Abstract
Lotus seed is well known as traditional food and medicine, but its skin is usually discarded. Recent studies have shown that lotus seed skin contains a high concentration of proanthocyanidins that have multi-functions, such as antioxidation, anti-inflammation, and anti-cancer effects. In the present study, we aimed to isolate and purify the proanthocyanidins from lotus seed skin by acetone extraction and rotary evaporation, identify their chemical structures by HPLC-MS-MS and NMR, and further investigate the antioxidant properties of the extract purified by macroporous resin (PMR) from lotus seed skin both in vitro and in vivo. The results showed that PMR mainly contained oligomeric proanthocyanidins, especially dimeric procyanidin B1 (PB1), procyanidin B2 and procyanidin B4. Although it had limited ability to directly scavenge radicals in vitro, PMR could significantly enhance the expressions of antioxidant proteins via activation of nuclear factor-E2-related factor 2 (Nrf2)–antioxidant response element (ARE) pathway in HepG2 cells. Molecular data revealed that PB1, a major component in PMR, stabilized Nrf2 by inhibiting the ubiquitination of Nrf2, which led to subsequent activation of the Nrf2–ARE pathway, including the enhancements of Nrf2 nuclear translocation, Nrf2–ARE binding and ARE transcriptional activity. Moreover, the in vivo results in high fat diet-induced mice further verified the powerful antioxidant property of PMR. These results revealed that lotus seed skin is a promising resource for functional food development.
Introduction
The nuclear factor-E2-related factor 2 (Nrf2) is a master redox-sensitive modulator to orchestrate the defense mechanisms of cells against excessive oxidative stress caused by obesity or other chronic diseases, by binding to the antioxidant response element (ARE) in the promoter regions of series of antioxidant proteins or detoxifying enzymes [1,2]. The major repressor of Nrf2, Kelch-like ECH-associated protein 1 (Keap1), has been reported to lose its function by chemopreventive phytochemicals. Therefore, Nrf2 can enter into the nuclear to bind to ARE for the induction of phase II antioxidant proteins and detoxifying enzymes [3].
Lotus seed is extensively used as traditional Chinese medicine and as dessert in East Asia. Lotus seed skin is by-products of the lotus processing and is generally discarded or used as feed stuff. Recent studies have revealed that lotus seed skin account for approximately 15% weight of the whole seed with a high concentration of oligomeric proanthocyanidins (OPCs) [4], which possess multiple biological functions, such as anti-inflammation [5], antioxidation [6], and anti-cancer effects [7]. OPCs are found to exhibit antioxidant properties by directly scavenging reactive oxygen species (ROS) in vitro [8]. However, relatively much lower plasma concentration of polyphenols is found in animals or human. Therefore, the beneficial effects of polyphenols have been investigated from direct antioxidant actions to indirect antioxidant actions, especially their effects on regulation of gene expression by modulating signaling pathways, such as activation of Nrf2–ARE pathway [9].
Chronic inflammation induced by excess ROS levels in the body is the crucial cause for chronic disease initiation. Phytochemicals are reported to exert chemopreventive effects on the interface against excessive oxidative stress and ROS via modulation of Nrf2 and NF-κB pathways [10]. Recently, similar actions of OPCs and other phytochemicals were reported in several studies, e.g. procyanidins and proanthocyanidins from wild grape seeds can activate Nrf2 pathway via p38 and PI3K/Akt and inhibit NF-κB via MAPK [11,12].
High fat diet (HFD)-induced obesity or diabetes causes prolonged imbalance of oxidative stress and inflammation in the body. HFD-induced mice are ideal models for the investigation of interventional action of OPCs against metabolic disorders.
In the present study, OPCs from lotus seed skin were isolated, purified and identified by macroporous resin AB-8, HPLC-MS-MS, and NMR. The direct antioxidant activities of the extract purified by macroporous resin AB-8 (PMR) or its major functional component procyanidin B1 (PB1) were investigated by total antioxidant activity (TAA), 1,1-diphenyl-2-picrylhydrazyl (DPPH) assay, and their indirect antioxidant activities were investigated in HepG2 cells, focusing on the Nrf2–ARE signaling by reporter gene assay, real-time RT-PCR, electrophoretic mobility shift assay (EMSA) and immunoprecipitation. Finally, the antioxidant properties of PMR were verified in HFD-induced mice.
