Pteryxin: a Coumarin in Peucedanum Japonicum Thunb Leaves Exerts Antiobesity Activity Through Modulation of Adipogenic Gene Network

Nutrition 30 (2014) 1177–1184

Contents lists available at ScienceDirect

Nutrition

journal homepage: www.nutritionjrnl.com

Basic nutritional investigation Pteryxin: A coumarin in Peucedanum japonicum Thunb leaves exerts antiobesity activity through modulation of adipogenic gene network

Ruwani N. Nugara M.Sc. a,b, Masashi Inafuku Ph.D. b, Kensaku Takara Ph.D. c, Hironori Iwasaki Ph.D. b, Hirosuke Oku Ph.D. b,* a United Graduate School of Agricultural Sciences, Kagoshima University, Kagoshima, Japan b Center of Molecular Biosciences, Tropical Biosphere Research Center, University of the Ryukyus, Okinawa, Japan c Department of Bioscience and Biotechnology, Faculty of Agriculture, University of the Ryukyus, Okinawa, Japan article info abstract

Article history: Objectives: Partially purified hexane phase (HP) of Peucedanum japonicum Thunb (PJT) was iden- Received 24 October 2013 tified as an antiobesity candidate. However, the compound responsible for the antiobesity activity Accepted 20 January 2014 remained unknown. Thus, in this study we isolated the active compound, to determine the mechanisms related to antiobesity activity in vitro. Keywords: Methods: The HP was fractionated, and the effect on the triacylglycerol (TG) content was evaluated Lipid metabolism in 3T3-L1 preadipocytes and HepG2 hepatocytes. On the basis of comprehensive spectroscopic Obesity analyses, the structure of the active compound was identified as pteryxin, a known compound in Peucedanum japonicum Thunb Pteryxin PJT. However, to our knowledge, its biological activities against obesity have not been reported 3T3-L1 adipocytes previously. The dose-dependent effect on the TG content, and the gene expressions related to HepG2 hepatocytes adipogenesis, fatty acid catabolism, energy expenditure, lipolysis, and lipogenesis due to pteryxin (10, 15, and 20 mg/mL) were examined in vitro. Results: Pteryxin dose dependently suppressed TG content in both 3T3-L1 adipocytes (by 52.7%, 53.8%, and 57.4%, respectively; P < 0.05) and HepG2 hepatocytes (by 25.2%, 34.1%, and 27.4%, respectively; P < 0.05). Sterol regulatory element-binding protein-1 (SREBP-1c), fatty acid synthase (FASN), and acetyl-coenzyme A carboxylase-1 (ACC1) were down-regulated in pteryxin-treated 3T3-L1 adipocytes (by 18%, 36.1%, and 38.2%, P < 0.05) and HepG2 hepatocytes (by 72.3%, 62.9%, and 38.8%, respectively; P < 0.05). The adipocyte size marker gene, paternally expressed gene1/ mesoderm specific transcript (MEST) was down-regulated (by 42.8%; P < 0.05), and hormone- sensitive lipase, a lipid catabolizing gene was up-regulated (by 15.1%; P < 0.05) in pteryxin- treated adipocytes. The uncoupling protein 2 (by 77.5%; P < 0.05) and adiponectin (by 76.3%; P > 0.05) were up-regulated due to pteryxin. Conclusion: Our study demonstrated that pteryxin in PJT plays the key role in regulating the lipid metabolism-related gene network and improving energy production in vitro. Thus, the results suggest pteryxin as a new natural compound to be used as an antiobesity drug in the pharmaceutical industry. Ó 2014 Elsevier Inc. All rights reserved.

Introduction

RNN and HO made substantial contributions to the conception and design of the study, performing the experiment, assembling, analyzing and interpreting Obesity and overweight are major health concerns worldwide the data, discussing the results, and drafting the manuscript. MI and HI partic- due to their role in numerous metabolic disorders such as dia- ipated in the experimental work and in collecting, assembling, analyzing, and betes, hypertension, and cardiovascular diseases [1–3]. Since interpreting the data. KT made substantial contribution in structure identifica- 2008, more than 10% of the world’s adult population tion, assembling, and analysing the data related to NMR. All authors read and was estimated to be obese [4]. The search for potent obesity approved the final manuscript. The authors have declared no conflict of interest. * Corresponding author. Tel.: þ81 98 895 8972; fax: þ81 98 895 8972. inhibitory compounds with fewer adverse effects attracts great E-mail address: [email protected] (H. Oku). attention.

