Isolation and Purification of Two Isoflavones from Hericium erinaceum Mycelium by High-Speed Counter-Current Chromatography

High-speed counter-current chromatography (HSCCC) was used to separate and purify two isoflavones for the first time from Hericium erinaceum (H. erinaceum) mycelium using a two-phase solvent system composed of chloroform-dichloromethane-methanol-water (4:2:3:2, v/v/v/v). These two isoflavones were identified as genistein (4′,5,7-trihydroxyisoflavone, C15H10O5) and daidzein (4′,7-dihydroxyisoflavone, C15H10O4), using infrared spectroscopy (IR), electro-spary ionisation mass (ESI-MS), 1H-nuclear magnetic resonance (NMR) and 13C-NMR spectra. About 23 mg genistein with 95.7% purity and 18 mg daidzein with 97.3% purity were isolated from 150 mg ethanolic extract of H. erinaceum mycelium. The results demonstrated that HSCCC was a feasible method to separate and purify genistein and daidzein from H. erinaceum mycelium.

Many phytochemicals, including polysaccharides, pyrones, terpenoids, phenols are present in the mycelium and fruiting bodies of H. erinaceum [9][10][11]. Flavonoids are found in many plant species and exhibit many bioactivities, including anti-oxidant, anti-inflammatory and anti-cancer effects [12,13]. It was reported that the methanol extract of H. erinaceum had strong antioxidant activities as it contained flavonoids and phenolic compounds [14]. However, specific flavonoid spices in H. erinaceum have still not been identified. In view of these beneficial properties, study on the separation and purification of flavonoids from H. erinaceum is necessary.
Many studies have been focused on the separation and purification of flavonoids from natural plants [15,16]. Most separation and purification methods are based on thin-layer chromatography and other chromatographic techniques based on a solid stationary phase, such as semi-preparative and preparative HPLC [17]. These methods were generally restricted to several disadvantages, including low-yielding, time-consuming, complex processing, high-cost, poor reproducibility, and irreversible adsorption [18,19]. High-speed counter-current chromatography (HSCCC) is an irreversible adsorption [18,19]. High-speed counter-current chromatography (HSCCC) is an advanced technique based on liquid-liquid partitioning and has played an important role in the preparation of phytochemicals in the last decade [20]. HSCCC provides advantages by eliminating the solid support, which may cause adsorption or degradation of target compounds [18,21]. Therefore, it has been widely used for separation and purification of flavonoids, alkaloids, terpenes, polyphenols, and other natural products [22][23][24][25][26]. To the best of our knowledge, there was no report about isolating and purifying of isoflavones from H. erinaceum mycelium by HSCCC.
In this study, we discussed the development of the HSCCC method for the separation and purification of pure isoflavones from H. erinaceum mycelium. The two-phase solvent system composed of chloroform-dichloromethane-methanol-water was established. In addition, chemical structures of two purified isoflavones were further identified by infrared spectroscopy (IR), electrospary ionisation mass (ESI-MS), 1 H-nuclear magnetic resonance (NMR) and 13 C-NMR spectra. As a result, two isoflavones, genistein (4′,5,7-trihydroxyisoflavone) and daidzein (4′,7-dihydroxyisoflavone), were isolated from H. erinaceum mycelium for the first time by HSCCC.

HPLC Analysis
Crude extract from H. erinaceum mycelium, named as HEM-E-E, was analyzed by HPLC. As showed in Figure 1, compounds 1 and 2 were two main components in HEM-E-E. Compounds 1 and 2 were set as target compounds in further HSCCC separation of HEM-E-E.