Materials and Methods
Reagents and antibodies
All Solvents (HPLC grade) used in HPLC and other analytical methods were purchased from Sinopharm Chemical Reagent Inc (Beijing, China). Standard samples of (-)-gallocatechin, (-)-epicatechin and procyanidin B1/B2/B4 were purchased from Sigma-Aldrich (St Louis, USA). The anti-Nrf2 (A-10, sc-365949), anti-HO-1 (H-105, sc-10789), anti-Keap1 (G-2, sc-365626), anti-GAPDH (6C5, sc-32233) and anti-ubiquitin (A-5, sc-166553) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, USA). Horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG secondary antibodies were purchased from Cell Signaling Technology (Beverly, USA). All other reagents used in this study were of the highest grade commercially available.
Preparation of PMR
Lotus seeds were freshly harvested by the local company, Hong Xinglong Co Ltd of Xiangtan County (Xiangtan, China), in July 2016. This Nelumbo nucifera Gaertn was named as ‘Furong Lian’ and approved by Cultivar Registration Committee of Hunan Province. After being harvested, the lotus seed skin was directly extracted from the raw lotus seed using a skinning machine and ground into powder using a grinding machine. The powder was stored at −20°C before use.
The scheme of extraction, isolation, purification, and identification of PMR from lotus seed skin is shown in Fig. 1A. The preparation of PMR was performed according to our previous study [13]. In brief, the lotus seed skin powder was milled with a 60-mesh screen and extracted in 67% aqueous acetone for 90 min, with the solid-liquid ratio of 1:54 (w/w, pH 2.7, 37°C). The supernatant was collected and concentrated using a rotary evaporator at 50°C in order to remove acetone. After that, proanthocyanidins in the concentrates were isolated by 3–4 times of extraction with an equal volume of ethyl acetate. The ethyl acetate phase was pooled and the ethyl acetate was removed by rotary evaporation at 50°C to yield a dry powder. The dried powder was dissolved in 60% acetone and loaded onto a macroporous resin AB-8 chromatographic column (Guangfu Chemical Industry Institute, Tianjin, China). The macroporous resin was first eluted with 500 ml distilled water at a flow rate of 10 ml/min and then eluted with 300 ml of 60% aqueous acetone at a flow rate of 5 ml/min. The purification of macroporous resin AB-8 (PMR) was concentrated by rotary evaporation under vacuum at 50°C for the further analysis.
Direct antioxidant ability assay of PMR (A) The workflow for preparation and antioxidant activity assay of PMR from lotus seed skin. (B,C) The total antioxidant ability (TAA) and DPPH scavenging ability of PMR were determined and compared with Vc and catechin at concentrations of 0.01–0.10 mg/ml and 0.01–0.40 mg/ml, respectively. (D) The DPPH radicals scavenging ability of PMR, M1, M2, and M3. Data are expressed as the mean ± SD of three independent experiments.
Structural identification by HPLC-MS-MS and NMR
Concentrated PMR was loaded on a polyamide column, and three fractions (M1, M2, and M3) were obtained by elution with 100 ml of 10%, 30% and 50% aqueous acetone, respectively. HPLC-MS-MS analysis was conducted using an Agilent UHPLC (1290)-Q/TOF (6530) MS system (Santa Clara, USA) equipped with an Agilent Single Quadrapole Mass-Selective Detector HP 1101. The HPLC conditions were similar to those of other related studies [14,15] with slight modifications. In brief, the fraction of PMR was separated by using a C18, reversed-phase column (250 mm × 4.6 mm; Latek Labortechnik, Eppelheim, Germany). Detection wavelength was set at 280 nm and injection volume was 20 μl. The mobile phase consisted of 0.25% acetic acid in water (solvent A) and acetonitrile (solvent B) with the following gradient: 7%–20% of solvent B in 40 min, keeping 20% of solvent B for 5 min at a flow rate of 0.8 ml/min. Mass spectra in the negative-ion mode were generated under the following conditions: fragmenter voltage is 100 V, capillary voltage is 2500 V, nebulizer pressure is 40 psi, drying gas temperature is 340°C, and mass range is 100–2500 m/z. NMR spectra of 13C were recorded at 100 MHz, on a INOVA-400 spectrometer (Varian, Palo Alto, USA). Further structural confirmation of the fractions was performed by HPLC-MS-MS analysis using standard samples.
TAA and DPPH radical scavenging ability assay
TAA was determined according to a previous study [16]. In brief, 1.0 ml sample/standard (0.01–0.10 mg/ml) was mixed with 1 ml reagent solution (3 M sulfuric acid, 0.14 M sodium phosphate, and 0.02 M ammonium molybdate) and then diluted with distilled water to 5 ml. The test tubes were incubated at 95°C for 90 min, and the absorbance was measured by using a spectrophotometer at 695 nm after cooling down. The DPPH radical scavenging ability was determined as reported previously [17]. In brief, 0.1 ml sample/standard solution (0.01–0.40 mg/ml) was mixed thoroughly with 3.9 ml DPPH solution (25 μg/ml), incubated for 30 min, and the absorbance was determined at 515 nm.
