Mutant enzyme

Abstract

The present invention relates to mutant enzymes, and more particularly to mutant enzymes with enhanced performance and methods of using such enzymes to oxidize organic compound substrates. The present invention provides a mutant CYP102A1 enzyme having enhanced monooxygenase activity and / or altered product selectivity and comprising a substitution at one or more amino acid residue positions 330, 191, 401, 403, 307, 377, 425, 276, 117, 131 or 215.

Description

[0001] The present application is a further divisional application of the divisional application with application number 201510867081.X and application date of October 8, 2008, and invention name of “Mutant Enzyme”, wherein the divisional application with application number 201510867081.X is a divisional application of Chinese patent application No. 200880119582.8 with application date of October 8, 2008 and title of “Mutant Enzyme”. Technical Field

[0002] The present invention relates to mutant enzymes with enhanced properties. Background Art

[0003] Bio-enzyme catalysts, such as P450 BM-3 Enzymes are finding increasing use in a variety of industrial applications ranging from the synthesis of fine chemicals, intermediates, pharmaceuticals and drug metabolites to the degradation of organic chemical pollutants and contaminants. Protein engineering, using directed evolution or site-directed mutagenesis, can be used to isolate variants of known enzymes, which can yield new insights and applications for their catalytic activities.

[0004] P450 of Bacillus megaterium (1) BM-3 Belong to the superfamily (or superfamily) of cytochrome P450 enzymes. There are more than 7,700 genes encoding P450 enzymes in various gene sequence databases. The nomenclature of P450 enzymes has also been systematized. The superfamily of enzymes is called CYP, followed by the numbering of the enzyme family (thus, CYP1, CYP51, CYP102, etc.), which is divided into subfamilies represented by letters (thus, CYP1A, CYP101B, etc.), and each subfamily member is identified by a number (thus, CYP1A1, CYP3A4, CYP101D3, etc.). The genes encoding CYP enzymes are represented by italics, such as the CYP101A1 gene. P450 BM-3 It has been identified as CYP102A1, that is, it is the first member of the CYP102 family. From now on, the systematic nomenclature of CYP102A1 will be used for P450 BM-3 .

[0005] CYP102A1 (1) is an enzyme of interest for biotransformation applications because it is catalytically self-sufficient. Unlike other P450 enzymes, where the P450 monooxygenase and electron transfer cofactor proteins are separate entities, CYP102A1 has a heme monooxygenase domain fused to a diflavin electron transfer reductase domain, which contains both FAD and FMN prosthetic groups in a single polypeptide. The natural substrates of CYP102A1 are believed to be straight or branched medium-chain fatty acids (1, 2). The crystal structure of the CYP102A1 heme domain became available in 1993 (3), revealing the active site structure and the presence of a substrate access channel. The crystal structure with bound substrate published four years later indicated changes in the side chain conformation of F87 upon substrate binding (4).

[0006] Protein engineering of CYP102A1 has been reviewed (5-7). Early studies focused on the active site residue F87, where mutations F87V, F87A, F87Y, and F87G have been shown to have different effects on activity and selectivity for fatty acid oxidation (8-11). Mutations on F87 have been found to favor oxidation of various substrates (7). Residues at the entrance of the substrate access pathway such as F42, R47, and Y51 have also been targeted. Although the F42A mutation reduced enzyme activity (10), neutralizing or reversing the charge at position 47 altered substrate specificity (8, 12), as did the hydrophobic substitution Y51A. WO0031273 discloses a mutant R47L / Y51F couplet to facilitate entry, binding, and oxidation of hydrophobic organic molecules such as polyaromatic hydrocarbons and terpenoids. This coupler has also been combined with F87A, I263A, A264G and M354A mutations to provide enhanced activity and / or product selectivity for substrate oxidation (13, 14). The R47L / Y51F combination, as well as the R47L and Y51F mutations on their own, is now commonly used in CYP102A1 engineering (15-19).

[0007] In addition to the rational selection of mutation sites, screening techniques have been used to identify other mutations and mutation sites that have the desired effect on activity and selectivity. Random or site saturation mutagenesis was applied to CYP102A1 as early as 1997 (20). NO20020380 discloses the use of indigo production via indole oxidation as a screening method for discovering CYP102A1 mutations with new activity. Saturation mutagenesis was applied to many residues that may affect substrate binding, and the mutant A74G / F87V / L188Q was reported to oxidize a wide range of organic molecules with enhanced activity and altered selectivity compared to the wild type (21-25). AT342351T discloses the generation of p-nitrophenol detected by spectroscopic detection via ω-p-nitrophenoxy-carboxylic acid oxidation as a screening procedure in a set of random mutagenesis experiments. Mutations V26T, R47F, S72G, A74G, F87A&V, L188A,G,N,K,Q,R,S&W, and M354T have all been published (26, 27).

[0008] The p-nitrophenol screening method was extended by using p-nitrophenoxyoctane as a surrogate substrate. WO2002083868, EP1470219 and US2005202419 (subsequently amended in WO2005017116, EP1660646 and US2005037411) disclose mutations L52I, I58V, F87A, H100R, S106R, F107L, A135S, M145A&V, A184V, N239H, S274T, L324I, V340M, I366V, K434E, E442K, V446I.

[0009] WO2003008563 and US2003100744 disclose multiple rounds of random mutagenesis, gene recombination and the like using the same method. The results of shuffling and screening were reported, and the mutations M30I, E64A, V78A, F87A, D, G, H, I, K, N, R, V&W, H138Y, F162S, H171Q, T175I, V178I, A184V, N186D, D217V, I220T, K224I, S226I, D232G, T235A, H236Q, E252G, R255S, I258T, I259V, T268Q, A290V, A295T, L353V, D370Q, E380G, G396M, T411A, and M416L were reported.

[0010] WO2005017105, US2005059128 and EP1639091 disclose the use of the same method and report mutations R47C, L75I&W, V78A,F&T, A82L,F,G,I,S&T, F87I,L&V, T88C, K94I, P142S, T175I, A184V, F205C, S226R, H236Q, E252G, R255S, T260,L,N&S, A290V, A328V&M, L353V.

[0011] Subsequently, WO2006105082 disclosed mutations R47C, V78F, A82S, K94I, P141S, T175I, A184V, F205C, S226R, H236Q, E252G, R255S, A290V, A291V, A328F, and L353V.

[0012] These series of mutants generated by random mutagenesis showed enhanced activity for alkane oxidation from ethane to mid-chain alkanes (28-30). There were also changes in selectivity, especially when the directed evolution variants were combined with mutations introduced into the active site by site-directed mutagenesis, for example, in octane oxidation, where the mutations shifted the site of oxidation to the terminal carbon (31), in the selective epoxidation of terminal olefins (32), and in the enantioselectivity in the oxidation of cyclopentanecarboxylic acid derivatives (33). Notably, better results were often obtained by combining directed evolution with rational redesign.

[0013] CYP102A3 is a P450 enzyme belonging to the same sub-family as CYP102A1. Random mutagenesis of CYP102A3, followed by alkane oxidation and monitoring of NADH formation in the presence of an alcohol dehydrogenase specific for terminal alcohols, generated a mutation that produced 50% 1-octanol from octane oxidation. This is the highest rate of terminal CH bond oxidation of straight-chain alkanes observed to date by a designed CYP102 family P450 enzyme (34).

[0014] There continues to be a need to isolate additional mutations of industrially useful enzymes, such as the CYP102A1 enzyme, to further understand the effects of structural changes on their catalytic mechanisms, improve their catalytic turnover, and expand their substrate and / or product ranges. Typically, P450 enzymes, such as CYP102A1, are engineered to enhance enzyme activity, with control of product selectivity and substrate specificity being secondary goals. Mutations and mutation sites that can couple selectivity control to enhanced monooxygenase activity are significantly lacking, such that enzymatic conversion of compounds may be rapid but not sufficiently selective, or have some selectivity but react very slowly, or the desired product is not formed. There is also a need for screening methods that can provide mutants with enhanced activity and / or desired selectivity. Summary of the invention

[0015] It has now been found that substitution mutations at specific positions in CYP102A1 have the desired effect in enhancing monooxygenase activity and also provide altered selectivity according to the present invention. These mutation sites were identified by using a novel screening method that provides enhanced activity and enhanced / altered selectivity.

[0016] The present invention provides a mutant CYP102A1 enzyme having enhanced monooxygenase activity and / or altered selectivity and comprising a substitution at one or more of positions 117, 131, 191, 215, 276, 307, 330, 377, 401, 403, 425 of CYP102A1. Also provided is a method for oxidizing a substrate as an organic compound, the method comprising oxidizing the organic compound using the mutant CYP102A1 enzyme of the present invention.

[0017] The claimed substitutions form part of the same inventive concept, as they share the effect of enhancing monooxygenase activity and / or altering the selectivity of CYP102A1. Substitutions at positions 330, 401 and 403 also exert their effects via a common structural and functional mechanism as outlined below.

[0018] SEQ ID NO: 1 is the sequence of CYP102A1. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 :Related residues in CYP102A1(P450 BM-3 Palm indicates the fatty acid substrate of the enzyme.

[0020] Figure 2: Comparison of key residues in the substrate entry channel and active site of wild-type CYP102A1 and the A330P mutant highlights the structural perturbations at P329 and A328 resulting from the A330P mutation. DETAILED DESCRIPTION

[0021] The present invention can be applied to natural and artificial homologues (or homologues) of CYP102A1, for example, comprising sequences having at least 40% amino acid sequence identity with CYP102A1. Such homologues typically comprise an amino acid sequence corresponding to (i.e., homologous to or identical to) the heme monooxygenase domain of CYP102A1 (represented by amino acid positions 1 to 480).

[0022] The enzyme of the present invention comprises (or consists of) a sequence having at least 40% identity to SEQ ID NO: 1 (sequence of CYP102A1). In a preferred embodiment, the sequence may have at least 55%, 65%, 80% or 90% homology thereto within at least 20, preferably at least 30, such as at least 40, 60, 100, 200, 300, 400 or more adjacent amino acids, or even within the entire sequence of the homolog, and more preferably at least 95%, 97% or 99%. In one embodiment, the enzyme of the present invention has any specified percentage homology when compared to amino acid residues 1 to 480 of CYP102A1. The adjacent amino acids may include the active site. The homology may alternatively be determined not within the adjacent amino acids but only within the amino acids of the active site. Therefore, homology based on amino acid identity is typically at least 40% homologous to CYP102A1. The enzyme of the invention may have a percentage identity with the CYP102A1 sequence that is the same as any of the specific percentage homology values ​​mentioned above over any sequence length (i.e., it may have at least 40%, 55%, 80% or 90% and more preferably at least 95%, 97% or 99% identity).

[0023] The homologous sequence may represent a mutated portion of the CYP102A1 sequence and / or may be present in the form of a full-length fusion polypeptide of the enzyme of the invention.

[0024] Any homologous protein mentioned herein (i.e., described as being homologous to another protein) is typically at least 40% homologous to the related protein. Homology can be determined using known methods. For example, the UWGCG software package provides the BESTFIT program (e.g., used with its default settings) that can be used to calculate homology (Devereux et al (1984) Nucleic Acids Research 12, 387-395).

[0025] The PILEUP and BLAST algorithms can be used to calculate homology or align sequences (typically according to their default settings), for example as described in Altschul SF (1993) J MoI Evol 36:290-300; Altschul, S, F et al (1990) J MoI Biol 215:403-10.

[0026] Software for performing BLAST analysis is publicly available through the National Center for Biotechnology Information (http: / / www.ncbi.nhn.nih.gov / ). The algorithm first involves identifying high-scoring sequence pairs (HSPs) by identifying short words of length W in a query sequence that match or satisfy some positive threshold score T when aligned with a word of the same length in a database sequence. T is called the neighborhood word score threshold (Altschul et al, supra). These initial neighborhood word hits serve as seeds for initial searches to find HSPs containing them. Word hits are extended in both directions along each sequence until the cumulative alignment score can be increased. When the cumulative alignment score drops by an amount X from its maximum reach value; the cumulative score becomes zero or lower due to the accumulation of one or more negatively scored residue alignments; or the end of either sequence is reached, the extension of the word hits in each direction stops. The BLAST algorithm parameters W, T, and X determine the sensitivity and speed of the alignment. The BLAST program uses a wordlength (W) of 11, a BLOSUM62 scoring matrix (see Henikoff and Henikoff (1992) Proc. Natl. Acad. Sci. USA 89:10915-10919) alignment (B) of 50, an expectation (E) of 10, M=5, N=4 and a double-stranded control as defaults.

[0027] The BLAST algorithm performs a statistical analysis of the similarity between two sequences; see, e.g., Karlin and Altschul (1993) Proc. Natl. Acad. Sci. USA 90: 5873-5787. One measure of similarity provided by the BLAST algorithm is the smallest sum probability (P(N)), which provides an indication of the probability by which a match between two nucleotide or amino acid sequences would occur by chance. For example, a sequence is considered similar to another sequence if the smallest sum probability in a comparison of a first sequence to a second sequence is less than about 1, preferably less than about 0.1, more preferably less than about 0.01, and most preferably less than about 0.001.

[0028] Typically, a homologous protein differs from the related protein by at least, or by fewer than, 2, 5, 10, 20, 40, 50 or 60 mutations (each of which can be a substitution, insertion or deletion) when comparing all proteins or any of the above mentioned contiguous amino acids in length.

[0029] The enzymatic activity of the CYP102A1 enzyme of the present invention is typically determined in vitro using any of the substrates or conditions mentioned herein and is given as the rate of NADPH oxidation, the rate of product formation and the coupling efficiency. The rate is the conversion frequency and is expressed as (nmol NADPH)(nmol CYP102A1) -1 (min) -1 or (nmol product)(nmolCYP1 02A1) -1 (min) -1 The binding efficiency is the percentage of NADPH consumed for product generation, i.e., the percentage of the theoretical maximum efficiency. The CYP102A1 enzyme of the invention (e.g., when used in the methods of the invention) can typically have a binding efficiency of at least 1%, such as at least 2%, 4%, 6%, 10%, 20%, 40%, 80% or more. The CYP102A1 enzyme (e.g., when used in the methods of the invention) typically has a binding efficiency of at least 2 min. -1 , such as at least 4, 10, 15, 20, 25, 50, 100, 200, 300, 500, 700, 1000, 2000 min -1 or greater. In the case where more than one product is formed (which is a common situation), the product generation rate represents the total amount of all oxidation products formed. In some embodiments, the product generation rate of a specific oxidation product is measured, that is, not all oxidation products may be measured.

[0030] The mutant CYP102A1 enzymes of the present invention exhibit enhanced monooxygenase activity and / or altered selectivity relative to the corresponding wild-type CYP102A1 enzyme. The enhanced monooxygenase activity can be characterized by using one or more substrates for oxidation according to an increased binding efficiency or an increased product generation rate. The increased binding efficiency or increased product generation rate may or may not be shared among all substrates utilized by the mutant CYP102A1 enzyme. The mutant CYP102A1 enzyme typically exhibits a binding efficiency that is at least 10%, 20%, 50%, 100%, 500%, 1000% or 1500% greater than the binding efficiency of the wild-type enzyme. The mutant CYP102A1 enzyme may also have a product generation rate that is at least 50%, 100%, 150%, 500%, 1000%, 2000%, 5000%, 10000% greater than the product generation rate of the wild-type enzyme.

