Phenol profiles and antioxidant capacities of Bistort Rhizoma (Polygonum bistorta L.) extracts

Abstract

Bistort Rhizoma (BR) is consumed as a traditional medicinal herb in China, but there reports on its bioactive components are scarce. In the present study, a fingerprinting analysis was built in 35 batches of BR from various sources, and five major compounds, the content of total phenols (TP) as well as total extractable tannins (TET) were measured. The antioxidant properties were examined by three methods, and then the correlation coefficients between antioxidant activity and content of BR properties were evaluated. The results showed that the content of TP was significantly correlated with the herbal antioxidant activity (R² = 0.5501 in the DPPH assay; R² = 0.6387 in the ABTS assay and R² = 0.6722 in the FRAP assay) but the five major components are not, TET also partly influenced the herbal efficiency. In conclusion, the properties responsible for the antioxidant activity of BR are phenols, and the active constituents may be some trace phenols in BR, which need further study to unravel the truly active compounds.

---

1. Introduction

Antioxidants are an important group of medicinal preventive compounds as well as being food additives inhibiting detrimental changes of easily oxidizable nutrients.¹ Naturally occurring phenolic compounds have been recognized as an important class of antioxidants,² such as anthocyanins, tannins, flavanones, isoflavones, resveratrol and ellagic acid, and they are currently used as in scientific research and industry as pharmaceutical, nutraceuticals and/or functional foods.³ In recent years, considering the side effects of synthetic antioxidants such as carcinogenicity,⁴ interest has considerably increased in finding new natural, safe and economical antioxidant substances, especially from abundant and low-value raw materials.

Bistort Rhizoma (Polygonum bistorta L.), is a widely clinically-deployed herb for the treatment of dysentery with bloody stools, diarrhea in acute gastroenteritis and venomous snake bites.⁵ It has been reported that BR contains large amounts of phenolic compounds.^6,7 In our preliminary work, 31 trace polyphenols in BR were quickly identified using weak anion-exchange solid phase extraction (SPE) and high performance liquid chromatography-quadrupole time-of-flight mass spectrometry (HPLC-QTOF MS),⁸ which may be a potential antioxidant source. Several pharmacological studies demonstrated that BR has anti-oxidative and antimicrobial,⁹ and other biological activities. To the best of our knowledge, no report has revealed the constituents responsible for the anti-oxidative activity of BR extracts.

Due to the complexity of the composition of the herb, separating each antioxidant compound and studying each one is both costly and inefficient. However, synergistic interactions among the antioxidant compounds may be possible in a medicinal plant mixture. Many factors such as geographic and environmental differences of growing conditions could affect the batch-to-batch uniformity of herbal products. Unlike chemical drugs, herbal medicines contain relatively unrefined mixtures of phytochemical compounds and thus present a great challenge in discovering the active ingredients exactly responsible for certain biological activities. It is necessary to identify a single compound or a combination of chemical compounds which act together to elicit pharmacological effects and could be representative of the whole herbal medicine.¹⁰

In the present investigation, the content of total phenols (TP), the content of total extractable tannins (TET) and five major compounds of 35 batches of the BR sample were determined by spectrophotometry coupled with precipitation methods and high-performance liquid chromatography with variable wavelength detection (HPLC-VWD), respectively. Antioxidant activity against organic free radicals and reducing power were assayed by three tests: 1,1-diphenyl-2-picrylhydrazyl radical (DPPH), 2,2′-azinobis(3-ethylbenzothiazoline sulfonate) (ABTS) and ferric-reducing antioxidant power (FRAP) methods. This study was aimed at discovering and identifying the effective ingredients that are most relevant with antioxidant activity of BR.

2. Experimental

2.1. Chemicals and reagents

Acetonitrile (ACN) of HPLC grade was purchased from Tedia (Ohio, USA), and formic acid with a purity of 96% is of HPLC grade (Tedia, USA). Deionized water (18 MΩ) was prepared by a Milli-Q system (Millipore, Milford, MA, USA). The Folin–Ciocalteu reagent, ascorbic acid, 1,1-diphenyl-2-picrylhydrazyl (DPPH), 2,2′-azinobis-3-ethyl-benzothiazolin-6-sulfonic acid (ABTS) and 2,3,4-tris(2-pyridyl)-s-triazine (TPTZ) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Iron chloride hexahydrate, ferrous sulfate, sodium acetate trihydrate and potassium persulfate were purchased from Xilong Chemical Co., Ltd (Shantou, China). These solutions were wrapped in aluminum foil and stored at 4 °C. Methanol and acetic acid of analytical grade was purchased from Jiangsu Hanbang Chemical Reagent Co., Ltd. (Nanjing, China). Other reagents and chemicals were of analytical grade.

Three reference compounds including gallic acid (1), catechin (4) and chlorogenic acid (5) were purchased from the National Institute for the Control of Pharmaceutical and Biological Products with a purity of more than 95%. Standard stock solution 1, contains two accurately weighed reference compounds (1 and 4) directly prepared in methanol at final concentrations of 0.48 mg mL⁻¹ and 0.41 mg mL⁻¹, respectively. Standard stock solution 2 contains chlorogenic acid compounds (5) at a final concentration of 0.84 mg mL⁻¹ for high-performance liquid chromatography with variable wavelength detection (HPLC-VWD). These two kinds of standard stock solutions were prepared by diluting the stock solutions with methanol to a series of concentrations. These solutions were stored at 4 °C for further study.

2.2. Plant materials and extraction

2.2.1. Plant materials. A total of 35 BR samples (batch no. S1–S35) were collected from Gansu, Hunan, Jiangsu, Zhejiang, Yunnan, Sichuan, Guizhou, Jilin, Liaoning, Inner Mongolia, Hebei, Anhui and Henan Province of China (Table 1). Among them, 7 batches (batch no. S1, S5, S16, S24, S28, S29 and S30) were gathered by our staff and other batches were purchased from local drugstores. All of the BR samples were authenticated by Prof. Hui-Jun Li at China Pharmaceutical University. BR was powdered and sieved through a no. 60 mesh, then dried at 50 °C in an oven for 4 h.

