- Email: [email protected]
Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase
Journal Pre-proofs Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase Wang Ma, Lu Zhao, Yudi Ma, Yuqiang Li, Song Qin, Bingfang He PII: DOI: Reference:
S0141-8130(19)33207-6 https://doi.org/10.1016/j.ijbiomac.2019.09.208 BIOMAC 13504
To appear in:
International Journal of Biological Macromolecules
Received Date: Revised Date: Accepted Date:
30 April 2019 1 July 2019 20 September 2019
Please cite this article as: W. Ma, L. Zhao, Y. Ma, Y. Li, S. Qin, B. He, Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase, International Journal of Biological Macromolecules (2019), doi: https://doi.org/10.1016/j.ijbiomac.2019.09.208
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier B.V.
Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase
Wang Maa, Lu Zhaoa, Yudi Maa, Yuqiang Lia, Song Qina, Bingfang Hea,b,*.
a
College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University,
No. 30 Puzhu South Road, Nanjing, 211816, China. b
School of Pharmaceutical Sciences, Nanjing Tech University, No. 30 Puzhu South
Road, Nanjing, 211816, China. Corresponding author: Bingfang He E–mail address:[email protected]
Abstract Ginsenoside
Rh2
(3β-O-Glc-protopanaxadiol),
a
trace
but
an
important
pharmacological component of ginseng, has exhibited excellent medicinal potential. Many studies have found that the synthesis of Rh2 by UDP-glucosyltransferase (UGT) is an alternative production strategy. In this study, Yjic from B. subtilis 168 was found to synthesize ginsenoside F12 (3β,12β-Di-O-Glc-protopanaxadiol) and Rh2 at a ratio of 7:3. Yjic regioselectivity toward Rh2 synthesis was successfully improved using a semi-rational design including structure-guided alanine scanning and saturation mutations. As a result, mutant M315F was found to efficiently synthesize Rh2 (~99%) and block the further glycosylation of C12-OH. The circulation of UDPG was achieved by combining M315F with AtSuSy through a cascade reaction. Furthermore, an extraordinarily high yield of Rh2 (3.7 g/L) was attained in an aqueous solvent system with 17% DMSO (v/v) through the fed-batch feeding of PPD. This study presents the high potential for the oriented preparation of ginsenoside Rh2. Key words: Ginsenoside Rh2; UDP-glycosyltransferase; Yjic; Mutant M315F; Cascade reaction
1. Introduction Ginsenosides are the major pharmacologically active components of ginseng (Panax ginseng, C. A. Meyer) [1]. Natural ginsenosides exhibit various pharmacological effects, such as anticancer, antitumor, antistress, antiaging, and anti-inflammatory, which protect the central nervous system and enhance immune system activities [2-4]. Ginsenosides can be classified into the protopanaxadiol (PPD) or protopanaxatriol (PPT) groups depending on the hydroxylation of the C6 site [5]. Ginsenoside Rh2 belongs to the PPD group and is a trace but important saponin isolated from red ginseng [6]. Ginsenoside Rh2 also shows significant anticancer, anti-inflammatory, and antiallergic pharmaceutical activities [7-9]. The long growth periods of ginseng and difficulties in extracting certain ginsenosides has initiated research into artificially synthesizing rare ginsenosides. However, chemical approaches for synthesizing ginsenosides face enormous challenges, such as steric hindrance and acid labile dammarane derivatives [10]. In addition, by-products and low yield rates are obstacles in the development of chemical synthesis [11]. In recent years, synthetic biology has played a unique role in the study of natural products [12-13]. PPD (~ 1.2 g/L) has been creatively produced via a synthetic biology pathway using glucose as the raw material in yeast [14]. This method of production has dramatically solved PPD resourcing issues and has laid a solid foundation for the enzymatic synthesis of ginsenosides. UDP-glycosyltransferases (UGTs) generally transfers glucose from UDP-glucose (UDPG) to various metabolites, such as hormones and secondary metabolites [15-16].
Considerable numbers of UGTs with multiple-site selectivity have been identified [17]. At present, the functions of key UGTs from ginseng for ginsenoside synthesis have been illuminated [18-19]. Compared with plant UGTs, some microbial UGTs show high aglycon promiscuity, high catalytic proficiency, and can be efficiently expressed in engineered bacteria [20-22]. For example, UGT51 from Saccharomyces cerevisiae, which exhibits broad acceptor tolerance and can transfer a glucosyl moiety to the free C3-OH of PPD. Through semi-rational design, UGT51 presents an ~1800-fold enhancement of catalytic efficiency (kcat/Km) for converting PPD to ginsenoside Rh2 in vitro (~300 mg/L) [23]. UDP-glycosyltransferases Yjic (NP389104) from Bacillus subtilis has been used for extensive structurally diverse compounds [24]. Bs-Yjic can transfer a glucosyl moiety to the C3-OH and C12-OH of protopanaxadiol (PPD), and successfully synthesize ginsenosides F12 and Rh2 (~7%) in 12 h [25]. In this study, a glycosyltransferase from Bacillus subtilis 168 was selected after the homologous evolutionary analysis used for Rh2 synthesis (Fig. S1). It was then confirmed that the sequence of amino acids was the same as that of Yjic. In subsequent experiments, attempts were made to obtain a mutant with increasing C3 specificity of PPD by semi-rational design. A cascade reaction system of mutant and sucrose synthase from Arabidopsis thaliana (AtSuSy) was then constructed to alleviate the inhibition of high concentrations of UDP and realize the circulation of UDPG (Fig. 1). Furthermore, the final yield of Rh2 was increased by fed-batch synthesis.
2. Material and methods 2.1 Bacterial strains and substrates The Bacillus subtilis 168 (Bio-82277) strain was purchased from the China Center for Type Culture Collection. Authentic PPD and PPD-type ginsenosides (Rh2 and CK) were purchased from Nanjing Spring & Autumn Biological Engineering Co., Ltd. (Nanjing, China) and dissolved in dimethyl sulfoxide (DMSO) for use. All other chemicals were of analytical grade. 2.2 The molecular cloning, expression, and semi-rational design of the Yjic from Bacillus subtilis 168 The Yjic gene (AAC46318.1) excluding 392 amino acids was amplified from the genomic DNA of B. subtilis 168, using the primer sets Yjic-F (5’-ATGAGATTTCCTT CAATTTT-3’)
and
Yjic-R
(5’-TTAACAACTACCGACGCCAA-3’),
which
incorporated Nco I and Xho I sites, respectively. Recombinant plasmid pET-28a-Yjic was then transformed into E. coli BL21 (DE3) for expression. SWISS-MODEL was employed to conduct structural modeling for the prediction of the substrate binding region of Yjic, and CalG2 (PDB:3iaa) from Calicheamicin was used as a template [26]. 2.3 Site-directed mutagenesis and saturation mutagenesis Alanine scanning mutagenesis, a method of systematic alanine substitution, was particularly useful for the identification of functional residues. The primers used for all chosen residues are shown in Table S1. Then the mutant that increases Rh2
production for saturation mutation was selected, and the mutant plasmid was used as the initial template for saturation mutagenesis. All primers for saturation mutagenesis are shown in Table S2. All mutants were created in the recombinant vector pET-28a-UGT. PCR was conducted with Kodaq 2X PCR MasterMix. The product was subsequently treated with Dpn I for 3 h at 37 °C and transformed into E. coli BL21 cells (DE3). The targeted mutation was confirmed by sequencing. After the induced expression of recombinant E. coli, soluble fractions of the enzymes were obtained by cell sonication. The relative activities were measured using wild type Yjic as the control. 2.4 Analysis of selected mutant properties The effect of pH on the synthesis activity of ginsenoside was assayed at 30 ℃ and 200 rpm in the following buffer solutions (each at 50 mM); citrate buffer (pH 5.0), phosphate buffer (pH 6.0-7.0), Tris-HCl (pH 8.0-9.0), and glycine-NaOH (pH 10.0). The pH stability of the enzymes was examined by measuring residual activity after incubation in each buffer for 2 h at 4 ℃. The results were expressed as the relative conversion rate of the PPD substrate compared to the wild type Yjic. The effect of temperature on the synthesis activity of ginsenoside was tested at different temperatures ranging from 20 ℃ to 60 ℃ at pH 8.0. The temperature stability of the enzyme was determined by measuring the residual activity after preincubation in 50 mM Tris-HCl (pH 8.0) for 6 h at different temperatures (20-60 ℃). The results were expressed as the relative conversion rate of the PPD substrate. 2.5 The effect of the cofactor and organic solvent on enzyme activities
