Journal of Chromatography B, 945–946 (2014) 68–74
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Enrichment and purification of total flavonoids from Flos Populi extracts with macroporous resins and evaluation of antioxidant activities in vitro Pengfei Wan a , Zunlai Sheng a,∗ , Qiang Han a , Yulin Zhao a , Guangdong Cheng b , Yanhua Li a,∗ a b
College of Veterinary Medicine, Northeast Agricultural University, Harbin 150030, PR China College of Life Science, Jiamusi University, Jiamusi, 148 Xuefu Road, Jiamusi 154007, PR China
a r t i c l e
i n f o
Article history: Received 30 July 2013 Accepted 17 November 2013 Available online 25 November 2013 Keywords: Total Flavonoids Flos Populi Macroporous resins Enrichment Antioxidant activity
a b s t r a c t Enrichment and purification of total flavonoids from Flos Populi extracts were studied using five macroporous resins. The static tests indicated that NKA-9 resin was appropriate and its adsorption data were well fitted to the Langmuir and Freundlich isotherms. To optimize the separation process, dynamic adsorption and desorption tests were carried out. The optimal adsorption parameters were initial concentrations in sample solution of 7.64 mg/mL, pH of 5.0, sample loading amount of 2.3 BV, flow rate of 2 BV/h, temperature of 25 ◦ C. The optimal desorption parameters were deionized water and 20% ethanol each 5 BV, then 60% ethanol of 10 BV, flow rate of 2 BV/h. After one run treatment with NKA-9 resin, the content of total flavonoids in the product increased from 11.38% to 53.41%, and the recovery yield was 82.24%. The results showed that NKA-9 resin revealed a good ability to enrichment total flavonoids from Flos Populi, and the method can be referenced for the enrichment of total flavonoids from other materials. The antioxidant activities of the purified flavonoids were further evaluated in vitro. It showed that the DPPH radical scavenging increased from 59.46% to 82.63% at different concentrations (0.06–0.14 mg/mL). At different concentrations (0.6–1.4 mg/mL), the hydroxyl radical scavenging increased from 35.39% to 74.12%. Moreover, the reducing ability and total oxidant capacity appeared to be dose-dependent of flavonoids. It indicated that the purified flavonoids can be used as a source of potential antioxidant. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Flos Populi is an important traditional Chinese medicine prepared from the male inflorescence of Populus tomentosa Carr. or Populus canadensis Moench (Salicaceae family) [1]. Traditionally, it is developed for detoxication and relief of fever in some Asian countries for many years. Currently, it is mainly used for the treatment of various inflammatory diseases and diarrhea in East Asian countries [2]. Comparatively, research on Flos Populi is scanty, and phytochemicals and possible medicinal uses are not fully explored. They are usually discarded as agricultural wastes or used as firewood by farmers, the total amount is huge. Hence, innovative technology concepts for the utilization of Flos Populi to provide the raw material for the manufacture of future-oriented products are desirable. It is reports that the major active ingredients in Flos Populi are flavonoids [1], and previous studies have optimized the extraction
∗ Corresponding authors. Tel.: +86 451 55191881; fax: +86 451 55190263. E-mail addresses: [email protected] (Z. Sheng), [email protected] (Y. Li). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.11.033
progress of the total flavonoids from the herb [3]. In view of these beneficial clinical and pharmacological effects, an efficient and lowcost technology for the enrichment of the total flavonoids from Flos Populi is needed. The preparative enrichment of active components from plant extracts is an important step in the manufacture of phytochemicalrich products. There are several methods, such as liquid–liquid extraction [4–6], membrane filtration [7–10], ion exchange [11], expanded bed adsorption [12–14] and resin adsorption [15], available in the literature for the enrichment and separation of active constituents. Comparatively, resin adsorption seems to be the most suitable method due to its low cost, high efficiency and simple procedure [16]. Therefore, there has been a growing interest in employing macroporous resins to separate bioactive components from crude extracts of herbal raw materials. For example, macroporous resins have been successfully used in the enrichment of rare stilbenes and coumarin from pigeon pea leaves extracts, astragalosides from Radix Astragali extracts, baicalin and wogonoside from Scutellaria baicalensis extracts, madecassoside and asiaticoside from Centella asiatica extracts [17–20]. However, there is no report on using macroporous resins to enrich and separate total flavonoids from Flos Populi extracts so far.
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
It is considered that oxidative damage is attributable to excess active oxygen species generated in the body. It is well known that reactive oxygen, as well as being naturally generated, is also generated by factors such as strong ultraviolet, radial ray irradiation, and stress [21,22]. In order to reduce damage to the human body and prolong the storage stability of foods, synthetic antioxidants are often used for industrial processing. However, the use of synthetic antioxidants in food products is being questioned [23]. Consumers have also become more cautious about the nutritional quality and safety of food additives. In response to the growing consumer demand, investigations on antioxidants from natural sources have gained interest. In general, the natural antioxidants mainly consist of many compounds including phenolic, nitrogen compounds and carotenoids. As a kind of phenolic compound, antioxidant activities of flavonoids have attracted extensive attention [24,25]. In the present study, five resins with different polarities were used to investigate the adsorption and desorption properties of total flavonoids, and to develop a simple and efficient process for preliminary enrichment and separation of total flavonoids from Flos Populi extracts with the optimal resin. The parameters influencing the adsorption and desorption properties of total flavonoids were optimized and the experimental equilibrium data at different temperatures were fitted to Langmuir and Freundlich isotherms. Additionally, the antioxidant activities of the purified flavonoids were evaluated in vitro, including 1, 1-diphenyl-2-picryl-hydrazyl (DPPH) free radical, hydroxyl free radical scavenging effects, reducing power and total antioxidant capability test. We hope that this study will be helpful to further exploit and utilize this resource.
2. Materials and methods 2.1. Materials and instruments Flos Populi was male inflorescence of P. tomentosa Carr., which was collected from Heilongjiang province (China) and authenticated by Professor Yanbing Li (Heilongjiang University of traditional Chinese Medicine, Harbin, PR China). Reference compound Rutin (No. MUST-11040302) was purchased from the Chinese Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Deionized water was purified by a Milli-Q academic water purification system (Millipore, Bedford, MA, USA). DPPH, Deoxyribose, trichloride ferric, ethylene diamine tetraacetic acid (EDTA), H2 O2 , ascorbate acid, and thiobarbituric acid (TBA) were purchased from Sigma Chemical Co. (St. Louis, MO). Sodium dihydrogen phosphate, disodium hydrogen phosphate, potassium ferricyanide and trichloroacetic acid were of analytical grade and purchased from Hangzhou Reagent Company (Hangzhou, PR China). Five macroporous resins including S-8, NKA-9, X-5, D101 and AB-8 were purchased from Nankai Hecheng S & T Co., Ltd. (Tianjin, China). Their physical and chemical properties were summarized in Table 1. The adsorbent beads were pretreated to remove the monomers and porogenic agents trapped inside the pores during the synthesis process. The resins were soaked in 95% ethanol for 24 h and then washed by circumfluence until there was no residue after distillation. Prior to use, the resins were wet with ethanol and then washed with deionized water until the ethanol was thoroughly replaced by deionized water. The moisture contents of the tested resins were determined by drying the beads at 70 ◦ C to constant weight in a drying oven for over 24 h. UV-1700 spectrophotometer (Shimadzu Corporation, Japan) was used for analysis of total flavonoids, RE-52AA rotary evaporator (Shanghai Yarong Biochemistry Instrument Factory, Shanghai, China) was used for concentration of samples.