Cell culture, treatment, and viability assay
HepG2 cells were purchased from ATCC (Rockville, USA) and maintained at 37°C with 5% CO2 in DMEM plus 10% FBS and PSG (100 μg/ml streptomycin, 100 units/ml penicillin and 1% glutamax). Cells were pre-cultured in dishes for 24 h and then treated with various concentrations of PMR or PB1 in 0.1% dimethyl sulfoxide (DMSO) for the indicated time periods. Then MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] assay was performed to determine the cytotoxicity of samples [18]. Briefly, cells were seeded onto 96-well plates at a density of 104 cells/well and pre-incubated at 37°C plus 5% CO2 for 24 h. Then cells were treated with different concentrations of samples for another 24 h, followed by incubation with MTT (5 mg/ml) at 37°C for an additional 4 h in the dark. Acidic isopropanol (0.04–0.1 M HCl in isopropanol) was added to dissolve the formazan crystals and the optical density (OD) was measured at 570 nm with a microplate reader (Thermo Scientific, Waltham, USA). Viability was determined by comparing the OD of sample-treated cells with those of the untreated cells.
Transient transfection and luciferase reporter gene assay
Transient transfection was performed by using pGL4-TK-encoding renilla luciferase plasmid (Promega, Madison, USA) and LipofectAMINE 2000 (Invitrogen, Carlsbad, USA) as reported previously [19]. Reporter gene assay was performed according to the manufacturer’s instruction by using Dual-Luciferase Reporter Assay System (Promega). In brief, HepG2 cells were plated into 12-well plates at a concentration of 1 × 105/well and pre-cultured for 24 h in DMEM plus 10% FBS. Cells were then transfected with ARE promoter-encoding firefly luciferase plasmid and pGL4-TK-encoding renilla luciferase plasmid. After 24 h of incubation, the cells were treated with samples in 0.1% DMSO, or 0.1% DMSO alone as a control, and further incubated for 24 h. The activities of firefly and renilla luciferase were measured in ARVOTMSX multilabel counter (Perkin Elmer, Santa Clara, USA). Luciferase activity values were normalized to transfection efficiency monitored by renilla expression, and ARE transcription activity was expressed as fold induction relative to the control cells.
Total RNA extraction and real-time RT-PCR
HepG2 cells (1 × 106) were pre-cultured in 10-cm dishes for 24 h and then treated with various concentrations of samples in 0.1% DMSO, or with 0.1% DMSO alone as a control, for 9 h. Total RNA was extracted with an Isogen RNA kit (Nippon Gene Co. Toyama, Japan) as described in the manufacturer’s manual. All primers were designed with the software PRIMER3 and synthesized as follows: HO-1, forward (5′-CCAGCGGGCCAGCAACAAAGTGC-3′) and reverse (5′-AAGCCTTCAGTGCCCACGGTAAGG-3′); Nrf2, forward (5′-AGACAAACATTCAAGCCGCT-3′) and reverse (5′-CCATCTCTTGTTTGCTGCAG-3′); Keap1, forward (5′-CCTTCAGCTACACCCTGGAG-3′) and reverse (5′-AACATGGCCTTGAAGACAGG-3′). Reverse transcription and real-time PCR were performed with DyNAmo™ SYBR® Green 2-Step qRT-PCR Kit (Finnzymes Oy, Espoo, Finland) according to the manufacturer’s manual. Briefly, RNA (200 ng) was reverse-transcribed to cDNA using Oligo dT and M-MuLV RNase at 37°C for 30 min, and the reaction was then terminated at 85°C for 5 min. The sequences of PCR primers and other reaction conditions used in the present study were described previously [19]. The results were presented as the relative expression levels normalized with those in control cells.
Immunoprecipitation and western blot analysis
For western blot analysis, HepG2 cells (3 × 106) were pre-cultured in 10-cm dishes for 24 h, and treated with various concentrations of samples for the indicated time periods. After that, cells were lysed with modified RIPA buffer and proteins quantification was performed using a protein assay kit (Bio-Rad Laboratories, Hercules, USA) as described previously [19]. For immunoprecipitation, whole-cell lysates containing 0.5 mg of proteins were precleared with protein A-Sepharose beads (Amersham Pharmacia Biotech, Buckinghamshire, UK) for 1 h and incubated with 1 μg of anti-Nrf2 or anti-Keap1 antibody for 4 h. Immunoprecipitated complexes were washed five times with RIPA buffer and then boiled in SDS sample buffer for 5 min. Either the immunoprecipitation products or the whole-cell lysates were subject to SDS-PAGE and electrophoretically transferred to PVDF membrane (Amersham Pharmacia Biotech). After incubation with primary and secondary antibodies, the membrane was detected using an ECL system, and the relative amounts of specific proteins were quantified using Lumi Vision Image software (TAITEC, Saitama, Japan).
