Anti-Obesity Effects of Petasites japonicus (Meowi) Ethanol Extract on RAW 264.7 Macrophages and 3T3-L1 Adipocytes and Its Characterization of Polyphenolic Compounds

Koreans have been consuming Petasites Japonicus (PJ) as food. Although the therapeutic effect of PJ on allergic or inflammatory reactions associated with asthma has been proven, its effect on obesity is unclear. Therefore, the present study was aimed to assess the obesity related anti-inflammatory and anti-adipogenic effects of ethanol extract PJ (EPJ) on the inflammatory response in RAW 264.7 macrophages and on differentiation in 3T3-L1 adipocytes. In addition, the polyphenolic compound was quantitatively characterized from the EPJ using ultra performance liquid chromatography coupled with diode array detector, quadrupole time-of-flight-mass spectrometry (UPLC-DAD-QToF-MS). In RAW 264.7 or 3T3-L1, reduction of nitric oxide (in macrophages) production as well as monocyte chemoattractant protein-1 and tumor necrosis factor-α were observed. Treatment of EPJ in adipocyte differentiation showed an improvement in adiponectin and lipid accumulation and a significant reduction of PPARγ and FABP-4 mRNA expression levels. On the other hand, mRNA expression of UCP-1, PPARα, and ACO increased in the EPJ treated group. In addition, a total of 26 polyphenolic compounds were detected and of which 12 are reported for the first time from PJ. The higher content of diverse polyphenolic compounds presented in EPJ might be responsible for the observed anti-inflammatory and anti-adipogenic effect. These results suggest that PJ is valuable in improving obesity-related inflammatory responses.


Cell Culture and Viability
The RAW 264.7 and 3T3-L1 cells were purchased from the Korea cell line bank (Seoul, Korea). RAW 264.7 macrophages were cultured in DMEM supplemented with 10% FBS, 100 U/mL of penicillin and 100 µg/mL of streptomycin at 37 • C in a humidified 5% CO 2 atmosphere.
3T3-L1 adipocyte differentiation, as an obese model, was carried out as previously described [24][25][26]. Once cells reached confluence, the cells were incubated in the differentiation medium containing EPJ or troglitazone (positive control reagent) with MDI (methyl-isobutylxanthine 0.1 mM, dexamethasone 0.4 µM, insulin 10 µg/mL) medium for 72 h. On day 3 (D-0), medium was changed to insulin (5 µg/mL) with EPJ or troglitazone medium. The medium was renewed every 2 days for 10 days [22]. Measurement of cell viability was performed by MTT assay. RAW 264.7 macrophages and 3T3-L1adipocytes were seeded at approximately 1 × 10 5 cells/200 µL/well and 2 × 10 3 cells/200 µL/well, respectively, in 96 well plates and incubated for 2 h. The sample was dissolved in DMSO to a concentration of 2 g/mL and diluted in culture medium before use. The final concentration of DMSO for all experiments remained constant at 0.1%. After incubating for 18 h, 0.5 mg/mL MTT solution was added in a concentration of 20 µL to each well and incubated for another 2 h. After MTT removal, the colored formazan was dissolved in 200 µL of DMSO. The color intensity was measured at 550 nm using a microplate reader (Softmax pro; Molecular Devices, Sunnyvale, CA, USA).

Measurement of Nitric Oxide
The RAW 264.7 macrophages were treated with diluted EPJ in Opti-mem (Gibco-Invitrogen; Cat: 31985) at different concentrations (0.01 and 0.1 mg/mL). At the same time, lipopolysaccharides (LPS) was added at a concentration of 100 ng/mL to make the obesity model. NO concentration in the culture medium was determined using the Griess reagent. The optical density (O.D.) was determined at 570 nm with a microplate reader (Softmax pro). NO values were corrected based on the cell viability.

Oil-Red O Staining
Oil-red O staining was performed as described previously [27]. Adherent cells were rinsed with phosphate buffer saline (PBS) and fixed with 10% formalin (neutral buffered) for 2 h, followed by rinsing with PBS. After rinsing, staining with Oil-red O was performed for 3 h. Plates were rinsed with distilled water (DW) and images of cells on the plates were taken. For quantification, the dye was extracted by DMSO, and absorbance was determined at 510 nm (Softmax pro).