69
2.2. Preparation of sample solutions The male inflorescences of P. tomentosa Carr. (200 g) were extracted by 2000 mL of deionized water at 100 ◦ C for 2 h, repeated twice. The extracts were combined and purified by membrane filtration and then vacuum concentrated by a rotary evaporator at 60 ◦ C, and later lyophilized by a freeze dry system (Labconco, Model Freezone 6, MO, USA). The content of flavonoids in extracts was 11.38%. Deionized water was added into crude extracts to get sample solutions. The initial concentrations of sample solutions were 2.55, 5.10, 7.64, 10.19 and 12.74 mg/mL, respectively. 2.3. Determination of total flavonoids content The content of total flavonoids was determined by the colorimetric method [26–28] with some modifications. 1 mL of diluted solution containing flavonoids and 1 mL of 5% (w/w) NaNO2 were mixed for 6 min, and then 1 mL of 10% AlCl3 (w/w) was added and mixed, 6 min later, 10 mL of 1 mol/L NaOH was added. With 15 min standing, the absorbance of the solution was measured at 510 nm with UV-1700 spectrophotometer against the same mixture, without the sample as a blank. The calibration curve (y = − 0.0184 + 0.01251x, where y was absorbance value of sample, x was sample concentration) ranged 8.16–48.96 g/mL (R2 = 0.9993). 2.4. Static adsorption and desorption tests 2.4.1. Adsorption and desorption capacities, desorption ratio The adsorption tests were performed using the method reported by Toor et al. [29] with minor modifications. Briefly, 1 g (dry weight) samples of hydrated test resins were put into flasks with a lid; 50 mL of sample solution was added (the initial concentrations of total flavonoids were 7.64 mg/mL). The flasks were then shaken (120 rpm) at 25 ◦ C for 24 h. After adsorption equilibrium was reached, the resins were washed by deionized water and then desorbed with 50 mL of 70% ethanol solution. The flasks were continually shaken (120 rpm) for 24 h. The selectivity of resins was based on the capacities of adsorption (Qe ), capacities of desorption (Qd ), and ratio of desorption (D), which were quantified according to the following equations: Qe = (C0 − Ce ) ×
Qd = D=
Vi W
Cd × Vd W
Cd × Vd × 100% (C0 − Ce ) × Vi
(1)
(2)
(3)
where Qe was the adsorption capacity at adsorption equilibrium (mg/g dry resin). Qd was the desorption capacity after adsorption equilibrium (mg/g dry resin). C0 , Ce , and Cd were the initial, absorption equilibrium and desorption concentrations of analyte in the solutions, respectively (mg/mL). Vi and Vd were the volume of the initial sample and desorption solution (mL), respectively. W was the dry weight of resin (g), and D was the desorption ratio (%). The impact of pH on the adsorption capacities of total flavonoids was carried out by mixing 1 g (dry weight) of hydrated selected resin with sample solutions (50 mL each) in the pH range of 3.0–7.0. The sample pH was adjusted to the desired value with hydrochloric acid or ammonia solution. Then, the flasks were shaken at 25 ◦ C for 4 h. 2.4.2. Adsorption kinetics The adsorption kinetics curves for total flavonoids on the selected NKA-9 and AB-8 resins were studied by mixing 50 mL
70
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
Table 1 Physical properties of test macroporous resins. Trade name
Surface area (m2 /g)
Average pore diameter (nm)
Particle diameter (mm)
Polarity
S-8 NKA-9 X-5 D-101 AB-8
100–120 250–290 500–600 480–520 480–520
28–30 15.5–16.5 29–30 25–28 13–14
0.3–1.25 0.3–1.25 0.3–1.25 0.3–1.25 0.3–1.25
Polar Polar Non-polar Non-polar Weak-polar
sample solutions (the initial concentration of total flavionids was 7.64 mg/mL) with pre-weighed amounts of hydrated resin (equal to 1 g dry resin) in 100 mL flasks. The flasks were then shaken (120 rpm) at 25 ◦ C for 8 h [18]. The concentrations of the total flavonoids in adsorption solution were monitored at different time intervals until equilibrium. 2.4.3. Adsorption isotherms In order to investigate the effect of initial concentration and temperature on the total flavonoids absorption, experiments of adsorption isotherm on NKA-9 resin were performed. Five aliquots of 50 mL sample solutions at different concentrations were contacted with pre-weighed amounts of hydrated resins (equal to 1 g dry resin) in a constant temperature shaker at 25, 35 and 45 ◦ C for 5 h. The Langmuir and Freundlich models were used to evaluate the adsorption behavior between adsorbate and adsorbent [18,29,30]. Langmuir isotherms: Qe =
Qm × KL × Ce 1 + KL × Ce
(4)
Freundlich isotherms: /1n
Qe = KF × Ce
(5)
where Qe and Ce were the same as those in Eqs. (1), KL (mg/mL) and Qm (mg/g resin) were the Langmuir constants. KL was the dissociation constant of the adsorption interaction and Qm was the maximum adsorption capacity. KF (mg/mL) and n were the Freundlich constants. 2.5. Dynamic adsorption and desorption tests Dynamic adsorption and desorption tests [30] were carried out on glass columns (300 mm × 15 mm i.d.) wet-packed with 5 g (dry weight) of NKA-9 resin. The bed volume (BV) of resin was 30 mL. The effect of pH value of sample solution and the dynamic breakthrough tests were carried out first. After adsorption equilibrium, the column was firstly washed with deionized water, and then eluted with different concentrations ethanol at the flow rate of 2.0 BV/h. The elution volume of each ethanol concentration was adjusted under the guidance of absorbance.
2.7. Evaluation of antioxidant activities in vitro of flavonoids 2.7.1. Scavenging activity on DPPH radical The free radical scavenging activity of the purified flavonoids from Flos Populi was measured by DPPH test according to the reported method with some modifications [31–33]. The 0.4 mmol/L solution of DPPH in 95% ethanol was prepared daily before UV measurements. One milliliter of the purified flavonoids of different addition quantity in water was thoroughly mixed with 2 mL of freshly prepared DPPH and 2 mL of 95% ethanol. The mixture was shaken vigorously and left to stand for 30 min in the dark, and the absorbance was then measured at 517 nm against a blank. Lower absorbance of the reaction mixture indicated higher free radical scavenging activity, which was analyzed from the graph plotted of inhibition percentage against compound concentration. Ascorbic acid was used as positive controls. The experiment was performed in three times and averaged. The capability to scavenge the DPPH radical was calculated by the following equation: I% =
1 − (Ai − Aj ) Ac
× 100%
(6)
where Ac is the absorbance of DPPH solution without sample (2 mL DPPH + 2 mL of 95% ethanol); Ai is the absorbance of the test sample mixed with DPPH solution (2 mL sample + 2 mL DPPH) and Aj is the absorbance of the sample without DPPH solution (2 mL sample + 2 mL of 95% ethanol). 2.7.2. Scavenging activity on hydroxyl radical assay Assessment of the scavenging ability of the purified flavonoids on hydroxyl radicals was performed by the reported method with some modifications [34]. Reaction mixtures in a final volume of 1.0 mL contained deoxyribose (60 mM), phosphate buffer (pH 7.4, 20 mM), ferric trichloride (100 M), EDTA (100 M), H2 O2 (1 mM), ascorbic acid (100 M) and different concentrations of the purified flavonoids (0, 0.25, 0.5, 1, 2, or 4 mg/mL). Solutions of ferric trichloride and ascorbic acid were made immediately before use. The reaction solution was incubated for 1 h at 37 ◦ C, and then 1 mL of 1% TBA and 1 mL of 20% (v/v) HCl were added to the mixture. The mixture was boiled for 15 min and cooled on ice. Deionized water and ascorbic acid served as blank and positive control, respectively. The absorbance of the resulting mixture was measured at 532 nm. The scavenging activity of hydroxyl radical (%) was calculated according to the following equation: A532(blank) − A532(sample)
2.6. Laboratory preparative-scale separation
Scavenging effect(%) =
Under the above optimal conditions, laboratory preparativescale separation could be performed [18]. The dried extract of Flos Populi (100 g) was dissolved in deionized water. The sample solution (pH 5.0) was applied to a glass column (100 cm × 7.5 cm i.d.) containing 1.0 kg of wet NKA-9 macroporous resin) with a bed volume of 2.2 L. Initially, the column was washed by 5 BV of deionized water, and then 5 BV of 20% ethanol was used to remove the high polar impurities. The column was then eluted with 10 BV of 60% ethanol to obtain the total flavonoids-rich fraction. The flow rate of each gradient elution was set at about 2 BV/h, and the eluate of 60% ethanol was collected, concentrated and freeze-dried.
where A532(blank) was the absorbance of the control (deionized water, instead of sample); A532(sample) was the absorbance of the test sample mixed with reaction solution.
A532(blank)
× 100
(7)
2.7.3. Reducing power The determination of reducing power was carried out as described by Zou et al. [35]. A 2.4 mL extract sample (varying concentration) was mixed with 2.5 mL of 0.2 mol/L phosphate buffer (pH 6.6) and 2.5 mL of 1% (w/v) potassium ferricyanide. After the mixture was incubated at 50 ◦ C for 20 min, 2.5 mL of 10% (w/v) trichloroacetic acid was added, and the mixture was centrifuged at
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
71
Fig. 1. Adsorption, desorption capacities and desorption ratio of the total flavonoids on different resins.
3000 rpm for 10 min. The upper layer of 5 mL solution was mixed with 4 mL distilled water and 1 mL of 0.1% (w/v) ferric chloride for 10 min, and then the absorbance was measured at 700 nm against a blank. Increasing absorbance of the reaction mixture indicates increasing reducing power. The experiment was performed in three times and averaged. 2.7.4. Total antioxidant capacity Total oxidant capacity was determined using commercial kit (Nanjing Jianchen Bioengineering Institute, Nanjing, China) [31]. The absorbance of sample was measured at 520 nm. One unit of total antioxidant capacity was defined as per millilitre of extract sample that caused an increase of 0.01 in the absorbance per min at 37 ◦ C. The experiment was performed in three times and averaged. The total antioxidant capacity (TAC) was described by the following equation: TAC (U/mL) =
(As − Ao ) × Vt × Vs × C 30 × 0.01
(8)
where As is absorbance of the sample, Ao is absorbance of control, Vt (mL) is total volume of reaction mixture, Vs (mL) is volume of sample, and C is diluted times of sample. 3. Results and discussion
but also because of higher surface area, and which may correlate with the capability of the resins and the chemical features of the adsorbed substance. The adsorption capacities of the total flavonoids on NKA-9 or AB-8 resins was 150.19 or 150.10 mg/g dry resin with desorption ratio of 90.37% or 92.21%, respectively. Therefore, adsorption kinetics experiments were further carried out on NKA-9 and AB-8 resins. 3.1.2. Adsorption kinetics on resins The kinetics of adsorption that describes the solute uptake rate governing the contact time of the sorption reaction is one of the important characteristics that define the efficiency of sorption [36]. Hence, in the present study, the adsorption kinetics of the total flavonoids was determined to understand the adsorption behavior of NKA-9 or AB-8 resins. Generally, the adsorption capacity of the total flavonoids increased transiently with adsorption time before reaching the adsorption equilibrium [17]. The equilibrium time for the total flavonoids was 5 h on NKA-9 and was 20 h on AB-8 resin, respectively. Comparing the two resins in a comprehensive consideration of the adsorption capacity and desorption ratio, NKA-9 resin possessed a great advantage for the total flavonoids. Thus, NKA-9 resin was selected as the most suitable resin for the enrichment of the total flavonoids and was used in the following tests.
