LC-Q-Orbitrap-MS/MS Characterization, Antioxidant Activity, and α-Glucosidase-Inhibiting Activity With In Silico Analysis of Extract From Clausena Indica (Datz.) Oliv Fruit Pericarps

Clausena indica (Datz.) Oliv fruit pericarps (CIOPs) is an important agro-industrial by-product rich in active components. In this article, the effects of traditional and green deep eutectic solvents (DESs) on the high-performance liquid chromatography (HPLC) characterization, antioxidant activities, and α-glucosidase-inhibitory activity of phenolic extracts from CIOPs were investigated for the first time. The results showed that ChCl-Gly and Bet-CA had higher extraction efficiency for the total phenolic content (TPC, 64.14–64.83 mg GAE/g DW) and total flavonoid content (TFC, 47.83–48.11 mg RE/g DW) compared with the traditional solvents (water, methanol, and ethyl acetate). LC-Q-Orbitrap-MS/MS was adopted to identify the phenolic compositions of the CIOPs extracts. HPLC-diode array detection (HPLC-DAD) results indicated that arbutin, (–)-epigallocatechin, chlorogenic acid, procyanidin B1, (+)-catechin, and (–)-epicatechin were the major components for all extracts, especially for deep eutectic solvents (DESs). In addition, ChCl-Xyl and ChCl-Gly extracts showed higher antioxidant activities against 2,2-diphenyl-1-picrylhydrazyl (DPPH•), 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulphonic acid (ABTS+•), ferric reducing antioxidant power (FRAP), reducing power (RP), and cupric ion reducing antioxidant capacity (CUPRAC) than extracts extracted by other solvents. A strong α-glucosidase-inhibiting activity (IC50, 156.25-291.11 μg/ml) was found in three DESs extracts. Furthermore, in silico analysis of the major phenolics in the CIOPs extracts was carried out to explore their interactions with α-glucosidase. Multivariate analysis was carried out to determine the key factors affecting the antioxidant activity and α-glucosidase-inhibiting activity. In short, DES can be taken as a promising solvent for valorization and recovery of bioactive compounds from agro-industrial by-products. The results verified that CIOPs can be used as a prospective source rich in bio-active compounds applied in the food and pharmacy industries.


INTRODUCTION
Clausena indica (Datz.) Oliv, which belongs to the wild evergreen arbor plant of the Rutaceae family, is mainly distributed in the Southern and Southeastern Asian countries (1). In China, it is mainly distributed in Hainan Island, Guangxi, and Yunnan, etc. The fruit of C. indica is commonly known as "Jipi fruit, " a rare and special sour-sweet berry with an intense aroma that can be used as a type of traditional Chinese medicine as well. In the folk, C. indica fruit is widely used for strengthening the spleen and improving human immunity (2). Many researchers have confirmed that C. indica fruit is rich in polyphenols, coumarins, alkaloids, terpenoids, and has various biological activities, such as anti-oxidation, anti-diabetics, anti-bacteria, anti-inflammation, lipid-modifying, and liver-protective effects (3). At present, the processed products of C. indica fruit mainly include beverage, jam, and preserved fruit, but a large amount of C. indica fruit pericarps (CIOPs) are often discarded as waste in the process of industrial processing, causing considerable environmental pollution. Therefore, it is of paramount importance to realize the valorization and recovery of bioactive compounds from CIOPs.
In general, recovering compounds from the agro-industrial by-products or plant matrix is performed using organic solvents. Although organic solvents extract bio-active components, they have inevitable shortcomings, such as toxicity, non-degradability, low boiling points, and high flammability (4)(5)(6)(7)(8). Currently, deep eutectic solvents (DESs) are synthesized with hydrogen bond acceptors (HBAs) and hydrogen bond donors (HBDs) under a low melting temperature. DESs are favored for the simple preparation process, lower synthesis cost, biodegradability, negligible volatility, non-flammability, favorable stability, and renewability (5,(9)(10)(11). It is particularly worth mentioning that DESs can badly damage the structure of plant cell walls by dissolving lignocellulose and lignin and thereby significantly increasing the extraction yields of bioactive components from the agro-industrial by-products (12,13). de Almeida Pontes et al. (2021) have reported that DESs are innovative, environmentally friendly, and high-performance solvents for extracting the phenolic compounds from olive leaves (14). Shang et al. (2019) adopted 20 types of DESs to extract isoflavones from chickpea sprouts and found that a mixture of choline chloride and propylene glycol (1:1, mol/mol) showed excellent extraction efficiency for isoflavone compounds (15). Marcos et al. (2020) found that a tailor-made eutectic solvent yielded the highest contents of phenolic compounds, anthocyanins, total sugars, and acid sugars from strawberry or raspberry, and the extract also presented better antioxidant activity (12). To date, there are few reports on the chemical characterization and green valorization of bio-active compounds from CIOPs.
This study aimed to systematically investigate the impacts of conventional and greenly solvents on the high-performance liquid chromatography (HPLC) characterizations, antioxidant activities, and α-glucosidase-inhibiting activity of phenolic extracts from CIOPs. The phenolic compositions of the CIOPs extracts were identified and quantified by LC-Q-Orbitrap-MS/MS for the first time. In silico analysis was carried out to investigate the binding mechanisms of major phenolic compounds to α-glucosidase. A multivariate analysis was performed to determine the main contributors to the antioxidant activity and α-glucosidase-inhibiting activity of the CIOPs extracts. This study may provide important evidence for the valorization and utilization of C. indica fruit pericarps.

