Efficient separation and purification of anthocyanins from saskatoon berry by using low transition temperature mixtures

Abstract

Different kinds of low transition temperature mixtures (LTTMs) were synthesized with DL-lactic acid and five different quaternary ammonium salts respectively, and their viscosity and solubility parameters change with different molar ratios of quaternary ammonium salt to DL-lactic acid were investigated and compared with anthocyanins. Then the LTTMs synthesized were used to extract anthocyanins from saskatoon berry (Amelanchier alnifolia Nutt.). It turned out that the choline chloride/DL-lactic acid (ChCl/LA) showed the best solubility of anthocyanins compared with the other LTTMs, which was in line with solubility parameter theory. Under the optimized process condition with 1 : 15 of molar ratio, 1 : 20 of solid to liquid ratio, 50 °C of extraction temperature and 45 min of extraction time, the extraction amount of anthocyanins reached 525.27 ± 0.82 mg per 100 g and the extracted rate of anthocyanins was 87.4%. Then the macroporous resin AB-8 adsorbent was used to separate and purify the anthocyanin compounds from saskatoon berry, with a recovery yield of 65.7%. Further, the confirmation of the purified anthocyanins was carried out using nuclear magnetic resonance (NMR), Fourier transform infrared spectroscopy (FT-IR) and high performance liquid chromatography-mass spectrometry (HPLC-MS). Finally, the purified anthocyanins were proved to be good antioxidants through the comparison with vitamin C and butylated hydroxytoluene.

---

1 Introduction

The interest in high bioactive natural products has dramatically increased in recent decades. For example, artemisinin has already been proved to be great antimalarial drug^1,2 and goniothalamin shows anti-inflammatory activity and antinociceptive effects reported by some researchers.^3,4 Nowadays, there has been a sharp increase of research activities in saskatoon berry for their high content of anthocyanins (higher than blue berry) and potential commercial value.⁵ Anthocyanins are glycosides of polyhydroxy/methoxy derivatives of 2-phenylbenzopyrylium or flavylium salts which confer the characteristic colors to flowers, fruits, vegetables and some plant tissues.^6–8 In 1986, Mazza et al. found that saskatoon berry contains 13 kinds of phenolic compounds, of which the two most abundant anthocyanins were cyanidin-3-O-galactoside (a) and cyanidin-3-O-glucoside (b) (Fig. 1), and the latter accounted for 61 wt% of the total anthocyanin content.⁹ Later, it is reported that more than 4 cyanidin-based anthocyanins existed in mature saskatoon fruits.¹⁰ Besides, it has been proved that anthocyanins are outstanding antioxidant compounds and shows potential application value for various diseases, due to their excellent free radical scavenging properties.¹¹

Fig. 1 The two most abundant anthocyanins cyanidin-3-O-galactoside (a) and cyanidin-3-O-glucoside (b) in saskatoon berry.

In general, methanol and ethanol are the most commonly used solvents in anthocyanins extraction. Lataoui¹² reported that the extracted content of anthocyanins in Vitex agnus-castus Linn. leaves reached 360 mg per 100 g by high pressure and temperature extraction. Chen¹³ also studied the extraction of anthocyanins in sugar beet molasses and found that the content of anthocyanins was 31.81 mg per 100 g. However, the traditional extraction solvents have inevitable shortcomings, such as high toxicity, easy volatility, chemical instability, recycle difficulty, etc.¹⁴

Deep eutectic solvent (DES) has come upon scene as a new generation of solvents in 2001.¹⁵ DES is a new class of solvent obtained by the combination of a quaternary ammonium salt with a metal salt or hydrogen bond donor (HBD), which results in a eutectic mixture with a melting point much lower than the individual components.^16,17 However, DES does not cover the complete class of solvents, because part of them do not have eutectic melting point but glass transitions temperature instead, which we coined them “low transition temperature mixtures” (LTTM). In other words, DES is a special kind of LTTM.¹⁸ Many kinds of LTTMs (synthesized also by hydrogen bond acceptor (HBA) and HBD) in recent years exhibited lots of excellent properties superior to traditional organic solvents. Their performances are very impressive: negligible volatility, low melting point, and high solubilization power strength for a wide range of compounds.¹⁹ For natural products, LTTMs (or DESs) offer the prospect of being novel extraction solvents with a great variety of compounds.^20,21 For instance, phenolic compounds and bioactive flavonoids like Flos carthami²² and rutin²³ have been extracted with considerable results in some DESs. There are also some researches on anthocyanins extraction using DESs, but the purity and extraction efficiency of anthocyanins extracted cannot be guaranteed at the same time.^24,25 However, there hasn't been any research on the anthocyanins extraction using LTTM, the much broader ionic liquid analogues than DESs.

