Near-perfect kinetic resolution of racemic p-chlorostyrene oxide by SlEH1, a novel epoxide hydrolase from Solanum lycopersicum with extremely high enantioselectivity

Journal Pre-proofs Near-perfect kinetic resolution of racemic p-chlorostyrene oxide by SlEH1, a novel epoxide hydrolase from Solanum lycopersicum with extremely high enantioselectivity Bo-Chun Hu, Die Hu, Chuang Li, Xiong-Feng Xu, Zheng Wen, Min-Chen Wu PII: DOI: Reference:

S0141-8130(19)36202-6 https://doi.org/10.1016/j.ijbiomac.2019.10.091 BIOMAC 13585

To appear in:

International Journal of Biological Macromolecules

Received Date: Revised Date: Accepted Date:

6 August 2019 8 October 2019 8 October 2019

Please cite this article as: B-C. Hu, D. Hu, C. Li, X-F. Xu, Z. Wen, M-C. Wu, Near-perfect kinetic resolution of racemic p-chlorostyrene oxide by SlEH1, a novel epoxide hydrolase from Solanum lycopersicum with extremely high enantioselectivity, International Journal of Biological Macromolecules (2019), doi: https://doi.org/10.1016/ j.ijbiomac.2019.10.091

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.

Near-perfect kinetic resolution of racemic p-chlorostyrene oxide by SlEH1, a novel epoxide hydrolase from Solanum lycopersicum with extremely high enantioselectivity

Bo-Chun Hu a,1, Die Hu b,c,1, Chuang Li a, Xiong-Feng Xu a, Zheng Wen a, Min-Chen Wub,*

a

Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of

Biotechnology, Jiangnan University, Wuxi 214122, PR China b

Wuxi School of Medicine, Jiangnan University, Wuxi 214122, PR China

c

School of Food Science and Technology, Jiangnan University, Wuxi 214122, PR China

1

Bo-Chun Hu and Die Hu, the two first authors, contributed equally to this work.

*

Corresponding Author E-mail address: [email protected] (M.-C. Wu).

1

ABSTRACT An open reading frame of sleh1, a gene encoding for a novel epoxide hydrolase from Solanum lycopersicum (SlEH1), was amplified by RT-PCR and expressed in E. coli BL21(DE3). The substrate spectrum assay showed that E. coli/sleh1 had EH activities towards all tested substrates except for racemic (rac-) 5a, and the highest enantiomeric ratio (E > 200) towards rac-2a, retaining (R)-2a with 99.1% ees and 49.2% yields and affording (R)-2b with 89.8% eep and 46.7% yieldp. Besides, E. coli/sleh1 also hydrolyzed of rac-7a–9a with moderate regioselectivities, producing (S)- or (R)-7b–9b with 40.551.3% eep and 69.475.2% yieldp. The pH optimum and stability of the purified SlEH1 were 7.5 and at a range of 6.58.5, and it was thermostable at or below 40 °C. Its catalytic efficiency (kcatS/KmS = 7.49 mM-1 s-1) for (S)-2a was much higher than that for (R)-2a. The gramscale kinetic resolution of 150 mM rac-2a was carried out by E. coli/sleh1 at 20 °C for 8 h, producing (R)-2a with 98.2% ees and 45.3% overall yields after purification by silica gel column chromatography. Furthermore, the source of extremely high enantioselectivity of SlEH1 towards rac-2a was analyzed by molecular docking simulations. Keywords:

Solanum lycopersicum; Epoxide hydrolase; Enantioselectivity; Kinetic resolution; p-Chlorostyrene

oxide

2

1. Introduction Epoxide hydrolases (EHs, EC 3.3.2.-), which are cofactor-independent biocatalysts and ubiquitously exist in organisms, can catalyze the conversion of epoxides into their corresponding vicinal diols through the addition of water molecule into oxirane ring [1]. The vast majority of characterized EHs are classified into α/β-hydrolase fold superfamily, which share an α/β domain, that is, a β-sheet surrounded by a cluster of α-helices, and a cap domain harboring a variable cap-loop. Both the α/β and cap domains are connected by a variable NC-loop in residue composition and chain length [2]. Based on the catalytic mechanisms of given EH-epoxide pairs, the asymmetric conversion of racemic (rac-) epoxides was divided into two main pathways: kinetic resolution and enantioconvergent hydrolysis [3]. The former preferentially hydrolyzes one epoxide enantiomer, retaining the other one with an intrinsic limitation of 50% yields, while the latter enantioconvergently hydrolyzes racepoxides, affording enantiopure diols up to 100% theoretical yieldp [4]. Along with the green waves of global industrialization, the hydrolysis of rac-epoxides using resting cells or EHs was recognized as an alternative or complement to chemocatalysis that required expensive chiral ligands and hazardous metals [5] for producing chiral epoxides and diols. Enantiopure epoxides and diols, highly value-added and versatile building blocks, have been widely applied in pharmaceutical industry. For examples, (R)-p-chlorostyrene oxide (2a) was used for the synthesis of (R)-Eliprodil, a neuroprotective agent for the treatment of ischemic stroke and (R)-1,2epoxyoctane (9a) for a therapeutic agent, ONO-2506, of Parkinson’s disease [6,7]. To date, numerous EHs have been isolated and characterized from microorganisms, plants and mammals, while various EH-encoding genes cloned, heterologously expressed and modified [8]. The near-perfect kinetic resolution of rac-epoxides by single EHs with extremely high enantio- and regioselectivity can simultaneously produce epoxides and their corresponding diols with high enantiomeric excess (ee) values and yield approaching 3