[0031] It should be understood that the mutant CYP102A1 enzymes of the present invention may also exhibit other altered properties relative to the corresponding wild-type enzyme and mutants disclosed in the literature, such that these effects may include, but are not limited to, enhanced monooxygenase activity. For example, the mutant enzymes may exhibit altered substrate specificity, allowing for preferential utilization of specific substrates, or may exhibit monooxygenase activity where the wild-type enzyme or known mutants are unable to oxidize substrate organic compounds.

[0032] Mutant enzymes of the present invention may also show altered product selectivity, wherein the product formed in a smaller proportion by the wild type becomes the dominant product of the mutant, or a new product formed in a smaller proportion or not at all by the wild type becomes the majority or dominant product. Other altered properties of the mutant enzyme and the oxidation process implemented by the mutant enzyme are described below.

[0033] The mutant CYP102A1 enzymes include substitutions at one or more of positions 117, 131, 191, 215, 276, 307, 330, 377, 401, 403 and 425 of CYP102A1. Typically, they may include substitutions at 2 or more, 3 or more, 4 or more, 5 or more, 6 or more of the above defined positions. In a preferred embodiment, where there is a substitution at position 330, there are less than 5 other substitutions, such as less than 3, or in one embodiment, no other positions are substituted.

[0034] In the case of describing specific mutations of CYP102A1, the letter of the amino acid residue present in the native form of CYP102A1 precedes the position, and the position is followed by the amino acid in the mutant. These positions can be related to the numbering shown in SEQ ID NO:1. In order to indicate multiple mutations in the same protein, each mutation is listed separated by a slash. Moreover, particularly preferred mutants can be described using the internal nomenclature outlined below.

[0035] Although mutations are defined with reference to positions in CYP102A1, equivalent substitution mutations at homologous or corresponding positions in the polypeptide chain of a CYP102A1 homologue that shares at least 40% amino acid identity with SEQ ID NO: 1 are also encompassed by the present invention. Equivalent positions are determined by reference to the amino acid sequence of SEQ ID NO: 1. Homologous or corresponding positions can be readily deduced by comparing the sequence of CYP102A1 (SEQ ID NO: 1) and the sequence of the homologue (or homologue) based on the homology between the sequences. PILEUP and BLAST algorithms can be used to align these sequences. Where the homologous or corresponding amino acid referred to is an active site residue, it will generally be in a similar position in the active site as the homologue of any particular amino acid discussed herein.

[0036] It is well known to those skilled in the art that the P450 enzyme superfamily, despite having a highly conserved tertiary structure, is uncommon in proteins and enzymes with a primary structure of low homology (35-37). There are now >6500 CYP genes in the genomic database, and >150 structures in the Protein Data Bank. All P450 structures determined to date exhibit a characteristic morphology of a helical-rich domain packaged against the main β-strand domain, as described in the crystal structure of the P450 enzyme (CYP101A1) first reported by Poulos and coworkers (38). The helices are named AL, while the β-strands β1-β5, with the entire morphology now referred to as the "P450 fold" (38-42). Among the secondary structural elements, the B and B' helices, the BC loop, the F and G helices, and the FG loop on the distal side of the heme form a substrate binding pocket. Sequence alignments readily identify residues within these helices and loops, but there is a high degree of variability within these general frameworks based on amino acid sequence and structural alignment, and this gives rise to the diverse specificities, activities, and selectivity patterns of P450 catalysis.

[0037] P450 enzymes in different families have as little as 20% homology (amino acid identity) (35-37). An example of an alignment between CYP102A1 and structurally characterized P450 enzymes is shown in Table A. Until recently, sequential sequence analysis has suggested that as few as three residues in a typical 400-460 in a P450 enzyme or domain are absolutely conserved: a proximal cysteine ​​ligand for the heme iron, and an EXXR motif in the K helix that may play a role in heme ligation and binding (43). However, results on the CYP157 family published in 2006 indicate that even the EXXR motif is not conserved, leaving the proximal cysteine ​​as the only conserved residue in the entire P450 superfamily. In the systematic classification of the P450 superfamily (44), enzymes that share exactly 40% amino acid identity are therefore placed in the same family, while closely related family members (>55% identity) are grouped into subfamilies (see, e.g., Table B).

[0038] In fact, detailed molecular structures, substrate specificities, and product selectivities are conserved within a family, whereas sequence identity is often low. The most striking example is the CYP51 family of sterol 14α-demethylases found across kingdoms of life. These play a key role in the oxidative demethylation of the C14 methyl group of intermediates formed following cyclization of squalene oxide. Sequence alignments have shown that homology between known CYP51 family genes across kingdoms of life averages 30%, rising to 95% in closely related species such as mammals and falling to 23% among lower organisms (45). There is increasing recognition that the 40% cutoff for assigning enzymes to the same family may be too high in some cases, and that the enzyme activity and higher homology often observed for active site residues may need to be given greater consideration in the future.

[0039] Thus, typically based on the "P450 fold" homologs that are at least 40% homologous to CYP102A1 based on amino acid identity can also be readily identified, while alignment of homolog sequences with equivalent mutations introduced at corresponding or homologous positions can be aided by knowledge of the conserved nature of the alpha helical and beta strand arrangement of the P450 fold that is shared across the enzyme family.

[0040] It should be understood that CYP102A1 is a fusion of an electron transfer reductase domain and a heme monooxygenase domain. These domains can be excised by proteolysis or by truncation of the full-length gene. The active site (substrate binding pocket) is located in the heme domain. Some members of the CYP102 family are not fusion proteins, but have a sequence homology of 40% with the CYP102A1 heme domain. Therefore, sequence homology can be determined solely within these environments over the entire heme domain. Equivalent residues in these enzymes that are equivalent to CYP102A1 disclosed in the present invention can be identified by sequence homology and structural analysis known to those skilled in the art.

[0041] The amino acids in the active site are those that arrange or define the sites to which the substrate is bound during catalysis or those that arrange or define the sites through which the substrate must pass before reaching the catalytic site. Thus, such amino acids typically interact with the substrate during entry into the catalytic site or during catalysis. Such interactions typically occur through electrostatic interactions (between charged or polar groups), hydrophobic interactions, hydrogen bonds, or van der Waals forces. The active site amino acids can be identified by sequence alignment and reference to the known crystal structure of the heme domain of wild-type CYP102A1, or the crystal structure of a homologue.

[0042] In cases where the mutated residue is not an active site residue, a computerized or manual alignment of the homolog and CYP102A1 sequences is performed to infer the homologous or corresponding position, which can be aided by knowledge of the residues flanking the mutated position in CYP102A1. Thus, for example, the 10 N- and C-terminal flanking residues flanking the residues to the following positions in CYP102A1 are:

[0043] FSQQAMKGYH(A117)MMVDIAVQLV;

[0044] DIAVQLVQKW(E131)RLNADEHIEV;

[0045] LDEAMNKLQR(A191)NPDDPAYDEN;

[0046] FQEDIKVMND(L215)VDKIIADRKA;

[0047] HETTSGLLSF(A276)LYFLVKNPHV;

[0048] VLVDPAPSYK(Q307)VKQLKTVGMV;

[0049] EALRLWPTAP(A330)FSLYAKEDTV;

[0050] GDDVEEFRP(E377)RFENPSAIPQ;

[0051] KPFGNGQRAC(I401)GQQFALHEAT;

[0052] FGNGQRACIG(Q403)QFALHEATLV;

[0053] GMMLKHFDFE(D425)HTNYELDIKE.

[0054] Conservation of 2, 3 or more N- and / or C-terminal flanking residues may allow inference of homologous or corresponding positions of introduced mutations.

[0055] Similar analyses can be performed for any other position in CYP102A1 referred to in this description to identify homologous or corresponding sites in naturally occurring CYP102A1 homologs.

[0056] Functional fragments of the CYP102A1 enzyme are also encompassed within the present invention. These fragments may therefore contain only those amino acids required for the oxidative activity. Thus, with reference to the polypeptide sequence of CYP102A1, up to 20 residues of the reductase domain and / or monooxygenase domain can be deleted at the N-terminal or C-terminal portion without significantly affecting the folding of the active site or the intrinsic substrate oxidation capacity of the monooxygenase domain. In homologues of CYP102A1, similar truncations are possible, and the extent of possible truncations may be determined by the methods described herein for monitoring oxidative activity. Truncated forms of the enzyme may have favorable properties in terms of stability, expression level, and protein activity.

[0057] The nature of the amino acid substituted at the positions of CYP102A1 described herein (or equivalent positions as defined above) is determined primarily by the mutation conditions that exhibit enhanced monooxygenase activity. Thus, the introduced amino acid will typically enhance monooxygenase activity. In any case where a specific substitution mutation in CYP102A1 is mentioned, it is to be understood that any substitution of another amino acid residue at the same position which has an effect greater than or similar to that of the specific substitution mutation on the oxidation activity of the CYP102A1 enzyme is contemplated according to the present invention. Similarly, where a specific substitution mutation also has an effect on another parameter of the CYP102A1 enzyme, such as substrate specificity, or the range or ratio of oxidation products obtained in a given substrate oxidation, it is to be understood that substitution of those other amino acid residues which also cause greater or similar effects are also contemplated for use according to the present invention.

[0058] In some embodiments, substitutions introduce conservative changes, which is the replacement of an amino acid with another amino acid of similar chemical structure, similar chemical properties, or similar side chain volume. The introduced amino acid can have a similar polarity, hydrophilicity, or hydrophobicity to the amino acid it replaces. Conservative amino acid changes are well known in the art and can be selected based on the changes defined in Table C. In the case of amino acids with similar polarity, this can also be determined by reference to the hydropathy scale of the amino acid side chains (Table D).

[0059] Conservative amino acid changes can also be determined with reference to the scoring matrix Point Accepted Mutation (PAM) or BLOcks Substitution Matrix (BLOSUM) family for amino acid sequence conservation. Therefore, conservative amino acid changes can be members of an equivalence group, which is a group of amino acids with a positive score for mutation in the similarity description of the scoring matrix selected for reference and the comparison of the mutant polypeptide chain.

[0060] It should be understood that the definitions of physical properties provided in Table C are not considered to be limiting of the present invention, and non-polar amino acids include amino acids with aliphatic side chains and amino acids with aromatic side chains. The amino acid proline is classified as a non-polar amino acid, but it also has fixed properties and can change in the secondary structure. For example, proline is often found at the end of a helix. Moreover, depending on the specific circumstances of a given amino acid residue side chain, for example, the amino acid tyrosine, which is generally classified as non-polar due to its aromatic ring, can have similar functional effects of polar amino acid residues (such as threonine via its hydroxyl group). Therefore, for the purposes of the present invention, tyrosine can be considered to be a non-polar and polar amino acid. Moreover, amino acids that can be described as polar or hydrophilic can be uncharged or charged, and can also be alkaline or acidic. The amino acid histidine is well known to have a pKa value close to 7, so that depending on the protein environment at neutral pH, it may or may not be protonated on its side chain, and therefore may or may not carry a charge. Therefore, for the purposes of the present invention, histidine can be considered to be a polar charged or polar uncharged amino acid residue.

[0061] Specific examples of conservative amino acid changes at positions 117, 131, 215, 307, 330, 401 and 403 (or their equivalents) in CYP102A1 include, but are not limited to:

[0062] A117V, A117I, A117L, A117P, A117M, A117F, A117W, A117Y;

[0063] E131D;

[0064] L215I, L215V, L215P, L215F, L215W, L215Y;

[0065] Q307H, Q307N, Q307S, Q307T, Q307Y;

[0066] A330P, A330I, A330L, A330M, A330V, A330F, A330W, A330Y;

[0067] I401P, I401I, I401L, I401M, I401V, I401F, I401W, I401Y;

[0068] Q403N, Q403H, Q403A, Q403T, Q403Y.

[0069] In other preferred embodiments, the amino acid substitution introduces a polar amino acid at a given position of the wild-type enzyme, typically where the existing residue is a non-polar residue, thereby changing the polarity. Specific examples of polar amino acid substitutions at positions 191 and 276 (or their equivalents) in CYP102A1 include, but are not limited to:

[0070] A191T, A191S, A191C, A191Y, A191H, A191K; A191R, A191N, A191Q;

[0071] A276T, A276S, A276C, A276Y, A276H, A276K, A276R, A276N, A276Q.

[0072] In contrast, in other preferred embodiments, the amino acid substitution introduces a non-polar amino acid at a given position of the wild-type enzyme, typically where the existing residue is a polar residue. For example, a non-polar amino acid can be introduced at position 377 or 403 or its equivalent. Specific examples include, but are not limited to:

[0073] E377A, E377V, E377L, E377I, E377P, E377F, E377Y, E377W;

[0074] Q403P, Q403W, Q403F, Q403Y;

[0075] In other embodiments, the amino acid substitution results in the loss of a charged side chain group at a given position of the wild-type enzyme. Thus, this substitution introduces an uncharged amino acid at the relevant position. This may or may not result in a loss of polarity at the position, resulting in the introduction of a polar uncharged or non-polar (aromatic or aliphatic) residue. For example, a non-polar or polar uncharged residue may be introduced at position 425 or its equivalent. Specific examples include, but are not limited to:

[0076] D425N, D425Q, D425H, D425S, D425T, D425A, D425L, D425V, D425I, D425P, D425W, D425Y, D425F.

[0077] In another embodiment, an amino acid with increased side chain volume is introduced at a position of the present invention. In a preferred embodiment, an amino acid with increased side chain volume, typically a bulky non-polar amino acid, is introduced at position 330, 401, 403 or its equivalent. Particularly preferred substitutions for position 330 are A330P, A330V, A330L, A330I, A330W, A330F, A330Y. Particularly preferred substitutions for position 403 are Q403P, Q403W, Q403F. In other embodiments, for example at position 377, it may be preferred that the amino acid to be introduced has a reduced side chain volume, such as E377A or E377G.

[0078] Mutation discussed herein is generally introduced into the enzyme by adopting known methods in the art, such as site-directed mutagenesis, PCR and gene recombination (gene shuffling) method of enzyme or by using multiple mutagenic oligonucleotides in the cycle of site-directed mutagenesis. Therefore, mutation can be introduced in a directed or random manner. Mutagenesis method therefore produces one or more polynucleotides encoding one or more different mutants. Typically, the mutant gene library generated can be used to produce a mutant enzyme library.

[0079] In addition to the one or more mutations specified above, such as substitutions, insertions or deletions, the enzyme may also have 1, 2, 3, 4, 5 to 10, 10 to 20, 20 to 40 or more other mutations. These additional mutations may or may not enhance the monooxygenase activity of the mutant CYP102A1 enzyme. Other mutations may be in the active site or outside the active site. For example, the mutation may be in the second sphere, i.e., a residue that affects or contacts the position or orientation of one or more amino acids in the active site. Insertions will typically be at the N and / or C terminus. Therefore, the enzyme may contain a short peptide of up to 20 amino acids or a full-length protein fused to one or both ends, for example, for assisting affinity chromatography or protein purification fixed on a solid matrix. Deletions typically include deletions of amino acids that are not involved in catalysis, such as those outside the active site (therefore, the enzyme is a mutant fragment of a naturally occurring enzyme).

[0080] Other mutations in the active site typically change the position and / or conformation of the substrate when it binds in the active site. The mutation may make the site more accessible to the heme group on the substrate to be oxidized. Thus, the mutation may be a substitution of an amino acid with a smaller or larger, or more or less polar side chain.