Cell fractionation and EMSA
The nuclear and cytosolic proteins were prepared according to the modified method as described previously [19]. Protein concentration was determined using a protein assay kit (Bio-Rad Laboratories) according to the manufacturer’s instructions. In vitro protein-DNA interaction was examined using Lightshift® Chemiluminescent EMSA kit (Thermo Scientific) as described previously [20]. ARE-specific EMSA DNA probe was synthesized, gel purified and 5′-biotin labeled or unlabeled (5′-TTTTATGCTGTGTCATGGTT-3′). The biotin-labeled complexes were detected by chemiluminescence (TAITEC).
Animal experiment design
Male C57BL/6 mice (20 ± 1.6 g) used in the study were obtained from Hunan SJA Laboratory Animal Co., Ltd. The mice were maintained in a room at 23 ± 2°C, 55%–60% humidity, with 12 h light/12 h dark cycle, and provided with indicated diets and water ad libitum for about 1 week prior to the experiments. All animal care and experimental procedures followed the Guide for the Care and Use of Laboratory Animals (NIH publication No. 85–23, 1985 revised) and the protocols were approved by the Animal Care and Use Committee of Hunan Agricultural University. Mice (n = 30) were randomly assigned to five groups (n = 6 for each group) and treated as follows: group 1 (control), animals received a standard diet (STD) as vehicle; group 2 (high fat diet, HFD), animals received a high fat diet as the positive control; group 3 (HFD + low dose PMR), animals received HFD with a low dose of PMR (100 mg/kg diet); group 4 (HFD + high dose PMR), animals received HFD with a high dose of PMR (200 mg/kg diet); and group 5 (STD + high dose PMR), animals received STD with a high dose of PMR (200 mg/kg diet). After 8 weeks, all mice were anesthetized with diethyl ether following a 12-h fasting and then sacrificed. Samples of serum, liver, and other organs were immediately collected, weighed and stored at −80°C before use.
Hepatic antioxidant properties assay in mice
The hepatic antioxidant properties, including glutathione peroxidase activity (GPx), SOD activity and relative lipid malondialdehyde concentration (MDA) were measured using the corresponding assay kits (Nanjing Jiancheng BioEng, Nanjing, China).
Statistical analysis
Data were expressed as the means ± SD of at least three independent experiments. Statistical comparisons were evaluated using SPSS 11.0. Student’s t-tests were performed to compare the mean of two groups and selected data sets were analyzed by ANOVA followed by Tukey’s test. P < 0.05 was considered as significant.
Results
Preparation of PMR from lotus seed skin and structural identification of its major components
A total of 9 proanthocyanidin fractions (F1–F9) were identified according to their retention times (RT) using the analytical HPLC, and some of which were further confirmed and identified by HPLC-MS-MS and NMR. As shown in Supplementary Fig S1A, nine compounds, F1–F9, were separated by analytical HPLC. The further HPLC-MS-MS data in Table 1 revealed that, compounds F2, F3, F6, and F8 (m/z 577) yielded two main fragments, m/z 407 and m/z 289 (Supplementary Fig. S1B); F5 contained five fragments, including m/z 1001, 865, 695, 575, and 287; F9 was consisted of five pseudo molecular ions at m/z 865, 739, 575, 423, and 287. The quantity of each constituent of PMR was summarized in Table 1, catechin as the reference standard. Thus, the major proanthocyanidins quantified from Lotus seed skins were procyanidin B1 (33%); (+)-catechin (30.2%); procyanidin B4 (25.1%); procyanidin B2 (4.5%) and (+)-gallocatechin (1.6%) according to a series of previous studies [21–28].