ultimate value of 50% [9]. Previously, the asymmetric hydrolysis of 20 mM rac-styrene oxide (1a) using resting cells of E. coli transformant expressing PvEH2 was carried out, retaining (R)-1a with over 99.5% ees and 49.4% yields while producing (R)-phenyl-1,2-ethanediol (1b) with 96.2% eep and 49.7% yieldp [10]. However, considering the diversity of the epoxides, few EHs reported, so far, can catalyze all types of epoxides optimally. For example, a Vigna radiata EH (VrEH3) was used for the kinetic resolution of rac-phenyl glycidyl ether (6a), retaining (R)-6a with over 99% ees but affording (R)-6b with very low eep [11]. To provide more options for the perfect production of chiral epoxides and/or diols, it is necessary to modify some local structures of existing EHs by protein engineering or to find novel EHs with superior catalytic performance. In our previous studies, several EH genes, such as Aueh2 (GenBank no. KF061095) and pveh2 (MF409201), were cloned and heterologously expressed, while EH modification and application were also carried out [10–12]. In this work, a novel SlEH1-encoding gene (sleh1, MH553384) was amplified from Solanum lycopersicum total RNA by RT-PCR, and expressed in E. coli BL21(DE3). The substrate spectrum assay of E. coli/sleh1 towards nine rac-epoxides was carried out (Fig. 1), and enzymatic properties of the purified SlEH1 were measured. Then, the gram-scale kinetic resolution of rac-2a was carried out using the resting cells of E. coli/sleh1. Furthermore, the molecular mechanism of SlEH1 with extremely high enantioselectivity towards rac-2a was analyzed by molecular docking simulations.

2. Materials and methods 2.1. Materials Immature S. lycopersicum was obtained from a local farm (Wuxi, China) for total RNA extraction. All enzymes for gene manipulation were purchased from TaKaRa (Dalian, China). E. coli JM109 and plasmid pUCm-T (Sangon, Shanghai, China) were used for EH gene cloning and sequencing, while E. coli BL21(DE3) 4

and pET-28a(+) (Novagen, Madison, WI) for the expression of sleh1. Rac-1a, 6a, 8a and 9a were from Energy Chemical (Shanghai, China), while rac-2a–5a, 7a, (S)-2a and (R)-2a were synthesized by our lab (Fig. S1). 2.2. Cloning and expression of sleh1 S. lycopersicum total RNA was isolated using a UNIQ-10 column Trizol total RNA isolation kit (Sangon, Shanghai, China) according to the manufacturer’s instructions. A pair of PCR primers, SlEH1-F and SlEH1-R (Table S1), was designed based on the hypothetical EH-encoding mRNA of S. lycopersicum (XM_004239757) searched at NCBI website (https://ncbi.nlm.nih.gov/) by BLAST server. The first-strand cDNAs were reversely transcribed from the total RNA using a PrimeScript RT-PCR kit (TaKaRa, Dalian, China), from which the open reading frame (ORF) of sleh1, flanked by BamH I and Xho I sites, was PCR-amplified using SlEH1-F/-R, and ligated with pUCm-T, followed by DNA sequencing. Then, the multiple sequence alignment of SlEH1, which was deduced from ORF, with other six plant-derived EHs and the confirmation of their conserved motifs were performed using the ClustalW (https://www.genome.jp/tools-bin/clustalw) and ESPript 3.0 (http://espript.ibcp.fr/). The correct sleh1 was excised from a recombinant plasmid, pUCm-T-sleh1, by digestion with BamH I and Xho I, and inserted into pET-28a(+) digested with the same enzymes, followed by transforming it into E. coli BL21, thereby constructing an E. coli transformant, designated E. coli/sleh1. A single colony of E. coli/sleh1 was inoculated into LB medium harboring 100 μg/mL kanamycin, and cultured at 37 °C for 1214 h as the seed culture. Then, the same fresh medium was inoculated with 1% (v/v) seed culture, and cultured until OD600 reached 0.60.8. After induced by 0.05 mM IPTG at 25 °C for 8 h, the E. coli/sleh1 cells were collected and resuspended in 50 mM Na2HPO4-NaH2PO4 buffer (pH 7.0) to 200 mg wet cells/mL used as whole-cell biocatalyst. Comparatively, E. coli BL21(DE3) transformed with pET-28a(+), designated E. coli/pET-28a, was used as a negative control. 2.3. Enzyme activity assay 5

The hydrolytic conditions of substrate for SlEH1 activity assay were as follows: 450 μL cell suspension of E. coli/sleh1 or purified SlEH1 solution, suitably diluted with 50 mM phosphate buffer (pH 7.0), was mixed with 50 μL 200 mM rac-1a, incubated at 20 °C for 10 min, and 200 μL reaction sample was extracted with 1 mL ethyl acetate. Similarly, SlEH1 activities towards rac-2a–9a were handled by substituting rac-1a with them, respectively. The samples were subsequently analyzed by high-performance liquid chromatography (HPLC) or gas chromatography (GC) (Table S2). One unit (U) of SlEH1 activity was defined as the amount of whole wet cells or purified enzyme hydrolying 1 μmol substrate per minute under the given assay conditions. 2.4. Substrate spectrum assay Hydrolytic reactions, in 4 mL 50 mM phosphate buffer (pH 7.0) systems containing 10–20 mM rac-1a–9a and 20–100 mg wet cells/mL of E. coli/sleh1, were carried out, respectively, at 20 °C. During the reaction course, aliquots of 100 μL sample were periodically drawn out, extracted with 1 mL ethyl acetate and analyzed by HPLC or GC. The absolute configurations of enantiomers 1a–9a and 1b–9b were confirmed by comparing the retention times with those reported previously, respectively [6,13–21]. The conversion ratio (c value) of racsubstrate was calculated based on its depleted concentration. The ees of retained single epoxide and eep of produced chiral diol were calculated using equations: ees = [(Rs  Ss) / (Rs  Ss)] × 100% and eep = [(Rp  Sp) / (Rp  Sp)] × 100%. Rs and Ss were the concentrations of (R)- and (S)-epoxide, while Rp and Sp the concentrations of (R)- and (S)-diol. EH enantioselectivity, quantitatively described by its enantiomeric ratio (E value), was used to evaluate the degree of preferential hydrolysis of one enantiomer to its antipode, and was calculated: E  ln [(1  c)  (1  ees)] / ln [(1  c)  (1  ees)] [20]. The EH regioselectivity coefficients, αS and βR, were applied to quantitatively evaluate the preference attacking on Cα (a more hindered carbon in an oxirane ring) of (S)-enantiomer and on Cβ (a less hindered terminal carbon) of (R)-form, respectively [21]. In this work, the αS or βR of SlEH1 for (S)- or 6

