Compound K Production from Red Ginseng Extract by β-Glycosidase from Sulfolobus solfataricus Supplemented with α-L-Arabinofuranosidase from Caldicellulosiruptor saccharolyticus

Ginsenoside compound K (C-K) is attracting a lot of interest because of its biological and pharmaceutical activities, including hepatoprotective, antitumor, anti-wrinkling, and anti-skin aging activities. C-K has been used as the principal ingredient in skin care products. For the effective application of ginseng extracts to the manufacture of cosmetics, the PPD-type ginsenosides in ginseng extracts should be converted to C-K by enzymatic conversion. For increased yield of C-K from the protopanaxadiol (PPD)-type ginsenosides in red-ginseng extract (RGE), the α-l-arabinofuranoside-hydrolyzing α-l-arabinofuranosidase from Caldicellulosiruptor saccharolyticus (CS-abf) was used along with the β-d-glucopyranoside/α-l-arabinopyranoside-hydrolyzing β-glycosidase from Sulfolobus solfataricus (SS-bgly) because SS-bgly showed very low hydrolytic activity on the α-l-arabinofuranoside linkage in ginsenosides. The optimal reaction conditions for C-K production were as follows: pH 6.0, 80°C, 2 U/mL SS-bgly, 3 U/mL CS-abf, and 7.5 g/L PPD-type ginsenosides in RGE. Under these optimized conditions, SS-bgly supplemented with CS-abf produced 4.2 g/L C-K from 7.5 g/L PPD-type ginsenosides in 12 h without other ginsenosides, with a molar yield of 100% and a productivity of 348 mg/L/h. To the best of our knowledge, this is the highest concentration and productivity of C-K from ginseng extract ever published in literature.


Introduction
Ginseng (Panax ginseng C. A. Meyer), one of the most valuable herbs, has been used in traditional medicine in Asian countries for over 2000 years [1]. The active components of ginseng are ginsenosides, which possess diverse biological and pharmaceutical activities, including anti-fatigue [2], anti-allergic [3], anti-oxidant [4], anti-inflammatory [5], anti-cancer [6], and anti-skin aging [7] activities. Ginsenosides are classified into three groups on the basis of the types of the aglycone structures, namely, oleanane, protopanaxadiol (PPD), and attached to the resin were successively eluted with methanol, the eluant was evaporated to remove methanol, and the residue was dissolved in the same volume of distilled water as the original loading volume. Sugar-free RGE was used because the Maillard reactions between free sugars and the enzyme occurred at temperatures above 70°C.
Cloning and gene expression S. solfataricus DSM 1617 and C. saccharolyticus DSM 8903, Escherichia coli ER2566, and the plasmids pET-24a and pTrc99A were used as the sources of β-glycosidase and α-L-arabinofuranosidase genes, host cells, and expression vectors, respectively. The genomic DNA from S. solfataricus and C. saccharolyticus was extracted using the genomic DNA extraction kit (Qiagen, Hilden, Germany). The genes encoding SS-bgly and CS-abf were amplified from each genomic DNA by polymerase chain reaction (PCR) using Pfu DNA polymerase (Solgent, Daejeon, Korea). The sequences of the oligonucleotide primers used for gene cloning were based on the DNA sequences of SS-bgly (GenBank accession number, WP 009992676) and CS-abf (Gen-Bank accession number, YP 001180344). Forward (5'-GCGTCTGCATATGTACTCATTTCC AAATAGC-3') and reverse (5'-GCGAATGGATCCTTAGTGCCTTAATGGCTTTAC-3') primers were designed to introduce the NdeI and BamHI restriction sites (underlines), respectively, for SS-bgly gene insertion, and forward (5'-TTTGGATCCATGAAAAAAGCAAAAGT CATCTA-3') and reverse (5'-TTTCTGCAGTTAATTTTCTCTCTTCTTCAATCTG-3') primers were designed to introduce the BamHI and PstI restriction sites (underlined), respectively, for CS-abf gene insertion. Each amplified DNA fragment obtained by PCR was purified and digested with the corresponding restriction endonuclease [34,35]. The digested DNA fragments of SS-bgly and CS-abf were extracted using a gel extraction kit (Qiagen, Hilden, Germany), and then inserted into the pET-24a vector and pTrc99A vectors, respectively, and digested with the same restriction enzymes. E. coli ER2566 strain was transformed with the ligation mixture and plated on Luria-Bertani (LB) agar containing 50 μg/mL of ampicillin. Ampicillin resistant colony was selected, and plasmid DNA from the transformant was isolated using a plasmid purification kit (Promega, Madison, WI). DNA sequencing was performed at the Macrogen facility (Seoul, Korea). The expression of SS-bgly and CS-abf genes was evaluated by both sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and enzyme activity assay.