Table 1 Total phenol and tannin content in BR extracts from different regions^a

Sample	Source	TP^a (mg GAE per g)	NPP^a (mg GAE per g)	TET^a (mg GAE per g)	DPPH-IC50 (μg mL^−1)	ABTS-IC50 (μg mL^−1)	Fe^2+ (μmol L^−1)
a TP means total phenols, NPP means non-precipitable phenols, TET means total extractable tannins. ^a,bData are represented as mean ± SD.
S1	Gansu	71.1245	68.0329	3.0916	18.4792 ± 0.0741^b	2444.1021 ± 61.4602	0.6234 ± 0.0039
S2	Gansu	79.3686	39.1784	40.1902	19.2266 ± 0.1055	2059.7285 ± 36.4010	0.9563 ± 0.0033
S3	Zhejiang	109.9095	38.0542	71.8552	15.1230 ± 0.0474	1158.4977 ± 37.3661	0.8464 ± 0.0041
S4	Zhejiang	122.6504	50.0457	72.6047	11.6930 ± 0.0820	1119.5334 ± 35.8312	1.1886 ± 0.0362
S5	Gansu	80.8676	35.6185	45.2491	11.5191 ± 0.1268	1979.1563 ± 72.7460	0.7905 ± 0.0046
S6	Gansu	72.9982	33.1827	39.8155	18.1117 ± 0.1093	1756.8321 ± 14.9422	0.8428 ± 0.0177
S7	Guizhou	56.5099	24.9385	31.5713	24.1025 ± 0.0957	3029.5864 ± 159.2863	0.4566 ± 0.0059
S8	Hebei	102.0401	35.0564	66.9837	9.3385 ± 0.2790	1347.3250 ± 36.2033	0.7333 ± 0.0031
S9	Gansu	105.6000	67.2835	38.3165	10.3053 ± 0.0682	1527.6013 ± 8.6406	0.7243 ± 0.0030
S10	Hebei	110.3779	12.5911	97.7868	9.8618 ± 0.0713	842.8758 ± 29.2474	1.2570 ± 0.0100
S11	Inner Mongolia	50.1394	33.7448	16.3946	25.0544 ± 0.0901	3572.1481 ± 82.4181	0.4691 ± 0.0012
S12	Gansu	107.2863	68.2203	39.0660	10.5213 ± 0.0741	1329.0967 ± 18.0452	0.7382 ± 0.0019
S13	Guizhou	35.4311	22.0344	13.3967	35.6198 ± 0.1067	5895.9146 ± 94.8863	0.4294 ± 0.0040
S14	Henan	95.1075	40.4900	54.6175	15.7570 ± 0.0729	1515.4999 ± 13.9314	1.1787 ± 0.0372
S15	Zhejiang	65.8782	26.9996	38.8786	22.6743 ± 0.0593	1472.1550 ± 26.1829	0.5188 ± 0.0180
S16	Guizhou	85.3644	26.4375	58.9269	11.3237 ± 0.0344	1112.0322 ± 16.3812	0.6746 ± 0.0163
S17	Huabei	25.6880	15.5702	10.1178	90.5082 ± 0.7056	5992.9238 ± 208.3781	0.2441 ± 0.0011
S18	Guizhou	65.4098	41.9889	23.4209	20.1477 ± 0.1025	2103.1843 ± 62.1890	0.5832 ± 0.0020
S19	Guizhou	94.3580	36.5553	57.8027	15.9717 ± 0.2858	1497.9685 ± 23.8894	0.6513 ± 0.0034
S20	Guizhou	34.3069	20.6291	13.6778	52.9639 ± 0.3024	5753.4787 ± 24.4100	0.1935 ± 0.0151
S21	Liaoning	152.1607	49.2962	102.8645	8.8378 ± 0.0444	916.6795 ± 1.9447	1.2097 ± 0.0172
S22	Anhui	81.1486	33.7448	47.4038	22.7801 ± 0.1027	1753.7040 ± 54.2327	0.6917 ± 0.0173
S23	Jiangsu	25.8754	23.0649	2.8105	67.5449 ± 0.4322	5587.1468 ± 135.5964	0.2897 ± 0.0070
S24	Yunnan	88.4559	25.6880	62.7679	24.4970 ± 0.0881	1960.0167 ± 49.2870	0.7802 ± 0.0021
S25	Jilin	71.0308	30.3722	40.6586	24.0673 ± 0.1828	1771.4861 ± 54.9255	0.5754 ± 0.0043
S26	Yunnan	71.7803	30.7469	41.0334	24.7082 ± 0.4415	1734.4158 ± 23.4199	0.4573 ± 0.0031
S27	Sichuan	57.9151	24.7512	33.1640	41.8828 ± 0.1474	2086.4714 ± 41.4331	0.6986 ± 0.0260
S28	Dongbei	76.4645	33.7448	42.7197	31.8608 ± 0.1336	1417.5029 ± 34.7313	0.4199 ± 0.0392
S29	Dongbei	53.6057	32.2459	21.3598	39.1583 ± 0.3640	2451.5044 ± 45.4019	0.4482 ± 0.0024
S30	Yunnan	64.8477	24.3764	40.4713	32.8514 ± 0.7623	1835.0333 ± 18.8329	0.5332 ± 0.0021
S31	Sichuan	62.7867	28.6859	34.1008	31.6641 ± 0.7187	2150.6292 ± 11.8453	0.5372 ± 0.0021
S32	Dongbei	93.2338	36.3679	56.8659	27.3121 ± 0.2132	1432.8144 ± 13.3125	0.655 ± 0.0190
S33	Sichuan	150.4744	60.1635	90.3109	14.0262 ± 0.1094	861.2609 ± 25.8427	0.9903 ± 0.0033
S34	Hubei	102.1337	80.5865	21.5472	22.0211 ± 0.1297	1929.6395 ± 28.6538	0.8383 ± 0.0025
S35	Anhui	110.3779	107.0053	3.3726	26.4372 ± 0.2509	1228.8878 ± 10.4266	0.6775 ± 0.0151

---

2.2.2. Sample preparation. About 0.5 g of dried Bistort Rhizoma was transferred, finely powdered and accurately weighed, to a 100 mL glass-stoppered conical flask, followed by the addition of 20 mL of 70% methanol and after weighing the filled flask with a precision of ±0.01 g, sonicated for 40 minutes (140 W, 100 kHz). The extract was allowed to cool and adjusted to the initial weight by corresponding solvent as needed. After centrifuging at 13 000 rpm for 10 min, the supernatant was separated and stored at 4 °C in airtight containers prior to analysis by HPLC-VWD.