In the present work, a variety of LTTMs were synthesized DL-lactic acid (lower viscosity and cheaper than most of HBDs) with different quaternary ammonium salts, and investigated about their abilities of extracting anthocyanins from saskatoon berry. Then several key variables were systematically evaluated to improve the extraction efficiency with the LTTM which afforded the best results. Besides, the antioxidant activities of the purified anthocyanins were examined. This work has certain significance on the expansion of the application of LTTMs in the extraction of natural products to some degree.

2 Materials and methods

2.1 Materials

Choline chloride (ChCl) (≥99% mass fraction purity) was purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The other quaternary ammonium salts (Table 1) were obtained from Macklin Chemical Industry Co. Ltd. (Shanghai, China). DL-Lactic acid (LA) (≥99% mass fraction purity) and ethanol (≥99% mass fraction purity) were purchased from Tianjin Kermel Chemical Reagent Co., Ltd. (Tianjin, China). The saskatoon berries (Amelanchier alnifolia Nutt.) used in this study were obtained from the orchard (Xinjiang, China) in mid-July of 2015. The standards of cyanidin-3-O-glucoside and cyanidin-3-O-galactoside (≥98% mass fraction purity) were purchased from Macklin Chemical Industry Co., Ltd. (Shanghai, China). n-Octane, n-dodecane, n-tetradecane, and n-hexadecane were also purchased from Macklin Chemical Industry Co., Ltd. (Shanghai, China). All other chemicals were of analytical grade.

Table 1 Several quaternary ammonium salts and their abbreviation

Quaternary ammonium salt	Abbreviation	Mass fraction
Choline chloride	ChCl	≥99.0%
Tetramethylammonium chloride	TMAC	≥99.0%
Butyltrimethylammonium chloride	AMBTAC	≥99.0%
Tributylmethylammonium chloride	AETAC	≥99.0%
Benzyltriethylammonium chloride	TEBAC	≥99.0%

---

2.2 Preparation of LTTMs

LTTMs were synthesized by mixing HBA and HBD and heated at 70 °C at atmospheric pressure under sharply stirring until a stable homogeneous liquid was formed.²⁶ The mole ratio of HBA to HBD was in the range of 1 : 5–1 : 20. Then all LTTMs were required to cool down to room temperature and stored in sealed laboratory vials.

2.3 Solubility parameters (δ) of LTTMs and anthocyanins determination

Solubility parameters of LTTMs and anthocyanins determination were tested using the inverse gas chromatography (IGC). A GC-9A (Shimadzu Co., Ltd) gas chromatograph, equipped with a flame ionization detector, was used. The nitrogen carrier gas flow rate was 20 mL min⁻¹. The injector and detector temperatures were kept at 523.15 K during all experiments. Add each LTTM or anthocyanin standard into acetone, the mass ratio was 1 : 10, then add red supporter into it, and stir until blended. After the solvent evaporated, the coated supporter was packed into stainless steel columns with 2 mm inner diameter and 600 mm length and then conditioned at 80 °C under nitrogen for 6 h prior to use. Column temperature was 45 °C. The data was processed referring to literature.²⁷

2.4 LTTMs' viscosities measurement

LTTMs' viscosities were measured by capillary tube method.²⁸

2.5 Extraction of the crude anthocyanins from saskatoon berries

The saskatoon berries were pretreated via a process of washing, freeze-drying and grind to obtain powdered berries samples and the formed powders were stored in a desiccator for further use. For a typical extraction process, powdered berries samples (0.5 g) were mixed with different solvents (LTTMs or ethanol) in flasks. Then the flasks were placed in water bathes at a certain temperature with continuous stirring for definite time, and the mixtures were centrifuged at 12 000 rpm for 5 min to remove the solids. The supernatant was further applied to the determination of anthocyanins content by pH differential method.

2.6 Determination of anthocyanins content

The total anthocyanins content in extracts were directly determined using the pH differential method.^29,30 The absorbance of anthocyanins was measured at 514 nm and 700 nm, employing eqn (1):where A = [(A_{514 nm} − A_{700 nm})_pH1.0 − (A_{514 nm} − A_{700 nm})_pH4.5]. MW is the molecular weight of anthocyanins (449.2 g mol⁻¹ is the molecular weight of cyanidin-3-O-galactoside which accounted for the highest content in the saskatoon berries); DF is the dilution factor; ε is the molar extinction coefficient (26 900 L cm⁻¹ mol⁻¹); L is the path length (1 cm); V is the volume; W_t is the quality of berries.