Peucedanum japonicum Thunb (PJT), a subtropical plant Cell lines and reagents belonging to the family of Apiaceae, originates from southern Both 3T3-L1 and HepG2 cell lines were purchased from the American Type Japan, China, and Taiwan. The PJT plant is known as “Chomeiso” Culture Collection (Manassas, VA, USA). Dulbecco’smodified Eagle’s medium in Japanese denoting long life and its leaves are frequently used (DMEM), and fetal bovine serum (FBS) were purchased from Wako Pure as a leafy vegetable, garnish for raw fish, and as the main Chemical Industries Ltd. (Osaka, Japan). Bovine calf serum (BCS) was purchased ingredient in Chomeiso-noodle in the Okinawa Island, Japan. It from Life Technologies (Grand Island, NY, USA). Dexamethasone, 3-isobutyl-1- has been reported that some coumarins isolated from PJT were methylxanthine, and insulin were purchased from Sigma-Aldrich (St. Louis, MO, USA). found to have antiplatelet [5], antioxidant [6], and antagonistic effects [7]. Moreover, a recent study has shown its applicability as a medicinal herb for the treatment of rheumatic disorder [8]. Cell culture and treatments Our previous studies have reported on the antiobesity activity of PJT [9], and thus ethanol extract (EE) of PJT being characterized 3T3-L1 preadipocytes were grown in 10% BCS and avoided complete with antidiabetic activity [10]. In this context, we showed that confluence before initiating differentiation. For adipogenesis, preadipocytes were 4 fl the antiobesity activity in EE was partially purified into the cultured in 24-well plates at a density of 1 10 cells per well. The con uent fibroblasts were maintained for another 2 d (defined as day 0). Differentiation hexane phase (HP) demonstrating antiobesity activity in 3T3-L1 was induced with standard differentiation inducers 0.5 mM 3-isobutyl-1- adipocytes and HepG2 hepatocytes, with an improved energy methylxanthine, 0.25 mM dexamethasone, 10 mg/mL insulin, and 10% FBS for 48 h expenditure profile in C2C12 myotubes [11]. However, to our (from day 0–2). The culture medium was then changed to DMEM supplemented knowledge it has not been determined whether the active with 10% FBS, 10 mg/mL insulin, and HP (50 mg/mL), chlorogenic acid (CGA; 10 mg/ mL), or pteryxin (10, 15, or 20 mg/mL) from 2 to 6 d. The culture medium was compound in PJT is responsible for the antiobesity effect. replaced every 2 d. To investigate the time course of the effect of pteryxin on Thus, in the present study, we isolated the active compound adipogenesis, we triggered the differentiation in the presence of pteryxin (20 mg/ in PJT and identified it as pteryxin, which is a previously known mL) at different time intervals. Cells were harvested on day 6 for analyses. compound in PJT [5], and the molecular mechanisms related to HepG2 cells were maintained in DMEM supplemented with 100 U/mL penicillin, 100 mg/mL streptomycin, and 10% FBS in an atmosphere of 5% CO at pteryxin-dependent inhibition of lipid accumulation were eval- 2 37C. For experiments, cells were seeded in 24-well plates at a density of 5 uated in vitro. Pteryxin attenuated lipogenesis by direct modu- 104 cells per well, incubated in complete medium to 70% confluence, and lation of the lipogenic gene network and improved the lipid maintained in serum-free DMEM containing 1% BSA overnight, as described metabolism profile to suppress obesity. To the best of our elsewhere [12]. HepG2 cells were then subsequently treated with insulin knowledge, we are the first group to report on antiobesity (1 mM), HP (50 mg/mL), CGA (10 mg/mL), or pteryxin (10, 15, or 20 mg/mL) for another 24 h. properties of pteryxin. Cell viability was determined by using 3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium solution as previously described [13]. Materials and methods

