Polynucleotide, host cell and a method to recombinantly produce the protein encoded by said polynucleotide having peroxygenative activity

Abstract

The invention relates to an unspecific peroxygenase of the Agrocybe aegerita fungus, obtained by means of directed molecular evolution to facilitate the functional expression thereof in an active, soluble and stable form. The peroxygenase described in the invention shows a significant improvement in the functional expression thereof, improved monooxygenase activity and reduced peroxidase activity, in relation to the monooxygenase and peroxidase activities showed by the unspecific wild-type peroxygenase of A. aegerita. The peroxygenase of the invention is useful in chemical processes, including industrial transformations such as the selective oxyfunctionalisation of carbon-hydrogen bonds of various organic compounds.

Claims

1. A polynucleotide that encodes a polypeptide with peroxygenase activity, wherein the polypeptide comprises an amino acid sequence at least 95% identical to SEQ ID NO: 2 (AaeUP01), and comprises at least two amino acid substitutions at positions corresponding to positions 241 and 257 of the polypeptide of SEQ ID NO: 2, wherein the amino acid at position corresponding to position 241 of the polypeptide of SEQ ID NO: 2 is replaced with aspartic acid and the amino acid at position 257 of the polypeptide of SEQ ID NO: 2 is replaced with lysine.

2. The polynucleotide of claim 1, wherein the polypeptide with peroxygenase activity further comprises an amino acid substitution at the position corresponding to position 191 of the polypeptide of SEQ ID NO: 2, wherein the amino acid at the position corresponding to position to position 191 of the polypeptide of SEQ ID NO: 2 is replaced with serine.

3. The polynucleotide of claim 1, wherein the polypeptide with peroxygenase activity further comprises one or more substitutions selected from the group consisting of: a) the amino acid at the position corresponding to position 67 of the polypeptide of SEQ ID NO: 2 is replaced with phenylalanine, b) the amino acid at the position corresponding to position 248 of the polypeptide of SEQ ID NO: 2 is replaced with valine, c) the amino acid at the position corresponding to position 311 of the polypeptide of SEQ ID NO: 2 is replaced with leucine, d) the amino acid at the position corresponding to position 75 of the polypeptide of SEQ ID NO: 2 is replaced with isoleucine, and e) the amino acid at the position corresponding to position 57 of the polypeptide of SEQ ID NO: 2 is replaced with alanine.

4. The polynucleotide of claim 1, further comprising a nucleotide sequence encoding the signal peptide of SEQ ID NO: 26.

5. The polynucleotide of claim 1, further comprising a nucleotide sequence encoding a variant of the signal peptide of SEQ ID NO: 26, wherein said variant comprises one or more substitutions selected from the group consisting of: a) the replacement of the amino acid at the position corresponding to position 12 of the signal peptide of SEQ ID NO: 26 with tyrosine, b) the replacement of the amino acid at the position corresponding to position 14 of the signal peptide of SEQ ID NO: 26 with valine, c) the replacement of the amino acid at the position corresponding to position 15 of the signal peptide of SEQ ID NO: 26 with glycine, and d) the replacement of the amino acid at the position corresponding to position 21 of the signal peptide of SEQ ID NO: 26 with aspartic acid.

6. The polynucleotide of claim 1, wherein said polynucleotide comprises a nucleotide sequence selected from the group consisting of SEQ ID NO: 11, SEQ ID NO: 9, SEQ ID NO: 7, SEQ ID NO: 23, SEQ ID NO: 21, SEQ ID NO: 19, SEQ ID NO: 41, SEQ ID NO: 39 and SEQ ID NO: 37.

7. A method for obtaining a polypeptide with peroxygenase activity comprising the steps of: i. introducing a vector with a polynucleotide that encodes a polypeptide with peroxygenase activity, wherein the polypeptide comprises an amino acid sequence at least 95% identical to SEQ ID NO: 2 (AaeUP01), and comprises at least two amino acid substitutions at positions corresponding to positions 241 and 257 of the polypeptide of SEQ ID NO: 2, wherein the amino acid at position corresponding to position 241 of the polypeptide of SEQ ID NO: 2 is replaced with aspartic acid and the amino acid at position 257 of the polypeptide of SEQ ID NO: 2 is replaced with lysine, in a suitable host cell, ii. culturing the host cell in a suitable medium, and iii. purifying the synthesized polypeptide.

8. A host cell comprising the polynucleotide according to claim 1.

9. The host cell; according to claim 8, wherein the host cell is a yeast or fungus cell.

10. The host cell, according to claim 8, wherein the host cell is a yeast cell that belongs to the genus Saccharomyces sp or Pichia sp, or the host cell is a fungus cell that belongs to the genus Aspergillus sp.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 Directed evolution of AaeUPO1. From cycles 1 to 5, the enzyme was improved in terms of functional expression and activity (the accumulated mutations are detailed as light grey rectangles). Starting from the parental AaeUPO, it was subjected to five directed evolution cycles until obtaining the PaDa-I mutant, which was subjected to two more cycles of directed evolution, in this case to improve the production capacity of 1-naphthol (the new mutations appear as black rectangles), and three further cycle grouped together in a single generation to improve the production of 5-hydroxypropranolol. The activities (as a %) stem from measurements using microcultures of S. cerevisiae in 96-well microplates of the second re-screening. Thermostability (T.sub.50) was determined using flask culture supernatants: n.m. not measurable, n.d. not determined.

(2) FIG. 2 Biochemical characteristics of the variants of the invention. A) Spectroscopic characteristics of the PaDa-I (thin line) and JaWa (thick line) mutants at rest. AU, arbitrary units. B) Thermostability analysis (T.sub.50) of the PaDa-I (black circles) and JaWa (white circles) mutants. The experiments were carried out using culture supernatants and each point represents the average value and standard deviation of three individual experiments. C) Stability of the PaDa-I (black bars) and JaWa (grey bars) mutants at high acetonitrile concentrations. The stabilities were determined after 5 hours of incubation of the enzyme in increasing concentrations of the co-solvent (from 50% to 100%) at 20 C. in 10 mM pH 7.0 potassium phosphate buffer. After that time, aliquots were taken and analysed using ABTS substrate (100 mM pH 4.0 sodium phosphate/citrate buffer, 2 mM H.sub.2O.sub.2 and 0.3 mM ABTS). The error bars indicate standard deviations.

(3) FIG. 3 Transformation of naphthalene by means of the variants described in the invention. A) Products formed after 15 minutes of reaction stopped with 20 L of HCl 37% (PaDa-I, black bars; JaWa, grey bars). The reactions were carried out at room temperature using 6.6 nM of pure enzyme, 100 mM pH 7.0 of potassium phosphate buffer, 1 mM naphthalene, 20% acetonitrile and 1 mM H.sub.2O.sub.2 (1 mL of final volume). As can be observed in the figure, the products obtained were mainly naphthalene, 1-naphthol and 2-naphthol. B) Chromatograms of the naphthalene transformation reaction after 270 minutes (1: naphthalene; 2: 1-naphthol; 3: 2-naphthol and 4: 1.4-naphthoquinone (1.4-NQ)). C) and D) Monitoring of the reaction for 270 minutes (without adding HCl) for the PaDa-I (C) and JaWa (D) mutants. Black circles: naphthalene; white circles: 1.2-naphthalene oxide; white squares: 1-naphthol and black squares: 2-naphthol. Total turnover numbers (TTN, expressed as moles of product/moles of enzyme) were calculated using the production value of 1-naphthol after 270 minutes.

(4) FIG. 4 Conversion of naphthalene at 1-naphthol by means of the PaDa-I and JaWa variants. The reactions were performed at room temperature and their composition was as follows: 40 nM of pure enzyme, 100 mM pH 7.0 potassium phosphate buffer, 1 mM naphthalene, 20% acetonitrile and 1 mM H.sub.2O.sub.2 (1 mL of final volume). 1-N: 1-naphthol; 1,4-NQ: 1-4-naphthoquinone. Each reaction was performed in triplicate and were stopped with HCl (pH<1) at different times (between 60 and 600 s). Inset: polymeric colorimetric products derived from 1.4-naphthoquinone, 1: PaDa-I and 2: JaWa.

