Isoflavonoids from Crotalaria albida Inhibit Adipocyte Differentiation and Lipid Accumulation in 3T3-L1 Cells via Suppression of PPAR-γ Pathway

Two 2″-isopropenyl dihydrofuran isoflavonoids (1 and 3), one 2″-isopropenyl dihydrofuran chromone (2), as well as 13 known compounds were isolated from the herbs of Crotalaria albida. Their structures and relative configurations were elucidated via NMR and HRESIMS analyses. The 2″ S absolute configuration of 1 and 2 were deduced by comparing their NOESY spectra with that of 3, which was determined via single crystal X-ray diffraction (CuKα). The 3R absolute configuration of 1 was determined by CD. Compounds 1, 2, and 3 inhibit the adipocyte differentiation and lipid accumulation of 3T3-L1 through down-regulation of PPAR-γ activity.


Introduction
Crotalaria albida Heyne ex Roth is a member of the subfamily Papilionoideae within Fabaceae and is a shrubby perennial herbs mainly distributed in the Southeast provinces of China [1]. In our previous research, we reported two pyrrolizidine alkaloids [2] isolated from this plant.
The isoflavonoids are an important subclass of the flavonoids and are mostly distributed in the subfamily Papilionoideae of the Fabaceae [3]. The isoflavonoids are also called phytoestrogens [4] and are cardioprotective [5], anticarcinogenic [6], antidiabetic and hypolipidemic [7]. They also prevent bone loss [8] and arteriosclerosis [9].
Peroxisome proliferators-activated receptors (PPARs) belong to one of the nuclear receptor superfamilies and control expression of a set of genes that regulate lipid and glucose metabolism [10,11]. Of the PPARs, peroxisome proliferator-activated receptor gamma (PPAR-γ) is a ligand-activated transcription regulator of adipocyte differentiation. It has been a molecular target for combating obesity and diabetes for decades [12,13].
A natural products search found that the PPAR-γ antagonist is an important path in new drug discovery and is involved in type 2 diabetes, obesity and other metabolic diseases [14,15]. Currently, berberine [15], tanshinone IIA [16], mycophenolic acid [17] and some germacranolide compounds [18] that present PPAR-γ antagonism effects have been shown to inhibit adipocyte differentiation and lipid accumulation in 3T3-L1 cells, reduce fat mass and weight, improve the glucose tolerance, and ameliorate glucose and lipid metabolism in the blood and liver. Thus, they are being considered as potential drugs for the treatment of obesity and diabetes.
As a part of an ongoing research program for the discovery of natural PPAR-γ antagonists from C. albida [19], this paper focuses on the isolation, structural determination and the evaluation of PPAR-γ antagonist activity of isoflavonoids.
Rosiglitazone was purchased from Sigma-Aldrich Chemical Co. Fetal bovine serum (FBS) and Dulbecco's modified Eagle Medium (DMEM) were purchased from Life Technology Co. All compounds were dissolved in dimethysulfoxide (DMSO).

Plant material
C. albida were collected, identified and deposited similar to our previous reports [2].

Ethics
No specific permissions were required for the described field studies. The locations are neither privately owned nor protected by the Chinese government. No endangered or protected species were sampled.

X-ray crystallographic analysis[2]
The single-crystal X-ray diffraction data were collected with a Bruker APEX-II CCD area detector diffractometer employing graphite-monochromated CuKα radiation (λ = 1.54178 Å) at 133(2) K. Data collection and integration were performed using a Bruker APEX2 and Bruker SAINT.
Crotadihydrofuran C X-ray crystallography data included:

Proliferation assay
Cell viability was investigated with the MTT [3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide] assay. Briefly, 1 × 10 4 cells were seeded into 96-well plates and cultured in DMEM supplemented with 10% FBS and then incubated with compound 1, 2, and 3 (0, 12.5, 25, 50, 100 μM) for 24 or 48 h. After treatment, MTT was added to each well and incubated at 37 C. After 4 h, the medium was removed and then blue formazan crystal was dissolved in 100μL DMSO. Absorbance values were collected at 570 nm using a fluorescent plate reader. The data were presented as percent cell viability versus control group. In each experiment, proliferation was determined in six replicate wells, and the overall experiment was repeated at least three times.

Oil red O staining
Eleven days after the induction of differentiation, the cells were washed with phosphate-buffered saline (PBS) twice, fixed in fresh 10% formalin for 10 min at room temperature, and stained with Oil Red O (Sigma, St. Louis, MO) at 60°C for 30 min. The cells were washed once with PBS and then photographed [15].
Transfection of cultured cells and luciferase assays [15,21] HEK 293T cells were inoculated into a 48-well culture plate at 5 × 10 4 cells/well and incubated in 5% CO 2 at 37°C with DMEM and 10% FBS overnight. The expression plasmid pCMX-Gal-mPPAR-γ-LBD, the Gal4 reporter vector MH100×4-TK-Luc and Renilla-Luc were gifts from Dr. C. Huang. The reporter assay was conducted as described previously [21]. The transfection mixture contained total plasmids, and FuGENE HD (Roche, Germany) and was added to 293T cells overnight and the removed. The solution was then changed to fresh media containing PPAR-γ agonist rosiglitazone or the compounds of interest at the previously determined concentration. All measurements were performed according to the protocol of the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA). Luciferase activity was measured and displayed as values ± SE, which was corrected for transfection efficiency using the renilla luciferase activity. All transfection experiments were achieved in triplicate and repeated three times independently [21].
Quantitative real-time PCR [20] Total RNA from 3T3-L1 cells was extracted with a spin column (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The first-strand cDNA was synthesized from 3μg of total RNA using a cDNA kit (Fermentas, Madison, WI, USA). The RNA expression levels were then quantified with a quantitative real-time RT-PCR using SYBR Green PCR Master Mix (Applied Biosystems, USA) and ABI Step One Plus Real Time PCR system (Applied Biosystems, USA) [21]. The forward and reverse primer sequences used in the RT-PCR were given in S1 Table. The results were calculated relative to β-actin.

