Helminthostachys zeylanica alleviates hepatic steatosis and insulin resistance in diet-induced obese mice

Rhizomes of HZ were purchased from the Wanhua herbal market (Taipei, Taiwan) and identified by comparison with the voucher specimen (NRICM-99-003), which is already deposited at the herbarium of the National Research Institute of Chinese Medicine, Taiwan. HZ rhizomes (531 g) were heated and extracted with 2.5 l of EtOH-H₂O (1:1) under reflux for 1 h. The filtrate was concentrated and lyophilized to yield HZ extract (29 g, yield 5.46%).

The preparation of ugonins J and ugonin K were prepared as described previously [11]. Briefly, the rhizomes of HZ (12 kg) were extracted with EtOH (20 l × 3) at 50 °C for 24 h. The concentrated EtOH extract (460 g) was partitioned between EtOAc and H2O, and the EtOAc extract (153 g) was applied to a silica gel column eluted with gradient solvent systems of n-hexane–EtOAc (20:1–1:10) and EtOAc–MeOH (10:1–1:1) to yield 16 fractions (Fr-1–Fr-16). Fraction Fr-7, the eluate of n-hexane–EtOAc = 1:2, was further subjected to a silica gel CC (CH2Cl2–MeOH = 30:1) and Sephadex LH-20 (MeOH–H2O = 5:1) to give ugonin J (26.3 mg) and ugonin K (18.6 mg), respectively.

The HZ extract (1.0 g) was refluxed in 20 ml methanol for 30 min and filtered. The filtrate volume was then adjusted to 50 ml with the same solvent. A 10 μl portion of the solution was injected into the HPLC system, an Agilent 1100 series equipped with a G1311A Quat Pump, a G1379A degasser, a G1315B photodiode array detector, a 1200 series G1329A autosampler, and a column oven H-650 (Chrom Tech, TNC.). A Cosmosil 5C18-AR-II column was utilized with a mobile phase of MeOH-H₂O (0.1% phosphoric acid, v/v) using a linear gradient, which started from 70% MeOH for 35 min, increasing to 75% in 10 min, and finally reaching 100% at 65 min with flow rate of 1.0 ml/min. The column oven was set at 30 °C and the UV detection wavelength was set at 344 nm.

Palmitate, Oil red O, and luteolin were purchased from Sigma-Aldrich, St. Louis, MO. Antibodies against AMPK, pACC (Ser 79), ACC, SREBP-1c, CPT1, and tubulin were from Genetex. The anti-pAMPK (Thr 172) antibodies were obtained from Millipore, and the HRP-conjugated anti-mouse or anti-rabbit secondary antibodies were from Jackson ImmunoResearch Laboratories Inc. Western blot analysis was performed as described previously [16].

HuS-E/2 cells were kindly provided by Dr. Shimotohno (Kyoto University, Japan) and maintained as described previously [18]. The lyophilized HZ extract was solubilized in DMSO as stock at the concentration of 25 mM and diluted to the indicated concentration. DMSO was used as the vehicle for experimental control. For the fatty liver disease cell model, HuS-E/2 cells were cultured with 0.1 mM palmitate for 18 h. To measure lipid content in HuS-E/2 cells, the oil red O method was used as previously described [16].

For messenger RNA (mRNA) analysis, real-time polymerase chain reaction (RT-PCR) was performed as described previously [19]. The primer sets used in this study are listed in Additional file 1: Table S1.

The plasma, epididymis adipose, and liver tissues were collected from each sacrificed mouse. The biochemical analysis of plasma and histological analysis of the fat and liver tissues were performed as described previously [20].

The 12 h fasting blood glucose was measured with a glucose analyzer (EASYTOUCH, Taiwan). The plasma insulin and HOMA-IR were detected and calculated as described previously [20].

All data are expressed as the mean ± SD from for three separate experiments. More than two sets of data were accessed by one-way ANOVA with Dunnett’s multiple comparison test. The values significantly different from the control were indicated by asterisks (*, p < 0.05; **, p < 0.01; ***, p < 0.001.).