(5) FIG. 5 W24F variants obtained by means of directed mutagenesis. A) Model built on the crystal structure of the AaeUPO1 enzyme (PDB access number: 2YOR), comprising the mutations of the JaWa variant as well as the W24F modification with respect to wild AaeUPO1. The model is shown without a surface, with a transparent surface and with an opaque surface, showing position W24. B) Activity of the W24F variants using different substrates with respect to their respective parentals, relativised to the PaDa-I activity. The experiments were carried out using 100 mL flask culture supernatants. The buffer used was 100 mM pH 7.0 potassium phosphate buffer, except for the ABTS, in which case 100 mM pH 4.0 sodium phosphate/citrate was used. The components of the mixture were: 0.5 mM naphthalene, 1 mM NBD, 3 mM DMP and 0.3 mM ABTS. In all cases, 1 mM H.sub.2O.sub.2 and 15% acetonitrile were added to the mixtures. For the activity with naphthalene, the Fast Red method was applied (after 10 minutes of reaction, Fast Red was addedfinal concentration 0.5 mMand when the red colour appeared and became stabilised, final absorbance was measured). The molar extinction coefficients are: naphthalene+Fast Red, .sub.510=4,700 M.sup.1 cm.sup.1; NBD, .sub.425=9,700 M.sup.1 cm.sup.1; DMP, .sub.469=27,500 M.sup.1 cm.sup.1 and ABTS, .sub.418=36,000 M.sup.1 cm.sup.1.

(6) FIG. 6 Mutations in the UPO variants described in the invention. Model built on the structure of the AaeUPO1 crystal (PDB access number: 2YOR). A) PaDa-I; B) JaWa. The V248 mutant stems from the previous evolution pathway. The phenylalanine (Phe) residues are responsible for the accommodation of the substrates in the catalytic pocket, the Cys36 residue is the axial heme ligand; R189 is a component of the acid-base pair involved in the catalysis, and heme Fe.sup.3+ is represented as a sphere.

(7) FIG. 7 Protein model of A) PaDa-I and B) JaWa. The protein model for PaDa-I (A) was built on the structure of the AaeUPO1 crystal (PDB access number: 2YOR) and the software PyMOL Molecular Graphics System, Version 1.3 Schrdinger, LLC. The new mutations of the PaDa-I mutant with respect to the native UPO are shown underlined, while the residues with a zig-zag underline are those which have been changed in JaWa (B). The image shows the five Phe that participate in the accommodation of the substrate: Phe 69, Phe 76, Phe 121, Phe 191 and Phe 199; the two catalytic residues are R189 and E196.

(8) FIG. 8 B factors for the evolved UPOs of the present invention. Representation of the B factors (obtained using PyMOL Molecular Graphics System, Version 1.3 Schrdinger, LLC.) of the PaDa-I variant (left) and the JaWa variant (right). Said B factors make reference to the rigidity/flexibility of a protein region or of an amino acid. A) Detail of the mutation in position 257, located on the surface: darker shades indicated greater rigidity. B) Representation in putty mode of the complete structure of the PaDa-I and JaWa variants. The greater the thickness of the lines, the greater the flexibility.

(9) FIG. 9 Assay of 4-AAP (4-aminoantipyrine) with different pure UPO variants (AaeUPO1, PaDa-1 and JaWa). The reactions were performed at room temperature and their composition was as follows: 0.2 M of each pure UPO variant, 50 mM pH 7.0 potassium phosphate buffer, 5 mM propranolol, 2 mM H.sub.2O.sub.2 (0.05 mL of final volume) and, in the case of reactions with ascorbic acid, it was added to a concentration of 4 mM. Each reaction was performed in triplicate.

(10) FIG. 10 Molecular docking with JaWa and propranolol. Amino acids that interact with propranolol are indicated, with the distances therefrom. The zone selected for MORPHING experiments due to its proximity to the protein-substrate contact points is indicated in dark grey.

(11) FIG. 11 Mutations in SoLo variants with respect to the JaWa variant described in the invention. Model built on the structure of the PaDa-I crystal. A) JaWa; B) SoLo.

(12) FIG. 12 Thermostability analysis (T.sub.50) of the JaWa (black circles) and SoLo (white circles) mutants. The experiments were carried out using culture supernatants and each point represents the average value and standard deviation of three individual experiments.

(13) FIG. 13 Chromatogram showing the enzyme reactions. The reactions were performed at room temperature and their composition was as follows: 0.03 M of each pure UPO variant, 50 mM pH 7.0 of potassium phosphate buffer, 4 mM propranolol, 2 mM H.sub.2O.sub.2 (0.5 mL of final volume).

(14) FIG. 14 Turnover rates of AaeUPO, JaWa and SoLo. The reaction mixture contained 0.03 M of each pure UPO variant, 0.4 mM 5-hydroxypropranolol, and 2 mM H.sub.2O.sub.2 in 50 mM pH 7.0 potassium phosphate buffer (0.3 mL of final volume). The disappearance of the product 5-hydroxypropranolol can be observed due to the formation of its corresponding quinone by means of the peroxidase activity of the enzyme.

(15) FIG. 15 Calculation of the total turnover number (TTN) of AaeUPO and SoLo. The assay was carried out using 0.03 M of each pure enzyme, 4 mM propranolol and 2 mM H.sub.2O.sub.2 in 50 mM pH 7.0 potassium phosphate buffer and in the same manner, but also with 4 mM ascorbic acid. In both cases, 2 mM H.sub.2O.sub.2 was added every 10 minutes, monitoring the reaction in each addition point taking different aliquots.

EXAMPLES

(16) Following are examples of the invention by means of assays carried out by the inventors, which evidence the effectiveness of the product of the invention. The following examples serve to illustrate the invention and must not be considered to limit the scope thereof.

Example 1. Obtainment and Characterisation of the Variants of the Present Invention

(17) Materials and Methods

(18) Reagents and Enzymes

(19) ABTS (2,2-azino-bis(3-ethylbenzothiazolin-6-sulfonic) acid), DMP (2,6-dimetoxiphenol), benzyl alcohol, 1-naphthol, 2-naphthol, 1,4-naphthoquinone, Fast Red (Fast Red TR Salt hemi(zinc chloride) salt), Taq DNA polymerase and the Saccharomyces cerevisiae transformation kit were obtained from Sigma-Aldrich (Saint Louis, Mo., USO). NBD (5-nitro-1,3-benzodioxole) was acquired from TCI America (Portland, Oreg., USA), while the naphthalene is from Acros Organics (Geel, Belgium).

(20) The cDNA of upo1 (C1A-2 clone) of A. aegerita was provided by Dr. Martin Hofrichter (M. J. Pecyna, et al. Appl. Microbiol. Biotechnol. 2009, 84, 885-897).

(21) The competent Escherichia coli XL2-Blue cells and the Genemorph II Random Mutagenesis (Mutazyme II) kit were obtained from Agilent Technologies (Santa Clara, Calif., USA) and the iProof high-fidelity DNA polymerase was acquired from Bio-Rad (Hercules, Calif., USA). The BamHI and XhoI restriction enzymes were obtained from New England Biolabs (Ipswich, Mass., USA) and the protease-deficient strain of S. cerevisiae BJ5465 from LGCPromochem (Barcelona, Spain). The Zymoprep Yeast Plasmid Miniprep and Zymoclean Gel DNA Recovery kits are marketed by Zymo Research (Orange, Calif., USA). The NucleoSpin Plasmid kit is from Macherey-Nagel (Dren, Germany) and the oligonucleotides used were synthesised by Isogen Life Science (Barcelona, Spain). All the chemical compounds are of the highest purity available in the market.

(22) Directed Evolution

(23) The PaDa-I mutant (SEQ ID NO: 18) comprising the mutated signal peptide of SEQ ID NO: 28, was obtained as described in P. Molina-Espeja, et al. Appl. Environ. Microbiol. 2014. 80, 3496.-3507. After each evolution cycle, the PCR products were loaded in a semi-preparatory agarose gel and were purified using the Zymoclean Gel DNA Recovery kit. The DNA fragments recovered were cloned in the pJRoC30 plasmid under the control of the GAL1 promoter linearised with BamHI and XhoI (wherewith the parental or predecessor gene is also eliminated). The linearised plasmid was loaded in a low-melting-point preparatory agarose gel and was purified using the Zymoclean Gel DNA Recovery kit.

(24) First Generation (1G)

(25) In order to obtain the variants described in the present invention, an error-prone PCR was performed in a final volume of 50 L. This reaction contained 3% dimethyl sulfoxide (DMSO), 0.37 M of RMLN (SEQ ID NO: 33 5-cctctatactttaacgtcaagg-3), 0.37 M of RMLC (SEQ ID NO: 34 5-gggagggcgtgaatgtaagc-3), 0.8 mM deoxynucleotide triphosphate (dNTPs, 0.2 mM each), 0.05 U/L of Mutazyme II (Genemorph II kit, Stratagene) and 2.822 ng of template (pJRoC30 plasmid (from the California Institute of Technology (CALTECH, USA), which comprises the nucleotide sequence of the PaDa-I mutant of SEQ ID NO:17, 300 ng of the target DNA). This mutagenic PCR was performed in a gradient thermocyclator (Mycycler, Bio-Rad, USA), determining the following parameters: 95 C. 2 min (1 cycle); 94 C. 45 s, 53 C. 45 s and 74 C. 3 min (28 cycles); and 74 C. 10 min (1 cycle). 200 ng of the PCR product were mixed with 100 g of the linearised plasmid and competent S. cerevisiae cells were transformed so as to produce in vivo DNA shuffling and cloning (using the yeast transformation kit for such purpose). The volume resulting from the transformation was plated in (solid) minimal plates (for SC drop-out plates, said (solid) minimum consists of 100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L uracil-free amino acid supplement, 100 mL of 20% glucose, 20 g of bacto agar, 700 mL of distilled water and 1 mL of 25 g/L chloramphenicol) were incubated for three days at 30 C. The individual colonies that were formed were selected and subjected to a dual colorimetric High-Throughput Screening (HTS) assay, to efficiently explore mutant libraries without altering enzyme stability thereof, in addition to various re-screenings, as described below.