Plant material, extraction and analysis Triacylglycerol assay The origin of PJT used in this study, ground, and extracted was as described elsewhere [9]. The yield of EE was w11% of the starting material. EE was further At the end of the treatment period, 3T3-L1 and HepG2 cells were washed suspended in water and partitioned into n-hexane (41: 1 v/v). The approximate with phosphate saline buffer (Wako) and harvested into 10% Triton X-100 solu- yield of the HP (3.3%) was subjected to preparative silica gel open-column tion and lysed by brief sonication. The TG content was quantified using a com- chromatography (CC). The starting eluent was hexane (100%), thereby changing mercial enzymatic kit (Wako) according to the manufacturer’s instructions. The to hexane:dichloromethane:methanol (90:9.5:0.5, v/v), followed by a gradual content of cellular protein was determined using the Quant-iT protein assay kit shift in the mixing ratio to 50:47.5:2.5 (v/v), then enriched with dichlor- (Life Technologies). The TG content was expressed as total TG (mg) per mg omethane:methanol (95:5, v/v), and finally by methanol:formic acid (99.5:0.5, v/ cellular protein. v) to give 11 fractions (Fr1–11). The fractions of HP (Fr1–11) were subjected to total triacylglycerol (TG) assay using 3T3-L1 preadipocytes and Fr3 showed the highest inhibition of the lipid content (data not shown). mRNA extraction and quantitative real-time reverse transcriptase PCR The Fr3 (1.3%) was further subjected to high-performance liquid chromatography (HPLC) on a silica gel column into 12 fractions (Fr3 1–12). The mobile The 3T3-L1 preadipocytes were differentiated into adipocytes in 3-cm dishes phases were hexane as the starting eluent and 0.5% formic acid in dichlor- at a density of 3 104 cells per dish, and HepG2 at 15 104 cells per dish. Total omethane:methanol (94.5:5.0, v/v) as the limiting eluent. The gradient program RNA was extracted using RNeasy mini kit (Qiagen, Hilden, Germany) according to of the limiting eluent is as follows: 0 to 5 min, 10%, and 5 to 30 min, 10% to 50%, 30 manufacturer’s recommendations. First-strand cDNA was generated from 2 mgof to 40 min, 50% to 10% with a flow rate of 2 mL/min at a wavelength of 322 nm. All total RNA as a template using high-capacity RNA-to-cDNA kit (Applied Bio- isolated fractions were evaporated, dissolved in 99.5% ethanol 1:0.1 (w/v) and systems, CA, USA). The quantitative real-time polymerase chain reaction (qPCR) stored at 20C until further use. The fractions were examined against lipid was performed on the Step One PlusTM Real-Time PCR System (Applied Bio- accumulation by TG assay in 3T3-L1 adipocytes. The purified active component in Ò systems) using SYBR GREEN master mix (Applied Biosystems) with the Fr3 (Fr3–8; 0.15%) was subjected to direct electron ionization-mass spectrometry following parameters: one cycle of 95C for 20 sec, 40 cycles of 95C for 3 sec, and (EI-MS), 1 H-nuclear magnetic resonance (NMR), 13 C-NMR, heteronuclear mul- 60C for 30 sec. A melting curve analysis was performed starting at 95C for 15 tiple quantum coherence (HMQC), and heteronuclear multiple bond correlation sec, 60C for 60 sec and increasing by 0.3C every 15 sec to determine primer (HMBC), for the identification of the chemical structure. specificity (specific primers are listed in Supplementary Table 1). The mRNA levels of all genes were normalized using b-actin, and human glyceraldehyde-3- phosphate dehydrogenase (hGAPDH), as the internal controls for 3T3-L1 and Instrumentation HepG2, respectively. The open-CC was performed on silica gel 40 mm60A (26 100 mm; Yamazen Corporation, Japan) column. HPLC was performed on a Shimadzu HPLC 10 A system on a silica gel column (10 250 mm, 5 SL-II; Cosmosil, Nacalai Tesque, Statistical analyses INC., Japan). EI-MS was performed on a gas chromatography mass spectrometer (GCMS-QP 2010; Shimadzu Kyoto, Japan) using a direct inlet system. NMR spectra All results were expressed as the mean SEM. The statistical significance of were measured on a Bruker AVANCE 400 (Bruker Biospin, Rheinstetten, Ger- the difference between the means of control and treatment groups was deter- many). 1H-NMR, 13C-NMR, HMQC, and HMBC were measured using a 5-mm mined by Dunnett’s test. The significance of the difference between the means of probe. The operating frequencies were 400.13 MHz for 1H-NMR and 100.62 two groups was determined by Student’s t test. Differences were considered 13 MHz for C-NMR spectra. Samples were measured at 299 K in CDCl3 with tet- significant at P < 0.05 using the online statistical software available at the ramethylsilane as standard. genome research center of Osaka University, Japan [14]. R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184 1179

Fig. 1. Inhibitory effects of the Fr-3 fractions on lipid accumulation in 3T3-L1 adipocytes. During the differentiation maintenance, 3T3-L1 preadipocytes were treated with 12 fractions at 16.7 mL fraction per well on days 2 and 4. On day 6, cells were harvested and triacylglycerol (TG) content was measured and expressed as total TG per cellular protein. The results represent as the mean SEM from three independent experiments. The asterisk (*) indicates a signiﬁcant difference between control and treatment groups tested by Dunnett’s test. *P < 0.05 versus control.

Results Effect of pteryxin during the course of adipogenesis

Purification of the antiobesity active compound To gain insight on the mechanism of the suppressive effect of pteryxin on adipogenesis, we examined the time course of the The strongest antiobesity activity was observed in Fr3 of the adipocyte differentiation in the presence of pteryxin (Fig. 3B). first purification step, silica gel open-CC. Thus, the Fr3 was The results indicated that when pteryxin-treatment was further fractionated into 12 fractions (Fr3 1–12) using the HPLC administered between day 0 and 2, the lipid accumulation was system, and the suppression of the lipid accumulation in the significantly blocked by 50.6% (P < 0.05; Fig. 3C). In the presence adipocytes due to the isolated fractions was evaluated (Fig. 1). of pteryxin throughout the adipocyte maintenance period (day The minimum TG content was observed in Fr3, reducing the lipid 0–6) did not further potentiate the inhibitory effect, indicating content by 57.8% when compared with the control in the adi- a similar inhibition pattern to that of day 0–2 treatment. pocytes (P < 0.05). Nevertheless, the first half of fractions of Fr3 Importantly, pteryxin exerted the ability of suppressing the lipid (Fr3 1–6), rather increased the lipid content in the adipocytes. accumulation at any given time point during the adipogenic The HP showed a 21.2% reduction in the TG content (P < 0.05). Of process. the fractions of Fr3, Fr3 8 showed the highest inhibition by 45.9% in the lipid content of 3T3-L1 adipocytes. The Fr3 7, Fr3 9 and Fr3 Effect of pteryxin on the gene modulation pattern in adipocytes 10 suppressed the lipid accumulation by 27.5%, 12.7%, and 9.9%, respectively (P < 0.05). In the present investigation, pteryxin at a concentration of 20 mg/mL showed a lower peroxisome proliferator-activated receptor gamma (PPARg) expression by 24.9% than the HP treat- fi Identi cation of the chemical structure ment, yet significantly higher than the control (P < 0.05; Fig. 4). The mRNA expression level of CCAAT/enhancer binding protein 13 ’’’ ’’’ The C-NMR (CDCl3, 100 MHz): d 15.6 (C-4 ), 20.4 (C-5 ), (C/EBPa) was significantly up-regulated by the HP (P < 0.05), and ’’ ’ ’ ’ ’ 20.8 (C-2 ), 22.2 (Me-2 ), 25.3 (Me-2 ), 60.2 (C-4 ), 70.5 (C-3 ), was not modulated by the pteryxin treatment. Furthermore, 77.3 (C-2’), 107.4 (C-8), 112.6 (C-4 a), 113.3 (C-3), 114.4 (C-6), 127.5 (C-2’’’), 129.2 (C-5), 137.9 (C-3’’’), 143.2 (C-4), 154.1 (C-8 a), 156.7 (C-7), 159.7 (C-2), 166.9 (C-1’’’), 169.9 (C-1’’). EI-MS analysis of the þ active compound found the [M–1] ion at m/z 385. Interpretation of these spectral data identified the chemical structure of the active component as pteryxin (Fig. 2) [5,15].