Statistical analysis
The data are the mean ± SE. Significant differences between means were evaluated via one-way analysis of variance (ANOVA) and a two-tailed unpaired Student's test and results were considered significant when p < 0.05.

Results and Discussion
Crotadihydrofuran A (1) was obtained as a yellow oil. Its molecular formula was determined as C 20 [23] that was substituted at position 2@. The location of the 2@-isopropenyl dihydrofuran unit on ring B was determined based on the HMBC (S2 Table; S4 and S5 Figs) correlations from δ H 2.90 and 3.25 (Ha-1@ and Hb-1@) to δ C 153.3 (C-2 0 ), 114.8 (C-3 0 ), and 163.2 (C-4 0 ). Moreover, the R configuration of C-3 was determined based on its circular dichroism (CD) spectrum (S6 Fig), which gave a positive effect at 334 nm [22,24]. In the NOESY spectrum (S2 Table; (Table 1; S11 and S12 Figs) showed an α, βunsaturated carbonyl carbon resonance at δc 176.8 (C-4), an olefinic singlet proton signal at δ H 8.14 (1H, s, H-2) with a corresponding carbon atom signal at δ C 145.8 (C-2), as well as a typical ABX aromatic proton system at δ  (Table 1; S11 and S12 Figs). The 1 H and 13 C NMR data of 2 also closely resembled those of 3. The only significant difference in their 13 C NMR spectra were of the carbon signals at δc 115.1 (C-3) and 113.8 (C-1 0 ) in 3 shifted to δc 144.7 (C-3) and 140.3 (C-1 0 ) in 2. These were due to an electron-withdrawing effect of the oxygen atom at C-3. Comparing their molecular formula and weight gives further evidence for this assumption 2 has one more oxygen atom than 3. Thus, ring B was connected to ring C by an oxo-bridge. The configuration of H-2@ was the same α-orientation as 1 via similar NOESY correlations (S2 Table; S15 and S16 Figs). On the basis of this evidence, the structure of crotadihydrofuran B (2) was determined as 7-hydroxy-3-(2 0 -hydroxy-2@α-isopropenyl-dihydrobenzofuran-1 0 -oxy)-chromen-4-one (Fig 1).
We assayed whether compounds (1-16) isolated from C. albida could modulate transactivities of PPAR-γ on the Dual-Luciferase Reporter Assay System, because several studies have reported that regulation of PPAR-γ is key to the biological effects of flavones. The results showed that 1, 2, and 3 (crotadihydrofuran A, B, C) were potent suppressors of PPAR-γ ( Table 2). Compounds 1, 2, and 3 are undescribed compounds, and the strengths of inhibition for transactivity of PPAR-γ are the strongest among all compounds. Therefore, we focused on 1, 2, and 3.
The results indicated that the luciferase activity increased 4-fold when the PPAR-γ agonist rosiglitazone was used. Compounds 1, 2, and 3 also significantly inhibited the rosiglitazone-  stimulated PPAR-γ transactivity in a dose-dependent manner (Fig 3A). At 50 μM, transactivity of PPAR-γ induced by rosiglitazone was inhibited by 1 up to 64%, 2 up to 80%, and 3 up to 85%. These data suggest that 1, 2, and 3 inhibit PPAR-γ transactivity and may be an antagonist of PPAR-γ. Based on their inhibition of PPAR-γ transactivity, we postulated that 1, 2, and 3 might suppress adipocyte differentiation via direct targeting of PPAR-γ activity. To test this hypothesis, we used a 3T3-L1 adipocyte differentiation model [20]. Rosiglitazone could strongly promote 3T3-L1 adipocyte differentiation in the presence of insulin and dexamethasone in the culture medium. During the induction, the compounds (10, 25 and 50 μM) were added to the medium at day 0, and we observed their effects on 3T3-L1 adipocyte differentiation at day 10 [15]. As shown in Fig 3B, 1 inhibited the most adipocyte differentiation at doses of 50 μM and the effect of inhibition weakened as concentration decreased. Similar effects were seen in 3T3-L1 adipocytes treated with 2 and 3 (Fig 3C and 3D). The cell proliferation assay demonstrated that the suppression of adipocyte differentiation by isolated compounds was not due to cytotoxicity because compound treatment did not influence the cell viability and proliferation (S27 Fig). These findings suggest that three novel isoflavonoids inhibit the 3T3-L1 differentiation and adipogenesis by potentially suppressing PPAR-γ activity.