[0081] Additional mutations may include changes in amino acid residues that improve the stability of the enzyme. These mutations typically prevent oligomerization of the protein, such as P450 cam Dimerization of (CYP101A1) is eliminated by substitution of Cys344, preferably to alanine. Crystal structures of full-length CYP102A1 have not yet been obtained, only those of the isolated heme and FAD / FMN domains. Similar substitutions are not necessary for CYP102A1, since Cys334 of CYP101A1 is aligned with Asp370 in CYP102A1. However, crystal structures of full-length CYP102A1 and / or the reductase domain may reveal cysteine ​​residues that are removed by substitution with alanine to improve protein stability. Other mutations may also inhibit oligomerization resulting from contacts between hydrophobic patches on the surface of the protein. Still other mutations include insertions / deletions that aid enzyme purification and / or fixation, and that allow the protein to be prepared in soluble form, for example by introducing deletions or polyhistidine tags, or by mutations of the N-terminal membrane anchor sequence.

[0082] Preferably, the additional mutation is selected from one or more of the following mutations in CYP102A1: R47L, Y51F, A74G, A264G, N239H, I259V, L353I, F87A, F87L, F87G, H171L, L188Q, N319Y, I263A, A328P, or the same mutations at equivalent positions thereof. It should also be understood that the same considerations listed above with reference to the selection of other amino acid changes that have more or similar effects to those specifically listed apply to positions 117, 131, 191, 215, 276, 307, 330, 377, 401, 403 and 425. Thus, for example, depending on whether the specific additional mutations listed above are conservative changes, whether to change polarity or introduce uncharged amino acids, a similar range of amino acids will be suitable for introduction at each additional position.

[0083] In a particularly preferred embodiment, the mutant CYP102A1 enzyme of the present invention comprises one or more mutation groups selected from the following mutation groups of CYP102A1:

[0084] i) A330P

[0085] ii)A191T / N239H / I259V / A276T / L353I;

[0086] iii) F87A / H171L / Q307H / N319Y;

[0087] iv)F87A / A330P / E377A / D425N;

[0088] v)F87A / A117V / E131D / L215I;

[0089] vi) I401P;

[0090] vii) R47L / Y51F / I401P;

[0091] viii) F87A / I401P;

[0092] ix) R47L / Y51F / F87A / I401P;

[0093] x)R47L / Y51F / A330P / I401P;

[0094] xi) Q403P;

[0095] xii) R47L / Y51F / Q403P;

[0096] xiii)R47L / Y51F / F87A / Q403P.

[0097] or including its equivalent mutation group.

[0098] The A330P mutation as defined in i) is an unusual mutation in several respects, since, unlike other directed evolution variants (whose effects may depend on a mix of altered residues acting together), its activity is derived from a single point mutation. In CYP102A1, A330 is located close to the naturally occurring proline residue at position 329, and thus A330P juxtaposes two prolines at a major substrate contact point within the β-sheet (3). The crystal structure of the mutant ( Figure 2 ) suggest that this compresses the entry channel and makes the active site less accessible, resulting in a more closed active site and altered substrate binding conformation. As will be seen below, this has a characteristic effect on monooxygenase activity and product selectivity.

[0099] Mutation groups ii) and iii) enhance the activity of CYP102A1 while broadly reflecting the specificity characteristics exhibited by the wild-type enzyme.Except for residue 87 in groups iii) through v), none of the positions mutated in any mutation group is an active site residue.

[0100] For example, in group ii), position 353 is close to the substrate entry channel residue 354, while residues 191, 239, 259, 276 and 353 are all located on the protein surface. According to the crystal structure (4), A191 is significantly substituted for palmitate binding and is located on the outer edge (or outer lip) of the entry channel. It can be inferred that mutation of this residue may have an effect on substrate attraction and / or capture.

[0101] In group iii), position 171 is located on or close to the protein surface, while residues 307 and 319 are close to a region thought to be a docking site for the electron transfer reductase domain and could therefore potentially mediate their effect on the enhancement of monooxygenase activity through effects on electron transfer kinetics.

[0102] Mutation group iv) includes the A330P mutation at the substrate contact point, and mutations at positions 377 and 425 located peripherally in the enzyme structure close to the protein surface, such as positions 117, 131 and 215 in mutation group v).

[0103] In a further preferred embodiment, the mutant CYP102A1 enzyme of the invention of the type defined above is:

[0104] i) additionally comprising one or more of the following mutations in CYP102A1: R47L, Y51F, A74G, A264G or equivalent mutations thereof;

[0105] ii) additionally comprising one or more of the following mutations in CYP102A1: R47L, Y51F, F87A, F87L; and

[0106] iii) additionally comprising one or more of the following mutations in CYP102A1: F87A, F87G, I259V, I263A.

[0107] It should be understood that in addition to the specific mutation or specific additional mutations specified above, up to 1, 2, 3, 4, 5 to 10, 10 to 20 or more other mutations may also be included in these preferred embodiments of the mutant CYP102A1 enzymes of the present invention.

[0108] The substrate for the oxidation method is any organic compound, more typically any organic compound capable of being oxidized by a monooxygenase. The suitability of any organic compound for oxidation by a monooxygenase can be routinely determined by the methods described herein.

[0109] Oxidation processes result in the formation of C-O bonds in the compound, generally from the oxidation of carbon-hydrogen bonds to form alcohols, but from the oxidation of C=C bonds epoxides can be formed. Oxidation can thus introduce alcohols, aldehydes, ketones or epoxy groups. Alternatively, oxidation can result in further oxidation of oxygen-containing groups, such as conversion of alcohol groups to aldehydes or ketones. 1, 2 or more carbon atoms can be attacked in the same substrate molecule. Oxidation can also result in N- and O-dealkylation of substrate molecules.

[0110] Oxidation typically produces 1, 2 or more oxidation products. These different products may originate from different carbon atoms attacked and / or different degrees of oxidation occurring at a given carbon atom.

[0111] Oxidation can occur at a ring carbon atom or at a substituted carbon atom or at both. At least the initial oxidation will involve an attack on a C-H bond, which may be activated or unactivated, or an attack on a carbon-carbon double bond (typically giving an epoxide). Typically, an activated C-H bond is one in which the carbon atom is in a benzylic or allylic position. Aromatic rings and olefin double bonds activate C-H bond attacks by stabilizing free radical intermediates or any accumulation of charge generated during the reaction pathway. The carbon in the C-H bond can be primary, secondary or tertiary. Oxidation can occur to result in dehydrogenation, resulting in the formation of a C=C double bond rather than the insertion of an oxygen atom. This is most likely to occur when the alkyl substitution is branched or dehydrogenated to produce a C=C bond that is bound to an aromatic system or when dehydrogenation results in the formation of an aromatic system.

[0112] The substrate may be a natural substrate of the wild-type CYP102A1 enzyme or a substrate that is not a normal substrate of the wild-type enzyme but is available in the mutant enzyme. Examples of natural substrates of the CYP102A1 enzyme are branched and straight chain fatty acids that are hydroxylated at the sub-terminal position (ω-1 to ω-3) by the wild-type CYP102A1. Preferred examples are lauric acid, undecanoic acid, capric acid, nonanoic acid and caprylic acid.

[0113] In a preferred embodiment, the substrate is a short chain alkane or a medium chain alkane or an alkylbenzene. The term alkane refers to a hydrocarbon having the general formula C n H 2n+2 aliphatic branched or unbranched hydrocarbons.

[0114] The short chain alkanes typically have 1 to about 9 carbon atoms, more preferably 1 to 8, 1 to 6 or 1 to 4 carbon atoms. The C1-C8 alkyl group or moiety may be straight or branched. In the case where it is a C1-C4 alkyl moiety, for example, it may be methyl, ethyl, n-propyl, isopropyl, sec-butyl and tert-butyl.

[0115] Alkylbenzenes have one or more alkyl groups or moieties substituted at positions on the benzyl aromatic ring. The number of carbon atoms in the alkyl group or moiety can typically range from 1 to about 8 carbon atoms, more preferably 1 to 8, 1 to 6 or 1 to 4 carbon atoms.

[0116] In some embodiments, there are 1,2,3 or more substitutions on the short chain or medium chain alkane or directly on the benzyl ring substituted skeleton, or on the alkyl substitution of alkylbenzene. Any combination of the following substituents can be present. The substituent is typically a halogen atom or an alkyl or alkenyl group, which generally has 1 to 6 carbon atoms, and the substituent is optionally substituted by one or more halogens. The substituent may also include 1,2 or more oxygen atoms, halogens or nitrogen atoms, and may be, for example, an alcohol, an aldehyde, a ketone, an ether, an amine or an epoxy group.

[0117] Examples of preferred short chain alkane substrates include, but are not limited to, pentane, 3-methylpentane, 2-methylbutane, butane, propane, ethane and methane, octane and nonane. Examples of preferred alkylbenzene substrates include, but are not limited to, propylbenzene, ethylbenzene, butylbenzene, cumene, tert-butylbenzene, o-xylene, m-xylene, p-cymene and ethylanisole. Other preferred aromatic compounds are naphthalene and fluorene.

[0118] It should be noted that organic compounds such as butane, naphthalene, and especially propane, tert-butylbenzene and o-xylene are broadly classified as "non-natural" substrates for the wild-type CYP102A1 enzyme, but are capable of being oxidized by the mutant CYP102A1 enzymes of the present invention. Non-natural substrates can be defined as molecules that have no detectable on-rate and / or product formation when incubated with the wild-type CYP102A1 enzyme. Non-natural substrates can also include molecules whose oxidation rate is <10% of the rate of the wild-type CYP102A1 enzyme's natural substrates, such that they cannot be considered true substrates.

[0119] In other embodiments of the present invention, the substrate is a terpene, such as a monoterpene or a sesquiterpene. The substrate may also be a cyclic olefin. Although the terpene used in the present invention generally has the formula (C5H8) n , wherein n is 2 or more, especially 2 or 3, but it will be appreciated that the term "terpenoid" may extend to compounds strictly termed "terpenoids", involving the loss or migration of a fragment, generally a methyl group. Thus, for example, a sesquiterpene (where n is 3) which may be used in the present invention may contain only, for example, 14, instead of 15 carbon atoms. In general, terpenes are terpenes which may be built up from isoprene units. Terpenes may be cyclic or acyclic. It will also be appreciated that "terpenoid" may also extend to compounds related to terpenes and may contain one or more oxygen atoms, for example in the form of an alcohol or keto group, such as damascones and ionones, especially β-ionone.

[0120] Monoterpenes (where n is 2) will generally have 10 carbon atoms, typically have 1 to 3 double bonds, especially 1 or 2 ring double bonds, and typically have 0 to 2 rings. For one of the rings, a bridge may be formed containing typically 0 or 1 carbon atom. In other words, it may be formed by direct connection between 2 carbon atoms of an existing ring or with an intermediate methylene group. If the terpene is acyclic, it will generally contain at least 2 double bonds, and generally 3 double bonds.

[0121] Sesquiterpenes will generally contain 14 or 15 carbon atoms, typically have 0 to 2 double bonds and generally have 1 to 3 rings, and may have fused and / or bridged rings.

[0122] The rings which may be present in such terpenes will typically have from 3 to 9 carbon atoms, more particularly 5 or 6. Thus, in particular, the terpene will contain a cyclohexane, or cyclohexadiene ring.

[0123] The terpene typically contains a total of 3 or 4 exocyclic methyl or methylene groups, for example, 2 methyl groups and 1 methylene group or 3 methyl groups for a monoterpene and 3 methyl groups and 1 methylene group or 4 methyl groups for a sesquiterpene.

[0124] The monoterpene is typically a 1,8-terpene such as R-1,8-terpene, a pinene such as (+)-α-pinene, terpinene, sabinene, thujene, myrcene, ocimene, nerol or geraniol.

[0125] Sesquiterpenes are generally formed by a head-to-tail arrangement of three isoprene units. Sesquiterpenes are generally valencene, caryophyllene, longifolene, valencene, isobazzanene, silphinene, ishwarane, isopatchchoul-3-ene or isosesquicarene. It is particularly preferred that the sesquiterpene substrate is valencene.

[0126] Cyclic olefins generally contain up to 9 ring members, for example they may be 5, 6, 7, 8, 9 or more membered rings. The cycloolefin is typically cyclohexene.

[0127] Substituted derivatives of any of the terpenes or cycloolefins mentioned above may also be used. Typically, there are 1, 2, 3 or more substituents. Any combination of the following substituents may be present. The substituents are typically halogen atoms or oxygen- or nitrogen-containing groups or alkyl or alkenyl groups, which generally have 1 to 6 carbons, optionally substituted with one or more halogens.

[0128] The substituent typically has the formula C n H k X m , wherein X is a halogen, oxygen-containing or nitrogen-containing group, n is 1, 2, 3 or greater, m is 1, 2, 3, 4 or greater, and k is a suitable value such that the substituent C n H k X m The valence state of the alkyl group is satisfied. For an alkyl substituent k+m=2n+1. Typically, k is 1, 2, 3, 4 or more or can be 0, that is, the substituent is a perhaloalkyl group. The halogen is typically fluorine, chlorine or bromine. The substituent may also contain 1, 2 or more oxygen atoms and may be, for example, an alcohol, an aldehyde, a ketone or an epoxy group.

[0129] In another embodiment of the present invention, the substrate is a halogenated aromatic compound. The halogenated aromatic compound is typically a benzene or biphenyl compound. The benzene ring is optionally fused and can be substituted. The halogen is typically chlorine. In many cases, there are more than one halogen atom in the molecule, typically 2 to 5 or 6, for example, 3 halogen atoms. Generally speaking, 2 of the halogen atoms will be ortho or para to each other. The compound may or may not contain oxygen atoms such as hydroxyl groups, aryloxy groups or carboxyl groups. The compound may or may not be a chlorinated phenol or chlorinated phenoxyacetic acid compound.

[0130] Specific compounds that may be oxidized by the process of the present invention include 1,2-; 1,3- and 1,4-dichlorobenzene, 1,2,4-; 1,2,3- and 1,3,5-trichlorobenzene, 1,2,4,5- and 1,2,3,5-tetrachlorobenzene, pentachlorobenzene, hexachlorobenzene, and 3,3'-dichlorobiphenyl.

[0131] Other compounds that can be oxidized by the process of the invention include damascones, halogenated aromatic compounds, especially dioxins and halogenated dibenzofurans, and corresponding compounds in which one or both oxygen atoms are replaced by sulfur, especially dioxin compounds having at least one halogen substituent, such as dioxin itself, 2,3,7,8-tetrachlorodibenzodioxin.

[0132] The oxidation of halogenated aromatic compounds typically produces 1,2 or more oxidation products.The oxidized atom can be a ring carbon.These oxidation products will generally contain 1 or more hydroxyl groups.Therefore, in general, the oxidation product is a phenol, which can be degraded by various Pseudomonas (Pseudomonads) and other bacteria, and unoxidized halogenated aromatic compounds are difficult to oxidize.As described below, this makes enzyme of the present invention suitable for the site decontamination of halogenated aromatic compounds pollution.

[0133] Additional substrates contemplated for use in the methods of the present invention include, but are not limited to, chlorzoxazone, aniline, p-nitrophenol, nifedipine, thujone diastereomers, olefins (including propylene, 1-hexene and styrene), indoles, polycyclic aromatic hydrocarbons, propanolol, alkoxyresorufins (including 7-ethoxyresorufin, 7-methoxyresorufin, 7-pentyloxyresorufin, 7-benzyloxyresorufin), buspirone, testosterone, amodiaquine, dextromethorphan, acetaminophen, 3,4-methylenedioxymethamphetamine (MDMA).