Culture conditions
S. solfataricus and C. saccharolyticus were grown at 70°C under anaerobic conditions in the presence of 100% N 2 gas for 5 days on Sulfolobus medium (DSM Media Formulation No. 88) and Caldicellulosiruptor medium (DSM Media Formulation No. 640), respectively. The recombinant E. coli was cultivated with shaking at 200 rpm in a 2-L flask containing 500 mL of LB medium at 37°C with 20 μg/mL of kanamycin. When the optical density of the bacteria at 600 nm reached 0.8, isopropyl-β-D-thiogalactopyranoside (IPTG) was added to a final concentration of 0.1 mM to induce enzyme expression, and the culture was incubated at 16°C with shaking at 150 rpm for 16 h.

Enzyme purification
The induced E. coli cells were harvested from the culture broth by centrifugation at 6,000g for 30 min at 4°C, washed twice with 0.85% NaCl, and resuspended in 50 mM citrate/phosphate buffer (pH 6.0) containing 0.1 mM phenylmethylsulfonyl fluoride as the protease inhibitor. The resuspended cells were disrupted by sonication using Sonic Dismembrator (Model 100, Fisher Scientific, Pittsburgh, PA, USA) on ice for 10 min. The unbroken cells and cell debris were removed by centrifugation at 13,000×g for 20 min at 4°C, and the supernatant obtained was used as the crude extract. The crude extract was subsequently heated at 75°C for 10 min, and the suspension was centrifuged at 13,000×g for 20 min to remove insoluble denatured proteins. The supernatant obtained was used as the soluble enzyme.

Enzyme reactions
One unit (U) of SS-bgly or CS-abf activity was defined as the amount of enzyme required to liberate 1 nmol of C-K or ginsenoside Rd as a product from ginsenoside Rd or Rc as a substrate, respectively. Unless otherwise stated, the reactions were performed at 80°C in 50 mM citrate/ phosphate buffer (pH 6.0) containing 1.25 g/L PPD-type ginsenosides in RGE for 2 h.

Optimization of the supplementation ratio of SS-bgly and CS-abf
The effect of the supplementation ratio of SS-bgly and CS-abf on the conversion of ginsenoside Rc to C-K was investigated by the enzymatic reactions of SS-bgly supplemented with CS-abf at the ratios of 0:2 to 2:0 U/mL, at a total concentration of 2 U/mL. The reactions were performed at 80°C in 50 mM citrate/phosphate buffer (pH 6.0) containing 0.4 g/L ginsenoside Rc and two enzymes, at a total concentration of 2 U/mL, for 1 h. The time-course reactions for the conversion of PPD-type ginsenosides in RGE to C-K were performed at 80°C in 50 mM citrate/phosphate buffer (pH 6.0) containing 1.25 g/L PPD-type ginsenosides in RGE, 0.4 U/mL SS-bgly, and 1.6 U/mL CS-abf in 20 h. The effect of the supplementation ratio of CS-abf to SS-bgly on the production of C-Mc from PPD-type ginsenosides in RGE was conducted by supplementing CS-abf, ranging from 0.0 to 1.6 U/mL, with 0.4 U/mL SS-bgly. The reactions were performed with 1.25 g/L PPD-type ginsenosides in RGE for 20 h.

Effects of pH and temperature on C-K production
The effects of pH and temperature on the production of C-K from PPD-type ginsenosides in RGE were investigated by varying the pH from 4.5 to 7.5, and the temperature from 65 to 95°C, respectively. The reactions were performed in 50 mM citrate/phosphate buffer containing 0.4 U/mL SS-bgly, 0.6 U/mL CS-abf, and 1.25 g/L PPD-type ginsenosides in RGE for 2 h. The effect of temperature on the stability of SS-bgly supplemented with CS-abf was monitored as a function of incubation time by maintaining the solution of the enzymes at five different temperatures (70, 75, 80, 85, and 90°C) in 50 mM citrate/phosphate buffer (pH 6.0). Samples were withdrawn at time intervals and assayed.

