An Ethanol Extract of Artemisia iwayomogi Activates PPARδ Leading to Activation of Fatty Acid Oxidation in Skeletal Muscle

Although Artemisia iwayomogi (AI) has been shown to improve the lipid metabolism, its mode of action is poorly understood. In this study, a 95% ethanol extract of AI (95EEAI) was identified as a potent ligand of peroxisome proliferator-activated receptorδ (PPARδ) using ligand binding analysis and cell-based reporter assay. In cultured primary human skeletal muscle cells, treatment of 95EEAI increased expression of two important PPARδ-regulated genes, carnitine palmitoyl-transferase-1 (CPT1) and pyruvate dehydrogenase kinase isozyme 4 (PDK4), and several genes acting in lipid efflux and energy expenditure. Furthermore, 95EEAI stimulated fatty acid oxidation in a PPARδ-dependent manner. High-fat diet-induced obese mice model further indicated that administration of 95EEAI attenuated diet-induced obesity through the activation of fatty acid oxidation in skeletal muscle. These results suggest that a 95% ethanol extract of AI may have a role as a new functional food material for the prevention and/or treatment of hyperlipidermia and obesity.


Introduction
Skeletal muscle is an important organ in the whole body regulation of energy homeostasis and the main site of fatty acid and glucose oxidation [1,2]. PPARd plays a critical role in skeletal muscle metabolism via transcriptional regulation of downstream gene expression [3]. The reported in vivo effects of PPARd activation include improvement of dylipidemia and hyperglycemia, prevention of diet-induced obesity, enhancement of insulin sensitivity and modulation of muscle fiber type switching as demonstrated by systemic ligand administration or by generation of transgenic mice that over-express an active PPARd [4][5][6][7]. Most of the observed beneficial effects are believed to be mediated by increasing fatty acid catabolism and mitochondrial function in muscle and adipocytes [8]. Thus, it is proposed that activators of PPARd may have therapeutic utility in the treatment of metabolic disease [9].
Artemisia herbs, a member of the Compositae, have long been used in foods and in traditional medicine for treatment of diseases, including diabetes and hepatitis [10]. Artemisia herbs have been reported to have anti-diabetic and anti-hyperlipidemic activities in diabetic patients and rats [11,12]. However, molecular mechanisms whereby Artemisia exerts its benefit on lipid and glucose metabolism remain unknown.
In this study, we screened medicinal herbs to search for natural PPARd ligands. We found that a 95% ethanol extract of Artemisia iwayomogi (95EEAI) directly interacted with PPARd, enhanced the expression of genes involved in lipid catabolism and induced PPARd-dependent activation of fatty acid oxidation. Furthermore, administration of 95EEAI to mice fed a high-fat diet enhanced fatty acid oxidation in the skeletal muscle and protected against diet-induced obesity.

Ethics statement
All animal experiments were approved by the AmorePacific Institutional Animal Care and Use Committee and adhere to the OECD guidelines. Permit numbers: AP11-101-FR012. No specific permits were required for the described field studies. No specific permissions were required for these locations/activities. We confirmed that the location was not privately-owned or protected in any way and the field studies did not involve endangered or protected species.

Preparation of an ethanol extract of Artemisia Iwayomogi
Three hundred grams of the aerial parts of Artemisia iwayomogi was heated to 80uC with 70% ethanol for 3 h. The extract was then filtered through Whatman No. 1 (Whatman, Piscataway, NJ, USA) and loaded on a D-101 macroporous resin column, followed by elution of the column with water (WEAI), 50% ethanol (50EEAI) and 95% ethanol eluate (95EEAI) were obtained. After evaporation, the solutions were freeze-dried.

PPARd coactivator assay
PPARd coactivator assay was performed using Lanthascreen TM time-resolved fluorescence resonance energy transfer (TR-FRET) PPARd coactivator assay kit according to the instructions of the manufacturer (Invitrogen, Carlsbad, CA, USA). All assays were validated for their robustness by determining the respective Z9-factors [13]. Measurements were performed on a VICTOR3_V Multilabel Counter (WALLAC 1420; PerkinElmer Life and Analytical Sciences, Rodgau, Germany) with instrument settings as described in the manufacturer's instructions for LanthaScreen TM assays.

