non-canonical Notch ligand 1 and wingless-type MMTV integration site family, member 1, are involved in the process of adipogenesis. It was reported that GATA2 bound to PPARc and inhibits PPARc expression. 15 SREBP1 is a transcription factor, which binds to a sequence in the promoter of different genes, called sterol regulatory elements-1. SREBP1c regulates the expression of genes or enzymes for glucose metabolism and lipid production. 16 SREBP1c is an upstream molecule of fatty acid synthase. 17 As reported in other articles, p53 overexpression suppressed the transactivation of SREBP-1 promoter and the expression of SREBP-1 downstream genes. 18,19 In addition, adipokines, including leptin, adiponectin, tumour necrosis factor a, resistin and interleukin-6, also play a role in regulation of lipolysis. NR4A1, also known as Nur77, TR3 or NGFI-B, is a unique transcriptional activator belonging to the orphan nuclear receptor subfamily. NR4A1 expresses in many cell types and mediates diverse biological processes. 20,21 Recent studies showed that NR4A1 together with two other family members (NR4A2 and NR4A3) played important roles in maintaining cellular energy homeostasis. 22 It was reported that mice with genetic deletion of Nur77 exhibited increased susceptibility to diet-induced obesity and insulin resistance. 23 Some studies showed that NR4A1 had an inhibitory effect on obesity. [23][24][25] However, the underlying mechanisms remain elusive.
In this study, we investigated the role of NR4A1 in adipogenesis and explored the possible mechanisms.

| Animal study
Both NR4A1 KO mice (generated from C57BL/6J mice) and their comparable wild-type mice (WT mice) were purchased from the Jackson Laboratory (ME, USA) and were fed ad libitum and maintained in pathogen-free (SPF) condition. The genotypes of the mice were verified with PCR, in which the primers applied were designed and suggested by the vendor to exhibit the difference between WT and KO (Table 1) 26 Epididymal fat was isolated for haematoxylin and eosin staining (H&E staining). We also detected the mRNA and protein levels of NR4A1 in adipose tissues from both WT and KO mice.

| Cell culture
The 3T3-L1 pre-adipocytes were cultured with high glucose (4.5 g/L glucose) DMEM plus 10% bovine serum (from Gibco), 100 IU/mL penicillin and 100 lg/mL streptomycin at 37°C in a humidified atmosphere with 5% CO 2 . When the pre-adipocytes reached 100% conference, the medium was replaced with differentiation medium containing 10% foetal bovine serum (from Gibco), 1 lmol/L dexamethasone (from Sigma), 0.5 mmol/L 3-isobutyl-1-methylxanthie (IBMX) (from Sigma) and 10 lg/mL insulin (from Sigma). The day when the inducer was added was considered as day 0 (D0) of differentiation. After two days (D2), the differentiation medium was replaced with fresh medium (high glucose DMEM with serum and antibiotics as described in the beginning of this paragraph) plus 10 lg/mL insulin. After that, the cells were cultured with medium without any inducer and medium was changed every other day (D4, D6, D8). HeLa cells applied to dual-luciferase reporter gene assay were cultured in high glucose DMEM with 10% foetal bovine serum (from Gibco), 100 IU/mL penicillin and 100 lg/mL streptomycin.

| Lentiviral infection and stable cell line selection
Both the Lentivirus encoding full-length NR4A1 and the control lentivirus were generated by GenePharma as previously described. 27 Highlights • NR4A1 regulates adipocyte differentiation or maturation.
3T3-L1 pre-adipocytes were infected with recombinant NR4A1 lentiviral stocks or control lentiviral stocks, and stable cell clones were selected under puromycin drug pressure. NR4A1 overexpression cell lines were designated as OV, and control cell lines were designated as NC.

| Antibodies and western blotting
The primary antibodies used to detect mouse proteins were

| Quantitative real-time PCR (qPCR)
Total RNA preparation and quantitative real-time PCR were conducted as described. 28 The relative mRNA expression levels were normalized to 18S rRNA. The primer sequences for real-time PCR are listed in Table 1.

