Hydrolyzation of mogrosides: Immobilized β‐glucosidase for mogrosides deglycosylation from Lo Han Kuo

Abstract An immobilized enzyme system for bioconversion of Lo Han Kuo (LHK) mogrosides was established. β‐Glucosidase which was covalently immobilized onto the glass spheres exhibited a significant bioconversion efficiency from pNPG to pnitrophenol over other carriers. Optimum operational pH and temperature were determined to be pH 4 and 30°C. Results of storage stability test demonstrated that the glass sphere enzyme immobilization system was capable of sustaining more than 80% residual activity until 50 days, and operation reusability was confirmed for at least 10 cycles. The Michaelis constant (K m) of the system was determined to be 0.33 mM. The kinetic parameters, rate constant (K) at which Mogrosides conversion was determined, the τ 50 in which 50% of mogroside V deglycosylation/mogroside IIIE production was reached, and the τ complete of complete mogroside V deglycosylation/mogroside IIIE production, were 0.044/0.017 min−1, 15.6/41.1 min, and 60/120 min, respectively. Formation of the intermediates contributed to the kinetic differences between mogroside V deglycosylation and mogroside IIIE formation.

MG-III reduced pulmonary fibrosis and may have therapeutic potential for treating fibrosis (Tao et al., 2017).
Several studies have been carried out to enhance the bioactive form of mogrosides, and the most common conversion method used is acid hydrolysis . LHK extracts were treated by hydrochloric acid, generating different type of aglycones (Chen et al., 2011). Although acid hydrolysis method can easily produce mogrosides at low cost, several potential drawbacks need to be overcome, such as poor specificity, low yield, and unexpected byproducts (Girisuta, Danon, Manurung, Janssen, & Heeres, 2008).
Enzymatic method has been used to producing mogrosides. For example, cellulase is able to convert mogroside V into mogroside III and mogroside II E (Murata et al., 2010); maltase (an α-glucosidase) converts mogroside V into several types of mogroside (Takemoto, Arihara, Nakajima, & Okuhira, 1983). Compared to acid hydrolysis, enzymatic conversion method exhibits efficiency and specificity benefits. However, the cost of enzymatic conversion method is relatively high. Enzyme immobilization techniques provide several advantages, including easy separation of the enzyme from the reaction mixture, cost reduction by improving the reusability, continuous processing, and improvement of stability (Dervakos & Webb, 1991). In food industry, immobilized lactase catalytic system provides lactose-free milk for patients with lactose intolerance; hydrolysis of glycosylated-isoflavones of black soymilk using immobilized β-glucosidase was also reported (Chen, Lo, Su, Chou, & Cheng, 2012;Chen et al., 2013;Ko et al., 2018).
In this study, we evaluated the feasibility of mogrosides conversion using an immobilized β-glucosidase enzyme system. β-Glucosidase was immobilized on three different carriers (glass spheres, nylon pellets, cellulose beads) for mogrosides deglycosylation, and p-nitrophenyl β-D-glucuronide (pNPG) was employed as a chromogenic substrate to determine β-glucosidase activity. The carrier possessing the most efficient catalytic behavior among these three solid carriers was chosen and applied to mogrosides deglycosylation for Lo Han Kuo extract solution. The kinetic parameters of the chosen catalytic system for mogrosides deglycosylation were further determined.

| Materials
Lo Han Kuo extract powder (Mogroside V ≥25.0%) was purchased from Huitong bio-technology company (Guangxi, China). β-Glucosidase from Aspergillus niger was purchased from Sigma-Aldrich (St. Louis, MO). 4-Nitrophenyl β-D-glucuronide (pNPG) purchased from Alfa Aesar (Ward Hill, MA) was used as the substrate to determine the catalytic activity of immobilized enzyme system. Glass spheres, nylon pellets, and cellulose beads were separately utilized as solid carriers for enzyme immobilization. Glass spheres were purchased from Sigma-Aldrich with 9-13 μm particle size. Nylon pellets were provided by Keen Corp Company (Tainan, Taiwan). Porous cellulose beads were synthesized as previously described (Chen & Tsao, 1976). The particle size of nylon pellets and cellulose beads was about 1 mm.