[0134] The oxidation process performed using the mutant CYP102A1 enzyme of the present invention can be distinguished from the oxidation process performed by another wild-type or mutant CYP102A1 enzyme by the improved binding rate or product formation rate as defined above. The process of the present invention is also characterized by the formation of a specific product from the oxidized substrate, typically a product that is not formed by the wild-type CYP102A1 enzyme or another mutant CYP102A1 enzyme, or is formed in a negligible amount, i.e., less than 10%, 8%, 5%, 2%, 1% or less of the total amount of product. For example, the oxidation of propylbenzene can produce 2-propylphenol or 1-phenyl-2-propanol with high selectivity, or the oxidation of ethylbenzene can produce 2-phenylethanol and styrene.

[0135] Methods performed using the mutant CYP102A1 enzymes of the invention may also exhibit an altered ratio or number of oxidation products compared to oxidation methods performed by wild-type CYP102A1 enzymes or other mutant CYP102A1 enzymes. Where there is an altered ratio of products, the rate of product formation for a particular oxidation product is typically increased compared to a corresponding method performed by a wild-type CYP102A1 enzyme or other mutant CYP102A1 enzyme. The increased prevalence of a particular oxidation product may exceed at least 10%, 20%, 50%, more preferably 100%, 200%, 300%, 500% or more of the amount of said oxidation product in the product mixture formed by the wild-type CYP102A1 enzyme or other mutant CYP102A1 enzyme.

[0136] Specific examples of oxidation products that may exhibit enhanced oxidation in the process of the present invention include: i) 1-phenyl-2-propanol, wherein the oxidation substrate is propylbenzene; ii) 2-ethylphenol, wherein the oxidation substrate is ethylbenzene; iii) 1-phenyl-2-butanol or 4-phenyl-2-butanol, wherein the oxidation substrate is butylbenzene; iv) benzyl alcohol, wherein the oxidation substrate is toluene; v) 2-methyl-2-phenylpropan-1-ol, wherein the oxidation substrate is tert-butylbenzene; vi) 2-methylbenzyl alcohol, wherein the oxidation substrate is o-xylene; vii) carvacrol or thymol or 4-isopropylbenzyl alcohol, wherein the oxidation substrate is p-methylisopropylbenzene; viii) cymene, wherein the oxidation substrate is valencene; ix) 2-nonanol , wherein the substrate to be oxidized is nonane; x) 2-butanone or 2-butanol, wherein the substrate to be oxidized is butane; xi) 3-methyl-3-pentanol, wherein the substrate to be oxidized is 3-methylpentane; xii) 2-propanol, wherein the substrate to be oxidized is propane; xiii) trans-isopiperitoneol (trans-isopiperitenol), wherein the substrate to be oxidized is R-1,8-terpene; xiv) 2,3-pinene epoxide, cis-verbenol, trans-verbenol, wherein the substrate to be oxidized is α-pinene; xv) 9-fluorenol, wherein the substrate to be oxidized is fluorene; xvi) 8-hydroxydodecanoic acid and 7-hydroxydodecanoic acid, wherein the substrate to be oxidized is dodecanoic acid (lauric acid).

[0137] The method is typically carried out in the presence of a CYP102A1 enzyme, a substrate and the enzyme's natural cofactors, which are NADPH and molecular oxygen. In one embodiment, the method is carried out using an enzyme such as a dehydrogenase and its cosubstrate (cosubstrate, or cosubstrate) to oxidize NADP to NADP with the concomitant oxidation of the cosubstrate of the dehydrogenase. + Regeneration of NADPH. In another embodiment, the method is performed by regenerating the NADPH cofactor by electrochemical methods known to those skilled in the art.

[0138] It should be understood that the increased activity and altered selectivity result from substitutions in the heme domain of the CYP102A1 enzyme. Thus, the present invention also provides systems in which the substrate oxidation process is carried out in the presence of the heme domain (a) of the enzyme, a substrate, an electron transfer reductase (b), an electron transfer reductase (redoxin) (c), a cofactor of the enzyme, and an oxygen donor. In the system, the electron flow is generally: cofactor → (b) → (c) → (a).

[0139] (b) is generally an electron transfer reductase capable of mediating electron transfer from the cofactor to (c), such as a naturally occurring reductase or a protein homologous to a naturally occurring reductase, typically having at least 70% homology; or a fragment of a reductase or homolog. (b) is typically a reductase of any electron transfer chain found in a naturally occurring P450 enzyme system, and is typically a riboflavin-dependent reductase, such as pseudomonas reductase.

[0140] (c) is generally an electron transfer reductase capable of mediating electron transfer from a cofactor via (b) to (a), (c) is typically a naturally occurring electron transfer reductase or a protein homologous to a naturally occurring electron transfer reductase, typically having at least 70% homology; or a fragment of a reductase or homolog. (c) is typically a reductase of any electron transfer chain found in a naturally occurring P450 enzyme system, (c) is typically a 2Fe-2S reductase, such as a pseudomonad redox protein, or a flavodoxin protein.

[0141] Typically (a), (b) and (c) are present as separate proteins; however, they may be present in the same fusion protein. Typically, only two of them, preferably (b) and (c), are present in the fusion protein. Typically, these components are adjacent in the fusion protein without the presence of a linker peptide.

[0142] Alternatively, a linker may be present between components. Linkers generally include amino acids that do not have bulky side chains and therefore do not hinder the folding of protein subunits. Preferably, the amino acids in the linker are uncharged. The amino acids in the preferred linkers are glycine, serine, alanine or threonine. In one embodiment, the linker contains the sequence N-Thr-Asp-Gly-Gly-Ser-Ser-Ser-C. The linker typically has a length of at least 5 amino acids, such as a length of at least 10, 30 or 50 or more amino acids.

[0143] In this method, the concentration of enzyme, (b) or (c) is typically 10 -8 Up to 10 -2 M, preferably 10 -7Up to 10 -4 M. Generally speaking, the method is carried out at a temperature and / or pH at which the enzyme is a functional enzyme, for example when the enzyme has at least 20%, 50%, 80% or more peak activity. Typically, the pH is 3 to 11, such as 5 to 9 or 6 to 8, preferably 7 to 7.8 or 7.4. Typically, the temperature is 10°C to 90°C, such as 25 to 75°C or 30 to 60°C.

[0144] In this method, more than one different mutant CYP102A1 enzyme of the invention may be present. Typically, each mutant is capable of oxidizing a different substrate or may be better able to oxidize a given substrate than another enzyme, so a mixture of mutant CYP102A1 enzymes is employed to enable oxidation of a wider range of substrates. The method may also include wild-type CYP102A1 enzymes, other P450 enzymes or their homologs, any monooxygenases, and any other enzymes suitable for use in the desired synthesis or oxidation reaction.

[0145] In one embodiment, the method is carried out in the presence of a substance (e.g., catalase) capable of removing hydrogen peroxide byproducts. In another embodiment, the method is carried out in the presence of a full-length enzyme or a heme domain of the enzyme, a substrate, and an oxygen atom donor such as hydrogen peroxide or tert-butyl hydroperoxide, for example, using a peroxide shunt.

[0146] In other embodiments, the method is carried out in the presence of the full-length enzyme or only the heme domain of the enzyme, a substrate and oxygen in an electrochemical cell such that the two electrons required for oxygen activation and generation of the reactive intermediates are provided by an electrode, either by direct electron transfer from the electrode or indirectly via a small molecule modulator.

[0147] The method can be performed intracellularly or extracellularly. The cell is typically in culture, at a locus, in vivo or in situ (these aspects will be discussed below). The method is typically performed, for example, on land (e.g., in soil) or in water (e.g., fresh water or seawater) at a locus. When the method is implemented in culture, the culture medium typically contains different types of cells of the present invention, such as cells expressing different mutant CYP102A1 enzymes of the present invention. In general, such cells are cultured in the presence of an absorbable carbon source and a nitrogen source.

[0148] Typically, the cell in which the method is performed is one in which the mutant CYP102A1 of the invention or wild-type CYP102A1 is not naturally produced. In another embodiment, the mutant CYP102A1 enzyme is expressed in a cell that does not naturally produce wild-type CYP102A1, but at a higher level than that occurring in nature. The cell may produce 1, 2, 3, 4 or more different mutant CYP102A1 enzymes of the invention. These mutant CYP102A1 enzymes are capable of oxidizing different organic compound substrates, different short-chain alkanes or different alkylbenzenes.

[0149] This cell can be a prokaryotic or eukaryotic cell, and is generally any cell or any organism mentioned herein. Preferred cells are Escherichia coli, Pseudomonas sp., flavobacteria or fungal cells (e.g., Aspergillus and yeast, especially Pichia sp.). According to the purposes of the present invention, it is also conceivable that Rhodococcus sp. and Bacillus sp. are present. This cell may or may not be a cell capable of oxidizing any substrate or producing any oxidative product mentioned herein in its naturally occurring form. Typically, this cell is in a substantially isolated form and / or substantially purified form, in which case it generally contains at least 90%, for example at least 95%, 98% or 99% of the dry mass of the cell or preparation.

[0150] The cell is typically produced by introducing into the cell (i.e., transforming the cell with it) a vector comprising a polynucleotide encoding a mutant CYP102A1 enzyme of the present invention. It should be understood that due to the degeneration of nucleotide codes, more than one polynucleotide can encode each mutant CYP102A1 enzyme of the present invention. It should also be understood that the nucleotide sequence can be designed to exhibit a codon bias suitable for a particular cell or organism. The vector can be integrated into the genome of the cell or remain extrachromosomal. The cell can develop into an animal or plant discussed below. Typically, the coding sequence of the polynucleotide is operably linked to a control sequence that can provide expression of the coding sequence by the host cell. The control sequence is generally a promoter, typically a promoter of a cell in which a monooxygenase is expressed.

[0151] The term "operably linked" refers to a juxtaposition in which the elements described are in a relationship permitting them to function in their intended manner. An operating sequence "operably linked" to a coding sequence is joined in such a way that expression of the coding sequence is achieved under conditions compatible with the operating sequence.

[0152] The vector is typically a transposon, plasmid, virus or phage vector. It typically comprises an origin of replication. It typically comprises one or more selectable marker genes, for example, a gene for tolerance to ampicillin in the case of a bacterial plasmid. The vector is typically introduced into a host cell using conventional techniques (including calcium phosphate precipitation, DEAE-dextran transfection or electroporation).

[0153] The present invention further provides transgenic animals or plants, the cells of which are any of the cells of the present invention. Such animals or plants are transgenic for one or more mutant CYP102A1 genes. They can be homozygous or heterozygous for such genes, which are typically transiently introduced into the cell or stably integrated (e.g., in the genome). The animal is typically a worm (e.g., earthworm) or nematode. Such a plant or animal can be obtained by transforming suitable cells (e.g., embryonic stem cells, callus tissue, or reproductive cells), cultivating the cells if necessary, allowing the cells to develop into an animal or plant, and cultivating the animal or plant if necessary. The animal or plant can be obtained by sexual or asexual reproduction (e.g., cloning), proliferation of the animal or plant of the present invention or F1 organism (or any offspring removed from F1, or chimeras developed from transformed cells).

[0154] As discussed above, the method can be implemented at a site. Therefore, the present invention also provides a method for treating a site contaminated by a substrate of the present invention, such as a short-chain alkane, an alkylbenzene or a halogenated aromatic compound. The method comprises contacting the site with a mutant CYP102A1 enzyme, cell, animal or plant of the present invention. These organisms are then typically tolerant of oxidizing halogenated aromatic compounds. In one embodiment, the organisms used to treat the site are intrinsic to the site. Therefore, they can be obtained from the site (e.g., after contamination) and transformed / transfected (as discussed above) to express mutant CYP102A1 (and optionally appropriate electron transfer reductases and / or reductases (redoxin)).

[0155] In one embodiment, the site is treated with more than one type of organism of the invention, for example, with 2, 3, 4 or more types expressing different monooxygenases that oxidize different organic compound substrates, such as different short chain alkanes, alkylbenzenes or halogenated aromatic compounds. In one embodiment, such an aggregation of organisms among themselves is capable of oxidizing all substrates of a particular group present in the contaminated area, i.e., short chain alkanes.

[0156] The organisms (e.g., in the form of a collection) can be subjected to the method of the invention in a bioreactor (e.g., where they are present in an immobilized form). Thus, water or soil to be treated can be passed through such a bioreactor. The soil can be rinsed with water supplemented with a surfactant or ethanol and subsequently introduced into the bioreactor.

[0157] Screening design / isolation of mutants of the present invention

[0158] Methods disclosed to date for screening libraries of CYP102A1 mutants generated by random mutagenesis and genetic recombination tend to use surrogate substrates such as indole (indigo formation) and p-nitrophenol derivatives (detection of liberated p-nitrophenol). For some selected mutants with enhanced surrogate substrate oxidation activity, enhanced activity has been found for compounds of different structures, but changes in product selectivity are less common.

[0159] With regard to screening for product selectivity, WO2006105082 discloses a method for conjugating a product alcohol to a compound that can be detected spectroscopically. Similarly, improved selectivity for 1-octanol in octane oxidation by CYP102A3 was obtained by targeting this product in a dehydrogenase screening method (34). These methods tend to be searches for selectivity where the improvement in activity is small. Moreover, only those mutations that promote the formation of specific target compounds are found, while mutations at sites that affect product selectivity towards a wider range of compounds in different and potentially desirable ways have not been revealed.

[0160] On the contrary, the screening method for separating mutants according to the present invention utilizes the random library of screening mutations for indigo formation via indole oxidation and then searches for the enhanced activity and selectivity of the product of interest chemically. Therefore, the most active mutations from the indole oxidation screening are further screened via the in vivo oxidation of naphthalene, propylbenzene and octane. The product is analyzed by gas-liquid chromatography (see the embodiment part). Naphthalene is more hydrophobic than indole and is more similar to the hydrocarbon substrates often targeted in P450 catalysis. Propylbenzene is smaller than naphthalene but because of the competition between aromatic ring oxidation, benzyl oxidation and the attack at two unactivated aliphatic carbon centers, the test of product selectivity change is proposed. Octane shows different challenges, it is soft and less compact, and has 4 groups of different CH bonds that can be used for oxidation, so that the selectivity is biased towards the mutations of the end and the internal position. The variants whose product generation rate improves and / or product distribution changes are selected for in vitro activity research.

[0161] In this multi-step procedure, approximately 1500 colonies were screened in the initial indigo formation (activity) step, the number of which was reduced to approximately 800 colonies after genetic recombination of 11 first generation mutants. From among these 800 colonies, 130 colonies were selected for the in vivo substrate oxidation screening step. Of these 130 colonies, 5 variants were selected for further in vitro studies, all of which showed increased activity and / or altered product selectivity for a wide range of organic compounds, from naphthalene to pentane. Thus, the screening method used to isolate mutants according to the present invention allows for efficient and stringent selection of mutants with increased activity and / or product selectivity. The small number of initial libraries screened in the mutation discovery also suggests that the inventors' method for discovering variants of CYP102A1 has not yet reached its potential.

[0162] Another difference from previously disclosed directed evolution techniques is that the screen for CYP102A1 using indigo formation involved mutagenesis of specific sites in the CYP102A1 heme domain (e.g., site saturation), whereas the present screen involved random mutagenesis of the full-length heme domain gene. Thus, this screen has the potential to isolate an increased number of variants. Furthermore, it is apparent that site saturation mutagenesis can be applied to specific mutants isolated according to the present invention to isolate additional useful variants.