3.1. Optimization of operational parameters for total flavonoids 3.1.1. Adsorption and desorption capacities, desorption ratio According to the “like dissolve like” rule, given flavonoids contain non-polar phenyl group and polar multi-hydroxyl groups, either non-polar resins or polar resins were applicable to adsorption of flavonoids. So, five macroporous resins ranging from non-polarity to polarity were employed for enrichment the total flavonoids. Adsorption properties of five macroporous resins towards the total flavonoids were exhibited in Fig. 1. The adsorption and desorption performances of different resins for the total flavonoids were distinct. Although the adsorption capacity of S-8 resin towards the total flavonoids was higher than that of NKA-9 and AB-8 resins, the lower desorption capacity resulted in a lower desorption ratio. However, the desorption ratios of D-101 and X-5 resins for the total flavonoids were close to that of NKA-9 and AB-8 resins, but their adsorption and desorption capacities were lower. Both D-101 and X-5 resins have the following characteristics in common: they are polar and have a surface area similar. In a sum, compared with other resins, NKA-9 and AB-8 resins possessed better adsorption and desorption capabilities for the total flavonoids. The two resins exhibited better adsorption capacity and desorption ratio not only because of its appropriate polarity,
3.1.3. Adsorption isotherms To investigate the adsorption capacity and characterize the adsorption behavior of the total flavonoids, sample solutions with various concentrations of the total flavonoids (2.55–12.74 mg/mL) were shaken with NKA-9 resin at 25, 35 and 45 ◦ C. The adsorption capacity increased with the initial concentration, and reached the saturation plateau when the initial concentration of the total flavonoids was 7.64 mg/mL (Fig. 2a). Thus, the initial concentration of the total flavonoids in the sample solution for adsorption was selected as 7.64 mg/mL. Additionally, for interpretation of the adsorption experimental data, the Langmuir and Freundlich models were used. Table 2 listed the two isotherm equations at different temperatures and two important parameters: Qm value (obtained from the Langmuir isotherm) and 1/n value (obtained from the Freundlich isotherm). The correlations (0.9965–0.9969) for total flavonoids on NKA-9 indicated that the two models were suitable for describing the tested adsorption system in the concentration ranges studied. Generally, in the Freundlich equation, the adsorption was easy to carry out when 1/n value was between 0.1 and 0.5, and it was difficult to take place if 1/n value was above 1 [37]. It can be seen from Table 2, the 1/n values were all between 0.1 and 0.3, indicated that the adsorption of the total flavonoids on NKA-9 resin took place
72
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
Fig. 2. Adsorption isotherms (a) of the total flavonoids on NKA-9 resin at 25, 35 and 45 ◦ C and dynamic breakthrough curve (b) of the total flavonoids on column packed with NKA-9 resin. Table 2 Langmuir and Freundlich parameters of total flavonoids on NKA-9 resin at different temperature. Temperature (◦ C) 25 35 45
Langmuir equation Ce /Qe = 0.0044Ce + 0.0075 Ce /Qe = 0.005Ce + 0.0071 Ce /Qe = 0.0058Ce + 0.0056
R2 0.9968 0.9965 0.9969
easily. So, the NKA-9 resin was proper to the enrichment of the total flavonoids. Within the ranges of temperatures investigated, the adsorption capacities decreased with the temperature increase (Fig. 2a), which indicated the adsorption process was a thermopositive process [38]. Similar results were obtained for the enrichment other compounds using macroporous resin [39]. Therefore, 25 ◦ C was selected in the following experiments. 3.1.4. Effect of pH value of sample solution A critical parameter affecting the adsorption capacity is the pH value. The pH value influences the extent of ionization of solutes, thus affecting the affinity between the solutes and solutions [39]. As shown in Table 3, the highest adsorption capacity for the total flavonoids appeared at the pH values of 5.0, and then decreased with the increase of pH value. These results indicated that hydrogen bonding may play an important role in the adsorption process on NKA-9 resin. At a higher pH value, the hydrogen bonding interactions were reduced, because the phenolic hydroxyl groups in the total flavonoids dissociated to H+ and their corresponding anions, thus resulting in the lower adsorption capacity. Consequently, the pH value of sample solution was adjusted to 5.0 for the following tests. 3.1.5. Dynamic breakthrough curve on NKA-9 resin The dynamic breakthrough curve on NKA-9 resin was obtained based on the volume of effluent liquid and the concentration of solute herein. As shown in Fig. 2b, the total flavonoids in solution was almost completely adsorbed by the NKA-9 resin before 50 mL, and then the concentration of the total flavonoids in leak solution Table 3 Effect of sample solution pH value and flow rate on the adsorption capacity of the total flavonoids on NKA-9 resin. pH value
Adsorption capacity (mg/g)
Flow rate
Adsorption capacity (mg/g)
3 5 7
55.51 ± 1.06 148.26 ± 2.73 109.61 ± 2.16
1.0 BV/h 2.0 BV/h 3.0 BV/h
140.14 ± 2.39 138.96 ± 2.74 116.37 ± 2.68
Qm (mg/g resin)
Freundlich equation
R2
1/n
227.27 200 172.41
Qe = 105.15Ce0.2731 Qe = 105.14Ce0.2268 Qe = 105.81Ce 0.1739
0.9945 0.9988 0.9958
0.2731 0.2268 0.1739
increased rapidly until it reached a steady plateau in 200 mL. Generally, the 10% ratio of the exit to the inlet solute concentration was defined as breakthrough point [40]. According to this standard, the breakthrough volume of the total flavonoids on NKA-9 resin was 70 mL (2.3 BV). Therefore, the 70 mL (2.3 BV) of sample solution on NKA-9 resin was determined as saturated adsorption volume for dynamic adsorption. In addition, the effect of flow rate on dynamic adsorption capacity was carried out with constant loading volume 2.3 BV. As shown in Table 3, the adsorption capacity decreased a little as the flow rate increased from 1.0 to 2.0 BV/h, but decreased greatly as the flow rate increased from 2.0 to 3.0 BV/h, this was due to the interaction time of flavonoids with active sites of the resin surface was deceased. Thus, taking the efficiency into account, the flow rate for loading sample was maintained constantly at 2.0 BV/h. 3.1.6. Dynamic desorption on NKA-9 resin Dynamic desorption was carried out with gradient elution mode at the flow rate of 2.0 BV/h. Different elution solvents with the same volume of 5 BV were used to desorb the total flavonoids when the sample loading amount was 2.3 BV. At the 20% ethanol, the total flavonoids were hardly desorbed. When the ethanol concentration was over 20%, the desorption ability increased sharply and reached a peak value at 60% ethanol. Hence, a gradient elution procedure with 20% and 60% ethanol at a flow rate of 2 BV/h was applied for desorption of the total flavonoids. In addition, the elution volume of each ethanol concentration was modified under the guidance of absorbance. The dried extract of Flos Populi (2.82 g) was dissolved in deionized water. The total flavonoids were enriched and purified from the sample solution with AKA-9 resin. According to content analysis of total flavonoids, the elution scheme was modified as 5 BV deionized water, 5 BV 20% ethanol and 10 BV of 60% ethanol. As shown in Table 4, non-adsorption components were firstly washed out by 5 BV deionized water. Subsequently, some impurities were removed by elution of 5 BV 20% ethanol. The following 10 BV of 60% ethanol elution gave the product rich in the total flavonoids. After separation on NKA-9 resin by the gradient elution, the content of the total flavonoids reached 53.41%, which was 4.69-fold to those in extract of Flos Populi, and the recovery yields were 82.24%. Thus,
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
73
Fig. 3. Scavenging effects on (a) DPPH radical and (b) hydroxyl radical of different total flavonoids concentration and ascorbic acid. Values are presented as mean ± SD (n = 3). Table 4 Results of gradient elution of the total flavonoids on column packed with NAK-9 resin. Concentration of ethanol (%)
Mass of dried residue (mg)
Mass of the total flavonoids (mg)
Content of the total flavonoids (%)
Recovery of the total flavonoids (%)
0 20 60 70 80
45.1 86.3 493.3 33.1 20.6
0.07 0.24 263.47 0 0
0.15 0.28 53.41 0 0
82.24
