Enzymatic Preparation and Structure-activity Relationship of Sesaminol.

As a valuable natural antioxidant, sesaminol can be used in food and medicine industries, but it is trace in sesame seeds and oil, and it is feasible to prepare sesaminol from sesaminol triglucoside (STG) which is abundant in defatted sesame cake. Therefore, in order to establish an effective enzymatic preparation method and elucidate the antioxidant structure-activity relationship of sesaminol, a suitable glycosidase for preparing sesaminol from STG were screened, enzymatic hydrolysis was optimized by single-factor test and response surface methodology, and finally, the structure-activity relationship of sesaminol was illustrated by comparative molecular field analysis (CoMFA). These results suggested that βgalactosidase was the optimal glycosidase for enzymatic hydrolysis of STG to prepare sesaminol. Under the optimal conditions of a reaction temperature of 50°C, reaction time of 4.0 h, pH of 5.5, substrate concentration of 1.0 mg/mL, and enzyme dosage of 20 mg/mL, the conversion rate of sesaminol was 98.88±0.67%. Sesaminol displayed excellent antioxidant ability in 2,2-diphenyl-1-picrylhydrazyl (DPPH, IC50 = 0.0011 mg/mL), 2,2'-azinobis-(3-ethyl-benzothiazoline-6-sulfonate) (ABTS, IC50 = 0.0021 mg/mL) radical scavenging activities and Ferric reducing antioxidant power (FRAP, 103.2998 mol/g) compared to other sesaminol derivatives. According to －log (IC50 of DPPH) and －log (IC50 of ABTS), CoMFA models were successfully established based on Q2 >0.5 (QDPPH2 = 0.558, QABTS2 = 0.534). The active site of sesaminol tended to be located on the hydroxyl group of the benzene ring (R1 position). A positive correlation between the bulky and positively charged groups at the 1H, 3H-furo [3, 4-c] furan group, the small, negatively charged groups at the R1 position and the antioxidant activity of sesaminol. This study provides an effective method to prepare sesaminol, reveals the structure-activity relationship of sesaminol and provides theoretical basis to design the novel compound.

improve the usability of tocopherols in organisms and inhibit oxidative modification in DNA 14 . Therefore, sesaminol is considered one of the most valuable natural antioxidants. However, few studies have been carried out on the antioxidant activity of sesaminol compared with that of other natural antioxidants, such as flavonoids and polyphenols 15,16 . Moreover, little attention has been given to relevant studies of sesaminol and its structure-activity relationship. As a naturally occurring trace compound, the content of sesaminol is low in sesame seeds and oil and is difficult to synthesize, which limits relevant study and application 13 .
Some researchers separated sesaminol from roasted sesame oil or refined sesame oil 17,18 , sesaminol content was no more than 2 mg/100 g and 10 mg/100 g in sesame seed and sesame oil respectively. This method was unsatisfactory due to low yield and reduction in the nutrition and edible values of sesame oil. Other researchers have prepared sesaminol from sesamolin by chemical reactions in organic medium 19 . However, residues of organic solvents are a risk in the application of the food and pharmaceutical industry. Some attempts have been made to prepare sesaminol from water-soluble sesaminol triglycoside STG by enzymes. STG can be effectively extracted from defatted cake, which is a rich byproduct of sesame oil industries. However, low yield and long reaction time limit the preparation of sesaminol by enzymes. Peng et al. 20 reported that the yield of sesaminol prepared by enzymatic hydrolysis of STG for 24 h was only 48.9 . Therefore, it is urgent to solve the problem of the time-consuming and low conversion rate of sesaminol by optimizing tool enzyme and hydrolysis conditions. Moreover, previous research has not considered the degradation rule of STG. STG transforms into either sesaminol directly or sesaminol diglycoside SDG or sesaminol monoglycoside SMG indirectly during enzymatic hydrolysis, and SDG or SMG becomes sesaminol via further hydrolysis.
In this study, a suitable enzyme was selected from common glycosidases, and then its enzymatic conditions for the preparation of sesaminol from STG were optimized by a single-factor experimental design and the response surface method RSM . The transformation rules of SMG and SDG were analysed. Moreover, comparative molecular field analysis CoMFA was built to clarify the structure-activity relationship of sesaminol.

