WO2020009168A1 - Novel hydroxynitrile lyase mutant - Google Patents

Abstract

The activity of the wild-type hydroxynitrile lyase (HNL) obtained to date is insufficient for industrial use, and HNL having better activity has been demanded. The present inventors introduced various amino acid mutations predicted on the basis of the three-dimensional structures of HNL and (R)-mandelonitrile into wild-type HNL and measured the activities of the resulting HNL. As a result, it was discovered that amino acid modification of specific sites brings about elevation of activity common to millipede-derived HNL. Therefore, the present invention pertains to an (R)-HNL mutant protein having one or more amino acid substitutions in a millipede-derived (R)-HNL protein and having (R)-HNL activity. More specifically, the present invention pertains to HNL having amino acid substitutions in the third β-sheet structure (β3), the fourth β-sheet structure (β4), and the fifth β-sheet structure (β5) of (R)-HNL.

Description

Novel hydroxy nitrile lyase mutant

(4) The present invention relates to a modified hydroxynitrile lyase and use thereof.

Hydroxynitrile lyase (hereinafter, referred to as “HNL”) is an enzyme used for synthesizing a cyanohydrin compound. The present inventors have already succeeded in cloning the HNL gene from various millipedes (Patent Document 1, Patent Document 2, Non-Patent Document 1, and Non-Patent Document 2).

International Publication WO2015 / 133462 International Publication WO2017 / 150560

However, the activity of these natural HNLs is not sufficient for industrial use, and HNLs having even better activities have been demanded.

The present inventors introduced various amino acid mutations into HNL and measured their activities. As a result, they have found that amino acid modification at a specific site results in an increase in activity common to millipede-derived HNL, thereby completing the present invention.

In one aspect, the present invention relates to a (R) -HNL mutant protein having one or more amino acid substitutions in a millipede-derived (R) -HNL protein and having (R) -HNL activity.

に お い て In the present specification, “millipedes-derived (R) -hydroxynitrile lyase” is an enzyme encoded by the millipede (Millipedes) gene, and is a compound having (R) -HNL activity. As used herein, the term “millipede” means an arthropod belonging to the polypod subdivision millipede (Diplopoda). As the millipede, for example, Shinoharahusayasade, Ryukyufusayasde, Tamayasde, Anetaytamayasde, Otamayamasde, Mykosyasde, Ezomikoshyades, Kroiwayasde, Futkesayas, Yaryasde, Takanejashyde, Ookeshyde , Tsukiyasade, Itoyasude, Hiratayasde, Akahirathayasade, Yamashinahirathayasde, Tamamohiratayades, Babayasde, Yamambayasde, Herababayasde, Midoribayayasde, Amabicayas, Yakayakeyasde, Komaya, Maayakosaya Amabicoyasude, Obibabayasude, Torideyasude, Kishayasude, Tammetomoyasude, Uchikakeyasude, Hira Millipede, Yakeyasude, Masakiyayasuede, Miitsuyasuede, Yamato Akayasde, Ryukyu Yayasude, Yanbarto Sakayasde, Tosakasasade, Moriyasde, Nejiashiasude, Akayas, Umagaesiakayasade, Nayoyayasade, Nayoyasadeas Yoroyasde, Sawglyjasde, Tsunokogiyasude, Ishiiobiasde, Obiasde, Motoobiyasde, Higasiobiasde, Fuziobiasde, Shirohadayasde, Makuragiyasde, Tibijasde, Giboshejasde, Paraogibojade , Fujisade, Fujisademodoki, Firefly Yasude, Ikahohimesude, Firefly Himedesue, Senbu Millipede, Uenoyasude, Prince of Changyi millipedes, giant cell Prince of Changyi millipede, Torii Prince of Changyi millipedes, vertical Une Hora millipede, Ichihashiyasude, Nenjuyasude, Hiroumiyasude Yoshida Hime millipede, Kurohimeyasude, Ezohimeyasude, Kudayasude, Minamiyasude, Magaimaruyasude, Kaguyayasude, Maruyasude, Higeyasude, Himoyasude, P. Tokaiensis and P. japonica.

に お い て In the present specification, “(R) -hydroxynitrile lyase activity ((R) -HNL activity)” means an activity of catalyzing the following reaction for synthesizing an optically active cyanohydrin from a ketone or aldehyde and a cyanide. As used herein, both a millipede-derived (R) -HNL and a millipede-derived (R) -HNL mutant have (R) -HNL activity.

Whether a protein has (R) -HNL activity or not can be determined by measuring the catalytic activity of the (R) -mandelonitrile synthesis reaction and the activity of the decomposition reaction of mandelonitrile to benzaldehyde. it can. The catalytic activity of the (R) -mandelonitrile synthesis reaction can be determined by performing the following reaction using benzaldehyde as a substrate. The benzaldehyde DMSO solution and the test protein solution are added to a citrate buffer (pH 4.2), mixed, and 1 M @KCN is added to start the synthesis reaction. The reaction is performed at 15 to 25 ° C. for 5 minutes to 1 hour. The liquid is collected, a mixed liquid of n-hexane: 2-propanol = 85: 15 is added, the mixture is vigorously stirred, and the mixture is centrifuged at 4 ° C. and 16,000 g for 3 minutes to collect an organic layer. The obtained organic layer was analyzed using HPLC having a stationary phase in which a cellulose derivative (Cellulose @ tris (4-methylbenzoate)) was coated on a silica gel carrier, and (R) -mandelonitrile and (S) -mandelonitrile were analyzed. The peak detected at 254 nm at each retention time is observed. When the synthesis of (R) -mandelonitrile can be confirmed and the production amount of (R) -mandelonitrile is larger than that of (S) -mandelonitrile, the protein has (R) -HNL activity Can be determined. The activity of the decomposition reaction of mandelonitrile to benzaldehyde can be determined by adding a test protein solution to a citrate buffer (5.0-5.5) containing racemic mandelonitrile and gently stirring. The reaction can be performed at 15 to 25 ° C. for 1 minute to 1 hour, and the amount of benzaldehyde produced can be confirmed by measuring the absorbance at 280 nm. When benzaldehyde is produced, the protein can be determined to have (R) -HNL activity.

に お い て In the present specification, the millipede-derived (R) -HNL preferably has eight back-equilibrium β-sheet structures as three-dimensional structures. More preferably, the amino acid sequences constituting the eight β sheet structures are X15X16FX17X18VL (β1) (SEQ ID NO: 1), TX19RX20YVX21P (β2) (SEQ ID NO: 2), TAX1DI (β3) (SEQ ID NO: 3), respectively. X2X3X4 (X7) X2X3X3 (X7X3) X2X3X3 (X7) X3 (X7) X3 (X7) X3 (X7) X3 (X7) X3 (X7) X3 (X7) X3 (X7X8) (X7X3X8) (X7X3X8) (X7X3X8) (X7X3X8X8) (SEQ ID NO: 8) is there. Where X1 is L or F, X2 is Q, H or R, X3 is I or V, X4 is M, I, T or D, and X5 is A, T or I , X6 is Y or N, X7 is V, T or L, X8 is G or I, X9 is G or A, X10 is P, A or S, X11 is S, L , M or I, X12 is T or absent, X13 is H, I, Y or F, X14N or T, X15 is F or L, X16 is E, Q or L , X17 is E, A, S or T, X18 is Y or F, X19 is A or T, X20 is V or I, X21 is Q or R, X22 is G or D X23 is E, K, D or A, X24 is Q, T or A, X25 is V, I or T, X26 is Y H or N, X27 is T, V or I, X28 is N or D, X29 is A or S, X30 is N or S, X31 is R, T or S; X32 is N or D, X33 is E, A, Q, N or S, X34 is I, A or V, X35 is A or T, and X36 is S, N or T.

Furthermore, the millipede-derived (R) -HNL herein may have one α-helix structure. The amino acid sequence constituting the α-helix structure is preferably VPNX37KIH (SEQ ID NO: 9) (where X37 is D or Y).

In one example, the millipede-derived (R) -HNL herein may be classified as a protein belonging to the lipokine superfamily based on its structural characteristics.

As a more specific example, (R) -HNL derived from a millipede in the present specification is
ChuaHNL from Chamberlinius hualienensis (SEQ ID NO: 10, signal peptide 1-221, mature protein 22-183);
NttHNL from Nedyopus tambanus tambanus (SEQ ID NO: 11, signal peptide 1-20, mature protein 21-182);
NtmHNL from Nepodus tambanus mangaesinus (SEQ ID NO: 12, signal peptide 1-20, mature protein 21-182);
OgraHNL derived from Oxidus gracilis (SEQ ID NO: 13, signal peptide 1-18; mature protein 19-184);
PlamHNL (SEQ ID NO: 14, signal peptide 1-20, mature protein 21-183) derived from the yellow croaker (Parafontaria laminata armigera);
Pton1HNL (Parafontaria tonominea species complex 1) derived from Parafantaria tonominea (SEQ ID NO: 15, signal peptide 1-26th, mature protein 27-189th), Pton2HNL ealone Signal peptide 1 to 25, mature protein 26 to 188), and Pton3HNL (Parafontaria tonominea species complex 3) (SEQ ID NO: 17, signal peptide 1 to 26, mature protein 27 to 189);
PfalHNL from Parafantaria falcipera (SEQ ID NO: 18, signal peptide 1-26, mature protein 27-189);
Ptok HNL from Parafontaria tokaiensis (SEQ ID NO: 19, signal peptide 1 to 25, mature protein 26 to 188);
RssHLN (SEQ ID NO: 20, signal peptide 1-26, mature protein 27-188) derived from Rubiaria millipede (Riukiaria semicircularis semicircularis); and RspHN1 peptide derived from one species of Rubiaria millipede (Riukiaria sp.) (26th, mature protein 27th to 189th).

輸送 Transport and localization of proteins are controlled by signal sequences. The (R) -HNL or a mutant protein thereof in the present specification may be a protein having a signal sequence or a mature protein in which the signal sequence has been cleaved.

