Z. mioga Roscoe flower buds were collected from a traditional market on Jeju Island (South Korea) in September 2013 and authenticated by Professor Seong-gyu Ko of the College of Oriental Medicine, Kyunghee University (Seoul, South Korea). A voucher specimen was deposited at the KFRI Herbarium (K00234) on September 23, 2012. Freshly picked plants were freeze-dried, ground using a mill and passed through a 50-mesh sieve. The ground powder was subjected to extraction with water or 70% ethanol twice using repeat sonication for 30 min at room temperature. The extract was filtered using a Whatman No. 2 paper filter (Advantec MFS, Inc., Dubulin, CA, USA) and concentrated using a vacuum evaporator (R-200; BÜCHI Labortechnik AG, Flawil, Switzerland). The extracts were freeze-dried and stored at −80°C.

Mouse anti-β-actin monoclonal antibody (sc-47778), rabbit anti-sterol regulatory element-binding protein (SREBP)-1c polyclonal antibody (sc-366), rabbit anti-CCAAT-enhancer-binding protein α (C/EBPα) polyclonal antibody (sc-61), mouse anti-peroxisome proliferator-activated receptor (PPAR)-γ monoclonal antibody (sc-7273), rabbit anti-phosphorylated (p)-nuclear factor (NF)-κB polyclonal antibody (sc-33039), rabbit anti-NF-κB polyclonal antibody (sc-109) and horseradish peroxidase (HRP)-conjugated anti-rabbit (sc-2004) and anti-mouse (sc-2005) secondary antibodies, were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Rabbit anti-p-AMP-activated kinase (AMPK; 2531), anti-AMPK (2532) and anti-FAS (318) polyclonal antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Oil Red O (O0625) and 3-(4,5-dime thylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), penicillin-streptomycin, bovine calf serum (CS) and phosphate-buffered saline (PBS) were obtained from Gibco (Thermo Fisher Scientific, Inc., Grand Island, NY, USA).

3T3-L1 cells were purchased from the American Type Culture Collection (Manassas, VA, USA). Cells were cultured in DMEM containing 10% CS, 100 U/ml penicillin and 100 µg/ml streptomycin at 37°C under 5% CO₂. The cells were seeded onto a 6-well plate (4×10⁵ cells/well) and maintained for 2 days. To induce differentiation, 2-day post-confluent 3T3-L1 cells (day 0) were incubated in DMEM containing 0.5 mM 3-isobutyl-1-methylxanthine (IBMX; Sigma-Aldrich), 1 µM dexamethasone (Sigma-Aldrich), 1 µg/ml insulin (Sigma-Aldrich) and 10% FBS for 3 days. Following induction, the medium was replaced with DMEM containing 10% FBS and 1 µg/ml insulin, and the cells were incubated for 2 days. Cells were then maintained in DMEM containing 10% FBS until they reached maturity. Between days 0 and 3, the cells were exposed to Z. mioga extract.

3T3-L1 cells were seeded in a 96-well plate (2×10³ cells/well). After 24 h of preconditioning, the cells were exposed to various concentrations of Z. mioga extracts for 24 or 48 h. Subsequently, 50 µl cell counting kit-8 solution (#CK04; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) was added to each well, and the plate was incubated for an additional 3 h at 37°C, to detect cell survival. Cell viability was calculated by measuring the absorbance on a microplate reader (Infinite M200; Tecan US, Inc., Morrisville, NC, USA) at 450 nm.

Cells were washed with PBS, fixed with 10% formalin for 1 h at room temperature, washed with PBS again and stained with filtered 0.5% Oil Red O solution in 60% isopropanol for 1 h. After the stained lipid droplets were washed with distilled water, they were observed under a fluorescence microscope (IX71; Olympus Corporation, Tokyo, Japan). Oil Red O was extracted from the cells with 100% isopropanol, and the optical density was determined using a microplate reader (Infinite M200) at an excitation wavelength of 490 nm.

Four-week-old male C57BL/6J mice were purchased from Orient Bio, Inc. (Seoul, South Korea) and acclimatized for 1 week prior to random assignment into experimental groups. The animals were housed in individual cages with free access to a regular diet and water in a room with a 12-h light cycle, a temperature of 24°C and a humidity of 55%. The mice were divided into four groups: Normal group fed the American Institute of Nutrient-76 diet (AIN-76; N group); a group fed a high-fat diet (HFD; HFD group); and two groups fed a HFD with 0.25 or 0.5% ZMW (HFD + ZMW 0.25 or 0.5% groups). The experimental diets were based on the AIN-76 diet, which contained 20% fat and 0.5% cholesterol (w/w). Animal studies were conducted in accordance with institutional and national guidelines, and all experimental procedures were approved by the Korea Food Research Institute Animal Care and Use Committee (#2014-0120). After consuming the experimental diets for 8 weeks, the mice were subcutaneously anesthetized with Zoletil 50 (30 mg/kg; Virbac Philippines Inc., Taguig, Philippines) in PBS and then sacrificed by cervical dislocation. Blood samples were collected from the abdominal aorta. The liver and epididymal fat tissues were harvested in liquid nitrogen and stored at −80°C for protein and RNA extraction.