[0163] Example

[0164] Materials and methods

[0165] General reagents and chemical substrates of analytical grade or higher were from Alfa-Aesar, Fisher Scientific and Sigma-Aldrich or their subsidiaries. HPLC quality solvents were from Rathburn Chemicals (UK) and subsidiaries of Sigma-Aldrich and Merck. Buffer components were from Anachem, UK. NADPH (tetrasodium salt) was from ApolloScientific and Melford Laboratories. Isopropyl-β-D-thiogalactopyranoside (IPTG) was from Melford Laboratories. Restriction enzymes, T4 DNA ligase and related buffers were from New England Biolabs. Taq and KOD polymerases were from Merck Biosciences. Competent and supercompetent Escherichia coli strains were from Stratagene. Site-directed mutagenesis was performed using the PCR method described in the Stratagene Quik-Change mutagenesis kit. The appropriate length of the oligonucleotides flanking the codons changed in the mutagenic oligonucleotides was designed according to the manufacturer's instructions. Oligonucleotides were from MWG Biotech. General molecular biology procedures were performed according to literature methods (Sambrook, J., Fritsch, EF, and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2 nd Ed., Cold Spring Harbor Laboratory Press, New York). All mutant genes were fully sequenced by the Oxford University Department of Biochemistry on an automated ABI377XL Prism DNA sequencer. UV / visible spectroscopy and enzyme activity analysis were performed on a Varian Gary 50 spectrophotometer at 30°C. 1 H NMR spectra were acquired on a Varian Unity Plus 500 MHz spectrometer. Gas chromatography (GC) was performed on a Thermo Finnigan Trace and 8000 Top instrument equipped with a flame ionization detector (FID) using a DB-1 fused silica capillary column and helium as carrier gas. The injector was maintained at 200°C or 250°C, while the FID was maintained at 250°C.

[0166] Mutagenesis, design, directed evolution, and screening procedures

[0167] SpeI restriction site using oligonucleotide: 5′ (The SpeI recognition sequence is underlined) and its reverse complement were introduced downstream of the pGLW11 heme domain-coding region (13) of the CYP102A1 gene, resulting in silent mutations at residues 482 and 483.

[0168] Error-prone PCR using forward and reverse primers between this site and the EcoRI site upstream of the heme domain-coding region: 5′ (EcoRI recognition sequence is underlined); and (The SpeI recognition sequence is underlined).

[0169] Libraries were constructed from wild-type CYP102A1 (WT) and mutant F87A templates under conditions designed to introduce 1 to 3 mutations / 1000 bp according to the adopted Stratagene GeneMorph protocol. The genes were amplified for 30 cycles, each cycle consisting of 60 s of strand separation at 94°C, 90 s of annealing at 45°C, and 110 s+2 s of extension at 68°C. After digestion with EcoRI and SpeI, the short fragments were reintroduced into pGLWl1 (SpeI WT variant) using T4 DNA ligase, transformed into E. coli DH5α competent cells and cultured on Luria-Bertani (LB) agar plates for 36 h.

[0170] About 1500 colonies were screened. Those showing indigo formation (Gillam, E.MJ et al (1999) Biochem.Biophys.Res.Commun.265,469-472; Li, QS, et al (2000) Chem.Eur.J.6,1531-1536) were isolated, transferred to new plates and cultured for another 36 h to minimize false positives before sequencing. Eleven variants showing 16 new mutations of potential interest were subsequently recombined by random-priming recombination (Shao, Z., et al (1998) Nucleic Acids Res.26,681-683).

[0171] The protocol given by Volkov and Arnold (Volkov, AA, and Arnold, FH (2000) Methods Enzymol 328, 447-456) was modified at stage 9, using Taq and KOD polymerase and 2 μL MgSO4 instead of Pfu polymerase. PCR was performed on the assembly strands as described above, but using KOD polymerase. Samples were digested, ligated and plated as before.

[0172] Of the approximately 800 colonies, approximately 130 exhibited indigo formation. These were grown on a 5-10 mL scale and screened in vivo for naphthalene and propylbenzene oxidation activity using gas chromatography. Twelve variants that showed an increase in the maximum product peak area or an altered product distribution relative to the WT were sequenced, grown on a larger scale and screened against the same two substrates. Of these, five were selected for in vitro studies.

[0173] We also prepared various single-site mutants and their combinations with the five mutants from the random mutagenesis screening program and tested them in vivo for indigo formation and naphthalene, propylbenzene, and octane oxidation activities using gas chromatography. These single-site mutations are R47L, Y51F, E267V, I263A, A74G, L188Q, M177V / K, A399P, I401P, G402P, Q403P, V302I, A264G, A99T, S270I, R179H, and F87L / A / G. Most of these are either previously known mutations (at R47, Y51, I263, A74, L188, M177, A264, F87) or belong to those we discovered in earlier rounds of the random mutagenesis / screening process (at E267, V302, A99, S270, R179) that we have selected for additional rounds based on their position in the structure and that they may have some effect on substrate binding. Proline mutations at A399, I401, G402, and Q403 are based on mutations in the A330P mutant ( Figure 2 ) was selected based on the structural changes observed in the crystal structure of , which was one of the five mutants selected for the screening program. These mutants were then also screened for evidence of increased activity by in vivo screening methods.

[0174] The R47L and Y51F mutations, either by themselves or in combination with other mutations, were not as effective as the R47L / Y51F couplet. Other single site mutations showed varying effectiveness in the indigo screen, while the I401P and Q403P single site mutations showed increased product formation in the in vivo oxidation screen relative to wild type, and these mutants were prepared and their activities studied in vitro. Variants R47L / Y51F (RLYF) and F87A were prepared as described (13). The RLYF couplet was introduced into variants KT2 and A330P using standard cloning procedures using NcoI and AflII restriction sites (Sambrook, J., Fritsch, EF, and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2 ndEd., Cold Spring Harbor Laboratory Press, New York). The F87A mutation was introduced into the KT2, I401P, RYLF / I401P, and RYLF / Q403P variants by site-directed mutagenesis. The I401P and Q403P mutations were introduced into the RLYF and RLYF / A330P variants by site-directed mutagenesis.

[0175] Protein expression and purification

[0176] The variant of interest was transferred into the pET28 vector using the NcoI and BamHI restriction sites, allowing tighter control of expression levels in the pGLW11 vector using the T7 promoter compared to the tac promoter. -1 An overnight culture of E. coli JM109 (DE3) containing the plasmid was grown to contain 0.4% (v / v) glycerol and 30 mg.L -1 The culture medium was added with kanamycin and incubated at 37°C with shaking at 180 rpm until the OD 600 >1. Protein expression was induced by adding isopropyl-β-D-thiogalactopyranoside (PTG) to 0.4 mM. The temperature was lowered to 30°C and the cells were harvested by centrifugation after another 12 h of incubation at 30°C. The reddish-brown pellet per 1 L of growth was resuspended in 25 mL of 40 mM potassium phosphate, buffered in pH 7.4, 1 mM dithiothreitol (phosphate buffer). The cells were lysed by sonication and the cell debris was removed by centrifugation at 37500 g for 30 min at 4°C. The supernatant was loaded onto an Amersham-Pharmacia DEAE fast flow Sepharose column (200×50 mm) pre-equilibrated with phosphate buffer, whereby the protein was eluted using a linear gradient of 80 to 400 mM ammonium sulfate in phosphate buffer. The red P450 fraction was collected and concentrated by ultrafiltration, desalted using a Sephadex G-25 column pre-equilibrated with phosphate buffer, and reconcentrated by ultrafiltration. The solution was centrifuged at 9250 g for 5 min at 4°C and filter sterilized. FPLC anion-exchange purification was performed on an Amersham-Pharmacia Source-Q column (120×26 mm) using a 0-30% linear gradient of 15× phosphate buffer. 418 M 280 Fractions >0.35 were concentrated by ultrafiltration and stored in 50% (v / v) glycerol at -20°C after filter sterilization. Glycerol and salts were removed from the protein immediately prior to the experiment using an Amersham Pharmacia 5 ml PD-10 column pre-equilibrated with 50 mM pH 7.4 Tris buffer.

[0177] NADPH conversion rate determination

[0178] NADPH turnover (except butane and propane) was performed at 30°C in 1250 μL 50 mM Tris (pH 7.4) containing 0.1 or 0.25 μM enzyme, 125 μg bovine liver catalase and 1 mM substrate added as a 100 mM stock in DMSO. Protein concentration was determined as described (13) or by CO difference spectroscopy (Omura, T and Sato, R (1964) J. Biol. Chem. 239, 2379-85). The analysis was performed at 20 mg ml NADPH. -1 The stock solution was kept at 30°C for 1 min before adding to a final concentration of ~160μM or ~320μM (equivalent to 1 or 2AU). In the butane and propane conversions, the substrate was bubbled into 3000μL Tris on ice for a minimum of 30min while oxygen was bubbled into 1000μL Tris also on ice. CYP102A1 (0.25μM) and catalase (concentrations as above) were gently added to the oxygenated portion, followed by the addition of substrate-saturated Tris. The entire container was quickly sealed, inverted several times, and kept at 30°C for 2min before NADPH was added to 1AU.

[0179] In all transformations, the absorbance decay at 340 nm was monitored and ε 340 =6.22mM -1 cm -1 NADPH consumption rates were extrapolated. To ensure accurate coupling determinations, enzyme concentrations up to 2.5 μM were used, driving slower WT and F87A conversions to completion when necessary. Data from at least 3 experiments were averaged to ±5% (aromatic compounds) or ±10% (alkanes and all NADPH rates below 200 min -1 conversion).

[0180] Product analysis

[0181] For substrates other than lauric acid, 3 μL of internal standard (100 mM in DMSO) was added to 1000 μL of each completed turnover before extraction into 400 μL of ethyl acetate or chloroform. Centrifugation was performed at 21000 g in 1500-μL microcentrifuge tubes for 3 min. 1 / 2min. Products were identified by comparing GC elution times observed with authentic equivalents. FID responses were calibrated using representative equivalents for each product group as detailed in the table below, using the assumption that isomeric monooxidized products give similar responses. Samples containing a known concentration range of the selected products and 1 mM in DMSO were prepared in Tris and extracted as above. The derived integrated peak areas are expressed as a ratio of the internal standard peak area and plotted against product concentration. 2-Methyl-2-phenyl-propan-1-ol, which is unlikely to be of commercial origin, was produced in vivo, isolated and identified by MS: M149.00; and 1 H NMR: d 1.38 (6H, s, gem dimethyl), 3.59 (2H, s, CH2), 7.24 (1H, m, p-phenyl), 7.34 (2H, m, m-phenyl), 7.37 (2H, m, o-phenyl). For lauric acid oxidation of CYP102A1 and its mutants, 990 μL of the incubation mixture was mixed with 10 μL of internal standard solution (25 mM decanoic acid in ethanol) and 2 μL of concentrated HCl. The mixture was extracted three times with 400 μL of ethyl acetate, and the organic extracts were combined and dried over MgSO4. The solvent was evaporated under a stream of nitrogen and the sample was dissolved in 200 μL of acetonitrile. Excess (25 μL) N,O-bis(trimethylsilyl)trifluoroacetamide and trimethylchlorosilane (BSTFA+TMCS, 99:1) were added and the mixture was maintained for at least 120 min to generate trimethylsilyl esters of carboxylic acid groups and trimethylsilyl ethers of alcohols, if generated. The reaction mixture was used directly for GC analysis.

[0182] Calibration (galibrants), internal standards and oven temperature in GC product analysis

[0183]

[0184]

[0185] result

[0186] Of the variants generated by random mutagenesis, all of which were screened in vivo for enhanced activity and altered product selectivity, five specific variants were selected for in vitro studies. These are:

[0187] (i) A330P

[0188] (ii) A191T / N239H / I259V / A276T / L353I (mutant KT2)

[0189] (iii) F87A / H171L / Q307H / N319Y (mutant KSK19)

[0190] (iv) F87A / A330P / E377A / D425N (mutant KT5)

[0191] (v) F87A / A117V / E131D / L215I (mutant LO25)

[0192] All five mutants were readily expressed and purified by standard procedures (13). None showed evidence of production of the inert "P420" form after storage in 50% v / v glycerol at -20°C for at least 15 months. Mutant LO25 was less studied than the other four variants, but had the highest activity and selectivity for damascone / ionone molecules among the five initial mutants, e.g., 86% production of hydroxydamascone product. The activities of A330P, KT2, KSK19, and KT5 were assayed using a wide range of substrates. Data for some of these are given in Tables 1-20, where NADPH oxidation and product formation rates (PFRs) are reported in nmol (nmol P450). -1 min -1 Given and abbreviated as min in this article -1 The binding efficiency is the product yield based on the consumed NADPH and is given as a percentage, in some cases the catalytic parameters were compared with the A74G / F87V / L188Q (GVQ) mutant disclosed in NO20020380.

[0193] The four variants of the present invention (A330P, KT2, KT5, KSK19) enhanced the PFR of WT with naphthalene (3.1 min -1 ) at least one order of magnitude, A330P in 155min -1 The most effective, but none could match the 487 min recorded by the GVQ variant (Table 1). -1 In all cases, 1-naphthol was the only GC detectable product (23).

[0194] WT at 606min -1 , 70% binding to p-propylbenzene is considerably more active. Three of the four variants of the present invention have similar activity to the GVQ variant (943 min -1 ) of PFR, and the fourth, KT2 with 2205min -1 significantly exceeded it (Table 2). WT and KT2 obtained >99% 1-phenyl-1-propanol, but other novel variants directed oxidation away from the activated benzylic position. Variant A330P produced 30% ortho-phenol, a product type not previously reported in CYP102A1 conversions, while variant KT5, in which mutations F87A and A330P appeared in combination, gave 80% 1-phenyl-2-propanol. Mutation F87A is known to promote the formation of l-phenyl-2-propanol, but when acting alone gave only 54% (46).

[0195] The R47L / Y51F (RLYF) couple, which has been shown to increase the activity of CYP102A1 for several substrates (13, 14, 17), was introduced into the variants A330P and KT2 with the goal of further increasing the rate of product formation. Two highly active second-generation variants were obtained, RLYF / KT2 and RLYF / A330P. Single-site mutations were also introduced at various residues around the heme surface as well as in the active site, and these were also combined with the five mutations identified in the first round. The activity of these second-generation mutants was screened by indigo formation and in vivo oxidation screening procedures. The I401P and Q403P mutants were identified as promising new variants from the in vivo oxidation procedure and were prepared. The combination mutants R47L / Y51F / I401P, F87A / I401P, and R47L / Y51F / A330P / I401P were also prepared.

[0196] RLYF / KT2 is similar to the GVQ variant in PFR in naphthalene conversion 496min -1 , RLYF / A330P is 666min -1 , exceeding it by about 35%, while I401P is still more active, at 1183 min -1 Q403P showed 121min -1 The PFR of RLYF / KT2 for propylbenzene was 2688min -1 , a 22% improvement over KT2. This rate is close to the reported value for WT on natural substrates (47), while the 3578 min for I401P -1 The rate value exceeded that of the natural substrate. The product distribution was less changed relative to the first generation variants, but RLYF / A330P produced some p-phenylphenol (Table 2). KT2 was also combined with mutant F87A to obtain an F87A-directed product distribution at a higher rate. Variant F87A / KT2 oxidized propylbenzene to 1-phenyl-1-propanol and l-phenyl-2-propanol in approximately equal amounts, consistent with mutant F87A, but with a PFR of 566 min. -1 And F87A is 241min -1 .