Selection of Two-Phase Solvent System
Choosing a suitable two-phase solvent system is a very important step in HSCCC experiment. The partition coefficient (K) and retention of the stationary phase are key factors for HSCCC separation [20]. The K value usually reflects the distribution between two mutually equilibrated solvent phases. A small K value elutes the solute close to the solvent front with lower resolution. A large K value tends to give better resolution but broader, more dilute peaks [20,27]. The retention of the stationary phase is accomplished by a combination of coiled column configuration and the planetary motion of the column holder. Therefore, the successful separation of HSCCC mainly depends on the selection of the two-phase solvent system. For the target compounds, suitable K values for HSCCC are 0.5 ≤ K ≤ 2.0 and the separation factor (α) between two components should be greater than 1.5 [20,28]. To select a suitable solvent system, several solvent systems were tested in this study. The K and α values in different solvent systems were listed in Table 1. At first, the n-hexaneethyl acetate/methanol water (HEMWat) system was tested, which can be used for analyses over a wide range of polarity [29,30]. As shown in Table 1, the K values of compounds 1 and 2 in HEMWat

Selection of Two-Phase Solvent System
Choosing a suitable two-phase solvent system is a very important step in HSCCC experiment. The partition coefficient (K) and retention of the stationary phase are key factors for HSCCC separation [20]. The K value usually reflects the distribution between two mutually equilibrated solvent phases. A small K value elutes the solute close to the solvent front with lower resolution. A large K value tends to give better resolution but broader, more dilute peaks [20,27]. The retention of the stationary phase is accomplished by a combination of coiled column configuration and the planetary motion of the column holder. Therefore, the successful separation of HSCCC mainly depends on the selection of the two-phase solvent system. For the target compounds, suitable K values for HSCCC are 0.5 ≤ K ≤ 2.0 and the separation factor (α) between two components should be greater than 1.5 [20,28]. To select a suitable solvent system, several solvent systems were tested in this study. The K and α values in different solvent systems were listed in Table 1. At first, the n-hexane-ethyl acetate/methanol water (HEMWat) system was tested, which can be used for analyses over a wide range of polarity [29,30]. As shown in Table 1, the K values of compounds 1 and 2 in HEMWat (1:1:1:1, 3:2:3:2, 4:5:4:5, v/v/v/v) were lower than 0.5, which meant compounds 1 and 2 were mainly distributed in the lower phase. This phenomenon indicated that the system's polarity was too high, compared to the target compounds. We tried n-hexane-methanol-water and ethyl acetate-methanol-water in different composition, but the K values were still lower than 0.5 (data not shown). Then the chloroform/methanol/water (ChMWat) system, which is extremely useful for separations of various natural products with moderate hydrophobicity, was tested [29,30]. When the solvent system was changed to ChMWat (4:4.5:2.5, 5:4:2, 5:5:2, v/v/v), KU/L values of compounds 1 and 2 increased. The results showed that the solvent systems were suitable for the separation of compound 2. However, the K values for compound 1 were too large, which caused a long time for elution and low resolution. Based on the above data, systems of chloroform-dichloromethane-methanol-water (4:1.5:2:2, 4:2:3:2, 4:4.5:2:5, v/v/v/v) were tested. The K values (0.71~0.87 for compound 1 and 0.56~0.91 for compound 2) were suitable for separation of compounds 1 and 2. These three solvent systems were selected for further research. respectively. b α expressed as: separation factor between two target compounds K 1 /K 2 or K 2 /K 1 . "-" stand for that the K value was too small.
In order to improve the retention of the stable phase, the settling time of the solvent system should be less than 20 s [20]. The settling time of three solvent systems was 24, 14 and 19 s, respectively (Table1). The shorter settling time means a higher retention of the stationary phase. In this study, the solvent system chloroform-dichloromethane-methanol-water (4:2:3:2, v/v/v/v) with a settling time of 14 s was selected for further HSCCC separation. Several flavonoids were reported to be separated from the seeds of Vernonia anthelmintica Willd by HSCCC using a two-step operation. The two solvent systems were chloroform-dichloromethane-methanol-water (2:2:3:2, v/v/v/v) and 1,2 dichloroethane-methanol-acetonitrile-water (4:1.1:0.25:2, v/v/v/v) [31], respectively. In this study, we used chloroform-dichloromethane-methanol-water (4:2:3:2, v/v/v/v) and separated two isoflavones from H. erinaceum mycelium using one-step operation.