Measurement of Chemokine and Adipocytokines
Cytokines released from RAW 264.7 and 3T3-L1 were measured in their culture supernatants using the sandwich ELISA method. Analysis was performed with mouse TNFα (Cat: 555179), MCP-1(Cat: 555212) from BD Pharmingen (San Diego, CA, USA) and adiponectin (Cat: DY1119) from R&D system kits (Minneapolis, MN, USA). All data were analyzed as recommended by the manufacturer.

Extraction of Phenolic Acids and Flavonoids
Crude extract was prepared by extracting 100 g of leaves and stalk part of PJ powder using 80% ethanol by soxhlet extraction method for 24 h. Then, the crude extract solvent was removed by rota evaporator and freeze-dried for complete drying of EPJ. Phenolic acid and flavonoid purification from the crude extract were conducted according to the method described by Kim et al. (2012) with minimum modification [29]. A total of 100 mg of EPJ powder was mixed with 10 mL of extraction solvent (methanol: water: formic acid = 80:15:5, v/v/v) containing 250 mg/L of 2,4,5-trimethoxycinnamic acid as the internal standard for phenolic acid analysis. The mixture was vortexed, stirred with an orbital shaker for 30 min at 200 rpm and centrifuged (LABOGENE 1580R, Korea) for 15 min at 3000 rpm, 10 • C. The supernatant was filtered using a syringe filter (0.45 µm, PVDF, Whatman, Kent, England). A total of 0.5 mL of the filtrate was diluted with water to 5 mL of final volume. The crude phenolic extract was then isolated by solid phase extraction method using Sep-Pak C18 cartridge (Waters Co., Milford, MA, USA). The cartridge was activated by washing with 2 mL of methanol, followed by 2 mL of water for conditioning. Then the diluted phenolic extract was loaded into the cartridge, and impurities were removed by washing with 2 mL of water. Finally, the crude phenolic acid was eluted from the cartridge using 3 mL of methanol. The purified extract was concentrated using N 2 gas, and then re-dissolved with 0.2 mL of the extract solvents without internal standard prior to instrumental analysis. For flavonoids analysis, 100 mg of crude ethanol extract was taken and extracted with the solvent system (methanol: water: formic acid = 50:45:5, v/v/v) containing 20 µg/mL of galangin as an internal standard. The extraction and purification steps were followed as described above. All experimental analyses were performed in triplicates.

Statistical Analysis
Data were expressed as mean ± standard deviation of four independent determinations. The statistical significance of any difference was determined by Student's t-test. The SPSS 12.0 software package (SPSS, Chicago, IL, USA) was used for statistical analysis. A significant value was defined as * p < 0.05, ** p < 0.01, and *** p < 0.001.

EPJ Inhibits NO Production in LPS-Stimulated Macrophages
Cell viability and NO production of RAW 264.7 macrophages are presented in Figure 1. NO values were corrected for cell viability. Cell viability increased significantly in 0.01 and 0.1 mg/mL of EPJ groups compared with the C group (LPS treated only). NO was significantly reduced in the PC group compared with the C group (p < 0.001, 38%). NO values were not statistically different in the EPJ 0.01 mg/mL group but were significantly reduced in the EPJ 0.1 mg/mL group compared with the C group (p < 0.001, 59%).

EPJ Inhibits NO Production in LPS-Stimulated Macrophages
Cell viability and NO production of RAW 264.7 macrophages are presented in Figure 1. NO values were corrected for cell viability. Cell viability increased significantly in 0.01 and 0.1 mg/mL of EPJ groups compared with the C group (LPS treated only). NO was significantly reduced in the PC group compared with the C group (p < 0.001, 38%). NO values were not statistically different in the EPJ 0.01 mg/mL group but were significantly reduced in the EPJ 0.1 mg/mL group compared with the C group (p < 0.001, 59%).