The mutant (R) -HNL protein of the present invention has one or more amino acid substitutions in the millipede-derived (R) -HNL protein. Such amino acid substitutions can be the following amino acid substitutions:
(A) substitution of another amino acid, A, which is the second amino acid in the amino acid sequence represented by TAX1DI (SEQ ID NO: 3) constituting the β sheet structure (β3) of the (R) -HNL, wherein , X1 is L or F;
(B) substitution of another amino acid, X6, which is the fifth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of (R) -HNL, wherein , X2 is Q, H or R, X3 is I or V, X4 is M, I, T or D, X5 is A, T or I, X6 is Y or N, X7 Is V, T or L, X8 is G or I, X9 is G or A, and X10 is P, A or S;
(C) substitution of another amino acid F which is the seventh amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β sheet structure (β4) of the (R) -HNL, wherein , X2 to X10 are as defined in (b);
(D) substitution of another amino acid, X8, which is the ninth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, Wherein X2 to X10 are as defined in (b);
(E) substitution of another amino acid, X12, which is the second amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -HNL, wherein , X11 is S, L, M or I, X12 is T or absent, X13 is H, I, Y or F and X14N or T; and (f) the (R) -HNL Substitution of another amino acid, A, which is the third amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β sheet structure (β5), wherein X11 to X14 are defined in (d) As it was done.

Preferably, the amino acid substitution is one or more substitutions selected from the following (a) to (e):
(A) substitution of C or H for the second amino acid A in the amino acid sequence represented by TAX1DI (SEQ ID NO: 3) constituting the β-sheet structure (β3) of the (R) -hydroxynitrile lyase;
(B) H, Y, M, V of X6 which is the fifth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase , L or W substitutions;
(C) substitution of I, which is the seventh amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase;
(D) substitution of X8, which is the ninth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, with G;
(E) substitution of A for X12, which is the second amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase; and f) The hydrophobic residue (I, L) of the third amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase , M, F, W, Y or V), C, T, E, Q or S (preferably I, L, M, F, V, C, T, E, Q or S).

These amino acid substitutions may be made in only one place, or may be made in two or more places, for example, two places, three places, four places, or five places.

More specifically, the substitution is a substitution in an OgraHNL mutant protein, and is selected from the following (i) to (iii): 1, 2, 3, 1-2, or One to three substitutions can be mentioned.
(I) a cysteine residue at the 76th alanine residue in SEQ ID NO: 13 (corresponding to the 58th alanine residue in the OgraHNL mature protein (protein consisting of amino acids 19 to 184 of SEQ ID NO: 13; the same applies hereinafter)); Or substitution with a histidine residue (A76C or A76H) (A58C or A58H in the mature protein)
(Ii) Substitution of the phenylalanine residue at position 89 (corresponding to the phenylalanine residue at position 71 in the mature OgraHNL protein) with an isoleucine residue in SEQ ID NO: 13 (F89I) (F71I in the mature protein)
(Iii) cysteine residue, phenylalanine residue, isoleucine residue, leucine residue, methionine residue, serine at the 97th alanine residue (corresponding to the 79th alanine residue in the mature OgraHNL protein) in SEQ ID NO: 13 (A97C, A97F, A97I, A97L, A97M, A97S, or A97V) (A79C, A79F, A79I, A79L, A79M, A79S, or A79V in the mature protein)

More specifically, the substitution is a substitution in a PlamHNL mutant protein, and is selected from the following (iv) to (vi): 1, 2, 3, 1-2, or One to three substitutions can be mentioned. For example, in the case of having two mutations, N65H and T95A, N85Y and T95A, N85Y and I89G (N65H and T75A, N65Y in PlamHNL mature protein (protein consisting of amino acids 21 to 183 of SEQ ID NO: 14; the same applies hereinafter)) T75A, N65Y and I69G).
(Iv) Cysteine residue, serine residue, threonine residue, glutamic acid residue, glutamic acid residue, glutamine residue, valine at the 96th alanine residue (corresponding to the 76th alanine residue in the PlamHNL mature protein) in SEQ ID NO: 14 Residue or methionine residue (A96C, A96S, A96T, A96E, A96Q, A96V, or A96M) (A76C, A76S, A76T, A76E, A76Q, A76V, or A76M in mature protein)
(V) a histidine residue, a tyrosine residue, a methionine residue, a valine residue, a leucine residue, or the asparagine residue at position 85 in SEQ ID NO: 14 (corresponding to the asparagine residue at position 65 in the PlanHNL mature protein) Substitution to tryptophan residues (N85H, N85Y, N85M, N85V, N85L, or N85W) (N65H, N65Y, N65M, N65V, N65L, or N65W in mature protein)
(Vi) Substitution of glycine residue for isoleucine residue 89 (corresponding to isoleucine residue 69 in PlamHNL mature protein) in SEQ ID NO: 14 (I89G) (I69G in mature protein)

Further, in the present specification, the variant protein of the millipede-derived (R) -HNL of the present invention is characterized in that amino acids other than the above amino acid substitutions are derived from the original or wild type millipede (R) -HNL as long as it has (R) -HNL activity. ) -HNL protein may be substituted with a different amino acid. For example, a variant protein of a millipede-derived (R) -HNL of the present invention has 1 to 10 amino acids other than the above amino acid substitutions in the amino acid sequence of the original or wild type millipede-derived (R) -HNL protein. 1 to 8, 1 to 5, 1 to 4, 1 to 3, 1 to 2, or 1 may be substituted. Alternatively, a variant of a millipede-derived (R) -HNL protein of the present invention is 60%, 70%, 80%, 85%, 90%, 95% of the original or wild type millipede-derived (R) -HNL protein. , 98%, or 99% identity. Thus, for example, when the millipede-derived (R) -HNL protein has the amino acid sequence described in any one of SEQ ID NOS: 10 to 21 (or the amino acid sequence constituting the mature protein in the sequence), the present invention In addition to the amino acid substitution, the variant protein of the millipede-derived (R) -HNL is the same as the amino acid sequence described in any one of SEQ ID NOS: 10 to 21 (or the amino acid sequence constituting the mature protein in the sequence). Has a sequence in which 1 to 10, 1 to 8, 1 to 5, 1 to 4, 1 to 3, 1 to 2, or 1 amino acid is substituted with another amino acid; and , (R) -HNL protein. Alternatively, the variant protein of the millipede-derived (R) -HNL of the present invention comprises the amino acid sequence described in any one of SEQ ID NOS: 10 to 21 (or the amino acid sequence constituting a mature protein in the sequence). %, 70%, 80%, 85%, 90%, 95%, 98% or 99% identity and may be a protein having (R) -HNL activity. .

ア ミ ノ 酸 As used herein, amino acids are represented by one-letter or three-letter codes commonly used in the art. Further, the amino acid represented by “XN” (N is a natural number) means any one amino acid selected from a plurality of defined amino acids. The notation represented by XaNuXb (Xa and Xb are single-letter amino acid codes, and Nu is a natural number) indicates that the Nuth amino acid is substituted from Xa to Xb. For example, A79M means that the alanine at position 79 is replaced with methionine.

In the amino acid sequences described herein, A is an alanine residue, R is an arginine residue, N is an asparagine residue, D is an aspartic acid residue, C is a cysteine residue. , Q is a glutamine residue, E is a glutamic acid residue, G is a glycine residue, H is a histidine residue, I is an isoleucine residue, L is a leucine residue. , K is a lysine residue, M is a methionine residue, F is a phenylalanine residue, P is a proline residue, S is a serine residue, T is a threonine residue, W is Is a tryptophan residue, Y is a tyrosine residue, and V is a valine residue.

In another embodiment, the present invention relates to a nucleic acid molecule encoding the above-mentioned millipede-derived (R) -HNL mutant protein. The nucleic acid molecule of the present specification is a nucleic acid molecule capable of expressing the variant protein of the millipede-derived (R) -HNL of the present invention in a target transformed cell. The nucleic acid may be DNA, RNA, artificial nucleic acid or a modified product thereof. The nucleic acid of the present invention may be an expression cassette containing a region necessary for expression (promoter, enhancer, terminator, etc.) as necessary.

に お い て In yet another aspect, the present invention relates to a vector having the nucleic acid molecule. The vector of the present invention is not particularly limited as long as it is a vector capable of expressing a target protein in combination with a transformed cell to be used. The vector may be a plasmid vector or a viral vector. For example, when Escherichia coli is used as a host, a pET vector can be used.

The present invention further includes a transformed cell transformed with the vector. As the transformed cells, any cells such as Escherichia coli (for example, Escherichia coli SHuffle @ T7 strain), yeast (for example, Pichia @ pastoris), insect cells, and animal cells can be used.

The variant of the millipede-derived HNL of the present invention has excellent specific activity, enantioselectivity, and / or productivity, and thus is suitable for industrial production of optically active cyanohydrin utilizing HNL activity. The mutants of the present invention show high enantioselectivity with less enzyme than before (150 U / ml). Also, for (R) -2-Cl-Man, Chua HNL and other reported HNLs showed only very low enantioselectivity in the buffer system (ee <21%), but the mutants of the present invention Also shows high enantioselectivity for (R) -2-Cl-Man. Also, unlike the conventional (R) -HNL, which catalyzes the reaction in a two-phase system containing an organic solvent, the mutant of the present invention has a high enantiopurity (96.3%) in a single-phase buffer system (96.3%). An asymmetric synthesis of R) -2-Cl-Man is possible.