(R)-2a was directly calculated based on concentration ratio of its corresponding produced (R)- and (S)-2b [22], while the αS or βR values for other (S)- or (R)-epoxides were derived by linear regression: eep  (αS  βR  1)  [(βR  αS)  ees  (1  c)] / c [6]. 2.5. Purification of expressed SlEH1 The induced and collected E. coli/sleh1 cells were resuspended in binding buffer (20 mM Tris-HCl, 500 mM NaCl and 50 mM imidazole, pH 7.5) to 100 mg wet cells/mL, disrupted by sonication in an ice-water bath, and then purified by affinity chromatography using the nickel-nitrilotriacetic acid (Ni-NTA) column (Tiandz, Beijing, China) [12]. Aliquots of eluent containing the target protein were pooled, dialyzed against 20 mM phosphate buffer (pH 7.0) and concentrated. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) was conducted on a 12% agarose gel. The isolated proteins were visualized by staining with Coomassie Brilliant Blue R-250, and their apparent molecular weights were estimated using Quantity One software based on the standard marker proteins. The protein concentration was measured using the BCA-200 protein assay kit (Pierce, Rockford, IL). 2.6. Enzymatic properties of purified SlEH1 The optimal pH of purified SlEH1 was investigated under the standard enzyme activity assay conditions, except for 20 mM rac-2a in varied pH value buffers (50 mM Na2HPO4-citric acid buffer: pH 5.07.0 and TrisHCl buffer: pH 7.58.5). To evaluate pH stability, aliquots of SlEH1 solution were incubated at 20 °C for 1 h, in the absence of substrate, at pH 5.08.5. Its pH stability here was defined as a pH range, over which the residual SlEH1 activity retained more than 85% of its original activity. For estimating the thermostability of SlEH1, aliquots of EH solution were treated, at optimal pH, at temperatures ranging from 15 to 45 °C for 1 h. Its temperature stability was defined as the temperature, at or below which the residual activity was over 85%. The hydrolytic rate (mol/min/mg protein) of (S)- or (R)-2a catalyzed by SlEH1 was measured under the 7

standard EH activity assay conditions at pH 7.5, except for concentrations of substrate ranging from 2.0 to 20 mM. Both Km and Vmax values were calculated by non-linear regression analysis using Origin 9.0 software (http://www.orignlab.com/). The turnover number (kcat) of SlEH1 was deduced from its Vmax and apparent molecular weight, while the ratio of kcat to Km was defined as the catalytic efficiency. 2.7. Gram-scale kinetic resolution of rac-2a The kinetic resolution of rac-2a, in the 50 mL phosphate buffer (50 mM, pH 7.5) system containing 150 mM rac-2a and 200 mg wet cells/mL of E. coli/sleh1, was carried out at 20 °C. During the hydrolytic course, aliquots of 50 μL sample solution were periodically drawn out, extracted with 950 μL ethyl acetate, and analyzed by chiral HPLC to monitor the c of rac-2a and ees of (R)-2a. When the ees reached over 98%, the total reaction solution was extracted with 25 mL n-hexane twice. The n-hexane fractions were pooled, dried over anhydrous sodium sulphate, and purified by silica gel column chromatography, followed by concentrating under reduced pressure. 2.8. Molecular docking simulations Selecting the known crystal structure of Solanum tuberosum EH (StEH, PDB: 2CJP) as a template, sharing 88.6% primary structure identity with SlEH1, the three-dimensional (3-D) structure of SlEH1 was homologically modeled using the MODELLER 9.21 program (https://salilab.org/modeller/), and then subjected to molecular dynamics simulation using the CHARMM27 force in GROMACS 4.5 package (https://www.gromacs.org/) [23]. Meanwhile, the 3-D structure of (S)- or (R)-2a was handled using ChemBio3D Ultra 14.0 software (http://www.cambridgesoft.com/). The interaction between 3-D structures of SlEH1 and (S)- (or (R)-2a) was predicted using the AutoDock 4.2 program (https://autodock.scripps.edu/). In detail, firstly, the rotatable bond in (S)- or (R)-2a was automatically identified and all the catalytic residues were set as flexible residues. Docking was run and 10 enzyme-substrate complexes were obtained (Table S3). Then, considering the binding energy, a 8

representative complex was selected to optimize by GROMACS 4.5 package to locate the most suitable binding sites and steric orientations [11]. The GRPMOS 96 force field and solutions encapsulated by 1.5 nm SPC/E water were used for each complex. Counter ions Na+ and Cl− were added at random positions into the solvent to neutralize the system. Following this, each complex being fixed and unfixed was energy minimized by 500 step steepest descent and 1000 steps conjugate gradient algorithms, respectively. Subsequently, each energyminimized complex was subjected to a 100 ps equilibration stage and 2 ns molecular dynamics simulation processes at a temperature of 300 K. Finally, based on the conformation of SlEH1 docking with (S)- (or (R)-2a), the through-space distances (d and dβ) between nucleophilic oxygen atom of Asp105 side-chain and C and Cβ of the oxirane ring as well as the hydrogen bond lengths (l1 and l2) from hydroxyl groups of two proton donors (Tyr154 and Tyr235) to O-atom of the oxirane ring were measured using PyMol software (http://pymol.org/).