Materials
Sesame seeds were obtained from a local supermarket. STG, SDG, SMG and sesaminol were prepared in laboratory. Sesamin, sesamolin, sesamol, asarinin, paulownin, cellulase 0.7 U/mg , β-glucanase 2.0 U/mg , β-glucosidase 2.0 U/ mg , β-galactosidase 8.0 U/mg , 2,2 -azinobis-3-ethylbenzothiazoline-6-sulfonate ABTS , 2,2-diphenyl-  21 with some modifications. Defatted sesame cake was extracted with 75 ethanol v/v at a ratio of 1:8 w/v and stirred at 20 for 8 h 3 times. The supernatant was merged, concentrated and lyophilized to obtain a crude extract rich in STG. 2 g crude extract was dissolved in distilled water 10 mL , and then the solution was added to an open column 5.0 50 cm package with polyamide 100-200 mesh . Distilled water as the mobile phase eluted the column and flowing at 1.0 mL/min. The eluate was collected with an automatic DBS 100 collector Huxi Analysis Instrument Factory Co., Ltd., Shanghai, China in 10 mL portions and tested with HPLC and LC-ESI/MS. According to the determined results, the fraction was collected and freeze-dried to achieve purified STG.
The preparation methods of SDG, SMG and sesaminol were developed according to the results in this study. SDG and SMG were prepared from STG by β-galactosidase, and their enzymatic hydrolysis conditions were the same except for the pH. Two amounts of 50 mg STG were dissolved in a 25 mL of 0.05 mol/L citrate buffers at a pH of 3.0 and a 0.05 mol/L phosphoric acid buffer at a pH of 7.0. SDG and SMG were prepared by enzymatic hydrolysis under the same reaction conditions of 50 , an enzyme dosage of 20 mg/mL and a reaction time of 3.0 h, and then the reaction solutions were boiled for 5 min and centrifuged at 12000 rpm for 15 min. The supernatants were collected and loaded onto a Sephadex LH-20 column 2.0 50 cm . Distilled water was used as the eluent, and the flow rate was 1.0 mL/min. The eluate was assembled in 10 mL increments with an automatic collector, and the fractions containing SDG/SMG were tested, gathered and lyophilized.
S e s a m i n o l w a s a l s o p r e p a r e d f r o m S T G b y β-galactosidase, and the enzymatic hydrolysis conditions were the same as the enzymatic hydrolysis conditions of SDG/SMG except that the pH was 5.5. After the reaction was completed, the reaction mixture was centrifuged, 50 mL ultrapure water was added to the precipitate, and then the mixture was shaken for 2 min and centrifuged again at 12000 rpm for 15 min. Then, 50 mL ethanol was added to the precipitate, the mixture solution was shaken for 2 min and separated by centrifugation, and the supernatant was collected. This step was repeated 3 times. All the supernatants were combined and enriched by rotary evaporation. After vacuum freeze-drying for 36 h, sesaminol was tested and collected.

Determination and identi cation of STG, SDG, SMG
and sesaminol The analysis of STG, SDG, SMG and sesaminol was performed on a Thermo Fisher Scientific Ultimate 3000 series HPLC system Thermo Fisher Scientific, Waltham, MA, USA , a UV detector and an Agilent 5 TC-C18 2 250 mm 4.6 mm, 5 μm column Agilent, CA, USA . The temperature of column oven is set at 30 . The mobile phase are methanol solvent A and water solvent B , the gradient system is as follows: 0-10 min 1 mL/min, 40 A ; 15-30 min 0.8 mL/min, 40 A ; 40-50 min 1 mL/min, 80 A , 55 min 1 mL/min, 80 A and 60 min 1 mL/min, 40 A . The absorbance was monitored at 287 nm for STG, SDG and SMG and 293 nm for sesaminol. The injection volume was 5 μL.
The molecular weights of STG, SDG, SMG and sesaminol were identified by LC-ESI/MS. LC-ESI/MS analysis was carried out by using an Ultimate 3000-Q Exactive-Orbitrap Thermo, MA, USA and a positive model.