2.3. Evaluation of the antioxidant capacity

All reactive oxygen species (ROS) scavenging capacities studied in this research were evaluated spectrophotometrically. Ascorbic acid was used as a positive control. A control group was performed without adding BR extract or the standard. Samples were analyzed in triplicate. The percentage of scavenging effect was calculated by using the following formula:¹¹

ROS scavenging effect (%) = [(A₀ − A₁)/A₀] × 100 (1)

where A₀ is the absorbance of the control (without antioxidant), and A₁ is the absorbance of the solutions in the presence of antioxidant. For comparison purposes, the antiradical activity (three replicates per treatment) was expressed as IC₅₀ (mg mL⁻¹) or 1/IC₅₀ in DPPH assay and ABTS assay, so that the higher antioxidant efficiency value would correspond to a better scavenging activity. The IC₅₀ value, which is the concentration of the sample required to scavenge 50% of reactive species, was determined graphically from the curve plotted for BR extracts or standards concentration (mg mL⁻¹) vs. inhibition (%).

2.3.1. DPPH assay. The electron donation ability of the obtained methanol BR extracts was measured by bleaching the purple-colored solution of 1,1-diphenyl-2-picrylhydrazyl radical (DPPH) according to the method described in the literature¹² with slight modification. For DPPH radicals assay, methanol solutions (0.1 mL) of BR extracts with concentrations of 0.75–40 μg mL⁻¹ were mixed with a freshly prepared DPPH solution (0.1 mL, 0.1 mM in methanol).¹³ The mixture was shaken vigorously and left to stand for 30 min in the dark at room temperature and the absorbance was read at 517 nm. The ability to scavenge the DPPH radical was calculated using eqn (1). Ascorbic acid was used as a positive control. Samples were analyzed in triplicate.

2.3.2. ABTS assay. This assay is based on 2,2′-azinobis(3-ethylbenzothiazoline sulfonate) (ABTS) incubated with a peroxidase (metmyoglobin) and hydrogen peroxide (H₂O₂) to produce the radical cation ABTS˙⁺ (ref. 14) which has a relatively stable blue-green colour. Antioxidants in the added sample cause suppression of the appearance of the colour to a degree proportional to their concentration. This analytical procedure has been applied to physiological antioxidant compounds and radical-scavenging drugs, and an antioxidant ranking based on their reactivity relative to ascorbic acid standard has been established. The ABTS˙⁺ stock solution was diluted with methanol to an absorbance of 0.70 ± 0.02 at 734 nm as a working solution. An aliquot (5 μL) of aqueous solution containing 62.5 to 4000 μg mL⁻¹ of BR extract was mixed thoroughly with the ABTS˙⁺ solution (150 μL) and after 6 min in the dark at room temperature; the absorbance was read at 734 nm. The radical-scavenging activity for ABTS radical was calculated using the same formula as the DPPH radical. Samples were analyzed in triplicate.

2.3.3. FRAP assay. The ferric-reducing antioxidant power (FRAP) assay was carried out according to the procedure described in the literature.¹⁵ In the FRAP assay, reductants/antioxidants in the sample reduced Fe³⁺/tripyridyltriazine complex, present in stoichiometric excess, to the blue coloured ferrous form, with an increase in absorbance at 593 nm. Since the FRAP assay has a limit of detection of <2 μmol L⁻¹ reducing/antioxidant power, precision of this assay is excellent. Briefly, the FRAP reagent was prepared according to the description of total antioxidant capacity assay.¹⁶ The FRAP reagent was freshly prepared daily and warmed to 37 °C in a water bath before use. 5 μL of the diluted sample was added to 150 μL of the FRAP reagent. The absorbance of the mixture was measured at 593 nm using a Synergy 2 Multi-Mode Microplate Reader (BioTek, USA) after 4 min. The standard curve was constructed using FeSO₄ solution, and the results are expressed as μM Fe(II) per g dry weight of herb. Additional dilution was needed if the FRAP value measured was over the linear range of the standard curve. A higher absorbance indicates a higher reducing power. In a similar way to the DPPH and ABTSS assays, these samples were analyzed in triplicate.

2.4. Chromatography conditions

Chromatographic analysis was performed on an Shimadzu LC-20A (Shimadzu Corporation, Japan) HPLC system equipped with a quaternary solvent delivery system, an on-line degasser, an auto-sampler, a column temperature controller and a variable wavelength detector coupled with an analytical workstation (Lab solution for LC). Chromatographic separation was achieved at 40 °C on a Shimadzu Inertsil ODS-SP C18 (4.6 × 150 mm, 5 μm) with a Shimadzu Inertsil ODS-SP C18 guard column (4∅ × 10 mm, 5 μm). The separation was obtained using a gradient mobile phase consisting of 0.04% phosphoric acid water (A) and acetonitrile (B). The gradient elution was set at: 0–5 min, 0–1% B; 5–10 min, 1–5% B; 10–40 min, 5–8% B; and the column was finally reconditioned with 0% B isocratic for 10 min. The flow-rate was kept at 0.8 mL min⁻¹. The detection wavelength was set at 270 nm for the first 32 min, and 325 nm from 32 to 40 min. The injection volume was set at 5 μL. Experiments were performed in sextuplicate for each batch of BR sample.