Preparation of DESs
Deep eutectic solvents were synthesized according to the procedure described in the previous study (16). The starting components were added to a flask with a suitable molar ratio and heated at 80 • C until the formation of transparent and homogeneous liquid. Then, 30% ultra-pure water (w/w) was added to reduce the DES viscosity for subsequent extraction. Table 1 shows the information of the starting components for DESs preparation.

Extraction of Phenolic Compounds From CIOPs
Clausena indica (Datz.) Oliv fruit pericarps powder (0.5 g) was mixed with 5 ml of extraction solvents (H 2 O, 50% MeOH, EtAc, and 18 types of DESs) in 10 ml tubes, respectively. The extraction procedure was performed in an ultrasonic bath at 320 W, 40 • C for 30 min, followed by centrifugation at 10,000 × g for 5 min to collect supernatants.

Determination of Total Phenolic Content (TPC) and Total Flavonoid Content (TFC)
Total phenolic content (TPC) in the CIOPs extracts was determined with the colorimetric Folin-Ciocalteu method (17). Briefly, 20 µl of the CIOPs extracts were incubated with 200 µl of Folin-Ciocalteu reagent at 25 • C for 5 min, followed by the addition of 400 µl of saturated Na 2 CO 3 solution for another 30 min of incubation, and then, the absorbance was tested at 765 nm. Data were denoted as milligrams of gallic acid equivalents (GAE) per gram dry weight (DW) of CIOPs (mg GAE/g DW). The calibration curve of gallic acid (Y = 0.0032 X -0.0004, R 2 = 0.996) was drawn. Total flavonoid content (TFC) was tested with the aluminum chloride method (18,19). Let 100 µl of extract react with 50 µl of 5% NaNO 2 (w/v) for 5 min and then, add 50 µl of 10% AlCl 3 (w/v) for another 6 min of reaction. Finally, 400 µl of 1 M NaOH and 400 µl of water were mixed and incubated for 15 min before reading the absorbance at 510 nm. The calibration curve for rutin (Y = 0.0006 X -0.0143, R² = 0.997) was drawn. Data were denoted as milligrams of rutin equivalents (RE) per gram dry weight (DW) of CIOPs (mg RE/g DW).

Identification and Quantification of Phenolic Compositions
All the extracts were subjected to a 0.22 µm filter before being analyzed. The identification of phenolic compounds was performed by using an Agilent 1,  (17). A linear standard curve was plotted using series dilutions of standards with known concentrations (Supplementary Table 1). The contents of the analytes were expressed as milligram per gram dry weight (mg/g DW).

