Antidiabetic Stilbenes from Peony Seeds with PTP1B, α-Glucosidase

Jun 4, 2019 - PDF (3 MB). Supporting Info .... Peony seeds (10 kg) were pulverized and extracted with petroleum ether (PE) to remove the oil. The resi...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/JAFC

Cite This: J. Agric. Food Chem. 2019, 67, 6765−6772

Antidiabetic Stilbenes from Peony Seeds with PTP1B, α‑Glucosidase, and DPPIV Inhibitory Activities Chen-Chen Zhang,†,‡,§ Chang-An Geng,*,†,‡ Xiao-Yan Huang,†,‡ Xue-Mei Zhang,†,‡ and Ji-Jun Chen*,†,‡,§ †

Downloaded via UNIV OF SOUTHERN INDIANA on July 20, 2019 at 10:14:33 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

State Key Laboratory of Phytochemistry and Plant Resources in West China, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650201, P. R. China ‡ Yunnan Key Laboratory of Natural Medicinal Chemistry, Kunming 650201, P. R. China § University of Chinese Academy of Sciences, Beijing 100049, P. R. China S Supporting Information *

ABSTRACT: One unusual resveratrol tetramer, paeonilactiflorol (1), and 14 known compounds (2−15) were isolated from peony seeds (Paeonia lactiflora) under the guidance of bioassay. Paeonilactiflorol (1) was determined by extensive HRESIMS, UV, IR, 1D and 2D NMR spectroscopic analyses. Most of the stilbenes showed obvious inhibition on PTP1B and αglucosidase, superior to the monoterpene glycosides. Especially, the stilbene tetramer (1) and trimer (8) exhibited high activity inhibiting both PTP1B with IC50 values of 27.23 and 27.81 μM and α-glucosidase with IC50 values of 13.57 and 14.39 μM. Two trans-dimers (4 and 5) also showed dipeptidyl peptidase-4 (DPPIV) inhibitory activity (55.35% and 61.26%, 500 μM) in addition to PTP1B and α-glucosidase. Enzyme kinetic study indicated that the types of inhibition on PTP1B were noncompetitive for 3 and 5 and mixed for 8 and 10. Quantitative analysis suggested that the stilbene trimers 8 (23.17 ± 0.36 mg/g) and 10 (15.24 ± 0.25 mg/g) were the main contents in peony seeds and should be responsible for the antidiabetic effects. This investigation supports the therapeutic potential of peony seeds in the treatment of diabetes with stilbenes as the active constituents. KEYWORDS: peony seeds, Paeonia lactif lora, paeonilactiflorol, stilbenes, antidiabetic activity

■

INTRODUCTION Diabetes mellitus (DM) is a group of metabolic disorders characterized by the high levels of blood glucose (hyperglycemia) and has become a major public health concern. It is estimated that 425 million people are suffering from diabetes, of which type 2 diabetes accounts for about 90%.1 According to the World Health Organization (WHO), diabetes will reach the seventh cause of death by 2030. Although a number of antidiabetic drugs with diverse mechanism of action are commercially available on the market, inevitable side effects including weight gain, hypoglycemia, gastrointestinal reactions, metabolic and nutritional disorders, liver damage, and allergic reactions hinder their application.2 Thus, new antidiabetic agents with multiple targets and slight side effects are further needed. With the development of diabetic pharmacology, many receptors involving protein tyrosine phosphatase 1B (PTP1B),2 glucogen-like peptide-1 (GLP-1),3 dipeptidyl peptidase-4 (DPPIV),4 sodium glucose cotransporter-2 (SGLT-2),5 and α-glucosidase6 are considered as promising antidiabetic targets. Especially, PTP1B that can negatively regulate the insulin and leptin signal pathway is a critical target for the treatment of diabetes.7 Currently, many kinds of PTP1B inhibitors involving phenolics, terpenoids, steroids, alkaloids, and fatty acids have been reported, providing valuable clues for the search of novel antidiabetic drugs.8 Peony seeds have received increasing attention for the production of peony seed oil which has been approved as new food resource by the Ministry of Health of China.9 A recent © 2019 American Chemical Society

study manifested that peony seed oil had the potency to alleviate postprandial hyperglycemia in streptozotocin-induced diabetic mice.10 There are about 90 000 tons of peony seeds consumed for oil extraction every year in China, which produces 50 000 tons of seed shells and cakes as underutilized byproducts except for fertilizers.11 Therefore, it is particularly important to explore new applications for peony seeds after oil extraction. In order to verify the antidiabetic potency of peony seeds and the active constituents, three types of diabetes-related proteins, PTP1B, α-glucosidase, and DPPIV, were applied to assess the antidiabetic effects of different fractions and compounds from peony seeds. As a result, one new resveratrol tetramer, paeonilactiflorol (1), and 14 known compounds were obtained from the seeds of Paeonia lactiflora. Herein, we report their isolation, structural elucidation and antidiabetic effects on PTP1B, α-glucosidase, and DPPIV.