2.5. Determination of the contents of total phenols and total extractable tannins

The content of total phenols (TP) of BR extract was estimated according to Folin–Ciocalteu’s method^11,17 with slight modification. 0.2 mL of sample solution at a concentration of 400 μg mL⁻¹ was mixed with 1.0 mL of the Folin–Ciocalteu reagent (diluted 10 times by distilled water) and 1.2 mL of 29% (w/v) aqueous sodium carbonate solution. The absorbance of the final mixture was read at 760 nm against the blank solution (solution with no extract added). Gallic acid was used as the standard and TP content was expressed as milligrams of gallic acid equivalents (GAE) per gram of dry BR samples (mg GAE per g of dry BR samples) through the calibration curve with gallic acid. The calibration curve range was 0.001–0.01 mg mL⁻¹ (R² = 0.9985). Total extractable tannins (TET) were estimated indirectly by spectrophotometric measurement of the absorbance of the solution obtained after the precipitation of the insoluble tannins based on phosphomolybdium tungstic acid–casein reaction, described in Chinese Pharmacopoeia 2010.¹⁸ Briefly, 2.5 mL sample solution was quickly mixed with 0.6 g casein and incubated at 30 °C for 1 h. After filtration, the supernatant was collected. The phenol content of supernatant, which corresponds to the non-precipitable phenols (NPP), was measured by Folin–Ciocalteu’s method, and the total extractable tannins (TET) were calculated as the following formula:¹¹

TET = TP − NPP (2)

2.6. Statistical analysis

All analyses were carried out in triplicate, and the data were expressed as the mean ± standard deviation (SD). Differences were considered to be significant for p < 0.05, and were considered to be very significant for p < 0.01. Graphpad Prism 6.0 (San Diego, CA, USA) and Microsoft Excel 2013 (Roselle, IL, USA) were used for the statistical and graphical analysis.

3. Results and discussion

3.1. Fingerprint analysis and quantitation of five major compounds

3.1.1. Optimization of HPLC conditions. The HPLC conditions, including chromatographic column and mobile phases were optimized to simultaneously separate different types of compounds in BR samples. Finally, the assay was performed on a Shimadzu Inertsil ODS-SP C18 column (4.6 × 150 mm, 5 μm) and an elution program of 0.1% aqueous formic acid (A) and acetonitrile (B) was used as the mobile phase system. All major compounds could be eluted with baseline separation in a run time of 40 min. Increasing the column temperature can significantly promote the resolution of gallic acid, galloylglucoside isomers, catechin and chlorogenic acid. To achieve appropriate resolution and a short analysis time, 40 °C was selected as a compromise.

The fingerprints of 35 batches of BR samples were obtained by HPLC-VWD. The common peaks were defined by the software approved by China Pharmacopeia, named as Similarity Evaluation System for the analysis of chromatographic fingerprint of TCM (2004A Version). The known common peaks were used in a “multiple point correction” step to align the spectra, and five common peaks were found accordingly, they are gallic acid (1), galloylglucoside isomer 1 (2), galloylglucoside isomer 2 (3) and catechin (4) and chlorogenic acid (5). The five common compounds all belong to the phenolic compounds. Typical chromatograms of the samples are shown in Fig. 1. The five compounds analyzed have different UV λ_max values, for example, gallic acid, galloylglucoside isomer 1, galloylglucoside isomer 2 and catechin showed distinct maximal UV absorption at 270 nm, while chlorogenic acid had maximal absorptions at 325 nm. To analyze them at one run, a segmental wavelength method was used. The UV wavelength program was set as follows: 270 nm for 0–32 min; 325 nm for 32–40 min.

Fig. 1 Chromatograms of 35 batches of BR. Common peaks: 1, gallic acid; 2, galloylglucoside isomer 1; 3, galloylglucoside isomer 2; 4, catechin; 5, chlorogenic acid.

3.1.2. Method validation and quantitative analysis of five major compounds in BR. The method was validated in terms of linearity, sensitivity, precision, stability and accuracy. The linearity, test range, limit of detection (LOD) and limit of quantification (LOQ) of the five compounds are shown in Table 2. It was found that the calibration curves for all compounds showed good linearity (R² = 0.9995 for gallic acid; R² = 0.9997 for catechin and R² = 0.9999 for chlorogenic acid) within the test range. The LODs and LOQs of peak 1 were 7.21 and 23.90 ng, respectively; those of peak 4 were 82.80 and 310.49 ng, respectively; those of peak 5 were 12.56 and 42.11 ng, respectively for VWD, indicating that this method is sufficiently sensitive. The relative standard deviations (RSDs) for intra- and inter-day repeatability are also shown in Table 2. It was found that overall intra- and inter-day variations were not more than 3% (1.766% and 1.226% for gallic acid; 2.072% and 1.157% for catechin; 1.716% and 1.226% for chlorogenic acid), suggesting that the developed method is precise. The spike recoveries of three components were 95.36–100.11% and RSD were not more than 1% (0.182%, 0.888% and 0.690%, respectively), demonstrating that this method is also accurate.

Table 2 Results of the method validation for quantification of five major compounds in BR^a

Compound	Regression equation	Linear range (μg ml^−1)	R^2	LOD (ng)	LOQ (ng)	Precision (RSD, %)		Recovery
						Intra-day	Inter-day	Mean, %	RSD, %
a LOD: limits of detection, LOQ: limits of quantification.
Gallic acid	Y = 20 569 814X − 31 281	4.78–191.20	0.9995	7.21	23.90	1.766	1.226	95.36	0.182
Catechin	Y = 3333.6X − 6507.1	4.14–165.60	0.9997	82.80	310.49	2.072	1.157	100.11	0.888
Chlorogenic acid	Y = 17 322X − 88 254	8.42–505.2	0.9999	12.56	42.11	1.716	0.970	95.36	0.69

---

With reference compounds, the contents of three known compounds (peaks 1, 4 and 5) were quantified (standard curves and validation data are shown in Table 2). Peaks 2 and 3, due to lack of reference compounds, were tentatively identified as galloylglucoside isomer 1 and galloylglucoside isomer 2, respectively, and semi-quantified using a standard curve of gallic acid, considering their similar maximum absorption wavelength. As is evidenced from the data from the chemical profile and quantitative analysis, the five major chemical components and their contents in different batches of BR samples varied greatly (Table 3), raising interest in exploring the effective marker of antioxidant activity in BR.

