Potential lipase inhibitors from Chinese medicinal herbs

Abstract

Context: Obesity has become a major health concern, and it places both personal and economic burdens on the world’s population. Traditional Chinese medicinal herbs are rich source of lead compounds and are possible drug candidates, which may be used to treat this condition.

Objective: This study screened potent pancreatic lipase inhibitors found in traditional Chinese medicinal herbs for ability to treat obesity.

Materials and methods: A porcine pancreatic lipase inhibition assay was established, and the inhibitory activity of 35 traditional Chinese medicinal herbs was evaluated at a concentration of 200 μg/mL. Two elutions of herbal extracts with strong lipase inhibitory activity were further fractionated by preparative high-performance liquid chromatography into 22 sub-fractions each, and these sub-fractions were tested for anti-lipase activity. Sub-fractions, which exhibited strong lipase inhibitory activity, were continuously fractionated into individual compounds. Two active compounds with potent anti-lipase activity were finally isolated and identified from two traditional Chinese medicinal herbs, respectively.

Results: Among 35 traditional Chinese medicinal herbs, the 95% ethanol elutions of Panax notoginseng (Burk.) F.H. Chen (Araliaceae) and Magnolia officinalis Rehd. et Wils (Magnoliaceae) showed strong anti-lipase activity. Two compounds, including 20(S)-ginsenoside Rg3 and honokiol were identified using bioactivity-guided isolation with IC₅₀=33.7 and 59.4 μg/mL, respectively.

Conclusion: 20(S)-ginsenoside Rg3 and honokiol might be suitable candidates for the treatment of obesity.

Keywords:

• 20(S)-ginsenoside Rg3
• bioactivity-guided isolation
• honokiol
• obesity

Introduction

Obesity has become a serious public health problem worldwide (Stevens et al. 2012). Obesity not only causes diseases and other painful conditions, such as hyperlipidemia, hypertension, arteriosclerosis, non-insulin-dependent diabetes mellitus, and coronary heart disease (Jung 1997; Kopelman 2000) but also places an enormous economic burden on individuals and society. In 2005, about 190 billion had been expensed for obesity-related healthcare in the U.S. alone (Lehnert et al. 2013). Medical costs have increased dramatically and may continue to grow (Wang et al. 2008; Mora et al. 2015). In addition to these direct expenses, obesity also causes loss of productivity (Trogdon et al. 2008). The research and development of new anti-obesity drugs has recently drawn attention from both the pharmaceutical industry and academic research centers.

Obesity involves the accumulation of excessive amounts of body fat (Hill et al. 2000). The subjects were fed high-fat diets, and triacylglycerols from those fats degraded by lipase and absorbed by intestinal cells (Shi & Burn 2004). In the digestion process, pancreatic lipase is a key lipid-digesting enzyme (Hadvary et al. 1988). Inhibiting pancreatic lipase is one method to prevent and treat obesity (Birari & Bhutani 2007). For instance, one pancreatic lipase inhibitor, orlistat, was approved by U.S. Food and Drug Administration in 1999 to treat obesity (Leung et al. 2003). Although the drug has a significant anti-obesity effect, it also causes non-negligible gastrointestinal side effects (Cheung et al. 2013). It is important to find and develop more pancreatic lipase inhibitors that are safe and effective and do not share orlistat’s drawbacks.

Natural products from traditional medicinal plants and microbial sources are recognized as an important source of novel drug leads (Molinari 2009). About 50% of approved drugs are derived from natural products (Newman & Cragg 2007). In recent years, natural products have been found to exhibit promising activity toward the inhibition of pancreatic lipase (Birari & Bhutani 2007; de la Garza et al. 2011). This has drawn the interest of a large number of researchers (Kim & Kang 2005; Sharma et al. 2005; Slanc et al. 2009; Zheng et al. 2010; Ramirez et al. 2012; Roh & Jung 2012). Their efforts are likely to render the identification and development of potent pancreatic lipase inhibitors from traditional Chinese medicinal herbs, a viable alternative strategy for safe and effective anti-obesity drug discovery (de la Garza et al. 2011).

In this study, 35 traditional Chinese medicinal herbs that showed promising inhibitory activity against obesity, hyperlipidemia, or both under molecular docking analysis and literature mining were selected for discovery of new lead compounds and pancreatic lipase inhibitors through bioassay-guided isolation.