Cofactors, especially metal ions, play an important role in enzymatic reactions [27]. The cofactors for this test include: sodium dodecyl sulfonate (SDS), dithiothreitol (DTT), β-mercaptoethanol (β-ME), K+, Na+, NH4+, Ni2+, Ba2+, Zn2+, Ca2+, Mg2+, Mn2+, and Fe3+. All the ions originate from chloride. The reaction mixture consisted of 500 μL of 50 mM Tris-HCl buffer (pH 8.0) containing PPD (2 mM), UDP-Glucose (4 mM), 10% (v/v) DMSO, and 5 mM cofactors. The mixture was incubated at 30 °C and 200 rpm, and the effects of cofactor concentrations were studied in subsequent experiments. The effects of organic solvents were estimated due to the low solubility of PPD in the water phase. The concentration of the organic solvents in the reaction system was 10% (v/v), and the effects of this concentration were optimized. 2.6 Enzyme activity assays Activity assays (0.5 mL) were conducted with 1 mM PPD, 2 mM UDPG, 50 mM Tris-HCl (pH 8.0), and a certain amount of mutant for 0.5 h at 30 °C. The reactions were terminated by adding two times of methanol (v/v). The reactants were subsequently centrifuged at 12000 rpm for 2 min. The supernatant was then filtered through a 0.22 μm filter, and the ginsenosides and residual PPD were analyzed by HPLC. The standard curve of PPD concentration is shown in Fig. S2. A single unit of enzyme activity was defined as the amount of mutant that glycosylated 1 μmol of PPD per minute. There were three parallel experiments for all operations. Activity assays of AtSuSy were conducted in a 1 mL volume containing 0.5 mM UDP, 300 mM sucrose, 50 mM Tris-HCl (pH 8.0), and various amounts of AtSuSy for
30 min at 30 °C. The reaction was terminated by boiling water bath for 5min. The AtSuSy activity was measured by determining the amount of fructose released from sucrose and analyzed using 3,5-dinitrosalicylic acid colorimetry (DNS) [28]. The standard curve of fructose concentration is shown in Fig. S3. A single unit of enzymatic activity was defined as the amount of enzyme that releases 1 μmol of fructose from sucrose per minute. There were three parallel experiments for all operations. 2.7 Optimization of the mutant/AtSuSy cascade reaction The Mutant/AtSuSy cascade reaction system contained 1 mM PPD, 0.2 mM UDP, 50 mM Tris-HCl (pH 8.0), 400 mM sucrose, 10% DMSO (v/v), 21 mU/mL mutant, and 48 mU/mL AtSuSy. The cascade reaction was incubated at 35°C and 200 rpm for 2 h. The effect of temperature was confirmed using a range of 25-45°C. In addition, the effects of UDP (0.05−0.5 mM), sucrose (0.1−0.8 M), and PPD (0.5−2 mM) were analyzed at different concentrations individually. Finally, the effects of the mutant and AtSuSy ratio was determined. 2.8 Fed-batch synthesis of ginsenosides by the mutant/AtSuSy cascade reaction The reaction mixtures (20 mL) contained 1 mM PPD, 0.3 mM UDP, 50 mM Tris-HCl (pH 8.0), 0.3 M sucrose, 10% DMSO, 5 mM Mg2+, 21 mU/mL Mutant, and 48 mU/mL AtSuSy. The reaction was performed at 35°C and 200 rpm. Next, 200 μL PPD (100 mM) was periodically added to the reaction mixtures at 1, 2, 4, 6, 8, 10, and 12 h and 200 μL of the reactant solution was collected and terminated by adding 3-fold the volume of methanol. After centrifugation at 12000 rpm for 2 min, the
supernatant was filtered through 0.22 μm filters and analyzed by HPLC. Fresh enzymes (11.5 mU/mL mutant and 24 mU/mL AtSuSy) were added at 4, 8 and 12 h. 2.9 HPLC, LC-MS, and NMR analysis of the glycosylated products The
reactant
(20
μL)
was
examined
by
high-performance
liquid
chromatography (HPLC). The ZORBAX SB-C18 reverse-phase column (4.6 × 150 mm, 5 μm particles) connected to an UltiMate 3000 HPLC system with UV detection at 203 nm was eluted with solvent A (water) and solvent B (methanol) using an equal-degree program of 80% B over 55 min. The solvent flow rate was 1.0 mL/min, and the column temperature was set at 30 °C. LC-MS analysis was performed in positive ion mode on an Agilent 6520 Accurate-Mass Q-TOF LC-MS platform (Palo Alto, CA, U.S.A.). The glycosylation products of PPD were purified using an Agilent 1200 preparative HPLC system coupled with a reverse phase Ultimate C18 column (21.2 × 250 mm2, 5 μm particles, Welch, Shanghai, China). The products were then analyzed using the proton and carbon nuclear magnetic resonance (1H NMR,
13
C
NMR) spectrum with samples dissolved in DMSO-d6 (Bruker Avance IIII 400). 3. Results and Discussion 3.1 The selection and semi-rational design of Yjic towards ginsenoside Rh2 synthase A glycosyltransferase (UGT) was selected from B. subtilis through the analysis of protein homology evolution and comparison of the PSPG (Plant Secondary Product Glycosyltransferase) motif (Fig. S1). It was found that this UGT could glycosylate PPD (1 mM) to the unnatural ginsenoside F12 (~69%) and the ginsenoside Rh2 (~31%) after 6 h of conversion (Fig. S4). The catalytic products are consistent with
the Yjic reported (NP389104) [25]. The spectrum of the proton and carbon nuclear magnetic resonance analysis of glycosylated product 1 is shown in Fig S5 and S6. The 1
H-NMR and
13
C-NMR data (Table S3) of product 1 is identical to that of the
3-O-β-D-glucopyranosyl-20(S)-protopanaxadiol (Rh2) [25]. Semi-rational design guided by structural information is usually an effective protein-engineering strategy for modifying catalytic properties or improving catalytic activity such as UGT51 [23].The PPD substrate was docked into the acceptor-binding pocket using the automated docking program Autodock 4.2.6 [29]. Some residues were found to be mutagenesis sites around the substrate binding pocket when the positions of the residues were mapped to the 3D structure of the protein (Fig. S7). 3.2 The site-specific modification of Yjic The alanine scanning mutation on the selected 28 residues nearing the substrate binding site was carried out to find the hotspot that influences the specificity of Yjic. As Fig. 2 shows, changing the majority of selected residues can dramatically change the relative conversion rates and regioselectivity (Rh2/total products) of Yjic. Among these residues, about half of the mutants experienced a promoted relative conversion rate, and the mutant R68A showed a 2.4-times higher conversion rate than the wild-type (WT). The mutants of some residues also significantly improved the regioselectivity of Yjic for Rh2 synthesis. About 70% of the mutants prefer to synthesize Rh2 compared with WT Yjic. Mutants Y14A, F168A, and M315A show a tendency to glycosylate the C3-OH of PPD. Fortunately, the mutant M315A showed nearly 99% regioselectivity (Fig. S8) for Rh2 synthesis. Further studies have found
that M315 residues are located in the PSPG motif, where the amino residues play an essential role in determining the substrate specificity of the enzymes [21]. Saturation mutation at residue 315 was then performed, and the relative conversion rates and regioselectivity of most mutants varied greatly (Fig. 3). The eight mutants exhibited a higher relative conversion rate than the WT Yjic. Surprisingly, ten mutants, M315A, M315F, M315V, M315S, M315E, M315K, M315Y, M315C, M315W, and M315G all exhibited perfect regioselectivity (~99%). Mutant M315F was selected for the following experiments due to both its relative conversion rate and regioselectivity. 3.3 Characterization and preliminary optimization for glycosylation of PPD by mutant M315F The effect of pH on the glycosylation of PPD was assayed. The M315F mutant shows higher activity and stability at pH 8 and could retain more than 90% of its vitality between pH 8 and 9 for two hours (Fig. S9A). The increased activity of M315F was observed at 37 °C. The enzymes also maintained excellent stability in the range of 20-37 °C for six hours, but the activity decreased sharply above 42 °C (Fig. S9B). The metal ions showed the most significant effect on the glycosylation rate (Fig. 4A). Mn2+ and Mg2+ are significantly promoted in Rh2 synthesis. Coincidentally, Dai et al [25]. report that Mg2+ strengthens the glycosylation of PPD by Bs-Yjic. The relative conversion rate of the mutant M315F was elevated by 1.6-times compared with the addition of 8 mM Mg2+ (Fig. 4B).