Rhizomes of HZ were extracted and the chemical components were analyzed. HPLC analysis was performed on the HZ extract and two of the individual ingredients, ugonins J and K were isolated [11] and used as standard markers for quality control of HZ material. Both standard markers were well separated and their purities were determined by HPLC to be more than 98%. The HPLC chromatogram of the HZ extract showed two major peaks at 44.484 and 60.466 min. (Fig. 1a), corresponding to ugonin J (44.588 min.) (Fig. 1b) and ugonin K (60.276 min.) (Fig. 1c) under the same conditions.

Fatty liver disease is mainly attributable to triglyceride accumulation in the hepatocytes [21]. To determine the effect of HZ extract on esterification in human liver cells and deposition of fatty acid as lipid droplets, HuS-E/2 immortalized human primary hepatocytes were used as a human fatty liver cell model [16]. HuS-E/2 cells were incubated with palmitate and 100 μg/ml of HZ extract for 18 h. The lipid content of the cells was observed by Oil-Red O staining and quantified. As shown in Fig. 2a, compared to HuS-E/2 cells with only palmitate, cells incubated with the HZ extract showed significantly less lipid accumulation. The reduction of cellular lipid accumulation to 39% with the treatment with HZ extract was confirmed by image quantification (Fig. 2b). Because we found HZ extract had an inhibitory effect on lipid deposition in human hepatocytes, possible molecular mechanisms were explored. AMP-activated protein kinase (AMPK) was reported to regulate fat metabolism in the liver and change with cellular energy status [22]. To determine whether HZ extract increased levels of AMPK and its activation, HuS-E/2 cells were incubated with 0.1 mM palmitate in the presence or absence of HZ extract and the expression of AMPK was assessed by western blotting. Levels of phospho-AMPK (pAMPK) at Thr-172 were assessed to evaluate AMPK activation. Ugonins J and K are structurally related to natural flavonoid luteolin [23], which has been demonstrated to attenuate hepatic steatosis [24]. Therefore, the agent luteolin was used as a positive control drug in the subsequent experiment. As shown in Fig. 2c, HZ extract increased AMPK phosphorylation at Thr-172 in palmitate-treated HuS-E/2 cells. In addition, the activation of AMPK’s downstream target enzyme, acetyl-CoA carboxylase (ACC), by phosphorylation at Ser-79 was also measured. HZ extract significantly increased ACC protein phosphorylation. The results indicated that HZ extract facilitated AMPK and ACC activation in HuS-E/2 cells under high-fat conditions. HZ extract showed a stronger effect on AMPK and ACC activation than luteolin in HuS-E/2 cells.

We also identified the changes in proteins associated with fatty acid synthesis and β oxidation. Compared to the palmitate group, HZ extract substantially decreased sterol regulatory element-binding transcription factor 1c (SREBP-1c) protein expression, which is involved in fatty acid synthesis (Fig. 2d). Carnitine palmitoyltransferase I (CPT1) functions as catalyzing fatty acids by β oxidation [25]. Treatment with HZ extract greatly increased CPT1 protein expression, compared to the palmitate group. We determined whether lipid metabolism-related genes in the hepatocytes were influenced by HZ intervention. The expression of transcription factors, peroxisome proliferator-activated receptor alpha (PPARα) and peroxisome proliferator-activated receptor delta (PPARδ), associated with fatty acid β oxidation were markedly increased by treatment with HZ extract, compared to the palmitate and luteolin groups (Fig. 2e). The activities of genes involved in de novo lipogenesis in the hepatocytes, SREBP-1c and peroxisome proliferator-activated receptor gamma (PPARγ) were substantially higher in the palmitate group than the non-treated group, while all the genes were expressed at greatly lower levels after treatment with HZ extract and luteolin, compared with the palmitate group (Fig. 2f). Taken together, the results suggest HZ extract had a better effect than luteolin on inhibition of fatty acid synthesis and activation of fatty acid β oxidation in palmitate-treated HuS-E/2 hepatocytes.