Materials and methods

Plant materials and chemicals

All herbs were supplied by a Kangji drugstore from May to June 2013 and identified by Dr. Zhenzhong Wang (Jiangsu Kanion Pharmaceutical Co. Ltd.). A collection of voucher specimens is available for confirmation in the research center of Jiangsu Kanion Pharmaceutical Co. Ltd. (Lianyungang, P.R. China). Porcine pancreatic lipase (PPL) type VI-S and p-nitrophenylpalmitate (PNP) were purchased from Sigma Chemical (Shanghai, China). 20(S)-Ginsenoside Rg3 and honokiol were obtained from National Institutes for Food and Drug Control (Beijing, China). High-performance liquid chromatography-grade (HPLC) acetonitrile was provided by Tedia (Anhui, China). All other chemicals and solvents were of analytical grade.

Preparation of samples

All herbs were air-dried and ground into fine powder. The powders (50 g) were extracted with distilled water (10 times volume per weight) under reflux twice, 2 h each time. After filtering, the water extract was adsorbed onto HP-20 macroporous resin (100 mL) (Diaion HP-20, Mitsubishi Chemical Ind. Ltd., Tokyo, Japan). Then, the HP-20 column was washed with H₂O (400 mL), 20% ethanol (400 mL), 40% ethanol (400 mL), 95% ethanol (400 mL) in sequence. Then, the solutions were dried on a rotary vacuum evaporator at 50 °C. Samples were stored under dry conditions for further studies.

The 95% ethanol elution of Magnolia officinalis Rehd. et Wils (Magnoliaceae) and Panax notoginseng (Burk.) F.H. Chen (Araliaceae) were fractionated by preparative HPLC (Agilent 1260 equipped with multiple wavelength detector and a Fuji-C-18 column). The mobile phase consisted of water (A) and acetonitrile (B) at a flow rate of 30 mL/min: 20–30% B at 0–20 min; 30–45% B at 20–40 min; 45–90% B at 40–55 min; 90–100% B at 55–60 min; 100% B at 60–70 min. The detection wavelength was set at 210 nm. The sub-fractions were collected every 3 min and concentrated by distillation under reduced pressure. Then the sub-fractions were lyophilized and stored at room temperature.

PPL inhibition assay

Lipase activity was measured using PNP as substrate (Winkler & Stuckmann 1979; Lee et al. 1993). PPL stock solution (1.2 mg/mL) was prepared in 50 mM Tris-HCl buffer pH 8.0 (combined with 0.2% sodium deoxycholate and 0.1% gum arabic). This mixture was aliquoted and stored at −70 °C. PNP was dissolved in isopropanol at a stock solution of 25 mM and stored at 4 °C. To determine inhibitory activity against lipase, the samples and compounds were pre-incubated with the enzyme (5 μg/mL) for 10 min in Tris-HCl buffer at 37 °C. The reaction was then started by adding 0.1 mL PNP (0.5 mM). The final volume was 0.2 mL. After incubation at 37 °C for 60 min, the absorbance at 405 nm was measured using a micro-plate reader (Molecular Devices, Sunnyvale, CA).

The inhibitory rate was calculated using the following formula (Zheng et al. 2010):

Here, A is the activity in the absence of the inhibitor; a is the negative control in the absence of inhibitor; B is the activity in the presence of the inhibitor; and b is the negative control in the presence of the inhibitor.

HPLC (LC-DAD/Q-TOF/MS) analysis

Agilent 1290 HPLC (Santa Clara, CA) equipped with diode array detector and quality Time-of-Flight 6538 mass spectrometer was used. Chromatographic separation was performed on a Phenomenex Luna C18 column (Torrance, CA) for the 19th sub-fraction of Magnolia officinalis (MO-19) and the 17th sub-fraction of Panax notoginseng (PN-17). The mobile phase was composed of water (A) and acetonitrile (B). The gradient elution was performed as follows: 0–25 min, 30–35% B; 25–55 min, 35–50% B; 55–80 min, 50–90% B at a flow rate of 1 mL/min for MO-19; 0–30 min, 35–50% B; 30–35 min, 50% B; and 35–50 min, 50–90% B at a flow rate of 1 mL/min for PN-17. The detection wavelengths were set at 230 and 210 nm for MO-19 and PN-17, respectively. For MS detection, the parameters were set as follows: column oven temperature, 30 °C; collision gas, ultra-high-purity helium; nebulizing gas, high-purity nitrogen; capillary voltage, 4000 V; drying gas flow rate, 10 L/min; gas temperature, 350 °C; pressure of nebulizer, 50 psi; skimmer voltage, 65 V; and fragmentor voltage, 100 V. The mass range was set to m/z 100–3200 Da. Each sample was analyzed in positive modes to determine its structure.

