Comparison of Various Solvent Extracts and Major Bioactive Components from Unsalt-Fried and Salt-Fried Rhizomes of Anemarrhena asphodeloides for Antioxidant, Anti-α-Glucosidase, and Anti-Acetylcholinesterase Activities

The rhizome of Anemarrhena asphodeloides Bunge (AA, family Liliaceae) is a famous and frequently used herbal drug in the traditional medicine of Northeast Asia, under vernacular name “zhimu”. A. asphodeloides has been used as an anti-inflammatory, antipyretic, anti-platelet aggregation, anti-depressant, and anti-diabetic agent in traditional Chinese medicine. We examined the antioxidant, anti-acetylcholinesterase (AChE), and anti-α-glucosidase activities of various solvent extracts and the main bioactive compounds from the rhizome of A. asphodeloides. Acetone extract exhibited comparatively high antioxidant activities by 2,2-diphenyl-1-(2,4,6-trinitrophenyl)hydrazyl (DPPH), 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical scavenging, and ferric-reducing antioxidant power (FRAP) assays. A water extract exhibited relatively strong antioxidant activity by superoxide radical scavenging test. Furthermore, dichloromethane, chloroform, and n-hexane extracts showed significant anti-α-glucosidase activities. Finally, ethanol and dichloromethane extracts exhibited relatively strong AChE inhibitory activity. HPLC analysis was used to examine and compare various solvent extracts for their compositions of isolates. We isolated four major chemical constituents and analyzed their antioxidant, anti-α-glucosidase, and AChE inhibitory activities. The bioactivity assays showed that mangiferin displayed the most potential antioxidant activities via FRAP, ABTS, DPPH, and superoxide assays and also exhibited the most effective anti-AChE and anti-α-glucosidase activities among all the isolates. The present study suggests that A. asphodeloides and its active extracts and components are worth further investigation and might be expected to develop as a candidate for the treatment or prevention of oxidative stress-related diseases, AChE inhibition, and hyperglycemia.


Introduction
Diabetic treatment is based on carbohydrate enzymes (such as α-glucosidase and α-amylase) inhibition as well as other mechanisms, such as reduction of glucose level and maintenance of β-cell performance. It has been shown that the inhibition of α-glucosidase and α-amylase correlates with an increase in postprandial glucose level and reduces the absorbance of glucose in the intestine, limiting the excursion of glucose in plasma. The control of activity is an important feature in diabetes treatment [1].
Oxidative stress and reactive oxygen species (ROS) play an important role in some diseases, such as Alzheimer's disease (AD) and diabetes. Oxidative stress creates an excess of ROS/nitrogen species, leading to injury of the cell composition (such as lipids, protein enzymes, and nucleic acids) that may have a predictable aftermath in neurodegenerative disorders such as Alzheimer's [2] and metabolic ailments such as diabetes [3]. The use of acetylcholinesterase inhibitory drugs is the most appropriate modern and primary treatment method to confront neurodegenerative diseases [4]. Because the major enzyme in the pathogenesis of Alzheimer's disease is AChE, the suppression of AChE raises the signal transfer in cholinergic pathways and decreases the symptoms of Alzheimer's disease [5]. The use of synthetic drugs in therapy for AD such as galantamine has adverse reactions, including weight loss, diarrhea, loss of appetite, nausea, vomiting, dizziness, headache, stomach pain, and muscle weakness [6].
However, treatment of diabetes and AD with drugs such as galantamine is no longer reassuring and without side effects. In fact, many synthetic drugs may induce several undesirable symptoms, like gastrointestinal troubles or abdominal disorders [7]. Assuredly, natural products are favorable drug candidates because they are easy to get and relatively safe [8].
Anemarrhena asphodeloides, known as "zhimu" in Taiwan, which is listed in the Taiwan herbal pharmacopeia, is used to clear heat and resolve toxins. A. asphodeloides has also been found to have antifungal activity [9], decrease the blood glucose concentration [10], inhibit platelet agglutination and neoplastic activity [11,12], act as an antiallergic drug [13], and be an anti-dementia agent [14]. Furthermore, it has been reported that the pharmacological effects of A. asphodeloides are tightly connected to the methods of medicinal material processing. In other words, different processing methods will get different pharmacological effects [15]. In this study, the antioxidant, anti-α-glucosidase, and anti-AChE activities of A. asphodeloides ( Figure 1) and its active extracts and components were examined.