(26) Second Generation (2G)

(27) Mutagenic StEP (Staggered Extension Process) was performed using the best mutants obtained in the first generation (H. Zhao, et al. Nat Biotechnol. 1998. 16, 258-261; E. Garcia-Ruiz, et al. Biochem. J. 2012. 441, 487-498) combined with in vivo shuffling. The conditions of the StEP PCR were: 3% DMSO, 90 nM RMLN (SEQ ID NO: 33 5-cctctatactttaacgtcaagg-3), 90 nM RMLC (SEQ ID NO: 34 5-gggagggcgtgaatgtaagc-3), 0.3 mM dNTPs (0.075 mM each), 0.05 U/L Taq DNA polymerase and 16 ng of the templates (pJRoC30 with the four best mutants of the first generation). The PCRs were performed in a gradient thermocyclator using the following parameters: 95 C. 5 min (1 cycle); 94 C. 30 s, 55 C. 20 s (90 cycles). 200 ng of the PCR products were mixed with 100 ng of the linearised plasmid and transformed into competent S. cerevisiae cells). The rest of the procedure was followed as explained previously to obtain the first generation. In this evolution cycle a new variant, JaWa, was obtained, wherein the two new mutations took place: G241D and R257K, with respect to any of the enzymes AaeUPO1 or PaDa-I.

(28) W24F Variants

(29) Two individual high-fidelity PCRs were performed for each PaDa-I variant (PaDa-I of SEQ ID NO: 18, encoded by SEQ ID NO: 17) and JaWa (SEQ ID NO: 24, encoded by SEQ ID NO: 23), using the nucleotide sequences that encode both as a template and thereby introducing the change required in their sequence. Starting the numbering of the upo1 gene of SEQ ID NO: 1 from the start of the mature protein of SEQ ID NO: 2, the two nucleotide changes made were G71T and G72T (change in codon: TGG-W to TTT-F). Two primers were designed for these PCRs, wherein the aforementioned changes were included. Said primers were the F24FOR primer of sequence SEQ ID NO: 35 (F24FOR: 5-ctcacccatttaagccgcttcgacctggcgatattcgtggac-3) and the F24REV primer of sequence SEQ ID NO: 36 (5-gtccacgaatatcgccaggtcgaagcggcttaaatgggtgag-3). The changes made to said primer to perform the mutagenesis appear underlined in the nucleotide sequence thereof.

(30) The conditions of these PCRs were: (i) in a final volume of 50 L, 3% DMSO, 0.5 M RMLN (SEQ ID NO: 33), 0.5 M F24REV of SEQ ID NO: 36, 1 mM dNTPs (0.25 mM each), 0.02 U/L of iProof high-fidelity DNA polymerase and 10 ng of the templates; or (ii) in a final volume of 50 L, 3% DMSO, 0.5 M F24FOR of SEQ ID NO: 35, 0.5 M RMLC of SEQ ID NO: 34, 1 mM dNTPs (0.25 of each), 0.02 U/L of iProof high-fidelity DNA polymerase and 10 ng of the templates. The following parameters were used: (i) 98 C. 30 s (1 cycle), 98 C. 10 s, 47 C. 25 s, 72 C. 15 s (28 cycles) and 72 C. 10 min (1 cycle); or (ii) 98 C. 30 s (1 cycle), 98 C. 10 s, 58 C. 25 s, 72 C. 45 s (35 cycles) and 72 C. 10 min (1 cycle). 200 ng of the two PCR products corresponding to their respective template were mixed with 100 g of the linearised plasmid and were transformed into S. cerevisiae in order to perform the in vivo assembly of the genes and cloning using the In Vivo Overlap Extension (IVOE) technique (M. Alcalde. Methods Mol. Biol. 2010. 634, 3, -14).

(31) Preparation of the Mutant Libraries

(32) Individual colonies corresponding to clones were selected and inoculated in 96 sterile wells (Greiner Bio-One GmbH, Germany), hereinafter mother plates, with 200 L/minimal medium for expression per well (100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L, 67 mL of 1M pH 6.0 potassium phosphate buffer, 111 mL of 20% galactose, 22 mL of 0.1 M MgSO.sub.4, 31.6 mL of absolute ethanol, 1 mL of 25 g/L chloramphenicol and ddH.sub.2O up to 1,000 mL). Column 6 of each column was inoculated with the corresponding parental and well H1 with untransformed S. cerevisiae. The plates were sealed to avoid evaporation and were incubated at 30 C., 220 RPM and 80% of relative humidity (in a Minitron, INFORS, Switzerland) for five days.

(33) Dual Colorimetric High-Throughput Screening (HTS)

(34) The mother plates were centrifuged (Eppendorf 5810R centrifuge, Germany) for 10 minutes at 3,500 RPM and 4 C. 20 L of supernatant were transferred from these mother plates to two replica daughter plates with the help of a Freedom EVO liquid handling robot (Tecan, Switzerland). 180 L of reaction mixture were added with 2,6-dimethoxyphenol (DMP) or naphthalene to the daughter plates using a pipetting robot (Multidrop Combi Reagent Dispenser, Thermo Scientific, USA).

(35) The DMP reaction mixture was composed of 100 mM pH 7.0 potassium phosphate buffer, 3 mM DMP and 1 mM H.sub.2O.sub.2. Simultaneously, this same screening assay was carried out but adding 10% acetonitrile to the reaction mixture in order to determine changes in the activity caused by the appearance of resistance to this organic co-solvent (present in the naphthalene screening reaction mixture, necessary so it remains dissolved). The reaction mixture with naphthalene contained 100 mM pH 7.0 potassium phosphate buffer, 0.5 mM naphthalene, 10% acetonitrile and 1 mM H.sub.2O.sub.2. The plates were briefly agitated and initial absorbance was measured at 469 nm and 510 nm, respectively, using a plate reader for such purpose (SPECTRAMax Plus 384, Molecular Devices, USA). After a reaction time of 10 minutes, 20 L of Fast Red (Fast Red TR Salt hemi(zinc chloride) salt) were added to each naphthalene screening well (so that its final concentration in each well was 0.5 mM). The plates were kept at room temperature until they turned orange (DMP) or red (naphthol-Fast Red), at which time the absorbance was newly measured. The values were normalised against the parental of each plate. In order to rule out false positives, two re-screenings were carried out, in addition to a third re-screening wherein kinetic stability was determined (T.sub.50) (P. Molina-Espeja, et al. Appl. Environ. Microbiol. 2014. 80, 3496-3507). The Fast Red compound was specifically coupled to the 1-naphthol to form an azo-type red dye that can be measured at 510 nm (.sub.510=4,700 M.sup.1 cm.sup.1), wavelength at which the interference in the measurement produced by the culture medium is minimal.

(36) First Re-Screening

(37) The best screening clones were selected (50 clones), of which 5 L aliquots were taken and transferred to sterile plates containing of 200 L minimal medium for expression per well. Columns 1 and 12 plus rows A and H were not inoculated, for the purpose of avoiding evaporation and, thus, the appearance of false positives. They were incubated for 5 days at 30 C. and 220 RPM. The parental was treated in the same manner (row D, wells 7-11). The plates were treated following the same protocol as the previously described screening.

(38) Second Re-Screening

(39) An aliquot with the 10 best clones of the first re-screening was inoculated in 3 mL of YPD culture medium (10 g of yeast extract, 20 g of peptone, 100 mL of 20% glucose, 1 mL of 25 g/L chloramphenicol and ddH.sub.2O up to 1,000 mL) at 30 C. and 220 RPM for 16 hours. The plasmids of those cultures were extracted using the Zymoprep Yeast Plasmid Miniprep kit. Due to the impurity and low concentration of the DNA extracted, the plasmids were transformed into supercompetent E. coli XL2-Blue cells and plated in LB-amp plates (Luria-Bertani medium is composed of 5 g of yeast extract, 10 g of peptone, 10 g of NaCl, 100 mg of ampicillin and ddH.sub.2O up to 1,000 mL). An individual colony was selected from each clone, inoculated in 5 mL of LB and grown for 16 hours at 37 C. and at 250 RPM. The plasmids were extracted using the NucleoSpin Plasmid kit and transformed into competent S. cerevisiae cells (as well as with the parental). Five individual colonies of each clone were selected and inoculated to undergo the same previously described screening protocol.