Statistical analysis

All results are expressed as the means ± standard deviation (n = 3). The significance of differences from the control was determined by Duncan’s test and Kruskal–Wallis test, and p < 0.05 was considered significant. The IC₅₀ values were calculated using GraphPad Prism (La Jolla, CA).

Results

Anti-lipase activity of herbal elutions

In the current study, 35 traditional Chinese herbal medicines were extracted and eluted with 20%, 40% and 95% ethanol. Then the pancreatic lipase inhibitory activities of the elutions were tested at a concentration of 200 μg/mL. The 95% ethanol elution of most herbal extracts exhibited stronger anti-lipase activity than the 20% and 40% ethanol elutions (data not shown). Among the thirty-five 95% ethanol elutions, the inhibitory rates of five herbal elutions were greater than 20%, including Panax notoginseng (58.1%), Magnolia officinalis (44.2%), Arnebia euchroma (Royle) Johnst (Boraginaceae) (28.3%), Prunus persica (L.) Batsch (Rosaceae) (28.7%) and Paeonia suffruticosa Andr (Ranunculaceae) (20.4%). The other elutions exhibited weak inhibitory activities against lipase, ranging from 1.1% to 18.6% (Table 1).

Table 1. PPL inhibitory activities of various Chinese medicinal herbs.

No	Scientific name	Plant part	Family	Inhibition%
1	Atractylodes macrocephala Koidz	Rhizome	Compositae	11.0 ± 0.7
2	Panax notoginseng (Burk) F.H.Chen	Root	Araliaceae	58.1 ± 0.5
3	Pueraria lobata (Willd.) Ohwi	Root	Leguminosae	7.4 ± 1.1
4	Salvia miltiorrhiza Bunge	Root	Labiatae	12.4 ± 0.3
5	Alisma plantago-aquatica Linn	Tuber	Alismataceae	12.3 ± 0.5
6	Phellodendron chinense Schneid	bark	Rutaceae	11.5 ± 0.8
7	Rubia cordifolia L.	Root	Rubiaceae	5.7 ± 1.2
8	Paeonia lactiflora Pall.	Root	Ranunculaceae	14.4 ± 0.4
9	Astragalus membranaceus (Fisch) Bge. var. mongholicus (Bge.) Hsiao	Root	Leguminosae	5.2 ± 1.1
10	Ligusticum chuanxiong Hort.	Rhizome	Umbelliferae	1.1 ± 0.3
11	Acorus tatarinowii	Rhizome	Araceae	13.6 ± 0.7
12	Liquidambar formosana Hance	Ripe fruit	Hamamelidaceae	12.6 ± 0.5
13	Gleditsia sinensis Lain.	Spines	Leguminosae	10.0 ± 1.4
14	Eugenia caryophyllata Thunb.	Bud	Myrtaceae	16.4 ± 0.9
15	Clinopodium polycephalum (Vaniot) C. Y. Wu et Hsuan	Aboveground	Labiatae	5.7 ± 0.7
16	Trichosanthes kiriowii Maxim.	Ripe fruit	Cucurbitaceae	3.8 ± 0.8
17	Allium macrostemon Bge.	Bulbs	Liliaceae	6.4 ± 1.6
18	Arnebia euchroma (Royle) Johnst.	Root	Boraginaceae	28.3 ± 0.8
19	Erigeron breviscapus (Vant.) Hand.-Mazz	The whole plant	Compositae	2.2 ± 0.4
20	Glycyrrhiza uralensis Fisch.	Root	Leguminosae	9.5 ± 1.1
21	Citrus aurantium L.	Young fruit	Rutaceae	12.5 ± 1.4
22	Epimedium brevicornu Maxim.	leaf	Berberidaceae	12.2 ± 1.7
23	Prunus persica (L.)Batsch	Mature seeds	Rosaceae	28.7 ± 1.4
24	Eucommia ulmoides Oliv.	Bark	Eucommiaceae	8.6 ± 0.8
25	Curculigo orchioides Gaertn.	Rhizome	Amaryllidaceae	16.0 ± 0.9
26	Magnolia officinalis Rehd. et Wils	Bark	Magnoliaceae	44.2 ± 1.2
27	Corydalis yanhusuo W. T. Wang ex Z. Y. Su et C. Y. Wu	Tubers	Papaveraceae	9.1 ± 1.5
28	Carthamus tinctorius L.	Flower	Compositae	14.7 ± 0.5
29	Ligustrum lucidum Ait.	Fruit	Oleaceae	18.6 ± 1.3
30	Cistanche deserticola Ma	Stems	Orobanchaceae	16.0 ± 1.4
31	Eclipta prostrata (L.) L.	The whole plant	Compositae	9.7 ± 0.1
32	Angelica sinensis (Oliv.) Diels	Root	Umbelliferae	4.4 ± 0.4
33	Paeonia suffruticosa Andr.	Root bark	Ranunculaceae	20.4 ± 1.1
34	Dipsacus asperoides C. Y. Cheng et T. M. Ai	Root	Dipsacaceae	15.6 ± 1.8
35	Evodia rutaecarpa (Juss.) Benth.	Fruit	Rutaceae	17.7 ± 0.5