Pteryxin inhibits lipid accumulation in adipocytes

To investigate the concentration course of pteryxin on lipid accumulation, adipocytes were treated with HP, CGA, or different doses of pteryxin from day 2 to 6 (Fig. 3A). We observed a dose- dependent inhibition in the lipid accumulation by 52.7%, 53.8%, and 57.4% in the presence of 10, 15 and 20 mg/mL pteryxin, respectively (P < 0.05). The TG in CGA was comparable to that of the lowest dose of pteryxin. Under the conditions of our study, the treatments produced no detectable cell toxicity for 3T3-L1 cells. Fig. 2. Chemical structure of () pteryxin. 1180 R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184

increasing tendency in the expression of adiponectin, also known as AdipoQ [17] due to HP, and pteryxin treatments (P > 0.05). Moreover, the peroxisome proliferator-activated receptor g coactivator-1 (PGC1) a showed comparable values to that of the control, regardless of the treatment. The increased level of adiponectin expression prompted us to investigate the insulin sensitivity for pteryxin-treated adipocytes. Thus, glucose transporter 4 (GLUT4) and insulin receptor substrate-1 (IRS-1) expressions were examined. GLUT4 was suppressed by 43.1% (P < 0.05) and IRS-1 by 36.6% (P > 0.05) due to pteryxin.