■

MATERIALS AND METHODS

General Experimental Procedures. IR (KBr) spectrum was obtained on a Nicolet iS10 spectrometer (Thermo Fisher, California, U.S.A.). UV spectra were determined by a UV2401PC spectrophotometer (Shimadzu, Kyoto, Japan). Optical rotations were measured Received: Revised: Accepted: Published: 6765

February 20, 2019 May 8, 2019 June 4, 2019 June 4, 2019 DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry

Paeonilactiflorol (1). Brown powder; [α]21D + 81 (c 0.1, MeOH); UV (MeOH) λmax(log ε): 221 (3.23), 279 (2.39) nm; IR (KBr) νmax3443, 1632, 1515, 1457, 1384, 1343, 1312, 1249, 1180, 1163, 1131, 1112, 1098, 1078, 1014, 874, 840 cm−1; 1H and 13C NMR spectral data see Tables 1 and S1; (−) HRESIMS m/z 879.2801 [M − H]− (calcd. for C55H43O11, − 1.14 ppm), (+) HRESIMS m/z 881.2948 [M + H]+ (calcd. for C55H45O11, − 0.91 ppm).

on an Autopol VI automatic polarimeter (Rudolph Research Analytical, Hackettstown, NJ, U.S.A.). 1D and 2D NMR spectra were performed on an Advance III-600 spectrometer (Bruker, Bremerhaven, Germany). Thin-layer chromatography (TLC) analyses were performed on silica gel GF254 plates (Yantai Jiangyou Silicon Development Company, Yantai, China), and spots were detected under UV light or by heating after spraying with 10% H2SO4 in EtOH. Chromatographic silica gel (200 to ∼300 mesh) was purchased from Qingdao Makall Chemical Company (Makall, Qingdao, China). Sephadex LH-20 gel (20 to ∼50 μm) for chromatography was bought from Pharmacia Fine Chemical Co., Ltd. (Pharmacia, Uppsala, Sweden). HPLC purification was achieved on a Shimadzu HPLC system equipped with LC-20AR pumps and a model SPD-M20A UV detector (Shimadzu, Kyoto, Japan). HRESIMS data were recorded on a LCMS-IT-TOF mass spectrometer (Shimadzu, Kyoto, Japan). Chemicals. α-Glucosidase enzyme (Saccharomyces cerevisiae) was purchased from Sigma Aldrich (St. Louis, MO, U.S.A.). Recombinant human DPP4 (10688-HNCH) and human PTP1B (10304-H07E) were purchased from Sino Biological (Wayne, PA, U.S.A.). 3-(NMorpholino) propanesulfonic acid (MOPS), N,N,N′,N′-ethylenediaminetetraacetate (EDTA) and p-nitrophenyl phosphate (p-NPP) were purchased from Solarbio (Beijing, China). Dithiothreitol (DTT), Gly-Pro p-nitroanilide (Gly-Pro-PNA), and hydrochlorideand tris(hydroxymethyl)-aminomethane (Tris) were purchased from Meilunbio (Dalian, China). Bovine serum albumin (BSA) was purchased from Amrecso (Ohio, USA). p-Nitrophenyl α-D-glucopyranoside (p-NPG) was purchased from Yuanye Bio-Technology (Shanghai, China). Plant Material. The seeds of Paeonia lactif lora Pall. were purchased from Bozhou, Anhui Province of China, in July 2017, which were authenticated by Dr. Li-Gong Lei (Kunming Institute of Botany, Chinese Academy of Sciences). A voucher specimen (PLS2017-010) was deposited in the Laboratory of Antivirus and Natural Medicinal Chemistry, Kunming Institute of Botany, CAS. Extraction and Isolation. Peony seeds (10 kg) were pulverized and extracted with petroleum ether (PE) to remove the oil. The residue was extracted with 70% EtOH at room temperature for two times to afford the total extraction. The combined extraction (2.8 kg) was fractionated on a silica gel column chromatography (Si CC, 11 kg, 22 × 80 cm) and eluted with CHCl3-MeOH-H2O gradient (90:10:1, 80:20:2, 70:30:3, 60:40:4, v/v) to yield four fractions, Frs. 1−4. Based on the bioassay, the most active Frs. 1 and 2 were chosen for the following study. Fr. 1 (677 g) was chromatographed on a silica gel column (7 kg, 22 × 45 cm) and eluted with acetone-PE gradient (30:70, 40:60, 50:50 and acetone) to give four fractions. Purification of Fr. 1-1 over silica gel column (EtOAc-CHCl3) gave rise to compound 2 (1.5 g). Fr. 1-2 was separated by silica gel CC (MeOHCHCl3) to yield three fractions, Frs. 1-2-1 to 1-2-3. Fr. 1-2-1 (27 mg) was purified by HPLC on an Agilent Zorbax SB-C18 column (5 μm, 10 × 250 mm) using the mobile phase of MeOH-H2O (48:52) to provide compound 6 (12 mg). Similarly, HPLC purification of Fr. 12-2 (220 mg) with the elution of MeOH-H2O (45:55) to give compounds 3 (50 mg) and 4 (15 mg). Compound 5 (36 mg) was obtained from Fr. 1-2-3 (48 mg) by HPLC purification using the elution of MeOH-H2O (52:48). Fr. 1-3 (200 mg) was purified by HPLC with MeOH-H2O (48:52) as the mobile phase to yield compounds 7 (3 mg) and 8 (23 mg). A portion (100 g) of Fr. 2 (1.1 kg) was initially separated by a silica gel column (1 kg, 16 × 25 cm) eluted with EtOAc-MeOH gradient (100:0, 90:10, 80:20, 0:100) to afford seven fractions. Fr. 2-1 (1.1 g) was divided into four fractions by RP-C18 CC eluted with MeOH-H2O gradient (5:95, 10:90, 20:80, 30:70, 50:50, MeOH). After HPLC purification, compound 9 (22 mg) was obtained from Fr. 2-1-1, and compounds 10 (421 mg) and 11 (5 mg) were obtained from Fr. 2-1-2, respectively. Compound 1(3 mg, yield 3.3 × 10−6) was isolated from Fr. 2-2 after Sephadex LH-20 CC (MeOH) and HPLC purification (MeCN-H2O, 30:70). Fr. 2-3 was purified by repeated Si CC (CHCl3-MeOH) and Sephadex LH20 CC (MeOH) to give compounds 12 (35 mg), 13 (27 mg), 14 (12 mg), and 15 (2.7 mg).