Nylon pellets
Nylon pellets modification procedure has been reported previously (Diano et al., 2008).
In brief, a total volume of 40 cm 3 nylon pellets was immersed in 30 ml of dimethyl sulfate reagent for 4 min at 100°C. A cleaning step was followed immediately by cold methanol, and then O-alkylated nylon pellets were immersed in 50 ml of 10% hexamethylenediamine (HMDA) aqueous solution at room temperature for 90 min.

Cellulose beads
An organic solution (acetone/DMSO = 6:4) was used to dissolve cellulose acetate (6.25 g). The solution was then homogenized and added drop-wisely into cold water by syringe, so that cellulose beads will precipitate and form (Chen & Tsao, 1976). Finally, the beads were washed with distilled water and dried overnight at room temperature to attain a constant weight. To enhance the hydroxyl groups on cellulose beads' surface, a pretreatment was carried out in 0.2 M NaOH for 90 min.

Glass spheres
First, a total volume of 40 cm 3 glass microspheres was incubated with a concentration of 10% nitric acid (HNO 3 ) at 90°C for 1 hr. Next, microspheres were continuously washed with distilled water and were subsequently immersed in a 10% 3-aminopropyltriethoxysilane (APES) aqueous solution (50 ml, pH 3.4) at 70°C for another 3 hr.
Treated aminopropyl-glass was cleaned with distilled water and dried overnight in an oven at 80°C. The procedure has been reported previously (Chen et al., 2012).

Enzyme immobilization
After preparation and activation, each activated carrier was treated with 2.5% glutaraldehyde aqueous solution. Cellulose beads were subjected to submersion at 100°C for 30 min, while nylon pellets and glass spheres were both treated under a gentler condition (room temperature, 60 min) (Chen & Tsao, 1977). The carrier was washed with double-distilled water at room temperature, and each type of carrier was incubated separately with 20 mg of β-glucosidase (dissolved in 50 ml of 0.1 M phosphate buffer at pH 6) at 4°C for 16 hr.
Uncross-linked enzymes were removed by the washing with 0.1 M phosphate buffer, and the enzyme-linked carriers were stored at 4°C until use. The washing solutions were collected to detect the amount of uncrossed-linked enzyme, β-glucosidase. The amount of β-glucosidase being successfully immobilized onto carrier was calculated as the difference between the initial and the uncrosslinked β-glucosidase. Bradford dye-binding method was adopted to determine the concentration of β-glucosidase (Bradford, 1976). Five milligrams of pNPG was dissolved in 0.1 M phosphate buffer (50 ml, pH 4). Each type of enzyme immobilized beads was incubated in 50 ml of pNPG solution in a batch reactor at 30 and 50°C, respectively. A volume of 800 μl of pNPG solution was withdrawn, and 1 ml of 2 M sodium carbonate was added to terminate the reaction. The absorbance of pNPG was read at 425 nm and recorded at regular time intervals. All of the experimental studies were performed in triplicates.

Morphology characterization
The surface morphologies of glass spheres with and without βglucosidase immobilization were investigated using scanning electron microscopy (SEM) at an accelerating voltage of 10 kV (Model: JSM-6300, JEOL; Tokyo, Japan). Prior to measurement, samples were coated with a thin layer of gold.

Verification of enzyme immobilization
The successful enzyme immobilization was confirmed by X-ray photoelectron spectroscopy (XPS), also known as electron spectroscopy for chemical analysis (ESCA), a widely used surface analysis technique. 16 XPS measurements were carried out with a ESCA system (VG MICROTECH, MT-500, British) using A1Ka radiation (hm = 1,486.6 eV); Thermo Scientific Avantage Data System software (Thermo Fisher Scientific Inc., Herts, UK) was utilized to perform curve fitting and calculate the atomic concentrations.

Mogrosides hydrolyzation, purification, and analysis
Hydrolyzation of Lo Han Kuo extract aqueous solution was performed with the free and immobilized enzymes. The hydrolyzation was carried out at pH 4.0, 30 or 50°C. Five hundred microliter solution of sample were withdrawn at regular intervals, and an equal volumes of methanol (100%) were added (into the sample) to terminate the reaction. To purify mogrosides from the above mixture solution, a separation was performed on a reversed-phase C-18 solid phaseextraction cartridges (500 mg/3 ml, Chrome expert, Sacramento, CA, USA) to elute out (under 45%-80% of methanol) the impurities and collect different mogrosides in certain region. The collected sample was concentrated by vacuum evaporator and resolve in the 1 ml methanol which contained 40 ppm hydrocortisone (internal standard) solution.
To determine the kinetic parameters, the glucose detection in the reaction solution was accomplished by Megazyme D-Glucose (glucose oxidase/peroxidase; GOPOD) assay kit (Wicklow, Ireland).
Addition of 1 ml of GOPOD Reagent to 0.1 ml of sample solution followed by an incubation at 45°C for 20 min. The absorbance of each solution was then read at 510 nm.