Evaluation of Antioxidant Activities In vitro
Antioxidant activities in vitro of the CIOPs extracts were determined using five methods, such as DPPH • , ABTS +• , FRAP, RP, and CUPRAC assay. The DPPH • and ABTS +• radicalscavenging activity of the CIOPs extracts were determined using the method proposed by Wu et al. (20). CUPRAC assay was performed according to the procedure reported by Wang et al. with 100 µl of the extracts or individual phenolics at 37 • C for 10 min, followed by the addition of 100 µl of 5 mM p-NPG solution for 20 min of reaction at 37 • C, and then the addition of 500 µl of 1 M Na 2 CO 3 solution for termination. The absorbance was measured at 405 nm. The α-GIA of the extracts was repressed as half inhibit concentration (IC 50 ) value. The α-GIA was calculated based on the following equation using acarbose and PBS as the positive and negative control, respectively:

In silico Analysis
To provide deep insight into the interaction between the main phenolics and α-glucosidase, an in silico docking study in Surflex-Dock Geom (SFXC) mode was performed using SYBYL-X 2.0 software (Tripos, Inc., St. Louis, MO, USA). The 2D structures of the main phenolics and acarbose were plotted using ChemBio3D Ultra software (MA, USA). The homologous structure of αglucosidase (PDB ID: 3A4A) was obtained from the RCSB Protein Data Bank (RCSB PDB). Acarbose, as a well-known inhibitor for α-glucosidase, was adopted as the positive control.
After docking with α-glucosidase, the key parameters of the docking of the ligand with α-glucosidase were generated. Based on the previous reports, a C-score ≥ of 4 was deemed as a credible docking result. T-score value, as a weighted sum of nonlinear functions, indicates van der Waals surface distance of the interaction of the ligand with α-glucosidase in docking analysis. The number and distances of hydrogen bonds and residues of amino acid active sites may clarify the interactive effects between the main phenolics and α-glucosidase (23).

Statistical Analysis
All results were measured three times and denoted as the mean and SD values (mean ± SD). The IC 50 value was obtained via Probit analysis. The multivariate analysis and statistical analysis were carried out by SPSS v26.0 software using one-way One Factor ANOVA. The differences were significant at a level of p < 0.05.

Total Phenolic Content and Total Flavonoid Content
Deep eutectic solvents, as a mixture of different types of HBDs and HBAs, have different physicochemical properties, such as viscosity, polarity, pH, and solubility. Consequently, DESs greatly affect the extraction efficiency for bio-active components from agro-industrial by-products (24). In this work, the traditional solvents and 18 types of DESs were used to extract TPC and TFC from CIOPs, and their extraction efficiencies were analyzed (Figures 1A,B). It was found that ChCl-Xyl and ChCl-Gly were more effective in extracting TPC (64.14-64.83 mg GAE/g DW) than other DESs. Intriguingly, except ChCl-Glu, ChCl-MA-Pro, and LA-Glu (13.05-24.72 mg RE/g DW, respectively), other DESs had high efficiency in the extraction of TFC. Regarding the traditional solvents, we found that 50% MeOH extract showed the highest TPC (50.75 mg GAE/g DW), followed by water (26.38 mg GAE/g DW) extracts. EtAc extract had the lowest TPC (1.84 mg GAE/g DW) and TFC (3.93 mg RE/g DW). It was observed that acidic-or polyalcohol-based DESs had higher extraction efficiencies for phenolics/flavonoids compounds from CIOPs than amide-and sugar-based DESs did, which was consistent with the results of the previous studies (24,25). As we know, the viscosity of DES is a critical factor affecting solid-liquid extraction. Excessively high viscosity greatly influenced massand energy-transfer, thereby affecting the extraction efficiency of active compounds from agro-industrial by-products. The viscosities of sugar-based DESs were significantly higher than those of other types of DESs (26,27). In addition, the polarity of the solvents also affected the extraction efficiency of phenolic compounds significantly. DESs with a wide range of polarity were reported to have high efficiency in the extraction of bioactive compounds from natural products (28,29). Taken together, ChCl-Xyl, ChCl-Gly, and Bet-CA could remarkably enhance the extraction of TPC and TFC from CIOPs, so they were chosen as the extraction solvents, and their compositions and bio-activities were comparatively analyzed. Sarikurkcu et al. (30) determined the yields of total phenolic and flavonoids from Onosma pulchra by using EtAc, MeOH, and water as the extraction solvents, and verified EtAc had the lowest extraction efficiency, which was in line with the result of this study. Zhu et al. (16) investigated the effects of different solvents (water, organic solvents, and DESs) on the extraction of phenolic compounds from Morinda citrifolia L. leaves, and found that DESs extracts yielded higher contents of phenolic compounds and had stronger biological activities, which was in accordance with the results of the current studies.