The effects of organic solvents on the glycosylation of PPD were analyzed due to its extremely low solubility. The addition of hydrophilic organic solvents (10%, v/v) promoted the dissolution of the PPD substrate and facilitated the catalytic reaction (Fig. 4C). However, the addition of hydrophobic organic solvents hindered the catalytic reaction. This result may have been caused by the presence of enzymes and substrates in different solvent layers. DMSO, DMF, and MeOH increased by more than twice compared with the initial conversion rate, with DMSO showing the most significant increase. The effects of solvent concentration on the glycosylation rates were further investigated. DMSO at a specific concentration (5%~20%, v/v) is beneficial for the glycosylation of PPD (Fig. 4D). The effects of UDPG on ginsenoside synthesis by mutant M315F were also studied. The production rate of Rh2 increased continuously with increasing UDPG concentration, and finally reached 1.08 mM (Fig. S10). The increased concentration of UDPG was found to be beneficial to the synthesis of ginsenoside F12 (Fig. S11). On the other hand, the conversion rate of total ginsenosides decreased when a higher substrate ratio (UDPG∶ PPD) over four was used. 3.4 Glycosylation of PPD using cascade reactions of M315F and AtSuSy A coupling reaction recombining AtSuSy and the mutant M315F was constructed to create in situ UDPG regeneration and reduce the accumulation of UDP in the reaction system. This reaction was used to synthesize ginsenoside Rh2 using PPD and sucrose as the aglycon and sugar donor, respectively. The ratio of M315F/AtSuSy on the synthesis of ginsenoside Rh2 was preliminarily optimized
(Table S4). The activity ratio 1:2.1 of M315F/AtSuSy (mU/mL) was selected due to the conversion rate of PPD and the higher proportion of Rh2 in total ginsenosides. A higher conversion of PPD was achieved at 35 °C through the elementary optimization of cascade catalysis conditions (Fig. 5A). UDP is used for UDPG supplementation in the UGT-AtSuSy cascade reaction, while UDP at higher concentrations can inhibit UGT activity [30]. The conversion of PPD increased by 1.72-fold when the ratio of UDP:PPD increased from 0.05 to 0.3 (1 mM PPD) and slightly decreased when the concentration of UDP increased further (Fig. 5B). An initial concentration of 0.3 M sucrose was selected for cascade catalysis to obtain efficient in situ UDPG regeneration (Fig. 5C). In addition, initial PPD concentration had a significant impact on the reaction. The glycosylation rate (Fig. 5D) gradually reduced as the initial concentration of PPD increased. Finally, 1 mM PPD at the initial concentration was selected for the catalytic reaction. The conditions for the optimization of the fed-batch cascade reaction were chosen as 35 oC, 0.3 mM UDP, 0.3 M sucrose, and 1 mM PPD. 3.5 Ginsenoside Rh2 preparation using fed-batch cascade reactions Ginsenoside Rh2 preparation was carried out via the fed-batch feeding of PPD based on the preliminary optimization of reaction conditions and the inhibition of high concentrations of PPD. The glycosylation of PPD occurred relatively quickly, at about 0.44 mM ginsenoside Rh2 from 1 mM PPD after 0.5 h of the reaction (Fig. 6). Only a trace amount of ginsenoside F12 was detected in the first 2 h, and ginsenoside Rh2 was the primary product, which is consistent with the in vitro enzyme reaction with
UDPG as the substrate. After 24 h, 5.92 mM (3.7 g/L) ginsenoside Rh2 and 0.64 mM ginsenoside F12 (0.5 g/L) were obtained through the periodic feeding of PPD with a total conversion yield of 82% (Fig. S12) in systems with 17% DMSO. The production intensity of Rh2 reached 0.154 g/h·L and benefited from the high efficiency of the mutant and DMSO aqueous reaction system to facilitate PPD dissolution and the glycosylation rate, the residual DMSO can be effectively controlled in the post-processing of the reaction. To our knowledge, the synthetic capacity of ginsenoside Rh2 in the present study was much higher than is reported [18,23]. The production of PPD via a yeast cell factory reduces the cost of PPD substrates and makes the production of high-value ginsenoside Rh2 by M315F-AtSuSy cascade reaction in vitro more commercially valuable [1]. 4. Conclusion In summary, the mutant M315F of Yjic from Bacillus subtilis 168 was found to efficiently synthesize Rh2 (~99%) by semi-rational design. Furthermore, the M315F/AtSuSy cascade reaction was constructed which reduces costs through the cyclic regeneration of UDPG and relieves the inhibition of UDP at high concentrations. Finally, fed-batch feeding of PPD was used to increase the yield of ginsenoside Rh2. This method resulted in the achievement of 5.92 mM (3.7 g/L) of Rh2 in DMSO aqueous solution with a final DMSO concentration of about 17%, which shows the practical potential of ginsenoside Rh2 production (Table 1).
Acknowledgements
This work was financially supported by the National Natural Science Foundation of China (81673321, 21776135, 21506099). We also thank the Jiangsu Synergetic Innovation Center for Advanced Bio-Manufacture (NO. XTC1812).
Reference [1] Z.B. Dai, B.B. Wang, Y. Liu, M.Y. Shi, D. Wang, X.N. Zhang, T. Liu, L.Q. Huang, X.L. Zhang, Producing aglycons of ginsenosides in bakers' yeast, Scientific Reports, 4 (2014) 6. [2] L. Li, S.-Y. Shin, S.J. Lee, J.S. Moon, W.T. Im, N.S. Han, Production of Ginsenoside F2 by Using Lactococcus lactis with Enhanced Expression of beta-Glucosidase Gene from Paenibacillus mucilaginosus, Journal of agricultural and food chemistry, 64 (2016) 2506-2512. [3] J. Pei, A. Chen, L. Zhao, F. Cao, G. Ding, W. Xiao, Enzymatic transformation of ginsenoside Rb1 to ginsenoside 20(S)-Rg3 by GH3 β-glucosidase from Thermotoga thermarum DSM 5069 T, Journal of agricultural and food chemistry, 65 (2017) 6042-6048. [4] J.Y. Seo, S.H. Ju, J. Oh, S.K. Lee, J.-S. Kim, Neuroprotective and Cognition-Enhancing Effects of Compound K Isolated from Red Ginseng, Journal of agricultural and food chemistry, 64 (2016) 2855-2864. [5] L.P. Christensen, Ginsenosides chemistry, biosynthesis, analysis, and potential
health effects, Advances in food and nutrition research, 55 (2009) 1-99. [6] X. Jin, D.B. Che, Z.H. Zhang, H.M. Yan, Z.Y. Jia, X.B. Jia, Ginseng consumption and risk of cancer: A meta-analysis, Journal of ginseng research, 40 (2016) 269-277. [7] M.K. Choo, E.K. Park, M.J. Han, D.H. Kim, Antiallergic activity of ginseng and its ginsenosides, Planta Medica, 69 (2003) 518-522. [8] G.-w. Liu, Y.-h. Liu, G.-s. Jiang, W.-d. Ren, The reversal effect of Ginsenoside Rh2 on drug resistance in human colorectal carcinoma cells and its mechanism, Human cell, 31 (2018) 189-198. [9] X.W. Xie, A. Eberding, C. Madera, L. Fazli, W. Jia, L. Goldenberg, M. Gleave, E.S. Guns, Rh2 synergistically enhances paclitaxel or mitoxantrone in prostate cancer models, Journal Of Urology, 175 (2006) 1926-1931. [10] J. Liao, J. Sun, Y. Niu, B. Yu, Synthesis of ginsenoside Rh2 and chikusetsusaponin-LT8
via
gold(I)-catalyzed
glycosylation
with
a
glycosyl
ortho-alkynylbenzoate as donor, Tetrahedron Letters, 52 (2011) 3075-3078. [11] J. Yu, J. Sun, Y. Niu, R. Li, J. Liao, F. Zhang, B. Yu, Synthetic access toward the diverse ginsenosides, Chemical Science, 4 (2013) 3899-3905. [12] Y.J. Kim, D. Zhang, D.C. Yang, Biosynthesis and biotechnological production of ginsenosides, Biotechnol Adv, 33 (2015) 717-735. [13] C. Zhao, X. Gao, X. Liu, Y. Wang, S. Yang, F. Wang, Y. Ren, Enhancing Biosynthesis of a Ginsenoside Precursor by Self-Assembly of Two Key Enzymes in Pichia pastoris, Journal of agricultural and food chemistry, 64 (2016) 3380-3385. [14] Z. Dai, Y. Liu, X. Zhang, M. Shi, B. Wang, D. Wang, L. Huang, X. Zhang,
Metabolic engineering of Saccharomyces cerevisiae for production of ginsenosides, Metabolic Engineering, 20 (2013) 146-156. [15] B.K. Hou, E.K. Lim, G.S. Higgins, D.J. Bowles, N-glucosylation of cytokinins by glycosyltransferases of Arabidopsis thaliana, Journal Of Biological Chemistry, 279 (2004) 47822-47832. [16] W. Wei, P. Wang, Y. Wei, Q. Liu, C. Yang, G. Zhao, J. Yue, X. Yan, Z. Zhou, Characterizations
of
Panax
ginseng
UDP-glycosyltransferases
catalyzing
protopanaxatriol and biosyntheses of bioactive ginsenosides F1 and Rh1 in metabolically engineered yeasts, Molecular Plant, 8 (2015) 1412-1424. [17] J.A. Campbell, G.J. Davies, V. Bulone, B. Henrissat, A classification of nucleotide-diphospho-sugar glycosyltransferases based on amino acid sequence similarities, Biochemical Journal, 326 (1997) 929-939. [18] S.-C. Jung, W. Kim, S.C. Park, J. Jeong, M.K. Park, S. Lim, Y. Lee, W.-T. Im, J.H. Lee, G. Choi, S.C. Kim, Two Ginseng UDP-Glycosyltransferases Synthesize Ginsenoside Rg(3) and Rd, Plant And Cell Physiology, 55 (2014) 2177-2188. [19] Y. Xing, F. Yun, W. Wei, W. Pingping, L. Qunfang, W. Yongjun, Z. Lei, Z. Guoping, Y. Jianmin, Z. Zhihua, Production of bioactive ginsenoside compound K in metabolically engineered yeast, Cell Research, 24 (2014) 770-773. [20] H.-H. Chiu, M.-Y. Shen, Y.-T. Liu, Y.-L. Fu, Y.-A. Chiu, Y.-H. Chen, C.-P. Huang, Y.-K. Li, Diversity of sugar acceptor of glycosyltransferase 1 from Bacillus cereus and its application for glucoside synthesis, Applied microbiology and biotechnology, 100 (2016) 4459-4471.