Selection of optimum enzyme for preparation of sesa-
minol and qualitative analysis of enzymatic hydrolysis products According to related literature 22,23 , cellulase, β-glucanase, β-glucosidase and β-galactosidase were tested in the experiment, and their optimum conditions are shown in Table 1. 10 mg STG was mixed with 10 mL buffer solution. Glycosidase was added to the reaction solution at the same amount of enzyme activity. The reaction solution was shaken in a water bath for 8.0 h at the optimal temperature. The sample was boiled for 5 min followed by centrifugation at 10,000 g for 20 min at 4 and filtered through a 0.22 μm microfiltration membrane. HPLC qualitatively analysed the composition of the hydrolysate. The HPLC method was in line with the method described in 2.2. A suitable glycosidase was selected based on the retention rate of STG calculated by 1 and hydrolysate yields calculated by 2 . In equations 1 and 2 , Y 1 represents the retention rate of STG, Y 2 represents the conversion rate of enzymatic hydrolysis products, m 1 represents the weight of STG in the reaction solution at 0 h, m 2 represents the weight of STG in the reaction solution at the end of the reaction, m 3 represents the weight of sesaminol in the reaction solution at the end of the reaction, M 1 represents the molecular weight of STG, and M 2 represents the molecular weight of enzymatic hydrolysis products.

Optimization of the enzymatic hydrolysis conditions
The optimal conditions for the preparation of sesaminol from STG were determined by a single-factor test and RSM 24,25 . Reaction time, temperature, pH, substrate concentration and enzyme concentration were selected as variables in the single-factor test of the experimental design. The levels of these five factors were set as follows: reaction times of 0.5-12.0 h; temperatures of 20-60 ; pH of 3.0-7.0; substrate concentration of 1.0-3.0 mg/mL; and enzyme concentration of 10-30 mg.
According to the results of the single-factor test, pH x 1 , temperature x 2 and reaction time x 3 were chosen for optimization by RSM. The levels of independent variables are shown in Table 2.

DPPH free radical scavenging activity assays
The abilities of 10 compounds to scavenge DPPH free radicals were assessed according to the method of Yeo et al. 26 . First, 100 μL samples of different concentrations were added to a 96-well microplate, and 100 μL DPPH in 80 ethanol 0.2 mmol/L was prepared in each well and mixed through the flat plate oscillation period. The mixture was incubated at room temperature for 30 min in the dark. The absorbance was measured using an Infinite M Nano absorbance plate reader Tecan Group Ltd., Zürich, Switzerland at 517 nm. The DPPH radical scavenging capacity was calculated using the following equation: where A sample is the absorbance of the sample and DPPH, Table 1 The optimum conditions of the enzymatic hydrolysis of STG by four enzymes. A sample background is the absorbance of the sample background sample and ethanol instead of DPPH , and A control is the absorbance of the control DPPH and 100 μL of ethanol instead of sample .

ABTS free radical scavenging activity assays
The method of Re et al. was used to determine ABTS assay 27 . 20 μL sample solution and 200 μL of working solution were added to the 96-well plate. The mixture was then mixed on the plate vibrator and reacted for 6 min in the dark at room temperature. An Infinite M Nano absorbance plate reader was used to observe the absorbance of the sample solution at 734 nm. The sample solution with a certain gradient concentration was used to determine the ABTS scavenging activity. The IC 50 value was calculated by SPSS analysis.
where A control is the absorbance of methanol and ABTS radicals and A sample is the absorbance of ABTS radicals in the sample.