[0197] Other alkylbenzenes were also tested as substrates for WT CYP102A1 and the new variants. Significant activity enhancements were observed with toluene (Table 3), especially with RLYF / KT2, RLYF / A330P, and I401P. The effect of the Q403P mutant was again similar to that of variant KT2. Large selectivity changes were also significant, although the variants had reduced product formation rates compared to the fastest mutant. WT CYP102A1 oxidized toluene primarily to o-cresol (98%). This epoxidation was unexpected because the benzyl CH bond is highly activated. For example, WT CYP101A1 from Pseudomonas putida attacks the benzyl position to produce >95% benzyl alcohol. Variant RLYF / A330P increased the PFR by 60 to 189 min. -1 The binding efficiency increased from 9% to 52% with a factor of 0.1, while maintaining minimal side chain oxidation. Addition of the I401P mutant produced another increase in activity, with NADPH and substrate oxidation rates increasing to 3732 min, respectively. -1 and 1824mn -1 It is apparent that the 49% binding efficiency is essentially unchanged from that of the RLYF / A330P (52%) and actual A330P (45%) variants, indicating that the main effect of the I401P mutation is to enhance the rate of NADPH conversion. The other variants showed both increased conversion activity and altered selectivity, e.g., variant F87A / KT2 produced 48% benzyl alcohol, while variant KT5 yielded 95% benzyl alcohol and only 5% o-cresol (Table 3).

[0198] NADPH rates and binding were generally lower for butylbenzene than for propylbenzene, with a PFR of 229 min recorded for RLYF / KT2 for both WT and the fastest variant. -1 and 1670min -1 (Table 4). Hydrogenation was no longer exclusive to benzyl groups, even in the WT-RLYF / KT2 subgroup, with ~10% occurring at each of the next two positions of the side chain. Most of the other specificity changes mirrored those observed with propylbenzene. The F87A variant increased oxidation at non-benzyl positions—up to 80% of KT5—while variant F87L / KT2 produced overwhelming amounts of o-butylphenol (32%). However, A330P favored the production of p-butylphenol, especially in combination with RLYF (26%), while simultaneously reducing the level of benzyl oxidation to only 10%-13%.

[0199] Using tert-butylbenzene, NADPH conversion rates were consistent with those of butylbenzene for most variants, but incorporation levels were greatly reduced except in the case of the F87A variant (Table 5). The GVQ variants, F87A / KT2, and KSK19 all increased PFR by two orders of magnitude relative to WT, with the highest rate being achieved by the new variant F87A7KT2 at 234 min -1 vs.WT2.4min -1 The F87A and F87V variants exclusively hydroxylated the non-activated CH bonds of the side chain to produce 2-methyl-2-phenyl-l-propanol, a compound that is difficult to synthesize by conventional methods. Most of the products formed by WT and other variants were phenols, with para-hydroxylation taking precedence over ortho-hydroxylation, but the A330P mutation again increased the tendency of aromatic oxidation, with the product further shifted to p-phenol.

[0200] For ethylbenzene (Table 6), the variants of the present invention generally showed a higher -1 ) enhanced activity. RLYF / KT2, the fastest variant, gave 1098 min -1 RLYF / A330P also gives a high PFR (1062min -1 ), in part because the A330P variants were able to bind ethylbenzene better than the WT (55%-62% vs. 28%). In contrast, the binding rates of the Phe87 variants remained lower than those of the WT (22%-30%), which again showed a lower specificity for the benzylic position in the conversion of propylbenzene, forming 10% of o-ethylphenol. Mutants containing A330P produced a higher percentage of this product (21%-27%), while the F87A and F87V variants, such as KSK19 and GVQ, excluded it from the product mixture and produced 100% 1-phenylethanol. The KT5 conversion produced small amounts of two other products: 2-phenylethanol and styrene. The former originates from a non-activated primary CH bond at the ethyl substituent, while the latter represents the inventors' first observation of the dehydrogenation of simple hydrocarbons catalyzed by CYP102A1.

[0201] Preferential attack at the methyl group next to the activated benzylic carbon is difficult to achieve because if both types of bonds were equally accessible to the P450 ferryl intermediate from the same distance, the more activated CH bond would be attacked more rapidly. Thus, KT5 can bind ethylbenzene in an orientation where the methyl CH bond is closer to the ferryl than the benzylic CH bond. In the dehydrogenation reaction, the ferryl intermediate captures the hydrogen atom to form the Fe IV -OH intermediate is then replaced by IVThe hydroxyl group of the -OH moiety undergoes recombination cleavage, and the substrate radical abstracts the second hydrogen atom from the substrate to form an alkene and water. The dehydrogenation of 3-methylindole by mammalian P450 enzymes was first reported in 1996 and has since been extended to human P450 enzyme dehydrogenation of indoline, capsaicin, and drugs (48-52). Aromatization of nifedipine via CYP102A1 dehydrogenation has also been reported (53). However, the reaction is driven by the formation of two delocalized aromatic systems.

[0202] The naturally occurring hydrocarbon, p-cymene (4-isopropyltoluene), is a substrate of interest because it is a precursor to four flavoring compounds. WT gave 82% p-α,α-trimethylbenzyl alcohol, which originates from the oxidation of the methine C-H bond of the isopropyl side chain. Small amounts of two possible aromatic hydroxylation products, thymol (3%) and carvacrol (7%), and only 2% of 4-isopropylbenzyl alcohol were also formed. In contrast, variant A330P gave 19% thymol, 17% carvacrol, 22% 4-isopropylbenzyl alcohol, and only 37% p-α,α-trimethylbenzyl alcohol (Table 7). F87L / KT2 gave 74% 4-isopropylbenzyl alcohol and 21% carvacrol, with p-α,α-trimethylbenzyl alcohol accounting for only 5% of the product mixture for most of the products in general. Enhanced activity was observed, with variant KSK19 giving 1442 min -1 The PFR of WT was 168min. -1 p-α-Dimethylstyrene is formed via dehydrogenation of the isopropyl ring substituent. Moreover, once formed, this compound is also a substrate for the enzyme, and small amounts (1% to 3% of total product) of the corresponding styrene oxide are also observed. Variants containing F87A and F87V minimize or eliminate phenol formation, but give significant amounts (>20%) of p-α-dimethylstyrene.

[0203] The variants of the present invention show enhanced cumene oxidation activity. Single site mutants A330P and Q403P, as well as KSK19 variants show similar activity enhancement relative to wild type, mainly due to increased NADPH conversion rate, which is also higher than KT5 variant. WT mainly produces benzyl oxidation products, while KT5 gives 27% 1-methylstyrene by dehydrogenation and also gives 1% styrene oxide product by further oxidation (Table 8a). These results indicate that aromatic compounds with primary CH bonds at the 2-position of the alkyl substituent can undergo dehydrogenation. This reaction pathway can form the basis for a method for synthesizing substituted styrenes as polymer precursors (e.g., preparing vinyl anisole from ethyl anisole).

[0204] KT2 and A330P showed greatly enhanced activity towards short chain alkanes relative to the wild type, especially in combination with RLYF (Tables 9-13). The I401P mutant also proved to be highly active. RLYF / KT2 and RLYF / A330P showed similar product formation rates towards pentane (1206 min -1 and 1183min -1 , based on 60% and 67% binding efficiencies, respectively), which is 75-fold higher than WT. The increase in binding efficiency in these mutants is important because it indicates a greatly improved fit between the active site topology and the substrate.

[0205] These variants showed high activity in 3-methylpentane (all >1000 min -1 vs.WT~20min -1 , Table 10), 2-methylbutane (RLYF / KT2 and RLYF / A330P both >1000min -1 , while I401P has a lower activity of 721min -1 vs.WT51min -1 , Table 11) and butane. The I401P mutation itself mainly increased the NADPH conversion rate, while it combined well with the RLYF coupler to increase the PFR of 3-methylpentane oxidation to 2980 min -1 On the other hand, the triple mutant RLYF / A330P showed significantly better binding to propane than RLYF / KT2, with a peak of 46 min. -1 vs 5.8min -1 The addition of the I401P mutant increased the NADPH conversion rate, while moderately increasing the binding effect, resulting in a PFR of 430 min for propane oxidation by the RLYF / A330P / I401P mutant. -1 These rates are comparable to the 23 min of previously reported CYP102A1 variants 9-10A, 1-12G, and 53-5H, which each contain 13 to 15 mutant alkane hydroxylases. -1 , 160min -1 and 370min -1 The GVQ variant used as a control gave a higher NADPH conversion rate for propane than RLYF / A330P (400 min - 1 vs180min -1). To our knowledge, its ability for this substrate has been reported. However, binding was only 0.7% vs 21% for RLYF / A330P. This is an extreme, but characteristic example of how A330P and KT2 compare to GVQ across the entire range of substrates studied. Although the NADPH rates for these two variants are often lower, the overall product formation rates are generally higher due to more efficient binding. When this property is combined with the I401P mutation, which primarily increases conversion rates, the resulting mutations, such as R47L / Y51F / A330P / I401P, can achieve significant activity increases over wild type for a range of non-natural substrates.

[0206] Long-chain alkanes are poor substrates for most mutants, with RLYF / KT2 and RLYF / A330P giving PFRs of 246 min for octane, respectively. -1 and 230min -1 , WT is 53min -1 (Table 13). Q403P showed a modest 2-fold improvement in activity over the wild type, to 104 min -1 The PFR of I401P was 709min. -1 RLYF / KT2 and I401P showed similar product selectivity to the wild type, but variant A330P significantly enhanced the production of 2-octanol at the expense of 3- and 4-octanol (53% compared to 15% for WT, Table 13) (31,54). This effect of direct oxidation of alkanes in the terminal position is a useful advantage for combination with other mutations in a comprehensive search for terminal alcohol synthesis via direct alkane oxidation.

[0207] The variants of the invention also showed increased activity for the oxidation of chlorinated aromatic compounds, as exemplified by 1,4-dichlorobenzene (1,4-DCB). This increase was mainly due to higher NADPH conversion activity, while the binding was not much higher than the WT. The highest binding was 15% for the A330P variant (Table 14). Only one product was detected by gas chromatography, but it was not possible to determine whether it was 2,4- or 2,5-dichlorophenol under the conditions used. These results confirm the effectiveness of the variants of the invention for chlorinated benzene and aromatic oxidation.

[0208] We have previously reported on the oxidation of the sesquiterpene valencene by WT CYP102A1 and mutants such as F87A (14). This enzyme produces a variety of products including nootkaytol, valencene, and epoxidation products. The F87A mutation has been shown to slightly shift product selectivity toward valencene (approximately 20%). In contrast, variant KSK19 increased the rate of valencene oxidation by 30-fold relative to WT (Table 15). More importantly, it doubled the rate of grapefruit-scented valencene production relative to mutant F87A. Variants F87A / KT2 and KT5 increased the proportion of valencene to approximately 30%, while also themselves increasing the PFR relative to F87A.

[0209] WO0031273 discloses monoterpene oxidation by CYP102A1. The new variants exhibited higher activity than WT and previously reported mutants. The catalytic performance of the new variants was compared with that of WT for R- and S-1,8-terpenoid oxidation (Table 16; 16a). Of particular note, the two enantiomers produced different products using the two enzymes, but the conversion activity and binding were closely similar, indicating the asymmetric nature of the CYP102A1 substrate pocket. 1,8-terpenoid oxidation activity was increased in all variants, with Q403P, I401P and the R47L / Y51F / I401P triple mutations showing activity equivalent to or higher than that of wild-type CYP102A1 for the oxidation of a fatty acid natural substrate (lauric acid) (1439 min -1 , see Table 20). Again, these mutants showed similar product selectivity to the wild type, producing mainly carveol, while mutations containing A330P and F87A had altered selectivity, with isopiperitenol being the major product. It is also apparent that both KT2 and I401P functioned well as general accelerator (catalyst) mutations. For (+)-α-pinene, WT had little activity while the F87A / KT2 variant had 206 min -1 The PFR of 1400 μmol / L was increased by 1.1% and the product selectivity was shifted to verbenol (Table 17). The effect of introducing the F87G mutation will be of interest. The I401P and F87A / I401P mutations are more active while maintaining verbenol as the major product. The combination of I401P and the R47L / Y51F couplet increased the PFR to 1146 min. -1 However, the product shifted to 56.5% of cis-l,2-oxide, and the selectivity was further enhanced by the A330P mutation.

[0210] Fluorene is a more stereostringent substrate than naphthalene and was used in the in vivo screening procedure. The R47L / Y51F combination slightly improved activity, and the addition of the A330P mutation increased the NADPH conversion rate to 510 min -1 , but the binding efficiency was lower (Table 18). The Q403P and I401P mutations had more important effects, and the R47L / Y51F / I401P combination was particularly active, with a 0.1 min increase compared to the wild-type. -1 , PFR is 582min -1 Ionones are precursors to flavor compounds. Table 19 shows the β-ionone oxidation activity of the wild type and some of the novel variants, highlighting the potential increase in activity with combination variants such as R47L / Y51F / I401P.

[0211] Lauric acid (dodecanoic acid) is a known natural substrate for CYP 102A1, with an NADPH conversion rate of 2777 min -1 , and PFR is 1439min -1 . The structural perturbations introduced by the A330P mutation essentially abolished lauric acid oxidation by CYP102A1, while the I401P mutation enhanced the NADPH conversion rate while maintaining binding efficiency, resulting in a 40% increase in activity of the mutant over the wild type for oxidation of the natural substrate. Clearly, the rate of first electron transfer was increased, suggesting changes in the redox potential, heme spin state, and reorganization energy for the process in this mutant. The R47L / Y51F / I401P mutant showed an even higher NADPH conversion rate but reduced binding, presumably because the R47 and Y51 side chains were unavailable for carboxylate anchoring, altering the linkage and therefore binding. The F87A mutation is known to shift lauric acid oxidation to sub-terminal carbons, and the I401P mutation increased activity while maintaining the selectivity-altering effect of the F87A mutation.

[0212] These findings demonstrate that despite the significant advances in the prior art, the CYP102A1 system remains fertile ground for the application of directed evolution and site-directed mutagenesis techniques, and variants with improved activity and product selectivity can be characterized. All variants of the present invention include many variants that have not been previously disclosed to the knowledge of the inventors.

[0213] The A330P mutation has the welcome effect of increasing activity towards a wider range of compounds, while also altering the product distribution when acting on its own and when combined with another selectivity-altering mutation (F87A). On the other hand, I401P acts as a general rate accelerator with little effect on selectivity. I401 is close to the heme on the β-bulge and can affect electron transfer. A330P is an unlikely product of directed evolution, for its potency depends on individual substitutions of a concert party of residues derived from single-site mutations rather than the usual alterations. A proline was introduced directly next to an existing proline at position 329, the substrate contacting residue at the end of the β-strand. The resulting loss of backbone flexibility is predictable and is supported by the crystal structure ( Figure 2 ) appears to restrict the active site pocket, allowing tighter substrate binding and enhancing activity. This interpretation is consistent with the unusual and potentially useful observed selectivity effects, which are typically a reversal of those produced by F87A, where the active site pocket is more open than WT. It will be interesting to see whether the strategy of juxtaposing proline residues can be usefully expanded elsewhere in CYP102A1 or indeed in the redesign of other enzyme systems.