Luciferase Reporter Assay
PPARd-responsive luciferase reporter assay was performed using Human Peroxisome Proliferator-Activated Receptor Delta Reporter Assay System (Indigo Biosciences, PA, USA). PPARd reporter cells (provided with assay system) are non-human mammalian cells stably transfected with human PPARd and PPARd-responsive luciferase reporter genes. Mock reporter cells which contain only the PPARd-responsive luciferase vector were also purchased from Indigo Biosciences. PPARd reporter cells and mock reporter cells were cultured in Cell Recovery Medium 1 (CRM-1) for 4 h. The cells were treated with indicated concentration of AI extracts for 24 h. Luciferase activity was measured using luciferase detection reagent (Indigo Biosciences, PA, USA) and Tecan infinite M200 Pro (Tecan, Grodig, Austria), following the manufacturer's recommended procedures. The protein concentration of the cell lysate was determined using the BCA protein assay kit (Pierce, Rockford, Illinois, USA). Luciferase activity was normalized to the protein concentration of each sample. The fold induction of normalized luciferase activity was calculated relative to DMSO (vehicle)-treated cells, and represents the mean of three independent samples per treatment group.

Small Interfering RNA Treatment
The PPARd small interfering RNA (siRNA) pool (a mixture of three siRNAs for PPARd, cat. no. sc-36305) and control siRNA (cat. no. sc-37007) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). For the transfection procedure, cells were grown to 70% confluence, and PPARd and control siRNAs were transfected using the Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Briefly, Lipofectamine 2000 reagent was incubated with serum-free medium for 10 min. Subsequently, a mixture of the respective siRNAs was added. After incubation for 15 min at room temperature, mixture was diluted with medium and added to each well. The final concentration of PPARd siRNA in each well was 100 nM. After culturing for 48 h, cells were washed and treated with GW501516 or AI extracts for an additional 24 h.

Western blot analysis
Myotubes were lysed in RIPA buffer (PBS, pH 7.4, containing 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS, and a protease inhibitor cocktail). Forty micrograms of proteins were resolved on 10% NuPAGE gels run in an MES buffer system (Invitrogen, Carlsbad, CA) and transferred to nitrocellulose membranes according to the manufacturer's protocol. Immunoreactive proteins were revealed by enhanced chemiluminescence with ECL Plus(Amersham, GE healthcare, Buckinghamshire, UK). Antibodies against PPARd and GAPDH were purchased from Santa Cruz Biotechnology. Blots were analyzed with a LAS-3000 imaging system (Fujifilm, Japan).

Fatty Acid Oxidation
Myotubes placed in a 12-well plate, or 400 mg of intact mouse quadriceps muscles were washed and incubated in low glucose DMEM (Invitrogen) containing 2% (w/v) fatty acid-free BSA,

Glucose Uptake
Differentiated primary human muscle myotubes were incubated in serum-free, low-glucose DMEM containing 0.1% BSA for 16 h at 37uC. Cells were treated with GW501516, AI extracts and vehicle for 24 h at 37uC and then stimulated with or without 100 nM insulin for 1 h at 37uC. Glucose uptake was initiated by the addition of 2-deoxy-D-[ 14 C]glucose (PerkinElmer Life) at a final concentration of 3 mmol/l for 10 min in HEPES buffer-saline (140 mM NaCl, 5 mM KCl, 2.5 mM MgCl 2 , 1 mM CaCl 2 , 20 mM HEPES, pH 7.4). The reaction was terminated by separating cells from the HEPES buffer saline and 2-deoxy-D-[ 14 C]glucose. After three washes in ice-cold PBS, the cells were extracted with 0.1% SDS and subjected to scintillation counting for 14 C radioactivity. The protein concentration was determined with a BCA assay kit (Pierce, Rockford, IL, USA), and the radioactivities were normalized by determining each total protein concentration.