Ferric reducing antioxidant power FRAP
The FRAP capacity was estimated following Benzie et al. 28 as described with slight modification. Fresh FRAP reagent 150 μL was pipetted into a well of a 96-well plate and mixed with 30 μL of 1 mg/mL sample solution. The absorbance of the blank sample 150 μL FRAP reagent and 30 μL distilled water was recorded after incubation for 30 min at 37 . A standard curve was plotted by measuring the absorbance of 0-2 mmol FeSO 4 at 593 nm. The results are presented in μmol FeEq/g of sample, according to a standard curve of FeSO 4 .

3D-QSAR of sesaminol
3D-QSAR helps to gain insight into the relationship between specific bioactivity and molecular structure. As a classic 3D-QSAR technique, CoMFA illuminates the effect of steric and electrostatic fields of a set of aligned compounds on their biological activity by associating these 3D fields with the corresponding experimental activities of ligands interacting with a common target receptor 29 . CoMFA has been applied to demonstrate the structure-activity relationships 30 .
Sesaminol, STG, SDG, SMG, asarinin, sesamin, sesamolin, paulownin, methylophiopogon flavone A and methylophiopogonanone A were utilized to construct the CoMFA model by SYBYL-X 2.1.1 software. The molecular structures of the 10 compounds were drawn with a sketch module and energy minimization using a conjugate-gradient minimizer. The MMFF94 method was used to calculate the partial atomic charges of all the compounds, and then the Powell conjugated gradient algorithm method was utilized to optimize their geometry. Moreover, 0.0005 kcal/ mol Å was selected as the energy convergence criterion, and the maximum iteration was set as 10,000 to construct a steady conformation 7 .
To construct the CoMFA model based on the pIC 50 of DPPH, sesaminol was chosen as a template molecule in the molecule alignment stage because it is the strongest DPPH scavenging compound; other compounds were aligned on the framework structure of 1,3-benzodioxole in the tem- plate molecule. To build the CoMFA model based on the pIC 50 of ABTS, methylophiopogonanone A was selected as the template molecule in the molecule alignment stage because it is the strongest ABTS scavenging compound Fig. 4 a , another operation is to refer to the method of the CoMFA model based on the pIC 50 of DPPH.
The linear correlation between the biological activity and the CoMFA model was assessed by Partial least squares PLS analysis. In this study, leave-one-out LOO cross-validation procedures were used to determine the optimal number of components and cross-validation correlation coefficient Q 2 , while non-cross-validated analysis was used to calculate the non-cross-validated correlation coefficient R 2 , standard error of estimate ESS , and F-value 31 . The CoMFA model was regarded as acceptable when Q 2 and R 2 were greater than 0.5 and 0.6, respectively 30 .

Statistical analysis
All the tests were performed in triplicate. The statistical significance was determined by the level of p 0.05. There was no significant difference between the values with the same letters in one curve. SAS software Version 8.5, North Carolina, USA and SPSS software version 20.0; SPSS, Chicago, IL, USA were applied to the experimental design and statistical analysis.

Preparation and identification of STG, SDG, SMG
and sesaminol STG, SDG, SMG and sesaminol were successfully obtained by column chromatography and liquid-liquid extraction. Their purity was determined by HPLC to be 98. 17 , 98.19 , 98.53 and 99.35 , respectively. The contents of the solution mixture were also determined by HPLC, including the levels of STG, SDG, SMG and sesaminol. The HPLC chromatogram is shown in Fig. 1A 24,32 . The results showed that STG could be obtained from sesame cake. STG could be a raw material that can be used for the preparation of SMG, SDG and sesaminol. Otherwise, the four compounds obtained could be used in the following test as standard and raw material.