Table 1. 1H (600 MHz) and 13C (150 MHz) NMR Data of Compound 1 in Acetone-d6. position 1a 2a (6a) 3a (5a) 4a 7a 8a 9a 10a 11a 12a 13a 14a 1b 2b (6b) 3b (5b) 4b 7b 8b 9b 10b 11b 12b 13b 14b

δH 7.64 (d, 8.6) 6.95 (d, 8.6) 5.96 (d, 11.5) 5.26 (d, 11.5)

6.09 (d, 2.1) 6.27 (d, 2.1) 6.92 (brs) 6.62 (brs) 4.42 (d, 11.7) 4.17 (m)

6.29 (s)

δC

position

130.7 130.6

1c 2c (6c)

δH 6.21 (d, 8.7)

134.3 129.0

116.3

3c (5c)

6.25 (d, 8.7)

115.2

158.8 90.6

4c 7c

3.98 (d, 6.1)

156.5 57.1

48.5

8c

3.75 (s)

61.5

142.2 121.7 158.4 103.7

9c 10c 11c 12c

157.4 105.3

5.85 (s)

δC

163.2 129.1 201.3 50.5

13c

3.22 (dd, 14.0, 5.5) 2.85 (overlapped) 3.28 (m) 3.09 (d, 12.8)

41.0

133.0 132.6

1d 2d (6d)

6.90 (d, 8.6)

136.9 129.6

115.2

3d (5d)

6.66 (d, 8.6)

115.9

156.3 45.8

4d 7d

3.38 (m)

156.7 47.6

47.3 147.0 118.4 160.5 96.4 155.7 120.6

8d 9d 10d 11d 12d 13d 14d

3.28 (m) 6.33 (d, 2.1) 6.19 (t, 2.1) 6.33 (d, 2.1)

54.5 149.6 107.0 159.4 101.8 159.4 107.0

PTP1B Inhibitory Assay. The PTP1B inhibition assay was performed according to the previous report with minor modification.12 In brief, working buffer containing MOPS (34.5 mM), DTT (2 mM), EDTA (0.69 mM), BSA (2 mg/mL) and NaCl (2 M), was prepared before the assay. A mixture containing 10 μL of tested samples (solved in DMSO), 70 μL of working buffer, and 10 μL of PTP1B enzyme (12.5 mg/L in working buffer) was transferred into 96-well plates and preincubated at 37 °C for 15 min. The reaction was started by the addition of 10 μL of substrate solution (p-NPP, 100 mM in working buffer) and further incubated for 30 min. The reaction was stopped by adding 100 μL of Na2CO3 (0.1 M), and the absorbance was measured at 405 nm using a Bio-Rad model 680 microplate reader. Oleanolic acid (OA) was used as the positive control, and the incubation without enzyme and samples were used as the control. The inhibition rate (%) was calculated as [(ODcontrol− OD c o n t r o l b l a n k ) − (OD s a m p l e − OD sa m p le b l an k )]/(OD c o n t r o l − ODcontrol blank) × 100%. Compounds showing the inhibition rate higher than 50% were further assayed for their dose−response relationships at five different concentrations. IC50 values were calculated through nonlinear regression using Graphpad Prism 6 Software. 6766