Phenolic Compositions
As shown in Table 2 and Figure 2, identification of chemical constituents in the CIOPs extracts was implemented by comparing retention time (RT), mass fragmentation pattern, and accurate mass information with public databases and the standards.   According to data given in Table 3, 10 identified phenolic compounds from the CIOPs extracts were quantified by commercial standards (Supplementary Table 1). Three DES extracts showed high contents of individual phenolics, such as arbutin (0.80 mg/g DW in ChCl-Xyl), (-)-epicatechin (0.72 mg/g DW in ChCl-Xyl), (-)-epigallocatechin (11.43 mg/g DW in Bet-CA), chlorogenic acid (2.70 mg/g DW in ChCl-Gly), procyanidin B1 (2.35 mg/g DW in ChCl-Gly), (+)-catechin (3.41 mg/g DW in ChCl-Gly), and taxifolin (0.42 mg/g DW in ChCl-Gly). Protocatechuic acid, caffeic acid, and taxifolin with relatively low contents can be detected in the extracts extracted with traditional solvents. In addition, taxifolin can be only detected in the extracts extracted using water, 50% MeOH, and ChCl-Gly. Vanillin was only detected in the three DESs extracts. Regarding the extracts extracted using traditional solvents, water, and 50% MeOH extracts showed high contents of (-)-epigallocatechin, chlorogenic acid, procyanidin B1, and (+)-catechin. Only four phenolic compounds [(-)-epigallocatechin, chlorogenic acid, procyanidin B1, (-)-epicatechin] with the lowest contents can be detected in the EtAc extract. Meanwhile, all identified phenolic compounds existed in water extract, except for vanillin, as compared to extracts by ChCl-Xyl, ChCl-Gly, and Bet-CA. The present work confirmed that extraction solvents significantly affected the contents of phenolic compounds in the CIOPs extracts (34,35). Simultaneously, eco-friendly DESs exhibited higher efficiency in extracting the individual phenolic compounds from CIOPs than traditional solvents, so they can be used for valorization and recovery of special high-value compounds from agro-industrial by-products.

Antioxidant Activities In vitro
In this article, various assays (DPPH • , ABTS • , FRAP, RP, and CUPRAC) were adopted to comprehensively evaluate the antioxidant activities of CIOP extracts. As shown in Table 4, three DESs extracts showed higher antioxidant abilities in DPPH • , ABTS • , FRAP, RP, and CUPRAC. Especially, ChCl-Xyl extracts exerted the strongest abilities in DPPH • (109.00 µmol TE/g DW), ABTS • (411.08 µmol TE/g DW), FRAP (1145.07 µmol TE/g DW), and RP (808.90 µmol TE/g DW) and CUPRAC (325.36 µmol TE/g DW), respectively. Better antioxidant activities were found in 50% MeOH extract as compared with water and EtAc extracts. EtAc extract had the lowest antioxidant activities. Remarkably, antioxidant activities of the CIOPs extracts varied with the type of solvent. The extracts with larger TPC/TFC had significantly better antioxidant activities. In addition, it can be found that ChCl-Xyl extract without the largest TPC/TFC exhibited the strongest antioxidant activities. It was probably due to the high contents of individual phenolics (arbutin, procyanidin B1, (+)-catechin, and (-)epicatechin) in the ChCl-Xyl extracts led to strong antioxidant activities. Ma et al. (2021) reported that the polarity of solvent greatly affected the extraction yields and the compositions of phenolic compounds from Huangshan Gongju (a kind of chrysanthemum in China), thereby influencing the antioxidant activities of the extracts (36). A similar trend can be also observed by Wan et al. (37), who reported that the chemical compositions and antioxidant activity of Chlorella vulgaris extracts were greatly influenced by the extraction solvents. Simultaneously, Rafińska et al. (38) also found that the solvents had obvious effects on the phenolic profile and antioxidant capacity of Moringa oleifera leaves extracts.

Inhibitory Activity Against α-Glucosidase
As we know, α-glucosidase can hydrolyze polysaccharides or starch into glucose or disaccharides, thereby increasing the blood  sugar level. Consequently, suppressing the α-glucosidase activity in the small intestine can effectively control the postprandial blood sugar level. At present, the natural α-glucosidase inhibitors from natural products have attracted increasing interest in the management of high blood sugar (39,40).