It is a figure which shows the whole structure of ChuaHNL. (A) A stereoscopic image of the monomer structure of ChuaHNL is shown pictorially. Helices (alpha-of α1-2 and _{3 10} helix helix η1,3), β sheets (β1-8) and loop structure is shown. The symbols “N” and “C” indicate the N-terminal and C-terminal of ChuaHNL, respectively. (B) shows the dimer structure of ChuaHNL. Disulfide bonds are represented as spheres with residue numbers. The N-acetylglucosamine moieties binding to Asn109 and Asn123 are designated NAG1 and NAG2, respectively. Asn109, Asn123, and the sugar moiety are shown as sticks. (C) Diagram showing the dimer interface of ChuaHNL. Disulfide bonds, hydrogen bonds, and salt bridges at the interface are shown. Interface residues and β-sheets are shown. FIG. 3 is a diagram showing the active site of ChuaHNL conjugated with various ligands. The active site amino acid residues and bound ligands are shown as sticks in the CPK color scheme (protein carbon atoms and ligands are shown in green and cyan, respectively). Hydrogen bonding and electrostatic interactions between amino acid residues and ligands are indicated by dotted lines and bond lengths are indicated by Å. The σ _A -weighted omit map of the ligand and the water molecule is shown in mesh representation and is outlined at the 3.0σ level except for D (5.0σ). (A) is a view showing a wall-eye stereo view of an active site bound to acetate. The surface of the active site cavity is shown as a transparent gray surface. The lower part of the image is the entrance of the active site. (B) is a diagram showing a ligand-free form. Water molecules bound to the active site are represented by spheres. (C) A diagram showing a complex with cyanide ions. (D) is a diagram showing a complex with iodoacetate. The anomalous difference map was shown as a mesh and outlined at the 4.0σ level. It is a figure which shows the complex with (E) thiocyanate. Two alternative binding forms of thiocyanate are shown (SCN1 and SCN2). It is a figure showing the result of docking simulation of ChuaHNL. 2 shows the docked (R) -MAN surface at the active site of ChuaHNL. (R) -MAN and active site residues are shown in sticky representation in the CPK color scheme. The alignment of the deduced amino acid sequences of the twelve millipede HNL is shown (SEQ ID NOS: 10 to 21). Asterisks indicate conserved regions that are thought to be involved in the binding pocket of millipede HNL. Underlines indicate signal sequences. Arrows represented by βn (n is a natural number of 1 to 8) indicate amino acids constituting the inverse equilibrium β sheet structure, respectively. In this figure, amino acid numbers are represented by numbers in the sequence including the signal sequence (underlined). FIG. 4 is a diagram showing the intended catalytic mechanism determined by the crystal structure of OgraHNL formed in a complex with (R) -2-Cl-Man impregnated with 2-chlorobenzaldehyde and KCN. (A) Dimer model of OgraHNL is shown. Two α- helices shows two _{3 10} helices and eight antiparallel β- sheet. C-terminal and N-terminal are indicated. The ligand (R) -2-Cl-Man in the active site was shown as a yellow stick model. (B) It is an enlarged view of a binding pocket. The entrance tunnel is indicated by the dotted blue circle. Residues exposed to the binding pocket containing hydrophobic residues (green) and hydrophobic residues (positive side chains: blue; negative side chains: pink; and uncharged side chains: light orange) It was shown as a stick model and displayed the amino acid type and number. (C) Shows the active site of OgraHNL bound to (R) -2-Cl-Man. The hydrogen bond is indicated by a red dotted line, and the distance is indicated. (D) Represents the desired catalytic mechanism of OgraHNL. (1) Standby state. The catalyst two molecules of Arg42 and Lys121 are shown in blue. (2) Complex formation with (R) -Man. The isolated electron pair at the nitrogen of Lys121 abstracts a proton from the hydroxyl group of (R) -Man, and the electron released from hydrogen is received by the carbon atom of the nitrile group, triggering the release of cyanide ion. (3) Cyanide protonation. The released cyanide ions remove protons from Arg42 to generate hydrogen cyanide. The aldehyde group of 2-chlorobenzaldehyde forms a hydrogen bond with Lys121 and Arg42. Subsequently, the generated 2-chlorobenzaldehyde and hydrogen cyanide are released, and the active residues return to the standby state (1). (A) and (b) The binding mode of (R) -2-Cl-Man designed such that different ortho positions (a, b) of the phenyl ring in the binding pocket of OgraHNL are replaced by chlorine atoms. (C) and (d) Open (c) and closed (d) structures of the substrate entry tunnel of the OgraHNL mutant. The predicted structure of PlamHNL by (R) -2-chloromandelonitrile by docking simulation is shown. a) Represents the overall structure of homology modeling prediction of PlamHNL (blue) using OgraHNL (violet) as template. b) Superposition of the side-chain residues linked to 2-chloromandelonitrile (green) in the binding pocket indicates significant similarity between the two proteins. 4 is a graph showing the influence of the amount of enzyme (a) and pH (b) on the enantiomeric excess of (R) -2-chloromandelonitrile. Solid bar: wild type; gray bar: T75A; crossbar: N65Y; open bar: T75A / N65Y.

(Wild-type millipede-derived hnl gene sequence)
The millipede-derived hnl gene can be easily obtained from a millipede-derived gene using primers encoding a conserved amino acid sequence, as described in WO2017 / 150560. As an example, it can be obtained by performing a PCR reaction using a gene obtained from a millipede as a template and primers having the following sequences. By analyzing the base sequence of the obtained millipede hnl gene, the sequence of the wild-type millipede-derived hnl gene can be determined.
HNL-FW: CTGCAACTGCATTGGAMATTCAAGG (SEQ ID NO: 76),
HNL-RV: ATGAATCTTRTCRCCGTTTGGAAC (SEQ ID NO: 77)
HNL-FW2: SSAACTGCATTGGAYATMMRMAGG (SEQ ID NO: 78)
HNL-RV2: ATGAATCTRTCRCCRTTTGGRAC (SEQ ID NO: 79)
Alternatively, when using the millipede HNL specifically disclosed in the present specification, SEQ ID NOs: 10 to 21 of the present specification can be used as the wild-type millipede-derived hnl gene sequences.

(Wild type millipede-derived HNL protein)
The millipede-derived HNL protein is obtained by inserting the millipede hnl gene obtained above into an expression cassette suitable for expression in a host, if necessary, inserting it into a vector, transforming the host cell, and culturing the host cell. Thus, it can be obtained by being expressed. In addition, an amino acid sequence encoded from the determined wild-type millipede hnl gene sequence can be obtained.

(A mutant millipede-derived hnl gene and a mutant millipede-derived HNL protein)
A mutant millipede-derived HNL protein can be obtained by the following procedure. First, the amino acid sequence constituting the wild-type millipede HNL protein determined as described above is aligned with the already obtained amino acid sequence of the millipede HNL (FIG. 4). From the alignment results, the amino acid sequences that make up the eight anti-equilibrium β sheets known for the other millipede HNL are determined. Further, the amino acid sequence constituting the β-sheet structure located at the third (β3), fourth (β4), and fifth (β5) counting from the N-terminal side is determined. selected from the second amino acid in the amino acid sequence constituting β3, the fifth and seventh amino acids in the amino acid sequence constituting β4, and the second and third amino acids in the amino acid sequence constituting β5 Design a structure that replaces one or more amino acids with another amino acid. Based on the designed amino acid sequence, a primer containing the mutated amino acid is designed, and PCR is performed using the primer and the original wild-type hnl gene as a template, thereby obtaining a mutated millipede-derived hnl gene (mutated HNL protein Encoding nucleic acid molecule). The vector having the nucleic acid molecule encoding the mutant HNL protein is obtained by inserting the obtained mutant millipede hnl gene into an expression cassette suitable for expression in a host, if necessary, and then inserting it into a vector. be able to. By transforming the obtained vector into a host cell, a transformed cell transformed with the vector can be obtained. A transformed HNL protein can be obtained by culturing the transformed cells to express the target protein.

(Cyanohydrin synthesis method)
In one aspect, the present invention relates to a method for producing cyanohydrin, comprising reacting a ketone or aldehyde with a cyanide compound in the presence of a (R) -HNL mutant protein. The synthesis of cyanohydrins from aldehydes or ketones can be performed, for example, with reference to Dadashour et al. (2015) (supra). Specifically, the (R) -HNL mutant protein is added to a citrate buffer containing an aldehyde and a cyanide, mixed, reacted at 25 ° C. for 3 minutes, and mixed with n-hexane and 2-propanol. By vigorous mixing, cyanohydrin can be obtained in the organic phase.

有機 If necessary, an organic solvent may be added to a citrate buffer (pH 4.0) to cause a reaction. Such organic solvents include ethyl acetate (EA), diethyl ether (DEE), methyl-t-butyl ether (MTBE), 2-isopropyl ether (DIPE), dibutyl ether (DBE), methanol (Met), and hexane. (Hex), preferably DIPE, DEE, MTBE, DBE and Hex, more preferably DIPE, DEE, MTBE and DBEE.

As the (R) -HNL mutant protein, a purified protein can be used, or a crushed cell or a crude product thereof can also be used. The amount of the enzyme used for cyanohydrin synthesis is not particularly limited as long as it is an enzyme capable of catalyzing the reaction. For example, the amount is 1 to 100 U, 1 to 50 U, 1 to 10 U, 2 to 8 U, and 3 to 5 U. Can be. Further, the pH of the citrate buffer can be, for example, 3 to 7, 3 to 6, 3 to 5, 3.5 to 5, and 3.5 to 4. Further, the reaction temperature is preferably a temperature suitable for the enzymatic reaction, in which the production of racemic cyanohydrin independent of the enzymatic reaction is suppressed, for example, 0 to 50 ° C, 15 to 35 ° C.

The aldehyde or ketone can be selected according to the structure of the cyanohydrin to be synthesized. For example, R ¹ and R ² are a hydrogen atom (however, only one of R ¹ and R ² ), an optionally substituted C1-18 linear or branched alkyl group, or an optionally substituted It may be a 5- to 22-membered aromatic group (including a heteroaromatic group having 1 to 4 atoms selected from N, O and S). Examples of the substituent which may be substituted include an amino group, an imino group, a hydroxyl group, a C1-22 linear or branched alkyl group (only in the case of an aromatic group substituent), a C1-8 alkoxy group, and a halogen atom. Atom, allyloxy group, carboxyl group, C3-20 cycloalkyl group (halogen atom, hydroxyl group, C1-8 linear or branched alkyl group, and / or C2-8 linear or branched alkenyl group A 5- to 22-membered heteroaromatic group having at least one atom selected from N, O and S (a halogen atom, a hydroxyl group, a C1-8 straight-chain or branched Alkyl group and / or C2-8 linear or branched alkenyl group). The number of substituents when they may be substituted may be one or more, and may be two, three, four, five or more. For example, aldehydes or ketones include formaldehyde, acetaldehyde, propionaldehyde, butanal, pentanal, hexanal, heptanal, octanal, nonanal, decanal, formic acid, vinyl aldehyde, benzaldehyde, 2-chlorobenzaldehyde, cinnamaldehyde, perylaldehyde, vanillin, glyoxal And the like. The concentration of the aldehyde or ketone can be 0.01 mM to 5M, 0.1 mM to 1M, or 1 mM to 100 mM.

ナ ト リ ウ ム As the cyanide compound, sodium cyanide and potassium cyanide can be used. The amount of cyanide used can be, for example, 0.1 mM to 10 M, 0.2 mM to 2 M, or 2 mM to 200 mM.

The obtained cyanohydrin can be further purified by chiral HPLC or the like, if necessary.

変 換 The formed (R) -Man and (R) -2-Cl-Man conversion and ee can be analyzed by chiral HPLC. The conversion can be calculated using the standard curve of the corresponding substrate. ee can be determined by calculating the peak area of the two enantiomers using equation (1) below. In the following formula, R represents the concentration of R-form and S represents the concentration of S-form.

Hereinafter, the present invention will be described more specifically based on examples. However, the present invention is not limited to these examples. All documents cited throughout the application are incorporated herein by reference in their entirety. In the following examples, the amino acid numbers of HNL indicate amino acid positions in the mature protein not containing the signal sequence. Therefore, the numbers indicating the same amino acids are different between FIG. 4 and the sequence listing and the following examples.