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry

Figure 1. Key 2D NMR correlations of compound 1. Enzyme Kinetic Analysis. The inhibitory mode of active stilbenes (3, 5, 8, and 10) on PTP1B was measured using three different concentrations of p-NPP (15, 17.5, 25 mM for 3; 10, 12.5, 18.75 mM for 5; 12.5, 18.75, 25 mM for 8 and 10) as the substrate, and four different concentrations of tested compounds (0, 180, 210 and 240 μM for 3; 0, 80, 100 and 120 μM for 5; 0, 20, 30 and 40 μM for 8; and 0, 50, 60 and 70 μM for 10) to obtain a Lineweaver−Burk double reciprocal plot. Using Dixon plots (single reciprocal plot), enzymatic reactions at various concentrations of 3, 5, 8, and 10 were evaluated by monitoring the effects of different concentrations of the substrate. The inhibition constants (Ki) were determined by the intersection of Dixon plots, where the value of the x-axis was taken as Ki.13,14 α-Glucosidase Inhibitory Activity. The α-glucosidase inhibition assay was performed according to the reported protocols.15,16 At first, acarbose was dissolved in phosphate buffer solution (PBS, pH 6.8), and samples were dissolved in MeOH-PBS (1:1).17 A mixture containing 20 μL of α-glucosidase enzyme (0.2 U/mL in PBS) and 30 μL of tested compounds (in triplicate) was preincubated at 37 °C for 5 min. The reaction was started by adding 20 μL of substrate solution (p-NPG, 2.5 mM in PBS) and incubated for additional 15 min. Afterward, the reaction was stopped by adding 40 μL of Na2CO3 (0.1 M). The absorbance was measured at 405 nm using a BIO-RAD Model 680 microplate reader. Blank was prepared by adding PBS instead of α-glucosidase, and control was prepared by adding MeOHPBS (1:1) instead of the samples using the same method. Compounds showing the inhibition rate higher than 50% were further assayed for their dose−response relationships at five or six different concentrations. The percentage inhibition and IC50 values were calculated using the same formula as described above. DPPIV Inhibitory Activity. The DPPIVinhibition assay was performed in accordance with the previous literature.18 In brief, a mixture containing 10 μL of samples dissolved in DMSO, 70 μL of Tris-HCl buffer, and 10 μL of DPPIV enzyme (10 mg/L in Tris-HCl buffer) was added in 96-well plates and preincubated at 37 °C for 10 min. The reaction was started by the addition of 10 μL of substrate solution (Gly-Pro-PNA, 10 mM in Tris-HCl buffer) and incubated for 30 min. The reaction was stopped by adding 100 μL of NaHCO3 (0.1 M), and the absorbance was measured at 405 nm using a Bio-Rad model 680 microplate reader. Sitagliptin was used as the positive control. The percentage inhibition and IC50 values were calculated using the same formula as described above. Quantitation of the Active Constituents. An HPLC-PDA quantitation was performed according to a method described previously.19,20 Stock solutions of compounds 4, 8, and 10 were prepared in MeOH, each at 1 mg/mL. Work solutions were obtained after series dilutions with methanol to achieve five concentration levels in the range of 6.25 to 100.00 μg/mL. The linearity was plotted using linear regression analysis by the integrated peak areas (Y) vs concentration of each standard (X, μg/mL) at five concentrations. Peony seed powder (1.0 g) was accurately weighed in triplicate and

each was extracted with MeOH (20 mL) by ultrasonication at room temperature for 30 min. The MeOH extraction was 10-fold diluted and filtered through 0.22 μm membrane before analyses. Statistical Analysis. All data were presented as mean ± standard deviation (SD) from three independent experiments. Statistical analysis was performed using GraphPad Prism 6.0 (GraphPad Software, Inc., San Diego, CA). For statistical comparisons, the data were analyzed using one-way analysis of variance (ANOVA) followed by Dunnett’s posthoc test, and p < 0.05 was considered statistically significant.