(40) Third Re-Screening. Thermostability Assay

(41) An individual S. cerevisiae colony was selected with the corresponding clone (grown in a SC drop-out minimal medium plate: 100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L uracil-free amino acid supplement, 100 mL of 20% glucose, 1 mL of 25 g/L chloramphenicol and ddH.sub.2O up to 1,000 mL) was inoculated in 2 mL of selective minimal medium (as in the SC plate medium, but with 20 g of bacto agar and rafinose instead of galactose) and was incubated for 48 hour at 30 C. and 220 RPM. An aliquot of this culture was taken such that, upon inoculating it in 5 mL of new minimal medium, optical density at 600 nm would have a value of 0.25 (optical density, OD.sub.600=0.25). This starter was incubated until completing two full growth cycles (between 6 and 8 hours), at which time 1 mL of cells were taken to inoculate 9 mL of expression medium in a 100 mL flask (OD.sub.600=0.1). This culture of each clone was incubated for 72 hours at 25 C. and 220 RPM (at peak UPO activity; OD.sub.600=25-30), the cells were separated by centrifugation (10 minutes at 4,500 RPM and 4 C.) and supernatant was filtered (using a glass and nitrocellulose filter with a pore size of 0.45 m). Appropriate dilutions of the supernatants were prepared so that aliquots of 20 L would give rise to a linear response in kinetic mode. 50 L of supernatant were used for each point in a temperature gradient created by means of thermocyclator, from 30 to 80 C. After incubating for 10 minutes, the aliquots were cooled in ice for 10 minute and tempered at room temperature for 5 minutes. Lastly, these supernatants were subjected to the colorimetric assay using ABTS (100 mM pH 4.0 sodium phosphate/citrate buffer, 0.3 mM ABTS and 2 mM H.sub.2O.sub.2). The thermostability values were calculated in accordance with the ratio between the residual activities incubated at different temperatures and the value of initial activity at room temperature. The value of T.sub.50 was determined as the value of the temperature at which the protein loses 50% of it initial activity after incubating for 10 minutes.

(42) Production of UPO Recombinant Variants in S. cerevisiae

(43) An independent S. cerevisiae colony that comprised the corresponding variant of the invention was selected from a SC drop-out minimal medium plate and inoculated in 20 mL of liquid SC minimal medium, cultures which were incubated at 48 h at 30 C. and 220 RPM. An aliquot of this culture was taken so that, upon inoculating it in 100 mL of new minimal medium, OD.sub.600 would have a value of 0.25. This starter was incubated until completing two full growth cycles (between 6 and 8 hours), at which time 100 mL of cells were taken to inoculate 900 mL of minimal medium for expression in a 2,000 mL flask (OD.sub.600=0.1). This culture of each clone was incubated for 72 hours at 25 C. at at 220 RPM (at peak UPO activity; OD.sub.600=25-30), the cells were separated by centrifugation (10 minutes at 4,500 RPM and 4 C.) and the supernatant was filtered (with glass and nitrocellulose filter with a pore size of 0.45 m).

(44) Purification of Recombinant AaeUPO1 Variants

(45) The purification of the recombinant AaeUPO variants described in the present invention was carried out by means of ion-exchange chromatography (KTA purifier, GE Healthcare). The raw extract was firstly treated by fractional precipitation with ammonium sulphate (55%, first cut) and, after eliminating the pellet, the supernatant was newly subjected to precipitation with ammonium sulphate (85%, second cut). The final pellet was re-suspended in the 10 mM pH 4.3 sodium phosphate/citrate buffer (buffer A) and the sample was filtered and loaded on a strong cation-exchange column (HiTrap SP FF, GE Healthcare), pre-balanced with buffer A. The proteins were eluded by means of a linear gradient of 0 to 25% of buffer A with 1 M of NaCl in 55 mL and of 25 to 100% of buffer A with 1 M NaCl in 5 mL, at a flow rate of 1 mL/min. The fractions with UPO activity were recovered, concentrated and dialysed in 10 mM pH 6.5 Bis Tris buffer (buffer B) and loaded on a high-resolution anion-exchange column (Biosuite Q, Waters), pre-balanced with buffer B. The proteins were eluded by means of a linear gradient of 0 to 15% of buffer B with 1 M of NaCl in 40 mL y de 15 a 100% de buffer B with 1 M NaCl in 5 mL, at a flow rate of 1 mL/min. The fractions with UPO activity were recovered, concentrated and dialysed in 50 mM pH 7.0 potassium phosphate buffer and stored at 4 C. Reinheitszahl [Rz] [A.sub.418/A.sub.280] values of 2 were obtained. The fractions of the different purification steps were analysed in a 12% SDS/PAGE acrylamide gel, dyed with Coomassie blue. The concentrations of the raw extracts of these steps were determined by means of Bradford reagent and BSA as standard.

(46) Kinetic Constants Values

(47) The kinetic constants of the variants of the invention for ABTS were estimated in 100 mM pH 4.0 sodium phosphate/citrate buffer and 2 mM H.sub.2O.sub.2; and for the rest of the substrates, in 100 mM pH 7.0 potassium phosphate buffer, 2 mM H.sub.2O.sub.2 (DMP) or 1 mM H.sub.2O.sub.2 (NBD and naphthalene, in 20% of acetonitrilefinal concentration). For H.sub.2O.sub.2, benzyl alcohol was used as substrate at the corresponding saturation conditions. The reactions were performed in triplicate and the oxidations of the substrates were followed by spectrophotometric changes (ABTS: .sub.418=36,000 M.sup.1 cm.sup.1; DMP: .sub.469=27,500 M.sup.1 cm.sup.1; NBD: .sub.425=9,700 M.sup.1 cm.sup.1, naphthalene: .sub.303=2,010 M.sup.1 cm.sup.1, and benzyl alcohol: .sub.280=1,400 M.sup.1 cm.sup.1). The kinetics for naphthalene were performed following the protocol described in M. G. Kluge, et al. Appl. Microbiol. Biotechnol. 2007. 75, 1473-1478. In order to calculate the values of K.sub.m and k.sub.cat, values of V.sub.max were represented at substrate concentrations and the hyperbole function was adjusted (using SigmaPlot 10.0, wherein the parameter a is equal to k.sub.cat and the parameter b, to K.sub.m).

(48) HPLC Analysis

(49) The reactions were analysed by means of chromatography in reverse phase (HPLC). The equipment is composed by a tertiary pump (Varian-Agilent Technologies, USA) coupled to an autosampler (Merck Millipore, MA, USA); an ACE C18 PFP column was used for separation (pentafluorophenyl, 15 cm4.6 cm) at 45 C. and detection was performed using a photodiode detector (PDA) (Varian-Agilent Technologies, USA). The mobile phase selected was 70% methanol and 30% ddH.sub.2O (in both cases with 0.1% of acetic acid) at a flow rate of 0.8 mL/min. The reaction was quantified at 268 nm (based on standard HPLCs). For the 15 minute reaction, the mixture contained 6.6 nM of pure enzyme, 1 mM naphthalene, 20% acetonitrile and 1 mM H.sub.2O.sub.2 in 100 mM pH 7.0 potassium phosphate buffer (1 mL of final volume). The reaction started with the addition of H.sub.2O.sub.2 and stopped with 20 L of 37% HCl. For long reaction times, the conditions used were those described earlier but without stopping the reaction with HCl. A sample of 10 L was injected and analysed at different reaction times (from 1 to 270 minutes).

(50) For the kinetic values of the 1-naphthol, the reaction was performed using 40 nM of pure enzyme, 1 mM 1-naphthol, 20% acetonitrile and 1 mM H.sub.2O.sub.2 in 100 mM pH 7.0 potassium phosphate buffer (0.2 mL of final volume).

(51) The standard deviations were less than 5% in all cases.

(52) Analysis Using MALDI-TOF-MS and Determination of the Isoelectric Point

(53) The analyses were performed using an Autoflex III MALDI-TOF-TOF unit with smartbeam laser (Bruker Daltonics). The samples were evaluated in positive mode. The method was calibrated using BSA with standard, thereby covering a range of 15,000 to 70,000 Da. In order to determine the isoelectric point of the UPO variants, 8 g of pure enzyme were subjected to two-dimensional electrophoresis. These experiments were carried out at the Proteomic and Genomic Service of the Biological Research Centre (CIB-CSIC, Spain).

(54) Analysis by Liquid Chromatography/Mass Spectrometry (LC/MS)

(55) These analyses were performed using a mass spectrometer with a Q-TOF hybrid analyser (QSTAR, ABSciex, MA, USA). Electrospray (ESI) was used as an ionisation source and, as ionising phase, methanol. In this case, the entrance system was direct injection in a HPLC 1100 (Agilent Technologies, USA). The resolution of the assay corresponds to 9,000 FWHM (Full Width at Half Maximum), accuracy, 5-10 ppm and was performed in negative mode.