Preparation of A. asphodeloides Extract
The rhizomes of A. asphodeloides were purchased from Wanhua Dist., Taipei City, Taiwan, in July 2021 and identified by Prof. J.-J. Chen. A voucher specimen was deposited in the Department of Pharmacy, National Yang Ming Chiao Tung University, Taipei, Taiwan. Preparation of A. asphodeloides extract was performed as described previously [16].

Reverse-Phase HPLC
The assay for measuring reverse-phase HPLC of four components was carried out as described previously with a slight modification [16]. In order to obtain a suitable chromatographic condition, the mobile phase conditions were optimized. The mobile phase consisting of 0.2% acetic acid in water and acetonitrile showed better separating selectivity than that consisting of methanol and water. Gradient separation using 0.2% acetic acid in water (v/v) (solvent A) and acetonitrile (solvent B) as mobile phase was as follows: 0-3 min, 100% A with isocratic elution; 3-9 min, linear gradient from 0 to 5% B; 9-22 min, linear gradient from 5 to 13% B; 22-32 min, linear gradient from 13 to 16% B; 32-45 min, linear gradient from 16 to 35% B; 45-60 min, linear gradient from 35 to 65% B; 60-80 min, linear gradient from 65 to 80% B; 80-85 min, linear gradient from 80 to 100% B; 85-95 min, back to initial conditions at 0% B; and 95-105 min, at 0% B. The flow rate was 1.0 mL/min and the injection volume was 10 µL. Peaks were detected at 280 nm. Quantification of four components from Anemarrhena asphodeloides in each solvent extract was performed as described above.

Determination of Total Phenolic Content
Total phenolic content (TPC) of various solvent extracts and major bioactive components was performed according to the method previously reported [16].

Determination of Total Flavonoid Content
The total flavonoid content (TFC) was conducted by the reference method with a slight modification [17]. In short, the extracted sample was diluted with MeOH to reach a concentration of 100 µg/mL. Quercetin was diluted in MeOH (0-100 µg/mL) as standard. The diluted extract sample or quercetin (400 µL) was mixed with 10% (w/v) aluminum chloride solution (200 µL) and 0.1 mM potassium acetate solution (200 µL). The mixture was reacted at room temperature for 30 min. Then the absorbance of the mixture was measured at 415 nm. The TFC of the extracts was determined from a standard calibration curve using quercetin. TFC was expressed as milligram quercetin equivalent per gram A. asphodeloides (mg QCE/g AA).

DPPH Radical Scavenging Activity
The DPPH radical scavenging assay was measured by the reference method [18].

ABTS Radical Scavenging Activity
ABTS radical scavenging activity of each extract was determined as previously described [19].

Ferric-Reducing Antioxidant Power (FRAP) Assay
The FRAP assay was determined by the reference method with a slight modification [21,22]. The working solution was mixed with acetate buffer (pH 3.6), ferric chloride solution (20 mM), and TPTZ solution (10 mM TPTZ in 40 mM HCl) in a proportion of 10:1:1, respectively, and freshly prepared before being used. A total of 900 µL of the working solution was warmed to 37 • C and then mixed with 100 µL of the diluted sample, blank or standard, in a microcentrifuge tube. The tubes were vortexed in a dry bath at 37 • C for 40 min. The absorbance was measured at 593 nm. The standard curve was linear between 0 and 100 mM Trolox. Results are expressed in mM TE/g dry weight. Additional dilution was needed if the FRAP value measured was over the linear range of the standard curve.

α-Glucosidase Inhibitory Activity Assay
The inhibition assay of α-glucosidase was conducted using the procedure previously reported [23].

Acetylcholinesterase Inhibitory Activity Assay
The inhibition assay of acetylcholinesterase was conducted using the conditions previously reported with slight modifications [24]. Briefly, 0.1 M sodium phosphate buffer (pH 8.0, 140 µL), test compound solution (20 µL), enzyme solution (15 µL of AChE 0.2 units/mL solution), and DTNB (15 mM, 10 µL) were mixed and incubated for 10 min at room temperature. Substrate (10 µL of AchI 15 mM solution) was then added and the reaction was initiated.