HSCCC Separation
We expect that the suitable two-phase solvent system, chromatographic parameters including flow rate, rotary speed and column temperature may also affect the separation of HSCCC [32]. In this study, the upper and lower phase of chloroform-dichloromethane-methanol-water (4:2:3:2, v/v/v/v) system was used as the stationary phase and mobile phase, respectively. The effects of flow rate on separation of the target compounds by HSCCC were shown in Table 2. Both separation time and stationary phase retention decreased with the increase of the mobile phase flow rate. Low flow rate (1 mL/min) could increase to the retention of the stationary phase (67.8%), but it also extended the separation time (320 min). High flow rate (3 mL/min) could decrease the separation time (200 min). However, it decreased the retention of stationary phase (54.3%) and the purity of the target compounds. The retention levels of the stationary phase for a given flow-rate of the mobile phase will greatly contribute to the application of HSCCC. The flow rate is a key parameter that influences the chromatographic behavior after all other conditions are set [33]. Based on the separation time, the stationary phase retention and the purity of the target compounds, 2 mL/min was selected as the optimized flow rate. In this condition, the purity of the target compounds was high, the retention of the stationary phase was 65% and the separation time was 250 min. Our results are consistent with other research that shows that HSCCC is an effective method to isolate isoflavones or flavonoids from raw material [34,35]. Table 2. Separation time, stationary phase retention and purities of the two target compounds by high-speed counter-current chromatography (HSCCC) as affected by flow rate.

Flow-Rate (mL/min) Separation-Time (min) Retention (%)
Purity (%) Compound The rotary speed could also affect the separation time and the stationary phase retention. Low rotary speed reduces the volume of the stationary phase that is retained in the column, which leads to low chromatographic resolution and purity of targeted compounds. However, the high rotary speed might produce excessive sample band broadening due to the violent pulsation in the column [36]. The optimized HSCCC condition for separation of HEM-E-E was 900 rpm (rotary speed) and 20 • C (column temperature).
Under the optimized HSCCC conditions, an appropriate retention percentage of the stationary phase was 64.5%, and the purified target compounds 1 and 2 were obtained ( Figure 2). About 23 mg compound 1 and 18 mg compound 2 were yield from 150 mg HEM-E-E. As shown in Figure 3, the purity of compounds 1 and 2 were 95.7% and 97.3%, respectively. It was reported that four isoflavones including daidzein and genistein was separated by HSCCC under a linear gradient elution, using a solvent system composed of n-hexane-ethyl acetate-1-butanol-methanol-water [37]. In another research, daidzein and genistein was separated from Trifolium pratense L. by HSCCC using the solvent system of n-hexane-ethyl acetate-ethanol-water [38]. In this manuscript, we developed a new solvent system (chloroform-dichloromethane-methanol-water 4:2:3:2 (v/v/v/v)) for the separation of genistein and daidzein. It offered higher yields of genistein and daidzein compared to previews research. chromatographic behavior after all other conditions are set [33]. Based on the separation time, the stationary phase retention and the purity of the target compounds, 2 mL/min was selected as the optimized flow rate. In this condition, the purity of the target compounds was high, the retention of the stationary phase was 65% and the separation time was 250 min. Our results are consistent with other research that shows that HSCCC is an effective method to isolate isoflavones or flavonoids from raw material [34,35]. The rotary speed could also affect the separation time and the stationary phase retention. Low rotary speed reduces the volume of the stationary phase that is retained in the column, which leads to low chromatographic resolution and purity of targeted compounds. However, the high rotary speed might produce excessive sample band broadening due to the violent pulsation in the column [36]. The optimized HSCCC condition for separation of HEM-E-E was 900 rpm (rotary speed) and 20 °C (column temperature).
Under the optimized HSCCC conditions, an appropriate retention percentage of the stationary phase was 64.5%, and the purified target compounds 1 and 2 were obtained ( Figure 2). About 23 mg compound 1 and 18 mg compound 2 were yield from 150 mg HEM-E-E. As shown in Figure 3, the purity of compounds 1 and 2 were 95.7% and 97.3%, respectively. It was reported that four isoflavones including daidzein and genistein was separated by HSCCC under a linear gradient elution, using a solvent system composed of n-hexane-ethyl acetate-1-butanol-methanol-water [37]. In another research, daidzein and genistein was separated from Trifolium pratense L. by HSCCC using the solvent system of n-hexane-ethyl acetate-ethanol-water [38]. In this manuscript, we developed a new solvent system (chloroform-dichloromethane-methanol-water 4:2:3:2 (v/v/v/v)) for the separation of genistein and daidzein. It offered higher yields of genistein and daidzein compared to previews research.