[21] B. Fan, T. Chen, S. Zhang, B. Wu, B. He, Mining of efficient microbial UDP-glycosyltransferases by motif evolution cross plant kingdom for application in biosynthesis of salidroside, Scientific Reports, 7 (2017). [22] R.P. Pandey, P. Parajuli, N. Koirala, J.W. Park, J.K. Sohng, Probing 3-Hydroxyflavone for In Vitro Glycorandomization of Flavonols by YjiC, Applied And Environmental Microbiology, 79 (2013) 6833-6838. [23] Y. Zhuang, G.-Y. Yang, X. Chen, Q. Liu, X. Zhang, Z. Deng, Y. Feng, Biosynthesis of plant-derived ginsenoside Rh2 in yeast via repurposing a key promiscuous microbial enzyme, Metabolic Engineering, 42 (2017) 25-32. [24] L. Dai, J. Li, P. Yao, Y. Zhu, Y. Men, Y. Zeng, J. Yang, Y. Sun, Exploiting the aglycon promiscuity of glycosyltransferase Bs-YjiC from Bacillus subtilis and its application in synthesis of glycosides, Journal Of Biotechnology, 248 (2017) 69-76. [25] L. Dai, C. Liu, J. Li, C. Dong, J. Yang, Z. Dai, X. Zhang, Y. Sun, One-Pot Synthesis of Ginsenoside Rh2 and Bioactive Unnatural Ginsenoside by Coupling Promiscuous Glycosyltransferase from Bacillus subtilis 168 to Sucrose Synthase, Journal of agricultural and food chemistry, 66 (2018) 2830-2837. [26] A. Chang, S. Singh, K.E. Helmich, R.D. Goff, C.A. Bingman, J.S. Thorson, G.N. Phillips, Jr., Complete set of glycosyltransferase structures in the calicheamicin biosynthetic pathway reveals the origin of regiospecificity, Proceedings Of the National Academy Of Sciences Of the United States Of America, 108 (2011) 17649-17654. [27] L. Wang, Q.-M. Liu, B.-H. Sung, D.-S. An, H.-G. Lee, S.-G. Kim, S.-C. Kim,
S.-T. Lee, W.-T. Im, Bioconversion of ginsenosides Rb1, Rb2, Rc and Rd by novel β-glucosidase hydrolyzing outer 3-O-glycoside from Sphingomonas sp. 2F2: Cloning, expression, and enzyme characterization, Journal Of Biotechnology, 156 (2011) 125-133. [28] Y. Jian-Xiong, L.U. Hong, X. Qiang, D. Jin-Ling, L.I. Ni-Hang, A study on rapid colorimetric determination of water soluble total sugar,reducing sugar and starch in tobacco with 3,5-dinitrosalicylic acid, Journal of Yunnan Agricultural University, (2007). [29] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility, Journal Of Computational Chemistry, 30 (2009) 2785-2791. [30] A. Gutmann, B. Nidetzky, Unlocking the Potential of Leloir Glycosyltransferases for Applied Biocatalysis: Efficient Synthesis of Uridine 5'-Diphosphate-Glucose by Sucrose Synthase, Advanced Synthesis & Catalysis, 358 (2016) 3600-3609. [31] L. Liu, X.-M. Zhu, Q.-J. Wang, D.-L. Zhang, Z.-M. Fang, C.-Y. Wang, Z. Wang, B.-S. Sun, H. Wu, C.-K. Sung, Enzymatic preparation of 20(S, R)-protopanaxadiol by transformation of 20(S, R)-Rg3 from black ginseng, Phytochemistry, 71 (2010) 1514-1520. [32] P. Wang, Y. Wei, Y. Fan, Q. Liu, W. Wei, C. Yang, L. Zhang, G. Zhao, J. Yue, X. Yan, Production of bioactive ginsenosides Rh2 and Rg3 by metabolically engineered yeasts, Metabolic Engineering, 29 (2015) 97-105. [33] L. Yu, Y. Chen, J. Shi, R. Wang, Y. Yang, L. Yang, S. Zhao, Z. Wang,
Biosynthesis of rare 20(R)-protopanaxadiol/protopanaxatriol type ginsenosides through Escherichia coli engineered with uridine diphosphate glycosyltransferase genes, Journal of ginseng research, 43 (2019) 116-124.
Fig. 1 Schematic diagram of PPD glycosylation reaction. The solid line represents the reaction pathway constructed in this experiment, the dotted line represents the reaction pathway catalyzed by Yjic. Fig. 2 Relative activities of alanine-substituted mutants of Bs-Yjic (Regioselectivity = Rh2 / total products). Reaction system (1 mL): 1 mM PPD, 2 mM UDPG, 10% DMSO, pH 8.0 Tris-HCl (50 mM). Reactions end at 6 h. Fig. 3 Relative activity of the saturated mutants at M315 (M = Methionine). Reaction system (1 mL): 1 mM PPD, 2 mM UDPG, 10% DMSO, pH 8.0 Tris-HCl (50 mM). Reactions end at 6 h. Fig. 4 Catalytic properties of M315F mutant under different conditions. (A) and (B) Effects of cofactors and ion concentration (Mg2+, Mn2+, Ca2+) on the activity of M315F. (C) and (D) Effects of solvents and concentration (DMF, DMSO, MeOH) on the activity of M315F. Fig. 5 Optimization of the reaction conditions for the M315F/AtSuSy cascade reaction. (A) The influence of temperature on cascade reaction. (B) The influence of UDP concentration on cascade reaction (1 mM PPD). (C) The influence of sucrose
concentration on cascade reaction (UDP: PPD=0.3). (D) The influence of PPD concentration on cascade reaction. Fig. 6 Fed-batch synthesis of ginsenosides Rh2 reaction. 1 mM PPD was added to the reaction mixtures at 1, 2, 4, 6, 8, 10 and 12 h. Fresh enzymes (11.5 mU/mL M315F and 24 mU/mL AtSuSy) were added at 4 h, 8 h and 12 h.
Table 1 Production status of ginsenoside Rh2 synthesized by biological method
Strategies
Enzyme source
Ginsenoside Rh2 yield
Reference
Hydrolysis
Aspergillus niger
~125 mg/L
[31]
Biosynthetic
UGTPg45 from P. ginseng
17 mg/L
[32]
Biosynthetic
Mutant of UGT51 from S.cerevisiae S288c
~300 mg/L
[23]
Enzymatic
Bs-Yjic from Bacillus subtilis 168
200 mg/L
[25]
Enzymatic
UGT from Bacillus subtilis 1.1470
0.77 mg/L
[33]
Enzymatic
Mutant M315F from Bs-Yjic
3700 mg/L
This study
S0141-8130(19)33207-6 https://doi.org/10.1016/j.ijbiomac.2019.09.208 BIOMAC 13504
To appear in:
International Journal of Biological Macromolecules
Received Date: Revised Date: Accepted Date:
30 April 2019 1 July 2019 20 September 2019
Please cite this article as: W. Ma, L. Zhao, Y. Ma, Y. Li, S. Qin, B. He, Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase, International Journal of Biological Macromolecules (2019), doi: https://doi.org/10.1016/j.ijbiomac.2019.09.208
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier B.V.