3. Results and discussion 3.1. Analysis of the primary structure of SlEH1 The ORF of sleh1 was amplified by RT-PCR consisting of 963 bp and encoding 321 amino acids (aa) of SlEH1 (AXM43803) which sharing over 50% identities with other six plant EHs: StEH (AAA81891, 88.6%), PvEH2 (ASS33914, 58.6%), VrEH1 (ADP68585, 57.6%), VrEH2 (AIJ27456, 57.5%), PvEH1 (AKJ75509, 56.9%) and GmEH (CAA55294, 53.2%). The multiple sequence alignment indicated that SlEH1 containing two feature regions of α/β-hydrolase: one was the SmXNuXSmSm (residues 103–108 in SlEH1) motif [24], where X, Sm and Nu represented any residue, small residue and nucleophilic residue, respectively, and the other one was the catalytic triad of Asp105-Asp265-His300. Furthermore, it contained two typical conserved motifs of EHs (Fig. 2): one was the HGXP (residues 31–34) motif forming an oxyanion hole by the X residue to stabilize the negative charge of the nucleophilic Asp105 during hydrolysis and the other one was GXSmXS/T (residues 61–65) 9

motif which function was unknown [25,26]. 3.2. Expression of SlEH1 in E. coli/sleh1 After the E. coli/sleh1 transformants were induced by 0.05 mM IPTG at 25 °C for 8 h, the EH activity of E. coli/sleh1 resting cells towards rac-1a was assayed to be 5.1 U/g wet cells, while no EH activity in E. coli/pET28a under the same conditions. The SDS-PAGE analysis results indicated that the SlEH1 was expressed as a soluble form in E. coli cells (Fig. S2, lane 2) and its apparent molecular weight was 40.1 kDa, which was in accordance with the predicted value (40,974 Da). To make better understanding of the applications of E. coli/sleh1 resting cells towards different substrates, a range of epoxides (rac-1a9a) were employed for the enzyme activity assay. As shown in Table 1, SlEH1 possessed significant differerces in catalytic activities towards these substrates (0.5 to 28.8 U/g wet cells), except for rac-5a for which no activity was observed. 3.3. Enantioselectivity of SlEH1 towards rac-1a–4a The substrate spectrum assay showed that SlEH1 exhibited high enantiomeric ratio towards rac-1a–4a (E values from 20.2 to > 200) affording (R)-1a–4a with ees > 99 %, yields from 37.5 to 49.2% and their corresponding (R)-1b–4b with eep from 65.0 to 89.8% (Table 1, Fig. S3A–D). All these results approved that the EH enantioselectivities were obviously affected by the positions and electronic properties of the substituents of epoxides [27]. Notably, SlEH1 displayed the extremely high E value (> 200) resulting in the near-perfect kinetic resolution towards rac-2a which was obviously higher than those of EHs previously reported (Table 2) for preparing (R)- or (S)-2a, such as Kau2 (E = 35), StEH (E = 40), Sphingomonas sp EH (E = 14) and A. radiobacter mutant, Y215F (E = 130) [28–31]. On the other hand, SlEH1 exhibiting the highest EH activity towards rac-4a of 28.8 U/g wet cells could hydrolyze 250 mM of this substrate obtaining useful (R)-4a with 36.7% yields and > 98% ees (Fig. S5A), while most of EHs previously reported (Table 2) would lead to the formation of (S)-4a, for example A. radiobacter mutant, F108C, S. racemosum EH and A. niger EH, except for 10

B. sulfurescens EH [32–35]. 3.4. Regioselectivity of SlEH1 towards rac-6a–9a The regioselectivity coefficients, αS and βR, were applied to interpret the enantioconvergent hydrolysis of EHs towards rac-epoxides [36]. Unexpectedly, SlEH1 possessed the enantioconvergency towards rac-6a–9a (Fig. S3E–H), owing to its moderate regioselectivities and low enantioselectivities towards these substrates obtaining their corresponding chiral vicinal diols with eep from 14.3 to 51.3% and yieldp from 57.2 to 75.2%. Specifically, SlEH1 simultaneously displayed αS value of 55.3%, βR of 96.0% and low E of 2.3 towards rac-9a producing its corresponding (R)-9b with highest eep and yieldp at 100% c (Fig. S3H, Table 1). To our best known, the hydrolysis of phenyl glycidyl ether derivatives or epoxyalkanes by single EH in a partial enantioconvergent manner was firstly observed among all EHs reported. These phenomena suggested that SlEH1 had a potential to obtain their corresponding enantiopurity vicinal diols with higher eep and yieldp by protein engineering. In our previous work, a three-site mutant, PvEH1L105I/M160A/M175I, with enhanced αS and βR value from 91.1 to > 99% and 53.3 to 86.4% towards rac-1a, respectively, was obtained using directed modification affording (R)-1b with increased eep from 33.6 to 87.8% [36]. 3.5. Purification of SlEH1 and its enzymatic properties The SlEH1 was purified by affinity chromatography on a Ni-NTA column with a purification fold of 72.3 and yield of 43.1% (Fig. S2, lane 3) and its enzymatic properties towards rac-2a were further investigated, owing to its high enantioselectivity. The pH optimum and stability of SlEH1 were investigated over a pH range of 5.0–8.5. The optimal pH for SlEH1 obtaining maximum enzyme activity was 7.5, and more than 85% of its original enzyme activity retained at pH values ranging from 6.5–8.5 (Fig. 3A). The temperature stability assay of SlEH1 shown that it was thermostable at or below 40 °C until 1 h at pH optimum 7.5 (Fig. 3B), which was higher than that of A. mediolanus EH [37] and similar to A. usamii EH2 [12] while lower than VrEH3 [11]. Although SlEH1 11

