N,N′-diphenethylurea, a product of phenylalanine metabolism that has been reported to have an anti-depressant effect, was isolated from Streptomyces sp. AM-2498, but little is known about its effect on adipocyte differentiation.1

White adipose tissue is an important organ for energy storage. In addition, adipose tissue is an endocrine organ that secretes a number of hormones, adipocytokines, which regulates metabolic homeostasis. The formation of new adipocytes from precursor cells by differentiation is important for normal adipose tissue function. Indeed, impaired adipocyte differentiation may contribute to the pathogenesis of obesity-associated conditions including insulin resistance, hyperlipidemia and type 2 diabetes. 3T3-L1 cells are commonly used as an in vitro model system to study differentiation of adipocytes. The differentiation of pre-adipocytes into adipocytes is a complex process, involving coordinated activation of a cascade of transcription factors that regulate the expression of genes responsible for adipocyte function. Several key transcription factors have been reported to be involved in adipogenesis, peroxisome proliferator-activated receptor-γ (PPARγ) is the major transcription factor that is expressed in many tissues but predominantly found in adipose tissue and regulates expression of various genes involved in adipogenesis and lipogenesis.2 Transcriptional activity of PPARγ is regulated by binding to ligands such as thiazolidinediones.3 Thiazolidinediones, known as synthetic PPARγ ligands, stimulate adipocyte differentiation and enhance insulin sensitivity by stimulating transcriptional activity of PPARγ.3, 4

Many external hormones, such as insulin and insulin-like growth factor-1, also affect the process of differentiation in 3T3-L1 pre-adipocytes. Insulin receptor substrate phosphorylated by the insulin receptor activates the Ras/extracellular signal-regulated kinase (ERK) 1/2 and phosphoinositide-3-kinase/Akt pathways. The Ras/ERK 1/2 and phosphoinositide-3-kinase/Akt pathways have been shown to be responsible for adipocyte differentiation.5, 6, 7, 8, 9

By screening for compounds that promote adipocyte differentiation with several marine natural products from soft coral, marine sponge, sea cucumber, feather-stars, acidian, brown algae, red algae and green algae, we identified a small molecule, N,N′-diphenethylurea (Figure 1a), isolated from the Okinawan ascidian Didemnum molle, that promotes adipocyte differentiation.

Figure 1
figure 1

The effect of N,N′-diphenethylurea on adipocyte differentiation (a) The chemical structure of N,N′-diphenethylurea. (b) PPARγ ligand-binding activity of N,N′-diphenethylurea. The binding affinity of N,N′-diphenethylurea toward the PPARγ-LBD at the indicated concentration was measured using the Lanthascreen TR-FRET Peroxisome Proliferator Activated Receptor-γ Co-activator Assay (left) and the PolarScreen PPAR Competitor Assay (right). All values are presented as means±s.d. (n=3). The data shown are representative of two independent experiments. *P<0.05, **P<0.005, ***P<0.001 versus the control group. R, 1 μM rosiglitazone (c) The effect of N,N′-diphenethylurea on adipocyte differentiation in 3T3-L1 cells and C3H10T1/2 cells. All values are presented as means±s.d. (n=3). *P<0.05, **P<0.005, ***P<0.001 versus the insulin-only group. C, vehicle; I, insulin 1 μg ml−1; N10, 10 μM N,N′-diphenethylurea; N30, 30 μM N,N′-diphenethylurea; N100, 100 μM N,N′-diphenethylurea; R, 1 μM rosiglitazone. All experiments were carried out independently in triplicate. (d) and (e) The effect of N,N′-diphenethylurea on the expression of adipocyte-related genes (d) and protein (e) in 3T3-L1 cells.

Full size image

Therefore, in the present study, we investigated the mechanism of N,N′-diphenethylurea on enhancing adipocyte differentiation.

To investigate the effect of N,N′-diphenethylurea on adipocyte differentiation, 3T3-L1 cells, pre-adipocytes and C3H10T1/2 cells, and pluripotent stem cells, were treated with insulin in the presence of N,N′-diphenethylurea. As shown in Figure 1c, treatment of 3T3-L1 cells (left) and C3H10T1/2 cells (right) with N,N′-diphenethylurea significantly enhanced adipocyte differentiation in a dose-dependent manner. The OD value of the Oil Red O eluted solution increased by 2.6-fold and 2.5-fold at a concentration of 100 μM N,N′-diphenethylurea in 3T3-L1 and C3H10T1/2 cells, respectively. Rosiglitazone (1 μM), a synthetic PPARγ ligand, also significantly promoted adipocyte differentiation in both cell types (P<0.001 in 3T3-L1 cells, P<0.005 in C3H10T1/2 cells).