Oriented efficient biosynthesis of rare ginsenoside Rh2 from PPD by compiling UGT-Yjic mutant with sucrose synthase
Wang Maa, Lu Zhaoa, Yudi Maa, Yuqiang Lia, Song Qina, Bingfang Hea,b,*.
a
College of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University,
No. 30 Puzhu South Road, Nanjing, 211816, China. b
School of Pharmaceutical Sciences, Nanjing Tech University, No. 30 Puzhu South
Road, Nanjing, 211816, China. Corresponding author: Bingfang He E–mail address:[email protected]
Abstract Ginsenoside
Rh2
(3β-O-Glc-protopanaxadiol),
a
trace
but
an
important
pharmacological component of ginseng, has exhibited excellent medicinal potential. Many studies have found that the synthesis of Rh2 by UDP-glucosyltransferase (UGT) is an alternative production strategy. In this study, Yjic from B. subtilis 168 was found to synthesize ginsenoside F12 (3β,12β-Di-O-Glc-protopanaxadiol) and Rh2 at a ratio of 7:3. Yjic regioselectivity toward Rh2 synthesis was successfully improved using a semi-rational design including structure-guided alanine scanning and saturation mutations. As a result, mutant M315F was found to efficiently synthesize Rh2 (~99%) and block the further glycosylation of C12-OH. The circulation of UDPG was achieved by combining M315F with AtSuSy through a cascade reaction. Furthermore, an extraordinarily high yield of Rh2 (3.7 g/L) was attained in an aqueous solvent system with 17% DMSO (v/v) through the fed-batch feeding of PPD. This study presents the high potential for the oriented preparation of ginsenoside Rh2. Key words: Ginsenoside Rh2; UDP-glycosyltransferase; Yjic; Mutant M315F; Cascade reaction
1. Introduction Ginsenosides are the major pharmacologically active components of ginseng (Panax ginseng, C. A. Meyer) [1]. Natural ginsenosides exhibit various pharmacological effects, such as anticancer, antitumor, antistress, antiaging, and anti-inflammatory, which protect the central nervous system and enhance immune system activities [2-4]. Ginsenosides can be classified into the protopanaxadiol (PPD) or protopanaxatriol (PPT) groups depending on the hydroxylation of the C6 site [5]. Ginsenoside Rh2 belongs to the PPD group and is a trace but important saponin isolated from red ginseng [6]. Ginsenoside Rh2 also shows significant anticancer, anti-inflammatory, and antiallergic pharmaceutical activities [7-9]. The long growth periods of ginseng and difficulties in extracting certain ginsenosides has initiated research into artificially synthesizing rare ginsenosides. However, chemical approaches for synthesizing ginsenosides face enormous challenges, such as steric hindrance and acid labile dammarane derivatives [10]. In addition, by-products and low yield rates are obstacles in the development of chemical synthesis [11]. In recent years, synthetic biology has played a unique role in the study of natural products [12-13]. PPD (~ 1.2 g/L) has been creatively produced via a synthetic biology pathway using glucose as the raw material in yeast [14]. This method of production has dramatically solved PPD resourcing issues and has laid a solid foundation for the enzymatic synthesis of ginsenosides. UDP-glycosyltransferases (UGTs) generally transfers glucose from UDP-glucose (UDPG) to various metabolites, such as hormones and secondary metabolites [15-16].
Considerable numbers of UGTs with multiple-site selectivity have been identified [17]. At present, the functions of key UGTs from ginseng for ginsenoside synthesis have been illuminated [18-19]. Compared with plant UGTs, some microbial UGTs show high aglycon promiscuity, high catalytic proficiency, and can be efficiently expressed in engineered bacteria [20-22]. For example, UGT51 from Saccharomyces cerevisiae, which exhibits broad acceptor tolerance and can transfer a glucosyl moiety to the free C3-OH of PPD. Through semi-rational design, UGT51 presents an ~1800-fold enhancement of catalytic efficiency (kcat/Km) for converting PPD to ginsenoside Rh2 in vitro (~300 mg/L) [23]. UDP-glycosyltransferases Yjic (NP389104) from Bacillus subtilis has been used for extensive structurally diverse compounds [24]. Bs-Yjic can transfer a glucosyl moiety to the C3-OH and C12-OH of protopanaxadiol (PPD), and successfully synthesize ginsenosides F12 and Rh2 (~7%) in 12 h [25]. In this study, a glycosyltransferase from Bacillus subtilis 168 was selected after the homologous evolutionary analysis used for Rh2 synthesis (Fig. S1). It was then confirmed that the sequence of amino acids was the same as that of Yjic. In subsequent experiments, attempts were made to obtain a mutant with increasing C3 specificity of PPD by semi-rational design. A cascade reaction system of mutant and sucrose synthase from Arabidopsis thaliana (AtSuSy) was then constructed to alleviate the inhibition of high concentrations of UDP and realize the circulation of UDPG (Fig. 1). Furthermore, the final yield of Rh2 was increased by fed-batch synthesis.
2. Material and methods 2.1 Bacterial strains and substrates The Bacillus subtilis 168 (Bio-82277) strain was purchased from the China Center for Type Culture Collection. Authentic PPD and PPD-type ginsenosides (Rh2 and CK) were purchased from Nanjing Spring & Autumn Biological Engineering Co., Ltd. (Nanjing, China) and dissolved in dimethyl sulfoxide (DMSO) for use. All other chemicals were of analytical grade. 2.2 The molecular cloning, expression, and semi-rational design of the Yjic from Bacillus subtilis 168 The Yjic gene (AAC46318.1) excluding 392 amino acids was amplified from the genomic DNA of B. subtilis 168, using the primer sets Yjic-F (5’-ATGAGATTTCCTT CAATTTT-3’)
and
Yjic-R
(5’-TTAACAACTACCGACGCCAA-3’),
which
incorporated Nco I and Xho I sites, respectively. Recombinant plasmid pET-28a-Yjic was then transformed into E. coli BL21 (DE3) for expression. SWISS-MODEL was employed to conduct structural modeling for the prediction of the substrate binding region of Yjic, and CalG2 (PDB:3iaa) from Calicheamicin was used as a template [26]. 2.3 Site-directed mutagenesis and saturation mutagenesis Alanine scanning mutagenesis, a method of systematic alanine substitution, was particularly useful for the identification of functional residues. The primers used for all chosen residues are shown in Table S1. Then the mutant that increases Rh2
production for saturation mutation was selected, and the mutant plasmid was used as the initial template for saturation mutagenesis. All primers for saturation mutagenesis are shown in Table S2. All mutants were created in the recombinant vector pET-28a-UGT. PCR was conducted with Kodaq 2X PCR MasterMix. The product was subsequently treated with Dpn I for 3 h at 37 °C and transformed into E. coli BL21 cells (DE3). The targeted mutation was confirmed by sequencing. After the induced expression of recombinant E. coli, soluble fractions of the enzymes were obtained by cell sonication. The relative activities were measured using wild type Yjic as the control. 2.4 Analysis of selected mutant properties The effect of pH on the synthesis activity of ginsenoside was assayed at 30 ℃ and 200 rpm in the following buffer solutions (each at 50 mM); citrate buffer (pH 5.0), phosphate buffer (pH 6.0-7.0), Tris-HCl (pH 8.0-9.0), and glycine-NaOH (pH 10.0). The pH stability of the enzymes was examined by measuring residual activity after incubation in each buffer for 2 h at 4 ℃. The results were expressed as the relative conversion rate of the PPD substrate compared to the wild type Yjic. The effect of temperature on the synthesis activity of ginsenoside was tested at different temperatures ranging from 20 ℃ to 60 ℃ at pH 8.0. The temperature stability of the enzyme was determined by measuring the residual activity after preincubation in 50 mM Tris-HCl (pH 8.0) for 6 h at different temperatures (20-60 ℃). The results were expressed as the relative conversion rate of the PPD substrate. 2.5 The effect of the cofactor and organic solvent on enzyme activities