Molecular Modeling Docking Study
The in silico evaluation was conducted in AutoDock Vina software [25]. The crystal structure (PDB: 3A4A) was retrieved from the Protein Databank and hydrogen atoms were added to prepare the docked receptor. The 3D structures of ligands were constructed in the Chem3D program. The hydrogen supplement, Gasteiger charge measurement for protein atoms, and selection of flexible torsions for ligands were conducted by Autodock-Tools (ADT ver. 1.5.6). The size of the grid was designed at 20 Å × 26 Å × 22 Å of mangiferin, 30 Å × 30 Å × 30 Å of timosaponin A-III, 20 Å × 25 Å × 20 Å of neomangiferin, 18 Å × 18 Å × 18 Å of isomangiferin, and 20 Å × 20 Å × 20 Å of acarbose. Additionally, the grid center at dimensions (x, y, and z, respectively) 21.2, −0.8, and 18.6 was determined. The binding affinity energy was provided as docking scores and is shown in kcal/mol. The best interaction was considered only the top-scoring pose. The visualization of the best docking interactions was analyzed in Biovia Discovery Studio client 2021 [26].

Statistical Analysis
All data are expressed as mean ± SEM. Statistical analysis was carried out using Student's t-test. A probability of 0.05 or less was considered statistically significant. All the experiments were performed at least 3 times.

Determination of Total Phenolic Content (TPC), Total Flavonoid Content (TFC), and Yields in Each Solvent Extract
We studied the TPC, TFC, and yields in various solvent extracts of A. asphodeloides (AA) and salt-fried Anemarrhena asphodeloides (salt-fried AA). Table 1 displays TPC, TFC, and extraction yields of n-hexane, chloroform, dichloromethane, ethyl acetate, acetone, methanol, ethanol, and water extracts from A. asphodeloides. The yields of different solvent extracts ranged from 0.3 ± 0.1% (n-hexane extract) to 76.7 ± 4.4% (water extract) of AA, and 0.3 ± 0.1% (n-hexane extract) to 72.9 ± 2.5% (water extract) of salt-fried AA. The water extract exhibited the largest yield among all extracts possibly due to its abundant amounts of high polar compounds. Significant differences were found in TPC among all various solvent extracts, of which acetone extract from AA contained the largest amount of TPC (45.3 ± 1.2 mg/g). In addition, ethyl acetate extract from salt-fried AA contained the highest amount of TPC (78.2 ± 3.0 mg/g). The TFC among different solvent extracts ranged from 5.8 ± 2.2% (ethyl acetate extract) to 42.9 ± 4.9% (n-hexane extract) of AA, and 8.3 ± 1.4% (n-hexane extract) to 45.4 ± 4.2% (acetone extract) of salt-fried AA.

DPPH Free-Radical Scavenging Activity
The DPPH radical scavenging activity of each extract is shown in Table 2, and butylated hydroxytoluene (BHT) was used as the positive control. From the results of our tests, acetone extract (IC 50 = 123.6 ± 3.9 µg/mL) of AA and water extract (IC 50 = 113.9 ± 8.6 µg/mL) of salt-fried AA exhibited relatively strong antioxidant activities by DPPH radical scavenging assay.

Ferric-Reducing Antioxidant Power (FRAP) Assay
The FRAP assay of each extract is shown in Table 2, and butylated hydroxytoluene (BHT) was used as the positive control. The FRAP assay is expressed as millimolar (mM) of Trolox equivalents (TE) per gram of extract. From the results of our tests, acetone extract (381.5 ± 8.3 mM TE/g) of AA and ester acetate extract (485.5 ± 4.3 mM TE/g) of salt-fried AA exhibited relatively high antioxidant powers via FRAP assays.
Based on the above data of ABTS, DPPH radical scavenging activity, and FRAP assays, acetone extract displayed the strongest antioxidant activity of all the solvent extracts. Additionally, water extract exhibited relatively strong antioxidant activity by superoxide radical scavenging test.

Acetylcholinesterase (AChE) Inhibitory Activity Assay
The AChE inhibitory activity of each extract is shown in Table 4, and galanthamine was used as the positive control. From the results of our tests, ethanol extract (IC 50 = 117.1 ± 2.6 µg/mL) of AA and water extract (IC 50 = 73.6 ± 2.8 µg/mL) of salt-fried AA exhibited relatively high AChE inhibitory activities among all solvent extracts.