EPJ Inhibits MCP-1 and TNFα in LPS-Stimulated Macrophages
Protein secretion and mRNA expression of MCP-1 and TNFα are presented in Figure 2A-D. MCP-1 and TNFα secretion levels were similar to the NO secretion patterns. Moreover, the PC and EPJ 0.1 mg/mL groups were significantly decreased compared to the C group (p < 0.001, Figure 2A,B). The mRNA expression of MCP-1 was significantly decreased in the PC group compared to the C group ( Figure 2C, p < 0.01). Expression in the EPJ 0.01 mg/mL group was significantly increased. However, there were no significant differences in the EPJ 0.1 mg/mL group ( Figure 2C). The mRNA expression of TNFα was significantly decreased in the EPJ 0.1 mg/mL group compared with the C group ( Figure 2D, p < 0.01). There were no differences in the PC and EPJ 0.01 mg/mL groups compared with the C group. However, the protein production and mRNA expression of MCP-1 and TNFα showed generally similar patterns. Cytokines secretion (MCP-1 and TNFα) as inflammatory markers were found to be improved in the EPJ 0.1mg/mL group than the EPJ 0.01mg/mL group.

EPJ Inhibits MCP-1 and TNFα in LPS-Stimulated Macrophages
Protein secretion and mRNA expression of MCP-1 and TNFα are presented in Figure 2A-D. MCP-1 and TNFα secretion levels were similar to the NO secretion patterns. Moreover, the PC and EPJ 0.1 mg/mL groups were significantly decreased compared to the C group (p < 0.001, Figure 2A,B). The mRNA expression of MCP-1 was significantly decreased in the PC group compared to the C group ( Figure 2C, p < 0.01). Expression in the EPJ 0.01 mg/mL group was significantly increased. However, there were no significant differences in the EPJ 0.1 mg/mL group ( Figure 2C). The mRNA expression of TNFα was significantly decreased in the EPJ 0.1 mg/mL group compared with the C group ( Figure 2D, p < 0.01). There were no differences in the PC and EPJ 0.01 mg/mL groups compared with the C group. However, the protein production and mRNA expression of MCP-1 and TNFα showed generally similar patterns. Cytokines secretion (MCP-1 and TNFα) as inflammatory markers were found to be improved in the EPJ 0.1mg/mL group than the EPJ 0.01mg/mL group.

EPJ Improves MCP-1 and Adiponectin in Differentiated 3T3-L1 Adipocytes
Protein and mRNA expression of MCP-1 and adiponectin in differentiated 3T3-L1 adipocytes following EPJ treatment is shown in Figure 3. MCP-1 protein levels were significantly decreased in the PC, EPJ 0.01, and 0.02 mg/mL groups compared with the C group (differentiated cells without PC or EPJ, p < 0.001, Figure 3A). Especially, EPJ 0.02 mg/mL showed a reduction effect similar to the PC group. MCP-1 protein levels were decreased by 64% in the PC and EPJ 0.02 mg/mL group compared with the C group. Also, the mRNA expression levels of MCP-1 were significantly decreased in the PC (98%), EPJ 0.01 mg/mL (94%), and 0.02 mg/mL (95%) groups compared with the C group (p < 0.001, Figure 3B).

EPJ Improves MCP-1 and Adiponectin in Differentiated 3T3-L1 Adipocytes
Protein and mRNA expression of MCP-1 and adiponectin in differentiated 3T3-L1 adipocytes following EPJ treatment is shown in Figure 3. MCP-1 protein levels were significantly decreased in the PC, EPJ 0.01, and 0.02 mg/mL groups compared with the C group (differentiated cells without PC or EPJ, p < 0.001, Figure 3A). Especially, EPJ 0.02 mg/mL showed a reduction effect similar to the PC group. MCP-1 protein levels were decreased by 64% in the PC and EPJ 0.02 mg/mL group compared with the C group. Also, the mRNA expression levels of MCP-1 were significantly decreased in the PC (98%), EPJ 0.01 mg/mL (94%), and 0.02 mg/mL (95%) groups compared with the C group (p < 0.001, Figure 3B).