Effect of reaction time on the conversion rate of sesaminol
The retention of STG and the conversion of its hydrolysis products are shown in Fig. 2B. A dramatic change in the content of each component occurred in the first 1.0 h during enzymatic hydrolysis. The STG rapidly disappeared after 0.5 h. The SDG conversion rate reached a peak at 84.69 at 0.25 h and then decreased to 0 at 1.0 h. The conversion rate of SMG reached a maximum of 19.91 at 0.5 h and subsequently gradually disappeared at 3.0 h. The conversion rate of sesaminol significantly increased to 84.15 within 1.0 h and continued to increase to a high point at 4.0 h. Therefore, the reaction path from STG to sesaminol included two intermediate products, SDG and SMG, glucose. The study results were consistent with those of Nair et al. 36 . The conversion rate of sesaminol did not significantly increase over an 8 h period. Thus, 4.0 h was selected as the proper reaction time. The reaction takes less time than those using β-glucosidase and microbial fermentation to prepared sesaminol 13, 20, 21 .

Effect of reaction temperature on the conversion rate of sesaminol
The effect of temperature 20-60 on the conversion rate of sesaminol from STG 1.5 mg/mL by β-galactosidase at pH 5.0 after incubation for 4 h is presented in Fig. 2C. The STG and SDG disappeared at the end of the reaction. The lowest conversion rate of sesaminol was 29.4 at 20 . At the same temperature, the SMG conversion rate reached 69.0 . As the reaction temperature gradually in-creased to 50 , the conversion rate of sesaminol significantly increased to a maximum of 95. 3 p 0.05 , while the conversion rate of SMG decreased to a minimum of 3.0 . The conversion rate of sesaminol significantly decreased as the temperature exceeded 60 , which was related to the enzyme denaturation caused by high tem-perature. This result was consistent with the report that the optimum temperature of β-galactosidase from Aspergillus oryzae was in the range of 45 to 55 33 . Therefore, 50 was selected as the optimum reaction temperature. 3.5 Effect of enzyme dosage on the conversion rate of sesaminol The influence of the enzyme dosage on the conversion rate of sesaminol is shown in Fig. 2D. When the enzyme dosage was increased from 10 mg/mL to 20 mg/mL, the STG and SDG disappeared, the conversion rate of SMG ranged from 21.3 to 2.7 , and the conversion rate of sesaminol significantly increased to 95.4 . However, when the enzyme dosage increased from 20 mg/mL to 30 mg/mL, the conversion rate of sesaminol did not significantly change. Considering the cost of the enzyme, the enzyme dosage was set at 20.0 mg/mL. 3.6 Effect of substrate concentration on the conversion rate of sesaminol Substrate concentration is one of the main affecting factors for enzymatic hydrolysis reaction 37 . As shown in Fig. 2E, when the substrate concentration was 1.0 mg/mL, STG was completely enzymatically hydrolysed, and the conversion rate of sesaminol was up to 95.8 with a small amount of residual SMG. As the substrate concentration increased, the conversion rate of sesaminol significantly decreased, and the conversion rates of the SDG and SMG slightly increased. Overall, these results indicate that 1.0 mg/mL was the optimal substrate concentration.
3.7 Effect of reaction pH on the conversion rate of sesaminol As shown in Fig. 2F, the conversion rate of sesaminol ranged from 7.2 to 95.9 as the pH increased from 3.0 to 5.0. There was no significant difference between the conversion rate of sesaminol at pH 5.0 and 6.0 p 0.05 . However, the conversion rate of sesaminol remarkably decreased at pH 7.0 p 0.05 . The highest conversion rates of SMG and SDG were 74.2 at pH 7.0 and 79.6 at pH 3.0. Therefore, it was suitable for enzymatic hydrolysis of STG into sesaminol at pH 5.0-6.0. The optimal pH for sesaminol production with β-galactosidase was in agreement with the optimal pH range of 5.0-6.0 for ginsenoside Rg 3 hydrolysis 38 .