■

RESULTS AND DISCUSSION

Structure Elucidation. Compound 1 was assigned the molecular formula of C55H44O11 by the [M+H]+ ion at m/z 881.2948 (−0.91 ppm) and [M−H]− ion at m/z 879.2801 (−1.14 ppm), with 34 unsaturation degrees. The IR spectrum showed absorptions for hydroxy (3443 cm−1), aromatic (1632, 1515, and 1457 cm−1), and 1,4-disubstituted phenyl (840 cm−1) groups. The UV maximum absorptions at 279 and 221 nm were consistent with the presence of aromatic chromophores which are characteristic for oligostilbenes.21 In the 1H NMR spectrum, six sets of AA′BB′ coupling protons at δH 7.64 (2H, d, J = 8.6 Hz, H-2a/6a), 6.95 (2H, d, J = 8.6 Hz, H-3a/5a), 6.21 (2H, d, J = 8.7 Hz, H-2c/6c), 6.25 (2H, d, J = 8.7 Hz, H-3c/5c), 6.90 (2H, d, J = 8.6 Hz, H-2d/6d), and 6.66 (2H, d, J = 8.6 Hz, H-3d/5d) suggested the involvement of three para-substituted phenyls in the structure. Two characteristic broad singlets at δH 6.92 (2H, H-2b/6b) and 6.62 (2H, H3b/5b) were ascribed to a para-substituted phenyl due to the restricted rotation, and this case has been reported by Sarker et al. in determining the structure of suffruticosol B.21 From the above analysis, this compound was proposed as a stilbene tetramer for the presence of four para-substituted phenyls. In addition, one set of aromatic protons in AB2 coupling at δH6.33 (2H, d, J = 2.1 Hz, H-10d) and 6.19 (1H, t, J = 2.1 Hz, H12d), one set of meta-coupled protons at δH 6.27 (1H, d, J = 2.1 Hz, H-14a) and 6.09 (1H, d, J = 2.1 Hz, H-12a), an aromatic singlet at δH 6.29 (1H, s, H-12b), and an olefinic singlet at δH 5.85 (1H, s, H-10c) were also recognized. In the shielded region, eight aliphatic methines at δH5.96 (1H, d, J = 11.5 Hz, H-7a), 5.26 (1H, d, J = 11.5 Hz, H-8a), 4.42 (1H, d, J = 11.7 Hz, H-7b), 4.17 (1H, m, H-8b), 3.98 (1H, d, J = 6.1 Hz, H-7c), 3.75 (1H, s, H-8c), 3.38 (1H, m, H-7d), and 3.28 (1H, m, H-8d) further supported the presence of four resveratrol moieties. Besides the protons mentioned above, two aliphatic methylenes at δH3.22 (1H, dd, J = 14.0, 5.5 Hz, H-12c), 2.85 6767

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry

Figure 2. Structures of compounds 1−15.

cross-peaks of H-8b with H-10c, H-7c with H-8d, and H-8c with H-7d suggested the α-orientation of H-8c and H-12c and β-orientation of H-8d. Thus, the structure of 1 was totally constructed and named as paeonilactiflorol. The known compounds (Figure 2) were identified as transresveratrol (2),23 cis-ε-viniferin (3),24 trans-ε-viniferin (4),23 (−)-7a,8a-cis-ε-viniferin (5),25 carasiphenol A (6),22 cis-gnetin H (7),26 trans-gnetin H (8),26 suffruticosol A (9),21 suffruticosol B (10),21 suffruticosol C (11),21 paeoniflorin (12),27 benzoylpaeoniflorin (13),27 desbenzoylpaeoniflorin (14),28 and albiflorin (15)29 by the comparison of their MS and 1H- and 13C NMR data with the previous reports. Antidiabetic Tests. Each fraction from peony seeds was assayed for the PTP1B inhibitory activities in vitro. As shown in Figure 3, the EtOH extract could obviously inhibit PTP1B by more than 70% at the concentration of 62.5 μg/mL, whereas the oil was inactive. In order to clarify the active part, the EtOH extract was further separated into four fractions. Frs. 1 and 2 showed increased activity with the inhibition rates higher than 90%, whereas Fr. 3 only showed moderate activity (50%) and Fr. 4 showed no inhibition at the concentration of 62.5 μg/mL. Therefore, Frs. 1 and 2 were selected for the following study. Preliminary LC-MS analyses suggested that Frs. 1 and 2 contained mainly stilbene oligomers, whereas paeoniflorinrelated monoterpene glycosides were mainly present in Frs. 3 and 4. As a result, 15 compounds were obtained from Frs. 1 and 2 by repeated CC, and further assessed for their inhibition on PTP1B. The stilbene oligomers 1 and 4−10 showed remarkable inhibition on PTP1B at the concentration of 125

(1H, overlapped, H-12c), 3.28 (1H, m, H-13c), and 3.09 (1H, d, J = 12.8 Hz, H-13c) and one olefinic proton δH 5.85 (1H, s, H-10c) were further characterized. Taking the carbons ascribed to a carbonyl (δC 201.3, C-11c) and a trisubstituted double bond (δC 163.2, C-9c and 129.1, C-10c) into consideration, a cyclohept-2-enone moiety was proposed.22 Compared with suffruticosol B, the highly matched 1H and 13 C NMR data in 1 established the partial structure I, which was well supported by the detailed 2D NMR analyses: correlations of 7a/8a and 7b/8b/7c in 1H 1H COSY spectrum (the missing correlations between 7c and 8c was in accordance with the previous report21); correlations from H-8c to C-7c and C-8b in the HMBC spectrum. The stereochemistry of part I was determined to be identical with suffruticosol B by the high agreement in chemical shifts and coupling constants (JH‑7a/8a = 11.5 Hz, JH‑7b/8b = 11.7 Hz, JH‑7c/8b = 6.1 Hz, and the single-like peak of H-8c), and detailed ROESY analysis: correlations of H-8a with H-7a and 7b, and of H-8b with H-7c (Figure 1). The part II was quite similar to carasiphenol A22 in both 1H and 13C NMR data except that the 4-hydroxystyryl was changed to I. Detailed 2D NMR spectroscopic analyses involving 1H 1H COSY correlations of H-12c/7d/8d/13c, and HMBC correlations from H-10c to C-12c and C-13c; from H-8d to C-9c, C-10d, and C-14d; and from H-7d to C-11c, C2d, and C-6d fully supported the above analyses. The connection of parts I and II was established by the HMBC correlations from H-8c to C-10c and C-13c and from H-10c and H-13c to C-8c. In the ROESY experiment, the obvious 6768