(Example 1) Structural analysis of Chua HNL (enzyme assay for synthesis of mandelonitrile (MAN))
Racemic MAN was purchased from Sigma-Aldrich (St. Louis, MO, USA). (R) -MAN synthesis activity by HNL was measured as previously reported (Dadashipour, M. et al., (2015) Proc. Natl. Acad. Sci. USA 112: 10605-10610. ). Briefly, enzyme samples were added to 1 ml of reaction buffer (400 mM citrate buffer (pH 4.2), 50 mM benzaldehyde and 100 mM potassium cyanide) and the mixture was incubated at 22 ° C. for 5 minutes. The reaction was then stopped by mixing with 9 volumes of n-hexane: 2-propanol (85:15). Finally, an HPLC apparatus (UFLC Promence Liquid Chromatograph LC-20AD, Shimadzu, Kyoto, Kyoto) equipped with a chiral column (CHIRALCEL OJ-H, Daicel, Tokyo, Japan) (inner diameter 4.6 mm × 200 mm long, particle size 5 μm). Used to analyze the organic layer. One unit of activity was defined as the amount of enzyme that synthesizes 1 μmol of (R) -MAN per minute from benzaldehyde and hydrogen cyanide.

(Construction of ChuaHNL in Pichia pastoris expression system)
ChuaHNL cDNA was prepared by using KOD-Plus-DNA polymerase (TOYOBO, Osaka, Japan), His-ChuaHNL-Fw primer and His-ChuaHNL-Rv primer, and using ChuaHNL cDNA (Dadashipour et al., (2015)) as a template. Amplified by PCR. The resulting PCR product was digested with XhoI and XbaI (TaKaRa) and cloned under the AOX1 promoter of the Pichia expression vector pPICZαA to obtain pPICZαA-His-ChuaHNL.

His-ChuaHNL-Fw primer (SEQ ID NO: 22):
GCGCTCGAGAAAAGAGCACATCATCATCATCATCATCATCATCATGAAAATTTTACTTCCAAGGGTCACTGACTTGTGATCAACTTCCC
His-ChuaHNL-Rv primer (SEQ ID NO: 23):
CGCTCTAGATTAGTAAAAGCAAAGCAACCGTGGGTTTTC

P. Pastoris-derived protein disulfide isomerase (PDI) -encoding gene (PpPDI, Genbank ID: AJ302014) was obtained by combining a primer pair (PpPDI-InFu-Fw primer and PpPDI-InFu-Rv primer), and KOD-Plus-Neo DNA polymerase (PDI). TOYOBO)). Pastoris genomic DNA was used as a template for amplification by PCR. The obtained DNA fragment was cloned into an EcoRI (TaKaRa) site under the AOX1 promoter in pAO815 of the Pichia expression vector using an In-Fusion HD cloning kit (Clontech Laboratories, Palo Alto, CA, USA). The obtained expression vector was linearized with StuI (TaKaRa), and the P.p. pastoris strain GS115. The obtained expression vector was linearized with SacI (TaKaRa) and transformed into cells. Transformants were confirmed by colony PCR using 5'AOX1 and 3'AOX1 primers.

PpPDI-InFu-Fw primer (SEQ ID NO: 24):
TCGAAACGAGGAATTCCATGCAATTCAACTGGGAATT
PpPDI-InFu-Rv primer (SEQ ID NO: 25):
TGTCTAAGGC GAATTCTTAAAGCTCGTGCGTGAGCCGTC
5'AOX1 primer (SEQ ID NO: 26):
GACTGGTTCCAATTGACAAGC
3'AOX1 primer (SEQ ID NO: 27):
GCAAATGGCATCTCTGACATCC

(Pichia pastoris medium and culture conditions)
Pichia pastoris GS115 strain was added to a YPDS medium (2% glucose, 2% peptone, 1% yeast extract, 2% agar and 1M) supplemented with 100-2000 μg / ml zeocin (Invitrogen, Carlsbad, CA, USA) as required. (Sorbitol) at 28 ° C. and transformed. Buffered minimal glycerol (BMG) medium (1.34% yeast nitrogen base without amino acids, 4 × 10 ⁻⁵ % biotin, 100 mM, supplemented with 0.004% histidine as needed for protein expression) Potassium phosphate buffer (pH 7.0) and 1% glycerol) or BMM medium (BMG medium in which 1% glycerol was replaced with 1% methanol) was used. Cultivation was performed at 28 ° C. under aerobic conditions with reciprocal shaking, and yeast growth was monitored by measuring optical density at a wavelength of 600 nm. E. coli HST08 strain (TaKaRa Bio, Otsu, Japan) was used for plasmid amplification. In low salt Luria-Bertani (LB) medium (1% tryptone, 0.5% yeast extract and 0.5% NaCl) supplemented with ampicillin (50 mg / ml) or zeocin (25 mg / ml) as needed, 37 E. coli was grown at ℃.

(Purification of Chua HNL protein)
Native ChuaHNL was purified by a previously reported protocol. Briefly, first, a millipede homogenate solubilized in buffer-A (20 mM potassium phosphate, pH 7.0) was fractionated by ammonium sulfate precipitation. Next, TOYOPEARL Butyl-650M (Tosoh, Tokyo, Japan), TOYOPEARL DEAE-650M (Tosoh), Q Sepharose FF (GE Healthcare, Chicago, IL, USA, Gazette, IL, USA) Of the ChuaHNL protein were purified. Finally, active fractions were collected and subjected to SDS-PAGE to evaluate purity. For protein crystallization, the buffer was replaced with 50 mM citrate buffer (pH 5.0) and stored at 4 ° C. until use.

Recombinant His-tagged ChuaHNL is a P.p. and expressed by the protein secretion system in P. pastoris. Initially, the medium after 6 days of incubation was adjusted to pH 7.5 by adding 2M sodium hydroxide. Next, the His-tag protein was purified using Ni Sepharose 6 FF resin (GE Healthcare), Mono Q 5/50 GL (GE Healthcare), and Superdex 200/10/300 GL (GE Healthcare). Finally, the active fraction was dialyzed against 10 mM citrate buffer (pH 5.5) and stored at −20 ° C. until use.

(Crystallization, data collection, structure determination)
All chemicals for protein crystallization were purchased from Hampton Research (Aliso Viejo, CA, USA). Native and recombinant ChuaHNL protein solutions were concentrated to 10 mg / ml using Amicon Ultra Centrifugal Filter Units NMWL, 10 kDa (Merck Millipore, Billerica, MA, USA). The protein concentration was measured using Quick Start Protein Assay (Bio-Rad Laboratories, Hercules, CA, USA) using bovine serum albumin (Sigma-Aldrich) as a standard. Initial crystallization conditions were screened using CrystalScreen I and II (Hampton Research), 0.2 M ammonium sulfate, 0.1 M sodium acetate (pH 4.6), 30% (w / v) polyethylene glycol (PEG) monomethyl Dihedral crystals were obtained under ether 2,000 conditions. Finally, hanging at 20 ° C. for 3 days under the conditions of 0.2 M ammonium sulfate, 0.1 M sodium acetate (pH 5.0), 28-32% (w / v) PEG monomethyl ether 2000, and 0.3 M NDSB-195. The best crystals were obtained using the drop vapor diffusion method.

For the determination of the phase, before freezing and data collection, native Chua HNL crystals were immersed in a storage solution containing 0.5 M sodium iodide at 20 ° C. for 30 minutes. For native and recombinant structures complexed with acetate, the resulting crystals were used directly for data collection. For the ligand-free form, the crystals obtained from the native protein were immersed in immersion solution-1 (32% (w / v) PEG monomethyl ether 2,000, 0.3 M @ NDSB-195, 50 mM bis-tris-propane, 50 mM The solution was immersed in citric acid (pH 4.5) at 20 ° C. for 20 minutes to remove the acetate. For complex structures with thiocyanate, iodoacetate and cyanide ions, the ligand-free form was included in a dipping solution-1 containing 80 mM potassium thiocyanate, 10 mM sodium iodoacetate and 0.5 M potassium cyanide, respectively. The crystals were soaked.

結晶 All crystals were cryoprotected with perfluoroether and the data set was collected at 100K under a stream of nitrogen. X-ray diffraction data for the single anomalous dispersion (SAD) phase was collected on an X-ray generator and imaging plate (MicroMAX-007 and R-AXIS @ VII, Rigaku, Tokyo, Japan). Other data include Photon Factory Beamline BL-1A and BL-5A (Tsukuba, Japan) silicon pixel detectors (Pilatus 2M-F, DECTRIS, Baden-Daettwil, Switzerland) and CCD detectors (Quantum 315r, Area Detector). , Poway, CA, USA).

All data sets were integrated using XDS (Kabsch, W. (2010) Acta Crystallogr. D Biol. Crystallogr. 66, 125-132) and SCALA (Winn, MD et al. (2011) Acta Crystallogr. .D @ Biol.Crystallogr. 67, 235-242). The initial phase was determined using SHELX suite (Sholdrick, GM (2010) Acta Crystalgraphica Section D D Biological Crystallography 66, 479-485). All models were modified using COOT (Emsley, P. et al., Acta Crystallographica Section D D Biological Crystallography 66, 486-501), and REFMAC5 (Murshudov, G.N. alt. $ 53, $ 240-255) or PHENIX (Afone, PV, et al., (2012) Acta Crystallogr. D D Biol. Crystallogr. 68, 352-367).

(Structural analysis)
All structures were verified using MolProbity (Chen, VB et al., (2010) Acta Crystallographic Section D Biological Crystallography 66, 12-21), and the presence of Ramachandran plots where residues were not permitted in Ramachandran plots. Did not. The statistics of the phase analysis are shown in Table 1 and the statistics of data collection and structure improvement are shown in Table 2. Structural analysis was performed using PISA (Krissinel, E. et al., (2007) J. Mol. Biol. 372, 774-797). A diagram of the protein structure was created using the PyMOL program (http://www.pymol.org).

(Overall structure of Chua HNL)
The crystal structure of native Chua HNL purified from millipede was determined by diffracting to an internal X-ray source with an iodine-impregnated Chua HNL crystal to a resolution of 2.1 ° by the SAD method (Table 1). One crystal was present in the P unit (Table 2 and FIG. 1A). Amino acid residues were numbered from the N-terminal Leu residue of mature ChuaHNL with the signal peptide removed. The electron density maps of Leu1 at the N-terminus and Tyr162 at the C-terminus of mature ChuaHNL were clearly confirmed. ChuaHNL contained two α- helices, three _{3 10} helix, and eight β sheet (Figure 1). Eight β-sheets formed an anti-parallel β-barrel containing a central active site cavity (FIG. 1B).