The purified anthocyanins content were measured by HPLC. Samples were analyzed by RP-HPLC on a 4.6 mm × 250 mm (5 μm) Zorbax SB-C18 column (Agilent Technologies Industries Co. Ltd.) using a Waters HPLC system equipped with two Waters 1525 pumps and a Waters 2489 UV detector at 514 nm. The samples were eluted at a flow rate of 1 mL min⁻¹ using a linear gradient of 5% phosphoric acid (solvent A) and 100% methanol (solvent B) as follows: 86% A at 0 min; 77% A by 25 min; isocratic at 77% A from 25 to 45 min; 53% A by 55 min; isocratic at 53% A from 55 to 58 min; 86% A by 59 min; and isocratic at 86% A from 59 to 64 min. Quantification was made according to the linear calibration curves of standard compounds. Cyanidin-3-O-glucoside and cyanidin-3-O-galactoside ranging from 0.01 to 0.5 mg mL⁻¹ were used for anthocyanin quantification. The results were expressed as milligrams of commercial standards per 100 g of fresh weight. The definite contents of cyanidin-3-O-galactoside and cyanidin-3-O-glucoside are calculated according to the linear calibration curves of standard compounds. The cyanidin-3-O-galactoside (CG1) was calculated from a calibration curve, [CG1 (mg mL⁻¹)] = 620.17x (peak area) + 2.9828 (R² = 0.9969). And the cyanidin-3-O-glucoside (CG2) was calculated from a calibration curve, [CG2 (mg mL⁻¹)] = 629.76x + 1.6151 (R² = 0.9997).

2.7 Effect of several factors on extraction yield via LTTMs

The factors influencing the extraction efficiency of the LTTMs were examined systematically. The key variables included mole ratio of HBA to HBD (1 : 3–1 : 19), solid to liquid ratio (1 : 10–1 : 50 mL g⁻¹), extraction time (20–180 min) and extraction temperature (40–70 °C). The next experimental procedures were the same as the typical extraction process as above.

2.8 Separation and purification of saskatoon berry anthocyanins using macroporous resin

The separation and purification of anthocyanins were performed with macroporous resin (AB-8), which was activated according to previous report.^31–34 Activated AB-8 (60.0 g dry weight) was added to 300 mL of anthocyanins extract in a 1000 mL flask, and then the mixture were agitated at 25 °C for 12 h to reach adsorption equilibrium.³⁵ Then the resin was washed with deionized water for 2–3 times and desorbed with 500 mL 80% ethanol containing 0.1% HCl in a 1000 mL flask while agitating on a vibratory shaker at 25 °C for 45 min to reach desorption equilibrium. The content of anthocyanins was then measured using the pH differential method. The adsorption and desorption ratio were quantified as follows:³⁶where C_o and C_e are the initial and equilibrium concentrations of anthocyanins in the solution, respectively (mg mL⁻¹); Q_re (%) is the adsorption ratio; C_d is the concentration of solutes in the desorption equilibrium solution (mg mL⁻¹); V_d is the volume of the desorption solution (mL); V_i is the volume of the initial sample solution (mL); D (%) is the desorption ratio.

The crude anthocyanins extract was dissolved in distilled water and passed through a pretreated resins column (2.8 cm × 40 cm). The column was first eluted with distilled water to remove the majority of sugars, organic acids, proteins and ions, and then eluted of anthocyanins by 80% (v/v) ethanol solution containing 0.1% HCl. The collected fraction was concentrated and lyophilized to afford the purified anthocyanins powder. The recovery of anthocyanins was determined by calculating the amount of anthocyanins before and after purification on column packed with the selected resin to evaluate the efficiency of the method.

2.9 FT-IR analysis

The FT-IR analysis was carried out using a Tensor 27 spectrometer (Bruker, Germany). The FT-IR spectra was the averages of 32 scans and recorded at a resolution of 2 cm⁻¹ in the range of 400–4000 cm⁻¹ to determine the functional groups of purified anthocyanins.

2.10 ¹H-NMR analysis

All NMR spectra were recorded on a Bruker AVANCE III HD 600 MHz spectrometer (Bruker) at 25 °C. For ¹H-NMR analysis, 2–5 mg of purified was dissolved in 2 mL of dimethyl sulfoxide-d6 (DMSO-d6). The relaxation delay was 1 s. The number of scans was 128 with an acquisition time of 3.98 s. The peak (2.49 ppm) of DMSO-d6 was used as an internal reference.

2.11 HPLC-MS analysis

The condition of HPLC refers to 2.6 above. The MS parameters were as follows: positive mode; ESI source voltage, ±4500 V; capillary voltage, 36 V; sheath gas flow rate, 40 arb; aux gas flow rate, 5 arb; sweep gas flow rate, 0 arb; capillary temperature, 350 °C; and scan range, 100–900 m/z. Prior to HPLC analysis, each sample was filtrated through a syringe filter (0.22 μm).³¹

2.12 Determination of the antioxidant property

The antioxidant ability of the purified anthocyanins was determined according to the DPPH radical scavenging activity and hydroxyl radical (˙OH) scavenging activity.