Kinetic parameters
The Michaelis constant (K m ) and maximal velocity (V max ) for free and immobilized β-glucosidase were determined at 30°C for 10 min. The Michaelis constant (K m ) reflects the effective characteristics of the enzyme and the affinity between enzyme and substrate, which was determined using the Michaelis-Menten equation.
Here, [S] represents for various concentration of the substrate, pNPG. Maximal velocity (V max ) stands for the maximum rate achieved by the enzymatic system, at saturating substrate concentration (The intrinsic characteristic of the enzyme, at which all the enzyme active sites are all bound to substrate.) The Michaelis constant (K m ) is the substrate concentration where the reaction rate is half of V max .

HPLC-UV analysis
Analysis Resolution of mogrosides was performed using Luna C18 (2) 100A column (5 μm, 250 × 4.6 mm) and maintained at room temperature. The column was attached to a HPLC workstation containing two LDC pumps and a LDC analytical mixer. The injection volume was 20 μl with an elution rate at 0.6 ml/min. Detection was monitored at 210 nm, and the acquired data were analyzed by SISC chromatography data system (SISC, Taiwan).

Statistical analysis
All experiments, including carrier comparison, evaluation of reaction condition, absorbance measurements, and HPLC analysis, were performed in triplicates. For fraction analyses, data are presented as the means ± standard deviations. Differences among the variants were analyzed using one-way analysis of variance with the Dunnett test; p < 0.01 was considered statistically significant. We used a twosample t test (F < 0.001) to evaluate the effect of focus on bioactivity.

| Selection of carrier for immobilized enzyme system
To select a reliable immobilized enzyme system for mogrosides conversion, the immobilized enzyme content was calculated. The immobilized enzyme content of nylon pellets, cellulose beads, and glass spheres was 16.60, 3.52, and 18.00 mg, giving protein binding ratio at 83.0%, 17.6%, and 90.0%, respectively (Table 1). There was no significant difference in the amount of immobilized enzyme between nylon pellets and glass spheres. Figure 1 summarized  50°C) where pNPG was used as a substrate to measure enzyme activity. Results indicated that the immobilized β-glucosidase on the glass sphere exhibited higher relative reactivity (30°C, 60 min, 93%; 50°C, 60 min, 99%) than those on the nylon pellets (30°C, 60 min, 24%; 50°C, 60 min, 29%) and cellulose beads (30°C, 60 min, 20%; 50°C, 60 min, 28%). The saturated conversion of pNPG to pNP by immobilized glass spheres enzyme system can be achieved within 30 min, and its catalytic effectiveness was close to the suspension enzyme system.
The particle size is described as an important factor to catalytic activity for the immobilized enzyme system. System with smaller particles possesses higher catalytic activity due to larger specific area or minimized diffusion limitation (Alagöz, Tükel, & Yildirim, 2016;Mendes et al., 2012). The particle size of glass spheres used in this study was 9-13 μm, which is smaller than the nylon pellets and cellulose beads (about 1 mm). Therefore, glass spheres were selected as the immobilized carrier for the rest of study.

| Morphology of glass spheres
The surface morphologies of glass spheres with and without β- the observation of previous study (Chen et al., 2012;Ko et al., 2018).
The granulated areas on the surface of glass spheres represented the regions of β-glucosidase attachment (as pointed by arrows) providing evidence that enzyme was successfully immobilized onto the glass spheres.

| Chemical modification of β-glucosidase onto glass spheres
β-Glucosidase immobilized on the glass spheres was applied to hydrolyzation of mogrosides. When β-glucosidase was  Figure 3) (Longo, Vasapollo, Guascito, & Malitesta, 2006). The peak at 399.8 eV in the N 1s ESCA spectrum can be verified that the amino acid of β-glucosidase cross-linked with glutaraldehyde through covalent binding (Chen et al., 2013). As a result, the successful enzyme immobilization onto glass spheres was confirmed.