Table 3 Contents (mg g⁻¹, n = 6)^a of 5 major compounds in BR collected from different regions

Sample	Gallic acid	Galloylglucoside 1	Galloylglucoside 2	Catechin	Chlorogenic acid	Total content
a Data are represented as the mean ± SD.
S1	1.817 ± 0.01	1.655 ± 0.003	1.703 ± 0.003	1.995 ± 0.016	12.66 ± 0.065	19.83 ± 0.094
S2	2.429 ± 0.008	1.535 ± 0.096	1.365 ± 0.007	0.777 ± 0.038	10.724 ± 0.066	16.831 ± 0.047
S3	0.618 ± 0.004	0.174 ± 0.001	0.169 ± 0.002	0.525 ± 0.011	4.602 ± 0.036	6.088 ± 0.053
S4	1.851 ± 0.017	2.059 ± 0.071	2.036 ± 0.01	4.012 ± 0.027	16.95 ± 0.067	26.909 ± 0.07
S5	1.593 ± 0.007	1.549 ± 0.023	1.424 ± 0.013	0.838 ± 0.007	14.07 ± 0.051	19.474 ± 0.053
S6	1.073 ± 0.004	1.656 ± 0.004	1.714 ± 0.002	2.489 ± 0.038	15.601 ± 0.023	22.533 ± 0.043
S7	0.304 ± 0.003	0.085 ± 0.001	0.075 ± 0.000	0.8 ± 0.025	4.908 ± 0.02	6.173 ± 0.043
S8	0.899 ± 0.006	0.572 ± 0.002	0.502 ± 0.009	0.715 ± 0.004	6.503 ± 0.012	9.192 ± 0.024
S9	3.107 ± 0.018	1.412 ± 0.006	1.347 ± 0.005	1.919 ± 0.019	14.947 ± 0.056	22.732 ± 0.102
S10	1.035 ± 0.002	4.361 ± 0.016	4.232 ± 0.014	0.721 ± 0.002	6.326 ± 0.013	16.675 ± 0.039
S11	0.764 ± 0.001	0.112 ± 0.001	0.107 ± 0.001	0.697 ± 0.002	1.404 ± 0.002	3.085 ± 0.006
S12	1.28 ± 0.004	1.486 ± 0.043	1.476 ± 0.02	1.42 ± 0.028	13.196 ± 0.461	18.858 ± 0.545
S13	0.281 ± 0.011	0.091 ± 0.001	0.08 ± 0.001	0.786 ± 0.014	4.677 ± 0.039	5.914 ± 0.053
S14	1.217 ± 0.002	1.092 ± 0.002	1.000 ± 0.005	0.961 ± 0.028	11.193 ± 0.033	15.464 ± 0.049
S15	2.285 ± 0.007	1.733 ± 0.009	1.585 ± 0.008	1.173 ± 0.01	12.908 ± 0.021	19.684 ± 0.026
S16	0.15 ± 0.000	0.082 ± 0.000	0.077 ± 0.000	0.542 ± 0.006	4.555 ± 0.009	5.405 ± 0.006
S17	0.387 ± 0.001	0.078 ± 0.001	0.067 ± 0.000	0.641 ± 0.005	3.065 ± 0.005	4.237 ± 0.006
S18	0.291 ± 0.000	0.077 ± 0.000	0.078 ± 0.000	0.76 ± 0.001	5.631 ± 0.014	6.837 ± 0.014
S19	0.266 ± 0.004	0.077 ± 0.000	0.076 ± 0.000	0.641 ± 0.017	3.374 ± 0.027	4.435 ± 0.041
S20	0.257 ± 0.003	0.081 ± 0.002	0.075 ± 0.001	1.779 ± 0.033	5.372 ± 0.083	7.564 ± 0.07
S21	1.761 ± 0.001	2.341 ± 0.009	2.138 ± 0.01	1.096 ± 0.005	15.44 ± 0.023	22.776 ± 0.038
S22	0.729 ± 0.006	1.621 ± 0.022	1.581 ± 0.02	1.433 ± 0.013	12.623 ± 0.169	17.987 ± 0.223
S23	0.428 ± 0.003	0.101 ± 0.001	0.095 ± 0.000	0.941 ± 0.019	1.393 ± 0.008	2.958 ± 0.026
S24	0.288 ± 0.003	0.154 ± 0.001	0.148 ± 0.001	0.806 ± 0.016	1.409 ± 0.018	2.807 ± 0.039
S25	0.404 ± 0.003	0.083 ± 0.000	0.086 ± 0.004	0.156 ± 0.003	2.04 ± 0.006	2.768 ± 0.015
S26	0.276 ± 0.001	0.162 ± 0.003	0.163 ± 0.001	0.727 ± 0.004	1.269 ± 0.129	2.597 ± 0.132
S27	1.281 ± 0.006	0.788 ± 0.002	0.698 ± 0.001	0.846 ± 0.006	6.135 ± 0.027	9.749 ± 0.026
S28	1.798 ± 0.008	0.755 ± 0.009	0.665 ± 0.004	0.435 ± 0.001	4.934 ± 0.007	8.587 ± 0.02
S29	1.000 ± 0.02	1.08 ± 0.006	0.949 ± 0.006	0.44 ± 0.001	5.14 ± 0.044	8.609 ± 0.037
S30	0.608 ± 0.001	0.604 ± 0.002	0.525 ± 0.002	0.944 ± 0.008	5.11 ± 0.003	7.79 ± 0.008
S31	0.445 ± 0.02	0.141 ± 0.001	0.131 ± 0.000	0.548 ± 0.006	2.792 ± 0.03	4.057 ± 0.037
S32	1.138 ± 0.003	0.226 ± 0.001	0.199 ± 0.001	1.371 ± 0.004	2.849 ± 0.027	5.781 ± 0.035
S33	1.968 ± 0.01	1.659 ± 0.002	1.514 ± 0.003	2.065 ± 0.009	14.526 ± 0.009	21.732 ± 0.024
S34	2.595 ± 0.002	1.467 ± 0.006	1.343 ± 0.008	2.322 ± 0.014	12.714 ± 0.067	20.44 ± 0.095
S35	0.899 ± 0.007	1.375 ± 0.004	1.253 ± 0.007	1.382 ± 0.012	13.33 ± 0.298	18.24 ± 0.32