Identification of Chemical Structure
The chemical structures of the compounds 1 and 2 were analyzed by IR, MS, UV and NMR chromatography.
The Their spectroscopic data for 1 H-NMR (400MHz) and 13 C-NMR (100MHz) were summarized in Table 3.

Identification of Chemical Structure
The chemical structures of the compounds 1 and 2 were analyzed by IR, MS, UV and NMR chromatography.
The Their spectroscopic data for 1 H-NMR (400MHz) and 13 C-NMR (100MHz) were summarized in Table 3. The molecular formulas of compounds 1 and 2 were established as C 15 H 10 O 5 and C 15 H 10 O 4 , respectively. Based on the IR and 1 H and 13 C-NMR spectra data, compounds 1 and 2 were identified as genistein (4 ,5,7-trihydroxyisoflavone) and daidzein (4 ,7-dihydroxyisoflavone), respectively. Our results were consistent with the NMR spectra data of genistein and daidzein in other studies [38][39][40]. Their chemical structures were shown in Figure 4. Genistein and daidzein were two isoflavones isolated from H. erinaceum mycelium.
Molecules 2018, 23, x FOR PEER REVIEW 6 of 10 The molecular formulas of compounds 1 and 2 were established as C15H10O5 and C15H10O4, respectively. Based on the IR and 1 H and 13 C-NMR spectra data, compounds 1 and 2 were identified as genistein (4′,5,7-trihydroxyisoflavone) and daidzein (4′,7-dihydroxyisoflavone), respectively. Our results were consistent with the NMR spectra data of genistein and daidzein in other studies [38][39][40]. Their chemical structures were shown in Figure 4. Genistein and daidzein were two isoflavones isolated from H. erinaceum mycelium.

HSCCC Apparatus
HSCCC was performed on a TEB 300A (Tauto Biotechnique Company, Shanghai, China) highspeed counter-current chromatography apparatus. The apparatus consisted of three preparative coils connected in series (the inner diameter of tube, 1.5 mm; total volume, 280 mL) and a 10 mL sample loop. The revolution radius was 5 cm, and the β value was varied from 0.5 at the internal terminal to 0.8 at the external terminal (β = r/R, where r is the distance from the coil to the holder shaft, and R is the distance between the holder axis and the central axis of the centrifuge ) [41]. The rotation speed was ranged from 0 to 1000 rpm. The system was equipped with a model TBP-50A constant-flow pump (Tauto Bioteh, Shanghai, China), a model UV-500 detector (XUYUKJ Instruments, Hangzhou, China) operating at 254 nm, and a model N2000 workstation (Zhejiang University, Hangzhou, China). DC-2010 constant temperature-circulating implement (Hanagzhou Dawei Instrument, Hangzhou, China) was used to adjust the experimental temperature.

Reagents and Materials
All organic solvents for HSCCC (analytical grade) were purchased from Shanghai Lingfeng Chemical Reagent Co. Ltd. (Shanghai, China). Acetonitrile used for HPLC analysis (chromatographic grade) were purchased from Tianjin Shiled Excellence Technology Co., Ltd. (Tianjin, China). Water was commercial ultrapure water. H. erinaceus (Bull.: Fr.) Pers. mycelium powder was purchased from Beijing Fuerkang Biotechnology research institute (Beijing, China) in August 2016. The scientific name was identified by one of the authors (Peilong Sun). The voucher specimen (ZJUT13000) was deposited at the Herbarium College of Pharmacy in Zhejiang University of Technology.

Preparation of H. erinaceum Mycelium Extracts
H. erinaceum mycelium powder (1000 g) was extracted three times with 95% ethanol (4L) under reflux for 4 h and at 80 °C using a stir bar. After removing ethanol by vacuum distillation at 55 °C, 76 g obtained ethanol extract was suspended in water (100 mL), and was extracted by petroleum ether and ethyl acetate in sequence. Ethyl acetate fraction was vacuum distilled, and 25 g crude extract was yield. The crude extract was named as HEM-E-E and stored at 4 °C for the HSCCC separation.