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry

Figure 3. (A) PTP1B inhibitory activities of different fractions (62.5 μg/mL); (B) PTP1B inhibitory activities of the isolates (125 μM); (C) αGlucosidase inhibitory activities of the isolates (62.5 μM); (D) DPPIV inhibitory activities of the isolates (500 μM). Data were expressed as mean ± SD (n = 3). One-way ANOVA was carried out followed by posthoc Dunnett’s multiple comparison test. *P < 0.05, ***P < 0.001, ****P < 0.0001, compared with control group.

μM, and thus, their dose−response relationships were further studied to provide their respective IC50 value (Table 2). The stilbene tetramer (1), trimers (8, 9, and 10), and dimers (4 and 5) displayed potent activity on PTP1B with IC50 values of 27.23, 27.81, 86.26, 53.93, 85.03, and 97.74 μM, respectively. Interestingly, the trans-forms were always manifested with

higher activity than the cis-forms (4 and 5 vs 3 and 8 vs 7), suggesting that the trans-configuration of double bond was preferable to the cis-configuration. This deduction was consistent with the previous report in which the transoligostilbenes showed obvious PTP1B inhibitory activity.13 Compounds 4 and 5 with trans-double bonds maintained similar activity, indicating that the configuration of the phenyl groups on dihydrofuran played little role in inhibiting PTP1B. When comparing to 9 and 10, compound 11 exhibited obviously decreased activity indicating the importance of seven-member ring. In contrast, the stilbene monomer (2), cisdimer (3), and monoterpene glycosides (12-15) were less active ( 0.999) was achieved for each of the quantitated constituents, based on which their contents in peony seeds were determined (Table 3). As a result, compounds 10 and 8 were the major stilbene trimers in peony seeds with the content of 23.17 ± 0.36 and 15.24 ± 0.25 6770

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry Funding

Table 3. Quantitative Analysis of 4, 8, and 10 in Peony Seeds (per Dry Weight). compounds

retention time (min)

4

6.0

8

8.8

10

4.0

regression equation (Y = aX + b, R2)a Y = 70616X + 134980, 0.9992 Y = 58868X − 134878, 0.9997 Y = 61177X − 201082, 0.9991

a

linear range (mg/L)

content (mg/g)b

6.25 ∼ 100.00 6.25 ∼ 100.00 6.25 ∼ 100.00

0.63 ± 0.02 15.24 ± 0.43 23.17 ± 0.62

This study was financed by the National Natural Science Foundation of China (81773609 and 81773612), the Program of Yunling Scholarship, and the Applied Basic Research Programs of Yunnan Province (2017FB137). Notes

The authors declare no competing financial interest.

■

Y represents peak area and X represents concentration; Values are expressed as the mean ± SD (n = 3).

mg/g, respectively, and compound 4 was present in low content (0.63 ± 0.01 mg/g). In this investigation, peony seeds were first revealed with PTP1B inhibitory effects with the high-polar extract instead of oil as the active part. Under the guidance of bioassay, 15 compounds involving 11 stilbenes and four paeoniflorinrelated monoterpene glycosides were isolated from the active parts. Most of the stilbenes showed obvious inhibition on PTP1B and α-glucosidase, superior to the monoterpene glycosides. Especially, the stilbene tetramer (1) and trimer (8) exhibited high activity inhibiting both PTP1B with IC50 values of 27.23 and 27.81 μM and α-glucosidase with IC50 values of 13.57 and 14.39 μM. Two trans-dimers 4 and 5 also exhibited DPPIV inhibitory activity in addition to PTP1B and α-glucosidase. In this study, the trans-form stilbenes always exhibited higher activity than the cis-form, suggesting the importance of the trans-double bond. Enzyme kinetic study on PTP1B suggested that the inhibition types of 3 and 5 were noncompetitive and of 8 and 10 were mixed. Quantitative analysis manifested that the active stilbene trimers 8 (23.17 ± 0.36 mg/g) and 10 (15.24 ± 0.25 mg/g) were the main contents in peony seeds. It is concluded that peony seeds especially the seed cake after oil extraction may be considered as a potentially antidiabetic functional foods or food additives. This investigation provided valuable clues for expanding the application of peony seeds and searching new PTP1B inhibitors from natural sources.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.9b01193. 1 H NMR, 13C NMR, HSQC, 1H 1H COSY, HMBC, ROESY, IR, UV, HRESIMS, [α]D, ECD spectra of 1; 1H NMR, 13C NMR, and HRESIMS data of 2−15; quantitative analyses; and PTP1B inhibitory activities of different fractions (PDF)