Cofactors, especially metal ions, play an important role in enzymatic reactions [27]. The cofactors for this test include: sodium dodecyl sulfonate (SDS), dithiothreitol (DTT), β-mercaptoethanol (β-ME), K+, Na+, NH4+, Ni2+, Ba2+, Zn2+, Ca2+, Mg2+, Mn2+, and Fe3+. All the ions originate from chloride. The reaction mixture consisted of 500 μL of 50 mM Tris-HCl buffer (pH 8.0) containing PPD (2 mM), UDP-Glucose (4 mM), 10% (v/v) DMSO, and 5 mM cofactors. The mixture was incubated at 30 °C and 200 rpm, and the effects of cofactor concentrations were studied in subsequent experiments. The effects of organic solvents were estimated due to the low solubility of PPD in the water phase. The concentration of the organic solvents in the reaction system was 10% (v/v), and the effects of this concentration were optimized. 2.6 Enzyme activity assays Activity assays (0.5 mL) were conducted with 1 mM PPD, 2 mM UDPG, 50 mM Tris-HCl (pH 8.0), and a certain amount of mutant for 0.5 h at 30 °C. The reactions were terminated by adding two times of methanol (v/v). The reactants were subsequently centrifuged at 12000 rpm for 2 min. The supernatant was then filtered through a 0.22 μm filter, and the ginsenosides and residual PPD were analyzed by HPLC. The standard curve of PPD concentration is shown in Fig. S2. A single unit of enzyme activity was defined as the amount of mutant that glycosylated 1 μmol of PPD per minute. There were three parallel experiments for all operations. Activity assays of AtSuSy were conducted in a 1 mL volume containing 0.5 mM UDP, 300 mM sucrose, 50 mM Tris-HCl (pH 8.0), and various amounts of AtSuSy for
30 min at 30 °C. The reaction was terminated by boiling water bath for 5min. The AtSuSy activity was measured by determining the amount of fructose released from sucrose and analyzed using 3,5-dinitrosalicylic acid colorimetry (DNS) [28]. The standard curve of fructose concentration is shown in Fig. S3. A single unit of enzymatic activity was defined as the amount of enzyme that releases 1 μmol of fructose from sucrose per minute. There were three parallel experiments for all operations. 2.7 Optimization of the mutant/AtSuSy cascade reaction The Mutant/AtSuSy cascade reaction system contained 1 mM PPD, 0.2 mM UDP, 50 mM Tris-HCl (pH 8.0), 400 mM sucrose, 10% DMSO (v/v), 21 mU/mL mutant, and 48 mU/mL AtSuSy. The cascade reaction was incubated at 35°C and 200 rpm for 2 h. The effect of temperature was confirmed using a range of 25-45°C. In addition, the effects of UDP (0.05−0.5 mM), sucrose (0.1−0.8 M), and PPD (0.5−2 mM) were analyzed at different concentrations individually. Finally, the effects of the mutant and AtSuSy ratio was determined. 2.8 Fed-batch synthesis of ginsenosides by the mutant/AtSuSy cascade reaction The reaction mixtures (20 mL) contained 1 mM PPD, 0.3 mM UDP, 50 mM Tris-HCl (pH 8.0), 0.3 M sucrose, 10% DMSO, 5 mM Mg2+, 21 mU/mL Mutant, and 48 mU/mL AtSuSy. The reaction was performed at 35°C and 200 rpm. Next, 200 μL PPD (100 mM) was periodically added to the reaction mixtures at 1, 2, 4, 6, 8, 10, and 12 h and 200 μL of the reactant solution was collected and terminated by adding 3-fold the volume of methanol. After centrifugation at 12000 rpm for 2 min, the
supernatant was filtered through 0.22 μm filters and analyzed by HPLC. Fresh enzymes (11.5 mU/mL mutant and 24 mU/mL AtSuSy) were added at 4, 8 and 12 h. 2.9 HPLC, LC-MS, and NMR analysis of the glycosylated products The
reactant
(20
μL)
was
examined
by
high-performance
liquid
chromatography (HPLC). The ZORBAX SB-C18 reverse-phase column (4.6 × 150 mm, 5 μm particles) connected to an UltiMate 3000 HPLC system with UV detection at 203 nm was eluted with solvent A (water) and solvent B (methanol) using an equal-degree program of 80% B over 55 min. The solvent flow rate was 1.0 mL/min, and the column temperature was set at 30 °C. LC-MS analysis was performed in positive ion mode on an Agilent 6520 Accurate-Mass Q-TOF LC-MS platform (Palo Alto, CA, U.S.A.). The glycosylation products of PPD were purified using an Agilent 1200 preparative HPLC system coupled with a reverse phase Ultimate C18 column (21.2 × 250 mm2, 5 μm particles, Welch, Shanghai, China). The products were then analyzed using the proton and carbon nuclear magnetic resonance (1H NMR,
13
C
NMR) spectrum with samples dissolved in DMSO-d6 (Bruker Avance IIII 400). 3. Results and Discussion 3.1 The selection and semi-rational design of Yjic towards ginsenoside Rh2 synthase A glycosyltransferase (UGT) was selected from B. subtilis through the analysis of protein homology evolution and comparison of the PSPG (Plant Secondary Product Glycosyltransferase) motif (Fig. S1). It was found that this UGT could glycosylate PPD (1 mM) to the unnatural ginsenoside F12 (~69%) and the ginsenoside Rh2 (~31%) after 6 h of conversion (Fig. S4). The catalytic products are consistent with
the Yjic reported (NP389104) [25]. The spectrum of the proton and carbon nuclear magnetic resonance analysis of glycosylated product 1 is shown in Fig S5 and S6. The 1
H-NMR and
13
C-NMR data (Table S3) of product 1 is identical to that of the
3-O-β-D-glucopyranosyl-20(S)-protopanaxadiol (Rh2) [25]. Semi-rational design guided by structural information is usually an effective protein-engineering strategy for modifying catalytic properties or improving catalytic activity such as UGT51 [23].The PPD substrate was docked into the acceptor-binding pocket using the automated docking program Autodock 4.2.6 [29]. Some residues were found to be mutagenesis sites around the substrate binding pocket when the positions of the residues were mapped to the 3D structure of the protein (Fig. S7). 3.2 The site-specific modification of Yjic The alanine scanning mutation on the selected 28 residues nearing the substrate binding site was carried out to find the hotspot that influences the specificity of Yjic. As Fig. 2 shows, changing the majority of selected residues can dramatically change the relative conversion rates and regioselectivity (Rh2/total products) of Yjic. Among these residues, about half of the mutants experienced a promoted relative conversion rate, and the mutant R68A showed a 2.4-times higher conversion rate than the wild-type (WT). The mutants of some residues also significantly improved the regioselectivity of Yjic for Rh2 synthesis. About 70% of the mutants prefer to synthesize Rh2 compared with WT Yjic. Mutants Y14A, F168A, and M315A show a tendency to glycosylate the C3-OH of PPD. Fortunately, the mutant M315A showed nearly 99% regioselectivity (Fig. S8) for Rh2 synthesis. Further studies have found
that M315 residues are located in the PSPG motif, where the amino residues play an essential role in determining the substrate specificity of the enzymes [21]. Saturation mutation at residue 315 was then performed, and the relative conversion rates and regioselectivity of most mutants varied greatly (Fig. 3). The eight mutants exhibited a higher relative conversion rate than the WT Yjic. Surprisingly, ten mutants, M315A, M315F, M315V, M315S, M315E, M315K, M315Y, M315C, M315W, and M315G all exhibited perfect regioselectivity (~99%). Mutant M315F was selected for the following experiments due to both its relative conversion rate and regioselectivity. 3.3 Characterization and preliminary optimization for glycosylation of PPD by mutant M315F The effect of pH on the glycosylation of PPD was assayed. The M315F mutant shows higher activity and stability at pH 8 and could retain more than 90% of its vitality between pH 8 and 9 for two hours (Fig. S9A). The increased activity of M315F was observed at 37 °C. The enzymes also maintained excellent stability in the range of 20-37 °C for six hours, but the activity decreased sharply above 42 °C (Fig. S9B). The metal ions showed the most significant effect on the glycosylation rate (Fig. 4A). Mn2+ and Mg2+ are significantly promoted in Rh2 synthesis. Coincidentally, Dai et al [25]. report that Mg2+ strengthens the glycosylation of PPD by Bs-Yjic. The relative conversion rate of the mutant M315F was elevated by 1.6-times compared with the addition of 8 mM Mg2+ (Fig. 4B).