DPPH scavenging activity assay was examined as described previously with some modifications.³⁷ Briefly, 2.0 mL of purified anthocyanins solution was added to 2.0 mL of 0.20 mmol L⁻¹ DPPH solution in ethanol. The mixtures were shaken and allowed to stand at room temperature in the dark for 30 min before measuring the absorbance at 517 nm. At the same time, a mixture of 2.0 mL ethanol and 2.0 mL DPPH solution was used as the control. And a mixture of 2.0 mL ethanol and 2.0 mL of purified anthocyanins solution was used as the blank. DPPH scavenging activity assay was repeated on vitamin C (Vc) and butylated hydroxytoluene (BHT) then. All the data reported were the averages of test performed in triplicate. The DPPH scavenging activity was calculated using the formula:

where A_i is the absorbance value of the sample, A_j is that of blank and A_c is that of control.

The measurements of ˙OH scavenging activity were performed using a salicylic acid/ferrous iron(II) oxidation assay with some modifications.³⁸ Briefly, 16.0 mL of 6 mmol L⁻¹ salicylic acid solution in ethanol, 2.0 mL of 2 mmol L⁻¹ ferrous sulfate solution and 2.0 mL of 6 mmol L⁻¹ H₂O₂, were mixed with 5.0 mL of purified anthocyanins solution which had a concentration range from 0.02 mg mL⁻¹ to 0.80 mg mL⁻¹. The mixtures were incubated at 37 °C for 15 min before the absorbance was measured at 510 nm. Meanwhile, 5.0 mL distilled water instead of 5.0 mL solution of purified anthocyanins was used as the blank and 2.0 mL distilled water instead of 2.0 mL H₂O₂ (6 mmol L⁻¹) was used as the control. The ˙OH scavenging activity was calculated by the equation:

where A_x, A_o, and A_xo represent the absorbance of the mixture solution, the blank and the control, respectively.

3 Results and discussion

3.1 Solubility parameters of LTTMs with molar ratio

The solubility parameter (δ) can provide guidelines for estimating the miscibility of one solute in various solvents.^39,40

The solute can easily dissolve into solvent theoretically when they have similar δ value. Some studies suggested that solubility parameter have little changes with temperature.⁴¹ So in this work, to make a better comparison, solubility parameters of all LTTMs were investigated about different molar ratios at the same temperature. The solubility parameters of cyanidin-3-O-glucoside and cyanidin-3-O-galactoside were 20.73 and 20.69 MPa^1/2 respectively through the test. As shown in Fig. 2, solubility parameters of all DESs increased with the molar ratio decrease, when the molar ratio was 1 : 15, ChCl/LA and TEBAC/LA's solubility parameters (20.97 and 20.13 MPa^1/2, respectively) were the two closest to 20.73 and 20.69 MPa^1/2. So we predicted that ChCl/LA and TEBAC/LA can better dissolve anthocyanins.

Fig. 2 The changes of each LTTM's solubility parameter with molar ratio.

3.2 Effect of molar ratio on the viscosities of LTTMs

Viscosity is another key factor that influences dissolution rate. Changes of viscosities of LTTMs with molar ratio were shown in Fig. 3. As Fig. 3 depicted, all LTTMs' viscosities fell down with more LA existing, which was beneficial to mass transfer between two phases of liquid and solid. In addition, LTTM with higher molecular weight HBA had a higher viscosity than those lower in molecular volume.

Fig. 3 Changes of viscosities of LTTMs with molar ratio.

3.3 The changes of anthocyanins extracted amount with molar ratio in different LTTMs

As shown in Fig. 4, with the molar ratio decreasing, the anthocyanins extracted amount in five different LTTMs first rose and then dropped. Among the five LTTMs, the amount of anthocyanins extracted by ChCl/LA was the highest in any certain molar ratio. When the molar ratio was 1 : 15, the extracted amount reached to 501.26 ± 1.82 mg/100 g. In addition, TEBAC/LA also showed good ability to dissolve ability to dissolve source rock. As known, the solubility parameters of above two LTTM were closer to those of anthocyanins than others, so the anthocyanins extracted amount in them were higher. However, the extracted amount in TEBAC/LA was not as high as in ChCl/LA, possibly because the former had higher viscosity (80 mPa s vs. 73.2 mPa s at molar ratio of 1 : 15 and 45 °C), resulting in higher mass transfer resistance. Therefore, the LTTM synthesized by ChCl and LA was ideal and chosen as the solvent to extract anthocyanins in the following experiment.

Fig. 4 The changes of anthocyanins extracted amount with molar ratio in different LTTMs (solid to liquid ratio 1 g/10 mL; 30 min; 45 °C).

3.4 The effect of several crucial variables on anthocyanins extraction

In order to further improve the extraction efficiency with ChCl/LA, the effects of several key variables were investigated. The first aspect was the molar ratio of ChCl and LA in the extraction process. Nine different molar ratios were tested, from 1 : 3 to 1 : 19 (Fig. 5a). When the molar ratio was decreased from 1 : 3 to 1 : 15, the anthocyanins extraction yield increased up to 501.32 mg per 100 g, which could be caused by the gradually decreased viscosity and approaching solubility parameter. But when molar ratio decreased from 1 : 15 to 1 : 19, anthocyanins extraction yield reduced because the solubility parameter of ChCl/LA was far away from anthocyanin and the decrease of viscosities becomes slow. Therefore, the best molar ratio of ChCl/LA was 1 : 15.