HSCCC Apparatus
HSCCC was performed on a TEB 300A (Tauto Biotechnique Company, Shanghai, China) high-speed counter-current chromatography apparatus. The apparatus consisted of three preparative coils connected in series (the inner diameter of tube, 1.5 mm; total volume, 280 mL) and a 10 mL sample loop. The revolution radius was 5 cm, and the β value was varied from 0.5 at the internal terminal to 0.8 at the external terminal (β = r/R, where r is the distance from the coil to the holder shaft, and R is the distance between the holder axis and the central axis of the centrifuge ) [41]. The rotation speed was ranged from 0 to 1000 rpm. The system was equipped with a model TBP-50A constant-flow pump (Tauto Bioteh, Shanghai, China), a model UV-500 detector (XUYUKJ Instruments, Hangzhou, China) operating at 254 nm, and a model N2000 workstation (Zhejiang University, Hangzhou, China). DC-2010 constant temperature-circulating implement (Hanagzhou Dawei Instrument, Hangzhou, China) was used to adjust the experimental temperature.

Reagents and Materials
All organic solvents for HSCCC (analytical grade) were purchased from Shanghai Lingfeng Chemical Reagent Co. Ltd. (Shanghai, China). Acetonitrile used for HPLC analysis (chromatographic grade) were purchased from Tianjin Shiled Excellence Technology Co., Ltd. (Tianjin, China). Water was commercial ultrapure water. H. erinaceus (Bull.: Fr.) Pers. mycelium powder was purchased from Beijing Fuerkang Biotechnology research institute (Beijing, China) in August 2016. The scientific name was identified by one of the authors (Peilong Sun). The voucher specimen (ZJUT13000) was deposited at the Herbarium College of Pharmacy in Zhejiang University of Technology.

Preparation of H. erinaceum Mycelium Extracts
H. erinaceum mycelium powder (1000 g) was extracted three times with 95% ethanol (4L) under reflux for 4 h and at 80 • C using a stir bar. After removing ethanol by vacuum distillation at 55 • C, 76 g obtained ethanol extract was suspended in water (100 mL), and was extracted by petroleum ether and ethyl acetate in sequence. Ethyl acetate fraction was vacuum distilled, and 25 g crude extract was yield. The crude extract was named as HEM-E-E and stored at 4 • C for the HSCCC separation.

Selection of Two-Phase Solvent
Two-phase solvent systems were selected on the base of the partition coefficient value (K) of the two target compounds. Three solvent systems: (1) n-hexane-ethyl acetate-methanol-water systems (v/v/v/v, 1:1:1:1 or 3:2:3:2 or 4:5:4:5), (2) chloroform-methanol-water systems (v/v/v/v, 4:4.5:2.5 or 5:4:2 or 5:5:2) and (3) chloroform-dichloromethane-methanol-water systems (v/v/v/v, 4:1.5:2:2 or 4:2:3:2 or 4:4.5:2:5) were investigated. The K values were determined by HPLC as follows: HEMP-E-E (5 mg) was added to the equilibration of two-phase solvent system, followed by vigorous shaking for 1 min. After two phases were completely separated, 1 mL of each phase was evaporated to dryness, dissolved in 1 mL of acetonitrile and the K values were determined by HPLC analysis. The peak areas of the upper phase and the lower phase were recorded as A U and A L , respectively. K value was obtained by the equation: K = A U /A L . The HPLC conditions were described in Section 3.7.

Preparation of the Two-Phase Solvent System and Sample Solution
The selected two-phase solvent system composed of chloroform-dichloromethane-methanol-water (v/v/v/v, 4:2:3:2) was used for HSCCC separation. The solvent system in a separatory funnel was violently shaken for thorough mixing. After equilibration, the two phases were separated and degassed by sonication for 15 min before used. The lower and upper phase were used as the mobile and stationary phase, respectively. About 150 mg of HEM-E-E were dissolved in the solvent mixture containing 5 mL the lower phase and 5 mL upper phase.