The effects of organic solvents on the glycosylation of PPD were analyzed due to its extremely low solubility. The addition of hydrophilic organic solvents (10%, v/v) promoted the dissolution of the PPD substrate and facilitated the catalytic reaction (Fig. 4C). However, the addition of hydrophobic organic solvents hindered the catalytic reaction. This result may have been caused by the presence of enzymes and substrates in different solvent layers. DMSO, DMF, and MeOH increased by more than twice compared with the initial conversion rate, with DMSO showing the most significant increase. The effects of solvent concentration on the glycosylation rates were further investigated. DMSO at a specific concentration (5%~20%, v/v) is beneficial for the glycosylation of PPD (Fig. 4D). The effects of UDPG on ginsenoside synthesis by mutant M315F were also studied. The production rate of Rh2 increased continuously with increasing UDPG concentration, and finally reached 1.08 mM (Fig. S10). The increased concentration of UDPG was found to be beneficial to the synthesis of ginsenoside F12 (Fig. S11). On the other hand, the conversion rate of total ginsenosides decreased when a higher substrate ratio (UDPG∶ PPD) over four was used. 3.4 Glycosylation of PPD using cascade reactions of M315F and AtSuSy A coupling reaction recombining AtSuSy and the mutant M315F was constructed to create in situ UDPG regeneration and reduce the accumulation of UDP in the reaction system. This reaction was used to synthesize ginsenoside Rh2 using PPD and sucrose as the aglycon and sugar donor, respectively. The ratio of M315F/AtSuSy on the synthesis of ginsenoside Rh2 was preliminarily optimized
(Table S4). The activity ratio 1:2.1 of M315F/AtSuSy (mU/mL) was selected due to the conversion rate of PPD and the higher proportion of Rh2 in total ginsenosides. A higher conversion of PPD was achieved at 35 °C through the elementary optimization of cascade catalysis conditions (Fig. 5A). UDP is used for UDPG supplementation in the UGT-AtSuSy cascade reaction, while UDP at higher concentrations can inhibit UGT activity [30]. The conversion of PPD increased by 1.72-fold when the ratio of UDP:PPD increased from 0.05 to 0.3 (1 mM PPD) and slightly decreased when the concentration of UDP increased further (Fig. 5B). An initial concentration of 0.3 M sucrose was selected for cascade catalysis to obtain efficient in situ UDPG regeneration (Fig. 5C). In addition, initial PPD concentration had a significant impact on the reaction. The glycosylation rate (Fig. 5D) gradually reduced as the initial concentration of PPD increased. Finally, 1 mM PPD at the initial concentration was selected for the catalytic reaction. The conditions for the optimization of the fed-batch cascade reaction were chosen as 35 oC, 0.3 mM UDP, 0.3 M sucrose, and 1 mM PPD. 3.5 Ginsenoside Rh2 preparation using fed-batch cascade reactions Ginsenoside Rh2 preparation was carried out via the fed-batch feeding of PPD based on the preliminary optimization of reaction conditions and the inhibition of high concentrations of PPD. The glycosylation of PPD occurred relatively quickly, at about 0.44 mM ginsenoside Rh2 from 1 mM PPD after 0.5 h of the reaction (Fig. 6). Only a trace amount of ginsenoside F12 was detected in the first 2 h, and ginsenoside Rh2 was the primary product, which is consistent with the in vitro enzyme reaction with
UDPG as the substrate. After 24 h, 5.92 mM (3.7 g/L) ginsenoside Rh2 and 0.64 mM ginsenoside F12 (0.5 g/L) were obtained through the periodic feeding of PPD with a total conversion yield of 82% (Fig. S12) in systems with 17% DMSO. The production intensity of Rh2 reached 0.154 g/h·L and benefited from the high efficiency of the mutant and DMSO aqueous reaction system to facilitate PPD dissolution and the glycosylation rate, the residual DMSO can be effectively controlled in the post-processing of the reaction. To our knowledge, the synthetic capacity of ginsenoside Rh2 in the present study was much higher than is reported [18,23]. The production of PPD via a yeast cell factory reduces the cost of PPD substrates and makes the production of high-value ginsenoside Rh2 by M315F-AtSuSy cascade reaction in vitro more commercially valuable [1]. 4. Conclusion In summary, the mutant M315F of Yjic from Bacillus subtilis 168 was found to efficiently synthesize Rh2 (~99%) by semi-rational design. Furthermore, the M315F/AtSuSy cascade reaction was constructed which reduces costs through the cyclic regeneration of UDPG and relieves the inhibition of UDP at high concentrations. Finally, fed-batch feeding of PPD was used to increase the yield of ginsenoside Rh2. This method resulted in the achievement of 5.92 mM (3.7 g/L) of Rh2 in DMSO aqueous solution with a final DMSO concentration of about 17%, which shows the practical potential of ginsenoside Rh2 production (Table 1).
Acknowledgements
This work was financially supported by the National Natural Science Foundation of China (81673321, 21776135, 21506099). We also thank the Jiangsu Synergetic Innovation Center for Advanced Bio-Manufacture (NO. XTC1812).
Reference [1] Z.B. Dai, B.B. Wang, Y. Liu, M.Y. Shi, D. Wang, X.N. Zhang, T. Liu, L.Q. Huang, X.L. Zhang, Producing aglycons of ginsenosides in bakers' yeast, Scientific Reports, 4 (2014) 6. [2] L. Li, S.-Y. Shin, S.J. Lee, J.S. Moon, W.T. Im, N.S. Han, Production of Ginsenoside F2 by Using Lactococcus lactis with Enhanced Expression of beta-Glucosidase Gene from Paenibacillus mucilaginosus, Journal of agricultural and food chemistry, 64 (2016) 2506-2512. [3] J. Pei, A. Chen, L. Zhao, F. Cao, G. Ding, W. Xiao, Enzymatic transformation of ginsenoside Rb1 to ginsenoside 20(S)-Rg3 by GH3 β-glucosidase from Thermotoga thermarum DSM 5069 T, Journal of agricultural and food chemistry, 65 (2017) 6042-6048. [4] J.Y. Seo, S.H. Ju, J. Oh, S.K. Lee, J.-S. Kim, Neuroprotective and Cognition-Enhancing Effects of Compound K Isolated from Red Ginseng, Journal of agricultural and food chemistry, 64 (2016) 2855-2864. [5] L.P. Christensen, Ginsenosides chemistry, biosynthesis, analysis, and potential
health effects, Advances in food and nutrition research, 55 (2009) 1-99. [6] X. Jin, D.B. Che, Z.H. Zhang, H.M. Yan, Z.Y. Jia, X.B. Jia, Ginseng consumption and risk of cancer: A meta-analysis, Journal of ginseng research, 40 (2016) 269-277. [7] M.K. Choo, E.K. Park, M.J. Han, D.H. Kim, Antiallergic activity of ginseng and its ginsenosides, Planta Medica, 69 (2003) 518-522. [8] G.-w. Liu, Y.-h. Liu, G.-s. Jiang, W.-d. Ren, The reversal effect of Ginsenoside Rh2 on drug resistance in human colorectal carcinoma cells and its mechanism, Human cell, 31 (2018) 189-198. [9] X.W. Xie, A. Eberding, C. Madera, L. Fazli, W. Jia, L. Goldenberg, M. Gleave, E.S. Guns, Rh2 synergistically enhances paclitaxel or mitoxantrone in prostate cancer models, Journal Of Urology, 175 (2006) 1926-1931. [10] J. Liao, J. Sun, Y. Niu, B. Yu, Synthesis of ginsenoside Rh2 and chikusetsusaponin-LT8
via
gold(I)-catalyzed
glycosylation
with
a
glycosyl
ortho-alkynylbenzoate as donor, Tetrahedron Letters, 52 (2011) 3075-3078. [11] J. Yu, J. Sun, Y. Niu, R. Li, J. Liao, F. Zhang, B. Yu, Synthetic access toward the diverse ginsenosides, Chemical Science, 4 (2013) 3899-3905. [12] Y.J. Kim, D. Zhang, D.C. Yang, Biosynthesis and biotechnological production of ginsenosides, Biotechnol Adv, 33 (2015) 717-735. [13] C. Zhao, X. Gao, X. Liu, Y. Wang, S. Yang, F. Wang, Y. Ren, Enhancing Biosynthesis of a Ginsenoside Precursor by Self-Assembly of Two Key Enzymes in Pichia pastoris, Journal of agricultural and food chemistry, 64 (2016) 3380-3385. [14] Z. Dai, Y. Liu, X. Zhang, M. Shi, B. Wang, D. Wang, L. Huang, X. Zhang,
Metabolic engineering of Saccharomyces cerevisiae for production of ginsenosides, Metabolic Engineering, 20 (2013) 146-156. [15] B.K. Hou, E.K. Lim, G.S. Higgins, D.J. Bowles, N-glucosylation of cytokinins by glycosyltransferases of Arabidopsis thaliana, Journal Of Biological Chemistry, 279 (2004) 47822-47832. [16] W. Wei, P. Wang, Y. Wei, Q. Liu, C. Yang, G. Zhao, J. Yue, X. Yan, Z. Zhou, Characterizations
of
Panax
ginseng
UDP-glycosyltransferases
catalyzing
protopanaxatriol and biosyntheses of bioactive ginsenosides F1 and Rh1 in metabolically engineered yeasts, Molecular Plant, 8 (2015) 1412-1424. [17] J.A. Campbell, G.J. Davies, V. Bulone, B. Henrissat, A classification of nucleotide-diphospho-sugar glycosyltransferases based on amino acid sequence similarities, Biochemical Journal, 326 (1997) 929-939. [18] S.-C. Jung, W. Kim, S.C. Park, J. Jeong, M.K. Park, S. Lim, Y. Lee, W.-T. Im, J.H. Lee, G. Choi, S.C. Kim, Two Ginseng UDP-Glycosyltransferases Synthesize Ginsenoside Rg(3) and Rd, Plant And Cell Physiology, 55 (2014) 2177-2188. [19] Y. Xing, F. Yun, W. Wei, W. Pingping, L. Qunfang, W. Yongjun, Z. Lei, Z. Guoping, Y. Jianmin, Z. Zhihua, Production of bioactive ginsenoside compound K in metabolically engineered yeast, Cell Research, 24 (2014) 770-773. [20] H.-H. Chiu, M.-Y. Shen, Y.-T. Liu, Y.-L. Fu, Y.-A. Chiu, Y.-H. Chen, C.-P. Huang, Y.-K. Li, Diversity of sugar acceptor of glycosyltransferase 1 from Bacillus cereus and its application for glucoside synthesis, Applied microbiology and biotechnology, 100 (2016) 4459-4471.