[0214] The crystal structure of the A330P mutant was obtained and is shown in Figure 2 The Calpha positions of this mutant are largely superimposable to those of the wild type, but there are also significant rearrangements at positions 328, 329, and 330. Figure 2 As shown in , the introduced A330P mutation induced a significant migration of the Pro329 loop toward the substrate binding pocket, reducing the active site volume and compressing the substrate access channel in a critical region. These structural changes are most likely responsible for the anomalous effects observed with the A330P mutation, such as enhanced binding of non-natural substrates and altered product selectivity, due to the protrusion of the Pro329 loop into the substrate pocket. However, they did not result in abrupt termination of secondary structural elements such as helices.

[0215] Based on these unexpected findings, the potential for remodeling loop regions by introducing bulky residues such as proline was further investigated. Specifically, proline substitutions were implemented in the A1a399-Gln403 loop, which provides the closest heme ligand, Cys400. Arg398 is involved in electrostatic and hydrogen bonding interactions and may be important in stabilizing protein folding, while residues beyond Gln403 may be removed too far from the heme to have any effect. The mutations explored were therefore A399P, I401P, G402P and Q403P. As shown in the Examples section, both I401P and Q403P produced a significant enhancement of enzyme activity.

[0216] Mutation of residues 401 and 403 to proline can induce conformational changes in the nearest ring, such as changing the strength of the Fe-S bond interaction by changing the Fe-S distance. This can make the heme iron more easily reduced, and the reorganisation energy barrier for electron transfer will be lowered, which will facilitate the movement of the heme iron into the plane of the porphyrin ring during the catalytic cycle.

[0217] Reorganization of the loop regions in CYP102A1 via the introduction of bulky residues such as proline is a unique and specific structural mechanism common to the A330P, I401P and Q403P substitution mutants of the present invention.

[0218] Variant KT5, in which A330P and F87A appear together, promotes the selectivity-shifting properties of F87A over A330P. However, the conversions involved can be more dramatic than those produced by other F87A-containing variants (e.g., 95% benzyl alcohol from toluene vs. 48% for F87A / KT2, 23% for F87A, and 2% for WT), and often occur at an increased rate of product formation, yielding a range of complementary possibilities. It will be interesting to see how A330P combines with other known selectivity-directed mutations, such as A264G, A82L, and A328V.

[0219] Variants KT2 (A191T / N239H / I259V / A276T / L353I) and F87A / KT2 gave product distributions closely similar to WT and variant F87A, respectively. Similarly, I401P accelerated the rate over the same substrate range of product formation as the wild type. Thus, KT2 and I401P act as rate accelerators. The component mutations N239H and 1259V in KT2 have been previously reported (55). However, variant KSK19, which contains the original F87A, has the same selectivity pattern as F87A / KT2 and is slightly more active across the entire substrate range despite containing only three other mutations, suggesting that, when prepared, the F87A-free derivative of KSK19 may prove to be more potent than KT2 as a rate accelerator.

[0220] Among the mutations that appear to act as rate accelerators in KT2 and KSK19 (H171L, A191T, N239H, I259V, A276T, Q307H, N319Y, and L353I), Gln307 and Asn319 ( Figure 1 ) is close to a region thought to be the docking site for the reductase domain, from which the heme domain receives electrons (56) and could affect electron transfer kinetics. Leu353 is close to the substrate entry channel residue, Met354, while A1a191 (located on the outer edge of the entry channel) is significantly displaced upon palmitate binding (57) and could play a role in substrate entrapment and / or capture. The remaining four mutations (Hisl71, Asn239, Ile259, A1a276) are all on or close to the protein surface, and the detailed mechanisms of their action remain unknown. It can be speculated that the effects of at least some of the mutations can be rationalized in terms of their context in the CYP102A1 structure. The conversion activity of P450 enzymes is rate-limited by the rate at which the first electron transfer step of the catalytic cycle is initiated. The rates of electron transfer reactions are often discussed in terms of Marcus theory, which states that the activation energy for electron transfer depends on the thermodynamic driving force (the change in free energy of the reaction) and the reorganization energy (the energy input required to deform the reactant state to be consistent with the product state). Substrate binding plays a major role by altering the electronic properties of the heme (most often by substitution of the sixth ligand of the heme iron), thereby making the reaction thermodynamically more favorable (higher driving force) and lowering the reorganization barrier for electron transfer (requiring less deformation of the reactant state) (58). If the active site structure is altered, for example by substitution, binding of non-natural substrates can be enhanced and faster substrate oxidation is observed.

[0221] Another mechanism for altering substrate binding (and therefore the electronic properties of heme) may be the induction of changes in secondary structural elements surrounding the substrate pocket. In P450 enzymes, the substrate pocket is typically defined by residues of the B and B' helices, the BC loop, the F / G loop, the G helix, and the I helix. Amino acid substitutions at residues distal to the substrate pocket may alter substrate binding by inducing changes in the positions of these secondary structural elements. His171 is at the start of the F helix and contacts the G helix at L215. N239 is in the H helix, and this helix contacts the N-terminal end of the I helix. Substitutions at H171 and N239 will affect the positioning of the G and I helices, respectively, and may alter substrate binding. Ile259 and A1a276 are both in the I helix. Although neither of these residues contacts the substrate, amino acid substitutions may affect the active site structure, for example by inducing structural changes in an interfering residue located in the active site, 1263, and have been shown in earlier work by the inventors to alter the activity of CYP102A1 (13).

[0222] It is particularly interesting to note that variant LO25 contains the L215I mutation, which may affect the contact between the F and G helices at the close approach to H171 / L215. Overall, while there are no mutations in the active site, all of them are at residues that play a role in the packing / interactions between secondary structure elements. The tandem proline arrangement introduced by A330P is unique and shows a very unexpected, but highly beneficial effect.

[0223] The mutations disclosed in the present invention can be introduced into existing variants (eg, those comprising L188Q, R47L, Y51F mutations) for method development, and also serve as a starting point for further evolution.

[0224] References

[0225] 1. Miura, Y., and Fulco, AJ (1975) Biochim. Biophys. Acta 388, 305-317.

[0226] 2. Cryle, MJ, Espinoza, RD, Smith, SJ, Matovic, NJ, and De Voss, JJ (2006) Chem Commun, 2353-2355.

[0227] 3.Ravichandran,K.G.,Boddupalli,S.S.,Hasemann,C.A.,Peterson,J.A.,andDeisenhofer,J.(1993)Science 261,731-736.

[0228] 4.Li,H.,and Poulos,T.L.(1997)Nature Struct.Biol.4,140-146.

[0229] 5.Munro,A.W.,Leys,D.G.,McLean,K.J.,Marshall,K.R.,Ost,T.W.,Daff,S.,Miles,C.S.,Chapman,S.K.,Lysek,D.A.,Moser,C.C.,Page,C.C.,andDutton,P.L.(2002)Trends Biochem.Sci.27,250-257.

[0230] 6.Urlacher,V.B.,Lutz-Wahl,S.,and Schmid,R.D.(2004)Appl.Microbiol.Biotechnol.64,317-325.

[0231] 7.Bell,S.G.,Hoskins,N.,Whitehouse,C.J.C.,and Wong,L.-L.(2007)MetalIons Life Sci.3,437-476.

[0232] 8.Graham-Lorence,S.,Truan,G.,Peterson,J.A.,Falck,J.R.,Wei,S.,Helvig,C.,and Capdevila,J.H.(1997)J.Biol.Chem.272,1127-1135.

[0233] 9.Oliver,C.F.,Modi,S.,Sutcliffe,M.J.,Primrose,W.U.,Lian,L.Y.,andRoberts,G.C.K.(1997)Biochemistry 36,1567-1572.

[0234] 10.Noble,M.A.,Miles,C.S.,Chapman,S.K.,Lysek,D.A.,Mackay,A.C.,Reid,G.A.,Hanzlik,R.P.,and Munro,A.W.(1999)Biochem.J.339,371-379.

[0235] 11.Cowart,L.A.,Falck,J.R.,and Capdevila,J.H.(2001)Arch.Biochem.Biophys.387,117-124.

[0236] 12.Oliver,C.F.,Modi,S.,Primrose,W.U.,Lian,L.Y.,and Roberts,G.C.K.(1997)Biochem.J.327,537-544.

[0237] 13.Carmichael,A.B.,and Wong,L.L.(2001)Eur.J.Biochem.268,3117-3125.

[0238] 14.Sowden,R.J.,Yasmin,S.,Rees,N.H.,Bell,S.G.,and Wong,L.L.(2005)Org.Biomol.Chem.3,57-64.

[0239] 15.Lussenburg,B.M.,Babel,L.C.,Vermeulen,N.P.,and Commandeur,J.N.(2005)Anal Biochem 341,148-155.

[0240] 16.Wong,T.S.,Wu,N.,Roccatano,D.,Zacharias,M.,and Schwaneberg,U.(2005)J Biomol Screen 10,246-252.

[0241] 17.Urlacher,V.B.,Makhsumkhanov,A.,and Schmid,R.D.(2006)Appl.Microbiol.Biotechnol.70,53-59.

[0242] 18.van Vugt-Lussenburg,BM,Damsten,MC,Maasdijk,DM,Vermeulen,NP,and Commandeur,JN(2006)Biochem.Biophys.Res.Commun.346,810-818.

[0243] 19.van Vugt-Lussenburg,BM,Stjernschantz,E.,Lastdrager,J.,Oostenbrink,C.,Vermeulen,NP,and Commandeur,JN(2007)J.Med.Chem.50,455-461.

[0244] 20.Maves,SA,Yeom,H.,McLean,MA,and Sligar,SG(1997)FEBS Lett.414,213-218.

[0245] 21.Li,QS,Schwaneberg,U.,Fischer,P.,and Schmid,RD(2000)Chem.Eur.J.6,1531-1536.

[0246] 22.Appel,D.,Lutz-Wahl,S.,Fischer,P.,Schwaneberg,U.,and Schmid,RD(2001)J.Biotechnol.88,167-171.

[0247] 23.Li,QS,Ogawa,J.,Schmid,RD,and Shimizu,S.(2001)Appl.Environ.Microbiol.67,5735-5739.

[0248] 24.Schulze,H.,Schmid,RD,and Bachmann,TT(2004)Anal.Chem.76,1720-1725.

[0249] 25.Sulistyaningdyah,WT,Ogawa,J.,Li,QS,Maeda,J.,Yano,Y.,Schmid,RD,and Shimizu,S.(2005)Appl.Microbiol.Biotechnol.67,556-562.

[0250] 26.Schwaneberg,U.,Schmidt-Dannert,C.,Schmitt,J.,and Schmid,R.D.(1999)Anal.Biochem.269,359-366.

[0251] 27.Li,Q.S.,Schwaneberg,U.,Fischer,M.,Schmitt,J.,Pleiss,J.,Lutz-Wahl,S.,and Schmid,R.D.(2001)Biochim.Biophys.Acta 1545,114-121.

[0252] 28.Farinas,E.T.,Bulter,T.,and Arnold,F.H.(2001)Curr.Opin.Biotechnol.12,545-551.

[0253] 29.Glieder,A.,Farinas,E.T.,and Arnold,F.H.(2002)Nat.Biotechnol.20,1135-1139.

[0254] 30.Meinhold,P.,Peters,M.W.,Chen,M.M.,Takahashi,K.,and Arnold,F.H.(2005)ChemBioChem 6,1-4.

[0255] 31.Peters,M.W.,Meinhold,P.,Glieder,A.,and Arnold,F.H.(2003)J.Am.Chem.Soc.125,13442-13450.

[0256] 32.Kubo,T.,Peters,M.W.,Meinhold,P.,and Arnold,F.H.(2006)Chemistry 12,1216-1220.

[0257] 33.Munzer,D.F.,Meinhold,P.,Peters,M.W.,Feichtenhofer,S.,Griengl,H.,Arnold,F.H.,Glieder,A.,and de Raadt,A.(2005)Chem.Commun.,2597-2599.

[0258] 34.Lentz,O.,Feenstra,A.,Habicher,T.,Hauer,B.,Schmid,R.D.,and B.,U.V.(2005)ChemBioChem 7,345-350.

[0259] 35.Ortiz de Montellano,P.R.(1986)Cytochrome P450:Structure,Mechanism,and Biochemistry,Plenum Press,New York.

[0260] 36.Ortiz de Montellano,P.R.(ed)(1995)Cytochrome P450:Structure,Mechanism,and Biochemistry,2 nd Ed.,Plenum Press,New York.

[0261] 37.Ortiz de Montellano,P.R.(ed)(2005)Cytochrome P450:Structure,Mechanism,and Biochemistry 3rd Ed.,Kluwer Academic / Plenum Press,NewYork.

[0262] 38.Poulos,T.L.,Finzel,B.C.,and Howard,A.J.(1987)J.Mol.Biol.195,687-700.

[0263] 39.Poulos,T.L.(2005)Biochem.Biophys.Res.Commun.338,337-345.

[0264] 40.Poulos,T.L.(2005)Drug Metab Dispos 33,10-18.

[0265] 41.Poulos,T.L.(2003)Proc.Natl.Acad.Sci.USA 100,13121-13122.

[0266] 42.Poulos,T.L.(2003)Biochem.Biophys.Res.Commun.312,35-39.

[0267] 43.Hasemann,CA,Kurumbail,RG,Boddupalli,SS,Peterson,JA,andDeisenhofer,J.(1995)Structure 2,41-62.

[0268] 44.Rupasinghe,S.,Schuler,MA,Kagawa,N,Yuan,H,Lei,L,Zhao,B,Kelly,SL,Waterman,MR,and Lamb,DC(2006)FEBS Lett.580,6338-6342.

[0269] Biochem.Biophys.Acta 1770,467-747, 45. Lepesheva, GI, and Waterman, MR (2007).

[0270] 46. ​​Li, QS, Ogawa, J., Schmid, RD, and Shimizu, S. (2001)FEBS Lett.508,249-252.

[0271] 47.Capdevila,JH,Wei,S,Helvig,C,Falck,JR,Belosludtsev,Y,Truan,G,Graham-Lorence,SE,and Peterson,JA(1996)J.Biol.Chem.271,22663-22671.

[0272] 48.Skiles,GL,and Yost,GS(1996)Chem.Res.Toxicol.9,291-297.

[0273] 49.Lanza,DL,and Yost,GS(2001)Drug Metab Dispos 29,950-953.

[0274] 50.Reilly,CA,Ehlhardt,WJ,Jackson,DA,Kulanthaivel,P,Mutlib,AE,Espina,RJ,Moody,DE,Crouch,DJ,and Yost,GS(2003)Chem.Res.Toxicol.16,336-349.

[0275] 51.Kassahun,K.,Skordos,K.,McIntosh,I.,Slaughter,D.,Doss,GA,Baillie,TA,and Yost,GS(2005)Chem.Res.Toxicol.18,1427-1437.

[0276] 52.Sun,H.,Ehlhardt,WJ,Kulanthaivel,P.,Lanza,DL,Reilly,CA,andYost,GS(2007)J Pharmacol Exp Ther 322,843-851.

[0277] 53.Di Nardo,G.,Fantuzzi,A.,Sideri,A.,Panicco,P.,Sassone,C.,Giunta,C.,andGilardi,G.(2007)J Biol Inorg Chem 12,313-323.

[0278] 54.Lentz,O.,Feenstra,A.,Habicher,T.,Hauer,B.,Schmid,RD,andUrlacher,WB(2006)ChemBioChem 7,345-350.

[0279] 55.Cirino,PC,and Arnold,FH(2003)Angew.Chem.Int.Ed.42,3299-3301.