Response surface experimental results
Three factors and three levels Box-Behnken design in-cluding 15 trials was used to assess the effect of hydrolysis factors, including pH x 1 , temperature x 2 and reaction time x 3 , on the conversion rate of sesaminol; the results of these trials were used to optimize these factors. The experimental results are shown in Table 2. The experimental data were analysed by multiple regression analysis, and a second-order polynomial Eq. 5 was used to describe the relationship between the independent variables and survey the conversion rate of sesaminol. The second-order equation in coded format was presented: In Table 3 above, we can see the ANOVA results. The Model F-value was 362.8777. A p-value lower than 0.0001 indicates that the model was in line with the actual data. The determination coefficient R 2 was 0.9985, which suggested that only 0.15 of the total variation was not fitted with the model. The Lack of Fit was 0.082943 0.05 , indicating that the lack of fit was not significant. The p-values of the linear coefficient x 1 , x 2 and quadratic coefficients x 1 2 , x 2 2 , x 3 2 were less than 0.05, implying the significant effects of these factors on the conversion rate.

Response surface three-dimensional plots
To illustrate the intercorrelations between the three above factors and the conversion rate of sesaminol, a threedimensional response surface was built Fig. 3 . When the effect of two factors was plotted, the other factor was  Fig. 3 The effect of pH and temperature interaction on the conversion rate of sesaminol at different levels.
fixed. In the investigation range, pH and temperature were significant factors, so the curve of the response surface and contour map of the conversion rate of sesaminol at fixed reaction times were observed. Figures 3A and 3B shows that as the reaction temperature was fixed, and that the conversion rate of sesaminol gradually increased and then tended to be stable with increasing pH. The enzyme activity remained stable because the pH was near the optimum pH range for the enzyme. This result was in line with the single-factor test studies. When pH was the fixed variable, the conversion rate of sesaminol increased when the reaction temperature was raised from 45 to 50 and then decreased when the reaction temperature was raised from 50 to 55 because high temperature leads to enzyme denaturation, so enzyme activity decreased 38,39 . The trend of sesaminol conversion at 4.0 h and 4.5 h was similar to that at 3.5 h Figs. 3C-3F .

Verification of optimization
The optimum conditions predicted by SAS software were as follows: the pH value was 5.67, the reaction temperature was 49.66 , the reaction time was 3.91 h, the substrate concentration was 1.0 mg/mL, and the enzyme dosage was 20 mg/mL. Under these conditions, the conversion rate of sesaminol was 100.42 0.52 . For convenience of operation, the enzymatic hydrolysis process parameters were set to a pH of 5.5, temperature of 50 , enzymatic hydrolysis time of 4.0 h, substrate concentration of 1.0 mg/mL, and an enzyme dosage of 20 mg/mL. 3 parallel tests were conducted, and the conversion rate of sesaminol was 98.88 0.67 . The result conformed to the predicted data. Therefore, the process parameters of the enzymatic hydrolysis of STG into sesaminol based on the response surface method are accurate and reliable. Compared with a previous report 17,18,20 , the performance of the enzymatic preparation of sesaminol from STG improved due to the selection of the enzyme and the optimization of the hydrolysis conditions; the conversion rate of sesaminol rose from 48.90 to 98.88 , the reaction time decreased to 1/6 of previous reports 20 . The enzymatic hydrolysis technology not only reduced the nutrition and edible values of sesame oil, but also could improve the utilization rate and added value of sesame cake, which is a rich by-product of sesame oil industries 17,18 .