the final separation and purification conditions for total flavonoids were determined as follows: Adsorption: concentrations of total flavonoids in sample solution 7.64 mg/mL; pH 5.0; sample loading amount 2.3 BV; flow rate 2 BV/h; temperature 25 ◦ C. Desorption: deionized water and 20% ethanol each 5 BV, then 60% ethanol 10 BV, flow rate 2 BV/h. 3.1.7. Laboratory preparative-scale separation Laboratory preparative-scale separation was carried out on a NKA-9 resin (1 kg, wet weight) column using the above determined conditions. As shown in Table 5, the elution of 10 BV 60% ethanol gave the total flavonoids-rich fraction with a content of (55.12 ± 2.15) % and a recovery yield of (81.61 ± 2.69) %, which was close to the values of the pilot trial above, indicating that the model was adequate for the enrichment process. 3.2. Antioxidant activity analysis 3.2.1. Scavenging activity on DPPH radical DPPH is a free radical compound and has been used to test free radical scavenging ability of various samples widely in present studies. In the DPPH test, the antioxidants reduce the DPPH radical to a yellow-colored compound, diphenylpicrylhydrazine, and the extent of the reaction will depend on the hydrogen donating ability of the antioxidants [41]. The scavenging activities of the purified flavonoids on DPPH free radical were shown in Fig. 3a and compared with ascorbic acid as Table 5 Purification result of the total flavonoids on column packed with NKA-9 resin by final modified conditions. Process
Weight of solids (g)
Purity of the total flavonoids (%)
Recovery of the total flavonoids (%)
Crude extract 60% ethanol fraction
100.0 g 17.39 ± 0.57
11.38 55.12 ± 2.15
81.61 ± 2.69
control standards. It can be seen from Fig. 3a that the DPPH radical scavenging increased from 59.46 ± 0.72% to 82.63 ± 1.18% when the concentration of flavonoids increased from 0.06 to 0.14 mg/mL. The DPPH radical-scavenging effect of the purified flavonoids was observed to be 59.46 ± 0.72% in a concentration of 0.06 mg/mL, which was relatively more prominent than that of ascorbic acid (50.58 ± 0.86%) at the same concentration. The results indicated that the purified flavonoids had strong scavenging activity on DPPH radical. 3.2.2. Scavenging activity on hydroxyl radical of purified flavonoids The hydroxyl radical, generated by the Fenton reaction in the system, was scavenged by the purified flavonoids from Flos Populi extracts. The hydroxyl radical scavenging effects of the purified flavonoids and ascorbic acid were increased with the increase of concentration up to 1.4 mg/mL (Fig. 3b). It can be seen from Fig. 3b that the hydroxyl radical scavenging increased from 35.39 ± 1.45% to 74.12 ± 1.75% when the concentration of purified flavonoids increased from 0.6 to 1.4 mg/mL. However, ascorbic acid showed higher hydroxyl radical scavenging activity than purified flavonoids. The results demonstrated that purified flavonoids possessed noticeable scavenging activities on hydroxyl radical. 3.2.3. Reducing power of purified flavonoids from Flos Populi extracts As we have known, it is better to evaluate the antioxidant capacities on a selected antioxidant in a different test system. During the reducing power assay, the presence of purified flavonoids in the tested samples would result in reducing Fe3+ /ferricyanide complex to the ferrous form (Fe2+ ). Therefore, the Fe2+ can be monitored by measuring the formation of Perl’s Prussian blue at 700 nm [35]. It is common that higher absorbance value means stronger reducing ability of samples. In this test, the absorbance values of samples increased rapidly from 0.177 at 0.1 mg/mL to 0.856 at 0.9 mg/mL. The equation of absorbance value (y) and amount of the flavonoids (x) was y = 0.11115 + 0.8465x (R2 = 0.9957), indicating that absorbance value correlated well with amount of the
74
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
flavonoids. It indicated that the purified flavonoids from Flos Populi extracts exhibited effective reducing ability which increased with an increased concentration of samples. 3.2.4. Total antioxidant capacity of purified flavonoids from Flos Populi extracts Total oxidant capacity was determined using commercial kit. In this test, the total oxidant capacity of samples increased rapidly from 15/mL at 0.6 mg/mL to 77/mL at 1.4 mg/mL. The equation of total oxidant capacity (y) and amount of the purified flavonoids (x) was y = −31.43407 + 77.47833x (R2 = 0.9952), indicating that it exhibited a strong dependence of the total antioxidant capacity on the concentration of purified flavonoids. The results obtained from the test systems indicated that the purified flavonoids from Flos Populi extracts had significant antioxidant activities, and the effects increase with increasing the concentration of purified flavonoids. 4. Conclusion The present study reported experimental data on the enrichment and separation of the total flavonoids from extracts of Flos Populi at various operating parameters, using five macroporous resins differing in chemical and physical properties. The equilibrium experimental data of the adsorption of the total flavonoids on NKA-9 resin at 25, 30 and 45 ◦ C were well-fitted to Langmuir and Freundlich isotherms. The most effective resin (NKA-9) was successfully applied to obtain a product of the total flavonoids with higher content. The antioxidant activities of the purified flavonoids in vitro including DPPH radical scavenging activity, hydroxyl radical scavenging activity, reducing power and total antioxidant capacity were evaluated. Based on the results, it is revealed that the purified flavonoids from Flos Populi extracts exhibited significant antioxidant activities. The knowledge gained from this study should be useful for further exploitation and application of the resource. Acknowledgements This research was supported by the National Natural Science Foundation of China (Grant No. 31201944), Science and Technology Project Founded by the Education Department of Heilongjiang Province (Grant No. 12531047) and Universities Animal Medical Key Laboratory Open Project of Heilongjiang Province (Grant No. AMKL2012008). References [1] National Commission of Chinese Pharmacopoeia, Pharmacopoeia of People’s Republic of China, 2010 ed., Chemical Industry Press, Beijing, 2010.
[2] Q.Q. Xu, Z.Q. Shen, Y.B. Wang, S.J. Guo, F. Li, Y.P. Wang, C.F. Zhou, J. Ethnopharmacol. (2013), http://dx.doi.org/10.1016/j.jep.2013.05.021. [3] Z.L. Sheng, P.F. Wan, C.L. Dong, Y.H. Li, Ind. Crop. Prod. 43 (2013) 778–786. [4] T.S. Ho, S. Pedersen-Bjergaard, K.E. Rasmussen, J. Chromatogr. A 963 (2002) 3–17. [5] J.P.M. Biazus, J.C.C. Santana, R.R. Souza, E. Jordão, E.B. Tambourgi, J. Chromatogr. B 858 (2007) 227–233. [6] G.B. Ferreira, A.F. Evangelista, J.B. Severo Júnior, R.R. Souza, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 50 (2007) 557–564. [7] M. Linder, E. Matouba, J. Fanni, M. Parmentier, Eur. J. Lipid Sci. Technol. 104 (2002) 455–462. [8] J.B. Severo Jr., S.S. Almeida, N. Narain, R.R. Souza, J.C.C. Santana, E.B. Tambourgi, Process. Biochem. 42 (2007) 1416–1520. [9] J.B. Severo Júnior, L.S.S. Oliveira, F.S. Sardeiro, R.R. Souza, F.L.G. Lopes, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 50 (2007) 713–718. [10] F.L.G. Lopes, J.B. Severo Júnior, R.R. Souza, D.D. Ehrhardt, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 52 (2009) 457–464. [11] N.W. Smith, M.B. Evans, Chromatographia 41 (1995) 197–203. [12] J.P.M. Biazus, J.B. Severo Jr., J.C.C. Santana, R.R. Souza, E.B. Tambourgi, Process. Biochem. 41 (2006) 1786–1791. [13] G.S. Padilha, J.C. Curvelo-Santana, R.M. Alegre, A.B. Tambourgi, J. Chromatogr. B 877 (2009) 521–526. [14] A.L. Toledo, J.B. Severo Jr., R.R. Souza, E.S. Campos, J.C.C. Santana, E.B. Tambourgi, J. Chromatogr. B 846 (2007) 51–56. [15] L. Peng, Z.M. Li, L.H. Lu, Med. Plant 1 (2010) 51–53. [16] Y. Sun, H.Y. Yuan, L.L. Hao, C.Y. Min, J.N. Cai, J.Y. Liu, P.L. Cai, S.L. Yang, Food Chem. 141 (2013) 533–541. [17] G.T. Jia, X.Y. Lu, J. Chromatogr. A 1193 (2008) 136–141. [18] Z.Q. Du, K. Wang, Y. Tao, L.X. Chen, F. Qiu, J. Chromatogr. B 908 (2012) 143–149. [19] Z.F. Wei, Y.G. Zu, Y.J. Fu, W. Wang, C.J. Zhao, M. Luo, T. Efferth, Chem. Eng. J. 172 (2011) 864–871. [20] Y. Kong, M.M. Yan, W. Liu, C.Y. Chen, B.S. Zhao, Y.G. Zu, Y.J. Fu, M. Luo, M. Wink, J. Sep. Sci. 33 (2010) 2278–2286. [21] D.Y. Zhang, Y.G. Zu, Y.J. Fu, W. Wang, L. Zhang, M. Luo, F.S. Mu, X.H. Yao, M.H. Duan, Sep. Purif. Technol. 102 (2013) 26–33. [22] K. Sowndhararajan, J.M. Joseph, S. Manian, Int. J. Food Prop. 16 (2013) 1717–1729. [23] L. Zhang, J. Chen, Y. Wang, D. Wu, M. Xu, Molecules 15 (2010) 3567–3577. [24] A.S. Awaad, D.J. Maitland, R. Donia Ael, S.I. Alqasoumi, G.A. Soliman, Pharm. Biol. 50 (2012) 99–104. [25] C. Prommuak, W. De-Eknamkul, A. Shotipruk, Sep. Purif. Technol. 62 (2008) 444–448. [26] G.Y. Pan, G.Y. Yu, C.H. Zhu, J.L. Qiao, Ultrason. Sonochem. 19 (2012) 486–490. [27] J. Liu, S.Y. Lin, Z.Z. Wang, C.N. Wang, E. Wang, Y. Zhang, J.B. Liu, Food Bioprod. Process 89 (2011) 333–339. [28] X.S. Wang, Q.N. Wu, Y.F. Wu, G.Y. Chen, W. Yue, Q.L. Liang, Molecules 17 (2012) 6769–6783. [29] M. Toor, B. Jin, Chem. Eng. J. 187 (2012) 79–88. [30] Y.J. Yan, X.Y. Zhang, D.W. Chen, Bioresour. Technol. 131 (2013) 179–187. [31] G. Zhang, L. He, M. Hu, Innov. Food Sci. Emerg. Technol. 12 (2011) 18–25. [32] S. Gonc¸alves, D. Gomes, P. Costa, A. Romano, Ind. Crop. Prod. 43 (2013) 465–471. [33] R. Ho, A. Violette, D. Cressend, P. Raharivelomanana, P.A. Carrupt, K. Hostettmann, Nat. Prod. Res. 26 (2012) 274–277. [34] J. Gutteridge, Biochem. J. 243 (1987) 709. [35] Y. Zou, Y. Lu, D. Wei, J. Agric. Food Chem. 52 (2004) 5032–5039. [36] Y.G. Bi, Y.Q. Tan, Adv. Mater. Res. 550 (2012) 987–992. [37] X.J. Li, Y.J. Fu, M. Luo, W. Wang, L. Zhang, C.J. Zhao, Y.G. Zu, J. Pharmaceut. Biomed. 61 (2012) 199–206. [38] V. Ferraro, I.B. Cruz, R.F. Jorge, M.E. Pintado, P.M. Castro, J. Chem. Eng. Data 58 (2013) 707–717. [39] Z.Y. Zhao, L.L. Dong, Y.L. Wu, F. Lin, Food Bioprod. Process. 89 (2011) 266–272. [40] Y. Zhang, J.J. Jiao, C.M. Liu, X.Q. Wu, Food Chem. 107 (2008) 1326–1336. [41] J. Song, D. Li, C. Liu, Y. Zhang, Innov. Food Sci. Emerg. Technol. 12 (2011) 282–287.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Enrichment and purification of total flavonoids from Flos Populi extracts with macroporous resins and evaluation of antioxidant activities in vitro Pengfei Wan a , Zunlai Sheng a,∗ , Qiang Han a , Yulin Zhao a , Guangdong Cheng b , Yanhua Li a,∗ a b