The effect of HZ extract on lipemia syndrome and fatty liver was examined in an HFD mouse model. Five-week-old male C57BL/6 J mice were fed with normal diet (ND group, n = 10), HFD (HFD group, n = 10), or HFD along with 0.5% lyophilized HZ extract (HFD-HZ group, n = 10) for 12 weeks. The morphology of the ND, HFD, and HFD-HZ mice was observed, as shown in Fig. 3a. The size and waist were obviously smaller in the ND group and HFD-HZ group than the HFD mice. The weight of HFD-HZ mice was significantly lower than the HFD mice after 12 weeks of diet supplemented with HZ (Fig. 3b). The food efficiency ratio (FER) was much lower in the HFD-HZ group than the HFD group (Fig. 3d), although the amount of food consumed did not differ significantly (Fig. 3c). This suggests that HZ extract causes reduced food uptake, which may be the reason the mice gain less weight.

A feature of obesity is increasing lipid accumulation within adipocytes, leading to excessive visceral fat deposits. Therefore, the epididymis adipose tissue (EAT) was dissected and measured after 12 weeks of experimental diet. The mass of EAT in the HFD group was significantly higher than the ND and HFD-HZ groups (Fig. 4a). The adipocytes from the mice that had an HFD with HZ supplement were lower in diameter than those of the HFD group (Fig. 4b and c), which suggests that HZ extract lowers lipid deposition in the mice.

Alteration of the lipid composition of serum is one of the signs of metabolic problems [26] and plasma lipid levels were monitored after the diet to evaluate the degree of metabolic deficiency. TG, TC, HDL-C, and LDL-C were measured. Significantly higher levels of TG, TC, and LDL-C were expressed in the HFD group than the ND group (Fig. 5a, b, and d). Interestingly, the plasma TG, TC, and LDL-C levels in HFD-HZ group were significantly lower than the HFD group. A high HDL-C level was detected in both the HFD and HFD-HZ groups and may be a consequence of the high cholesterol diet. It is suggested that the presence of hypertriglyceridemia and the high cholesterol phenomenon in the HFD mouse model is consistent with the symptom of obesity in humans, which suggests the HZ extract has the potential to inhibit hyperlipidemia bioactivity.

Non-alcoholic fatty liver disease (NAFLD) is one of the criteria for the development of metabolic syndrome, which is mainly due to triglyceride accumulation in the hepatocytes [27, 28]. To examine the effect of HZ on fatty liver, the liver was weighed and markers of hepatic-steatosis were evaluated to determine the incidence of steatohepatitis. The livers in the HFD group were heavier than the ND group, but lower when the mice had an HFD diet with the HZ extract (Fig. 6a). The increasing weight of the liver may be attributable to lipid accumulation. In addition, the hepatocytes of the HFD mice became swollen with foaming morphology, showing a lack of staining by H&E and suggesting more lipid deposition (Fig. 6b). The cell morphology with H&E staining in the HFD-HZ mouse model was more similar to the ND group. The levels of plasma GOT and GPT help diagnose injury to hepatic tissue [29, 30]. These markers of hepatic injury, GOT and GPT, were upregulated in HFD mice but maintained at lower levels by treatment with the HZ extract (Fig. 6c and d). There were no significant effects on the pancreas inflammatory marker, LIP (Fig. 6e). These data suggest that HZ extract specifically prevented diet-induced hepatic steatosis and inflammatory response in the liver.

Previous studies have confirmed the association between dietary fat intake and deteriorated insulin function [31]. The fasting-glucose and insulin levels in HFD mice with or without HFD or HZ extract were measured, reflecting the incidence of insulin resistance syndrome. High fasting blood glucose levels in the HFD group suggested an abnormality of insulin function, while this was prevented by HZ extract treatment (Fig. 7a). The uncontrolled high-fasting insulin level is considered as the earliest sign of the onset of metabolic syndrome [32]. Interestingly, the increase in insulin levels was prevented in the HFD-HZ mice, compared to the HFD group (Fig. 7ba). Generally, insulin resistance is monitored using the homeostasis model assessment of insulin resistance (HOMA-IR) [33]. Therefore, the high HOMA-IR index reflected the increased incidence of diabetes symptoms in HFD mice with high fasting blood glucose and insulin levels. The HOMA-IR index was calculated and found to be at the control level with HZ extract treatment (Fig. 7c), indicating inhibition of insulin resistance.