To further characterize the effect of N,N′-diphenethylurea on adipocyte differentiation, we examined the expression levels of adipogenic genes at day 4 and day 9 during adipocyte differentiation in 3T3-L1 cells. As shown in Figure 1d, the expression of PPARγ2 was significantly increased in cells treated with N,N′-diphenethylurea compared with insulin-only-treated cells. In addition, N,N′-diphenethylurea-treated cells exhibited significantly upregulated mRNA expression of adipogenic marker genes, including aP2, adiponectin and FAS during adipocyte differentiation. At the same time, protein expression of aP2 and PPARγ was also increased and protein expression of β-catenin, which has a central role as a transcriptional co-activator in the Wnt/β-catenin signaling pathway that inhibits adipocyte differentiation,10 was decreased in N,N′-diphenethylurea-treated cells (Figure 1e). These results suggest that N,N′-diphenethylurea promotes adipocyte differentiation by regulating PPARγ expression in 3T3-L1 cells.

The activation of PPARγ by its ligands is a key process in adipocyte differentiation and some natural products have been reported to act as PPARγ ligands.2, 3 Therefore, as N,N′-diphenethylurea enhanced adipogenesis, we examined whether N,N′-diphenethylurea served as a PPARγ ligand. In the PPARγ-binding assay, as shown in Figure 1b (left), N,N′-diphenethylurea exhibited PPARγ affinity at concentrations of 300 μM (1.3-fold, P<0.001). In addition, we examined the binding of N,N′-diphenethylurea to the PPARγ ligand-binding domain (PPARγ-LBD)using a PPARγ competitor binding assay. As shown in Figure 1b (right), treatment with N,N′-diphenethylurea bound to the receptor with an IC50 of 80 μM.

Insulin signaling involved in ERK 1/2 and phosphoinositide-3-kinase/Akt is considered to have a critical role in the process of adipocyte differentiation.11 We thus examined the effect of N,N′-diphenethylurea on insulin signal pathways in 3T3-L1 pre-adipocytes. As shown in Figure 2, insulin treatment led to a dramatic increase in the phosphorylation level not only of ERK but also Akt. However, 100 μM of N,N′-diphenethylurea treatment with insulin showed no change in phospho-ERK 1/2 level but an increase in the phospho-Akt level without apparent alteration of the total Akt protein level.

Figure 2
figure 2

The effect of N,N′-diphenethylurea on the insulin signals in 3T3-L1 cells. Post-confluent 3T3-L1 cells were subjected to serum starvation and incubated with 100 μM N,N′-diphenethylurea for 3 h before treatment with insulin at a concentration of 10 μg ml−1 for 15 min. Total cell extracts were prepared and subjected to western blot analysis for phosphor-ERK, ERK, phospho-Akt, Akt and GAPDH. N, 100 μM N,N′-diphenethylurea; R, 1 μM rosiglitazone. All experiments were carried out independently in triplicate.

Full size image

Dysregulation of adipocyte differentiation has been suggested as one of the many causes of type 2 diabetes. Several previous studies showed that adipocytes in type 2 diabetes patients are insulin-resistant and are not capable of accumulating lipids to their full capacity.12 PPARγ is a master regulator of adipocyte differentiation. Together with C/EBPα, PPARγ promotes terminal differentiation via transactivation of downstream adipocyte-specific genes expression, which is involved in their downstream effect and regulates insulin signaling, glucose and lipid metabolism in mature adipocytes. Therefore, activation of PPARγ in adipose tissue improves insulin resistance and reduces lipotoxicity in muscle and liver.6, 13, 14 Thiazolidinediones, known as PPARγ ligands, which have been reported to increase insulin sensitivity in type 2 diabetes, activate PPARγ and promote adipocyte differentiation and induce apoptosis in large adipocytes.2, 3, 12

In this study, we found that N,N′-diphenethylurea promoted adipocyte differentiation and activated PPARγ as a weak ligand and an insulin signal, possibly via the phosphoinositide-3-kinase/Akt signal pathway, in 3T3-L1 cells. These findings provide a new insight into the mechanism and possibility of N,N′-diphenethylurea to improve insulin sensitivity in type 2 diabetes.

Experimental procedure

Materials

Rosiglitazone was purchased from Alexis Biochemicals (San Diego, CA, USA). Insulin and Oil Red O were purchased from Sigma-Aldrich (St Louis, MO, USA). Polyclonal antibodies against aP2, PPARγ, β-catenin, ERK 1/2, phopho-ERK 1/2, Akt, phospho-Akt and GAPDH were purchased from Cell Signaling Technology (Beverly, MA, USA).

Extraction and isolation of N,N′-diphenethylurea from D. molle

Okinawan ascidian D. molle (132 g) collected in the Okinawa Prefecture was extracted with methanol (500 ml) for 7 days. The extract was filtered, concentrated and partitioned between ethyl acetate and water. The ethyl acetate-soluble material was further partitioned between aqueous methanol and hexane. The material obtained from the aqueous methanol portion (704.6 mg) was subjected to fractionation using ODS silica gel (Cosmosil 75C18-OPN, 40% aqueous methanol to methanol) and a crystalline mixture (224.0 mg) obtained from the 80% aqueous methanol portion was resolved by fractional crystallization into N,N′-diphenethylurea (110.0 mg). Its structural identity was verified by comparison with published spectral data (Figure 1a).9