Quantitation of Active Components in Different Solvent Extracts
Figures S1-S15 display the quantification of active components in different solvent extracts from Anemarrhena asphodeloides by reverse-phase HPLC analyses. The contents of four active components in each solvent extract are shown in Table 5. The total quantity of four bioactive components in each extract of AA ranged from a maximum of 47.5 ± 1.5 mg/kg (chloroform extract) ( Figure 2) to a minimum of 9.6 ± 0.5 mg/kg (ethanol extract) in the following order: chloroform > n-hexane > acetone > dichloromethane > water > ethyl acetate > methanol > ethanol extract. Additionally, the total quantity of four bioactive components in each extract of salt-fried AA ranged from a maximum of 46.5 ± 1.4 mg/kg (chloroform extract) ( Figure 3) to a minimum of 12.6 ± 0.8 mg/kg (water extract) in the following order: chloroform > n-hexane > ethyl acetate > dichloromethane > acetone > methanol > ethanol > water extract. Chloroform (47.5 ± 1.5 mg/kg in AA and 46.5 ± 1.4 mg/kg in salt-fried AA) extracts exhibited larger amounts of four active components compared with other extracts. Isomangiferin was the most abundant among the four active compounds in organic solvent extract, followed by timosaponin A-III, neomangiferin, and mangiferin.

Antioxidant Activities of Isolated Components
The isolated compounds mangiferin, timosaponin A-III, neomangiferin, and isomangiferin ( Figure 4) were measured for their antioxidant activities, including DPPH, ABTS, superoxide radical scavenging activity, and FRAP assays. Results are shown in Table 6. Mangiferin (IC 50 = 5.4 ± 0.3 µg/mL) exhibited the highest DPPH radical scavenging activity, followed by isomangiferin (IC 50 = 16.7 ± 1.1 µg/mL), timosaponin A-III (IC 50 > 200 µg/mL), and neomangiferin (IC 50 > 200 µg/mL). Mangiferin (IC 50 = 3.7 ± 0.2 µg/mL) also exhibited stronger antioxidant activity than other isolates by ABTS and superoxide radical scavenging assays. Furthermore, mangiferin (9371.4 ± 183.9 mM TE/g) exhibited relatively higher antioxidant activity than others via FRAP assay. other extracts. Isomangiferin was the most abundant among the four active compounds in organic solvent extract, followed by timosaponin A-III, neomangiferin, and mangiferin.    other extracts. Isomangiferin was the most abundant among the four active compounds in organic solvent extract, followed by timosaponin A-III, neomangiferin, and mangiferin.     and neomangiferin (IC50, >200 μg/mL). Mangiferin (IC50 = 3.7 ± 0.2 μg/mL) also exhibited stronger antioxidant activity than other isolates by ABTS and superoxide radical scavenging assays. Furthermore, mangiferin (9371.4 ± 183.9 mM TE/g) exhibited relatively higher antioxidant activity than others via FRAP assay. According to the above results, the acetone extract of A. asphodeloides and ethyl acetate extract of salt-fried A. asphodeloides contained the highest amount of main phenolic components and therefore the highest antioxidant power.    According to the above results, the acetone extract of A. asphodeloides and ethyl acetate extract of salt-fried A. asphodeloides contained the highest amount of main phenolic components and therefore the highest antioxidant power.

Acetylcholinesterase (AChE) Inhibitory Assays of Isolated Components
For further evaluation of the AChE inhibitory activity, we conduct anti-AChE assays of the main components isolated from A. asphodeloides. Among the isolated compounds, mangiferin exhibited higher anti-AChE activity than other isolates. Table 8 shows the AChE inhibitory effects of the major active components from A. asphodeloides.