displayed high stability at or below 40 °C, the E value of SlEH1 were decreased to 52.5 after kinetic resolution of rac-2a under the same conditions expect that the temperature was increased to 30 °C. Such a phenomenon that temperature dependent E value decrease had been reported previously [12,29]. The enzymatic kinetic assay was separately carried out using (S)- and (R)-2a as substrate, and the results were plotted and fitted to the Michaelis-Menten equation (Fig. S4). The VmaxS of SlEH1 towards (S)-2a was 27.67 ± 0.86 mol/min/mg which was 26.4-fold higher than that of VmaxR towards (R)-2a (1.05 ± 0.12 mol/min/mg), while the KmR was 72.83 ± 1.18 mM, 28.9-fold higher than that of KmS (2.52 ± 0.12 mM) (Table 3). This kinetic mechanism analysis indicating that SlEH1 simultaneously hydrolyzing faster and had higher binding affinity for (S)-2a than (R)-2a [11,37]. The ratio of catalytic efficiencies, (kcatS/KmS)/(kcatR/KmR), represented the degree of preferential hydrolysis of (S)- over (R)-2a [12], was 832.2, which was in accordance with the result that the extremely high enantioselectivity of SlEH1 towards rac-2a. 3.6. Gram-scale kinetic resolution of rac-2a by E. coli/sleh1 In order to confirm the capability of E. coli/sleh1 for synthetic applications, preparative-scale resolution of 150 mM rac-2a was conducted and monitored (Fig. S5B). After the reaction incubated at 20 °C for 8 h, the remaining (R)-2a with > 98% ees and 46.7% analytical yields was obtained. Subsequently, 0.53 g of (R)-2a was produced in 45.3% isolated yields after being purified by silica gel column chromatography. A higher concentration of rac-2a was found to be deleterious to the EH activity of E. coli/sleh1 and further studies on improving the stability and tolerance of E. coli/sleh1 were in progress, such as using a biphasic system [38]. 3.7. Modeling and analysis of the 3-D structure of SlEH1 The 3-D structure of SlEH1 (Fig. S6A) shown that it was composed by a α/β domain and a cap domain between which the catalytic triad (Asp105-Asp265-His300) and two proton donors (Tyr154 and Tyr235) were found. The catalytic triad, two tyrosines and residues around the active site (Gly32, Phe33, Pro34, Trp106, Phe109, Val130, 12

Ile155, Met180, Phe189, Leu266, Val267 and Phe301) [39] constituting a substrate binding pocket (SBP), which entrance size was considered to influence the substrate binding [40,41]. A comparison of the entrances to the SlEH1 and StEH SBP was performed (Fig. S6B and C), indicating that the size of entrance to the SlEH1 SBP was obviously smaller than that of StEH. Catalytic activity was observed in StEH towards rac-5a, while no activity was observed in SlEH1 [42]. This may be owing to that the relatively tight entrance tunnel caused higher steric hindrance when the bulky substrate rac-5a enter the SlEH1 SBP. 3.8. Molecular docking simulation of SlEH1 with (S)- or (R)-2a Molecular docking simulations were performed to analyze the origin of extremely high enantioselectivity of SlEH1 and high regioselectivity towards rac-2a. As shown in Fig. 4, (S)-2a was docked with the phenyl substituent pointing towards the exterior of the SlEH1 SBP (Fig. 4A), while (R)-2a was towards the interior (Fig. 4B). This phenomenon was also observed in StEH docked with (S)- and (R)-1a [39]. Based on the general mechanism, briefly, the Tyr154 and Tyr235 would assist in epoxide ring opening by forming hydrogen bonds with O-atom of the oxirane ring while the Asp105 attack on one C-atom of the oxirane ring to generate a covalent enzyme-substrate intermediate [43]. In detail, the hydrogen bond lengths (l1 and l2) for typical hydrogen bond interactions (2.7–3.2 Å) were the requirements needed to be fulfilled for a smooth reaction activation [44]. The l1 and l2 values of SlEH1 for (R)-2a (4.55 and 3.72 Å) were longer than those for (S)-2a (2.74 and 2.92 Å), implying that the disfavored (R)-2a was much more difficult to experience activation by the two proton donors than (S)-2a. The other requirement was the O-atom of nucleophile Asp and C-atom of oxirane ring were present in near attack conformation [32]. The dα for (S)-2a was 2.62 Å while the dβ was 2.79 Å indicating that the Cα of (S)-2a was present in near attack conformation. Analogously, the Cβ of (R)-2a was easily attacked by Asp105 owing to the relatively shorter dβ (3.13 Å) for (R)-2a than dα (3.26 Å). These results were consistent with the experimental findings that SlEH1 displayed high αS (95.6%) and βR value (81.1%) towards (S)- and (R)-2a, 13

respectively (Table 1). 4. Conclusion In conclusion, the ORF of sleh1 was cloned and heterologously expressed in E. coli BL21(DE3). The multiple sequence alignment of SlEH1 with other six plant EHs shown that it belongs to the α/β fold hydrolase family. E. coli/sleh1 expressing SlEH1 possessed high enantioselectivities towards rac-1a–4a and extremely high E value (> 200) towards rac-2a resulting in the near-perfect kinetic resolution to produce (R)-2a. Meanwhile, it also exhibited moderate regioselectivities towards rac-7a–9a which was first report about the single EH had a potential to enantioconvergently hydrolyze these types of rac-epoxides. Preparative-scale resolution of rac-2a was performed using E. coli/sleh1 resting cells, making SlEH1 an attractive biocatalyst for synthetic applications.

Conflict of interest The authors declare that there are no conflicts of interest.

Acknowledgment This work was financially supported by the National Natural Science Foundation of China (21676117), the Natural Science Foundation of Jiangsu Province for Youth of China (No. BK20180622), China Postdoctoral Science Foundation (No. 2018M630522), the National First-Class Discipline of Food Science and Technology of China (JUFSTR20180101), and the Postgraduate Research and Practice Innovation Program of Jiangsu Province, China (KYLX19_1840). We are grateful to Prof. Xianzhang Wu (School of Biotechnology, Jiangnan University) for providing technical assistance.