[21] B. Fan, T. Chen, S. Zhang, B. Wu, B. He, Mining of efficient microbial UDP-glycosyltransferases by motif evolution cross plant kingdom for application in biosynthesis of salidroside, Scientific Reports, 7 (2017). [22] R.P. Pandey, P. Parajuli, N. Koirala, J.W. Park, J.K. Sohng, Probing 3-Hydroxyflavone for In Vitro Glycorandomization of Flavonols by YjiC, Applied And Environmental Microbiology, 79 (2013) 6833-6838. [23] Y. Zhuang, G.-Y. Yang, X. Chen, Q. Liu, X. Zhang, Z. Deng, Y. Feng, Biosynthesis of plant-derived ginsenoside Rh2 in yeast via repurposing a key promiscuous microbial enzyme, Metabolic Engineering, 42 (2017) 25-32. [24] L. Dai, J. Li, P. Yao, Y. Zhu, Y. Men, Y. Zeng, J. Yang, Y. Sun, Exploiting the aglycon promiscuity of glycosyltransferase Bs-YjiC from Bacillus subtilis and its application in synthesis of glycosides, Journal Of Biotechnology, 248 (2017) 69-76. [25] L. Dai, C. Liu, J. Li, C. Dong, J. Yang, Z. Dai, X. Zhang, Y. Sun, One-Pot Synthesis of Ginsenoside Rh2 and Bioactive Unnatural Ginsenoside by Coupling Promiscuous Glycosyltransferase from Bacillus subtilis 168 to Sucrose Synthase, Journal of agricultural and food chemistry, 66 (2018) 2830-2837. [26] A. Chang, S. Singh, K.E. Helmich, R.D. Goff, C.A. Bingman, J.S. Thorson, G.N. Phillips, Jr., Complete set of glycosyltransferase structures in the calicheamicin biosynthetic pathway reveals the origin of regiospecificity, Proceedings Of the National Academy Of Sciences Of the United States Of America, 108 (2011) 17649-17654. [27] L. Wang, Q.-M. Liu, B.-H. Sung, D.-S. An, H.-G. Lee, S.-G. Kim, S.-C. Kim,
S.-T. Lee, W.-T. Im, Bioconversion of ginsenosides Rb1, Rb2, Rc and Rd by novel β-glucosidase hydrolyzing outer 3-O-glycoside from Sphingomonas sp. 2F2: Cloning, expression, and enzyme characterization, Journal Of Biotechnology, 156 (2011) 125-133. [28] Y. Jian-Xiong, L.U. Hong, X. Qiang, D. Jin-Ling, L.I. Ni-Hang, A study on rapid colorimetric determination of water soluble total sugar,reducing sugar and starch in tobacco with 3,5-dinitrosalicylic acid, Journal of Yunnan Agricultural University, (2007). [29] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, AutoDock4 and AutoDockTools4: Automated Docking with Selective Receptor Flexibility, Journal Of Computational Chemistry, 30 (2009) 2785-2791. [30] A. Gutmann, B. Nidetzky, Unlocking the Potential of Leloir Glycosyltransferases for Applied Biocatalysis: Efficient Synthesis of Uridine 5'-Diphosphate-Glucose by Sucrose Synthase, Advanced Synthesis & Catalysis, 358 (2016) 3600-3609. [31] L. Liu, X.-M. Zhu, Q.-J. Wang, D.-L. Zhang, Z.-M. Fang, C.-Y. Wang, Z. Wang, B.-S. Sun, H. Wu, C.-K. Sung, Enzymatic preparation of 20(S, R)-protopanaxadiol by transformation of 20(S, R)-Rg3 from black ginseng, Phytochemistry, 71 (2010) 1514-1520. [32] P. Wang, Y. Wei, Y. Fan, Q. Liu, W. Wei, C. Yang, L. Zhang, G. Zhao, J. Yue, X. Yan, Production of bioactive ginsenosides Rh2 and Rg3 by metabolically engineered yeasts, Metabolic Engineering, 29 (2015) 97-105. [33] L. Yu, Y. Chen, J. Shi, R. Wang, Y. Yang, L. Yang, S. Zhao, Z. Wang,
Biosynthesis of rare 20(R)-protopanaxadiol/protopanaxatriol type ginsenosides through Escherichia coli engineered with uridine diphosphate glycosyltransferase genes, Journal of ginseng research, 43 (2019) 116-124.
Fig. 1 Schematic diagram of PPD glycosylation reaction. The solid line represents the reaction pathway constructed in this experiment, the dotted line represents the reaction pathway catalyzed by Yjic. Fig. 2 Relative activities of alanine-substituted mutants of Bs-Yjic (Regioselectivity = Rh2 / total products). Reaction system (1 mL): 1 mM PPD, 2 mM UDPG, 10% DMSO, pH 8.0 Tris-HCl (50 mM). Reactions end at 6 h. Fig. 3 Relative activity of the saturated mutants at M315 (M = Methionine). Reaction system (1 mL): 1 mM PPD, 2 mM UDPG, 10% DMSO, pH 8.0 Tris-HCl (50 mM). Reactions end at 6 h. Fig. 4 Catalytic properties of M315F mutant under different conditions. (A) and (B) Effects of cofactors and ion concentration (Mg2+, Mn2+, Ca2+) on the activity of M315F. (C) and (D) Effects of solvents and concentration (DMF, DMSO, MeOH) on the activity of M315F. Fig. 5 Optimization of the reaction conditions for the M315F/AtSuSy cascade reaction. (A) The influence of temperature on cascade reaction. (B) The influence of UDP concentration on cascade reaction (1 mM PPD). (C) The influence of sucrose
concentration on cascade reaction (UDP: PPD=0.3). (D) The influence of PPD concentration on cascade reaction. Fig. 6 Fed-batch synthesis of ginsenosides Rh2 reaction. 1 mM PPD was added to the reaction mixtures at 1, 2, 4, 6, 8, 10 and 12 h. Fresh enzymes (11.5 mU/mL M315F and 24 mU/mL AtSuSy) were added at 4 h, 8 h and 12 h.
Table 1 Production status of ginsenoside Rh2 synthesized by biological method
Strategies
Enzyme source
Ginsenoside Rh2 yield
Reference
Hydrolysis
Aspergillus niger
~125 mg/L
[31]
Biosynthetic
UGTPg45 from P. ginseng
17 mg/L
[32]
Biosynthetic
Mutant of UGT51 from S.cerevisiae S288c
~300 mg/L
[23]
Enzymatic
Bs-Yjic from Bacillus subtilis 168
200 mg/L
[25]
Enzymatic
UGT from Bacillus subtilis 1.1470
0.77 mg/L
[33]
Enzymatic
Mutant M315F from Bs-Yjic
3700 mg/L
This study