College of Veterinary Medicine, Northeast Agricultural University, Harbin 150030, PR China College of Life Science, Jiamusi University, Jiamusi, 148 Xuefu Road, Jiamusi 154007, PR China
a r t i c l e
i n f o
Article history: Received 30 July 2013 Accepted 17 November 2013 Available online 25 November 2013 Keywords: Total Flavonoids Flos Populi Macroporous resins Enrichment Antioxidant activity
a b s t r a c t Enrichment and purification of total flavonoids from Flos Populi extracts were studied using five macroporous resins. The static tests indicated that NKA-9 resin was appropriate and its adsorption data were well fitted to the Langmuir and Freundlich isotherms. To optimize the separation process, dynamic adsorption and desorption tests were carried out. The optimal adsorption parameters were initial concentrations in sample solution of 7.64 mg/mL, pH of 5.0, sample loading amount of 2.3 BV, flow rate of 2 BV/h, temperature of 25 ◦ C. The optimal desorption parameters were deionized water and 20% ethanol each 5 BV, then 60% ethanol of 10 BV, flow rate of 2 BV/h. After one run treatment with NKA-9 resin, the content of total flavonoids in the product increased from 11.38% to 53.41%, and the recovery yield was 82.24%. The results showed that NKA-9 resin revealed a good ability to enrichment total flavonoids from Flos Populi, and the method can be referenced for the enrichment of total flavonoids from other materials. The antioxidant activities of the purified flavonoids were further evaluated in vitro. It showed that the DPPH radical scavenging increased from 59.46% to 82.63% at different concentrations (0.06–0.14 mg/mL). At different concentrations (0.6–1.4 mg/mL), the hydroxyl radical scavenging increased from 35.39% to 74.12%. Moreover, the reducing ability and total oxidant capacity appeared to be dose-dependent of flavonoids. It indicated that the purified flavonoids can be used as a source of potential antioxidant. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Flos Populi is an important traditional Chinese medicine prepared from the male inflorescence of Populus tomentosa Carr. or Populus canadensis Moench (Salicaceae family) [1]. Traditionally, it is developed for detoxication and relief of fever in some Asian countries for many years. Currently, it is mainly used for the treatment of various inflammatory diseases and diarrhea in East Asian countries [2]. Comparatively, research on Flos Populi is scanty, and phytochemicals and possible medicinal uses are not fully explored. They are usually discarded as agricultural wastes or used as firewood by farmers, the total amount is huge. Hence, innovative technology concepts for the utilization of Flos Populi to provide the raw material for the manufacture of future-oriented products are desirable. It is reports that the major active ingredients in Flos Populi are flavonoids [1], and previous studies have optimized the extraction
∗ Corresponding authors. Tel.: +86 451 55191881; fax: +86 451 55190263. E-mail addresses: [email protected] (Z. Sheng), [email protected] (Y. Li). 1570-0232/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jchromb.2013.11.033
progress of the total flavonoids from the herb [3]. In view of these beneficial clinical and pharmacological effects, an efficient and lowcost technology for the enrichment of the total flavonoids from Flos Populi is needed. The preparative enrichment of active components from plant extracts is an important step in the manufacture of phytochemicalrich products. There are several methods, such as liquid–liquid extraction [4–6], membrane filtration [7–10], ion exchange [11], expanded bed adsorption [12–14] and resin adsorption [15], available in the literature for the enrichment and separation of active constituents. Comparatively, resin adsorption seems to be the most suitable method due to its low cost, high efficiency and simple procedure [16]. Therefore, there has been a growing interest in employing macroporous resins to separate bioactive components from crude extracts of herbal raw materials. For example, macroporous resins have been successfully used in the enrichment of rare stilbenes and coumarin from pigeon pea leaves extracts, astragalosides from Radix Astragali extracts, baicalin and wogonoside from Scutellaria baicalensis extracts, madecassoside and asiaticoside from Centella asiatica extracts [17–20]. However, there is no report on using macroporous resins to enrich and separate total flavonoids from Flos Populi extracts so far.
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
It is considered that oxidative damage is attributable to excess active oxygen species generated in the body. It is well known that reactive oxygen, as well as being naturally generated, is also generated by factors such as strong ultraviolet, radial ray irradiation, and stress [21,22]. In order to reduce damage to the human body and prolong the storage stability of foods, synthetic antioxidants are often used for industrial processing. However, the use of synthetic antioxidants in food products is being questioned [23]. Consumers have also become more cautious about the nutritional quality and safety of food additives. In response to the growing consumer demand, investigations on antioxidants from natural sources have gained interest. In general, the natural antioxidants mainly consist of many compounds including phenolic, nitrogen compounds and carotenoids. As a kind of phenolic compound, antioxidant activities of flavonoids have attracted extensive attention [24,25]. In the present study, five resins with different polarities were used to investigate the adsorption and desorption properties of total flavonoids, and to develop a simple and efficient process for preliminary enrichment and separation of total flavonoids from Flos Populi extracts with the optimal resin. The parameters influencing the adsorption and desorption properties of total flavonoids were optimized and the experimental equilibrium data at different temperatures were fitted to Langmuir and Freundlich isotherms. Additionally, the antioxidant activities of the purified flavonoids were evaluated in vitro, including 1, 1-diphenyl-2-picryl-hydrazyl (DPPH) free radical, hydroxyl free radical scavenging effects, reducing power and total antioxidant capability test. We hope that this study will be helpful to further exploit and utilize this resource.
2. Materials and methods 2.1. Materials and instruments Flos Populi was male inflorescence of P. tomentosa Carr., which was collected from Heilongjiang province (China) and authenticated by Professor Yanbing Li (Heilongjiang University of traditional Chinese Medicine, Harbin, PR China). Reference compound Rutin (No. MUST-11040302) was purchased from the Chinese Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Deionized water was purified by a Milli-Q academic water purification system (Millipore, Bedford, MA, USA). DPPH, Deoxyribose, trichloride ferric, ethylene diamine tetraacetic acid (EDTA), H2 O2 , ascorbate acid, and thiobarbituric acid (TBA) were purchased from Sigma Chemical Co. (St. Louis, MO). Sodium dihydrogen phosphate, disodium hydrogen phosphate, potassium ferricyanide and trichloroacetic acid were of analytical grade and purchased from Hangzhou Reagent Company (Hangzhou, PR China). Five macroporous resins including S-8, NKA-9, X-5, D101 and AB-8 were purchased from Nankai Hecheng S & T Co., Ltd. (Tianjin, China). Their physical and chemical properties were summarized in Table 1. The adsorbent beads were pretreated to remove the monomers and porogenic agents trapped inside the pores during the synthesis process. The resins were soaked in 95% ethanol for 24 h and then washed by circumfluence until there was no residue after distillation. Prior to use, the resins were wet with ethanol and then washed with deionized water until the ethanol was thoroughly replaced by deionized water. The moisture contents of the tested resins were determined by drying the beads at 70 ◦ C to constant weight in a drying oven for over 24 h. UV-1700 spectrophotometer (Shimadzu Corporation, Japan) was used for analysis of total flavonoids, RE-52AA rotary evaporator (Shanghai Yarong Biochemistry Instrument Factory, Shanghai, China) was used for concentration of samples.