Molecular Docking Study
The 3D crystal structure of α-glucosidase showed that it mainly contained numerous structural domains, including the N-terminal domain, the barrel domain where the active site is located, and the C-terminal domain. The α-glucosidase active site was primarily formed by numerous β-sheets and several loops or α-helices. Importantly, the active site in different species showed certain conformational similarities. It mainly contained hydrophilic residues and allowed compounds with distinct sizes to enter. To further study how compounds might interact with α-glucosidase of Saccharomyces cerevisiae to exhibit its antagonistic effect, the docking models of compounds were generated with the Discovery Studio 2021 (Accelrys, San Diego, CA, USA) modeling program. According to the anti-α-glucosidase experimental data, mangiferin (  Figure 9) were selected to determine their binding abilities to the crystal structure of isomaltase from Saccharomyces cerevisiae.
The 3D crystal structure for α-glucosidase of Saccharomyces cerevisiae is not available at this moment, so the crystal structure (PDB: 3A4A) of Saccharomyces cerevisiae (PDB: 3A4A) containing 72% sequence homology with α-glucosidase from Saccharomyces cerevisiae is usually used to perform the docking study and was also employed in this study. In the crystal structure (PDB: 3A4A), the configuration of its active site is quite similar to that of α-glucosidase from beta vulgaris, but it is deep and narrow. The crystal structure of this active site reveals that its co-crystallized ligand, α-D-glucopyranose, is located deep in the ligand-binding pocket and makes three essential H-bond interactions, including (1) the 4-hydroxyl group interacting with His 351 and Asp 352 by acting as the H-bond donor; (2) the 5-hydroxyl group making H-bond contact Glu 277 as well as Asp 352 by acting as the H-bond donor, and also serving as the H-bond acceptor to interact with Arg 213; and (3) the 6-hydroxymethyl group acting as the H-bond donor to interact with Glu 277 and Asp 352.
Before the docking simulation, the ligand (α-D-glucopyranose) included in the 3A4A PDB file was redocked for validation. The interactions between alpha-D-glucopyranose and 3A4A and the best pose of the calculated results showed high similarity and repeatability with native data (data not shown). The results indicate the high accuracy of the existing simulation system and supported further computing.

Discussion
Diverse methods have been utilized and developed for the extraction of natural products (i.e., herbs, plants, and fungi) for application as alternatives to modern medicines. The most common method that has been applied for many years is boiling or making a decoction with water, a very easy and economically feasible process. Nowadays, medicinal plant-based studies have been conducted and many are ongoing, as various metabolites with health benefits have been found in natural products. Recently, in more advanced studies, organic solvents are used to get natural product extracts, including a variety of metabolites, according to the polarity and property of the component of interest [27]. Other factors that may affect the process applied to natural product extraction comprise the kind of solvent to be used, the temperature set during extraction, the properties of the plant material, and the target metabolites [28]. The change in solvent polarity results in a significant difference in phytochemical components and bioactivities. Thus, we utilized solvents of various polarity to get the rhizome fractions from A. asphodeloides in an effort to assess these different metabolites. We found various metabolites with different degrees of bioactivity due to their differences in solvent polarity.
ABTS and DPPH assays have been widely utilized to assess the antioxidant activities of natural components. Both assays are mostly connected with the proton radical scavenging or hydrogen donating capacities of the target compounds [29]. Superoxide radical scavenging activity is measured by the PMS/NADH-NBT system. Superoxide anion radicals generated from dissolved oxygen by PMS/NADH coupling reaction reduce NBT. The decrease in absorbance at 560 nm with antioxidants indicates a reduction in superoxide anion radicals in the reaction mixture [30]. The ferric-reducing antioxidant power (FRAP) measures the antioxidant potential of each extract through the reduction of ferric iron (Fe 3+ ) complex to ferrous iron (Fe 2+ ) complex by antioxidants present in the samples [31]. In our study, ethyl acetate and acetone extract in salt-fried rhizomes of A. asphodeloides displayed high antioxidant activities among all solvent extracts via ABTS and FRAP assays, which may be connected to TPC in the extracts, and the ethyl acetate extract showed The lowest binding energy of each ligand was considered the best conformation. The binding affinities are listed in Table 9. In this study, acarbose was used as positive control. Compared with acarbose, the binding energies of mangiferin, timosaponin A-III, and isomangiferin were lower than −2.3 kcal/mol (Table 9). This suggests that mangiferin, timosaponin A-III, and isomangiferin could dock into the pocket of the crystal structure of isomaltase from Saccharomyces cerevisiae similar to or even better than that of acarbose. Table 9. Binding energies of active components and acarbose calculated in silico.

Compounds
Affinity (kcal/mol) a Acarbose used as positive control.
As show in Figure 5, mangiferin was bound with Ser 311, Asp 307, Gln 279, and Asp 215 through conventional hydrogen bonds, and other interactions (π-anion, π-π T-shaped, and π-alkyl) were also observed with Asp 352, Phe 178, Tyr 158, and Val 216. Arg 315 formed a carbon hydrogen bond of mangiferin. In addition, the two hydroxyl groups provided unfavorable donor-donor interactions with Arg 442 and His 280 residues. These allowed mangiferin and protein to form a stable complex.
Finally, isomangiferin was bound with Asp 242 and Arg 442 through conventional hydrogen bonds, and other interactions (π-anion and π-π stacked) were also observed with Glu 277 and Phe 303. In addition, unfavorable acceptor-acceptor interactions were observed with Glu 353 (Figure 7).
According to our data, the docking scores of mangiferin, timosaponin A-III, and isomangiferin were higher than those of neomangiferin and acarbose, which suggests better binding capability. In this study, the active ingredients mangiferin, timosaponin A-III, and isomangiferin possessed not only anti-α-glucosidase activity, but also better binding potential with the active sites of S. cerevisiae α-glucosidase. This indicates that these compounds may deserve further investigation as natural α-glucosidase inhibitors.