Effects of the concentrations of enzymes and substrate on C-K production
The optimal concentration of the two enzymes required for C-K production was determined by varying the concentrations of SS-bgly and CS-abf at the ratio of 1:1.5, ranging from 0.4 and 0.6 U/mL to 4 and 6 U/mL at a constant substrate concentration of 7.5 g/L PPD-type ginsenosides in RGE. To determine the optimal substrate concentration, the concentration of PPDtype ginsenosides was varied from 1.25 to 10 g/L at constant enzyme concentrations of 2 U/mL SS-bgly and 3 U/mL CS-abf. The reactions were performed in 50 mM citrate/phosphate buffer (pH 6.0) at 80°C for 2 h.

Production of C-K from PPD-type ginsenosides in RGE
The time-course reactions for the conversion of PPD-type ginsenosides in RGE to C-K were performed at 80°C in 50 mM citrate/phosphate buffer (pH 6.0) containing 7.5 g/L PPD-type ginsenosides in 17.4% (w/v) RGE and 2 U/mL SS-bgly or 2 U/mL SS-bgly supplemented with 3 U/mL CS-abf for 12 h.

HPLC analysis
A reaction solution containing digoxin as an internal standard was extracted with an equal volume of n-butanol. The n-butanol fraction was then evaporated until dry, and methanol was added. Ginsenosides were assayed using an HPLC system (1100, Agilent, Santa Clara, CA, USA) equipped with a UV detector at a detection wavelength of 203 nm and a C18 column (YMC, Kyoto, Japan). The column was eluted at 37°C with a linear gradient of acetonitrile/ water from 20:80 to 80:20 (v/v) for 80 min at a flow rate of 1 mL/min.

TLC analysis
Three microliters of the ginsenoside standards and the reaction products of ginsenoside Rc were spotted onto a TLC plate (silica gel 60 F254, Merck, Darmstadt, Germany). The compounds on the plate were developed using a solvent mixture of chloroform, methanol, and water (40:20:10, v/v/v). The spots on the plate were detected by spraying 10% sulfuric acid, followed by incubation at 110°C in a dry oven for 5 min.

Results and Discussion
Conversion of major PPD-type ginsenosides to C-K by SS-bgly supplemented with CS-abf SS-bgly is a useful enzyme for C-K production because the enzyme simultaneously hydrolyzes the outer β-D-glucopyranoside or α-L-arabinopyranoside linked to C-20, and the inner and outer β-D-glucopyranosides linked to C-3 in PPD-type ginsenosides [32]. However, the enzyme exhibits very low hydrolytic activity for α-L-arabinofuranoside in PPD-type ginsenosides, indicating that SS-bgly does not convert ginsenosides Rc and C-Mc to Rd and C-K, respectively. Thus, CS-abf, which showed hydrolytic activity for α-L-arabinofuranoside [36], was supplemented to SS-bgly for the complete conversion of all PPD-type ginsenosides to C-K (Fig 1).
SS-bgly and CS-abf were purified as soluble proteins by heat treatment of crude enzymes extracted from harvested cells. The conversion of ginsenoside Rc to C-K by the enzymatic reaction of SS-bgly supplemented with CS-abf was analyzed and confirmed by TLC using the ginsenoside standards Rc, Rd, F 2 , C-Mc, and C-K (S1 Fig). Ginsenosides Rd, F 2 (trace), C-Mc, and C-K were obtained from ginsenoside Rc by SS-bgly supplemented with CS-abf, whereas only C-Mc was obtained by SS-bgly alone.
These results indicate that the supplementation of CS-abf to SS-bgly is an effective method for C-K production by introducing the α-L-arabinofuranoside-hydrolyzing activity, which caused the conversion of Rc and C-Mc to Rd and C-K, respectively.