69
2.2. Preparation of sample solutions The male inflorescences of P. tomentosa Carr. (200 g) were extracted by 2000 mL of deionized water at 100 ◦ C for 2 h, repeated twice. The extracts were combined and purified by membrane filtration and then vacuum concentrated by a rotary evaporator at 60 ◦ C, and later lyophilized by a freeze dry system (Labconco, Model Freezone 6, MO, USA). The content of flavonoids in extracts was 11.38%. Deionized water was added into crude extracts to get sample solutions. The initial concentrations of sample solutions were 2.55, 5.10, 7.64, 10.19 and 12.74 mg/mL, respectively. 2.3. Determination of total flavonoids content The content of total flavonoids was determined by the colorimetric method [26–28] with some modifications. 1 mL of diluted solution containing flavonoids and 1 mL of 5% (w/w) NaNO2 were mixed for 6 min, and then 1 mL of 10% AlCl3 (w/w) was added and mixed, 6 min later, 10 mL of 1 mol/L NaOH was added. With 15 min standing, the absorbance of the solution was measured at 510 nm with UV-1700 spectrophotometer against the same mixture, without the sample as a blank. The calibration curve (y = − 0.0184 + 0.01251x, where y was absorbance value of sample, x was sample concentration) ranged 8.16–48.96 g/mL (R2 = 0.9993). 2.4. Static adsorption and desorption tests 2.4.1. Adsorption and desorption capacities, desorption ratio The adsorption tests were performed using the method reported by Toor et al. [29] with minor modifications. Briefly, 1 g (dry weight) samples of hydrated test resins were put into flasks with a lid; 50 mL of sample solution was added (the initial concentrations of total flavonoids were 7.64 mg/mL). The flasks were then shaken (120 rpm) at 25 ◦ C for 24 h. After adsorption equilibrium was reached, the resins were washed by deionized water and then desorbed with 50 mL of 70% ethanol solution. The flasks were continually shaken (120 rpm) for 24 h. The selectivity of resins was based on the capacities of adsorption (Qe ), capacities of desorption (Qd ), and ratio of desorption (D), which were quantified according to the following equations: Qe = (C0 − Ce ) ×
Qd = D=
Vi W
Cd × Vd W
Cd × Vd × 100% (C0 − Ce ) × Vi
(1)
(2)
(3)
where Qe was the adsorption capacity at adsorption equilibrium (mg/g dry resin). Qd was the desorption capacity after adsorption equilibrium (mg/g dry resin). C0 , Ce , and Cd were the initial, absorption equilibrium and desorption concentrations of analyte in the solutions, respectively (mg/mL). Vi and Vd were the volume of the initial sample and desorption solution (mL), respectively. W was the dry weight of resin (g), and D was the desorption ratio (%). The impact of pH on the adsorption capacities of total flavonoids was carried out by mixing 1 g (dry weight) of hydrated selected resin with sample solutions (50 mL each) in the pH range of 3.0–7.0. The sample pH was adjusted to the desired value with hydrochloric acid or ammonia solution. Then, the flasks were shaken at 25 ◦ C for 4 h. 2.4.2. Adsorption kinetics The adsorption kinetics curves for total flavonoids on the selected NKA-9 and AB-8 resins were studied by mixing 50 mL
70
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
Table 1 Physical properties of test macroporous resins. Trade name
Surface area (m2 /g)
Average pore diameter (nm)
Particle diameter (mm)
Polarity
S-8 NKA-9 X-5 D-101 AB-8
100–120 250–290 500–600 480–520 480–520
28–30 15.5–16.5 29–30 25–28 13–14
0.3–1.25 0.3–1.25 0.3–1.25 0.3–1.25 0.3–1.25
Polar Polar Non-polar Non-polar Weak-polar
sample solutions (the initial concentration of total flavionids was 7.64 mg/mL) with pre-weighed amounts of hydrated resin (equal to 1 g dry resin) in 100 mL flasks. The flasks were then shaken (120 rpm) at 25 ◦ C for 8 h [18]. The concentrations of the total flavonoids in adsorption solution were monitored at different time intervals until equilibrium. 2.4.3. Adsorption isotherms In order to investigate the effect of initial concentration and temperature on the total flavonoids absorption, experiments of adsorption isotherm on NKA-9 resin were performed. Five aliquots of 50 mL sample solutions at different concentrations were contacted with pre-weighed amounts of hydrated resins (equal to 1 g dry resin) in a constant temperature shaker at 25, 35 and 45 ◦ C for 5 h. The Langmuir and Freundlich models were used to evaluate the adsorption behavior between adsorbate and adsorbent [18,29,30]. Langmuir isotherms: Qe =
Qm × KL × Ce 1 + KL × Ce
(4)
Freundlich isotherms: /1n
Qe = KF × Ce
(5)
where Qe and Ce were the same as those in Eqs. (1), KL (mg/mL) and Qm (mg/g resin) were the Langmuir constants. KL was the dissociation constant of the adsorption interaction and Qm was the maximum adsorption capacity. KF (mg/mL) and n were the Freundlich constants. 2.5. Dynamic adsorption and desorption tests Dynamic adsorption and desorption tests [30] were carried out on glass columns (300 mm × 15 mm i.d.) wet-packed with 5 g (dry weight) of NKA-9 resin. The bed volume (BV) of resin was 30 mL. The effect of pH value of sample solution and the dynamic breakthrough tests were carried out first. After adsorption equilibrium, the column was firstly washed with deionized water, and then eluted with different concentrations ethanol at the flow rate of 2.0 BV/h. The elution volume of each ethanol concentration was adjusted under the guidance of absorbance.
2.7. Evaluation of antioxidant activities in vitro of flavonoids 2.7.1. Scavenging activity on DPPH radical The free radical scavenging activity of the purified flavonoids from Flos Populi was measured by DPPH test according to the reported method with some modifications [31–33]. The 0.4 mmol/L solution of DPPH in 95% ethanol was prepared daily before UV measurements. One milliliter of the purified flavonoids of different addition quantity in water was thoroughly mixed with 2 mL of freshly prepared DPPH and 2 mL of 95% ethanol. The mixture was shaken vigorously and left to stand for 30 min in the dark, and the absorbance was then measured at 517 nm against a blank. Lower absorbance of the reaction mixture indicated higher free radical scavenging activity, which was analyzed from the graph plotted of inhibition percentage against compound concentration. Ascorbic acid was used as positive controls. The experiment was performed in three times and averaged. The capability to scavenge the DPPH radical was calculated by the following equation: I% =
1 − (Ai − Aj ) Ac
× 100%
(6)
where Ac is the absorbance of DPPH solution without sample (2 mL DPPH + 2 mL of 95% ethanol); Ai is the absorbance of the test sample mixed with DPPH solution (2 mL sample + 2 mL DPPH) and Aj is the absorbance of the sample without DPPH solution (2 mL sample + 2 mL of 95% ethanol). 2.7.2. Scavenging activity on hydroxyl radical assay Assessment of the scavenging ability of the purified flavonoids on hydroxyl radicals was performed by the reported method with some modifications [34]. Reaction mixtures in a final volume of 1.0 mL contained deoxyribose (60 mM), phosphate buffer (pH 7.4, 20 mM), ferric trichloride (100 M), EDTA (100 M), H2 O2 (1 mM), ascorbic acid (100 M) and different concentrations of the purified flavonoids (0, 0.25, 0.5, 1, 2, or 4 mg/mL). Solutions of ferric trichloride and ascorbic acid were made immediately before use. The reaction solution was incubated for 1 h at 37 ◦ C, and then 1 mL of 1% TBA and 1 mL of 20% (v/v) HCl were added to the mixture. The mixture was boiled for 15 min and cooled on ice. Deionized water and ascorbic acid served as blank and positive control, respectively. The absorbance of the resulting mixture was measured at 532 nm. The scavenging activity of hydroxyl radical (%) was calculated according to the following equation: A532(blank) − A532(sample)
2.6. Laboratory preparative-scale separation
Scavenging effect(%) =
Under the above optimal conditions, laboratory preparativescale separation could be performed [18]. The dried extract of Flos Populi (100 g) was dissolved in deionized water. The sample solution (pH 5.0) was applied to a glass column (100 cm × 7.5 cm i.d.) containing 1.0 kg of wet NKA-9 macroporous resin) with a bed volume of 2.2 L. Initially, the column was washed by 5 BV of deionized water, and then 5 BV of 20% ethanol was used to remove the high polar impurities. The column was then eluted with 10 BV of 60% ethanol to obtain the total flavonoids-rich fraction. The flow rate of each gradient elution was set at about 2 BV/h, and the eluate of 60% ethanol was collected, concentrated and freeze-dried.
where A532(blank) was the absorbance of the control (deionized water, instead of sample); A532(sample) was the absorbance of the test sample mixed with reaction solution.
A532(blank)
× 100
(7)
2.7.3. Reducing power The determination of reducing power was carried out as described by Zou et al. [35]. A 2.4 mL extract sample (varying concentration) was mixed with 2.5 mL of 0.2 mol/L phosphate buffer (pH 6.6) and 2.5 mL of 1% (w/v) potassium ferricyanide. After the mixture was incubated at 50 ◦ C for 20 min, 2.5 mL of 10% (w/v) trichloroacetic acid was added, and the mixture was centrifuged at
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
71
Fig. 1. Adsorption, desorption capacities and desorption ratio of the total flavonoids on different resins.
3000 rpm for 10 min. The upper layer of 5 mL solution was mixed with 4 mL distilled water and 1 mL of 0.1% (w/v) ferric chloride for 10 min, and then the absorbance was measured at 700 nm against a blank. Increasing absorbance of the reaction mixture indicates increasing reducing power. The experiment was performed in three times and averaged. 2.7.4. Total antioxidant capacity Total oxidant capacity was determined using commercial kit (Nanjing Jianchen Bioengineering Institute, Nanjing, China) [31]. The absorbance of sample was measured at 520 nm. One unit of total antioxidant capacity was defined as per millilitre of extract sample that caused an increase of 0.01 in the absorbance per min at 37 ◦ C. The experiment was performed in three times and averaged. The total antioxidant capacity (TAC) was described by the following equation: TAC (U/mL) =
(As − Ao ) × Vt × Vs × C 30 × 0.01
(8)
where As is absorbance of the sample, Ao is absorbance of control, Vt (mL) is total volume of reaction mixture, Vs (mL) is volume of sample, and C is diluted times of sample. 3. Results and discussion
but also because of higher surface area, and which may correlate with the capability of the resins and the chemical features of the adsorbed substance. The adsorption capacities of the total flavonoids on NKA-9 or AB-8 resins was 150.19 or 150.10 mg/g dry resin with desorption ratio of 90.37% or 92.21%, respectively. Therefore, adsorption kinetics experiments were further carried out on NKA-9 and AB-8 resins. 3.1.2. Adsorption kinetics on resins The kinetics of adsorption that describes the solute uptake rate governing the contact time of the sorption reaction is one of the important characteristics that define the efficiency of sorption [36]. Hence, in the present study, the adsorption kinetics of the total flavonoids was determined to understand the adsorption behavior of NKA-9 or AB-8 resins. Generally, the adsorption capacity of the total flavonoids increased transiently with adsorption time before reaching the adsorption equilibrium [17]. The equilibrium time for the total flavonoids was 5 h on NKA-9 and was 20 h on AB-8 resin, respectively. Comparing the two resins in a comprehensive consideration of the adsorption capacity and desorption ratio, NKA-9 resin possessed a great advantage for the total flavonoids. Thus, NKA-9 resin was selected as the most suitable resin for the enrichment of the total flavonoids and was used in the following tests.