Fig. 5 (a) The effect of molar ratio on anthocyanins extraction (solid to liquid ratio 1 : 15; 45 °C; 30 min), (b) the effect of solvent to solid ratio on anthocyanins extraction (ChCl/LA 1 : 15; 45 °C; 30 min), (c) the effect of time on anthocyanins extraction (ChCl/LA 1 : 15; solid to liquid ratio 1 : 20; 45 °C), (d) the effect of temperature on anthocyanins extraction (ChCl/LA 1 : 15; solid to liquid ratio 1 : 20; 45 min).

The solid to liquid ratio was another important factor in anthocyanins extraction process. As depicted in Fig. 5b, the amount of anthocyanins extracted rose obviously up to 504.48 mg per 100 g, when the solid to liquid ratio decreased from 1 : 10 to 1 : 20. Further decreasing the solid to liquid ratio, however, resulted in little increase in the amount of anthocyanins extracted. It could be speculated that the extracted amount of anthocyanins was close to saturation at a ratio of 1 : 20. Thus, the optimal solid to liquid ratio was 1 : 20.

The effect of extraction time on the amount of anthocyanins extracted was presented in Fig. 5c. It could be observed that the amount of anthocyanins extracted increased expectedly with an increase in process time. But when the process time was beyond 45 min, the amount of anthocyanins extracted had no significant increase. It indicated that the LTTM was basically saturated with anthocyanins after 45 minutes. In general, 45 min was the most ideal reaction time.

The temperature was also an extremely significant factor affecting the amount of anthocyanins extracted (Fig. 5d). As evident, anthocyanins extraction yield was notably affected by the reaction temperature from 35 °C to 50 °C, reaching 525.27 mg per 100 g. Because in this heating period, saturated solubility of anthocyanins in LTTM was increased according to Le Chatelier principle,⁴² and the viscosity of LTTM was decreased. When the temperature was above 50 °C, there was no obvious increase of the amount of anthocyanins extracted. Interestingly, it was found that the amount of anthocyanins extracted in the temperature range (60–70 °C) showed a little decrease. A possible reason for the phenomenon was that the anthocyanins denatured under high temperature (above 60 °C). In summary, 50 °C was appropriate as reaction temperature.

In conclusion, the result showed, 50 °C of temperature, 1 : 20 of solid to liquid ratio, 45 min of time and 1 : 15 of ChCl/LA molar ratio were the optimum processing condition of anthocyanins extraction, when anthocyanins extraction amount reached 527.27 mg per 100 g, much higher than extraction amount using methanol (397.80 mg per 100 g) or ethanol (277.05 mg per 100 g) which has been used as extracted solvents and Ozga's methods.¹⁰ In consideration of economizing, the recycle of solvent is necessary, and the diluted LTTM was dehydrated by rotary evaporation to obtain the concentrated LTTM. The obtained LTTM was applied to a secondary extraction and the extraction efficiency still reached 82.6%. It further demonstrates LTTM is a good extractant.

3.5 Separation and purification of anthocyanins

The static adsorption and desorption capacities of several types of resins been investigated by some researchers.^43–45 A comparison of several resins for separating anthocyanins was conducted. It could be observed that the adsorption capacity of AB-8 was pretty well, which could be attributed not only to its similar polarity with the anthocyanins, but also to its appropriate pore size and high surface area which were the significant variables reflecting the adsorption. One of the most decisive factors that affected the desorption capacity of resin was polarity which means the higher the polarity of resin, the weaker the desorption capacity.⁴⁶ And AB-8 was consistently more efficient at desorption than some other resins. The adsorption ratio (Q_re) of AB-8 could reach 87.0% higher than H-1020 (80.8%) and S-8 (82.1%), and the desorption ratio (D) (80.5%) of AB-8 was higher than H-1020 (62.5%) and S-8 (67.3%), too.

After separated with the selected resin column chromatography, the recovery yield of anthocyanins purified was only 65.7% which was obtained by HPLC (Fig. 6) because of the operational loss. And the AB-8 resin showed good reusability and was proved to be the perfect sorbent for anthocyanins purification from saskatoon berry.

Fig. 6 HPLC profile of anthocyanins compounds.