Although a sequence homologous to ChuaHNL was not found by BLAST search, ChuaHNL was found to be a lipocalin protein by a structural comparison search using a Dali server (Holm, L. et al., (2010) Nucleic Acids Res. $ 38, @ W545-549). It turned out to be similar. Many typical lipocalins give a Z score higher than 10, indicating significant structural similarity. The overall amino acid sequence identity of ChuaHNL to a typical lipocalin was less than 8%. Lipocalins are generally three structurally and sequencely conserved known as structurally-conserved regions (SCR) 1-3 (Flower DR (1996) Biochem. J. 318 (Pt1), 1-14). Including motifs. Despite the very low amino acid similarity in the SCR, the secondary structure of ChuaHNL overlapped well with that of human retinol binding protein 4.

(Binding of ligand to the active site of Chua HNL)
The acetate-binding form of the acetate-binding ChuaHNL structure was measured at 1.5 ° resolution. Acetate contained in the reservoir solution was observed at the active site. The carboxy oxygen formed a salt bridge with Arg38 and Tyr40 (FIG. 2A).
Ligand-free form The ligand-free form of the ChuaHNL structure was measured at 1.6 ° resolution. The crystals were soaked in a solution containing bis-trispropane-citrate buffer to remove acetate bound to the active site. Three water molecules were observed at the active site (FIG. 2B). These had formed hydrogen bonds with Arg38, Tyr103, and Lys117.
The crystal structure of ChuaHNL that formed a complex with cyanide ion-bonded cyanide ions was measured at a resolution of 2.1 ° (FIG. 2C). The orientation of the cyanide ion was determined by the temperature coefficient and possible electrostatic interactions between the negatively charged carbon and Arg38. The nitrogen atom of the cyanide ion formed a hydrogen bond with Try40-Oη at a distance of 3.3 °. The negatively charged carbon of the cyanide ion was electrostatically interacting with Arg38-Nη1 at a distance of 4.2 °.
Inhibitor-bound We have previously reported that iodoacetate and thiocyanate inhibit the ChuaHNL-catalyzed (R) -MAN synthesis reaction (Dadashipour et al. (2015) supra). The structure of each of ChuaHNL in which a complex of iodoacetate and thiocyanate was formed was measured at a resolution of 1.55 °. In the complex structure with iodoacetate, the carboxy oxygen formed a salt bridge with Arg38-N1η and N2η at distances of 3.1 ° and 3.0 ° and Lys117-Nε at a distance of 3.1 ° (FIG. 2D). ). Iodine atoms of iodoacetate may form weak π interactions with Phe67 at a distance of 4.3 °. A strong anomalous difference map between Phe67 and Arg38 indicated the presence of an iodine atom.
In the complex structure with thiocyanate, two alternative binding modes were observed. The orientation of the two thiocyanate molecules in the crystal structure was determined by the lower B-factor value after structure refinement. In one bonding mode (FIG. 2E, SCN1), the sulfur atom of the thiocyanate formed a hydrogen bond with Tyr103-Oη and Arg38-Nη1 at a distance of 3.0 ° and 3.2 °, respectively. In another mode of attachment (FIG. 2E, SCN2), the sulfur atom of the thiocyanate formed a salt bridge with Arg38-Nη2 and Lys117-Nε at a distance of 3.1 ° (FIG. 2E).

(Docking simulation of (R) -MAN)
To find out how ChuaHNL catalyzes the synthesis and cleavage reactions of cyanohydrin, MOE (Molecular Operating Environment, version 2016.8; (Chemical Computing Group, Montreal, Canada) to the active site of ChuaHNL by ChuaHNL). -A docking simulation of MAN was performed, showing the binding of (R) -MAN in the cavity of the active site in the model giving the most prominent affinity score (S = -5.44) (R)- The hydroxyl and nitrile groups of MAN interacted with the hydrophilic residues of ChuaHNL at the bottom of the active site.The benzene group of (R) -MAN had many hydrophobic and aromatic residues (Ile11, Phe17, Phe25, Ala54, P e67, Ala75, Leu77, Trp88, Phe90, Ala105 and Ala119) The deep hydrophobic cavities accept various bulky cyanohydrins, as previously observed (Dadasipour et al. (2015) supra). Thought to be helpful.

The interaction between the hydrophilic residue and (R) -MAN is shown in FIG. 3B. The hydroxyl group of (R) -MAN formed a hydrogen bond with Lys117-Nε and Arg38-Nη1 at a distance of 3.3 ° and 3.3 °, respectively. The nitrile group formed hydrogen bonds with Arg38-Nη1, Arg38-Nη2 and Tyr103-Oη at distances of 3.4, 3.3 and 3.3 °, respectively. Asp56 forms a salt bridge with Arg38-Nη2, Arg38-Nε and Lys117-Nε. Tyr40-Oη forms a hydrogen bond with Arg38-Nη1 and Tyr103-Oη. The bond lengths and orientations of the residues containing ligand binding were nearly identical in the five ChuaHNL structures.

(Example 2) Production of OgraHNL mutant and activity measurement (compound)
All compounds were purchased commercially. Benzaldehyde and (R / S) -Man were purchased from Sigma-Aldrich. 2-chlorobenzaldehyde was purchased from Tokyo Chemical Industry. (R / S) -2-Cl-Man was synthesized according to the method of Alagoz et al. (Alagoz D et al., Enzymatic (2014) 101: 40-46).

(Culture and expression of recombinant OgraHNL)
OgraHNL was produced using a previously constructed transformant of the E. coli SHuffle T7 strain carrying the plasmid pET28a containing the ograhn1 gene (Yamaguchi T et al., Scientific Reports (2018) 8 (1): 3051). The recombinant E. coli cells are inoculated into 5 ml of LB medium containing kanamycin (50 μg / ml), cultured at 30 ° C. for 16 to 18 hours while shaking at 250 rpm, and then cultured in LB medium containing kanamycin (50 μg / ml). Transferred to 500 ml. After culturing at 30 ° C. for 6 hours with shaking at 150 rpm (OD600 = 0.6-0.8), isopropyl β-thiogalactoside (IPTG) was added to a final concentration of 1 mM, and the culture was cooled to 16 ° C. For 20 hours. The cells were collected by centrifugation at 4,500 × g at 4 ° C. for 10 minutes. Harvested cells were resuspended in 20 mM potassium phosphate buffer (KPB, pH 7.4) containing 500 mM sodium chloride and 20 mM imidazole and disrupted by sonication. The supernatant obtained after centrifugation at 15,000 × g for 10 minutes at 4 ° C. was used as a crude enzyme.

(Purification of recombinant OgraHNL)
First, the crude enzyme was loaded onto a Ni ²⁺ Sepharose 6 Fast Flow column (GE Healthcare, Uppsala, Sweden) and eluted with 20 mM KPB (pH 7.4) containing 500 mM sodium chloride and 300 mM imidazole. . The active fraction was collected, concentrated, exchanged with phosphate buffered saline (PBS, pH 7.3), and subjected to thrombin digestion (22 ° C., 2 U thrombin / 100 μg tag-protein for 16 hours). The His tag was removed. The protein was then loaded onto a Mono Q5 / 50GL column (GE Healthcare) and eluted with a 0-50% gradient of elution buffer (20 mM KPB and 500 mM NaCl, pH 7.4) running at a flow rate of 1 ml / min. All purification steps were performed at 0-4 ° C. Protein purity was assessed by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The protein concentration was measured with a BCA assay kit (Thermo Fisher Scientific, MA, USA) using bovine serum albumin as a standard protein.

(Crystallization and structure determination)
Screening for optimal crystallization conditions was performed at 20 ° C. using 96-well Intelli-Plates (Art Robbins instruments, CA, USA) using the sitting drop vapor diffusion method. Droplets were prepared by mixing an equal volume (1 μL) of the reservoir solution with a 10 mg / mL protein solution of OgraHNL. Screening was performed using Index ^™ -HR2-144 (Hampton Research, CA, USA). Single crystals appeared after 1 day in 0.1 M BIS-TRIS (pH 5.5) containing 2.0 M ammonium sulfate. OgraHNL crystals were continuously immersed in stock solutions containing 10%, 20% and 25% (v / v) glycerol for approximately 30 seconds each. The crystals were then immersed in a 25% (v / v) glycerol cryoprotectant solution containing one drop of 2-chlorobenzaldehyde for 2 hours, immersed in potassium cyanide (KCN) for 5 minutes, and then flash-cooled in a stream of liquid nitrogen.

Diffraction data was collected at cryogenic temperature using a Rigaku Micro-Max007CuKα rotating anode X-ray generator and a Rigaku R-AXISVII image plate detector. Indexing and integration of the diffraction data is performed by XDS (Kabsch W (2006) Crystallographic of biological macromolecules: 218-225), and scaling is performed by Scala (Evans P (2006) Acta crytal sig. 72-82) with the CCP4 program suite (Winn MD et al. (2011) Acta Crystallographic Section D: Biological Crystallography; 67 (4): 235-242). Data were treated as hexagonal in space group P63. The initial stage was determined by the Molrep (Vagin A et al. (2000) Acta Crystallographic Section D: Biological Crystallography; 56 (12): 1622-1624) using the crystal structure obtained with Chua HNL as a template and the Molrep of the CCP4 program suite. . Model building and structural improvement were performed by Coot (Emsley P, Cowtan K (2004) Acta Crystallographica Section D: Biological Crystallogography; Biological Crystallography; 53 (3): 240-255). R _free value was calculated from 5% of randomly selected that were not used for the refinement reflection (Kleywegt GJ et al (1996) Structure; 4 (8 ): 897-904). Water molecules are located in different electron density map in both manual and automatic, are rejected either retained on the basis of the geometrical criteria and Purified factor B thereof (B <60Å ^2). All structural diagrams were prepared by PyMol (Schrodinger L: The PyMOL molecular graphics system, version 1.3 r1. Schrodinger, LLC, Portland, OR. In .; 2010.).

(Docking simulation using MOE program)
In order to predict the structure of the OgraHNL mutant complexed with (R) -2-Cl-Man, MOE was used using the crystal structure of OgraHNL complexed with (R) -2-Cl-Man as a template. Docking simulation was performed. The protein remained rigid. The position and orientation of (R) -2-Cl-Man and the two torsion angles including the chlorophenyl and OH groups were made variable during the simulation. The docking simulation was performed with a Compute-Simulation dock program. All parameters were according to MOE software default settings.