Effect of pteryxin on HepG2 hepatocytes

To gain insight on the effect of pteryxin on hepatocytes, HepG2 cells were treated with different doses of pteryxin (Fig. 5A). Insulin and HP treatments indicated a nonsignificant 4.4 and 8.9% suppression of the TG content, respectively. Pter- yxin showed 25.2%, 34.1%, and 27.4% inhibition in TG content at 10, 15 and 20 mg/mL, respectively (P < 0.05). CGA values were comparable to those of the control group. No cytotoxicity was observed due to treatments in HepG2 cells. To further clarify the mechanism of pteryxin in the hepatocytes, we examined the gene modulation pattern in pteryxin- treated HepG2 cells. The human SREBP1 (hSREBP1) expression was suppressed by 72.3% due to pteryxin (P < 0.05; Fig. 5B). Insulin-treated hepatocytes showed comparable values to that of the control. HP and pteryxin significantly reduced human FASN (hFASN) expression, with the highest degree of attenuation by pteryxin. Moreover, human stearoyl-coenzyme A desaturase (hSCD), and human ACC1 (hACC1) were significantly suppressed by 44.5% and 50.3%, respectively due to pteryxin. A lipolytic gene, human PPARa (hPPARa) was significantly up-regulated by the Fig. 3. Suppression of lipid content by pteryxin in 3T3-L1 cells. The 3T3-L1 pre- pteryxin treatment [18]. adipocytes were treated with hexane phase (HP; 50 mg/mL), chlorogenic acid (CGA; 10 mg/mL), or pteryxin (10, 15, 20 mg/mL) from days 2 to 6. On day 6, cells Discussion were harvested and triacylglycerol (TG) content was measured and expressed as total TG per cellular protein (A). 3T3-L1 preadipocytes were induced to differentiation in the absence or the presence of 20 mg/mL of pteryxin during the indicated Coumarins are a group of plant-derived polyphenolic com- time intervals represented in the (B) scheme. Differentiated cells were evaluated for pounds found in many plants and natural food products such as the TG content (C). Values are mean standard errors of mean of three indepen- seeds, nuts, green tea, vegetables, and fruits. In the present study, dent experiments. The asterisk (*) indicates a significant difference between control we demonstrate, to our knowledge for the first time, that and treatment groups tested by Dunnett’s test. *P < 0.05 versus control. pteryxin, a known natural compound in PJT, exerts a lipid- lowering effect by attenuating lipid accumulation in both of pteryxin decreased paternally expressed gene 1/mesoderm 3T3-L1 and HepG2 cells. The suppressive effect of pteryxin on specific transcript (MEST) expression by 42.9%, indicating a HepG2 was milder compared with that on 3T3-L1 cells. Pteryxin reduction in the size of the fat droplets in the adipocytes. The has been previously identified as an antiplatelet aggregation master regulator of fatty acid synthesis, sterol regulatory constituent, however, its biological activities against obesity element-binding protein-1 c (SREBP-1c) [16] was significantly have not been reported [5]. Our previous in vitro study showed reduced in pteryxin treatment when compared with the control. that HP directly modulated lipid metabolism in adipocytes, he- Furthermore, pteryxin significantly suppressed the acetyl- patocytes, and myotubes to inhibit adiposity [11], prompting us coenzyme A caboxylase-1 (ACC1) and pyruvate dehydrogenase to further fractionate the HP to identify the bioactive compound kinase 4 (PDK4) expression levels (P < 0.05). A nonsignificant responsible for antiobesity. 36.1% suppression in FASN expression was observed in pteryxin- The preadipocyte differentiation represents a fundamental treated adipocytes. No significant difference was noted with the process that is accompanied by activation of a cascade of farnesoid X receptor alpha (FXRa) expression between treat- transcription factors leading to gene modulation [19]. Thus, we ments. RAR-related orphan receptor C (RORC) expression was up- investigated the effect of pteryxin on the expression levels of regulated significantly by HP and pteryxin. Lipoprotein lipase genes involved in the adipogenic process. Previous studies have (LPL) expression showed a nonsignificant change due to pter- shown the inhibition of adipogenesis in 3T3-L1 preadipocytes yxin. Hormone-sensitive lipase (HSL) and fatty acid-binding by down-regulation of PPARg, the master regulator in adipo- protein 4 (FABP 4) showed 15.1% and 202% increase, respec- genesis [20,21]. Recent studies have revealed that white adi- tively, in the presence of pteryxin. The uncoupling protein (UCP) pose tissue (WAT) depots, which can be readily converted to a 2 was up-regulated (by 140.1% and 77.5%, respectively, P < 0.05), “brown-like” state (beige cells) by enhancing UCP1 and PGC1a, whereas there were no significant alterations in UCP3 levels due and that PPARg is a prerequisite for the development of brown to the HP and pteryxin treatments. Additionally, there was an fat in WAT [22,23].Wehaveconfirmed the up-regulation in R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184 1181

Fig. 4. Effect of pteryxin on the adipogenic gene network. 3T3-L1 preadipocytes were differentiated and treated with 20 mg/mL pteryxin during days 2 to 6. The quantitative polymerase chain reaction was performed with values expressed as fold-change over control. Values are mean standard errors of mean, in three independent experiments. y Comparisons between treatments with the control were analyzed by Dunnett’s test (*P < 0.05), and a particular treatment versus control was analyzed by Student’s t test ( P yy < 0.05, P < 0.01). ACC1, acetyl-coenzyme A carboxylase; AdipoQ, adiponectin; C/EBP, CCAAT-enhancer-binding protein; FABP, fatty acid-binding protein; FASN, fatty acid synthase; FXR, farnesoid X receptor; GLUT, glucose transporter; HP, hexane phase; HSL, hormone-sensitive lipase; IRS, insulin receptor substrate; LPL, lipoprotein lipase; MEST, mesoderm speciﬁc transcript; PDK, pyruvate dehydrogenase kinase; PGC, proliferator-activated receptor g coactivator; PPARg, Peroxisome proliferator-activated receptor gamma; RORC, RAR-related orphan receptor; SREBP, sterol regulatory element-binding protein; UCP, uncoupled protein. 1182 R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184

Fig. 5. The effect of pteryxin on HepG2 hepatocytes. Cells were treated with insulin (1 mM), hexane phase (HP; 50 mg/mL), chlorogenic acid (CGA 10 mg/mL), or pteryxin (10, 15, 20 mg/mL) for 24 h in serum-free medium. triacylglycerol content was measured in each treatment (A) and expression of lipogenic genes were assessed by quantitative polymerase chain reaction with values expressed as fold-change over control (B). Data are represented as mean standard errors of mean, in three independent experiments. The asterisk (*) indicates a signiﬁcant difference between control and treatment groups tested by Dunnett’s test. *P < 0.05 versus control. hGAPDH, glyceraldehyde 3-phosphate dehydrogenase; hACC1, human acetyl-coenzyme A carboxylase; hFASN, human fatty acid synthase; hFXR, human farnesoid X receptor; HP, hexane phase; hPPARg, human Peroxisome proliferator-activated receptor gamma; hSCD, human stearoyl-CoA desaturase; hSREBP, human sterol regulatory element-binding protein.