■

REFERENCES

(1) World Health Organization. Fact sheet: diabetes. URL (https:// www.who.int/news-room/fact-sheets/detail/diabetes). (2) Nagaraju, K.; Ashona, S. P.; Paul, A.; Parvesh, A. Current antidiabetic agents and their molecular targets: A review. Eur. J. Med. Chem. 2018, 152, 436−488. (3) Kesavadev, J.; Shankar, A.; Gopalakrishnan, G.; Jothydev, S. Efficacy and safety of liraglutide therapy in 195 Indian patients with type 2 diabetes in real world setting. Diabetes Metab. Syndr. 2015, 9, 30−33. (4) Baptist, G. Small molecule dipeptidylpeptidase IV inhibitors under investigation for diabetes mellitus therapy. Expert Opin. Invest. Drugs 2011, 20, 723−732. (5) Yabe, D.; Hamamoto, Y.; Seino, Y.; Kuwata, H.; Kurose, T.; Seino, Y. Sodium glucose co-transporter 2 inhibitor luseogliflozin in the management of type 2 diabetes: A drug safety evaluation. Expert Opin. Drug Saf. 2017, 16, 1211−1218. (6) Chiba, S. Molecular Mechanism in α-glucosidase and glucoamylase. Biosci., Biotechnol., Biochem. 1997, 61, 1233−1239. (7) Cho, H. Protein tyrosine phosphatase 1B (PTP1B) and obesity. Vitam. Horm. 2013, 91, 405−424. (8) Zhao, B. T.; Nguyen, D. H.; Le, D. D.; Choi, J. S.; Min, B. S.; Woo, M. H. Protein tyrosine phosphatase 1B inhibitors from natural sources. Arch. Pharmacal Res. 2018, 41, 130−161. (9) Ning, C. L.; Jiang, Y.; Meng, J. S.; Zhou, C. H.; Tao, J. Herbaceous peony seed oil: A rich source of unsaturated fatty acids and γ-tocopherol. Eur. J. Lipid Sci. Technol. 2015, 117, 532−542. (10) Su, J. H.; Wang, H. X.; Ma, C. Y.; Lou, Z. X.; Liu, C. X.; Rahman, M. T.; Gao, C. Z.; Nie, R. J. Anti-diabetic activity of peony seed oil, a new resource food in STZ-induced diabetic mice. Food Funct. 2015, 6, 2930−2938. (11) Sun, J.; Jiang, Y.; Tao, J. Research progress on edible value of Paeonia suff ruticosa and Paeonia lactif lora. Jiangsu Agric. Sci. 2018, 46, 10−14. (12) Tahtah, Y.; Wubshet, S. G.; Kongstad, K. T.; Heskes, A. M.; Pateraki, I.; Moeller, B. L.; Jager, A. K.; Staerk, D. High-resolution PTP1B inhibition profiling combined with high-performance liquid chromatography-high-resolution mass spectrometry-solid-phase extraction-nuclear magnetic resonance spectroscopy: proof-of-concept and antidiabetic constituents in crude extract of Eremophila lucida. Fitoterapia 2016, 110, 52−58. (13) Ha, M. T.; Park, D. H.; Shrestha, S.; Kim, M.; Kim, J. A.; Woo, M. H.; Choi, J. S.; Min, B. S. PTP1B inhibitory activity and molecular docking analysis of stilbene derivatives from the rhizomes of Rheum undulatum L. Fitoterapia 2018, 131, 119−126. (14) Ha, M. T.; Seong, S. H.; Nguyen, T. D.; Cho, W. K.; Ah, K. J.; Ma, J. Y.; Woo, M. H.; Choi, J. S.; Min, B. S. Chalcone derivatives from the root bark of Morus alba L. act as inhibitors of PTP1B and αglucosidase. Phytochemistry 2018, 155, 114−125. (15) Milella, L.; Milazzo, S.; Leo, M. D.; Saltos, M. B. V.; Faraone, I.; Tuccinardi, T.; Lapillo, M.; Tommasi, N. D.; Braca, A. α-Glucosidase and α-amylase inhibitors from Arcytophyllum thymifolium. J. Nat. Prod. 2016, 79, 2104−2112. (16) Bisio, A.; Mieri, D. M.; Milella, L.; Schito, A. M.; Parricchi, A.; Russo, D.; Alfei, S.; Lapillo, M.; Tuccinardi, T.; Hamburger, M.; Tommasi, N. D. Antibacterial and hypoglycemic diterpenoids from Salvia chamaedryoides. J. Nat. Prod. 2017, 80, 503−514. (17) Flores-Bocanegra, L.; Gonzaález-Andrade, M.; Bye, R.; Linares, E.; Mata, R. α-Glucosidase inhibitors from Salvia circinata. J. Nat. Prod. 2017, 80, 1584−1593.