Discussion
Diverse methods have been utilized and developed for the extraction of natural products (i.e., herbs, plants, and fungi) for application as alternatives to modern medicines. The most common method that has been applied for many years is boiling or making a decoction with water, a very easy and economically feasible process. Nowadays, medicinal plant-based studies have been conducted and many are ongoing, as various metabolites with health benefits have been found in natural products. Recently, in more advanced studies, organic solvents are used to get natural product extracts, including a variety of metabolites, according to the polarity and property of the component of interest [27]. Other factors that may affect the process applied to natural product extraction comprise the kind of solvent to be used, the temperature set during extraction, the properties of the plant material, and the target metabolites [28]. The change in solvent polarity results in a significant difference in phytochemical components and bioactivities. Thus, we utilized solvents of various polarity to get the rhizome fractions from A. asphodeloides in an effort to assess these different metabolites. We found various metabolites with different degrees of bioactivity due to their differences in solvent polarity.
ABTS and DPPH assays have been widely utilized to assess the antioxidant activities of natural components. Both assays are mostly connected with the proton radical scavenging or hydrogen donating capacities of the target compounds [29]. Superoxide radical scavenging activity is measured by the PMS/NADH-NBT system. Superoxide anion radicals generated from dissolved oxygen by PMS/NADH coupling reaction reduce NBT. The decrease in absorbance at 560 nm with antioxidants indicates a reduction in superoxide anion radicals in the reaction mixture [30]. The ferric-reducing antioxidant power (FRAP) measures the antioxidant potential of each extract through the reduction of ferric iron (Fe 3+ ) complex to ferrous iron (Fe 2+ ) complex by antioxidants present in the samples [31]. In our study, ethyl acetate and acetone extract in salt-fried rhizomes of A. asphodeloides displayed high antioxidant activities among all solvent extracts via ABTS and FRAP assays, which may be connected to TPC in the extracts, and the ethyl acetate extract showed the strongest antioxidant activity. The differences in antioxidant capacities of the extracts may be due to the different extents of TPC or the composition of antioxidant compounds in the extracts.
The comparative evaluation of total phenolic content (TPC), total flavonoid content (TFC), and antioxidant assays (DPPH, ABTS, superoxide, and FRAP) of various solvent extracts (n-hexane, chloroform, dichloromethane, EtOAc, acetone, EtOH, MeOH, and water) from the salt-fried and non-salt-fried rhizomes of A. asphodeloides was first mentioned in this study. This can provide a guide for the selection of appropriate solvents in TPC, TFC, and antioxidant extraction applications. Based on the antioxidant data, mangiferin displayed potent antioxidant properties. Similar experimental results could also be found in past studies [32]. Antioxidant assays of isomangiferin (FRAP), neomangiferin (superoxide), and timosaponin A-III (DPPH, ABTS, superoxide, and FRAP) were first evaluated in this study.
An anti-α-glucosidase agent could suppress the activity of glucosidases in the small intestine, which split the glycosidic bonds in carbohydrates so that it decreases the glucose release from food. The inhibitors studied were classified into non-sugar and sugar-mimicking types on the basis of their chemical structure. The anti-α-glucosidase drugs for clinical treatment such as voglibose, acarbose, and miglitol all belong to the sugarmimicking type. Nevertheless, the α-glucosidase inhibitors with the non-sugar type have received the attention of researchers due to the limitations of sugar-mimicking inhibitors. Mangiferin possesses a similar basic xanthone backbone, which is the molecular basis for anti-α-glucosidase [33]. In anti-α-glucosidase assay, mangiferin, timosaponin A-III, and isomangiferin showed more effective anti-α-glucosidase activities than acarbose (positive control). Our data show that mangiferin was about five-fold stronger than acarbose against α-glucosidase.
According to the results of the anti-α-glucosidase assay, mangiferin exhibited the most potent anti-α-glucosidase activity among all isolated compounds. Thus, the interaction between α-glucosidase and mangiferin was evaluated by molecular modeling docking. As a result of molecular docking, mangiferin exhibited high affinity with α-glucosidase. Similar experimental results could also be found in past studies [34,35]. Molecular docking study for timosaponin A-III and isomangiferin with S. cerevisiae α-glucosidase was first evaluated in this study. Furthermore, the binding energies of mangiferin, timosaponin A-III, neomangiferin, and isomangiferin with S. cerevisiae α-glucosidase were also first calculated in our study.
Alzheimer's disease (AD) is a progressive degenerative disease of the brain that is characterized by the deterioration of cognitive and memory functions. Treatments have focused on increasing brain cholinergic activity via AChE inhibitors. Actually, an AChE inhibitor, galanthamine, has been approved to treat AD [14,36]. In previous studies, mangiferin was supposed to possess some potential in postponing the progression of memory loss in AD. Mangiferin has drawn much attention from many researchers, as it may be able to improve memory function of the brain via diverse mechanisms [37]. In our study, mangiferin and neomangiferin also exhibited anti-AChE activity and deserve further study.