Total procyanidin composition and HPLC-MS-MS analysis data of PMR
| Peak No. | RT | HPLC-MS | HPLC-MS-MS | Area ratio | Formula | Compound |
|---|---|---|---|---|---|---|
| m/z | m/z | (%) | ||||
| F1 | 10.5 | 305.1 | / | 1.6 | C15H14O7 | (+)-gallocatechin |
| F2 | 17.1 | 577.1 | 407, 289 | 33.0 | C30H26O12 | epicatechin-(4β→8)-catechin (procyanidin B1) |
| F3 | 19.2 | 577.1 | 407, 289 | 25.1 | C30H26O12 | catechin-(4α→8)-epicatechin (procyanidin B4) |
| F4 | 21.5 | 289.1 | / | 30.2 | C15H14O6 | (+)-catechin |
| F5 | 22.9 | 1153.2 | 1001, 865, 695, 575, 287 | C60H50O24 | procyanidin tetramer | |
| F6 | 25.9 | 577.0 | 407, 289 | 4.5 | C30H26O12 | epicatechin-(4β→8)-epicatechin (procyanidin B2) |
| F7 | 28.3 | 289.1 | / | C15H14O6 | (-)-epicatechin | |
| F8 | 30.6 | 577.0 | 407, 289 | C30H26O12 | procyanidin dimer | |
| F9 | 35.8 | 1153.2 | 865, 739, 575, 423, 287 | C60H50O24 | procyanidin tetramer |
| Peak No. | RT | HPLC-MS | HPLC-MS-MS | Area ratio | Formula | Compound |
|---|---|---|---|---|---|---|
| m/z | m/z | (%) | ||||
| F1 | 10.5 | 305.1 | / | 1.6 | C15H14O7 | (+)-gallocatechin |
| F2 | 17.1 | 577.1 | 407, 289 | 33.0 | C30H26O12 | epicatechin-(4β→8)-catechin (procyanidin B1) |
| F3 | 19.2 | 577.1 | 407, 289 | 25.1 | C30H26O12 | catechin-(4α→8)-epicatechin (procyanidin B4) |
| F4 | 21.5 | 289.1 | / | 30.2 | C15H14O6 | (+)-catechin |
| F5 | 22.9 | 1153.2 | 1001, 865, 695, 575, 287 | C60H50O24 | procyanidin tetramer | |
| F6 | 25.9 | 577.0 | 407, 289 | 4.5 | C30H26O12 | epicatechin-(4β→8)-epicatechin (procyanidin B2) |
| F7 | 28.3 | 289.1 | / | C15H14O6 | (-)-epicatechin | |
| F8 | 30.6 | 577.0 | 407, 289 | C30H26O12 | procyanidin dimer | |
| F9 | 35.8 | 1153.2 | 865, 739, 575, 423, 287 | C60H50O24 | procyanidin tetramer |
Total procyanidin composition and HPLC-MS-MS analysis data of PMR
| Peak No. | RT | HPLC-MS | HPLC-MS-MS | Area ratio | Formula | Compound |
|---|---|---|---|---|---|---|
| m/z | m/z | (%) | ||||
| F1 | 10.5 | 305.1 | / | 1.6 | C15H14O7 | (+)-gallocatechin |
| F2 | 17.1 | 577.1 | 407, 289 | 33.0 | C30H26O12 | epicatechin-(4β→8)-catechin (procyanidin B1) |
| F3 | 19.2 | 577.1 | 407, 289 | 25.1 | C30H26O12 | catechin-(4α→8)-epicatechin (procyanidin B4) |
| F4 | 21.5 | 289.1 | / | 30.2 | C15H14O6 | (+)-catechin |
| F5 | 22.9 | 1153.2 | 1001, 865, 695, 575, 287 | C60H50O24 | procyanidin tetramer | |
| F6 | 25.9 | 577.0 | 407, 289 | 4.5 | C30H26O12 | epicatechin-(4β→8)-epicatechin (procyanidin B2) |
| F7 | 28.3 | 289.1 | / | C15H14O6 | (-)-epicatechin | |
| F8 | 30.6 | 577.0 | 407, 289 | C30H26O12 | procyanidin dimer | |
| F9 | 35.8 | 1153.2 | 865, 739, 575, 423, 287 | C60H50O24 | procyanidin tetramer |
| Peak No. | RT | HPLC-MS | HPLC-MS-MS | Area ratio | Formula | Compound |
|---|---|---|---|---|---|---|
| m/z | m/z | (%) | ||||
| F1 | 10.5 | 305.1 | / | 1.6 | C15H14O7 | (+)-gallocatechin |
| F2 | 17.1 | 577.1 | 407, 289 | 33.0 | C30H26O12 | epicatechin-(4β→8)-catechin (procyanidin B1) |
| F3 | 19.2 | 577.1 | 407, 289 | 25.1 | C30H26O12 | catechin-(4α→8)-epicatechin (procyanidin B4) |
| F4 | 21.5 | 289.1 | / | 30.2 | C15H14O6 | (+)-catechin |
| F5 | 22.9 | 1153.2 | 1001, 865, 695, 575, 287 | C60H50O24 | procyanidin tetramer | |
| F6 | 25.9 | 577.0 | 407, 289 | 4.5 | C30H26O12 | epicatechin-(4β→8)-epicatechin (procyanidin B2) |
| F7 | 28.3 | 289.1 | / | C15H14O6 | (-)-epicatechin | |
| F8 | 30.6 | 577.0 | 407, 289 | C30H26O12 | procyanidin dimer | |
| F9 | 35.8 | 1153.2 | 865, 739, 575, 423, 287 | C60H50O24 | procyanidin tetramer |
The major compound F2 was further purified by semi-preparative HPLC and identified by 13C-NMR spectra (100 MHz, DMSO-d6), as shown in Supplementary Fig. S2: Upper unit: δ75.701 (C-2), δ71.444 (C-3), δ35.558 (C-4), δ102.417 (C-4a), δ156.002 (C-5), δ95.933 (C-6), δ156.841 (C-7), δ93.964 (C-8), δ156.841 (C-8a), δ131.620 (C-1′), δ115.104 (C-2′), δ144.841 (C-3′), δ144.444(C-4′), δ115.104 (C-5′), δ117.995 (C-6′); Lower unit: 78.884 (C-2), δ66.303 (C-3), δ27.353 (C-4), δ99.192 (C-4a), δ154.705 (C-5), δ94.773 (C-6), δ154.705 (C-7), δ107.399 (C-8), δ153.919 (C-8a), δ131.619 (C-1′), δ115.104 (C-2′), δ144.620 (C-3′), δ144.444(C-4′), δ113.929 (C-5′), and δ117.995 (C-6′). Thus, F2 was identified as PB1. The above results were similar to those of lotus seedpod, in which the base units are also catechin and epicatechin, the proanthocyanidins mainly included monomers, dimers, and tetramers of procyanidins, and the major constitutes were dimers [29].