(56) Results

(57) Taking the PaDa-I mutant enzyme of SEQ ID NO: 18 encoded by SEQ ID NO: 17 as parental to carry out the directed evolution experiments, UPO mutant libraries were built by means of random mutagenesis and recombination by StEP and in vivo DNA shuffling with the objective of obtaining a mutant enzyme or variant that shows less peroxidase activity on the 1-naphthol, while boosting peroxygenase activity on the naphthalene, also taking into account that said variant must be expressed robustly in heterologous organisms and secreted in an active, soluble and very stable form. To this end, each variant obtained in the mutant libraries was subjected to ad hoc double screening for the purpose of obtaining the variants with the aforementioned capabilities, greater peroxygenase activity against naphthalene and less peroxidase activity against 1-naphthol.

(58) After subjecting the PaDa-I mutant (SEQ ID NO: 17) to two cycles of directed evolution (4,000 clones analysed), a double mutant was identified which was called JaWa and which comprises the nucleotide sequence SEQ ID NO: 23, that encodes the variant of SEQ ID NO: 24. Said JaWa mutant (SEQ ID NO: 24) comprises the G241D and R257K mutations with respect to the PaDa-I mutant of SEQ ID NO: 18, with a peroxygenase activity on microplate that doubled that of its parental and a peroxidase activity that was reduced to half (FIG. 1).

(59) Both variants, PaDa-I and JaWa, were produced, purified at homogeneity (Reinheitszahl [Rz] [A.sub.418/A.sub.280] value 2) and biochemically characterised. No changes were detected with regard to general spectral characteristics, processing of the N-terminus, molecular mass or degree of glycosylation (Table 1).

(60) TABLE-US-00001 TABLE 1 Biochemical characteristics of wild-type AaeUPO (SEQ ID NO: 4) and of the PaDa-I (SEQ ID NO: 18) y JaWa (SEQ ID NO: 24) variants. Spectroscopic and biochemical characteristics Wild-type UPO PaDa-I JaWa Pm (Da).sup.1 46,000 52,000 52,000 Pm (Da).sup.2 n.d. 51,100 51,100 Pm (Da).sup.3 35,942 35,914 35,944 Degree of glycosylation (%) 22 30 30 Thermal stability, T.sub.50 ( C.).sup.4 n.d. 57.6 59.7 pI 4.9-5.7 5.5 5.3 Optimum pH for ABTS 4.0 4.0 4.0 Optimum pH for DMP 7.0 6.0 6.0 Optimum pH for naphthalene 6.5 6.0 6.0 Rz, (A.sub.418/A.sub.280) 2.4 1.8 2.3 Soret region (nm) 420 418 418 CT1 (nm) 572 570 570 CT2 (nm) 540 537 537 .sup.1Estimated by SDS-PAGE; .sup.2estimated using MALDI-TOF; .sup.3estimated according to the amino acid composition. .sup.4Estimated in culture supernatants. n.d. not determined.

(61) As can be observed in Table 1 and in FIG. 2, the JaWa mutant enzyme of SEQ ID NO: 24 showed greater kinetic thermostability than the PaDa-I variant of SEQ ID NO: 18 (2 C. higher T.sub.50-temperature at which the enzyme retains 50% of its activity after 10 minutes of incubation-), in addition to higher stability in the presence of acetonitrile, necessary for the bioavailability of the naphthalene (the solubility of the naphthalene in water is 31.7 mg/L) (FIG. 2).

(62) The naphthalene transformation reaction performed by the JaWa (SEQ ID NO: 24) and PaDa-I (SEQ ID NO: 18) mutants and that was analysed by means of HPLC-PDA has evidenced that the oxygenation of the naphthalene by AaeUPO occurs through an unstable intermediary compound, 1,2-naphthalene oxide (epoxide). It undergoes quick hydrolysis to naphthol (1- and 2-naphthol) when the pH is acid (M. Kluge, et al. Appl. Microbiol. Biotechnol. 2009. 81, 1071-1076). Therefore, the distribution of the resulting products after 15 minutes of reaction was firstly measured (stopped with HCl). Both the PaDa-I (SEQ ID NO: 18) and JaWa (SEQ ID NO: 24) variants demonstrated similar regioselectivity (92% 1-naphthol, 8% 2-naphthol), but the JaWa variant showed a significant increase in the production of 1-naphthol (156% more than PaDa-I) without detectable traces of 1,4-naphthoquinone, its oxidation product (FIG. 3A).

(63) When the long reaction times were monitored (270 minutes at pH 7.0 without stopping the reaction), a similar behaviour was observed, which indicates that the transformation of the 1,2-naphthalene oxide to naphtholes also occurs at neutral pH, although it is true that, at lower speed, traces of 1,4-naphthoquinone were also detected (FIG. 3B, C, D).

(64) While with both variants, PaDa-I and JaWa, the formation of the epoxide intermediary reached its maximum at 40 minutes (due to the oxidative damage caused by the H.sub.2O.sub.2 in all the peroxidases), regioselectivity increased to 97% of 1-naphthol. This result corresponds to the loss of selectivity observed in acid conditions given by a greater reactivity of the epoxide.

(65) The composition of the resulting products did not vary for any of the PaDa-I (SEQ ID NO: 18) and JaWa (SEQ ID NO: 24) variants, as observed in the mass spectrometry analysis performed, but the differences between the two mutants in terms of production performance were very significant, reaching values of 0.14 and 0.32 mM of 1-naphthol for PaDa-I and JaWa, respectively. The JaWa variant obtained total turnover numbers (TTN) of nearly 50,000 against the 20,000 of PaDa-I.

(66) Additionally, the kinetic values of the two variants were determined using substrates of both peroxygenase and peroxidase activity (Table 2), as described in the section on materials and methods. Briefly, the kinetic constants for the ABTS were measured in 100 mM pH 4.0 sodium phosphate/citrate buffer and 2 mM H.sub.2O.sub.2, while 100 mM pH 7.0 potassium phosphate and 2 mM H.sub.2O.sub.2 (DMP) or 1 mM (naphthalene or NBD, in 20% acetonitrilefinal concentration) was used for the other buffers. For the H.sub.2O.sub.2, benzyl alcohol was used as substrate to the corresponding saturation conditions.

(67) TABLE-US-00002 TABLE 2 Kinetic parameters for PaDa-I (SEQ ID NO: 18) and JaWa (SEQ ID NO: 24) variants. Kinetic Substrate constants PaDa-I JaWa ABTS K.sub.m (M) 48.0 4.5 181 22 k.sub.cat (s.sup.1) 395 13 125 5 k.sub.cat/K.sub.m 8.2 10.sup.6 6 10.sup.5 6.9 10.sup.5 6.3 10.sup.4 (s.sup.1 M.sup.1) DMP K.sub.m (M) 126 14 866 108 k.sub.cat (s.sup.1) 68 2 142 8 k.sub.cat/K.sub.m 5.4 10.sup.5 4.8 10.sup.4 1.6 10.sup.5 1.2 10.sup.4 (s.sup.1 M.sup.1) Naphthalene K.sub.m (M) 578 106 127 27 k.sub.cat (s.sup.1) 229 17 78 3 k.sub.cat/K.sub.m 4 10.sup.5 4 10.sup.4 6.2 10.sup.5 1.1 10.sup.5 (s.sup.1 M.sup.1) NBD K.sub.m (M) 483 95 769 80 k.sub.cat (s.sup.1) 338 22 154 8 k.sub.cat/K.sub.m 7 10.sup.5 9.9 10.sup.4 2.0 0.sup.5 1.2 10.sup.4 (s.sup.1 M.sup.1) H.sub.2O.sub.2 K.sub.m (M) 486 55 1,250 300 k.sub.cat (s.sup.1) 238 8 447 40 k.sub.cat/K.sub.m 5.0 10.sup.5 4.2 10.sup.4 3.6 10.sup.5 5.9 10.sup.4 (s.sup.1 M.sup.1)

(68) As can be observed in Table 2, the k.sub.cat/K.sub.m value (catalytic efficiency) for naphthalene was 1.5 times higher for the JaWa variant (SEQ ID NO: 24) with respect to the PaDa-I variant (SEQ ID NO: 18). Also, the peroxidase activity of the JaWa variant (SEQ ID NO: 24) was reduced (with a significant decrease in catalytic efficiencies of 3 to 11 times for the substrates of peroxidase activity DMP and ABTS, respectively). The k.sub.cat/K.sub.m value for H.sub.2O.sub.2 with benzyl alcohol as substrate was also affected. In the results obtained with NBD, another oxygen transfer substrate such as naphthalene, the trend is similar, i.e. k.sub.cat decreases in the JaWa variant while the affinity to the K.sub.m substrate improves, despite the fact that this entails higher k.sub.cat/K.sub.m for the PaDa-I variant. The fact that the catalytic efficiency of the JaWa variant for NBD has not improved is significant, since it is not a substrate used in the screenings of this part of the evolution. However, the fact that the tendency of the catalytic constant and affinity to the substrate is similar in two monooxygenase substrates indicates that there is an enzyme action mechanism acting in some way to favour the formation of 1-naphtol while reducing peroxidase activity.