Animal Experiments
C57BL/6J mice (male, 5 weeks of age) were purchased from Samtako (Osan, Korea). The mice were housed individually in a temperature-and humidity-controlled (26.5uC and 35%) facility with a 12-h light/dark cycle. All animals were allowed free access to water and diets. After acclimatization for 1 week, mice were fed a normal chow diet (D12450B, Research Diets, New Brunswick, NJ, USA) or a high-fat diet (60% kcal fat, D12492, Research Diets). Food consumption and body weight were recorded every week. A 95EEAI was given at 200 mg/kg of body weight by oral zoned needle once daily for 8 weeks. At the end of the experiment, blood and quadriceps muscle tissue samples were collected from the mice. Plasma triacylglycerol, total cholesterol and total ketone bodies were measured by an automated TBA-120FR biochemical analyzer (Toshiba) by using commercial assay kits (Dai-ichi Pure Chemicals, Tokyo, Japan). Plasma-free fatty acids and blood glucose levels were determined by a NEFA C-test (catalog no. 279-75401, Wako Pure Chemical, Osaka, Japan) and GLU neo SINO-Test (SINO-Test, Tokyo, Japan), respectively. Serum level of bhydroxybutyrate was measured with high sensitivity and specificity according to the manufacturer's directions (Autokit 3 b-hydroxybutyrate, Wako Diagnostics, Richmond, VA) by calculating the rate of Thio-NADH (b-thionicotinamide adenine dinucleotide) production spectrophotometrically at 405 nm upon oxidation of 3b-hydroxybutyrate.

Statistical Analysis
Experiments were performed at least three times. The data were reported as a mean 6 S.D. Results were analyzed by Student's ttest or ANOVA using the program SPSS 11.0 (SPSS, Chicago, IL, USA). Statistical significance was set at p#0.05.

A 95EEAI is PPARd ligand
We found that a 95EEAI interacted with the PPAR d ligand binding domain (LBD) (EC 50 = 7 ug/ml, Z9-Factor = 0.65, Figure 1) in repeated assay. We then determined the ability of a 95EEAI to activate PPARd using cell-based PPARd-responsive luciferase reporter assays. The synthetic PPARd agonist GW501516 (1 mM) caused a strong luciferase activity (Figure 2A). Similarly, a 95EEAI induced the luciferase activity in a dosedependent manner ( Figure 2B), while WEAI and 50EEAI had no effect (Figure 2A). Altogether, these results show that a 95EEAI is capable of activating PPARd via interaction with LBD of PPARd.

A 95EEAI induces the expressions of PPARd target genes in a PPARd-dependent manner
We next examined whether a 95EEAI could induce PPARdregulated genes in human skeletal muscle cells ( Figure S1). To ensure that effects are PPARd-dependent, we also determined the effect of 95EEAI on the expression of PPARd target genes in muscle cells in which endogenous PPARd expression was knocked down with RNAi. As shown in Figure 3A, mRNA expression of PPARd was reduced by treatment with RNAi. A 95EEAI treatment resulted in induction of several PPARd target genes involved in the fatty acid oxidation pathway, uncoupling and mitochondrial biogenesis including CPT1b, which cataliyzes the esterification of acyl-CoA to form acyl-carnitine, the rate-limiting step of fatty acid oxidation; PDK4, which plays an important role in switching the fuel source from glucose to fatty acids by inactivating pyruvate dehydrogenase; PGC1a, which is the key mitochondrial transcription regulators; UCP3 which is involved in energy expenditure; and LCAD, which is involved in mitochondrial fatty acid oxidation ( Figure 3B). The upregulation of these genes was completely abolished when PPARd expression was knocked down ( Figure 3B). Either WEAI or 50EEAI had no detectable effects on induction of these PPARd target genes in the same assay (data not shown). These results demonstrated that activation of PPARd by a 95EEAI leads to elevated expression of genes involved in lipid catabolism.