[0280] 56.Sevriukova,IF,Hazzard,JT,Tollin,G.,and Poulos,TL(1999)J.Biol.Chem.274,36097-36106.

[0281] 57.Paulsen,MD,and Ornstein,RL(1995)Proteins:Struct.,Funct.,Genet.21,237-243.

[0282] 58.Honeychurch,MJ,Hill,HAO,and Wong,LL(1999)FEBS Lett.451,351-353.

[0283] Table A. Sequence similarity between the CYP102A1 heme domain (amino acid residues 1-470) and various structurally characterized cytochrome P450 enzymes

[0284]

[0285] *CYP505 (P450foxy) has not been structurally characterized and only the heme domain was used for the alignment

[0286] Table B. Sequence similarity between the entire sequence of CYP102A1 and various cytochrome P450 enzymes (aligned against proteins in the Swissprot protein database) It can be noted that, despite belonging to the same subfamily, CYP102A2 and CYP102A3 are only 59% and 58% homologous to CYP102A1.

[0287]

[0288] Table C. Physical properties of amino acids

[0289]

[0290] Table D. Hydrophilicity scale

[0291]

[0292] Table 1: In vitro oxidation activity, binding efficiency and selectivity of CYP102A1 variants towards naphthalene. Rates are in nmol min -1 (nmol P450) -1 The only detectable product was 1-naphthol (1-ol).

[0293]

[0294]

[0295] Table 2: In vitro oxidation activity, selectivity and spin shift of CYP102A1 variants towards propylbenzene. Rates are in nmol min -1 (nmol P450) -1 Given. The products are 1-phenyl-1-propanol (1-alcohol), 1-phenyl-2-propanol (2-alcohol), 2-propylphenol (ortho) and 4-propylphenol (para). Small amounts of 1-phenyl-1-propanone (≤1%) and 3-phenyl-1-propanol (<0.5%, only KT5) are also formed. - Not detected. (*) Due to these and other minor products, the percentages do not sum to 100.

[0296]

[0297]

[0298] Table 3: In vitro oxidation activity and selectivity of CYP102A1 variants towards toluene. Rates are in nmol min -1 (nmol P450) -1 Given. The products are benzyl alcohol (1-alcohol), o-cresol (ortho position) and p-cresol (para position). — Not detected.

[0299]

[0300] Table 4: In vitro oxidation activity and selectivity of CYP102A1 variants towards butylbenzene. Rates are in nmol min -1 (nmol P450) -1 The products are 1-phenyl-1-butanol (1-ol), 1-phenyl-2-butanol (2-ol), 4-phenyl-2-butanol (3-ol), 2-butylphenol (ortho position) and 4-butylphenol (para position). (*) Includes 1% to 3% of 1-phenyl-1-butanone. — Not detected.

[0301]

[0302]

[0303] Table 5a: In vitro oxidation activity, selectivity and spin mobility of CYP102A1 variants for tert-butylbenzene. Rates are in nmol min -1 (nmol P450) -1 The products are 2-tert-butylphenol (ortho), 4-tert-butylphenol (para) and 2-methyl-2-phenyl-propan-1-ol (1-ol). — Not detected.

[0304]

[0305] Table 5b: In vitro oxidation activity / selectivity and spin mobility of CYP102A1 variants for ethylbenzene. Rates are in nmol min -1 (nmol P450) -1 Given. The products are 1-phenylethanol (1-alcohol), 2-phenylethanol (2-alcohol), 2-ethylphenol (ortho) and styrene. - Not detected.

[0306]

[0307] NB / results were also obtained for F87A, F87A / KT2, KSK19 and GVQ after detector response recalibration. NADPH conversion was identical. Binding efficiencies were 23 (F87A), 21 (F87A / KT2), ​​25 (KSK19), 26 (GVQ). Product formation rates were 32 (F87A), 120 (F87A / KT2), ​​178 (KSK19), 572 (GVQ). 1-ol was 99% (F87A), 96% (F87A / KT2), ​​95% (KSK19), 99% (GVQ). Styrene was now detected in the reactions for these variants at 2% (F87A), 4% (F87A / KT2), ​​5% (KSK19), 1% (GVQ).

[0308] Table 6a: In vitro oxidation activity and selectivity of CYP102 variants towards o-xylene. Rates are in nmol min -1 (nmol P450) -1 The products are 2-methylbenzyl alcohol (l-alcohol), 2,3-dimethylphenol (2,3-phenol), 3,4-dimethylphenol (3,4-phenol), 2,3-dimethyl-p-benzoquinone (hydroquinone) and pyrocatechol (catechol). (*) Due to a few products, the percentages do not sum to 100. — Not detected.

[0309]

[0310]

[0311] Table 6b: In vitro oxidation activity and selectivity of CYP102A1 variants for m-xylene. Rates are in nmol min -1 (nmol P450) -1 The products are 3-methylbenzyl alcohol (1-alcohol), 2,4-dimethylphenol (2,4-phenol) and 2,6-dimethylphenol (2,6-phenol). (*) Due to a small number of products, the percentages do not sum to 100.

[0312]

[0313] Table 7: In vitro oxidation activity / selectivity and spin mobility of CYP102A1 variants towards p-cymene. Rates are in nmol min -1 (nmol P450) -1Given. The products are p-α-dimethylstyrene (p-α-DMS), p-α-α-trimethylbenzyl alcohol (i-Pr-alcohol), 4-isopropylbenzyl alcohol (Me-alcohol), thymol, carvacrol and unidentified products. (*) Contains up to 3% of p-α-dimethylstyrene oxide. (**) Due to other minor products, the percentages do not sum to 100. — Not detected.

[0314]

[0315]

[0316] Table 8: In vitro oxidation activity and selectivity of some CYP102A1 variants towards cumene. Rates are in nmol min -1 (nmol P450) -1 The products are α-methylstyrene (styrene), α-methylstyrene oxide (styrene oxide), 2-phenyl-1-propanol (1-ol), 2-phenyl-2-propanol (2-ol), 2-isopropylphenol (ortho) and 4-isopropylphenol (para). — Not detected.

[0317]

[0318] Table 8a: In vitro oxidation activity and selectivity of some CYP102A1 variants towards cumene after further GC analysis. Rates are in nmol min -1 (nmol P450) -1 The products are α-methylstyrene (styrene), α-methylstyrene oxide (styrene oxide), 2-phenyl-1-propanol (1-ol), 2-phenyl-2-propanol (2-ol), 2-isopropylphenol (ortho) and 4-isopropylphenol (para). — Not detected.

[0319]

[0320] Table 9: In vitro oxidation activity, selectivity and spin mobility of CYP102A1 variants towards pentane. Rates are in nmol min -1 (nmol P450) -1 The products are 2-pentanol (2-ol), 3-pentanol (3-ol), 2-pentanone (2-ketone) and 3-pentanone (3-ketone).

[0321]

[0322] Table 10: In vitro oxidation activity / selectivity and spin mobility of CYP102A1 variants towards 3-methylpentane. Rates are in nmol min -1(nmol P450) -1 The products are 3-methyl-2-pentanol (2-ol) - two diastereomers labeled (A) and (B), and 3-methyl-3-pentanol (3-ol).

[0323]

[0324]

[0325] Table 11: In vitro oxidation activity and selectivity of CYP102A1 variants towards 2-methylbutane. Rates are in nmol min -1 (nmol P450) -1 The products are 2-methyl-1-butanol (1-ol), 2-methyl-2-butanol (2-ol), 3-methyl-2-butanol (3-ol), 3-methyl-1-butanol (4-ol) and 3-methyl-2-butanone (3-ketone). — Not detected.

[0326]

[0327] Table 12: In vitro oxidation activity and selectivity of some CYP102A1 variants towards butane and propane. Rates are in nmol min -1 (nmol P450) -1 The products are 1-butanol (1-alcohol), 2-butanol (2-alcohol) and 2-butanone (2-ketone) for butane, and only 2-propanol for propane.

[0328]

[0329]

[0330] Table 13: In vitro oxidation activity and selectivity of CYP102A1 variants towards octane. Rates are in nmol min -1 (nmol P450) -1 The products are 2-octanol (2-ol), 3-octanol (3-ol), 4-octanol (4-ol), 3-octanone (3-ketone) and 4-octanone (4-ketone). - Not detected.

[0331]

[0332] Table 14: In vitro oxidation activity and selectivity of some CYP102A1 variants towards 1,4-dichlorobenzene. Rates are in nmol min -1 (nmol P450) -1The products are 2,4-dichlorophenol and / or 2,5-dichlorophenol, which have the same GC elution number. Further oxidation to GC-undetectable semiquinone makes the coupling numbers unreliable.

[0333]

[0334] Table 15: In vitro oxidation activity and selectivity of CYP102A1 variants towards valencene. Rates are in nmol min -1 (nmol P450) -1 The products are cis- and trans-coredil alcohol (coredil alcohol), coredil ketone, cis- and trans-valencene epoxide (Val.epox.) and cis- and trans-coredil ketone epoxide (Noot.epox.).

[0335]

[0336] Table 16: In vitro oxidation activity and selectivity of some CYP102A1 variants towards R- and S-1,8-terpenes. Rates are in nmol min -1 (nmol P450) -1 The products are 1,2-1,8-terpenoid epoxide (epoxide), cis- and trans-isopiperidinol (isopiper), and cis- and trans-carveol (carveol).

[0337]

[0338] Table 16a: In vitro oxidation activity and selectivity of CYP102A1 variants towards R-1,8-terpenes. N = NADPH conversion rate. C = coupling. PFR = product formation rate. Rates are in nmol min-1 (nmol P450)-1. Products are cis-1,2-1,8-terpenes epoxide (cis-1,2), trans-1,2-1,8-terpenes epoxide (trans-1,2), unidentified product (U), cis-isopiperidinol (cis-3-ol), trans-isopiperidinol (trans-3-ol), carveol (2 isomers) (6-ol (A) and (B)), carvone (6-ketone) and perillyl alcohol (10-ol).

[0339]

[0340] Table 17: In vitro oxidation activity and selectivity of some CYP102A1 variants towards (+)-α-pinene. Rates are given in nmol min-1 (nmol P450)-1. Products are 2,3-pinene epoxide (epoxide), cis- and trans-verbenol (verbenol), verbenone and myrtenol. Data exaggerate coupling due to impurities in the substrate.

[0341]

[0342] Table 17a: In vitro oxidation activity and selectivity of some CYP102A1 variants towards (+)-α-pinene. Rates are given in nmol min-1 (nmol P450)-1. N = NADPH conversion rate. C = coupling. PFR = product formation rate. Rates are given in nmol min-1 (nmol P450)-1. Products are (-)-2,3-pinene epoxide ((-)-2,3), (+)-2,3-pinene epoxide ((+)-2,3), cis-verbenol (cis-4-ol), trans-verbenol (trans-4-ol), verbenone (4-one) and myrtenol (10-ol).

[0343]

[0344]

[0345] Table 18: In vitro oxidation activity and selectivity of CYP102A1 variants towards fluorene. N = NADPH conversion rate. C = coupling. PFR = product formation rate. All rates are given in nmol min-1 (nmol P450)-1. The products are 9-fluorenol (9-ol) and 2-fluorenone (9-ketone).

[0346]

[0347] Table 19: In vitro oxidation activity and selectivity of CYP102A1 variants for β-ionone. N = NADPH conversion rate. C = coupling. PFR = product formation rate. Rates are given in nmol·min-1·(nmol P450)-1. The product is 4-hydroxy-β-ionone (4-ol).

[0348]

[0349] Table 20: In vitro oxidation activity and selectivity of CYP102A1 variants for lauric acid. N = NADPH conversion rate. C = coupling. PFR = product formation rate. Rates are given in nmol min-1 (nmol P450)-1. The products are 11-hydroxydodecanoic acid (ω-1), 10-hydroxydodecanoic acid (ω-2), 9-hydroxydodecanoic acid (ω-3), 8-hydroxydodecanoic acid (ω-4) and 7-hydroxydodecanoic acid (ω-5).

[0350]

[0351] Sequence of the present invention (CYP102A1 wild type)

[0352]

[0353]

[0354]

[0355]

[0356]

[0357]

Claims

1. A method for oxidizing a substrate, the method comprising contacting the substrate with a mutant CYP102A1 (cytochrome P450 family 102A subfamily member 1) enzyme, wherein the mutant CYP102A1 enzyme has a substitution I401P at position 401 in the polypeptide chain of the wild-type CYP102A1 enzyme shown in SEQ ID NO:2, thereby enhancing the monooxygenase activity of the mutant enzyme and / or changing the product selectivity of the mutant enzyme, wherein the substrate is naphthalene, propylbenzene, toluene, 3-methylpentane, 2-methylbutane, propane, octane, R-1,8-terpenoid, (+)-α-pinene, fluorene, β-ionone, or lauric acid.

2. A method for oxidizing a substrate, the method comprising contacting the substrate with a mutant CYP102A1 (cytochrome P450 family 102A subfamily member 1) enzyme, wherein the mutant CYP102A1 enzyme has a substitution Q403P at position 403 in the polypeptide chain of the wild-type CYP102A1 enzyme shown in SEQ ID NO: 2, thereby enhancing the monooxygenase activity of the mutant enzyme and / or changing the product selectivity of the mutant enzyme, wherein the substrate is cumene, valencene, R-1,8-terpenoid or fluorene.

3. The method according to claim 1 or 2, wherein: The substrate is oxidized in cells expressing the mutant CYP102A1 enzyme.

4. A mutant CYP102A1 (cytochrome P450 family 102A subfamily member 1) enzyme, wherein the mutant CYP102A1 enzyme has a substitution I401P at position 401 in the polypeptide chain of the wild-type CYP102A1 enzyme shown in SEQ ID NO: 2, thereby enhancing the monooxygenase activity of the CYP102A1 enzyme.

5. A mutant CYP102A1 (cytochrome P450 family 102A subfamily member 1) enzyme, wherein the mutant CYP102A1 enzyme has a substitution Q403P at position 403 in the polypeptide chain of the wild-type CYP102A1 enzyme shown in SEQ ID NO: 2, thereby enhancing the monooxygenase activity of the CYP102A1 enzyme.

6. A polynucleotide comprising a sequence encoding a mutant CYP102A1 enzyme as defined in claim 4 or 5, optionally in the form of a vector.

7. A cell expressing the mutant CYP102A1 enzyme as defined in claim 4, said cell being not an animal or plant species.

8. A cell expressing the mutant CYP102A1 enzyme as defined in claim 5, said cell being not an animal or plant species.

9. The cell according to claim 7 or 8, which is a prokaryotic cell or a eukaryotic cell.

10. The cell according to claim 9, which is a strain of Escherichia coli, Pseudomonas, Pichia, Rhodococcus, or Bacillus.

11. The method according to claim 1, wherein: The substrate is oxidized in the cell according to claim 10.

12. A method for treating a site contaminated with a substrate, wherein the substrate is naphthalene, propylbenzene, toluene, 3-methylpentane, 2-methylbutane, propane, octane, R-1,8-terpene, (+)-α-pinene, fluorene, β-ionone, or lauric acid, the method comprising contacting the site with the mutant CYP102A1 enzyme according to claim 4 or the cell defined in claim 7.

13. A method for treating a site contaminated with a substrate, the substrate being cumene, valencene, R-1,8-terpene or fluorene, the method comprising contacting the site with the mutant CYP102A1 enzyme according to claim 5 or the cell defined in claim 8.

Images