Three measurements were carried out per elution. Data were expressed as an average ± standard deviation (n = 3). The final concentration of the elutions used in the screening was 200 μg/mL.

Anti-lipase activity of sub-fractions from Panax notoginseng and Magnolia officinalis

Both Panax notoginseng and Magnolia officinalis exhibited strong anti-lipase activity. Twenty-two sub-fractions from the 95% ethanol elution of Panax notoginseng and Magnolia officinalis were then prepared using preparative HPLC. The inhibitory activities of the sub-fractions were investigated on pancreatic lipase. As shown in Figure 1, PN-17 of Panax notoginseng exhibited the strongest anti-lipase activity of these 22 sub-fractions, and MO-19 of Magnolia officinalis showed the most potent anti-lipase activity out of the 22 sub-fractions.

Figure 1. PPL inhibitory activities of 22 sub-fractions from Panax notoginseng and Magnolia officinalis. Three measurements were carried out per sub-fraction. Data were expressed as an average ± standard deviation (n = 3). The final concentration of the sub-fractions was 200 μg/mL.

Identification of active compounds

PN-17 and MO-19 were separated by LC-DAD/Q-TOF/MS (Figures 2(a) and 3(a)). The retention time of the reference substances were compared, the three peaks of PN-17 were attributed to ginsenoside Rh4, 20(S)-ginsenoside Rg3 and 20(R)-ginsenoside Rg3 (Yao et al. 2011; Kim et al. 2013), and the only peak of MO-19 was attributed to honokiol (Lee et al. 2011). Their identities were then confirmed by mass spectrometry (Table 2). The result of the analysis of the compound activities further suggested that the second peak had the strongest anti-lipase effect of the three peaks of PN-17 (Figure 2(b)). The dose-response curve of 20(S)-ginsenoside Rg3 and honokiol revealed the two compounds to be strong pancreatic lipase inhibitors (Figures 2(c) and 3(c)).

Figure 2. PPL inhibitory activity of 20(S)-ginsenoside Rg3. (a) Chromatograms of (A) reference substance 20(S)-ginsenoside Rg3, (B) mixture sub-fraction PN-17 including (1) ginsenoside Rh4, (2) 20(S)-ginsenoside Rg3, (3) 20(R)-ginsenoside Rg3 at 210 nm. (b) PPL inhibitory activity of three compound components of PN-17. The final concentration of three compound components was 50 μg/mL. (c) Concentration dependent inhibition of PPL by 20(S)-ginsenoside Rg3. The inhibition rate was plotted against the log of the cumulative doses of 20(S)-ginsenoside Rg3 (Mean ± SD, n = 3).