14

15

References [1] C. Bendigiri, K. Harini, S. Yenkar, S. Zinjarde, R. Sowdhamini, A. RaviKumar, Evaluating Ylehd, a recombinant epoxide hydrolase from Yarrowia lipolytica as a potential biocatalyst for the resolution of benzyl glycidyl ether, Rsc Adv. 8 (2018) 12918–12926. [2] S.L. Mowbray, L.T. Elfstrom, K.M. Ahlgren, C.E. Andersson, M. Widersten, X-ray structure of potato epoxide hydrolase sheds light on substrate specificity in plant enzymes, Protein Sci. 15 (2006) 1628–1637. [3] M. Kotik, A. Archelas, R. Wohlgemuth, Epoxide hydrolases and their application in organic synthesis, Curr. Org. Chem. 16 (2012) 451–482. [4] R. Wang, D. Hu, X.C. Zong, J.P. Li, L. Ding, M.C. Wu, J.F. Li, Enantioconvergent hydrolysis of racemic styrene oxide at high concentration by a pair of novel epoxide hydrolases into (R)-phenyl-1,2-ethanediol, Biotechnol. Lett. 39 (2017) 1917–1923. [5] Y. Genzel, A. Archelas, Q.B. Broxterman, B. Schulze, R. Furstoss, Microbiological transformations. 47. A step toward a green chemistry preparation of enantiopure (S)-2-, -3-, and -4-pyridyloxirane via an epoxide hydrolase catalyzed kinetic resolution, J. Org. Chem. 66 (2001) 538–543. [6] B.C. Hu, C. Li, R. Wang, X.C. Zong, J.P. Li, J.F. Li, M.C. Wu, Improvement in the activity and enantioconvergency of PvEH3, an epoxide hydrolase from Phaseolus vulgaris, for p-chlorostyrene oxide by site-saturation mutagenesis, Catal. Commun. 117 (2018) 9–13. [7] H. Kato, R. Kurosaki, C. Oki, T. Araki, Arundic acid, an astrocyte-modulating agent, protects dopaminergic neurons against MPTP neurotoxicity in mice, Brain Res. 1030 (2004) 66–73. [8] P. Saini, D. Sareen, An overview on the enhancement of enantioselectivity and stability of microbial epoxide hydrolases, Mol. Biotechnol. 59 (2017) 98–116. [9] H.B. Zheng, D. Kahakeaw, J.P. Acevedo, M.T. Reetz, Directed evolution of enantioconvergency: the case of 16

an epoxide hydrolase-catalyzed reaction of a racemic epoxide, Chemcatchem 2 (2010) 958–961. [10] C. Li, D. Hu, X.C. Zong, C. Deng, L. Feng, M.C. Wu, J.F. Li, Asymmetric hydrolysis of styrene oxide by PvEH2, a novel Phaseolus vulgaris epoxide hydrolase with extremely high enantioselectivity and regioselectivity, Catal. Commun. 102 (2017) 57–61. [11] D. Hu, C.D. Tang, C. Li, T.T. Kan, X.L. Shi, L. Feng, M.C. Wu, Stereoselective hydrolysis of epoxides by reVrEH3, a novel Vigna radiata epoxide hydrolase with high enantioselectivity or high and complementary regioselectivity, J. Agr. Food Chem. 65 (2017) 9862–9871. [12] D. Hu, C.D. Tang, B.A. Yang, J.C. Liu, T. Yu, C. Deng, M.C. Wu, Expression of a novel epoxide hydrolase of Aspergillus usamii E001 in Escherichia coli and its performance in resolution of racemic styrene oxide, J. Ind. Microbiol. Biotechnol. 42 (2015) 671–680. [13] X. Jia, Z.S. Wang, Z. Li, Preparation of (S)-2-, 3-, and 4-chlorostyrene oxides with the epoxide hydrolase from Sphingomonas sp. HXN-200, Tetrahedron Asymmetry 19 (2008) 407–415. [14] G. Hasnaoui, J.H.L. Spelberg, E. de Vries, L.X. Tang, B. Hauer, D.B. Janssen, Nitrite-mediated hydrolysis of epoxides catalyzed by halohydrin dehalogenase from Agrobacterium radiobacter AD1: a new tool for the kinetic resolution of epoxides, Tetrahedron Asymmetry 16 (2005) 1685–1692. [15] J.A. Kalow, A.G. Doyle, Mechanistic investigations of cooperative catalysis in the enantioselective fluorination of epoxides, J. Am. Chem. Soc. 133 (2011) 16001–16012. [16] S.K. Wu, Y.Z. Chen, Y. Xu, A.T. Li, Z. Li, Enantioselective trans-dihydroxylation of aryl olefins by cascade biocatalysis with recombinant Escherichia coli co-expressing monooxygenase and epoxide hydrolase, ACS Catal. 4 (2014) 409–420. [17] N. Bala, S.S. Chimni, H.S. Saini, B.S. Chadha, Bacillus alcalophilus MTCC10234 catalyzed enantioselective kinetic resolution of aryl glycidyl ethers, J. Mol. Catal. B Enzym. 63 (2010) 128–134. 17