| Plasmids construction
Mice genomic DNA was obtained from C57BL/6J liver using an Ezup column genomic DNA extraction kit (Sangon Biotech, Shanghai, China) according to the manufacture's instruction. The promoters of PPARc (À2000 bp), FAS (À2100 bp) and GATA2 (À2100 bp) were amplified from the genomic DNA with PCR. 27,28 All the PCR products were sequenced by Sangon Biotech to make T A B L E 1 Primer sequences for PCR or qPCR sure the promoter sequences obtained were correct by aligning with the sequences from NCBI (GenBank), shown in Table 2. The primers used for amplifying the different promoters are listed in Table 1.

| Statistical analysis
Statistical data were expressed as mean AE SE. Comparisons were performed by one-way analysis of variance (ANOVA), followed by the Newman-Keuls test using SPSS 17.0 software package. A value of P < .05 was considered statistically significant.

| NR4A1 knockout mice are confirmed with PCR and Western blotting
The genotypes of mice were confirmed by ordinary PCR, qPCR and WT mice exhibited a target band about 180 bp that was amplified from a pair of primers named as oIMR2060 and oIMR6603 (Figure 1A). qPCR results showed that WT mice had higher mRNA level of NR4A1 ( Figure 1B), Western blotting results showed that KO mice had less NR4A1 protein signal compared to WT mice ( Figure 1C). volume, we confirmed that KO mice fed with high-fat diet were more likely to gain weight ( Figure 2C and D). GTT and ITT were conducted to detect the insulin sensitivity in different mice. GTT results showed that KO mice had prolonged and higher blood glucose levels after glucose injection when compared with WT mice ( Figure 2E). ITT results showed that blood glucose levels of WT and KO mice were both reduced after insulin injection, but there was no significant difference in blood glucose levels between KO and WT ( Figure 2F).

| PPARc and FAS expression is increased in adipose tissues from KO mice
We collected epididymal fat tissues to analyse the expression of obesity-related genes implicated in fat synthesis and degradation.
The mRNA levels of lipolytic genes, including HSL, ATGL and LPL in the fat tissues from KO mice showed no significant differences compared to those of control (WT) mice ( Figure 3A). In KO mice, the expression levels of C/EBPb, adiponectin, FABP4 in adipose tissues were similar to those in WT mice ( Figure 3B and C). We also detected the expression of C/EBPa, another marker of adipogenesis, in adipose tissues of WT and KO mice, and found no significant difference (data not shown). However, the protein and mRNA expression of lipid synthesis-related genes, PPARc and FAS, were increased in adipose tissues of KO mice ( Figure 3D, E and F).

| NR4A1 inhibits the adipogenesis of 3T3-L1
pre-adipoctes showed that NR4A1 had an inhibitory effect on PPARc transcription, but had no effect on FAS ( Figure 5B). ChIP-qPCR results showed that NR4A1 had no direct physical association with PPARc or FAS ( Figure 5C).
However, NR4A1 as a transcriptional activator inhibited the adipogenesis of pre-adipocytes, we speculated that NR4A1 might activate the expression of some specific inhibitory factors, which in turn inhibited the expression of essential factors for adipogenesis. In some F I G U R E 5 NR4A1 regulating the transcription of adipogenesisrelated genes. A, Putative NR4A1 binding sites (a putative NBRE) in promoters of PPARc, FAS and GATA2. B, Dual-luciferase reporter gene assay for PPARc, FAS and GATA2 promoters. C, ChIP-qPCR was exploited to analyse the physical association between NR4A1 and the promoter region of PPARc, FAS or GATA2. The data show the means of three independent experiments, *P < .05, **P < .01, ***P < .001 previous studies, 29 GATA2 was reported as a transcription factor that inhibited PPARc transcription by binding to its promoter. We found that NR4A1 was able to enhance the transactivation of GATA2 promoter ( Figure 5B). Furthermore, we found that NR4A1 was able to bind to the promoter of GATA2 by means of ChIP-qPCR test. Then we studied the NR4A1 binding site on GATA2 promoter and figured out that NR4A1 bound to one putative binding sequence between À1125 bp to À1120 bp (GATA2-1) on the GATA2 promoter but did not bind to another putative binding sequence between À936 bp to À931 bp (GATA2-2) ( Figure 5C). GATA2 expression was decreased in adipose tissues from NR4A1 KO mice ( Figure 6A and B) and increased in OV cells during adipogenesis in vitro ( Figure 6C).
As NR4A1 showed no direct effect on FAS, so we explored the upstream genes of FAS, especially, SREBP1c. Our data demonstrated that the expression of SREBP1c was also reduced in OV cells ( Figure 7A). Then we examined the expression of some inhibitory proteins for SREBP1, including AMRK and ATF6, but none of them was positively correlated with NR4A1 expression (data not shown).
As p53 was also reported as an inhibitory protein for SREBP1c expression, 30 we therefore checked the expression of p53. Our data showed that the expression of p53 at both protein ( Figure 7B) and mRNA levels ( Figure 7C) in adipose tissues of KO mice was reduced.