(69) To confirm the decrease in peroxidase activity with respect to the hydroxylation of the naphthalene, the values of the catalytic constant were measured by using HPLC (mol product mol enzyme.sup.1 min.sup.1) for the conversion of 1-naphthol into 1,4-naphthoquinone. Although the catalytic constant of the PaDa-I variant (SEQ ID NO: 18) for 1-naphthol was already low (200 min.sup.1), with the JaWa variant (SEQ ID NO: 24) this value decreased to 92 min.sup.1, in addition to a reduction of 1.5 times in the ratio 1,4-naphthoquinone:1-naphthol (FIG. 4). This effect can also be observed at first glance, since the polymeric products produced in the reaction with the PaDa-I variants (SEQ ID NO: 18) (due to non-enzymatic quinone regrouping processes) are coloured (FIG. 4). There are hypotheses in literature on the possibility that UPO is similar to CPO in the existence of different sites with peroxidase activity in its structure. To suppress these alternative peroxidation pathways, the structure of the AaeUPO1 crystal was closely examined and a variant was built by mean of directed mutagenesis in Trp24 (FIG. 5A), a highly oxidable residue, found on the protein surface, using the PaDa-I (SEQ ID NO: 18) and JaWa (SEQ ID NO: 24) variants as templates, as described in the section on materials and methods.

(70) Next, the activities of the PaDa-I-W24F (SEQ ID NO: 30) and JaWa-W24F (SEQ ID NO: 32) variants were determined. The W24F mutation reduced 60% of the peroxidase activity in both variants and with all the tested substrates, but caused a decrease in the peroxygenase activity, with a reduction of 50% in the activity on the naphthalene and NBD (FIG. 5B). This indicates that the Trp24 residue probably also affects the peroxygenase activity of the UPO.

Example 2. Mutational Analysis of the Variants of the Invention

(71) The mutations of the JaWa variant were mapped (SEQ ID NO: 24) onto the structure of the wild AaeUPO1 (SEQ ID NO: 4), which shows a very characteristic catalytic pocket wherein linkage with the substrate takes place, dominated by a Phe triad (Phe69-Phe121-Phe199) involved in the correct orientation of the aromatic compounds (FIG. 6 and FIG. 7). The G241D mutation is at the entrance to the heme channel. The dramatic change of a Gly, apolar and small, for an Asp, loaded and larger, seems to narrow the cavity, which can affect the accommodation of the naphthalene in the catalytic pocket. This theory is not consistent with the fact that the affinity to naphthalene was improved in the JaWa variant, with a decrease in its K.sub.m of 3 times (Table 2). On the contrary, the introduction of a negative charge in the heme-thiolate domain (in which there is a Glu196-Arg189 acid-base pair involved in the formation of the Compound I-porphyrin with a radical cation and oxo-Fe IV=O) may negatively affect the k.sub.cat value, depending on the chemical nature of the bound substrate. The R257K mutation is located on the surface of the protein, far from catalysis-relevant regions, but is at the start of a pathway towards the catalytic R189 residue. It is a known fact that some peroxidases show various surface-exposed entrances for electron-mediated substrate oxidation through a long-range electron transfer pathway towards the heme domain, as also described in the present work for W24F variants. In this regard, the R257K replacement may be affecting any of these circuits with a possible beneficial lateral effect on thermostability through localised remodelling in the secondary structure (the two mutations, G241D and R257K, vary the estimation of factor B (FIG. 8)). B factor makes reference to the rigidity/flexibility of a protein or amino acid region present in a protein or peptide.

(72) These results evidence that the UPO variants described herein show greater selectivity and the highest TTN known for the production of 1-naphthol for this enzyme superfamily to date. Additionally, as demonstrated, said variants are heterologously secreted in an active, soluble and very stable form, being capable of carrying out selective aromatic oxygenations in the absence of NAD(P)H cofactors and reductase domains. Their self-sufficient mono(per)oxygenase activity make this UPO variant a valuable biocatalyst for application in the field of organic synthesis.

Example 3. Obtainment and Characterisation of Variants of the Invention for the Synthesis of Human Drug Metabolites (HDMs)

(73) The most important HDMs include, namely, derivatives of propranolol, a beta-blocker drug commonly used for the treatment of hypertension, migraine prophylaxis in children and attenuation of physical manifestations of anxiety. This example shows how the UPO variants of the invention are capable of forming 5-hydroxypropranolol from propranolol oxygenation, without inorganic pollutants, at room temperature, atmospheric pressure and in the absence of organic solvents, in a single step, with catalytic concentrations of H.sub.2O.sub.2 and without requiring the addition of antioxidants such as ascorbic acid to the reaction.

(74) In addition to the variants described in Example 1, a new variant was built based on the JaWa variant, which even showed an improvement in the production of 5-hydroxypropranolol with respect to said JaWa mutant. Following is a description of the obtainment of a new variant called SoLo comprising SEQ ID NO: 42 and which is encoded by the nucleotide sequence SEQ ID NO: 41.

(75) Materials and Methods

(76) Reagents and Enzymes

(77) ABTS (2,2-azino-bis(3-ethylbenzothiazolin-6-sulfonic acid)), L-ascorbic acid, 4-aminoantipyrine, benzyl alcohol, Taq DNA polymerase and the Saccharomyces cerevisiae transformation kit were obtained from Sigma-Aldrich (Saint Louis, Mo., USA). NBD (5-nitro-1,3-benzodioxole) was acquired from TCI America (Portland, Oreg., USA), while the naphthalene, propranolol and potassium persulfate are from Acros Organics (Geel, Belgium). 5-hydroxypropranolol was acquired from Santa Cruz Biotechnology (Santa Cruz, Calif., USA).

(78) The competent Escherichia coli XL2-Blue cells and Pfu ultra DNA polymerase were obtained from Agilent Technologies (Santa Clara, Calif., USA) and iProof high-fidelity DNA polymerase was acquired from Bio-Rad (Hercules, Calif., USA). The BamHI and XhoI restriction enzymes were obtained from New England Biolabs (Ipswich, Mass., USA) and the protease-deficient strain of S. cerevisiae BJ5465 from LGCPromochem (Barcelona, Spain). The Zymoprep Yeast Plasmid Miniprep and Zymoclean Gel DNA Recovery kits are marketed by Zymo Research (Orange, Calif., USA). The NucleoSpin Plasmid kit is from Macherey-Nagel (Dren, Germany) and the oligonucleotides used were synthesised by Metabion (Bayern, Germany). All the chemical compounds are of the highest purity available in the market.

(79) Directed Evolution

(80) Based on the JaWa mutant comprising SEQ ID NO: 24, which is encoded by the nucleotide sequence SEQ ID NO: 23, after each evolution cycle, the PCR products were loaded on a semi-preparatory agarose gel and purified using the Zymoclean Gel DNA Recovery kit. The recovered DNA fragments were cloned in the pJRoC30 plasmid under the control of the GAL1 promoter linearised with BamHI and XhoI (also eliminating the parental gel or predecessor). The linearised plasmid was loaded in a low-melting-point preparatory agarose gel and was purified using the Zymoclean Gel DNA Recovery kit.