Anti-Lipogenic Effect of EPJ in Differentiated 3T3-L1 Adipocytes
In Figure 5, C is differentiated adipocytes without any treated and the other three groups are treated with PC or EPJ to differentiated adipocytes. The mRNA expression level of adipogenic gene PPARγ was significantly increased in the PC and EPJ 0.01 mg/mL groups (p < 0.05, Figure 5A). The PPARγ target gene; FABP-4, was significantly increased in the PC (p < 0.001) and EPJ 0.01 mg/mL groups compared to the C (p < 0.01). Conversely, PPARγ and FABP-4 mRNA expression in the EPJ 0.02 mg/mL group was not significantly different from the C group. In the EPJ 0.02 mg/mL group, UCP-1 (p = 0.059), PPARα (p = 0.067), and ACO (p = 0.062) were increased compared with the C group. However, UCP-2, UCP-3 and CPT-1 were not statistically different by both EPJ groups. In conclusion, EPJ not only inhibited the expression of genes related to adipocyte differentiation and lipid

Anti-Lipogenic Effect of EPJ in Differentiated 3T3-L1 Adipocytes
In Figure 5, C is differentiated adipocytes without any treated and the other three groups are treated with PC or EPJ to differentiated adipocytes. The mRNA expression level of adipogenic gene PPARγ was significantly increased in the PC and EPJ 0.01 mg/mL groups (p < 0.05, Figure 5A). The PPARγ target gene; FABP-4, was significantly increased in the PC (p < 0.001) and EPJ 0.01 mg/mL groups compared to the C (p < 0.01). Conversely, PPARγ and FABP-4 mRNA expression in the EPJ 0.02 mg/mL group was not significantly different from the C group. In the EPJ 0.02 mg/mL group, UCP-1 (p = 0.059), PPARα (p = 0.067), and ACO (p = 0.062) were increased compared with the C group. However, UCP-2, UCP-3 and CPT-1 were not statistically different by both EPJ groups. In conclusion, EPJ not only inhibited the expression of genes related to adipocyte differentiation and lipid accumulation but also increased expression of genes associated with thermogenesis. The EPJ 0.02 mg/mL group showed relatively potent activity compared to the EPJ 0.01 mg/mL group.

Characterization of Phenolic Acids
EPJ was characterized by UPLC-DAD-QToF-MS for polyphenolic compounds. Compound identification was done using a UV-absorbance pattern, mass spectral data under (+) ESI mode, literature data, and by comparing with an authentic standard, wherever available. A total of fourteen phenolic acids and twelve flavonols, which were identified from EPJ, are presented in Table 1 and  The three treatment groups (PC and both of EPJ groups) were compared with the control (C) group. Data are mean ± S.D. of four independent determinations. * p < 0.05, ** p < 0.01, *** p < 0.001.