PPARg expression in our previous studies due to PJT [9,10]. fatty acids in mature adipocytes and conﬁrming the suggested Concordantly, in our most recent investigation, we found that mechanism in our previous study [11,31]. partially puriﬁed HP treatment can accelerate the gene ex- The expression of lipogenic genes such as, ACC1/hACC1, and pressions of PPARg by 160% [11]. The present study showed that hSCD is transcriptionally regulated by SREBP-1c. Suppression of pteryxin enhanced the PPARg and AdipoQ expression levels as these lipogenic enzymes attenuates TG accumulation in both previously reported for ascofuranone [24] and suggests pter- pteryxin-treated adipocytes and hepatocytes [32,33]. A decrease yxin as a potential therapeutic agent for obesity [25,26]. in PDK4 might facilitate the activation of pyruvate dehydroge- Up-regulation of the HSL gene expression, which is the primary nase complex, which catalyzes the decarboxylation of pyruvate lipase responsible for hydrolysis of TG in the adipocytes sug- to acetyl-coenzyme A, thereby favoring the glucose metabolism gests accelerated lipolysis due to pteryxin [27]. It is noteworthy, in pteryxin-treated adipocytes [34,35]. Furthermore, pteryxin that pteryxin led to genesis of small adipocytes, by the accelerated the direct fatty acid b-oxidation via activation down-regulation of MEST, an adipocyte size marker gene [28], of hPPARa [36] in HepG2 cells. Taken together, these results un- which may further increase the lipolytic activity, and equivocally demonstrated that pteryxin play a crucial role in the decreasing TG storage in adipocytes [29]. In agreement, the gene modulation to attenuate the adiposity and lipogenesis elevated RORC expression explains the contribution of pteryxin in vitro. In addition to the adipogenesis-related gene expres- to reduce the TG in small-sized adipocytes via increased fatty sions, we investigated the effect of pteryxin on energy acid b-oxidation [30]. Increased FABP 4 expression due to metabolism. The up-regulated trend of UCP2 suggests the pteryxin, suggests an enhanced ability of cells to metabolize improvement of energy metabolism [37]. R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184 1183