3.1. Optimization of operational parameters for total flavonoids 3.1.1. Adsorption and desorption capacities, desorption ratio According to the “like dissolve like” rule, given flavonoids contain non-polar phenyl group and polar multi-hydroxyl groups, either non-polar resins or polar resins were applicable to adsorption of flavonoids. So, five macroporous resins ranging from non-polarity to polarity were employed for enrichment the total flavonoids. Adsorption properties of five macroporous resins towards the total flavonoids were exhibited in Fig. 1. The adsorption and desorption performances of different resins for the total flavonoids were distinct. Although the adsorption capacity of S-8 resin towards the total flavonoids was higher than that of NKA-9 and AB-8 resins, the lower desorption capacity resulted in a lower desorption ratio. However, the desorption ratios of D-101 and X-5 resins for the total flavonoids were close to that of NKA-9 and AB-8 resins, but their adsorption and desorption capacities were lower. Both D-101 and X-5 resins have the following characteristics in common: they are polar and have a surface area similar. In a sum, compared with other resins, NKA-9 and AB-8 resins possessed better adsorption and desorption capabilities for the total flavonoids. The two resins exhibited better adsorption capacity and desorption ratio not only because of its appropriate polarity,
3.1.3. Adsorption isotherms To investigate the adsorption capacity and characterize the adsorption behavior of the total flavonoids, sample solutions with various concentrations of the total flavonoids (2.55–12.74 mg/mL) were shaken with NKA-9 resin at 25, 35 and 45 ◦ C. The adsorption capacity increased with the initial concentration, and reached the saturation plateau when the initial concentration of the total flavonoids was 7.64 mg/mL (Fig. 2a). Thus, the initial concentration of the total flavonoids in the sample solution for adsorption was selected as 7.64 mg/mL. Additionally, for interpretation of the adsorption experimental data, the Langmuir and Freundlich models were used. Table 2 listed the two isotherm equations at different temperatures and two important parameters: Qm value (obtained from the Langmuir isotherm) and 1/n value (obtained from the Freundlich isotherm). The correlations (0.9965–0.9969) for total flavonoids on NKA-9 indicated that the two models were suitable for describing the tested adsorption system in the concentration ranges studied. Generally, in the Freundlich equation, the adsorption was easy to carry out when 1/n value was between 0.1 and 0.5, and it was difficult to take place if 1/n value was above 1 [37]. It can be seen from Table 2, the 1/n values were all between 0.1 and 0.3, indicated that the adsorption of the total flavonoids on NKA-9 resin took place
72
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
Fig. 2. Adsorption isotherms (a) of the total flavonoids on NKA-9 resin at 25, 35 and 45 ◦ C and dynamic breakthrough curve (b) of the total flavonoids on column packed with NKA-9 resin. Table 2 Langmuir and Freundlich parameters of total flavonoids on NKA-9 resin at different temperature. Temperature (◦ C) 25 35 45
Langmuir equation Ce /Qe = 0.0044Ce + 0.0075 Ce /Qe = 0.005Ce + 0.0071 Ce /Qe = 0.0058Ce + 0.0056
R2 0.9968 0.9965 0.9969
easily. So, the NKA-9 resin was proper to the enrichment of the total flavonoids. Within the ranges of temperatures investigated, the adsorption capacities decreased with the temperature increase (Fig. 2a), which indicated the adsorption process was a thermopositive process [38]. Similar results were obtained for the enrichment other compounds using macroporous resin [39]. Therefore, 25 ◦ C was selected in the following experiments. 3.1.4. Effect of pH value of sample solution A critical parameter affecting the adsorption capacity is the pH value. The pH value influences the extent of ionization of solutes, thus affecting the affinity between the solutes and solutions [39]. As shown in Table 3, the highest adsorption capacity for the total flavonoids appeared at the pH values of 5.0, and then decreased with the increase of pH value. These results indicated that hydrogen bonding may play an important role in the adsorption process on NKA-9 resin. At a higher pH value, the hydrogen bonding interactions were reduced, because the phenolic hydroxyl groups in the total flavonoids dissociated to H+ and their corresponding anions, thus resulting in the lower adsorption capacity. Consequently, the pH value of sample solution was adjusted to 5.0 for the following tests. 3.1.5. Dynamic breakthrough curve on NKA-9 resin The dynamic breakthrough curve on NKA-9 resin was obtained based on the volume of effluent liquid and the concentration of solute herein. As shown in Fig. 2b, the total flavonoids in solution was almost completely adsorbed by the NKA-9 resin before 50 mL, and then the concentration of the total flavonoids in leak solution Table 3 Effect of sample solution pH value and flow rate on the adsorption capacity of the total flavonoids on NKA-9 resin. pH value
Adsorption capacity (mg/g)
Flow rate
Adsorption capacity (mg/g)
3 5 7
55.51 ± 1.06 148.26 ± 2.73 109.61 ± 2.16
1.0 BV/h 2.0 BV/h 3.0 BV/h
140.14 ± 2.39 138.96 ± 2.74 116.37 ± 2.68
Qm (mg/g resin)
Freundlich equation
R2
1/n
227.27 200 172.41
Qe = 105.15Ce0.2731 Qe = 105.14Ce0.2268 Qe = 105.81Ce 0.1739
0.9945 0.9988 0.9958
0.2731 0.2268 0.1739
increased rapidly until it reached a steady plateau in 200 mL. Generally, the 10% ratio of the exit to the inlet solute concentration was defined as breakthrough point [40]. According to this standard, the breakthrough volume of the total flavonoids on NKA-9 resin was 70 mL (2.3 BV). Therefore, the 70 mL (2.3 BV) of sample solution on NKA-9 resin was determined as saturated adsorption volume for dynamic adsorption. In addition, the effect of flow rate on dynamic adsorption capacity was carried out with constant loading volume 2.3 BV. As shown in Table 3, the adsorption capacity decreased a little as the flow rate increased from 1.0 to 2.0 BV/h, but decreased greatly as the flow rate increased from 2.0 to 3.0 BV/h, this was due to the interaction time of flavonoids with active sites of the resin surface was deceased. Thus, taking the efficiency into account, the flow rate for loading sample was maintained constantly at 2.0 BV/h. 3.1.6. Dynamic desorption on NKA-9 resin Dynamic desorption was carried out with gradient elution mode at the flow rate of 2.0 BV/h. Different elution solvents with the same volume of 5 BV were used to desorb the total flavonoids when the sample loading amount was 2.3 BV. At the 20% ethanol, the total flavonoids were hardly desorbed. When the ethanol concentration was over 20%, the desorption ability increased sharply and reached a peak value at 60% ethanol. Hence, a gradient elution procedure with 20% and 60% ethanol at a flow rate of 2 BV/h was applied for desorption of the total flavonoids. In addition, the elution volume of each ethanol concentration was modified under the guidance of absorbance. The dried extract of Flos Populi (2.82 g) was dissolved in deionized water. The total flavonoids were enriched and purified from the sample solution with AKA-9 resin. According to content analysis of total flavonoids, the elution scheme was modified as 5 BV deionized water, 5 BV 20% ethanol and 10 BV of 60% ethanol. As shown in Table 4, non-adsorption components were firstly washed out by 5 BV deionized water. Subsequently, some impurities were removed by elution of 5 BV 20% ethanol. The following 10 BV of 60% ethanol elution gave the product rich in the total flavonoids. After separation on NKA-9 resin by the gradient elution, the content of the total flavonoids reached 53.41%, which was 4.69-fold to those in extract of Flos Populi, and the recovery yields were 82.24%. Thus,
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
73
Fig. 3. Scavenging effects on (a) DPPH radical and (b) hydroxyl radical of different total flavonoids concentration and ascorbic acid. Values are presented as mean ± SD (n = 3). Table 4 Results of gradient elution of the total flavonoids on column packed with NAK-9 resin. Concentration of ethanol (%)
Mass of dried residue (mg)
Mass of the total flavonoids (mg)
Content of the total flavonoids (%)
Recovery of the total flavonoids (%)
0 20 60 70 80
45.1 86.3 493.3 33.1 20.6
0.07 0.24 263.47 0 0
0.15 0.28 53.41 0 0
82.24
the final separation and purification conditions for total flavonoids were determined as follows: Adsorption: concentrations of total flavonoids in sample solution 7.64 mg/mL; pH 5.0; sample loading amount 2.3 BV; flow rate 2 BV/h; temperature 25 ◦ C. Desorption: deionized water and 20% ethanol each 5 BV, then 60% ethanol 10 BV, flow rate 2 BV/h. 3.1.7. Laboratory preparative-scale separation Laboratory preparative-scale separation was carried out on a NKA-9 resin (1 kg, wet weight) column using the above determined conditions. As shown in Table 5, the elution of 10 BV 60% ethanol gave the total flavonoids-rich fraction with a content of (55.12 ± 2.15) % and a recovery yield of (81.61 ± 2.69) %, which was close to the values of the pilot trial above, indicating that the model was adequate for the enrichment process. 3.2. Antioxidant activity analysis 3.2.1. Scavenging activity on DPPH radical DPPH is a free radical compound and has been used to test free radical scavenging ability of various samples widely in present studies. In the DPPH test, the antioxidants reduce the DPPH radical to a yellow-colored compound, diphenylpicrylhydrazine, and the extent of the reaction will depend on the hydrogen donating ability of the antioxidants [41]. The scavenging activities of the purified flavonoids on DPPH free radical were shown in Fig. 3a and compared with ascorbic acid as Table 5 Purification result of the total flavonoids on column packed with NKA-9 resin by final modified conditions. Process