Optimization of the supplementation ratio of SS-bgly and CS-abf for C-K production
The production of C-K from ginsenoside Rc was investigated by supplementing CS-abf to SSbgly at the ratios ranging from 0:2 to 2:0 U/mL, to make up a total concentration of 2 U/mL. The maximum production of C-K was observed at 0.4 U/mL SS-bgly and 1.6 U/mL CS-abf with a unit ratio of 1:4 (Fig 2). This ratio was used in the conversion of PPD-type ginsenosides in RGE to C-K (Fig 3). In the conversion, ginsenoside Rc and Mc disappeared after 2 h, whereas ginsenoside Rd remained at a constant concentration of approximately 100 mg/L even after 16 h. These results suggest that the concentration of CS-abf added to the reaction solution for C-K production was excess compared to that of SS-bgly. To optimize the supplementation ratio of SS-bgly and CS-abf for the production of C-K from PPD-type ginsenosides in RGE, the residual concentration of C-Mc was determined after 20 h by increasing the concentration of   (Fig 4). Therefore, the optimal unit ratio of SS-bgly and CS-abf was 0.4 U/mL: 0.6 U/mL (1:1.5) for the production of C-K from PPD-type ginsenosides in RGE, respectively. This optimal unit ratio was used in the subsequent reactions.

Effects of pH and temperature on C-K production by SS-bgly supplemented with CS-abf
The maximum production of C-K by the enzymatic reaction of SS-bgly supplemented with CS-abf at the optimal unit ratio was observed at pH 6.0 and 85°C (Fig 5), and the optimal pH and temperature of only SS-bgly were also 6.0 and 85°C, respectively. The thermal inactivation of SS-bgly supplemented with CS-abf at the optimal unit ratio for the production of C-K from PPD-type ginsenosides in RGE followed first-order kinetics, with half-lives of 346, 76, 25, 15, and 1.2 h at 70, 75, 80, 85, and 90°C (S2 Fig). The half-lives of SS-bgly at 70, 75, 80, and 85°C were 700, 100, 66, and 30 h [32]; and those of CS-abf were 130, 41, 2.5, and 0.12 h [36]. Thus, the thermostability of SS-bgly supplemented with CS-abf was greater than that of CS-abf alone, but less than that of SS-bgly alone. Because of the instability of SS-bgly supplemented with CSabf above 85°C, the reaction temperature for C-K production by SS-bgly supplemented with CS-abf was determined to be 80°C.

Effects of the concentrations of enzymes and substrate on C-K production by SS-bgly supplemented with CS-abf
The effect of the concentration of SS-bgly supplemented with CS-abf on C-K production was investigated with 7.5 g/L PPD-type ginsenosides in RGE as a substrate by varying the concentrations of SS-bgly and CS-abf at the optimal unit ratio of 1:1.5, ranging from 0.4 and 0.6 U/mL to 4 and 6 U/mL, respectively ( Fig 6A). C-K production increased with increasing enzyme concentrations, however, C-K production per enzyme concentration decreased at concentrations above 2 U/mL SS-bgly and 3 U/mL CS-abf. Thus, the optimal concentration of the enzymes required for the production of C-K from PPD-type ginsenosides in RGE was 2 U/mL SS-bgly and 3 U/mL CS-abf.
To investigate the effect of substrate concentration on C-K production, the concentration of PPD-type ginsenosides in RGE was varied from 1.25 to 10 g/L. The conversion yield decreased with increasing the substrate concentration ( Fig 6B). However, the maximal production of C-K was exhibited above 7.5 g/L PPD-type ginsenosides. Thus, the optimal concentration of substrate was 7.5 g/L PPD-type ginsenosides in RGE.