(Site-directed mutagenesis of OgraHNL)
To generate all OgraHNL mutants, use pET28a-OgraHNL as a template and replace the site of interest with mutagenic primers (Table 3) and QuikChange II kit (Agilent Technologies, Santa Clara, CA, USA). Was introduced to perform site-directed mutagenesis. The reaction mixture consisted of 1 μl of 10 × reaction buffer, 0.2 μl of dNTP mixture, 8.2 μl of distilled water, 0.5 U of PfuTurbo DNA polymerase, 10 ng / μl of sense and antisense primer 0.2 μl each, and 50 ng / μl of pET28a-OgraHNL vector (0.2 μl) was used as template DNA. Denaturation (95 ° C., 30 seconds), annealing (55 ° C., 1 minute), and extension (68 ° C., 6 minutes) were performed for 16 cycles. The product was treated with DpnI (10 U) for 1 hour at 37 ° C., and then transformed into E. coli SHuffle T7. Production and purification of all mutant enzymes were performed in the same manner as in the wild type described above.

(HNL activity assay)
The synthesis of mandelonitrile from benzaldehyde was analyzed by HPLC using a chiral column according to a previous report (Dadashipour et al. (2015) supra) with minor modifications. The enzyme sample (140 μl) was added to 110 μl of 300 mM citrate buffer (pH 4.0) containing 50 mM benzaldehyde and 100 mM KCN, mixed and incubated at 25 ° C. for 3 minutes. To stop the reaction, 100 μl of the reaction was mixed vigorously with 900 μl of n-hexane: 2-propanol = 85: 15 to extract the obtained (R) -Man. 10 microliters of the obtained organic phase was applied to a CHIRALCEL OJ-H column (particle size 5 μm, inner diameter 4.6 mm × 250 mm, Daicel Corporation, Tokyo, Japan) according to the method described in Dadashour et al. (2015) (supra). ) Were analyzed using a UFLC Prominence Liquid Chromatograph LC-20AD connected to a Prominence UV-Vis detector SPD-20A (Shimadzu, Kyoto, Japan) equipped with:
Mobile phase n-hexane: 2-propanol (85:15)
Flow rate 1 ml / min OJ-H column oven temperature 30 ° C
Absorbance Absorbance at 254 nm The reaction that proceeded linearly during the first 3 minutes was used for activity calculations. One unit of activity was defined as the amount of enzyme that produced 1 μmol of optically active mandelonitrile per minute from benzaldehyde under the assay conditions.

To synthesize 2-Cl-Man from 2-chlorobenzaldehyde, add enzyme sample (150 μl) to 100 μl of 300 mM citrate buffer (pH 3.5) containing 50 mM 2-chlorobenzaldehyde and 60 mM KCN, mix, Incubate at 25 ° C for 5 minutes. To stop the reaction, 1 ml of n-hexane: 2-propanol (95: 5) was added to the reaction mixture, then mixed vigorously and centrifuged at 15,000 × g for 5 minutes. 5 μl of the organic phase were analyzed by HPLC equipped with a CHIRALPAK IC column (particle size 5 μm; inner diameter 4.6 mm × 250 mm; Daicel) under the following conditions:
Mobile phase n-hexane: 2-propanol (95: 5)
Flow rate 1ml / min Column oven temperature 30 ℃
Absorbance 220nm
Enzyme activity was calculated from the linear curve obtained from the first 5 minutes of the reaction. One unit of synthetic activity was defined as the amount of enzyme that produced 1 μmol of optically active 2-chloromandelonitrile per minute from 2-chlorobenzaldehyde under the assay conditions.

変 換 The formed (R) -Man and (R) -2-Cl-Man conversion and ee were analyzed by chiral HPLC. The conversion was calculated using the standard curve of the corresponding substrate. ee was determined by calculating the peak area of the two enantiomers using the following formula:

Here, R and S represent the concentrations of (R) -Man or (R) -2-Cl-Man and (S) -Man or (S) -2-Cl-Man, respectively.

(Reaction rate analysis)
For kinetic analysis, the initial rates of enzymatic conversion of benzaldehyde to (R) -Man and 2-chlorobenzaldehyde to (R) -2-Cl-Man were determined according to the method described above. Assays were performed using various concentrations of substrate (0-60 mM) and a total amount of 0.125 U of enzyme in each reaction in a salt buffer system.

(Effect of organic solvent on activity and enantioselectivity)
Organic solvents of ethyl acetate (EA), diethyl ether (DEE), methyl-t-butyl ether (MTBE), 2-isopropyl ether (DIPE), dibutyl ether (DBE), methanol (Met), and hexane (Hex) 0.5 ml was mixed with an equal volume of citrate buffer (50 mM, pH 7.0) and equilibrated by shaking at 1200 rpm at 10 ° C. for 60 minutes. The enzyme (2.5 U) was added to the citrate buffer phase and mixed gently without disturbing the interface between the aqueous and organic phases. At time 0 and 12 hours after incubation, enzyme activity was measured by shaking at 1,200 rpm at 10 ° C. as described above.

(Crystal structure and reaction mechanism of OgraHNL)
The primary structure of OgraHNL has 73% amino acid identity with ChuaHNL (FIG. 4). Using ChuaHNL as a template structure, the crystal structure of OgraHNL complexed with (R) -2-Cl-Man was determined by a molecular replacement method at 2.05 ° resolution. This enzyme was a homodimer and the subunit fold was similar to that of ChuaHNL (FIG. 5a). The secondary structure of each subunit consisted of two α helices, two 3 ₁₀ α helices, and eight antiparallel β sheets, forming a beta barrel with an outer alpha helix. Three intramolecular and two intermolecular disulfide bonds were observed in the crystal structure. In each subunit, the active site is located at the center of the beta barrel surrounded by all eight beta sheets. The relatively large hydrophobic cavities within the active site of OgraHNL are: Val at position 15; Phe at positions 29, 71 and 94; Tyr at positions 44 and 107; Ala at positions 58, 79 and 109; Leu at position 81; It consisted of 11 hydrophobic residues of the 92nd Trp (FIG. 5b). Four hydrophilic residues of Arg42, Asn69, Asp60 and Lys121 were also exposed on the active site surface. The substrate entry tunnel introducing the external solvent into the active site is clearly shown in FIG. 5b, and there were certainly two electron densities in the tunnel corresponding to water. The tunnel entrance region was formed by eight residues Val15, Pro16, Phe21, Tyr44, Phe71, Ala79, Trp92, and Phe94. Some hydrophobic residues in the active site appeared to form a hydrophobic interaction with the benzene moiety to recognize the benzene ring of (R) -2-Cl-Man. Two of these residues, Phe71 and Phe94, are located near the phenyl ring of (R) -2-Cl-Man and have π-π in edge-to-face and face-to-face, respectively. Form stacking interactions. π-π stacking interactions are stronger than general hydrophobic interactions and may play an important role in substrate binding (Yang ST et al., Nanotechnology (2008) 19 (39): 395101; and Nakano S). Et al., Biochim Biophys Acta (2014) 1844 (12): 2059-2067). The existence of a π-π stacking interaction has also been reported by Nakano et al. (2014) (supra) in (S) -HNL from Baliospermum tantanum.

As shown in FIG. 5c, the side chains of Arg42 and Lys121 of OgraHNL form a hydrogen bond with a hydroxyl group of (R) -2-Cl-Man, and Tyr107 is a nitrile group of (R) -2-Cl-Man. Forms a hydrogen bond with Asp60 does not interact directly with (R) -2-Cl-Man, but forms a hydrogen bond with Arg42 and Lys121. From this complex structure analysis, other R-specific HNLs in which the conserved histidine acts as a general acid / base to deprotonate the hydroxyl group of the substrate (Dreveny I et al., Protein Sci (2002) 11 (2): 292) -300; unlike Zhu W et al., Proteins (2015) 83 (1): 66-77; and Motojima F et al., FEBS J (2018) 285 (2): 313-324), and the general acids / by Arg42 and Lys121. A reaction mechanism dependent on the base catalyst (FIG. 5d) was suggested. In the cleavage reaction, the proton of the hydroxyl group of (R) -2-Cl-Man is removed by Lys121. Electrons transfer from the lysine to the nitrile leaving group, and the cyanide ion then abstracts a proton from Arg42, releasing 2-chlorobenzaldehyde and hydrogen cyanide. In a synthetic reaction, a series of reactions takes place in the opposite direction of cyanohydrin cleavage. After hydrogen cyanide has been deprotonated by Arg42, the cyanide ion attacks the carbonyl carbon of 2-chlorobenzaldehyde. The carbonyl oxygen accepts electrons from cyanide and produces (R) -2-Cl-Man.

The hydrophobic cavity within the active site of OgraHNL is predicted to play a significant role in substrate selectivity, but away from the proton abstraction site, and its mobility is reported in (S) -HNL from Balospermum @ montanum. This allows the rearrangement of the phenyl group of the substrate in the catalyst (Nakano @ S et al. (2014) supra). This cavity binds to its phenyl group when (R)-and (S) -2-Cl-Man enter the active site and is split into (R) -specific and (S) -specific pockets, respectively. As reported, ChuaHNL has broad substrate specificity (Dadasipour @ M et al. (2015) supra) due to the flexible structure of ChuaHNL that allows various substrates to bind to the substrate entry tunnel. .

(Docking simulation of OgraHNL by (R) -2-Cl-Man)
Based on the crystal structure modeling of OgraHNL complexed with (R) -2-Cl-Man, amino acids that could affect substrate binding were identified by molecular docking simulation using MOE. Using two molecular structures of (R) -2-Cl-Man having chlorine atom substitution at different ortho positions of the benzene ring for docking (FIGS. 6a and b), a total of 300 kinds of bonding structures for each design were obtained. I derived. Substrate docking analysis found that Ala58, Ala79, Phe71 and Val15 were the most potential residues based on the binding affinity score. These hydrophobic residues were located in the hydrophobic region of the active site, with their side chains facing the substrate. Therefore, kinetics in this region is expected to affect the substrate specificity of OgraHNL. Ala79, Phe71 and Val15 are located at the substrate tunnel entrance and can play an important role in affinity for the substrate. On the other hand, the side chain of A58 was in close contact with the chlorine atom at the ortho position of the aromatic ring of (R) -2-Cl-Man (FIG. 6b), suggesting the possibility of steric interaction. . This steric interaction was probably responsible for the decrease in enzyme performance observed with this substrate. Thus, structural analysis of the enzyme suggested that these four residues could be important amino acids to improve the enzymatic properties and / or enantioselectivity of OgraHNL. Site-directed mutagenesis of the eight mutations A58C, A58H, A58R, A58V, A79M, F71A, F71I and V15W, obtained as preferred 3D docking structures with the substrate entrance tunnel (FIGS. 3c-d) open. Study was carried out.