(81) First Generation (1G)

(82) To obtain the SoLo mutant (SEQ ID NO: 42, encoded by SEQ ID NO: 41), docking studies were performed on the JaWa mutant (SEQ ID NO: 24, encoded by SEQ ID NO: 23) using the Molecular Operating Environment program (MOE, Chemical Computing Group Inc.) and propranolol as a substrate. Based on these, a region of the protein was selected to be subjected to random mutagenesis using the MORPHING technique (Mutagenic Organized Recombination Process by Homologous in vivo Grouping) (D. Gonzlez-Perez et al., PLoS ONE 2014. 9:e90919). To obtain the different variants additional to those described earlier, two error-prone PCRs were performed in a specific zone of the nucleotide sequence (SEQ ID NO: 23) that encodes that JaWa mutant (SEQ ID NO: 24), specifically in the coding zone from the D187-V248 region of the JaWa mutant of SEQ ID NO: 24 in a final volume of 50 L. These reactions contained 3% of dimethyl sulfoxide (DMSO), 90 nM MJaWa Fw (SEQ ID NO: 43; 5-gcgcattcaagactccattg-3), 90 nM MJaWa Rev (SEQ ID NO: 44; 5-gatcttgccgacattttttcc-3), 0.3 mM deoxynucleotide triphosphates (dNTPs, 0.075 mM of each), 0.1 mM or 0.2 mM MnCl.sub.2, 1.5 mM MgCl.sub.2, 0.05 U/L Taq DNA polymerase and 1 ng/l of the template (pJRoC30 plasmid from the California Institute of Technology (CALTECH, USA), comprising the nucleotide sequence of the JaWa mutant of SEQ ID NO: 23). This mutagenic PCR was performed in a gradient thermocyclator (Mycycler, Bio-Rad, EEUU), determining the following parameters: 94 C. 2 min (1 cycle); 94 C. 45 s, 48 C. 30 s and 72 C. 90 s (28 cycles); and 72 C. 10 min (1 cycle). Furthermore high-fidelity PCRs were performed in the fragments that must remain non-mutagenic in a final volume of 50 L. These reactions contained 3% of dimethyl sulfoxide (DMSO), 0.5 M HFJaWa Fw (SEQ ID NO: 45; 5-caggctcatcctatgcagccc-3) and 0.5 M RMLC (SEQ ID NO: 34; 5-gggagggcgtgaatgtaagc-3) or 0.5 M HFJaWa Rev (SEQ ID NO: 46; 5-caaaggagaaattggggttggtcg-3) and 0.5 M RMLN (SEQ ID NO: 33; 5-cctctatactttaacgtcaagg-3) for the other high-fidelity fragment, 1 mM dNTPs (0.25 mM of each), 0.05 U/L PfuUltra DNA polymerase and 2 ng/L of template. These reactions were performed in the same gradient thermocyclator, determining the following parameters: 95 C. 2 min (1 cycle); 95 C. 45 s, 48 C. 30 s and 72 C. 90 s (28 cycles); and 72 C. 10 min (1 cycle). 200 ng of PCR products were mixed with 100 ng of the linearised plasmid and competent S. cerevisiae cells were transformed such as to produce in vivo shuffling of the DNA and cloning (using the yeast transformation kit for such purpose). The volume resulting from the transformation was plated in minimal solid medium plates (for SC drop-out plates, said minimal solid medium consists of 100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L uracil-free amino acid supplement, 100 mL of 20% glucose, 20 g bacto agar, 700 mL of distilled water and 1 mL of 25 g/L chloramphenicol) and were incubated for 3 days at 30 C. The individual colonies that were formed were selected and subjected to the dual colorimetric High-Throughput Screening (HTS) assay to efficiently explore mutant libraries without altering the enzyme stability thereof, in addition to various re-screenings, as described below. In this evolution cycle, a new variant was obtained called SoLo, which comprises the nucleotide sequence SEQ ID NO: 41, that encodes the variant of SEQ ID NO: 42, wherein a new mutation took place: F191S, with respect to the JaWa variant (SEQ ID NO: 24).

(83) Second Generation (2G)

(84) Since the mutation that appeared in the SoLo variant (SEQ ID NO: 42) is found in one of the two phenylalanines that delimit the entrance to the heme channel, combinatorial saturation mutagenesis (CSM) was performed using the 22c-trick method, as described in S. Kille, et al. ACS Synth. Biol. 2013. 2.83-92, in positions S191 and F76.

(85) To this end, three PCRs were performed in a final volume of 50 L. All contained 3% of DMSO, 0.3 mM dNTPs (0.075 mM each), 0.05 U/L PfuUltra DNA polymerase and 2 ng/L of template, but each with different primers. PCR 1 with 0.25 M of RMLN (SEQ ID NO: 33), 0.25 M of F76 VHG R

(86) TABLE-US-00003 (SEQIDNO:47; 5-gcaagtccgtaatgagattgccgtccacaaggtgggccgcatatgtg gccdbgattgcggc-3),
0.25 M of F76 NDT R

(87) TABLE-US-00004 (SEQIDNO:48; 5-gcaagtccgtaatgagattgccgtccacaaggtgggccgcatatgt ggcahngattgcggc-3
and 0.25 M of F76 TGG R

(88) TABLE-US-00005 (SEQIDNO:49; 5-gcaagtccgtaatgagattgccgtccacaaggtgggccgcatatgtg gcccagattgcggc-3).
PCR 2 con 0.25 M of HF F

(89) TABLE-US-00006 (SEQIDNO:50; 5-gcggcccaccttgtggacggcaatctcattacggacttgc-3
0.25 M of S191 VHG R

(90) TABLE-US-00007 (SEQIDNO:51; 5-cccatccacaaaaagattcgcggggaaggtggtctcgccgtaagca gtcdbgaacctaaag-3
0.25 M of S191 NDT R

(91) TABLE-US-00008 (SEQIDNO:52; 5-cccatccacaaaaagattcgcggggaaggtggtctcgccgtaagca gtahngaacctaaag-3)
y 0.25 M of S191 TGG R

(92) TABLE-US-00009 (SEQIDNO:53; 5-cccatccacaaaaagattcgcggggaaggtggtctcgccgtaagca gtccagaacctaaag-3).
PCR 3 con 0.25 M de HF F-RMLC

(93) TABLE-US-00010 (SEQIDNO:54; 5-cggcgaciaccaccttccccgcgaatctttttgtggatggg-3)
and 0.25 M of RMLC (SEQ ID NO: 34). The underlined regions are those in which in vivo DNA assembly occurs and the region in italics is the changed codon (where N=A/T/C/G; D=no C; V=no T, H=no G; and B=no A). These reactions were performed in the gradient thermocyclator, determining the following parameters: 95 C. 2 min (1 cycle); 95 C. 45 s, 48 C. 45 s and 72 C. 60 s (28 cycles); and 72 C. 10 min (1 cycle). 200 ng of each of the PCR products were mixed with 100 ng of the linearised plasmid and transformed into competent S. cerevisiae cells. The rest of the procedure was followed as explained previously to obtain the first generation. No improved variant was obtained with respect to the SoLo mutant.
Third Generation (3G)

(94) There is a phenylalanine triad in the catalytic pocket of AaeUPO, PaDa-I and JaWa (F69-F121-F199). Due to the complex catalytic pocket and to the fact that these phenylalanines are in charge correctly orienting the aromatic substrates, it was decided to carry out mutagenesis on these residues with NNK degenerated codons (N=A/T/C/G; D; K=T/G, M=A/C) independently, i.e. creating three different libraries.

(95) Library F69: two PCRs were performed in a final volume of 50 L. The first contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLN (SEQ ID NO: 33), 0.5 M F69 R (SEQ ID NO: 55; 5-gaagattgcggcttgattgtcmnnattgaatc-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41). And the second contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLC (SEQ ID NO: 34), 0.5 M F69 F (SEQ ID NO: 56; 5-cgcggttcaggaaggattcaatnnkgacaatc-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41).

(96) F121 library: two PCRs were performed in a final volume of 50 L. The first contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLN (SEQ ID NO: 33), 0.5 M F121 R (SEQ ID NO: 57; 5-catactggcgtcgccttcmnnggtgccatgc-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41). And the second contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLC (SEQ ID NO: 34), 0.5 M F121 F (SEQ ID NO: 58; 5-ggactcaatgagcatggcaccnnkgaaggcg-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41).

(97) F199 library: two PCRs were performed in a final volume of 50 L. The first contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLN (SEQ ID NO: 33), 0.5 M F199 R (SEQ ID NO: 59; 5-ccacaaaaagattcgcgggmnnggtggtctcg-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41). And the second contained 3% of DMSO, 0.2 mM dNTPs (0.05 mM of each), 0.5 M RMLC (SEQ ID NO: 34), 0.5 M F199 F (SEQ ID NO: 60; 5-ctactgcttacggcgagaccaccnnkcccgcg-3), 0.02 U/L iProof DNA polymerase and 2 ng/L of template (SoLo comprising SEQ ID NO: 41).

(98) These reactions were performed in the gradient thermocyclator, determining the following parameters: 98 C. 30 s (1 cycle); 98 C. 10 s, 48 C. 30 s and 72 C. 30 s (28 cycles); and 72 C. 10 min (1 cycle). 200 ng of each of the PCR products were mixed with 100 ng of the linearised plasmid (each library separately) and transformed into competent S. cerevisiae cells. The rest of the method was followed as explained earlier to obtain the first and second generation. Neither was any variant better than SoLo found (SEQ ID NO: 42), due to which this mutant was selected, together with the JaWa mutant (SEQ ID NO: 24) and the parental AaeUPO1, to analyse the synthesis of HDMs, taking 5-hydroxypropranolol with each by way of example.