The oral glucose tolerance test (OGTT) was performed after 8 weeks of ZMF administration. OGTT was determined in response to oral glucose administration (1.2 g/kg body weight) after fasting 12 h. Blood glucose was collected from the tail vein 0, 14, 30, 60 and 120 min after oral glucose administration. Blood glucose concentrations were determined by the glucose oxidase method using a blood glucose monitoring meter (Accu-Chek^® Performa Nano; Roche Diagnostics, Basel, Switzerland).

TG, TC and HDL were measured enzymatically using commercial kits (cat. nos. 20267, 20186 and 20081, respectively; Shinyang Chemical Co., Ltd., Busan, South Korea).

Cells were harvested in lysis buffer (Sigma-Aldrich) containing 40 mM HEPES (pH 7.4), 120 mM NaCl, 1 mM EDTA, 50 mM NaF, 1.5 mM Na₃VO₄, 10 mM β-glycerophosphate and 1% Triton X-100, supplemented with protease and phosphatase inhibitor cocktails (78440; Pierce; Thermo Fisher Scientific, Inc., Rockford, IL, USA). Protein concentration was determined using the Bradford protein assay. The proteins were separated by 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride (PVDF) membranes, followed by blocking with 5% non-fat dry milk for 1 h. Subsequently, the membranes were incubated overnight at 4°C with mouse anti-β-actin monoclonal antibody (1:2,000), rabbit anti-SREBP-1c polyclonal antibody (1:1,000), rabbit anti-C/EBPα polyclonal antibody (1:1,000), mouse anti-PPAR-γ monoclonal antibody (1:1,000), rabbit anti-p-NF-κB polyclonal antibody (1:1,000), rabbit anti-NF-κB polyclonal antibody (1:1,000), rabbit anti-p-AMPK polyclonal antibody (1:1,000), rabbit anti-AMPK polyclonal antibody (1:1,000) and rabbit anti-FAS polyclonal antibody (1:1,000), followed by washing with tris-buffered saline containing Tween-20 and incubation with HRP-conjugated anti-rabbit and anti-mouse IgG (1:1,000; Santa Cruz Biotechnology, Inc.) at room temperature for 1 h. The bands were visualized using a chemiluminescence reagent (GE Healthcare, Little Chalfont, UK), and the relative intensities of the bands to β-actin signal was quantified by ImageJ software v1.45 (National Institutes of Health, Bethesda, MD, USA).

Total RNA was extracted from 20–30 mg liver tissue using a NucleoSpin RNA II kit (Macherey-Nagel GmbH & Co. KG, Düren, Germany) according to the manufacturer's protocol. Following genomic DNA elimination with the gDNA elimination buffer in the RT² First Strand kit (cat. no. 330401, Qiagen, Inc. Valencia, CA, USA), cDNA was synthesized using 1 µg RNA and the ReverTra Ace^® qPCR RT kit (Toyobo, Co., Ltd., Osaka, Japan). qPCR was conducted on an StepOnePlus Real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc., Foster City, CA, USA) using SYBR Green real-time PCR Master Mix (Toyobo, Co., Ltd.) as follows: Pre-denaturation at 95°C for 5 min, followed by 40 cycles of 95°C for 15 sec, 60°C for 15 sec and 75°C for 45 sec. The PCR reaction was performed in a volume of 20 µl, containing 2 µl cDNA template and 10 pmol primers. The specific primer sequences were as follows: Phosphoenolpyruvate carboxykinase (PEPCK) forward, AAAAGCCTTTGGTCAACAAC and reverse, AAACTTCATCCAGGCAATGT; G6Pase forward, GAGTCTTGTCAGGCATTGCT and reverse, GGTACATGCTGGAGTTGAGG; interleukin (IL)-6 forward, AGTTGCCTTCTTGGGACTGA and reverse, TCCACGATTTCCCAGAGAAC; and tumor necrosis factor (TNF)-α forward, ACGGCATGGATCTCAAAGAC and reverse, AGATAGCAAATCGGCTGACG. The experiment was conducted in duplicate with a negative control (no cDNA) and RT control (no reverse transcription). RNA levels were expressed as the ratio of the signal intensity for each gene relative to β-actin and were calculated using the 2^−ΔΔCq method (13).