[18] S.E. Schaus, B.D. Brandes, J.F. Larrow, T. Makoto, K.B. Hansen, A.E. Gould, M.E. Furrow, E.N. Jacobsen, Highly selective hydrolytic kinetic resolution of terminal epoxides catalyzed by chiral (salen)Co(III) complexes. Practical synthesis of enantioenriched terminal epoxides and 1,2-diols, J. Am. Chem. Soc. 124 (2002) 1307–1315. [19] A.L. Botes, C.A.G.M. Weijers, M.S. van Dyk, Biocatalytic resolution of 1,2-epoxyoctane using resting cells of different yeast strains with novel epoxide hydrolase activities, Biotechnol. Lett. 20 (1998) 421–426. [20] M. Kotik, P. Kyslik, Purification and characterisation of a novel enantioselective epoxide hydrolase from Aspergillus niger M200, BBA 1760 (2006) 245–252. [21] W. Xu, J.H. Xu, J. Pan, Q. Gu, X.Y. Wu, Enantioconvergent hydrolysis of styrene epoxides by newly discovered epoxide hydrolases in mung bean, Org. Lett. 8 (2006) 1737–1740. [22] Q.Q. Zhu, W.H. He, X.D. Kong, L.Q. Fan, J. Zhao, S.X. Li, J.H. Xu, Heterologous overexpression of Vigna radiata epoxide hydrolase in Escherichia coli and its catalytic performance in enantioconvergent hydrolysis of p-nitrostyrene oxide into (R)-p-nitrophenyl glycol, Appl. Microbiol. Biotechnol. 98 (2014) 207–218. [23] C. Li, J. Zhao, D. Hu, B.C. Hu, R. Wang, J. Zang, M.C. Wu, Multiple site-directed mutagenesis of a Phaseolus vulgaris epoxide hydrolase to improve its catalytic performance towards p-chlorostyrene oxide based on the computer-aided re-design, Int. J. Biol. Macromol. 121 (2019) 326–332. [24] M. Holmquist, Alpha/Beta-hydrolase fold enzymes: structures, functions and mechanisms, Curr. Protein Pept. Sci. 1 (2000) 209–235. [25] S. Barth, M. Fischer, R.D. Schmid, J. Pleiss, Sequence and structure of epoxide hydrolases: a systematic analysis, Proteins 55 (2004) 846–855. [26] B. van Loo, J. Kingma, M. Arand, M.G. Wubbolts, D.B. Janssen, Diversity and biocatalytic potential of epoxide hydrolases identified by genome analysis, Appl. Environ. Microb. 72 (2006) 2905–2917. 18

[27] J. Zhao, Y.Y. Chu, A.T. Li, X. Ju, X.D. Kong, J. Pan, Y. Tang, J.H. Xu, An unusual (R)-selective epoxide hydrolase with high activity for facile preparation of enantiopure glycidyl ethers, Adv. Synth. Catal. 353 (2011) 1510–1518. [28] M. Kotik, V. Stepanek, M. Grulich, P. Kyslik, A. Archelas, Access to enantiopure aromatic epoxides and diols using epoxide hydrolases derived from total biofilter DNA, J. Mol. Catal. B Enzym. 65 (2010) 41–48. [29] K.M. Manoj, A. Archelas, J. Baratti, R. Furstoss, Microbiological transformations. Part 45: A green chemistry preparative scale synthesis of enantiopure building blocks of Eliprodil: elaboration of a high substrate concentration epoxide hydrolase-catalyzed hydrolytic kinetic resolution process, Tetrahedron 57 (2001) 695– 701. [30] S.K. Wu, A.T. Li, Y.S. Chin, Z. Li, Enantioselective hydrolysis of racemic and meso-epoxides with recombinant Escherichia coli expressing epoxide hydrolase from Sphingomonas sp. HXN-200: preparation of epoxides and vicinal diols in high ee and high concentration, ACS Catal. 3 (2013) 752–759. [31] J.H.L. Spelberg, R. Rink, A. Archelas, R. Furstoss, D.B. Janssen, Biocatalytic potential of the epoxide hydrolase from Agrobacterium radiobacter AD1 and a mutant with enhanced enantioselectivity, Adv. Synth. Catal. 344 (2002) 980–985. [32] B. van Loo, J. Kingma, G. Heyman, A. Wittenaar, J.H.L. Spelberg, T. Sonke, D.B. Janssen, Improved enantioselective conversion of styrene epoxides and meso-epoxides through epoxide hydrolases with a mutated nucleophile-flanking residue, Enzyme Microb. Tech. 44 (2009) 145–153. [33] P. Moussou, A. Archelas, J. Baratti, R. Furstoss, Microbiological transformations. 38. Clues to the involvement of a general acid activation during hydrolysis of para-substituted styrene oxides by a soluble epoxide hydrolase from Syncephalastrum racemosum, J. Org. Chem. 63 (1998) 3532–3537. [34] H. Jin, Z.Y. Li, X.W. Dong, Enantioselective hydrolysis of various substituted styrene oxides with Aspergillus 19

Niger CGMCC 0496, Org. Biomol. Chem. 2 (2004) 408–414. [35] A. Archelas, Fungal epoxide hydrolases: new tools for the synthesis of enantiopure epoxides and diols, J. Mol. Catal. B Enzym. 5 (1998) 79–85. [36] H.H. Ye, D. Hu, X.L. Shi, M.C. Wu, C. Deng, J.F. Li, Directed modification of a novel epoxide hydrolase from Phaseolus vulgaris to improve its enantioconvergence towards styrene epoxides, Catal. Commun. 87 (2016) 32–35. [37] F. Xu, Z.Q. Liu, S.P. Zou, N.W. Wan, W.Y. Zhu, Q. Zhu, Y.G. Zheng, A novel enantioselective epoxide hydrolase from Agromyces mediolanus ZJB120203: cloning, characterization and application, Process Biochem. 49 (2014) 409–417. [38] C. Li, B.C. Hu, D. Hu, X.F. Xu, X.C. Zong, J.P. Li, M.C. Wu, Stereoselective ring-opening of styrene oxide at elevated concentration by Phaseolus vulgaris epoxide hydrolase, PvEH2, in the organic/aqueous biphasic system, Catal. Commun. 123 (2019) 1–5. [39] M.E.S. Lind, F. Himo, Quantum chemical modeling of enantioconvergency in soluble epoxide hydrolase, ACS Catal. 6 (2016) 8145–8155. [40] R. Chaloupkova, J. Sykorova, Z. Prokop, A. Jesenska, M. Monincova, M. Pavlova, M. Tsuda, Y. Nagata, J. Damborsky, Modification of activity and specificity of haloalkane dehalogenase from Sphingomonas paucimobilis UT26 by engineering of its entrance tunnel, J. Biol. Chem. 278 (2003) 52622–52628. [41] M. Kotik, A. Archelas, V. Famerova, P. Oubrechtova, V. Kren, Laboratory evolution of an epoxide hydrolase towards an enantioconvergent biocatalyst, J. Biotechnol. 156 (2011) 1–10. [42] M.I. Monterde, M. Lombard, A. Archelas, A. Cronin, M. Arand, R. Furstoss, Enzymatic transformations. Part 58: enantioconvergent biohydrolysis of styrene oxide derivatives catalysed by the Solanum tuberosum epoxide hydrolase, Tetrahedron Asymmetry 15 (2004) 2801–2805. 20