---

3.2. Total phenols (TP) and total extractable tannins (TET)

Many phenols are present in BR, however, there is no commercial standard available for most of them. Thus, the content of total phenols and total extractable tannins were assayed besides the five major ones. TP content is an important test index when evaluating antioxidant activity of herbal medicines due to their scavenging ability.¹⁹ TET is a specific class of TP. The content of TET is also an important test index for quality control in Chinese Pharmacopoeia¹⁸ as well as in European Pharmacopoeia.²⁰ Both TP and TET contents were measured in order to obtain more reliable and comparable results.

The content of TP in BR extracts varied between different sources and ranged from 25.6880 to 152.1607 mg GAE per g. The batch of S21 had the highest total phenol content (152.1607 mg GAE per g) whereas S17 had the lowest (25.6880 mg GAE per g), showing an almost 6-fold difference. Total tannin content of the BR extracts varied greatly and ranged from 2.8105 to 102.8645 mg GAE per g. The BR raw material from Liaoning Province (S21) had the highest total tannin amount (102.8645 mg GAE per g), whereas the BR material from Jiangsu Province (S23) had the lowest (2.8105 mg GAE per g), showing an almost 37-fold difference. To the best of our knowledge, the content of TP and TET of BR extract has not yet been evaluated. With the standard solutions of gallic acid equivalents, the TP and TET contents are given in Table 1. Based on these results, we can conclude that the content varies greatly from batch to batch.

3.3. Antioxidant properties of BR extracts

A total antioxidant capacity assay using one chemical reaction seems to be rather unrealistic and easy to come by, yet the biggest problem is the lack of a validated assay that can reliably measure the antioxidant capacity of foods, and plant and biological samples.²¹ Because the “antioxidant activity” measured by individual assay reflects only the chemical reactivity under the specific conditions applied in that assay, it is inappropriate and misleading to generalize the data as indicators of “total antioxidant activity”.^22,23 To be consistent with actual conditions, the “capacity” should be the results obtained by different assays. Considering the multifaceted factors of antioxidants and their reactivity,²⁴ the antioxidant properties of BR extracts were determined as free radical-scavenging ability (DPPH and ABTS methods), and reducing power (FRAP assay) in the present study (Table 1 and Fig. 2), however, no report is yet available on the antioxidant assay of BR extracts.

Fig. 2 The antioxidant capacity of BR extracts determined by DPPH assay (A), ABTS assay (B) and FRAP assay (C).

3.3.1. Scavenging of DPPH and ABTS radicals. Some methods such as free radical-scavenging assays might provide information on how capable an antioxidant is in preventing reactive radical species from reaching lipoproteins, polyunsaturated fatty acid, DNA, amino acids, proteins and sugars in biological and plant systems.²² DPPH radical is more stable than hydroxyl and superoxide radicals, therefore, DPPH has been widely used in evaluation of antioxidant activity single compound, as well as of plant extracts. As the reaction between antioxidant molecules and radicals progresses, the absorbance of the reaction system decreases. Hence, the change of absorbance is used as a measurement for the scavenging of DPPH radicals.²⁵ The DPPH radical scavenging ability of extracts and standard were expressed as IC₅₀ values (Fig. 2A). The BR extracts were able to interact with DPPH efficiently and quickly (with IC₅₀ < 91 μg mL⁻¹), and the more rapidly the absorbance changes, the more potential antioxidant activity the samples possess. Differences for DPPH IC₅₀ between various batches might be related to differences in effective components content of analyzed extracts.

The potential of BR extracts to scavenge free radicals was also assessed by their ability to quench ABTS˙⁺. Fig. 1B depicts the concentration-dependent decolorization of ABTS˙⁺, expressed as the IC₅₀ value. Compared with DPPH scavenging, these data indicated the higher capacity of BR extracts. According to the literature, the antioxidant activities against ABTS or DPPH were correlated with the concentration, chemical structures, and polymerization degree of organ antioxidants.¹ It was reported that the reaction of DPPH with eugenol was reversible. This would result in falsely low readings for antioxidant capacity of samples containing eugenol and other phenols bearing a similar structure type (O-methoxyphenol).²⁶ Accidentally, syringic acid and 2,6-dimethoxyphenol, compounds with similar structures to eugenol, exist in BR extracts.²⁷ In other words, this could be the reason why the correlation between TP and the antioxidant activity of the DPPH assay was not as significant as that of the ABTS and FRAP assays.