GLUT4 is found in adipose tissues and responsible for insulin- [6] Hisamoto M, Kikuzaki H, Ohigashi H, Nakatani N. Antioxidant compounds regulated glucose translocation into the cell and regulated by from the leaves of Peucedanum japonicum thunb. J Agric Food Chem 2003;5118:5255–61. PPARg [38]. Our previous study has reported a significant [7] Takeuchi N, Kasama T, Aida Y, Oki J, Maruyama I, Watanabe K, et al. increase in GLUT4 expression in the muscle tissue by supple- Pharmacological activities of the prenylcoumarins, developed from folk menting the diet with EE [10]. However, our present results usage as a medicine of Peucedanum japonicum THUNB. Chem Pharm Bull (Japan) 1991;396:1415–21. demonstrated of a down-regulation of GLUT4 and IRS-1 in [8] Adams M, Berset C, Kessler M, Hamburger M. Medicinal herbs for the pteryxin-treated 3T3-L1 adipocytes. Further studies on the treatment of rheumatic disordersda survey of European herbals from the metabolic signaling pathways related to pteryxin may unveil its 16th and 17th century. J Ethnopharmacol 2009;1213:343–59. [9] Nukitrangsan N, Okabe T, Toda T, Inafuku M, Iwasaki H, Oku H. Effect of activity in suppressing adipogenesis. Increasing tendency of Peucedanum japonicum Thunb extract on high-fat diet-induced obesity FXRa has not been able to enhance the insulin signaling as and gene expression in mice. J Oleo Sci 2012;612:89–101. mentioned elsewhere [39]. Alternatively, the down-regulation of [10] Nukitrangsan N, Okabe T, Toda T, Inafuku M, Iwasaki H, Oku H. Anti-obesity activity of Peucedanum japonicum Thunb extract in obese diabetic animal GLUT4 may be due to hypoglycemic effects by activating PPARg, model C57BL/6J Ham Slc-ob/ob mice. LSMR 2012;22:28–34. thereby inhibiting the glucose incorporation similar to that of [11] Nugara RN, Inafuku M, Iwasaki H, Oku H. Partially purified Peucedanum curcumin, which poses both antidiabetic [40] and antiobesity japonicum Thunb extracts exert anti-obesity effects in vitro. Nutrition – effects [41]. Validation of these findings in vivo is required, as it 2014;30(5):575 83. [12] Zang M, Zuccollo A, Hou X, Nagata D, Walsh K, Herscovitz H, et al. would suggest that activation of PPARg could alter the blood AMP-activated protein kinase is required for the lipid-lowering effect of glucose levels. Furthermore, a detailed study on molecular metformin in insulin-resistant human HepG2 cells. J Biol Chem 2004; – pathways at protein level may provide important information on 27946:47898 905. [13] Zhang XH, Huang B, Choi SK, Seo JS. Anti-obesity effect of resveratrol- the specificity of pteryxin toward antiobesity. amplified grape skin extracts on 3T3-L1 adipocytes differentiation. Nutr During the process of purification to identify the active con- Res Pract 2012;64:286–93. stituent, we found that the activity of Fr3 was potent over the HP, [14] Genome Research Center. MEPHAS. Japan: Osaka University; 2002. Avail- – able at: http://www.gen-infoosaka-u.ac.jp. Accessed on July 2, 2013. suggesting an occurrence of compounds (Fr3 1 6) available in HP [15] Akihisa T, Tokuda H, Ukiya M, Iizuka M, Schneider S, Ogasawara K, et al. to counteract the antiobesity activity of Fr3. Furthermore, CGA, a Chalcones, coumarins, and flavanones from the exudate of Angelica keiskei main constituent in coffee [42], was previously identified in PJT and their chemopreventive effects. Cancer Lett 2003;2012:133–7. [16] Yang LH, Chen TM, Yu ST, Chen YH. Olanzapine induces SREBP-1-related [9]. Previous studies have reported the activity of CGA in stim- adipogenesis in 3T3-L1 cells. Pharmacol Res 2007;563:202–8. ulating the glucose uptake in 3T3-L1 adipocytes without [17] Havel PJ. Control of energy homeostasis and insulin action by adipocyte inducing significant adipogenesis [43] and have exhibited anti- hormones: leptin, acylation stimulating protein, and adiponectin. Curr Opin Lipidol 2002;131:51–9. obesity properties in high-fat diet-induced obese mice [44]. [18] Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete Several reports have emphasized that supplementation of a program of cholesterol and fatty acid synthesis in the liver. J Clin Invest high-fat diet with CGA leads to insulin resistance and hepatic 2002;1099:1125–31. lipid accumulation [45,46]. In agreement, CGA showed no TG [19] Fu M, Sun T, Bookout AL, Downes M, Yu RT, Evans RM, et al. A nuclear receptor atlas: 3T3-L1 adipogenesis. Mol Endocrinol 2005;1910:2437–50. suppression in hepatocytes in the present investigation. [20] Jin L, Fang W, Li B, Shi G, Li X, Yang Y, et al. Inhibitory effect of androgra- pholide in 3T3-L1 adipocytes differentiation through the PPARgamma pathway. Mol Cell Endocrinol 2012;3581:81–7. Conclusion [21] Lii CK, Huang CY, Chen HW, Chow MY, Lin YR, Huang CS, et al. Diallyl trisulfide suppresses the adipogenesis of 3T3-L1 preadipocytes through – – Our study clearly demonstrates, to our knowldege for the first ERK activation. Food Chem Toxicol 2012;50(3 4):478 84. [22] Spiegelman BM. Banting Lecture 2012: Regulation of adipogenesis: to- time, that pteryxin inhibits the transcription factors for lipid ward new therapeutics for metabolic disease. Diabetes 2013;626: synthesis in the differentiated adipocytes and in hepatocytes, 1774–82. playing a crucial role in regulating lipid metabolism-related gene [23] Fisher FM, Kleiner S, Douris N, Fox EC, Mepani RJ, Verdeguer F, et al. FGF21 regulates PGC-1alpha and browning of white adipose tissues in adaptive expressions. The results provide an insight on the activity of thermogenesis. Genes Dev 2012;263:271–81. pteryxin involved in suppressing obesity. Thus, our study sug- [24] Chang YC, Cho HJ. Ascofuranone stimulates expression of adiponectin gests pteryxin as a new natural compound to be used as an an- and peroxisome proliferator activated receptor through the modulation of mitogen activated protein kinase family members in 3T3-L1, murine tiobesity drug in the pharmaceutical industry. pre-adipocyte cell line. Biochem Biophys Res Commun 2012;4223: 423–8. [25] Ohara K, Uchida A, Nagasaka R, Ushio H, Ohshima T. The effects of Supplementary Data hydroxycinnamic acid derivatives on adiponectin secretion. Phytomedicine 2009;162–3:130–7. Supplementary data related to this article can be found at [26] Alappat L, Awad AB. Curcumin and obesity: evidence and mechanisms. Nutr Rev 2010;6812:729–38. http://dx.doi.org/10.1016/j.nut.2014.01.015. [27] Nishino N, Tamori Y, Kasuga M. Insulin efficiently stores triglycerides in adipocytes by inhibiting lipolysis and repressing PGC-1alpha induction. Kobe J Med Sci 2007;533:99–106. References [28] Takahashi M, Kamei Y, Ezaki O. Mest/Peg1 imprinted gene enlarges adipocytes and is a marker of adipocyte size. Am J Physiol Endocrinol Metab [1] Jeong YS, Hong JH, Cho KH, Jung HK. Grape skin extract reduces adipo- 2005;2881:E117–24. genesis- and lipogenesis-related gene expression in 3T3-L1 adipocytes [29] Willson TM, Cobb JE, Cowan DJ, Wiethe RW, Correa ID, Prakash SR, et al. through the peroxisome proliferator-activated receptor-gamma signaling The structure-activity relationship between peroxisome proliferator- pathway. Nutr Res 2012;327:514–21. activated receptor gamma agonism and the antihyperglycemic activity of [2] Tan B, Li X, Yin Y, Wu Z, Liu C, Tekwe CD, et al. Regulatory roles for thiazolidinediones. J Med Chem 1996;393:665–8. L-arginine in reducing white adipose tissue. Front Biosci 2012;17:2237–46. [30] Raichur S, Lau P, Staels B, Muscat GE. Retinoid-related orphan receptor [3] Yahagi T, Daikonya A, Kitanaka S. Flavonol acylglycosides from flower of gamma regulates several genes that control metabolism in skeletal muscle Albizia julibrissin and their inhibitory effects on lipid accumulation in 3T3- cells: links to modulation of reactive oxygen species production. J Mol L1 cells. Chem Pharm Bull (Tokyo) 2012;601:129–36. Endocrinol 2007;391:29–44. [4] World Health Organization. Obesity and overweight, 2013. Available [31] Shan T, Liu W, Kuang S. Fatty acid binding protein 4 expression marks a at: http://wwwwhoint/mediacentre/factsheets/fs311/en/. Accessed on population of adipocyte progenitors in white and brown adipose tissues. 26 December, 2013. FASEB J 2013;271:277–87. [5] Chen IS, Chang CT, Sheen WS, Teng CM, Tsai IL, Duh CY, et al. Coumarins [32] Abu-Elheiga L, Oh W, Kordari P, Wakil SJ. Acetyl-CoA carboxylase 2 mutant and antiplatelet aggregation constituents from Formosan Peucedanum mice are protected against obesity and diabetes induced by high-fat/high- japonicum. Phytochemistry 1996;412:525–30. carbohydrate diets. Proc Natl Acad Sci USA 2003;10018:10207–12. 1184 R. N. Nugara et al. / Nutrition 30 (2014) 1177–1184