The direct antioxidant activity of PMR
The direct antioxidant activity was estimated by TAA and DPPH assays. As shown in Fig. 1B,C, all the TAA values and DPPH radical scavenging abilities of PMR, catechin and Vc increased dose-dependently with the concentration range of 0.01–0.10 mg/ml and 0.01–0.40 mg/ml, respectively. The direct antioxidant activities of PMR were lower than catechin but higher than Vc, which indicates that the extracts from lotus seed skin have limited direct antioxidant activity. DPPH radical scavenging assay was further performed for the extracts M1, M2, and M3. As shown in Fig. 1D, the DPPH radical scavenging ability increased in the extract isolated with higher concentration of acetone.
The indirect antioxidant activity of PMR
It has been found that the in vivo antioxidant activity of polyphenols is due to their enhancement effects on the expression of a series of genes including phase II antioxidant or detoxifying enzymes and proteins, rather than their direct antioxidant activities [30]. Moreover, the Nrf2–ARE pathway is considered to be a master system for the indirect antioxidant activity of polyphenols [31]. Thus, we used PMR and its major bioactive compound, PB1, to investigate whether they activate the Nrf2–ARE pathway. First, MTT assay was used to determine the sample concentration with no cell toxicity to cells. As shown in Fig. 2A, there was no significant effect on the growth of HepG2 cells treated with less than 100 μg/ml of PMR or 80 μM of PB1 for 48 h. Thus, 100 μg/ml of PMR and 20 μM of PB1 were used in the subsequent experiments. As shown in Fig. 2B, PMR and PB1 significantly induced the transcription activity of ARE at the indicated concentrations. Myricetin was used as a positive control [19]. Other main constituents of PMR had no significant effect in the same treatment (data not shown). These results indicated that PB1 is the major responsible constituent for the indirect antioxidant activity of PMR in HepG2 cells.
The indirect antioxidant property of PMR and PB1 (A) The cytotoxicity assay of PMR and PB1 in HepG2 cells. (B) Induction of ARE-reporter gene activity by PMR and PB1 in HepG2 cells. (C–E) The induction of Nrf2, Keap1 and HO-1 by PMR and PB1 in HepG2 cells. Data are expressed as the mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 versus control, respectively.
To examine whether PMR and PB1 induce protein expressions of antioxidant Nrf2 and HO-1 by the Nrf2–ARE pathway, we next detect these proteins by western blot analysis. As shown in Fig. 2C–E, treatment with 25–50 μg/ml PMR (Fig. 2C) or 10–20 μM PB1 (Fig. 2D) for 8 h increased the expressions of both Nrf2 and HO-1, but no effect on Keap1 expression. A detail time course from 2 to 48 h was further performed for PB1 treatment (Fig. 2E), PB1 treatment increased Nrf2 expression from 2 to 24 h and HO-1 expression from 4 to 48 h, but had no effect on Keap1 expression.
Molecular mechanism of PB1-induced Nrf2 activation
To further elucidate the molecular mechanism, we performed advanced cellular biochemical assays. Figure 2 showed that PB1 treatment increased the cellular level of Nrf2, but not the Keap1 level. Subsequently, RT-PCR was performed to investigate whether these events also occur at transcriptional level. As shown in Fig. 3A, treatment with 10–20 μM of PB1 did not significantly increase the mRNA levels of Nrf2 and Keap1, but significantly increased the mRNA level of HO-1, a downstream enzyme of Nrf2. These results suggested that PB1-induced expression of HO-1 is not caused by the transcription regulation on Nrf2 or Keap1.