Weight of solids (g)
Purity of the total flavonoids (%)
Recovery of the total flavonoids (%)
Crude extract 60% ethanol fraction
100.0 g 17.39 ± 0.57
11.38 55.12 ± 2.15
81.61 ± 2.69
control standards. It can be seen from Fig. 3a that the DPPH radical scavenging increased from 59.46 ± 0.72% to 82.63 ± 1.18% when the concentration of flavonoids increased from 0.06 to 0.14 mg/mL. The DPPH radical-scavenging effect of the purified flavonoids was observed to be 59.46 ± 0.72% in a concentration of 0.06 mg/mL, which was relatively more prominent than that of ascorbic acid (50.58 ± 0.86%) at the same concentration. The results indicated that the purified flavonoids had strong scavenging activity on DPPH radical. 3.2.2. Scavenging activity on hydroxyl radical of purified flavonoids The hydroxyl radical, generated by the Fenton reaction in the system, was scavenged by the purified flavonoids from Flos Populi extracts. The hydroxyl radical scavenging effects of the purified flavonoids and ascorbic acid were increased with the increase of concentration up to 1.4 mg/mL (Fig. 3b). It can be seen from Fig. 3b that the hydroxyl radical scavenging increased from 35.39 ± 1.45% to 74.12 ± 1.75% when the concentration of purified flavonoids increased from 0.6 to 1.4 mg/mL. However, ascorbic acid showed higher hydroxyl radical scavenging activity than purified flavonoids. The results demonstrated that purified flavonoids possessed noticeable scavenging activities on hydroxyl radical. 3.2.3. Reducing power of purified flavonoids from Flos Populi extracts As we have known, it is better to evaluate the antioxidant capacities on a selected antioxidant in a different test system. During the reducing power assay, the presence of purified flavonoids in the tested samples would result in reducing Fe3+ /ferricyanide complex to the ferrous form (Fe2+ ). Therefore, the Fe2+ can be monitored by measuring the formation of Perl’s Prussian blue at 700 nm [35]. It is common that higher absorbance value means stronger reducing ability of samples. In this test, the absorbance values of samples increased rapidly from 0.177 at 0.1 mg/mL to 0.856 at 0.9 mg/mL. The equation of absorbance value (y) and amount of the flavonoids (x) was y = 0.11115 + 0.8465x (R2 = 0.9957), indicating that absorbance value correlated well with amount of the
74
P. Wan et al. / J. Chromatogr. B 945–946 (2014) 68–74
flavonoids. It indicated that the purified flavonoids from Flos Populi extracts exhibited effective reducing ability which increased with an increased concentration of samples. 3.2.4. Total antioxidant capacity of purified flavonoids from Flos Populi extracts Total oxidant capacity was determined using commercial kit. In this test, the total oxidant capacity of samples increased rapidly from 15/mL at 0.6 mg/mL to 77/mL at 1.4 mg/mL. The equation of total oxidant capacity (y) and amount of the purified flavonoids (x) was y = −31.43407 + 77.47833x (R2 = 0.9952), indicating that it exhibited a strong dependence of the total antioxidant capacity on the concentration of purified flavonoids. The results obtained from the test systems indicated that the purified flavonoids from Flos Populi extracts had significant antioxidant activities, and the effects increase with increasing the concentration of purified flavonoids. 4. Conclusion The present study reported experimental data on the enrichment and separation of the total flavonoids from extracts of Flos Populi at various operating parameters, using five macroporous resins differing in chemical and physical properties. The equilibrium experimental data of the adsorption of the total flavonoids on NKA-9 resin at 25, 30 and 45 ◦ C were well-fitted to Langmuir and Freundlich isotherms. The most effective resin (NKA-9) was successfully applied to obtain a product of the total flavonoids with higher content. The antioxidant activities of the purified flavonoids in vitro including DPPH radical scavenging activity, hydroxyl radical scavenging activity, reducing power and total antioxidant capacity were evaluated. Based on the results, it is revealed that the purified flavonoids from Flos Populi extracts exhibited significant antioxidant activities. The knowledge gained from this study should be useful for further exploitation and application of the resource. Acknowledgements This research was supported by the National Natural Science Foundation of China (Grant No. 31201944), Science and Technology Project Founded by the Education Department of Heilongjiang Province (Grant No. 12531047) and Universities Animal Medical Key Laboratory Open Project of Heilongjiang Province (Grant No. AMKL2012008). References [1] National Commission of Chinese Pharmacopoeia, Pharmacopoeia of People’s Republic of China, 2010 ed., Chemical Industry Press, Beijing, 2010.
[2] Q.Q. Xu, Z.Q. Shen, Y.B. Wang, S.J. Guo, F. Li, Y.P. Wang, C.F. Zhou, J. Ethnopharmacol. (2013), http://dx.doi.org/10.1016/j.jep.2013.05.021. [3] Z.L. Sheng, P.F. Wan, C.L. Dong, Y.H. Li, Ind. Crop. Prod. 43 (2013) 778–786. [4] T.S. Ho, S. Pedersen-Bjergaard, K.E. Rasmussen, J. Chromatogr. A 963 (2002) 3–17. [5] J.P.M. Biazus, J.C.C. Santana, R.R. Souza, E. Jordão, E.B. Tambourgi, J. Chromatogr. B 858 (2007) 227–233. [6] G.B. Ferreira, A.F. Evangelista, J.B. Severo Júnior, R.R. Souza, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 50 (2007) 557–564. [7] M. Linder, E. Matouba, J. Fanni, M. Parmentier, Eur. J. Lipid Sci. Technol. 104 (2002) 455–462. [8] J.B. Severo Jr., S.S. Almeida, N. Narain, R.R. Souza, J.C.C. Santana, E.B. Tambourgi, Process. Biochem. 42 (2007) 1416–1520. [9] J.B. Severo Júnior, L.S.S. Oliveira, F.S. Sardeiro, R.R. Souza, F.L.G. Lopes, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 50 (2007) 713–718. [10] F.L.G. Lopes, J.B. Severo Júnior, R.R. Souza, D.D. Ehrhardt, J.C.C. Santana, E.B. Tambourgi, Braz. Arch. Biol. Technol. 52 (2009) 457–464. [11] N.W. Smith, M.B. Evans, Chromatographia 41 (1995) 197–203. [12] J.P.M. Biazus, J.B. Severo Jr., J.C.C. Santana, R.R. Souza, E.B. Tambourgi, Process. Biochem. 41 (2006) 1786–1791. [13] G.S. Padilha, J.C. Curvelo-Santana, R.M. Alegre, A.B. Tambourgi, J. Chromatogr. B 877 (2009) 521–526. [14] A.L. Toledo, J.B. Severo Jr., R.R. Souza, E.S. Campos, J.C.C. Santana, E.B. Tambourgi, J. Chromatogr. B 846 (2007) 51–56. [15] L. Peng, Z.M. Li, L.H. Lu, Med. Plant 1 (2010) 51–53. [16] Y. Sun, H.Y. Yuan, L.L. Hao, C.Y. Min, J.N. Cai, J.Y. Liu, P.L. Cai, S.L. Yang, Food Chem. 141 (2013) 533–541. [17] G.T. Jia, X.Y. Lu, J. Chromatogr. A 1193 (2008) 136–141. [18] Z.Q. Du, K. Wang, Y. Tao, L.X. Chen, F. Qiu, J. Chromatogr. B 908 (2012) 143–149. [19] Z.F. Wei, Y.G. Zu, Y.J. Fu, W. Wang, C.J. Zhao, M. Luo, T. Efferth, Chem. Eng. J. 172 (2011) 864–871. [20] Y. Kong, M.M. Yan, W. Liu, C.Y. Chen, B.S. Zhao, Y.G. Zu, Y.J. Fu, M. Luo, M. Wink, J. Sep. Sci. 33 (2010) 2278–2286. [21] D.Y. Zhang, Y.G. Zu, Y.J. Fu, W. Wang, L. Zhang, M. Luo, F.S. Mu, X.H. Yao, M.H. Duan, Sep. Purif. Technol. 102 (2013) 26–33. [22] K. Sowndhararajan, J.M. Joseph, S. Manian, Int. J. Food Prop. 16 (2013) 1717–1729. [23] L. Zhang, J. Chen, Y. Wang, D. Wu, M. Xu, Molecules 15 (2010) 3567–3577. [24] A.S. Awaad, D.J. Maitland, R. Donia Ael, S.I. Alqasoumi, G.A. Soliman, Pharm. Biol. 50 (2012) 99–104. [25] C. Prommuak, W. De-Eknamkul, A. Shotipruk, Sep. Purif. Technol. 62 (2008) 444–448. [26] G.Y. Pan, G.Y. Yu, C.H. Zhu, J.L. Qiao, Ultrason. Sonochem. 19 (2012) 486–490. [27] J. Liu, S.Y. Lin, Z.Z. Wang, C.N. Wang, E. Wang, Y. Zhang, J.B. Liu, Food Bioprod. Process 89 (2011) 333–339. [28] X.S. Wang, Q.N. Wu, Y.F. Wu, G.Y. Chen, W. Yue, Q.L. Liang, Molecules 17 (2012) 6769–6783. [29] M. Toor, B. Jin, Chem. Eng. J. 187 (2012) 79–88. [30] Y.J. Yan, X.Y. Zhang, D.W. Chen, Bioresour. Technol. 131 (2013) 179–187. [31] G. Zhang, L. He, M. Hu, Innov. Food Sci. Emerg. Technol. 12 (2011) 18–25. [32] S. Gonc¸alves, D. Gomes, P. Costa, A. Romano, Ind. Crop. Prod. 43 (2013) 465–471. [33] R. Ho, A. Violette, D. Cressend, P. Raharivelomanana, P.A. Carrupt, K. Hostettmann, Nat. Prod. Res. 26 (2012) 274–277. [34] J. Gutteridge, Biochem. J. 243 (1987) 709. [35] Y. Zou, Y. Lu, D. Wei, J. Agric. Food Chem. 52 (2004) 5032–5039. [36] Y.G. Bi, Y.Q. Tan, Adv. Mater. Res. 550 (2012) 987–992. [37] X.J. Li, Y.J. Fu, M. Luo, W. Wang, L. Zhang, C.J. Zhao, Y.G. Zu, J. Pharmaceut. Biomed. 61 (2012) 199–206. [38] V. Ferraro, I.B. Cruz, R.F. Jorge, M.E. Pintado, P.M. Castro, J. Chem. Eng. Data 58 (2013) 707–717. [39] Z.Y. Zhao, L.L. Dong, Y.L. Wu, F. Lin, Food Bioprod. Process. 89 (2011) 266–272. [40] Y. Zhang, J.J. Jiao, C.M. Liu, X.Q. Wu, Food Chem. 107 (2008) 1326–1336. [41] J. Song, D. Li, C. Liu, Y. Zhang, Innov. Food Sci. Emerg. Technol. 12 (2011) 282–287.