3.6 Characterization of the purified anthocyanins

The group of anthocyanins was verified by FT-IR spectroscopy of the freeze-dried precipitate (Fig. 7). The intensive band at about 1720 cm⁻¹ (peak A) in the FT-IR spectra of purified anthocyanins was most probably related with (C=O) vibration in the aromatic ring of benzopyran which was the anthocyanins skeleton molecule.⁴⁷ The band located at about 1617 cm⁻¹ (peak B) corresponded to the stretching vibrations (C=C) in the phenyl rings.⁴⁸ At the same time, the bands at about 1527 cm⁻¹ and 1469 cm⁻¹ (peak C) originated in the spectrum of the anthocyanins complex. These bands found in the spectra were attributed to a deformation vibration of the two phenyl rings in the anthocyanins. The intensive band in FT-IR spectrum which was observed at about 1284 cm⁻¹ (peak D) was related with (C–O) vibration in the aromatic ring. The band at about 680 cm⁻¹ (peak E) was the stretching vibrations of the (C–O–C) glycosides linkage.⁴⁷ In the spectrum of freeze-dried anthocyanins powder these peaks could be the primary peaks of cyanidin-3-glucoside and cyanidin-3-galactoside chlorides. They are the preliminary evidences for identifying composition in purified anthocyanins.

Fig. 7 FT-IR spectra of purified anthocyanins.

The chemical structures of anthocyanins were studied by ¹H NMR (Fig. 8). Signal assignment for ¹H NMR spectra was based on a previous publication.⁴⁹ In the extracted anthocyanins, the signals of 0–4.80 ppm were attributed to protons in hexose units, except signal of 2.21 ppm, assigned as the DMSO-d₆. The intense peaks between 5.75 and 6.25 ppm might be assigned as hydroxyl protons in side chains of benzene rings. And the signals at 6.25–7.25 ppm were due to aromatic rings protons. The signals of 8.04 ppm were attributed to protons at H-1 and the intense peak 8.94 ppm might be assigned as H-4 in the structure above. It can be seen that the ¹H NMR and FT-IR analysis were consistent with the chemical structure of anthocyanins.

Fig. 8 ¹H-NMR profile of anthocyanins compounds.

The major anthocyanins peaks were identified by comparing the retention times and the m/z of each anthocyanin molecule with the standards and compared with the previous value.⁴³ Full-scan UV/Vis analysis of the saskatoon berry anthocyanins was conducted and the wavelength of 514 nm had been monitored as the maximum absorbance wavelength (λ_max) which had been reported to be close to the result.^10,50 The HPLC profile of the saskatoon berry anthocyanins showed three major peaks (Fig. 6) when monitored at 514 nm. Two anthocyanin standards cyanidin-3-O-galactoside, cyanidin-3-O-glucoside and the purified anthocyanin powders, respectively, were chromatographed on HPLC and exhibited identical spectrum.

As depicted in Fig. 6, peak A detected at the retention time of 27.004 min was the major anthocyanin cyanidin-3-O-galactoside according to the standard which was observed at 26.875 min and it had a m/z of 449 (Fig. 9A). And it showed that the retention time of peak B (Fig. 6) at 31.862 min was similar to the standard cyanidin-3-O-glucoside which was observed at 31.520 min. Meanwhile, it produced a molecular ion at m/z of 449 shown in the Fig. 9B. A minor proportion of cyanidin-3-O-arabinoside (peak C) was also observed, whose assignation was made by comparison of their t_R with the previous value¹⁰ and on the base of molecular ion at m/z of 419 shown in the Fig. 9C. Hence the purified anthocyanins were in conformity with the anthocyanin standards. This result was in agreement with other published studies¹⁰ which found that cyanidin-3-O-galactoside and cyanidin-3-O-glucoside are the two main components in the saskatoon berry with content account for 77.9 wt% and 17.1 wt%, respectively.

Fig. 9 The mass spectra of anthocyanins compounds.

3.7 Free radical scavenging activities

In order to evaluate antioxidant activities of the purified anthocyanins, the DPPH radical and hydroxyl radical scavenging activities were analyzed.

As shown in Fig. 10, in the concentration range of 0.02–0.6 mg mL⁻¹, the purified anthocyanins, Vc and BHT could all scavenge free radicals to different degrees. And their scavenging activities increased with the increase of concentration. It was noteworthy that the purified anthocyanins exhibited better DPPH scavenging activity than Vc and BHT, which was consistent with recent empirical evidence.⁴⁴ When the concentration of purified anthocyanins by LTTM increased to 0.1 mg mL⁻¹, scavenging ratio of DPPH radical nearly reached 90%, and the change range of scavenging activity rate began diminishing with higher concentration of anthocyanins. The results implied that the ability of anthocyanins to remove DPPH was stronger than those of Vc and BHT. The study also found that the anthocyanins purified by LTTM exhibited better DPPH scavenging activity than by ethanol, suggesting the former's purity was higher.

Fig. 10 The DPPH scavenging rates of purified anthocyanins.

The measurements of ˙OH scavenging activity were performed using a salicylic acid/ferrous iron(II) oxidation assay. Hydroxyl free radicals are produced in this method using Fenton system, then captured by salicylic acid/ethanol and colored substance is formed. At the moment, maximum absorption peak of the colored substance is at 510 nm. But with the increase of the added free radical scavengers, the colored substance decreases, also, the absorbance decreases at 510 nm. Therefore, by this way, we can calculate the clearance rate of hydroxyl free radicals and evaluate the antioxidant capacity of the samples.