| Optimal pH and temperature for β-glucosidase bioconversion
The enzyme activity is significantly affected by the reaction temperature and pH. β-Glucosidase was stable in the acidic environment, and the optimum range of pH was between 4 and 5 (Kita et al., 1991; Vazquez-Ortega, Alcaraz-Fructuoso, Rojas-Contreras, López-Miranda, & Fernandez-Lafuente, 2018). In this study, the efficiency of the immobilized enzyme system was evaluated and compared with the one from suspension enzyme at different pHs and temperatures. First of all, the enzyme activity was evaluated at different pH conditions at 30°C. It was noted in Figure 4a that the highest catalytic activity of both free and immobilized enzyme systems was reached at pH 4.
Subsequently, evaluation of temperature on β-glucosidase activity was carried out at pH 4. Figure 4b showed that β-glucosidase exhibited the highest efficiency both in the free and immobilized enzyme systems at

| Determination of kinetic parameters for the enzyme systems
In this study, p NPG was used to evaluate the hydrolyzation efficiency and to calculate enzyme kinetics. The kinetic parameters, K m and V max , were 0.60 mM and 0.58 μmol/min for suspension enzyme and 0.56 mM and 0.35 μmol/min for immobilized enzyme system

| Bioconversion of mogrosides
The hydrolyzation progress of mogrosides in luo han kuo fruit extract by free and immobilized β-glucosidase systems can be seen in   (Chiu et al., 2013;Suzuki et al., 2005). Therefore, efforts have been made for mogrosides bioconversion. In the previous studies, microorganisms were used as bioconversion host for biotransformation of mogrosides (Chiu et al., 2013;Wang et al., 2015). Chiu et al. reported that Exg1 is a major enzyme for the initiation of mogroside V conversion and deletion of KRE6 gene of S. cerevisiae facilitated mogroside IIIE production (Chiu et al., 2013). Wang et al. also reported a Saccharomyces cerevisiae mutant which can hyperproduction of β-glucanase resulting in the enhanced bioconversion of mogrosides (Wang et al., 2015). However, it usually took more than 24 hr to reach complete mogrosides bioconversion. For our immobilized enzyme system, it took only 120 min to reach complete mogrosides bioconversion (Table 2). Moreover, the recovery and purification process is relatively easier for the samples from immobilized enzyme system than those from the fermentation ones.

| Reusability and storage stability
The reusable features of immobilized enzyme system enable the mogroside bioconversion for mass production. In this study, TA B L E 2 Kinetic parameters characterizing the suspension and the immobilized β-glucosidase system F I G U R E 8 Storage stability of immobilized β-glucosidase from 1 to 50 days the immobilized enzyme system can be used consecutively for at least 10 cycles which still kept 90% of its original bioconversion activity (data not shown). A continuous bioconversion system can be adopted through bioreactor design (Ko et al., 2018).
Although the conversion rate of suspension enzyme system is faster than that of immobilized enzyme system, the immobi-  (Suzuki et al., 2005). Besides, from an in vivo study, mogroside IIIE, a metabolite of Mogrosides, was widely distributed in the organ of treated rat (Chiu et al., 2013). Nevertheless, since the molecule structure is complex, it is difficult to synthesis or produce a specific type of mogrosides.
The storage stability of glass spheres catalytic system was also evaluated. As shown in Figure 8, more than 80% catalytic activity was retained up to 50 days. These results demonstrated that glass spheres catalytic system allowed a long-period operation and was capable of retaining its enzyme activity.

| CON CLUS ION
In this study, we proposed the program by controlling reaction time to produce a specific type of mogroside, mogroside IIIE. The glass spheres catalytic system where β-glucosidase was immobilized successfully converted mogroside V to mogroside IIIE. A reliable stability and reusability of the system was also reported.
To our knowledge, this is the first time that the immobilized enzymes system has been applied to Lo Han Ko mogrosides hydrolyzation, which offers an alternative solution to specifically produce mogroside IIIE and, meanwhile, lowers the enzyme cost.
Further study is still needed to improve the productivity of the system for practical applications. Different reaction profiles to produce specific mogrosides with different functions may be another task.

ACK N OWLED G M ENTS
This study was partly funded by the National Science Council, 106-2628-E-002-009-MY3.

CO N FLI C T O F I NTE R E S T
The authors declared no conflicts of interest.

E TH I C A L S TATEM ENT
The current study was not required to complete an ethical assessment.