Effects of PB1 on transcription and modification of Nrf2 and Keap1 in HepG2 cells (A) The mRNA levels of typical genes in Nrf2–ARE signaling pathway. (B) The ubiquitination status of Nrf2 and Keap1. (C) PB1 improved the post-transcriptional stability of Nrf2 protein. Data are expressed as the mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 versus control, respectively.
Next, immunoprecipitation was used to investigate the ubiquitination status of Nrf2 and Keap1. As shown in Fig. 3B, PB1 significantly inhibited the ubiquitination of Nrf2, but not the ubiquitination of Keap1. In addition, we examined the stability of Nrf2 protein after stopping the translation by adding the protein expression inhibitor CHX, and the results showed that PB1 increased the stability of Nrf2 protein with extension of the half life time (t1/2) from 19.3 to 38.7 min (Fig. 3C). These results indicated that PB1 increased Nrf2 protein by inhibiting the turnover of Nrf2 at the post-transcriptional level. The nuclear translocation and ARE-binding activity induction of PB1-stabilized Nrf2 protein in cellular fractionation were confirmed by EMSA assay. As shown in Fig. 4, treatment with PB1 significantly enhanced the nuclear translocation of Nrf2 from cytoplasm into the nucleus and enhanced the ARE-Nrf2 binding ability.
The molecular mechanism of Nrf2 activation by PB1 in HepG2 cells (A) PB1 increases Nrf2 nuclear translocation. (B) PB1 enhances the Nrf2–ARE binding ability. Data are expressed as the mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 versus control, respectively.
In vivo antioxidant properties of PMR
Animal studies were performed to confirm the efficacy of lotus seed skin extract. To verify the antioxidant effects of PMR in HFD-induced mice, we firstly detected the concentrations of major antioxidant parameters, including relative MDA concentration, GSH-Px activity, and SOD activity. As shown in Fig. 5A, HFD significantly increased relative MDA concentration and decreased GSH-Px and SOD activities, while intake of PMR restored these changes by the inhibition of relative MDA concentration and induction of the activities of GSH-Px and SOD, in a dose-dependent manner, although low concentration of PMR (1%) had no significant effect on GSH-Px and SOD activities.
In vivoantioxidant properties of PMR (A) Effects of PMR on hepatic antioxidant property in mice. (B) Effects of PMR on the hepatic antioxidant proteins in mice. Data are expressed as the mean ± SD of three independent experiments. *P < 0.05, **P < 0.01 versus control, respectively.
Next, we performed western blot analysis to detect the expressions of Nrf2 and its downstream antioxidant proteins in tissues of mice. As shown in Fig. 5B, HFD decreased the protein expressions of Nrf2, HO-1, and NQO1 effectively; HFD plus 2% PMR treatment not only restored their expressions, but also further significantly increased the protein expressions by 1.8, 1.7 and 1.7 folds, respectively. These results revealed that PMR activated the Nrf2–ARE pathway in mouse liver, and that is probably the mechanism underlying the protective effect of PMR in HFD-induced mice.
Discussion
In the present study, we isolated PMR from lotus seed skin, separated the fractions of its major components, and finally identified them. They were procyanidin dimers with a total content of 62.6% (m/m), as shown in Table 1. Further molecular data indicated that PB1 in PMR exerted powerful antioxidant effect via indirect activation of the Nrf2–ARE pathway, rather than direct action of scavenging ROS. The in vivo data verified its antioxidant property in HFD-induced fat mice.
Little attention has been paid to the direct antioxidant property of polyphenol, such as scavenging of ROS and chelating of metal ions because plant polyphenol supplementation shows limited effect on antioxidant capacity in healthy animals, which is probably caused by the lack of mechanistic investigation [32,33]. Instead, indirect antioxidant property of polyphenols is catching more eyes, such as the regulation of gene expressions, modulation of cell signaling pathways, and epigenetic modifications [34,35]. In the present study, we used DPPH method to detect the direct antioxidant ability of PB1. As was expected, the results showed that the ROS scavenging ability of PB1 is limited and the indirect antioxidant property of PB1 by activation of the Nrf2–ARE pathway is more powerful. In addition, we also detect other direct antioxidant property (such as hydroxyl radical scavenging ability and superoxide anion scavenging ability) with the same results as DPPH, as shown in our previous study [36].