[43] M.T. Reetz, M. Bocola, L.W. Wang, J. Sanchis, A. Cronin, M. Arand, J.Y. Zou, A. Archelas, A.L. Bottalla, A. Naworyta, S.L. Mowbray, Directed evolution of an enantioselective epoxide hydrolase: uncovering the source of enantioselectivity at each evolutionary stage, J. Am. Chem. Soc. 131 (2009) 7334–7343. [44] Z.T. Sun, L. Wu, M. Bocola, H.C.S. Chan, R. Lonsdale, X.D. Kong, S.G. Yuan, J.H. Zhou, M.T. Reetz, Structural and computational insight into the catalytic mechanism of limonene epoxide hydrolase mutants in stereoselective transformations, J. Am. Chem. Soc. 140 (2018) 310–318.

21

Figure captions

Fig. 1. Nine rac-epoxides, rac-1a–9a, used for substrate spectrum assay.

Fig. 2. The multiple alignment of SlEH1 (AXM43803) with other six plant EHs. StEH (AAA81891, 88.6%), PvEH2 (ASS33914, 58.6%), VrEH1 (ADP68585, 57.6%), VrEH2 (AIJ27456, 57.5%), PvEH1 (AKJ75509, 56.9%) and GmEH (CAA55294, 53.2%). Three conserved motifs: HGXP (residues 31–34), GXSmXS/T (residues 61–65) and SmXNuXSmSm (residues 103–108).

Fig. 3. The enzymatic properties of SlEH1 towards rac-2a: (A) pH optimum and pH stability; (B) temperature stability.

Fig. 4. Molecular docking of (S)-2a (A) or (R)-2a (B) into the SBP of SlEH1. The purple and blue sticks were represented the (S)-2a and (R)-2a, respectively. The hydrogen bond length (l1 or l2) was shown in pink line, while the distance (dα or dβ) shown in green line.

22

Tables Table 1 Substrate spectrum assay of E. coli/sleh1 resting cells towards rac-1a–4a and 6a–9a. Conc.

Dose

Activity

(mM)

(g/L) a

(U/g) b

Subs.

αS

βR

(%)

(%)

E

ees (%) conf.

a

Enantioconvergent hydrolysis c

Kinetic resolution eep (%)

d

99.1 (R)

yields e

eep (%)

yieldp

conf.

(%) f

conf.

(%)

89.6 (R)

39.6

—

— —

1a

20

80

5.1

24.0

96.9

82.8

2a

20

50

18.0

> 200

95.6

81.1

99.1 (R)

89.8 (R)

49.2

—g

3a

20

80

2.0

51.2

92.3

96.3

99.3 (R)

86.5 (R)

44.8

—

—

4a

20

20

28.8

20.2

91.9

44.5

99.0 (R)

65.0 (R)

37.5

—

—

—

<1

14.3 (R)

57.2

6a

10

100

1.8

1.0

77.5

34.9

50.1 (S→R)

7a

10

100

0.5

1.1

1.0

52.7

46.4 (S)

—

<1

46.5 (S)

69.4

8a

10

100

1.5

1.7

73.7

64.6

99.5 (R)

—

<1

40.5 (R)

70.2

9a

10

100

5.5

2.3

55.3

96.0

99.3 (R)

—

<1

51.3 (R)

75.2

h

Dose of E. coli/sleh1 wet cells. b Assayed at 20 °C. c 100% Conversion ratio. d Configuration. e Analytical epoxide

yield. f Analytical diol yield. g Not determined. h The enantioselectivity reversed during the hydrolysis [21].

Table 2 Kinetic resolutions of rac-2a and 4a by several EHs. Substrate 2a

4a

a

Enzyme source

Subs. conc.

E

ees (%)

(mM)

(temp.)

conf.

Catalytic form

Reference

Kau2

enzymatic extract

3.5

35 (28 ℃)

— a (R)

[28]

S. tuberosum

purified enzyme

4

40 (27 ℃)

> 97 (R)

[29]

Sphingomonas sp

E. coli cells

200

14 (30 ℃)

98.1 (S)

[30]

A. radiobacter mutant, Y215F

purified enzyme

3

130 (30 ℃)

— (S)

[31]

A. radiobacter mutant, F108C

purified enzyme

2

39 (30 ℃)

— (S)

[32]

S. racemosum

enzymatic extract

<5

1.1 (27 ℃)

— (S)

[33]

A. niger

whole cells

15

— (25 ℃)

70 (S)

[34]

B. sulfurescens

whole cells

7.5

— (27 ℃)

> 98 (R)

[35]

No information. b Analytical yield.

23

Table 3 Kinetic parameters of hydrolysis of (R)- and (S)-2a with purified SlEH1. Substrate

Km (mM)

Vmax (mol/min/mg)

kcat (s-1)

kcat/Km (mM-1s-1)

(R)-2a

72.83 ± 1.18

1.05 ± 0.12

0.72 ± 0.03

0.009

(S)-2a

2.52 ± 0.12

27.67 ± 0.86

18.91 ± 0.87

7.49

24

25

26

27

a

28