| DISCUSSION
In this study, we further confirmed that NR4A1 knockout mice were more likely to gain weight. 18 Moreover, we found that NR4A1 overexpression inhibited adipogenesis in 3T3-L1 pre-adipocytes. In vivo, it has been reported that NR4A1 had the effects on glucose and The data show the means of three independent experiments, *P < .05, **P < .01, ***P < .001 lipid metabolism in liver 31,32 and muscles. 30,33 Studies have shown that NR4A1 knockout mice fed with high-fat diet were more likely to develop obesity compared to WT mice. 23 In consistency with previous study, our results also showed that NR4A1 KO mice fed with high-fat diet were inclined to develop obesity compared to WT mice.
We examined the morphological changes of adipocytes and the confirmed that NR4A1 had physical association with GATA2 promoter by exploiting the ChIP-qPCR techniques, but no physical association with PPARc promoter. There are two putative NR4A1 binding sites in GATA2 promoter (from À2100 bp to 0 bp). We further narrowed down the effective binding site (from À1125 bp to À1120 bp) rather than another one (from À936 bp to À931 bp). We proposed that NR4A1 might inhibit adipogenesis through up-regulation of GATA2, which in turn could inhibit PPARc expression.
It was also reported that SREBP1c positively regulated the transcription and expression of lipogenic enzymes such as fatty acid synthase. It was a key transcriptional regulator of triglyceride synthesis. 35 In this study, overexpression of NR4A1 in 3T3-L1 cells resulted in reduced expression of SREBP1c. We searched some related research articles and found that p53 was reported to have an inhibitory effect on SREBP1c expression. 18,19 Our data exhibited that NR4A1 overexpression suppressed the expression of the SREBP-1c and its downstream FAS. While knockout NR4A1 resulted in reduced p53 expression, therefore, NR4A1 might indirectly modulate the expression of SREBP-1c via p53, then hinder excess fat accumulation in adipocytes. 18,19 It has been reported that NR4A1 could either enhance DNA-dependent protein kinase to increase p53 transcription activity 36 or to cause mouse 3T3 cell double minute 2 (MDM2) to separation from p53, thus to protect p53 from MDM2mediated ubiquitination and degradation. 37 Herein we confirmed the presence of p53 in adipose tissue and found that p53 expression was reduced in adipose tissues of KO mice compared to WT. Based on these findings, we proposed that NR4A1 might indirectly reduce the expression of FAS by up-regulating p53 expression and subsequently down-regulating SREBP1c expression. Therefore, we proposed that NR4A1 might regulate adipogenesis and bodyweight via two possible pathways as shown in Figure 7D.
In summary, this study demonstrated that NR4A1 inhibits adipogenesis or adipocyte maturation. NR4A1 directly up-regulates GATA2 expression, which reduces the transcriptional expression of PPARc; and NR4A1 indirectly down-regulates SREBP1c and subsequently down-regulates FAS through p53.

ACKNOWLEDG EMENT
We are thankful to Dr. Sumei Lu, Dr. Huichen Zhao and Dr. Yuanmei Wang for their kind help and suggestions in the experiments.

CONFLI CT OF INTEREST
The authors confirm that there is no conflict of interests.