Conclusions
Various solvent extracts of A. asphodeloides were investigated with various antioxidant systems, anti-α-glucosidase, and acetylcholinesterase (AChE) inhibitory activity assays. In our study, ethyl acetate extract of salt-fried A. asphodeloides displayed the strongest antioxidant activity among all solvent extracts via ABTS and FRAP assays. Dichloromethane extract of A. asphodeloides showed the highest anti-α-glucosidase activity among all solvent extracts. Water extract of salt-fried A. asphodeloides exhibited the most potential DPPH and superoxide radical scavenging and anti-AChE activities among all solvent extracts. Four isolated compounds from A. asphodeloides were quantified by HPLC and identified as mangiferin, timosaponin A-III, neomangiferin, and isomangiferin. Moreover, the comparative evaluation for the identification and quantification of the major active components (mangiferin, timosaponin A-III, neomangiferin, and isomangiferin) of different solvent extracts (n-hexane, chloroform, dichloromethane, EtOAc, acetone, EtOH, MeOH, and water) from the salt-fried and non-salt-fried rhizomes of A. asphodeloides by HPLC analysis was first conducted in our study.
The bioactivity assays showed that mangiferin displayed the most potential antioxidant activity via DPPH, ABTS, superoxide, and FRAP assays and also exhibited the most effective anti-α-glucosidase and anti-AChE activities among all isolates. Mangiferin, timosaponin A-III, and isomangiferin possess significant anti-α-glucosidase potential. Further molecular docking computing results supported mangiferin, timosaponin A-III, and isomangiferin exhibiting a binding affinity to the active pocket of S. cerevisiae α-glucosidase.
The above active extracts and their active components can be used as potential natural antioxidants. Furthermore, mangiferin also can be used as a natural anti-α-glucosidase and anti-AChE agent.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/antiox11020385/s1. Table S1: Retention time, LODs, LOQs, and regression analysis for three components in Anemarrhena asphodeloides (AA) in reverse phase. Figure S1: Reverse-phase HPLC chromatogram of isolated pure compounds. Figure S2: Reverse-phase HPLC chromatogram of methanol extract in AA. Figure S3: Reverse-phase HPLC chromatogram of ethanol extract in AA. Figure S4: Reverse-phase HPLC chromatogram of acetone extract in AA. Figure S5: Reverse-phase HPLC chromatogram of ethyl acetate extract in AA. Figure S6: Reverse-phase HPLC chromatogram of dichloromethane extract in AA. Figure S7: Reverse-phase HPLC chromatogram of n-hexane extract in AA. Figure S8: Reverse-phase HPLC chromatogram of water extract in AA. Figure S9: Reversephase HPLC chromatogram of methanol extract in salt-fried AA. Figure S10: Reverse-phase HPLC chromatogram of ethanol extract in salt-fried AA. Figure S11: Reverse-phase HPLC chromatogram of acetone extract in salt-fried AA. Figure S12: Reverse-phase HPLC chromatogram of ethyl acetate extract in salt-fried AA. Figure S13: Reverse-phase HPLC chromatogram of dichloromethane extract in salt-fried AA. Figure S14: Reverse-phase HPLC chromatogram of n-hexane extract in salt-fried AA. Figure S15: Reverse-phase HPLC chromatogram of n-hexane extract in salt-fried AA.