Although, there are five major constituents in PMR, the DPPH scavenging ability of M1, M2, and M3 differs only depending on the concentration of F2 (PB1) in each fraction (Fig. 2D). That is, the fraction with more powerful DPPH scavenging ability has higher concentration of PB1, by comparing the HPLC data of M1, M2, and M3 in a previous study [28]. Moreover, the results from HepG2 cell experiments revealed that PB2 had no significant effect on the expressions of Nrf2 and HO-1 (Supplementary Fig. S3). These data indicated that PB1 is the crucial constituent in PMR for its antioxidant property. Therefore, we chose PB1 for the molecular mechanism study in cell models.
The mechanism of polyphenol-induced Nrf2 activation can be classified into Keap1-dependent and Keap1-independent [37]. The molecular data indicated that the ubiquitination inhibition and stabilization enhancement of Nrf2 by PB1 are the main mechanisms of PB1-stimulated Nrf2 translocation, Nrf2–ARE binding and induction of its downstream antioxidant proteins. Thus, PB1 activates Nrf2 pathway mainly via Keap1-independent mechanism. Finally, we constructed the HFD-induced obese mice model, and the in vivo data verified the antioxidant properties of PMR, which revealed that PB1 restored the redox balance disturbed by HFD treatment in mice. We for the first time found that PB1 is the primary substance in PMR that shows a potent antioxidant activity both in vitro and in vivo, through the activation of the Nrf2–ARE pathway.
Although the biological properties of procyanidins have been extensively studied and proved at in vitro level, the in vivo bioavailability and efficacy are less investigated and confirmed [38,39]. Especially, the relatively much lower level of serum procyanidins compared with their concentration in cell model weakens the functional reputation of procyanidins, as well as other dietary polyphenols [40,41]. Another reason is the complication of the crosstalk between Nrf2 and inflammatory pathways, including NF-κB and AP-1, have attracted more eyes in the field, which may finally lead to the clinical application of phytochemicals in prevention or treatment of chronic diseases, especially proanthocyanidins [42,43]. The most possible mechanism underlying the antioxidant and anti-inflammatory effects of phytochemical is that Nrf2-mediated HO-1 induction plays a central role in the crosstalk between Nrf2 and inflammatory pathways [44]. In the present study, we also found the inhibitory effect of PB1 on inflammation, as shown in Supplementary Fig. S4. Therefore, this crosstalk needs to be further investigated, especially the full dynamics of this interaction, before its clinical application in humans.
Recent in vivo studies have shown that PB1 can be rapidly absorbed by the human body, indicated by its level detected in blood plasma and urine [45]. Another human randomized double-blind and crossover study with six volunteers found that the pharmacokinetic parameters of PB1 were 1.6 ± 0.3 ng/ml (Cmax) and 10 ± 5 ng·h/ml (AUC), after intake of a single dose of 1 mg/kg body weight PB1 in the form of gelatin capsule [46]. On the other hand, several other studies using in vitro cell models (human monocytes and THP1 cells) showed that PB1 (30–150 μM) exerted significant anti-inflammatory effects via the ERK/IKKβ pathway and p38MAPK-NF-κB pathway [47,48]. Although the reasons for the concentration gap between in vitro and in vivo levels remain obscured, a recent review article pointed out a possible explanation that polyphenols could be extensively biotransformed by classic detoxification pathways such as methylation, glucuronidation, and sulphation, to a large number of hydrophilic-conjugated metabolites [32]. The methylation of PB1 by gut microbiota was reported by another study [46], however, the quantity and detection of most conjugates of dietary polyphenols are hard to track, and that may be why polyphenols exert bioactivity with a higher concentration in in vitro cell model but with a relatively lower concentration in in vivo experiments. Moreover, our group also performed animal experiment in mouse model by i.p. injection with 100 mg PB1/kg mouse weight, and the results showed that PB1 significantly increased the expressions of Nrf2 and HO-1 in mouse liver, as shown in Supplementary Fig. S5. Therefore, the data from our lab and other groups indicated that PB1 could reach liver cells after dietary intake, however the conjugates and biotransformed metabolites of PB1 need to be clarified in future studies.
In summary, we found that the major components with potent antioxidant property in lotus seed skin extract are dimers of procyanidin, especially PB1. They function via indirect activation of the Nrf2–ARE signaling pathway, both in vitro and in vivo. Therefore, lotus seed skin may be a suitable source for antioxidant functional food development and discovery.
Acknowledgments
The authors would like to thank Key Laboratory for Food Science and Biotechnology of Hunan Province, College of Food Science and Technology, Hunan Agricultural University for providing the facilities to carry out this study.
Funding
This work was partially supported by the grants from the National Natural Science Foundation of China (No. 31101268) and Core Research Program 1515 of Hunan Agricultural University.
References
Author notes
Tao Li and Qili Li contributed equally to this work.