b

AUTHOR INFORMATION

Corresponding Authors

*Tel: +86-871-65223265. Fax: +86-871-65227197. E-mail: [email protected]. *E-mail: [email protected]. ORCID

Chen-Chen Zhang: 0000-0003-1567-4676 Chang-An Geng: 0000-0001-9834-0756 Ji-Jun Chen: 0000-0001-5781-7511 6771

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772

Article

Journal of Agricultural and Food Chemistry (18) Wang, F.; Yu, G. Y.; Zhang, Y. Y.; Zhang, B. L.; Fan, J. F. Dipeptidyl peptidase IV inhibitory peptides derived from oat (Avena sativa L.), buckwheat (Fagopyrum esculentum), and highland barley (Hordeum vulgare trif urcatum (L.) Trofim) proteins. J. Agric. Food Chem. 2015, 63, 9543−9549. (19) Li, J.; Yuan, C. H.; Pan, L.; Benatrehina, P. A.; Chai, H.; Keller, W. J.; Naman, C. B.; Kinghorn, A. D. Bioassay-guided isolation of antioxidant and cytoprotective constituents from a maqui berry (Aristotelia chilensis) dietary supplement ingredient as markers for qualitative and quantitative analysis. J. Agric. Food Chem. 2017, 65, 8634−8642. (20) Gamiotea-Turro, D.; Camaforte, N. A. P.; Valerino-Diaz, A. B.; Nunez, Y. O.; Rinaldo, D.; Dokkedal, A. L.; Bosqueiro, J. R.; Santos, L. C. D. Qualitative and quantitative analysis of ethanolic extract and phenolic fraction of Jatropha aethiopica (Euphorbiaceae) leaves and their hypoglycemic potential. J. Agric. Food Chem. 2018, 66, 1419− 1427. (21) Sarker, S. D.; Whiting, P.; Dinan, L. Identification and ecdysteroid antagonist activity of three resveratrol trimers (suffruticosols A, B and C) from Paeonia suff ruticosa. Tetrahedron 1999, 55, 513−524. (22) Wang, S. G.; Ma, D. Y.; Hu, C. Q. Two new oligostilbenes from Caragana sinica. J. Asian Nat. Prod. Res. 2004, 6, 241−248. (23) Tisserant, L. P.; Hubert, J.; Lequart, M.; Borie, N.; Maurin, N.; Pilard, S.; Jeandet, P.; Aziz, A.; Renault, J. H.; Nuzillard, J. M.; Cleément, C.; Boitel-Conti, M.; Courot, E. 13C NMR and LC-MS profiling of stilbenes from elicited grapevine hairy root cultures. J. Nat. Prod. 2016, 79, 2846−2855. (24) Kim, H. J.; Chang, E. J.; Bae, S. J.; Shim, S. M.; Park, H. D.; Rhee, C. H.; Park, J. H.; Choi, S. W. Cytotoxic and antimutagenic stilbenes from seeds of Paeonia lactif lora. Arch. Pharmacal Res. 2002, 25, 293−299. (25) Choi, C. W.; Choi, Y. H.; Cha, M. R.; Kim, Y. S.; Yon, G. H.; Hong, K. S.; Park, W. K.; Kim, Y. H.; Ryu, Y. S. In vitro BACE-1 inhibitory activity of resveratrol oligomers from the seed extract of Paeonia lactif lora. Planta Med. 2011, 77, 374−376. (26) He, C. N.; Peng, Y.; Xu, L. J.; Liu, Z. A.; Gu, J.; Zhong, A. G.; Xiao, P. G. Three new oligostilbenes from the seeds of Paeonia suff ruticosa. Chem. Pharm. Bull. 2010, 58, 843−847. (27) Braca, A.; Kiem, P. V.; Yen, P. H.; Nhiem, N. X.; Quang, T. H.; Cuong, N. X.; Minh, C. V. New monoterpene glycosides from Paeonia lactiflora. Fitoterapia 2008, 79, 117−120. (28) Stavri, M.; Mathew, K. T.; Bucar, F.; Gibbons, S. Pangelin, an antimycobacterial coumarin from Ducrosia anethifolin. Planta Med. 2003, 69, 956−959. (29) Shi, Y. H.; Zhu, S.; Ge, Y. W.; He, Y. M.; Kazuma, K.; Wang, Z. T.; Yoshimatsu, K.; Komatsu, K. Monoterpene derivatives with antiallergic activity from red peony root, the root of Paeonia lactif lora. Fitoterapia 2016, 108, 55−61.

6772

DOI: 10.1021/acs.jafc.9b01193 J. Agric. Food Chem. 2019, 67, 6765−6772