As shown in the Fig. 11, the purified anthocyanins by LTTM and ethanol, Vc and BHT all presented certain activities about scavenging hydroxyl free radical and the scavenging activity rate also increased with the increase of concentration in the range of 0.02–1.6 mg mL⁻¹. It should also be noted that the activities of the purified anthocyanins by LTTM were more effective than that of anthocyanins by ethanol, Vc and BHT under the same concentration. When the concentration was 0.8 mg mL⁻¹, the scavenging activity rate of the purified anthocyanins by LTTM was 92%, higher than others, and the change range of scavenging activity rate began diminishing with higher concentration of anthocyanins because hydroxyl radical scavenging activity reached a saturation point. In general, the scavenging effect of anthocyanins on ˙OH radical was better in comparison to Vc and BHT. Meanwhile, purified anthocyanins by LTTM could play better results than that of anthocyanins by ethanol, indicating the former are purer, too. Therefore, the purified anthocyanins via LTTM synthesized by ChCl/LA had high antioxidant property.

Fig. 11 The ˙OH scavenging rates of purified anthocyanins.

4 Conclusions

In conclusion, this work provides a new insight into the extraction of anthocyanins from saskatoon berry (Amelanchier alnifolia Nutt.) using a designed LTTM (ChCl/LA) based on solubility parameter theory. ChCl/LA has shown higher extraction capacities than other LTTMs or organic solvent. Under the optimum condition, the highest extraction rate of 525.27 mg per 100 g was observed. The ratios of adsorption and desorption with AB-8 were 86.5% and 80.8%, respectively. Additionally, the purified anthocyanins exhibited very excellent oxidation resistance property. Therefore, the LTTM shows great potential in natural product extraction, and will benefit the extraction of anthocyanins from other plants.

Acknowledgements

We wish to thank the National Natural Science Foundation of China (21336002; 21376096), the Key Program of Guangdong Natural Science Foundation (S2013020013049), and the Fundamental Research Funds for the Chinese Universities (2015PT002; 2015ZP009) for partially funding this work.