HSCCC Separation
Head-tail elution was performed for the separation of HEMP-E-E. The coiled column was first entirely filled with the upper phase of the solvent system. Then the apparatus was rotated at a speed of 900 rpm, and the lower phase was pumped into the column at a flow rate of 2 mL/min. When the hydrodynamic equilibrium was achieved, as indicated by a clear mobile phase eluting at the tail outlet. About 10 mL of HEMP-E-E solution was injected into the separation column through the injection valve. The separation temperature was controlled at 20 • C. The effluent was continuously monitored at 254 nm, and collected using a fraction collector set at 5 min for each tube. Each fraction was collected according to the chromatogram and evaporated under vacuum. Three HSCCC fractions were obtained, the first fraction contained some impurities (confirmed by HPLC analysis, data not shown), were no longer investigated. The other two fractions are set as two target compounds and named as compounds 1 and 2, respectively.

HPLC Analysis of HEM-E-E and Its HSCCC Fractions
The analytical HPLC equipment was a Waters 1525 system consisting of a Waters 1525 Binary pump, a Waters 2487 UV-vis Photodiode array detector, a Waters 2707 injection valve with a 20 µL loop, and a Waters HPLC workstation (Waters, Milford, MA, USA). The column applied in this work was a XTerra MS C18 column (250 mm × 4.6 mm, 5 µm, Waters, USA). The system run with a gradient program at 1 mL/min, and two solvents acetonitrile (A) and water (B) with the following gradient combinations: 0-10 min 30% B; 10-20 min, 30-60% B; 20-25 min, 60-90% B; 25-30 min, 90-60 % B; The eluent was monitored at 254 nm, and the purity was calculated by the target analytic peak area divided by the total peak area (unitary area method).

Identification of HSCCC Peak Fractions
The UV-vis spectra were recorded by a UV-1900 spectrophotometer (Puxi, Beijing, China) using a 1 cm path length cell with absorption wavelength at 254 nm.
The IR spectrum was recorded in KBr disc and the spectrum was scanned from 400 to 4000 cm −1 with a 6700 Nicolet Fourier transform-infrared spectrophotometer (Madison, WI, USA).
ESI-MS was performed by Waters SQD2 mass spectrometer (Waters, Milford, MA, USA), operating in positive mode. The MS conditions were as follows: capillary voltage 3 kV and temperature maintained at 300 • C, cone voltage 40 V, Mass-scan range were measured from m/z 50 to 1000, source temperature 120 • C, the gas flow rate for cone and desolvation (N 2 ) were 500 mL/min. Mass data in this manner provided for the collection of information of intact precursor ions.
In addition, two purified compounds were concentrated to dryness under reduced pressure and lyophilized, followed by dissolution in deuterated methanol (CH 3 DO) for NMR analysis. The 1 H and 13 C spectra were obtained on a Bruker Avance III 400 MHz NMR spectrometer (Bruker Biospin Co., Billerica, MA Rheinstetten, Germany). 1 H-nuclear magnetic resonance (NMR) and 13 C-NMR spectra were obtained at the center of analysis, Shanghai Microspectrcum Chemical Technology Service Co. Ltd. (Shanghai, China).

Conclusions
In this study, the upper and low phase of chloroform-dichloromethane-methanol-water at a volume ratio of 4:2:3:2 (v/v/v/v) was selected as the mobile and stationary phase, and the separation condition of H. erinaceum mycelium crude extract (named HEM-E-E) were selected as follow: flow rate 2.0 mL/min, rotary speed 900 rpm, column temperature 20 • C. Under the optimized HSCCC conditions, 23 mg compound 1 with the purity of 95.7% and 18 mg compound 2 with the purity of 97.5 % were isolated from 150 mg HEM-E-E. These two compounds were confirmed as genistein (4 ,5,7-Trihydroxyisoflavone) and daidzein (4 ,7-Dihydroxyisoflavone). To the best of our knowledge, this is the first report in which two isoflavones, genistein and daidzein, are isolated and discovered from H. erinaceum mycelium. The results also demonstrated that HSCCC method is a powerful tool for the quick and efficient separation and purification of bioactive compounds from natural products.