(Site-directed mutagenesis of OgraHNL)
Mutations in the hydrophobic cavity of the active site affected the activity and enantioselectivity of OgraHNL. When using the purified enzyme, mutation of Ala79 to methionine (A79M) resulted in a significant increase in both specific activity and selectivity. Mutant A79M exhibited the highest specific activity of (400 ± 13 U / mg) and (ee) value of 83.2 ± 0.1% with respect to (R) -2-Cl-Man, and showed the wild-type OgraHNL (288 The activity was higher than ± 11 U / mg and 69.5 ± 0.5% ee) (10 and 1.2 U / ml, respectively) (Table 4). The specific activity and enantioselectivity of A79M were significantly improved not only for (R) -2-Cl-Man but also for (R) -Man as compared to the wild type. Regarding (R) -Man synthesis, the specific activity of wild type was 2,780 ± 80 U / mg and ee of 82.2 ± 0.6%, while the specific activity of A79M was 3,310 ± 78 U / mg. And 93.6 ± 0.3% ee.

(Site-directed mutagenesis of Ala79 in OgraHNL)
To examine the effect of position 79 of OgraHNL on catalytic activity and enantioselectivity, an additional 18 amino acids were substituted for Ala residues by site-directed mutagenesis, and for all mutants, the HNL productivity, ratio Activity, enantioselectivity and conversion were determined. Mutation replacing Ala at position 79 with most hydrophobic residues with large side chains, including Ile, Leu, Met, Phe, Val and Ser, improved the productivity of OgraHNL for (R) -Man. In addition, the mutation replacing Ala at position 79 with Met, Val, Ser and Cys improved the productivity of (R) -2-Cl-Man (Table 4). For both (R) -Man and (R) -2-Cl-Man, mutants A79C, A79I and A79M showed the highest specific activity, ee and conversion (Table 4).

(Enantioselectivity of OgraHNL mutant)
To determine if the enzyme with the mutation in Ala79 has high enantioselectivity, synthesis of (R) -Man from benzaldehyde and KCN and (R) -2-Cl-Man from 2-chlorobenzaldehyde and KCN Was synthesized using purified enzymes (10 and 2 U / ml, respectively). The conversion of benzaldehyde and 2-chlorobenzaldehyde was the same in the wild type and the mutant (not shown). Of the three mutants, A79C showed the highest ee of 97.1 and 83.9% for (R) -Man and (R) -2-Cl-Man synthesis, respectively. The ee values for (R) -Man and (R) -2-Cl-Man synthesis were 85.5 and 69.0% for the wild type, respectively, while 93.3 for the A79M and A79I, respectively. 93.6 and 83.5-83.6%. These results indicate that mutation of Ala79 contributes to significantly improving ee in both (R) -Man and (R) -2-Cl-Man production.

The productivity and ee value of (R) -Man and (R) -2-Cl-Man increased with increasing amount of enzyme. The total production of mandelonitrile and 2-chloromandelonitrile was similar for wild type and mutant, approximately 35-38 and 24-26 mM, respectively. All mutants catalyzed the synthesis of (R) -Man with a maximum ee of 98.4-98.7%, which was higher than the wild type (95.6%). For (R) -2-Cl-Man synthesis, variant A79C showed a maximum ee of 96.3%, followed by A79M at 95.5% and A79I at 95.0%. All of these values were significantly higher than those obtained from the wild type (92.8%). This represents a surprising improvement of (R) -Man and (R) -2-Cl-Man by the mutant. Thereby, mutations in Ala79, especially mutations to cysteine, isoleucine or methionine, can increase the enantiomeric excess in the synthesis of (R) -Man and (R) -2-Cl-Man, and It has been shown to enable the production of pure cyanohydrin. In the best case reported previously, 99% ee (R) -Man was obtained by ChuaHNL in citrate buffer (pH 2.7) (Dadasipour @ M et al. (2015) supra). The amount of enzyme performed was higher (150 U / ml). For (R) -2-Cl-Man, Chua HNL and other reported HNLs show very low enantioselectivity in buffer systems (ee <21%) (Dadashipour @ M et al. (2015) supra; Yildirim). D et al., Biotechnol @ Prog (2014) 30 (4): 818-827). Thus, the present invention has a high enantiopurity (96.3%) in a single-phase buffer system, unlike other (R) -HNL, which catalyzes these reactions under a two-phase system that usually contains an organic solvent. Asymmetric synthesis of (R) -2-Cl-Man was realized for the first time.

As described above, among the mutants subjected to the site-directed mutagenesis, the ee values for (R) -Man were superior to those of the wild type in A58C, A58H, F71I, A79C, A79I, and A79M. In addition, A79C and A79M were better in specific activity than the wild strain, and A79F, A79I, A79L, A79M, A79S, and A79V were better in productivity than the wild strain. The ee values for (R) -2-Cl-Man were superior to the wild type in A79C, A79I, and A79M, and those having specific activity superior to the wild type were A79I. A79M, A79M, A79S, and A79V which were superior in productivity to the wild type were A79M.

(Effect on activity of organic solvent and enantiomeric activity)
For all wild-type and mutants, most organic solvents had a significant effect on enzyme activity and enantioselectivity, whereas DIPE alone had little effect on enzyme activity, giving the highest% ee. (Not shown). A79C and A79M were more stable than wild type under DEE, MTBE and Hex conditions. The residual activity of all wild-type and mutants was more than 50% in all organic solvents. All mutants showed more than 50% ee under DEE, MTBE, DIPE, DBE and Met conditions, but less than 50% ee under EA and Hex conditions. Since EA and Hex exhibited low values of the partition coefficient, benzaldehyde and 2-chlorobenzaldehyde were almost dissolved in the aqueous phase (Uetrongitch T et al., Journal of Molecular Catalysis B: Enzymatic (2009) 56 (4): 208-21). ). It is believed that this promoted the non-enzymatic reaction and caused the lower ee of (R) -Man and (R) -2-Cl-Man.

(Example 3) Production of PlamHNL mutant and measurement of activity (docking simulation using MOE program)
In order to predict the structure of PlamHNL complexed with (R) -2-chloromandelonitrile, OgraHNL complexed with (R) -2-chloromandelonitrile was used as a template for docking simulation using MOE analysis. The structure of was used. The homology modeling of PlamHNL complexed with (R) -2-chloromandelonitrile calculates about 14 amino acid residues in the binding pocket using the Alanine and Residue Scanning function and assigns the other 19 amino acids an affinity. Important residues. Mutants with the lowest dAffinity were selected for validation.

(Site-directed mutagenesis)
PlamHNL mutants were prepared by site-directed mutagenesis using the Quick-Change site-directed mutagenesis kit using forward and reverse primers (Table 5) and pET28aPlamHNL as a template. In the PCR reaction, 18 cycles of denaturation (95 ° C., 20 seconds; first cycle: 95 ° C., 2 minutes) and annealing at 52 ° C. for 20 seconds were performed. The PCR product was treated with DpnI (10 U) at 37 ° C. for 1 hour, and then transformed into E. coli Shuffle T7 strain.

(HNL activity assay)
Enzyme activity was assayed by measuring the amount of optically active (R) -2-chloromandelonitrile using 2-chlorobenzaldehyde and potassium cyanide as substrates. A reaction mixture (0.5 mL) was prepared in a microtube. To a sodium citrate buffer (400 mM, pH 4.0) was added 1.25 M 2-chlorobenzaldehyde (20 μL in methanol), followed by an enzyme solution and a KCN solution (1.0 M, 50 μL). The reaction was monitored by taking an aliquot (100 μL) of the reaction mixture and extracting with 900 μL of organic solvent (94% n-hexane, 6% isopropanol, 0.2% TFA, v / v). After centrifuging the mixture at 15,000 × g at 4 ° C. for 10 minutes, an organic layer containing benzaldehyde, (R)-and (S) -2-chloromandelonitrile was obtained. Thereafter, an aliquot (10 μL) of the organic phase was analyzed using chiral HPLC as described above. One unit of HNL activity was defined as the amount of enzyme that produced 1 μmol / min of optically active (R) -2-chloromandelonitrile from 2-chlorobenzaldehyde and KCN under the assay conditions.

(Expression and purification of recombinant PlamHNL)
A single colony of Escherichia coli Shuffle T7 strain Express having pET28a-PlamHNL was inoculated into 5 mL of LB medium containing kanamycin (80 μg / mL) and cultured overnight at 30 ° C. and a shaking speed of 300 rpm. 5 mL of the starter culture was transferred to 500 mL of LB medium containing kanamycin (80 μg / mL) in a 2 L Erlenmeyer flask, and cultured at 30 ° C. at a shaking speed of 150 rpm. After 12 hours, IPTG was added to a final concentration of 0.5 mM, and the cells were cultured at 18 ° C at the same shaking speed for 24 hours. The cells were centrifuged (8500 × g; 15 minutes) and resuspended in potassium phosphate buffer (KPB; 20 mM, pH 7.0) containing sodium chloride (0.5 M) and imidazole (25 mM). The resuspended cells were lysed by sonication, and the lysate was centrifuged (15000 × g; 4 ° C. for 15 minutes) to remove debris. The supernatant was loaded onto a Ni Sepharose 6 Fast Flow (GE Healthcare, Little Chalfont, UK) column (25 mm ID, 20 mL column capacity), washed with 50 mM imidazole, and then KPB (20 mM) containing sodium chloride (0.5 M). Eluting with a linear gradient of 50-300 mM imidazole (pH 7.0) at a flow rate of 0.5 mL / min. Fractions showing the highest specific activity were pooled, dialyzed and loaded on a MonoQ 5/50 GL (GE Healthcare) column. Enzyme activity was measured as described above, active fractions were pooled, dialyzed, concentrated and tested for purity by SDS-PAGE.

(Homology Modeling PlamHNL Structure)
Since HNL derived from Parafonteria laminate has 52% amino acid sequence homology to Oxidus gracilis HNL, the model homology of the PlamHNL structure using the OgraHNL structure complexed with (R) -2-chloromandelonitrile as a template is examined. Calculated. The overall structure of PlamHNL was very similar to OregHNL (FIG. 7a) and consisted of the same 10 hydrophobic and 4 hydrophilic amino acid residues in the binding pocket and active site. (Table 6). All these residues on the binding pocket and active site were conserved in all millipede HNLs (FIG. 4). Homology modeling of PlamHNL indicated that, similar to Lys121 in OgraHNL, Lys118 was a key residue in the interaction of the cyanide region of (R) -2-chloromandelonitrile (FIG. 7b). .