References

1. H. M. Ismail, V. Barton, M. Phanchana, S. Charoensutthivarakul, M. H. L. Wong, J. Hemingway, G. A. Biagini, P. M. O'Neill and S. A. Ward, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 2080–2085.
2. D. L. Klayman, Science, 1985, 228, 1049–1055.
3. D. B. Vendramini-Costa, H. M. Spindola, G. C. D. Mello, E. Antunes, R. A. Pilli and J. E. D. Carvalho, Life Sci., 2015, 139, 83–90.
4. V. C. Barbosa, A. Antonio, P. R. Karin Juliane, T. Elena, Á. R. Javier, G. M. Sofia, P. R. Aloise, C. João Ernesto and M. Virginia, Toxicol. Appl. Pharmacol., 2016, 300, 1–12.
5. F. S. Hosseinian and T. Beta, J. Agric. Food Chem., 2007, 55, 10832–10838.
6. A. Castañeda-Ovando, M. D. L. Pacheco-Hernández, M. E. Páez-Hernández, J. A. Rodríguez and C. A. Galán-Vidal, Food Chem., 2009, 113, 859–871.
7. S. D. Pascual-Teresa and M. T. Sanchez-Ballesta, Phytochem. Rev., 2008, 7, 281–299.
8. T. Boranbayeva, F. Karadeniz and E. Yılmaz, Food Bioprocess Technol., 2014, 7, 1894–1902.
9. G. Mazza, J. Food Sci., 1986, 51, 1260–1265.
10. J. A. Ozga, A. Saeed, W. Wismer and D. M. Reinecke, J. Agric. Food Chem., 2007, 55, 10414–10424.
11. X. Sui, Y. Y. Pei and W. Zhou, Food Bioprocess Technol., 2015, 8, 983–994.
12. M. Lataoui, M. Seffen, B. Aliakbarian, A. A. Casazza, A. Converti and P. Perego, Nat. Prod. Res., 2014, 28, 67–69.
13. M. Chen, Z. Yi and S. Yu, Food Chem., 2015, 172, 543–550.
14. M. H. Duan, M. Luo, C. J. Zhao, W. Wang, Y. G. Zu, D. Y. Zhang, X. H. Yao and Y. J. Fu, Sep. Purif. Technol., 2013, 107, 26–36.
15. A. Paiva, R. Craveiro, I. Aroso, M. Martins, L. R. Rui and A. R. C. Duarte, ACS Sustainable Chem. Eng., 2014, 2, 1063–1071.
16. M. Francisco, A. v. d. Bruinhorst and M. C. Kroon, Green Chem., 2012, 14, 2153–2157.
17. E. L. Smith, A. P. Abbott and K. S. Ryder, Chem. Rev., 2014, 114, 11060–11082.
18. M. Francisco, A. V. Bruinhorst and M. C. Kroon, Angew. Chem., Int. Ed., 2013, 52, 3074–3085.
19. M. Francisco and P. M. Kroon, Angew. Chem., Int. Ed., 2013, 52, 3074–3085.
20. X. L. Qi, X. Peng, Y. Y. Huang, L. Li, Z. F. Wei, Y. G. Zu and Y. J. Fu, Ind. Crops Prod., 2015, 70, 142–148.
21. X. H. Yao, D. Y. Zhang, M. H. Duan, Q. Cui, W. J. Xu, M. Luo, C. Y. Li, Y. G. Zu and Y. J. Fu, Sep. Purif. Technol., 2015, 149, 116–123.
22. Y. Dai, G.-J. Witkamp, R. Verpoorte and Y. H. Choi, Anal. Chem., 2013, 85, 6272–6278.
23. B. Zhao, P. Xu, F. Yang, H. Wu, M. Zong and W. Lou, ACS Sustainable Chem. Eng., 2015, 3, 2746–2755.
24. M. C. Bubalo, N. Ćurko, M. Tomašević, K. K. Ganić and I. R. Redovniković, Food Chem., 2016, 200, 159–166.
25. A. L. Jin, J. A. Ozga, J. A. Kennedy, J. L. Koernersmith, G. Botar and D. M. Reinecke, J. Agric. Food Chem., 2015, 63, 1601–1614.
26. C. Florindo, F. S. Oliveira, L. P. N. Rebelo, A. M. Fernandes and I. M. Marrucho, ACS Sustainable Chem. Eng., 2014, 2, 2416–2425.
27. W. Qiang, Y. Chen, L. Deng, J. Tang and Z. Zhang, J. Mol. Liq., 2013, 180, 135–138.
28. W. J. Kim, M. Lee, M. Jung and D. Lee, US Pat., US20160223448, 2016, vol. 926, pp. 12–19.
29. R. E. Wrolstad, R. W. Durst and J. Lee, Trends Food Sci. Technol., 2005, 16, 423–428.
30. J. Lee, C. Rennaker and R. E. Wrolstad, Food Chem., 2008, 110, 782–786.
31. C. Conidi, A. D. Rodriguez-Lopez, E. M. Garcia-Castello and A. Cassano, Sep. Purif. Technol., 2015, 144, 153–161.
32. Y. Chen, W. Zhang and T. Zhao, Food Chem., 2016, 194, 712–722.
33. Y. Wen, Y. T. Liu and J. M. Zhang, Adv. Mater. Res., 2013, 641, 988–992.
34. L. I. Chao, N. X. Wang and Y. Zheng, Food Sci., 2011, 32, 31–35.
35. L. Yao, N. Zhang, C. Wang and C. Wang, J. Agric. Food Chem., 2015, 63, 3543–3550.
36. C. Jampani, A. Naik and K. S. Raghavarao, Sep. Purif. Technol., 2014, 125, 170–178.
37. M. Ohata, S. Uchida, L. Zhou and K. Arihara, Food Chem., 2016, 194, 1034–1039.
38. N. Smimoff and Q. J. Cumbes, Phytochemistry, 1989, 28, 1057–1060.
39. C. M. Hansen, Hansen solubility parameters: a user's handbook, CRC Press, 2007.
40. J. C. Huang, J. Appl. Polym. Sci., 2004, 94, 1547–1555.
41. P. Weerachanchai, Z. Chen, S. S. Leong, M. W. Chang and J. M. Lee, Chem. Eng. J., 2012, 213, 356–362.
42. B. Yan, G. Shao, K. Li, Y. Ma, L. Wei and D. Zhao, RSC Adv., 2016, 6, 6285–6288.
43. C. Yao, W. Zhang, T. Zhao, L. Fang, Z. Min, L. Jing, Z. Ye, W. Wei, S. J. Cobbina and X. Wu, Food Chem., 2016, 194, 712–722.
44. L. Liang, X. Wu, T. Zhao, J. Zhao, F. Li, Y. Zou, G. Mao and L. Yang, Food Res. Int., 2012, 46, 76–82.
45. X. Liu, G. Xiao, W. Chen, Y. Xu and J. Wu, J. Biomed. Biotechnol., 2004, 2004, 326–331.
46. T. J. Buran, A. K. Sandhu, Z. Li, C. R. Rock, W. W. Yang and L. Gu, J. Food Eng., 2014, 128, 167–173.
47. R. Klimaviciute, V. Navikaite, V. Jakstas and L. Ivanauskas, Carbohydr. Polym., 2015, 129, 70–78.
48. S. Garmón, Talanta, 2012, 88, 303–310.
49. T. K. Mcghie, D. R. Rowan and P. J. Edwards, J. Agric. Food Chem., 2006, 54, 8756–8761.
50. A. M. Pawlowska, W. Oleszek and A. Braca, J. Agric. Food Chem., 2008, 56, 3377–3380.

---