A 95EEAI promotes fatty acid oxidation and glucose uptake
To determine whether 95EEAI has direct effects on fatty acid oxidation and glucose uptake, we treated differentiated human skeletal muscle cells with 95EEAI. Treatment of 95EEAI led to 2fold increase in fatty acid oxidation (p,0.05, Figure 4A). The effects of 95EEAI on fatty acid oxidation were completely prevented by the knockdown of PPARd ( Figure 4A).
PPARd agonist GW501516 is known to enhance basal and insulin-stimulated glucose uptake in human skeletal muscle [14]. As shown in Figure 4B, there was a 1.8-fold increase in insulinstimulated glucose uptake in 95EEAI-treated cells compared with untreated control cells. However, in the absence of insulin stimulation the increase in glucose uptake in 95EEAI treated cells was not statistically significant. The siRNA-mediated reduction of PPARd expression was without effect on the stimulation of glucose uptake by either GW501516 or 95EEAI ( Figure 4B).

A 95EEAI attenuates HFD-induced obesity in mice
We and others predicted that enhanced fatty acid utilization and energy expenditure would protect against diet-induced obesity [6]. To determine whether 95EEAI regulate the progression of obesity, mice fed a HFD were orally administered the vehicle or 200 mg/kg 95EEAI once daily for 8 weeks. At the end of 8 weeks high-fat diet feeding, male C57BL/6J mice showed a significant increase in the rate of body weight gain compared with animals fed a normal diet (terminal body weight: normal diet group, 26.361.8 g vs. HFD group, 37.662.1 g, p,0.01, n = 9; Figure 5A). In contrast, 95EEAI treatment of mice resulted in significantly reduced body weight gain and fat pad mass compared with the vehicle-treated mice on HFD ( Figure 5A, B) without reducing food consumption.
To verify whether 95EEAI stimulates fatty acid oxidation, we measured the levels of fatty acid oxidation in quadriceps muscle. 95EEAI-treated mice on an HFD displayed higher levels of fatty acid oxidation than that of vehicle-treated mice on an HFD ( Figure 6A). We then examined whether increased enhanced fatty acid oxidation in the skeletal muscle affected the levels of lipid-derived substrates, such as nonesterified fatty acids (NEFAs), triglyceride (TG), and ketone bodies in the circulation. The plasma levels of free fatty acids and ketone bodies were significantly lower in 95EEAI-treated mice on an HFD than vehicle-treated mice on an HFD (Table 1). However, there was no statistically significant decrease in the plasma TG levels (Table 1).
To confirm that 95EEAI-induced fatty acid oxidation in skeletal muscle in vivo was also associated with increased expression of PPARd target genes involved in fatty acid oxidation pathway, we analyzed the expression of some representative genes (CPT1, LCAD, UCP2, UCP3, and PGC1a) by real-time quantitative PCR assay. The levels of mRNA expression were determined in individual animals, and the average expression for each group was presented. Consistent with results from the primary human myotubes, 95EEAI induced mRNA levels of these genes in skeletal muscle ( Figure 6B).