(Verification of docking simulation with (R) -2-chloromandelonitrile of PlamHNL)
The MOE program generated a mutant of approximately 14 amino acid residues in the binding pocket by the Alanine and Residue Scanning function and identified residues that are important for affinity. Mutants were generated for the selected 13 hits. The activity of the six mutants N65H, N65Y, W89H, N65E, N65Q and T75A at the three positions N65, T75 and W89 on 2-chlorobenzaldehyde was evaluated. Was confirmed. In contrast, no activity was detected in mutants at positions R38, Y40, L78 and Y104, suggesting that these residues may be involved in the mechanism of PlamHNL activity. The best mutant T75A obtained from the alanine scanning function showed about 1.5 times higher specific activity than wild type, whereas N65H, N65Y, W89H, N65E and N65Q showed lower specific activity than wild type. . These results indicate that the mutation in Thr75 contributes to improved enzyme activity on 2-chlorobenzaldehyde. To investigate mutations that improve the enantiomeric excess of (R) -2-chloromandelonitrile, a synthetic reaction from 2-chlorobenzaldehyde and KCN was performed using 1.0 U of all purified variants. The enantiomeric excess of (R) -2-chloromandelonitrile was 80% ee for the wild type, whereas N65Y, N65H and T75A were 92% ee, 85% ee and 86% ee, respectively. Showed an increased enantiomeric excess (Table 7). Thus, mutations in Asn65 and Thr75, in particular, mutation of Asn65 to tyrosine or histidine and mutation of Thr75 to alanine contribute to an increased enantiomeric excess in the production of (R) -2-chloromandelonitrile. Further, even when two types of mutations are combined, such as N65H / T75A and N65Y / T75A, the effect of improving the specific activity and the enantiomeric excess of (R) -2-chloromandelonitrile as compared with the wild type can be obtained. (Not shown).

The loop connecting β4-β5 of PlamHNL (residues 70-75) is longer than that of OgraHNL and is located at the entrance of the pocket. This loop region (I69G) of PlamHNL is considered to greatly affect the activity of PlamHNL. Furthermore, the 69th amino acid residue of β4 at the entrance of the pocket is isoleucine in PlamHNL and glycine in OgraHNL. These residues are thought to affect the affinity for the substrate. Therefore, when this mutant was prepared and examined, the Km value of the wild-type enzyme PlamlNL for (R) -2-chloromandelonitrile was 56.8 mM, whereas the Km value of the mutant enzyme PlamlHNL-I69G was ( The Km value for R) -2-chloromandelonitrile was as low as 41.8 mM, and the mutant enzyme had an increased affinity for the substrate (Table 8). The ee (enantiomeric excess) of (R) -2-chloromandelonitrile synthesized using I69G and N65Y / I69G was 83% and 91%, respectively, and that of wild-type enzyme PlamHNL was 80%. % (Table 7). These data suggest that mutations in the β4-β5 region also play an important role in improving the properties of the enzyme.

(Enantioselectivity of PlamHNL mutant)
The best candidate mutants N65Y, T75A and the two-point mutant N65Y / T75A were further investigated to find the optimal conditions for (R) -2-chloromandelonitrile synthesis. Increasing the amount of enzyme to 4U increased the maximum enantiomeric excess of (R) -2-chloromandelonitrile produced by N65Y to 96.33%. The maximum ee produced using the wild type under the same conditions was 87% (FIG. 8a). Further, by adjusting the pH to 3.5, the N65Y mutant had a 98.2% e. e. (FIG. 8b). The production of (R) -2-chloromandelonitrile by the purified enzyme N65Y is higher than that of the wild type (76% conversion, 90% ee) by incubation at pH 3.5, 25 ° C. for 30 minutes (76% conversion, 90% ee). 91%, 98.2% ee) (not shown). These results indicated that the enantioselectivity of PlamHNL in (R) -2-chloromandelonitrile synthesis was significantly improved by the Asn65 mutation.

The HNL mutant derived from the millipede of the present invention can be industrially applied in the synthesis of cyanohydrin.

Claims (15)

1. A variant of (R) -hydroxynitrile lyase derived from a millipede, which has one or more amino acid substitutions selected from the following (a) to (e), and further comprises (R) -hydroxynitrile lyase: (R) -hydroxynitrile lyase mutant proteins having activity:
   (A) substitution of A, which is the second amino acid in the amino acid sequence represented by TAX1DI (SEQ ID NO: 3) constituting the β-sheet structure (β3) of the (R) -hydroxynitrile lyase, with another amino acid; Wherein X1 is L or F;
   (B) substitution of another amino acid, X6, which is the fifth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, Where X2 is Q, H or R, X3 is I or V, X4 is M, I, T or D, X5 is A, T or I, X6 is Y or N , X7 is V, T or L, X8 is G or I, X9 is G or A, and X10 is P, A or S;
   (C) substitution of another amino acid F, which is the seventh amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, Wherein X2 to X10 are as defined in (b);
   (D) substitution of another amino acid, X8, which is the ninth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, Wherein X2 to X10 are as defined in (b);
   (E) substitution of another amino acid, X12, which is the second amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase, Wherein X11 is S, L, M or I, X12 is T or absent, X13 is H, I, Y or F and X14N or T; and (f) the (R)- Substitution of A, which is the third amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of hydroxynitrile lyase, wherein X11 to X14 are ( As defined in d).
2. 変 異 The mutant protein according to claim 1, wherein the (R) -hydroxynitrile lyase derived from the millipede has eight reverse-equilibrium β-sheet structures.
3. The amino acid sequences constituting the eight β-sheet structures are X15X16FX17X18VL (β1) (SEQ ID NO: 1), TX19RX20YVX21P (β2) (SEQ ID NO: 2), TAX1DI (β3) (SEQ ID NO: 3), and X2X3X4X5X6DFX7X8X9X10 (β4), respectively. (SEQ ID NO: 4), X11X12AX13LX14 (β5) (SEQ ID NO: 5), X22X23KX24X25WX26FQYX27X28 (β6) (SEQ ID NO: 6), X29X30YCAYX31CX32 (β7) (SEQ ID NO: 7), and X33IX34EYKCX35X36 (β8). Item 7. The mutant protein according to Item 2,
   Where X1 is L or F, X2 is Q, H or R, X3 is I or V, X4 is M, I, T or D, and X5 is A, T or I , X6 is Y or N, X7 is V, T or L, X8 is G or I, X9 is G or A, X10 is P, A or S, X11 is S, L , M or I, X12 is T or absent, X13 is H, I, Y or F, X14N or T, X15 is F or L, X16 is E, Q or L , X17 is E, A, S or T, X18 is Y or F, X19 is A or T, X20 is V or I, X21 is Q or R, X22 is G or D X23 is E, K, D or A, X24 is Q, T or A, X25 is V, I or T, X26 is Y H or N, X27 is T, V or I, X28 is N or D, X29 is A or S, X30 is N or S, X31 is R, T or S; X32 is N or D, X33 is E, A, Q, N or S, X34 is I, A or V, X35 is A or T, and X36 is S, N or T.
4. (4) The mutant protein according to (2) or (3), wherein the (R) -hydroxynitrile lyase derived from the millipede further has one α-helix structure.
5. 変 異 The mutant protein according to claim 4, wherein the amino acid sequence constituting the α-helix structure is VPNX37KIH (SEQ ID NO: 9), wherein X37 is D or Y.
6. 変 異 The mutant protein according to any one of claims 1 to 5, wherein the (R) -hydroxynitrile lyase derived from the millipede is a protein belonging to the lipokine superfamily.
7. The (R) -hydroxynitrile lyase derived from the millipede is ChuaHNL, NttHNL, NtmHNL, OgraHNL, PlamHNL, Pton1HNL, Pton2HNL, Pton3HNL, PfalHNL, PtokHNL, and any one of the following: The mutant protein according to any one of claims 1 to 6.
8. The (R) -hydroxynitrile lyase mutant according to any one of claims 1 to 7, wherein the amino acid substitution is one or more substitutions selected from the following (a) to (e). protein:
   (A) substitution of C or H for the second amino acid A in the amino acid sequence represented by TAX1DI (SEQ ID NO: 3) constituting the β-sheet structure (β3) of the (R) -hydroxynitrile lyase;
   (B) H, Y, M, V of X6 which is the fifth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β sheet structure (β4) of the (R) -hydroxynitrile lyase , L or W substitutions;
   (C) substitution of I, which is the seventh amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase;
   (D) substitution of X8, which is the ninth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase, with G;
   (E) substitution of A for X12, which is the second amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase; and f) The hydrophobic residue (I, L) of the third amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase , M, F, W, Y or V), C, T, E, Q or S.
9. The mutant protein according to any one of claims 1 to 8, wherein the amino acid substitution is present at two or more positions.
10. The mutant protein according to claim 9, wherein the two amino acid substitutions are two substitutions selected from the following (b), (d), and (e).
    (B) H, Y, M, V of X6 which is the fifth amino acid in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β sheet structure (β4) of the (R) -hydroxynitrile lyase , L or W substitutions;
    (D) substitution of G for the ninth amino acid X8 in the amino acid sequence represented by X2X3X4X5X6DFX7X8X9X10 (SEQ ID NO: 4) constituting the β-sheet structure (β4) of the (R) -hydroxynitrile lyase;
    (E) substitution of A for X12, which is the second amino acid in the amino acid sequence represented by X11X12AX13LX14 (SEQ ID NO: 5) constituting the β-sheet structure (β5) of the (R) -hydroxynitrile lyase;
    Here, X2 to X12 are as defined in claim 1.
11. 核酸 A nucleic acid molecule encoding the (R) -hydroxynitrile lyase mutant protein according to any one of claims 1 to 10.
12. A vector comprising the nucleic acid molecule according to claim 11.
13. A cell transformed with the vector according to claim 12.
14. 方法 The method for producing a (R) -hydroxynitrile lyase mutant protein according to any one of claims 1 to 10, comprising culturing the cell according to claim 13.
15. A method for producing cyanohydrin, comprising reacting a ketone or aldehyde with a cyanide compound (R) in the presence of the (R) -hydroxynitrile lyase mutant protein according to any one of claims 1 to 10. Method.

Images