Figure 3. PPL inhibitory activity of honokiol. (a) Chromatograms of (A) reference substance honokiol, (B) sub-fraction MO-19 at 230 nm. (b) The structure of honokiol. (c) Concentration dependent inhibition of PPL by honokiol. The inhibition rate was plotted against the log of the cumulative doses of honokiol (Mean ± SD, n = 3).

Table 2. ESI mass analysis of PN-17-2 and MO-19.

				Molecular ions, m/z
Identified compounds	TR/min	Molecular formula	Molecular weight	(+)	(−)
PN-17-2	28.56	C42H72O13	784.49	807.48[M + Na]^+ 767.49[M + H-H2O]^+ 749.48[M + H-2H2O]^+	897.48[M + TFA-H]^− 829.49[M + HCOOH-H]^– 783.49[M-H]^−
MO-19	56.09	C18H18O2	266.14	267.14[M + H]^+ 284.16[M + NH4]^+ 289.12[M + Na]^+	265.14[M-H]^−

Discussion

Natural products are vastly diverse in chemical structure and biological activity. They have been an important source of lead compounds for drug discovery (Xie et al. 2012; Tsai et al. 2013; Deng et al. 2014; Zhou et al. 2014). The purpose of this study was to screen potential pancreatic lipase inhibitors derived from traditional Chinese medicinal herbs for utility in anti-obesity treatment. In this study, 35 herbs were extracted and eluted with 20%, 40% and 95% ethanol. Then their anti-lipase activity was assessed in vitro. The 95% ethanol elutions of Panax notoginseng and Magnolia officinalis exhibited strong anti-lipase activity, so these were further fractionated into 22 sub-fractions each. PN-17 and MO-19 were found to have the most potent anti-lipase effects. Then 20(S)-ginsenoside Rg3 and honokiol were found in PN-17 and MO-19, respectively.

Magnolia officinalis has been reported to contain several bioactive components, including magnolol, honokiol, 4-O-methylhonokiol, and obovatol (Lee et al. 2011). A recent study indicated that Magnolia officinalis extract and its active component, 4-O-methylhonokiol, could protect mice from high-fat diet-induced obesity by improving lipid metabolism (Zhang et al. 2014). The results of the current study showed that it is reasonable that honokiol, a lipase inhibitor, may be the anti-obesity component of Magnolia officinalis.

Reports have shown that saponins from Platycodi radix, Acanthopanax senticosus and Panax japonicus can inhibit pancreatic lipase and ameliorate the body weight gain induced by high-fat diets in mice (Han et al. 2002; Han et al. 2005; Li et al. 2007). A previous study suggested that 20(S)-ginsenoside Rg3 can inhibit lipid accumulation during 3T3-L1 adipocyte differentiation through AMPK and PPAR-γ signaling pathways (Hwang et al. 2009). The current study showed 20(S)-ginsenoside Rg3 to be the active component of Panax notoginseng against pancreatic lipase. Taken together, 20(S)-ginsenoside Rg3 might be a suitable lead compound for anti-obesity treatment and that it acts by reducing lipid absorption and accumulation (Cicero et al. 2003; Kim & Park 2003).

Two major, valuable compounds, 20(S)-ginsenoside Rg3 and honokiol, were found to have anti-lipase effects by bioassay-guided isolation. However, in this study, other compounds with strong anti-lipase activity may have been present and gone undetected. One limitation of our study is that some potent compounds in herbal elution may have been present in concentrations too low to be isolated and identified. To address this problem, it may be necessary to combine the compound library of herbal medicine and high-throughput screening.

Here, thirty-five traditional Chinese medicinal herbs were screened for anti-lipase activity. This is the first report to show 20(S)-ginsenoside Rg3 isolated from Panax notoginseng and honokiol isolated Magnolia officinalis to have anti-lipase activity. The findings of this and previous studies suggest that 20(S)-ginsenoside Rg3 and honokiol might be compounds with potent anti-obesity properties. Further investigations are in progress to evaluate the anti-obesity activity of honokiol and 20(S)-ginsenoside Rg3 in animal models.

Funding information

The authors would like to thank the National Science and Technology Major Project-Key New Drug Creation and Manufacturing Program (2013ZX09402203) for financial support.

Disclosure statement

The authors have no conflicts of interest to report. The authors claim responsibility for the content and composition of the paper.