Discussion
Artemisia species are widely used in traditional medicine in East Asia, and have been reported to show anti-obesity, anti-diabetic, anti-lipogenic and anti-hyperglycaemic effects [15][16][17][18][19][20]. However, the molecular mechanism involved in Artemisia-induced lipid/ carbohydrate metabolism is poorly understood. Since PPARd activators have been shown to improve insulin resistance and reduce plasma glucose in rodent models of type 2 diabetes and reduce serum triglycerides in sedentary human [21,22], we aimed to test whether AI extracts have the ability to activate PPARd as a potential mechanism of action in mediating its beneficial effects.
We showed that 95EEAI interacted with the PPARd LBD leading to its activation. A 95 EEAI increased the expression of genes involved in lipid catabolism, enhanced fatty acid oxidation and insulin-stimulated glucose uptake in vivo as well as in human skeletal muscle cells, protected against diet-induced obesity. Furthermore, in PPARd knockdown cells, the positive effects of 95EEAI on fatty acid oxidation and their related genes expression  were no longer observed, suggesting that 95EEAI-mediated lipid metabolism would be PPARd-dependent. However, knockdown of PPARd expression did not alter the 95EEAI-mediated increase in insulin-stimulated glucose uptake. This result is in line with the report by Kramer et al. suggesting that direct activation of PPARd itself is not necessary for the stimulation of glucose uptake [23]. In vivo study, administration of 95EEAI to mice fed a HFD had no effect on fasting levels of blood glucose (Table 1). Our finding is in line with the work of Tanka et al. [24] who was also unable to detect changes in blood glucose levels in PPARd agonist-treated mice, despite the marked improvement in glucose tolerance and insulin sensitivity. Brunmair et al. [25] have reported that activation of PPARd acts to suppress glucose utilization as a result of a switch in substrate preference from carbohydrates to lipids in skeletal muscle, thus PPARd agonist fails to exert any effect on glucose uptake. Lee et al. [26] suggest that the improved glucose tolerance and insulin sensitivity triggered by PPARd agonist is due to promoting an increase in glucose flux through the pentosephosphate pathway and enhancing hepatic fatty acid synthesis. More studies are needed to elucidate the exact relationship between glucose utilization and 95EEAI-induced PPARd activation During starvation, glucose uptake and oxidation are reduced rapidly in muscle, which shifts to use free fatty acids and ketone bodies. In this study, 95EEAI-treated mice on an HFD showed a significant decrease in the plasma levels of free fatty acids and ketone bodies (Table 1). Tanka et al. [24] showed that the changes in gene expression by PPARd agonist are very similar to the gene expression profile induced by fasting in skeletal muscle. Hence, we speculate that the changes in levels of ketone bodies may be attributed to, at least in part, an increased uptake of ketone bodies in muscle through an activation of PPARd by 95EEAI.
The major compounds isolated from Artemisia species include terpenoids, flavonoids, coumarins, acetylenes, caffeoylquinic acids, and sterols [27] (Table S1). Major compounds of 95EEAI had no detectable effect on activation of PPARd protein (data not shown). Saturated and unsaturated fatty acids, such as arachidonic acid and eicosapentaenoic acid, are reported to be natural ligands for PPARd [28,29]. These fatty acids bind and activate PPARd in the low micromolar range. Although the bioactive component from 95EEAI for activation of PPARd was not chemically characterized yet, its structure may be similar to that of fatty acids.
Since activation of PPARd has shown to exert beneficial effects on preventing obesity-related diseases [30], natural compounds that enhance the activity of PPARd will provide a potential to develop a functional food with anti-obesity and anti-diabetic efficacies.
In summary, our data provide experimental evidence that 95EEAI is a natural PPARd agonist that robustly induces genes involved in fatty acid metabolism and activates fatty acid oxidation in vitro and in vivo, suggesting its potential as interventive and preventive measures for the treatment of metabolic disorders. Figure S1 The effects of 95EEAI on PPARd target genes expression in primary human myotubes. Primary human myotubes were treated with different doses of 95EEAI (0, 10, 25, 100 mM) or DMSO for 24 h. Total RNA was extracted from cells, and the mRNA levels of CPT1 and PDK4 genes were quantified by a real-time RT-PCR. Data from 3 independent experiments are represented as means 6 S.D. of the relative calculated with GAPDH as standard. *, p,0.05; **, p,0.01 versus DMSO-treated cells.

(TIF)
Table S1 Major chemical compounds in 95% ethanol extracts of Artemisia iwayomogi tested identified by GC-MS. The 95% ethanol extracts of Artemisia iwayomogi were analyzed using the Thermo Scientific TRACE GC Ultra TM gas chromatograph. It was fitted with a split-splitless injector and connected to an MS PolarisQ-Quadrupole Ion Trap (Thermo Electron) fused silica column VB5 (5% phenyl, 95% methylpolyxiloxane, 30 m with 0.25 mm i.d. film thickness 0.25 mm) (J & W Scientific Fisons, Folsom, CA, USA). The injector and interface were operated at 250 and 300uC, respectively. The oven temperature was programmed as follows: 50uC raised to 250uC (4uC/min) and held for 3 min. Helium was the carrier gas at 1 ml/min. The sample (1 ml) was injected in the split mode (1:20). MS conditions were as follows: ionization voltage EI of 70 eV, mass range 10-350 amu. The components were identified by comparing their relative retention times and mass spectra with those of authentic samples (analytical standards from data base). (DOC)