Cell culture and adipocyte differentiation

Mouse 3T3-L1 cells and C3H10T1/2 cells, obtained from the RIKEN Bio Resource Center (Tsukuba, Japan), were grown in Dulbecco's modified Eagle's medium (Gibco BRL, Grand Island, NY, USA) supplemented with 10% bovine calf serum (Gibco BRL), antibiotics (100 U ml−1 penicillin and 100 μg ml−1 streptomycin (Gibco BRL) at 37 °C under a humidified 5% CO2 atmosphere. 3T3-L1 cells and C3H10T1/2 cells grown in Dulbecco's modified Eagle's medium supplemented with 10% bovine calf serum (day 0) were treated with insulin (1 μg ml−1) with/without N,N′-diphenethylurea in 10% fetal bovine serum/Dulbecco's modified Eagle's medium at various concentrations for 9 days. Fresh medium containing insulin (1 μg ml−1) and 10% fetal bovine serum with/without N,N′-diphenethylurea was replenished every 3 days.

Oil red O staining

After differentiation, the cells were fixed with 10% formalin in phosphate-buffered saline, and then stained with Oil Red O (0.5% in 60% isopropanol). Cells were photographed under a microscope and lipid and Oil Red O were extracted using isopropanol. Absorbance was measured using a spectrophotometer at a wavelength of 520 nm.

RNA preparation and reverse transcriptase-mediated PCR

Total RNA was isolated from 3T3-L1 cells using Isogen reagent (Nippon Gene, Tokyo, Japan) and 1 μg of total RNA from each sample was used as a template for each reverse transcriptase-mediated PCR using an ImProm-II Reverse Transcription System (Promega, Fitchburg, WI, USA) and Taq polymerase (Takara, Tokyo, Japan). Primer sequences used in PCR were as follows; aP2, 5′-CAACCTGTGTGATGCCTTTGTG-3′ and 5′-CTCTTCCTTTGGCTCATGCC-3′; β-actin, 5′-TGTTACCAACTGGGACGACA-3′ and 5′-CTCTCAGCTGTGGTGGTGAA-3′; PPARγ2, 5′-GCTGTTATGGGTGAAACTCTG-3′ and 5′-ATAAGGTGGAGATGCAGGTTC-3′; adiponectin, 5′-GTTGCAAGCTCTCCTGTTCC-3′ and 5′-CTTGCCAGTGCTGTTGTCAT-3′, FAS, 5′-CTTCGCCAACTCTACCATGG-3′ and 5′-TTCCACACCCATGAGCGAGT-3′.

PPARγ ligand-binding assay

For PPARγ ligand-binding activity measurement, we used two commercial kits, the Lanthascreen TR-FRET PPARγ Coactivator Assay, and the Polar Screen PPAR Competitor Assay (Invitrogen, Camarillo, CA, USA). LanthaScreen TR-FRET-based nuclear receptor co-activator recruitment assay used a terbium-labeled anti-glutathione-S-transferase antibody, a fluorescein-labeled co-activator peptide and a PPARγ-LBD tagged with glutathione-S-transferase. Terbium-anti-glutathione-S-transferase antibody indirectly labels the PPARγ-LBD by binding to the glutathione-S-transferase tag. Binding of the agonist to PPARγ-LBD causes a conformational change, which results in an increase in the affinity of the PPARγ for a co-activator peptide. The close proximity of the fluorescently labeled co-activator peptide to the terbium-labeled antibody causes an increase in the TR-FRET signal. The TR-FRET ratio of 520/495 was calculated by measurement with an EnVision multi-label reader (Perkin-Elmer, Waltham, MA, USA) with an excitation wavelength of 340 nm and emission wavelengths of 520 and 495 nm. The PolarScreen PPAR Competitor Assay was based on the purified recombinant human PPARγ-LBD and a selective fluorescent PPARγ ligand. The complex between the ligand and the ligand-binding domain exhibits high fluorescence polarization, which is lost on ligand displacement by non-labeled competitors. Millipolarization values for different competitors were determined with the SpectraMax-M5 Microplate Reader (Molecular Devices, Sunnyvale, CA, USA) with an excitation wavelength of 485 nm and emission wavelength of 535 nm.

Western blot analysis

3T3-L1 cells were washed three times with ice-cold phosphate-buffered saline and solubilized with lysis buffer (Sigma) and then centrifuged at 12000 rpm for 15 min at 4 °C. The supernatant was used for immunoblot analysis. In all, 30 μg proteins was separated and transferred onto PVDF membranes, followed by immunoblot analysis with respective antibodies against aP2, PPARγ, β-catenin, ERK 1/2, phospho-ERK 1/2, Akt, phospho-Akt and GAPDH. Immune complexes were detected with horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence (ECL; Amersham, Buckinghamshire, UK).

Statistical analysis

All qualitative data are representative of at least three independent experiments. Quantitative data are presented as means±s.d. and were compared with ANOVA, using Origin 6.0 software (Microcal Software, Northampton, MA, USA) and Graphpad prism 5 (Graphpad software, San Diego, CA, USA). A P<0.05 was considered statistically significant.