C-K production from PPD-type ginsenosides in RGE by SS-bgly supplemented with CS-abf under optimum conditions
The optimal reaction conditions for the production of C-K from PPD-type ginsenosides in RGE by SS-bgly supplemented with CS-abf were pH 6.0, 80°C, 2 U/mL SS-bgly, 3 U/mL CSabf, and 7.5 g/L PPD-type ginsenosides in RGE. Under these conditions, 2 U/mL SS-bgly without supplementation of CS-abf produced 3.1 g/L C-K from 7.5 g/L PPD-type ginsenosides in RGE along with other ginsenosides, 1.1 g/L C-Mc and 0.3 g/L Rd, after 12 h, with a molar yield of 75% and a productivity of 263 mg/L/h for the production of C-K from PPD-type ginsenosides (Fig 7A). To convert the remaining other ginsenosides to C-K, 3 U/mL CS-abf was   supplemented to 2 U/mL SS-bgly. Under the optimized conditions, SS-bgly supplemented with CS-abf converted 7.5 g/L PPD-type ginsenosides in RGE to 4.2 g/L C-K in 12 h, with a molar yield of 100% and a productivity of 348 mg/L/h (Fig 7B). The molar yield and productivity of C-K by SS-bgly supplemented with CS-abf were 25% and 1.3-fold higher, respectively, than those by SS-bgly alone. No C-K was formed when the reaction was run under the same experimental conditions without enzymes.
The production of C-K from PPD-type ginsenosides in ginseng extracts by enzymes and cells is presented in Table 1. C-K was also synthesized from glucose by metabolically engineered yeast, however, the produced concentration of C-K was low (1.4 mg/L) [37]. β-Glycosidase from Sulfolobus acidocaldarius (SA-bgly) produced 2.0 g/L C-K from 5 g/L PPD-type ginsenosides in ginseng root extract (GRE) in 22 h, with a molar yield of 69% and a productivity of 91 mg/L/h [38], which were 6% and 2.9-fold lower than those in RGE by SS-bgly. α-L-Arabinopyranoside linked to ginsenoside Rb 2 could be hydrolyzed by SS-bgly but not by SA-bgly. The content of Rc among the PPD-type ginsenosides such as Rb 1 , Rb 2 , Rc, and Rd in RGE (21.3%) was lower than that in GRE (26.9%), and the concentration of the PPD-type ginsenosides in 10% (w/v) RGE was 1.5-fold higher ( Table 2), indicating that RGE is a better substrate for C-K production. Thus, SS-bgly using RGE as the substrate exhibited higher conversion yield and productivity of C-K than SA-bgly using GRE. Diverse glycosidases from Fusobacterium sp. [39], Microbacterium esteraromaticum [28], S. acidocaldarius [29], Aspergillus niger [40], Aspergillus sp. [41], Esteya vermicola [42], and Terrabacter ginsenosidimutans [43] converted reagent-grade major ginsenosides Rb 1 , Rb 2 , Rc, and Rd to C-K. However, the quantitative production of C-K from Rc was conducted by only the combined use of threeenzymes [38]. The three-enzyme system of SA-bgly, and α-L-arabinofuranosidase and β-galactosidase from C. saccharolyticus produced 2.9 g/L C-K from 5 g/L PPD-type ginsenosides in GRE in 20 h, with a molar yield of 100% and a productivity of 144 mg/L/h [38]. To the best of our knowledge, this was the highest previously reported conversion yield and productivity of C-K from ginseng extract. The concentration and productivity of C-K as obtained by the two-enzyme system of SS-bgly and CS-abf using RGE, which contained the higher content of PPD-type ginsenosides than GRE, were 1.4-and 2.4-fold higher than those obtained by the This study three-enzyme system using GRE, respectively. The supplementation of CS-abf to SS-bgly resulted in the complete conversion of all PPD-type ginsenosides in RGE to C-K with the highest productivity ever reported via two transformation pathways: Rb 1 , Rb 2 , and Rc ! Rd ! F 2 ! C-K and Rc ! C-Mc ! C-K (Fig 1). The conversion of ginsenosides Rb 1 , Rb 2 , Rc, and Rd in RGE to C-K via F 2 , and C-Mc was analyzed and confirmed by HPLC with the ginsenoside standards (S3 Fig).

Conclusions
SS-bgly shows hydrolytic activity on the β-D-glucopyranoside and α-L-arabinopyranoside linkage in ginsenosides without the α-L-arabinofuranoside linkage. To compensate this defect, SSbgly was supplemented with α-L-arabinofuranoside-hydrolyzing CS-abf. This combination completely converted PPD-type ginsenosides in RGE to C-K. To the best of our knowledge, this is the highest concentration and productivity of C-K from ginseng extract ever published in literature. Our results would contribute toward improving the industrial production of C-K by biotransformation.   Fig. HPLC profiles of the C-K produced from PPD-type ginsenosides in RGE by SS-bgly supplemented with CS-abf. HPLC analysis of C-K (27.9 min) produced from ginsenosides Rb 1 (9.6 min), Rb 2 (11.9 min), Rc (11.4 min), and Rd (