[33] Flowers MT, Miyazaki M, Liu X, Ntambi JM. Probing the role of [40] Nishiyama T, Mae T, Kishida H, Tsukagawa M, Mimaki Y, Kuroda M, et al. stearoyl-CoA desaturase-1 in hepatic insulin resistance. J Clin Invest 2006; Curcuminoids and sesquiterpenoids in turmeric (Curcuma longa L.) sup- 1166:1478–81. press an increase in blood glucose level in type 2 diabetic KK-Ay mice. J [34] Holness MJ, Zariwala G, Walker CG, Sugden MC. Adipocyte pyruvate de- Agric Food Chem 2005;534:959–63. hydrogenase kinase 4 expression is associated with augmented PPAR- [41] Ejaz A, Wu D, Kwan P, Meydani M. Curcumin inhibits adipogenesis in 3T3- gamma upregulation in early-life programming of later obesity. FEBS Open L1 adipocytes and angiogenesis and obesity in C57/BL mice. J Nutr Bio 2012;2:32–6. 2009;1395:919–25. [35] Murase T, Misawa K, Minegishi Y, Aoki M, Ominami H, Suzuki Y, et al. Coffee [42] Shimoda H, Seki E, Aitani M. Inhibitory effect of green coffee bean extract polyphenols suppress diet-induced body fat accumulation by down- on fat accumulation and body weight gain in mice. BMC Complement regulating SREBP-1c and related molecules in C57BL/6J mice. Am J Physiol Altern Med 2006;6:9. Endocrinol Metab 2011;3001:E122–33. [43] Alonso-Castro AJ, Miranda-Torres AC, Gonzalez-Chavez MM, Salazar- [36] Bandyopadhyay GK, Vu CU, Gentile S, Lee H, Biswas N, Chi NW, et al. Olivo LA. Cecropia obtusifolia Bertol and its active compound, chlorogenic Catestatin (chromogranin A(352-372)) and novel effects on mobilization of acid, stimulate 2-NBDglucose uptake in both insulin-sensitive and insulin- fat from adipose tissue through regulation of adrenergic and leptin resistant 3T3 adipocytes. J Ethnopharmacol 2008;1203:458–64. signaling. J Biol Chem 2012;28727:23141–51. [44] Cho AS, Jeon SM, Kim MJ, Yeo J, Seo KI, Choi MS, et al. Chlorogenic acid [37] Schrauwen P, Hesselink M. UCP2 and UCP3 in muscle controlling body exhibits anti-obesity property and improves lipid metabolism in high-fat metabolism. J Exp Biol 2002;205(Pt 15):2275–85. diet-induced-obese mice. Food Chem Toxicol 2010;483:937–43. [38] Watson RT, Kanzaki M, Pessin JE. Regulated membrane trafﬁcking of the [45] Cerejeira J, Lagarto L, Mukaetova-Ladinska EB. Behavioral and psychologi- insulin-responsive glucose transporter 4 in adipocytes. Endocr Rev cal symptoms of dementia. Front Neurol 2012;3:73. 2004;252:177–204. [46] Frank J, Kamal-Eldin A, Razdan A, Lundh T, Vessby B. The dietary hydrox- [39] Cariou B, van Harmelen K, Duran-Sandoval D, van Dijk TH, Grefhorst A, ycinnamate caffeic acid and its conjugate chlorogenic acid increase vitamin Abdelkarim M, et al. The farnesoid X receptor modulates adiposity and e and cholesterol concentrations in Sprague-Dawley rats. J Agric Food peripheral insulin sensitivity in mice. J Biol Chem 2006;28116:11039–49. Chem 2003;519:2526–31.