These results indicated that all tested BR extracts possessed radical scavenging capacity (Fig. 2A and B). The extracts of batches 8, 10 and 21 were determined to have the strongest radical scavenging activity in the DPPH assay (IC₅₀ values were: 9.3395 ± 0.2790, 9.8618 ± 0.0713 and 8.8378 ± 0.0444 μg mL⁻¹, respectively), and the extracts of batches 10, 21 and 33 were determined to have the strongest radical scavenging activity in the ABTS assay (IC₅₀ values were: 842.8758 ± 29.2474, 916.6795 ± 1.9447 and 861.2609 ± 25.8427 μg mL⁻¹, respectively). While the extracts of batches 17, 20 and 23 were determined to have the weakest activity (IC₅₀ values in DPPH assay were: 90.5082 ± 0.7056, 52.9639 ± 0.3024 and 67.5449 ± 0.4322 μg mL⁻¹, respectively; IC₅₀ values in ABTS assay were: 5992.9238 ± 208.3781, 5753.4787 ± 24.4100 and 5587.1468 ± 135.5964 μg mL⁻¹, respectively). There might be monomeric phenolic compounds such as catechin, epicatechin-3-O-gallate and procyanidins, which are responsible for the antioxidant activity.⁴

3.3.2. Reducing power (FRAP). The reducing power test, in which the capacity of breaking radical chain reactions is reflected, is considered to be a good indicator of antioxidant capacity.²⁸ Herein a ferric salt, Fe(III)(TPTZ)₂Cl₃ (TPTZ = 2,4,6-tripyridyl-s-triazine), is used as an oxidant. The presence of reducer (i.e. antioxidants) causes the conversion of the Fe³⁺ complex to the ferrous form.⁴ In this assay, the yellow colour of the test solution changes to various shades of green depending on the reducing power of the extract. The reducing capacity of a BR extract may serve as an indicator of its potential antioxidant activity. The reducing power of BR extracts was expressed as the concentration of ferric ion which has been deoxidized. Among the samples, the reducing capacity of 35 batches of BR extract exhibited a similar tendency with the results from the DPPH and ABTS assays, with slight differences (Fig. 2C). The extracts of batches 4, 10 and 21 were determined to have the strongest reducing capacity (the concentration of Fe²⁺: 1.1886 ± 0.0362, 1.2570 ± 0.0100 and 1.2097 ± 0.0172 μmol mL⁻¹), while batches 17, 20 and 23 were determined to have the weakest reducing capacity (the concentration of Fe²⁺: 0.2441 ± 0.0011, 0.1935 ± 0.0151 and 0.2897 ± 0.0070 μmol mL⁻¹). To achieve a better understanding of the trends and relationships among the chemical profile and bioactivity of the BR extract, statistical analysis was applied.

3.4. Correlations between antioxidant activities and the content of total phenols (TP)

Principal phenols are always responsible for the antioxidant activity, contrary to expectation, the antioxidant capacity incorrelated with the five major phenolic compounds completely (Fig. 3). For example, the extract of S16 with the lowest content of five major compounds, had a high free-radical scavenging activity, while the extract of S1 with a much higher content of five major compounds, had a considerably lower activity. The results may indicate that other minor constituents might possess a strong antioxidant effect. The identification of minor phenolic compounds including three types of polyphenol skeleton present in the BR extracts was achieved using HPLC-QTOF MS in our preliminary work.⁸

Fig. 3 The correlation between the five principal compounds and antioxidant activity in the DPPH assay (A); the correlation between the five principal compounds and antioxidant activity in the ABTS assay (B); the correlation between the five principal compounds and antioxidant activity in the FRAP assay (C).

Fig. 4 showed that TP contents in BR extracts varied greatly from different batches. Generally, samples having a higher TP content were more effective in scavenging the ABTS and DPPH radical cation, such as S21 with the strongest antioxidant activity, while both have the highest TP content and TET content. Similar results were obtained in the reducing power assay, and the high antioxidant activity of the BR extract may be explained by its higher TP content. The correlation between phenols of BR extracts and the antioxidant activities were determined by a linear regression analysis. There is a very significant positive correlation between the content of TP and the IC₅₀ of DPPH (R² = 0.5501) (Fig. 4A-2), ABTS (R² = 0.6387) (Fig. 4B-2) and FRAP (R² = 0.6722) (Fig. 4C-2) in BR extracts. These results were consistent with some literature reports.^4,11,13 Some reports found phenolic compounds are likely to be responsible for the measured antioxidant capacity of foods and herb medicines.^29,30 Therefore, the high content of TP in most BR extracts might be responsible for the strong antioxidant properties of this herbal medicine. The finding could be explained by a synergism of various low molar mass phenols, or by a more important influence of other specific classes of trace phenols¹¹ which were not examined in this study. The antioxidant capacity of the analyzed samples was also influenced by the TET. The antioxidant activity tends to be higher in analyzed samples with a higher TET content. As shown in Fig. 5, the antioxidant activity also presented a significant positive correlation with TET content (R² = 0.3540 in DPPH assay; R² = 0.4334 in ABTS assay; R² = 0.5605 in FRAP assay). For example, S10 with an extremely high TET exhibited a strong antioxidant activity. However, compared with some other samples with less TET, S10 did not show a much better activity than them, such as S4. From the results, we found that phenols, exhibited antioxidant activity. Although both TP and TET are very important influencing factors for the antioxidant activity of BR extracts, the antioxidant capacity is more severely correlated with TP than with TET content.

Fig. 4 The correlation between total phenols content and antioxidant activity in the DPPH assay (A-1); the linear regression analysis of total phenol content and antioxidant activity in the DPPH assay (A-2); the correlation between the total phenol content and antioxidant activity in the ABTS assay (B-1); the linear regression analysis of the total phenol content and antioxidant activity in the ABTS assay (B-2); the correlation between the total phenol content and antioxidant activity in the FRAP assay (C-1); the linear regression analysis of the total phenol content and antioxidant activity in the FRAP assay (C-2).

Fig. 5 The correlation between total extractable tannin content and antioxidant activity in the DPPH assay (A-1); the linear regression analysis of the total extractable tannin content and antioxidant activity in the DPPH assay (A-2); the correlation between the total extractable tannin content and the antioxidant activity in the ABTS assay (B-1); the linear regression analysis of the total extractable tannin content and antioxidant activity in the ABTS assay (B-2); the correlation between the total extractable tannin content and the antioxidant activity in the FRAP assay (C-1); the linear regression analysis of the total extractable tannin content and antioxidant activity in the FRAP assay (C-2).

In general, the antioxidant capacity of BR was found to be significantly correlated (p < 0.01) with TP, and insignificantly correlated with the five major constituents. With regard to these results, the selection of TP content as a bioactive marker for quality control of BR may be more appropriate than the five principal compounds or TET content.