In Silico Analysis
In silico analysis was carried out to explain how the main phenolics [arbutin, chlorogenic acid, procyanidin B1, (+)catechin, and (-)-epicatechin] in the CIOPs extracts might interact with α-glucosidase. Figures 4A-F indicates the 3D structures of five main phenolics/acarbose docking with αglucosidase. In addition, all ligands could fit well into the binding pocket of the α-glucosidase homology model (Supplementary Figures 1A-F). Table 5 shows the generated docking information. It can be found that C-score values were ≥ 4 after all of the ligands were docked with the enzyme receptor. The T-score value of arbutin was 6.38, seven Hbonds were formed with 10 amino acid active residues of the α-glucosidase (ASP 215, ASP 352, ARG 213, GLN 182, GLU 277, GLU 411, HIS 112, and HIS 351). The distance of the H-bond ranged from 1.798 to 2.717 Å (Figure 4A; Table 5). Nine H-bonds (average distance of 2.155 Å) interactions with nine catalytic residues (ASP 69, ASP 215, ASP 307, ASP 352, ARG 213, GLU 277, HIS 112, HIS 351, and THR 306) were found in chlorogenic acid, with a relative high T-score of 8.78 (Figure 4B; Table 5). Procyanidin B1 with the lowest Tscore of 6.99 indicated nine H-bonds interactions with 12   411, HIS 280, and TYR 158) of the α-glucosidase ( Figure 4F; Table 5).

Multivariate Analysis
Principal component analysis (PCA) was carried to visualize the impact of solvent on phenolic constituents and the bioactivities of the CIOPs extracts. PC1 61.93% and PC2 16.03% took up 77.96% of the total variances, which indicated that these two principal components could load maximum information of the original data. For the PCA loading plot, the traditional solvents and DESs were, respectively, divided into G1 and G2 ( Figure 5A). With respect to the PCA score plot, the relationship between samples can be represented by the distance between the points, and the relationship between the variables can be reflected by the cosine values (Figure 5B), the small distance or cosine values of two loadings indicated these had a good correlation. Among them, TPC, TFC, arbutin (Arb), vanillin (Van), chlorogenic acid (CGA), procyanidin B1 (PB1), (+)-catechin (CE), epicatechin (EC), and vanillin (Van) were extremely correlated with DPPH • , ABTS +• , FRAP, RP, and CUPRAC. In addition, TPC, TFC, Arb, CGA, PB1, CE, and EC were strongly correlated with α-GIA. The PCA results verified that the extraction solvents had a great impact on the biological activities of the CIOPs extracts.
A heatmap analysis can better visualize the coherent matrix between the phenolic constituents and the bio-activities of the CIOPs extracts ( Figure 5C) (45,46). Arb, CGA, CE, and ECG were strongly correlated with α-GIA. Phenolic compounds from natural products, such as arbutin, chlorogenic acid, (+)-catechin, and (-)-epicatechin, have been previously reported to have strong antioxidant activities and α-GIA, which have been extensively used in the food and pharmacy industries (47). A multivariate analysis also verified the main contributors in the CIOPs extracts to antioxidant activities (Arb, CGA, VAN, CE, and ECG) and α-GIA (Arb, CGA, PB1, CE, and ECG).

CONCLUSION
Clausena indica (Datz.) Oliv fruit pericarps extracts extracted with the traditional solvents and eco-friendly solvents showed significant differences in the phenolic profiles, antioxidant activities, and α-GIA. Phenolic compositions of the CIOPs extracts were identified for the first time using LC-Q-Orbitrap-MS/MS. Arbutin, (-)-epigallocatechin, chlorogenic acid, procyanidin B1, (+)-catechin, and (-)-epicatechin were dominant components in the extracts, especially for the DESs extracts. In addition, ChCl-Xyl and ChCl-Gly extracts showed more excellent antioxidant activities than other solvents extracts. Three DESs extracts with higher TPC and TFC (especially for arbutin, chlorogenic acid, and procyanidin B1) indicated stronger α-GIA. Furthermore, in silico analysis was carried out to determine the α-glucosidase-inhibiting mechanisms of the main phenolics. Multivariate analysis also testified the main contributors in the CIOPs extracts to antioxidant activities and α-GIA. In conclusion, DES can be considered as a promising ecofriendly solvent for the valorization and recovery of high-value compounds from agro-industrial by-products. Furthermore, CIOPs can be applied as a prospective source of active compounds applied in the food and pharmaceutical industries.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.