Discussion
In this study, the cytokine release of macrophages and adipocytes were evaluated in vitro to investigate the inhibitory effects of EPJ on obesity-related inflammatory responses. The study also evaluated its anti-adipogenic effect of EPJ by measuring the inhibition of adipocyte differentiation and lipid accumulation related mRNA expression level. The results showed that EPJ reduced NO secretion ( Figure 1) in addition to the protein and mRNA expression levels of MCP-1 and TNFα in LPS-induced RAW 264.7 macrophages ( Figure 2) in a dose dependent manner. In the differentiated 3T3-L1 adipocytes, the EPJ showed adipocytokine improvements such as MCP-1 and adiponectin ( Figure 3) and anti-lipogenic effect (Figure 4).
The LPS-stimulated NO production in RAW 264.7 was reduced in the EPJ 0.1 mg/mL group without cell toxicity (Figure 1). NO is biosynthesized endogenously by various NOS enzymes [36]. Previous studies have shown that anti-allergic and anti-inflammatory effects of Bakkenolide B isolated from PJ leaves were found to inhibit the production of LPS-induced nitric oxide synthase (NOS) and cyclooxygenase-2 (COX-2) protein within peritoneal macrophage cells [22]. Despite not a single compound (i.e., Bakkenolide B), the results of this study showed the inhibitory effect of EPJ on NO production.
In the inflammatory condition, TNFα induces chemokines such as interleukin-8 (IL-8) and MCP-1 to mediate the initial inflammatory responses [37]. A study on EPJ reported that it reduces elevated levels of cytokines, including TNFα in the Jurkat T-cell line and decreased interleukin-6 (IL-6) in THP-1 cells (a human monocytic cell line) [23]. The current result revealed that MCP-1 and TNFα protein levels were significantly decreased in the PC and EPJ 0.1 mg/mL groups for LPS-treated RAW 264.7 macrophages, and also, the mRNA level was decreased by EPJ ( Figure 2).
As a PPARγ ligand, TZDs are known for adipocyte differentiation and have an anti-inflammatory effect [38]. Figure 5 shows the mRNA expression level of several genes involved in lipid accumulation and degradation. PPARγ is a key transcriptional factor of adipocyte differentiation, also involved in the treatment of type 2 diabetes and the improvement of metabolic syndrome. Interestingly, EPJ activated PPARγ like the TZD effect but inhibited differentiation in 3T3-L1 cells, as shown in Figure 4. Figure 4 shows the effect of EPJ on lipid accumulation in differentiated adipocytes without cell cytotoxicity, showing the possibility of the prevention of obesity due to the decrease in adipocyte accumulation of PJ. This result is attributed to the combination of EPJ, the PPARα and the activation of heat-related factor UCP-1.
This result suggests that EPJ increased not only the expression of the PPARγ and target genes (i.e., MCP-1, TNFα, Adiponectin and UCP-1) but also the PPARα activation, which affects energy metabolism in white adipose tissue (WAT) and its target genes [39]. Furthermore, activation of PPARγ influenced adiponectin and UCP-1 increase in adipocytes as well as NO, MCP-1, and TNFα reduction in macrophages or adipocytes. However, further investigation is necessary as to whether PJ acts as an agonist of PPARα and γ through direct ligand assay.
A previous study showed that treatment with EPJ (flower buds) 3T3-L1 adipocytes significantly decreased the expression of the adipogenic transcription factors, such as sterol regulatory element-binding proteins-1c (SREBP-1c), PPARγ, and CCAAT-enhancer-binding proteins (C/EBPα) [40]. On the other hand, our results showed that the PPARγ and its target gene FABP-4 were significantly increased in the PC group treated with TZD and in the EPJ 0.01 mg/mL treated group. However, compared with the PC group, FABP-4 was significantly decreased in the two groups treated with EPJ ( Figure 5A).
UCPs (UCP-1, -2, and -3), have been identified as mediators of thermogenesis, a group of mitochondrial proton transporters responsible for uncoupling respiration from oxidative phosphorylation [41]. According to a previous study, UCP-1 expression in beige adipocytes was increased by beta adrenergic receptor activation, and 3T3-L1 cells can differentiate into beige-like adipocytes with prolonged treatment of T3, IBMX, and rosiglitazone [42]. TZD treatment of db/db mice also induced UCP-1 and UCP-3 expression in brown adipose tissue [43]. In Figure 5A, PPARγ mRNA expression levels were increased in the PC and EPJ 0.01 mg/mL groups, indicating that EPJ also has PPARγ activation effects similar to TZD as a PPARγ agonist. The results indicated the increase of UCP-1 mRNA levels in the PC and EPJ 0.02 mg/mL groups by PPARγ activation ( Figure 5B). In differentiated 3T3-L1 cells treated with EPJ 0.02 mg/mL, PPARα, ACO, and CPT-1 genes increased compared with the C group ( Figure 5C). The target genes of PPARα at the cellular level are involved in the catabolism of lipids [44], and ACO and CPT-1 have been extensively linked to fatty acid oxidation [45].
The available evidence revealed that dietary polyphenols have many health benefits ranging from infectious disease to chronic metabolic syndromes and cancers. The chemical constituents presented in PJ have shown anti-allergic properties [46], anti-inflammatory [35], radical scavenging activities [35], and macrophage NO production inhibiting activities [47]. These activities are related with fukinolic acid, fukiic acid, and other hydroxycinnamoylquinic acid derivatives identified from the methanol extract of PJ [48]. The current biological activities observed can be justified strongly due to the presence of higher concentration of flavonoids and phenolic acids in the ethanol extract of PJ.
In conclusion, EPJ treatment decreased the inflammatory response and lipid accumulation but increased anti-inflammatory and thermogenesis response in macrophages and adipocytes. These results suggest that the consumption of PJ can contribute to the prevention and improvement of obesity and related inflammatory conditions. Further research is required about PJ and their mechanisms in obesity and the related inflammation. We are conducting animal experiments to reproduce the positive results shown in vitro and to determine the mechanism of PJ. The results of this experiment are expected to further support the health benefits of PJ as a Korean agrofood.