4. Conclusions

In this work, we quantified five major compounds obtained from 35 BR extracts by HPLC-VWD and examined the TP, NPP and TET contents by a combination of spectrophotometric and precipitation methods. Simultaneously, the antioxidant capacity was determined by three methods. This work demonstrated that crude BR extracts exhibit an antioxidant effectiveness similar to reference samples, which already find applications as additives in food against oxidative deterioration in humans. The antioxidant activity very significantly correlated with the TP content, showing that TP may be responsible for the antioxidant activity of BR. To understand their mechanism of action as bioactive components, further fractionation of the phenolic compounds in the BR extracts and determination of their biological activities in vitro and in vivo is needed. We look forward to finding the effective components of antioxidant activity from BR extracts in a follow-up research. The five principal compounds could be the chemical markers of BR extract, and the TP content could be the anti-oxidative marker for quality control.

Acknowledgements

This work was financially supported by the Special Scientific Research for Traditional Chinese Medicine of State Administration of Traditional Chinese Medicine of China (No. 201307002), the National Natural Science Foundation of China (No. 81322051), and the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD).

References

1. R. Ksouri, H. Falleh, W. Megdiche, N. Trabelsi, B. Mhamdi, K. Chaieb, A. Bakrouf, C. Magne and C. Abdelly, Food Chem. Toxicol., 2009, 47, 2083–2091.
2. A. Moure, J. M. Cruz, D. Franco, J. M. Domínguez, J. Sineiro, H. Domínguez, M. J. Núñez and J. C. Parajo, Food Chem., 2001, 72, 145–171.
3. J. C. Espin, M. T. Garcia-Conesa and F. A. Tomas-Barberan, Phytochemistry, 2007, 68, 2986–3008.
4. M. S. Tounsi, I. Ouerghemmi, W. A. Wannes, R. Ksouri, H. Zemni, B. Marzouk and M. E. Kchouk, Ind. Crops Prod., 2009, 30, 292–296.
5. X. Q. Liu, L. L. Du, W. W. Li, H. F. Guan and F. Liu, Chem. Pharm. Bull., 2008, 56, 75–78.
6. X. Q. Liu, W. W. Li, H. M. Hua, X. Sha, F. K. Chen and L. J. Wu, Chin. Tradit. Herb. Drugs, 2006, 37, 1476–1478.
7. X. Q. Liu, W. W. Li, K. X. Sheng, J. Liu and F. K. Chen, J. Shenyang Pharm. Univ., 2006, 23, 15–17.
8. S. T. Wang, H. Yang, W. Gao, H. J. Li and P. Li, J. Pharm. Biomed. Anal., 2016, 119, 91–98.
9. Z. Sroka and R. Franiczek, Adv. Clin. Exp. Med., 2008, 17, 275–283.
10. Z. Q. Shi, D. F. Song, R. Q. Li, H. Yang, L. W. Qi, G. Z. Xin, D. Q. Wang, H. P. Song, J. Chen, H. Hao and P. Li, J. Chromatogr. A, 2014, 1345, 78–85.
11. M. Royer, P. N. Diouf and T. Stevanovic, Food Chem. Toxicol., 2011, 49, 2180–2188.
12. Z. Xiang and Z. Ning, LWT--Food Sci. Technol., 2008, 41, 1189–1203.
13. L. L. Yu, S. Haley, J. Perret, M. Harris, J. Wilson and M. Qian, J. Agric. Food Chem., 2002, 50, 1619–1624.
14. R. Re, N. Pellegrini, A. Proteggente, A. Pannala, M. Yang and C. Rice-Evans, Free Radical Biol. Med., 1999, 26, 1231–1237.
15. P. Zhao, L. Duan, L. Guo, L. L. Dou, X. Dong, P. Zhou, P. Li and E.-H. Liu, Food Chem., 2015, 173, 54–60.
16. I. F. F. Benzie and Y. T. Szeto, J. Agric. Food Chem., 1999, 47, 633–636.
17. E. A. Ainsworth and K. M. Gillespie, Nat. Protoc., 2007, 2, 875–877.
18. The State Commission of Chinese Pharmacopoeia, Pharmacopoeia of Peoples Republic of China, Chemical Industry Press, Beijing, 2010.
19. F. Shahidi and P. Ambigaipalan, J. Funct. Foods, 2015, 18, 820–897.
20. European Pharmacopoeia Commission, European Pharmacopoeia 8.0, European Directorate for the Quality of Medicines, Strasbourg, 2014.
21. R. L. Prior, J. Funct. Foods, 2015, 18, 797–810.
22. D. J. Huang, B. X. Ou and R. L. Prionr, J. Agric. Food Chem., 2005, 53, 1841–1856.
23. K. M. Schaich, X. Tian and J. Xie, J. Funct. Foods, 2015, 18, 782–796.
24. F. Shahidi and Y. Zhong, J. Funct. Foods, 2015, 18, 757–781.
25. X. Wang, C. Li, D. Liang, Y. Zou, P. Li and F. Ma, J. Funct. Foods, 2015, 18, 1086–1094.
26. V. Bondet, W. Brand-Williams and C. Berset, LWT--Food Sci. Technol., 1997, 30, 609–615.
27. A. Intisar, L. Zhang, H. Luo, J. Boima Kiazolu, R. Zhang and W. Zhang, Afr. J. Tradit., Complementary Altern. Med., 2012, 10, 53–59.
28. R. Ksouri, W. Megdiche, H. Falleh, N. Trabelsi, M. Boulaaba, A. Smaoui and C. Abdelly, C. R. Biol., 2008, 331, 865–873.
29. E. M. Sánchez-Salcedo, P. Mena, C. García-Viguera, F. Hernández and J. J. Martínez, J. Funct. Foods, 2015, 18, 1039–1046.
30. N. Alshikh, A. C. de Camargo and F. Shahidi, J. Funct. Foods, 2015, 18, 1022.

---

Footnote

† These authors contributed equally to this work.

---