Differences between groups were evaluated using a one-way analysis of variance (ANOVA) with Prism5 software (GraphPad Software, Inc., San Diego, CA, USA). The Bonferroni post-hoc correction for multiple comparisons was used when significant differences were identified using ANOVA (P<0.05). Data are expressed as the mean ± standard deviation.

To establish the optimal extract concentration for use in vitro, the cells were treated with media containing 0–500 µg/ml ZMW or Z. mioga ethanol extracts (ZME) for 24 or 48 h. The viability of 3T3-L1 preadipocytes was not significantly affected by extract treatment up to 250 µg/ml (Fig. 1A and B). We next compared the inhibitory effects of the extract (0–250 µg/ml) on lipid accumulation during 3T3-L1 preadipocyte cell differentiation. Oil Red O staining was used to determine lipid droplet accumulation in 3T3-L1 adipocytes. ZMW significantly and dose-dependently inhibited lipid accumulation, whereas ZME had no effect (Fig. 1C and D). The expression of several proteins involved in adipogenesis, including C/EBPα and PPARγ, were markedly suppressed by ZMW (Fig. 1E). These results suggest that ZMW, but not ZME, may have anti-obesity effects.

We next investigated whether ZMW was able to prevent HFD-induced obesity in mice. Total body and liver weight were significantly decreased in the HFD + ZMW 0.5% group, as compared to the HFD group (Fig. 2A and B). Unexpectedly, ZMW treatment did not affect adipose tissue weight (Fig. 2C), although ZMW inhibited adipogenesis in vitro.

We next evaluated the effects of ZMW on the serum and liver lipid levels in mice fed HFD. ZMW significantly reduced the serum TG concentration (N, 37.5±2.3; HFD, 62.3±5.1; HFD + ZMW 0.5%, 48.3±4.2) and TC (N, 162.0±14.2; HFD, 243.2±14.3; HFD + ZMW 0.5%, 208.4±22.6) (Fig. 2D), although there was no significant difference in HDL cholesterol. Furthermore, hepatic TC and TG in the ZMW group were significantly reduced compared with the HFD group (Fig. 2E), although the total lipid levels in the ZMW group were not significantly different compared with the HFD group. These results indicate that ZMW effectively lowers serum and liver cholesterol.

Lipid accumulation in the liver can lead to insulin resistance by interfering with insulin signal transduction (14). We next examined the effect of ZMW on insulin resistance in HFD-induced obese mice. The OGTT was used to evaluate insulin resistance. The ZMW group showed significantly lower blood glucose levels compared with the HFD group (Fig. 3A). Consequently, the area under the curve (AUC) in the ZMW group was significantly reduced (P<0.05) (Fig. 3B). To test the effect of ZMW on the expression of gluconeogenic genes, including PEPCK and G6Pase, the mRNA expression levels of PEPCK and G6Pase were determined using RT-qPCR. ZMW supplementation significantly reduced hepatic mRNA expression of PEPCK and G6Pase compared with the control group (Fig. 3C and D), indicating that HFD impairs hepatic insulin signaling, and ZMW supplementation effectively restores hepatic insulin sensitivity.

The AMPK signaling pathway serves a crucial function in liver lipid metabolism; when AMPK is activated, it inhibits the fatty acid and sterol synthesis pathway, and activates the fatty acid oxidation pathway (15). Furthermore, activated AMPK suppresses SREBP-1c, a lipogenesis-associated gene (16). Consistent with the reduced liver weight, western blot analysis indicated that ZMW treatment increased AMPK activation and decreased the SREBP-1c (Fig. 4A and B), although it did not suppress adipogenesis (Fig. 4C and D). These results suggest that ZMW may reduce body weight via the downregulation of hepatic lipogenesis.

Diet-induced obesity triggers inflammatory responses (17). When hepatic proinflammatory cytokine mRNA levels were examined, HFD-fed mice displayed a significant increase in IL-6, although TNF-α levels were slightly decreased, as compared with normal mice (Fig. 5A and B). The increased IL-6 levels were significantly decreased by ZMW supplementation (Fig. 5A). However, ZMW had no effect on TNF-α levels (Fig. 5B). We also examined whether ZMW inhibits NF-κB activation (Fig. 5C and D). ZMW reduced NF-κB phosphorylation, indicating that ZMW ameliorates HFD-induced inflammatory enzymes, such as NF-κB. Collectively, the present results suggest that ZMF improves hepatic steatosis while decreasing liver inflammation.