3D-QSAR of sesaminol and its derivative
The antioxidant ability of sesaminol, STG, SDG, SMG, asarinin, sesamin, sesamolin, paulownin, methylophiopogonanone A and methylophiopogonone A was measured. The results of DPPH and ABTS radical scavenging ability tests and FRPA are shown in Table 4. Sesaminol exhibited excellent antioxidant activity in DPPH IC 50 0.0011 mg/mL , ABTS IC 50 0.0021 mg/mL radical scavenging activities and FRAP 103.2998 mol/g among sesaminol derivatives, which was also in accordance with a previous study 40 .
To elucidate the effects of sterics and electrostatic fields on the antioxidant activity of sesaminol, CoMFA models based on log IC 50 of DPPH and log IC 50 of ABTS were established Table 5 . As seen in Table 6, the CoMFA model for log IC 50 of DPPH and log IC 50 of ABTS was successfully established based on Q 2 0.5 Q DPPH 2 0.558, Q ABTS 2 0.534 . These models had good R 2 values R 2 1.000 with low standard errors of estimate 0.05 and a high F value F DPPH 171737.028, F ABTS 8955.818 . The CoMFA based on the DPPH scavenging activity has the highest Q 2 ; moreover, it was reliable to evaluate the antioxidant activity of fat-soluble sesaminol by DPPH scavenging assay. Therefore, this paper only analysed and discussed the CoMFA model based on the DPPH activity results.
In Figs. 4A-4B, there are large green areas in the 1H, 3H-furo 3,4-c furan group and C 2 and C 6 positions of the substituent R 2 , indicating that the introduction of substituents with larger groups here is beneficial to the improvement of activity. This observation also agrees with the experimental result, as the larger 5,7-dihydroxy-6,8dimethyl-4H-1-benzopyran-4-one group at the R 2 position of methylophiopogonone A presented higher DPPH inhibi- tory activity than sesamin. There is a yellow region at the R 1 position, indicating that smaller groups would enhance the antioxidant activity. The smaller -OH at the R 1 position of sesaminol presented better DPPH inhibitory activity than did SMG, SDG and STG. This conclusion obeyed the scavenging mechanism of the DPPH radical; that is, the larger the steric hindrance of the compound was, the more challenging it was for to come into contact with the un-paired electron of the intermediate N atom in the DPPH radical through the spatial barrier of the three benzene rings of DPPH 27 . In Figs. 4C and 4D, the red region appears at the R 1 position, indicating that the introduction of more electronegative groups here is conducive to the improvement of the compound activity. This finding is in line with the experimental data, as the negatively charged -OH group at the R 1 Table 5 The antioxidant activity and molecular formula of sesaminol and its derivatives. position of sesaminol resulted in higher DPPH inhibitory activity than that of sesamin, whose R 1 position was occupied by the less negatively charged -H. A large blue area appeared in the 1H, 3H-furo 3,4-c furan group of substituent R 2 , indicating that the introduction of positively charged groups in the region was beneficial to the increase in the antioxidant activity of the compound. This increase in activity is because the introduction of a positive charge in this region can increase the rejection of DPPH radicals and increase the probability of DPPH radicals coming into contact with the compound from R 1 -OH to improve the scavenging ability of DPPH radicals. Combined with the results of the steric field and electrostatic field analysis, the data indicated that -OH at the sesaminol R 1 position was the active centre of DPPH free radical scavenging by sesaminol, and the result was in good agreement with a previous report that found that -OH increased the activity of flavonoids 16 .

Conclusions
This study investigated the preparation method and structure-activity relationship of sesaminol. The results showed that β-galactosidase was a suitable enzyme to prepare sesaminol from STG. The optimal enzymatic hydrolysis conditions were established. In addition, the transformation rules of sesaminol, SMG and SDG were investi- The contribution of steric field 0.650 0.720 The contribution of electrostatic field 0.350 0.280 NC 1 represents optimal number of principal components Enzymatic Preparation and Antioxidant Activity of Sesaminol gated to further reveal the mechanism of enzymatic hydrolysis. Furthermore, compared to other sesaminol derivatives, sesaminol has excellent scavenging capacity for DPPH, ABTS and FRAP. CoMFA models were developed according to log IC 50 of DPPH and log IC 50 of ABTS . The active site of sesaminol tended to be located on the hydroxyl group of the benzene ring R 1 position . This study not only provides a short-time and effective method to prepare sesaminol, reveals the structure-activity relationship of sesaminol, but also provides theoretical basis to design the novel compound. The results are expected to provide more evidence for the utility of sesame as an antioxidant functional food.

Author Contributions
Gao