(99) Preparation of the Mutant Libraries

(100) Individual colonies corresponding to clones were selected and inoculated in sterile 96-well plates (Greiner Bio-One GmbH, Germany), hereinafter mother plates, with 200 L/minimal medium for expression per well (100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L uracil-free amino acid supplement, 67 mL of 1 M pH 6.0 potassium phosphate buffer, 111 mL of 20% galactose, 22 mL of 0.1 M MgSO.sub.4, 31.6 mL of absolute ethanol, 1 mL of 25 g/L chloramphenicol and ddH.sub.2O up to 1,000 mL). Column 6 of each column was inoculated with the corresponding parental and well H1 with S. cerevisiae transformed with the pJRoC30-MtL plasmid (laccase without functional expression). The plates were sealed to avoid evaporation and were incubated at 30 C., 220 RPM and 80% of relative humidity (in a Minitron, INFORS, Switzerland) for five days.

(101) Dual Colorimetric High-Throughput Screening (HTS)

(102) The mother plates were centrifuged (Eppendorf 5810R centrifuge, Germany) for 10 minutes at 3,500 RPM and 4 C. 20 L of supernatant of these mother plates were transferred to two replica daughter plates with the help of a Freedom EVO liquid-handling robot (Tecan, Switzerland). 50 L of reaction mixture with propranolol were added to the daughter plates using a pipetting robot (Multidrop Combi Reagent Dispenser, Thermo Scientific, USA).

(103) The reaction mixture with propranolol was composed of 50 mM pH 7.0 potassium phosphate buffer, 5 mM propranolol and 2 mM H.sub.2O.sub.2 to detect the peroxygenase activity of the enzyme on the substrate and its subsequent peroxidase activity on the product. This same screening assay was simultaneously carried out but adding ascorbic acid (4 mM) to the reaction mixture in order to exclusively detect the peroxygenase activity of the enzyme on propranolol and avoid the subsequent peroxidase activity. Without ascorbic, the plates were incubated for 30 minutes and with ascorbic for 60 minutes. Subsequently, by means of the 4 aminoantipyrine (4-AAP, C. R. Otey and J. M. Joern, Methods Mol. Biol. 2003. 230, 141-8) the amount of product formed per well was revealed. The plates were briefly agitated and absorbance measured at 530 nm, using a plate reader for such purpose (SPECTRAMax Plus 384, Molecular Devices, USA). The values were normalised against the parental of each plate. To rule out false positives, re-screenings were carried out, in addition to a third re-screening wherein kinetic stability was determined (T.sub.50) (P. Molina-Espeja, et al. Appl. Environ. Microbiol. 2014. 80, 3496-3507).

(104) Second Re-Screening

(105) An aliquot with the 10 best screening clones was inoculated in 3 mL of YPD culture medium (10 g of yeast extract, 20 g of peptone, 100 mL of 20% glucose, 1 mL of 25 g/L chloramphenicol and ddH2O up to 1,000 mL) at 30 C. and 220 RPM for 24 hours. The plasmids of those cultures were extracted using the Zymoprep Yeast Plasmid Miniprep kit. Due to the impurity and low concentration of the DNA extracted, the plasmids were transformed into supercompetent E. coli XL2-Blue cells and plated in LB-amp plates (Luria-Bertani medium is composed of 5 g of yeast extract, 10 g of peptone, 10 g of NaCl, 100 mg of ampicillin and ddH2O up to 1,000 mL). An individual colony was selected from each clone, inoculated in 5 mL of LB and grown for 16 hours at 37 C. and at 250 RPM. The plasmids were extracted using the NucleoSpin Plasmid kit and transformed into competent S. cerevisiae cells (as in the parental, which in the first generation is JaWa and in the second and third is SoLo). Five individual colonies of each clone were selected and inoculated to undergo the same previously described screening protocol.

(106) Third Re-Screening. Thermostability Assay

(107) An individual S. cerevisiae colony was selected with the corresponding clone (grown on a SC drop-out minimal medium plate: 100 mL of 6.7% yeast nitrogen base, 100 mL of 19.2 g/L uracil-free amino acid supplement, 100 mL of 20% glucose, 1 mL of 25 g/L chloramphenicol and ddH.sub.2O up to 1,000 mL), was inoculated in 3 mL of selective minimal medium (like the SC plate medium, but with 20 g of bacto agar and rafinose instead of galactose) and incubated for 48 hours at 30 C. and 220 RPM. An aliquot of this culture was taken such that, upon inoculating it in 5 mL of new minimal medium, optical density at 600 nm would have a value of 0.25 (optical density, OD.sub.600=0.25). This starter was incubated until completing two full growth cycles (between 6 and 8 hours), at which time 1 mL of cells were taken to inoculate 9 mL of expression medium in a 100 mL flask (OD.sub.600=0.1). This culture of each clone was incubated for 72 hours at 25 C. and 220 RPM (at peak UPO activity; OD.sub.600=25-30), the cells were separated by centrifugation (10 minutes at 4,500 RPM and 4 C.) and the supernatant was filtered (using a glass and nitrocellulose filter with a pore size of 0.45 m). Appropriate supernatant dilutions were prepared so that aliquots of 20 L would give rise to a linear response in kinetic mode. 50 L of supernatant were used for each point at a temperature gradient created using a thermocyclator, from 30 to 80 C. After incubating for 10 minutes, the aliquots were cooled in ice for 10 minutes and tempered at room temperature for 5 minutes. Lastly, these supernatants were subjected to the colorimetric assay using ABTS (100 mM pH 4.0 sodium phosphate/citrate buffer, 0.3 mM ABTS and 2 mM H.sub.2O.sub.2). The thermostability values were calculated in accordance with the ratio between the residual activities incubated at different temperatures and the value of initial activity at room temperature. The value of T.sub.50 was determined as as the temperature value at which the protein loses 50% of its initial activity after incubating for 30 minutes.

(108) Production of UPO Recombinant Variants in S. cerevisiae

(109) An independent S. cerevisiae colony that comprised the corresponding variant of the invention, on the one hand JaWa and on the other SoLo, was selected from a SC drop-out minimal medium plate and inoculated in 20 mL of liquid SC minimal medium, cultures that were incubated for 48 hours at 30 C. and 220 RPM. An aliquot of this culture was taken so that, upon inoculating it in 100 mL of new minimal medium, OD.sub.600 would have a value of 0.25. This starter was incubated until completing two full growth cycles (between 6 and 8 hours), at which time 100 mL of cells were taken to inoculate 900 mL of minimal medium for expression in a 2,000 mL flask (OD.sub.600=0.1). This culture of each clone was incubated for 72 hours at 25 C. and at 220 RPM (at peak UPO activity; OD.sub.600=25-30), the cells were separated by centrifugation (10 minutes at 4,500 RPM and 4 C.) and the supernatant was filtered (using a glass and nitrocellulose filter with a pore size of 0.45 m).

(110) Purification of Recombinant AaeUPO1 Variants

(111) The purification of the variants described in the present invention, JaWa and SoLo, was carried out using cation-exchange chromatography followed by anion-exchange chromatography (KTA purifier, GE Healthcare). The raw extract was concentrated and dialysed in 20 mM pH 3.3 sodium phosphate/citrate buffer (buffer A) by means of tangential ultrafiltration (Pellicon; Millipore, Temecula, Calif., USA) through a membrane with a pore size of 10 kDa (Millipore) using a peristaltic pump (Masterflex Easy Load; Cole-Parmer, Vernon Hills, Ill.). The sample was filtered and loaded on a strong cation-exchange column (HiTrap SP FF, GE Healthcare), pre-balanced with buffer A. The proteins were eluded by means of a linear gradient of 0 to 40% of buffer A with 1M NaCl in 60 mL and from 40 to 100% of buffer A with 1 M NaCl in 5 mL, at a flow rate of 1 mL/min. The fractions with UPO activity were recovered, concentrated and dialysed in 20 mM pH 7.8 Tris-HCl buffer (buffer B) and loaded on a high-resolution anion-exchange column (Biosuite Q, Waters), pre-balanced with buffer B. The proteins were eluded by means of a linear gradient of 0 to 20% of buffer B with 1 M NaCl in 40 mL and from 20 to 100% of buffer B with 1 M NaCl in 5 mL, at a flow rate of 1 mL/min. The fractions with UPO activity were recovered, concentrated and dialysed in 10 mM pH 7.0 potassium phosphate buffer and stored at 4 C. Reinheitszahl [Rz] [A.sub.418/A.sub.280] values of 2 were obtained. The fractions of the different purification steps were analysed in a 12% SDS/PAGE acrylamide gel, dyed with Coomassie blue. The concentrations of the raw extracts of these steps were determined by means of Bradford reagent and BSA as standard.

(112) Kinetic Constants Values

(113) The kinetic constants of the variants of the invention, AaeUPO, PaDa-I, JaWa and SoLo, for ABTS were estimated in 100 mM pH 4.0 sodium phosphate/citrate buffer and 2 mM H.sub.2O.sub.2; and for the other substrates, in 100 mM pH 7.0 potassium phosphate buffer and 2 mM H.sub.2O.sub.2 (propranolol). For H.sub.2O.sub.2, benzyl alcohol was used as substrate at the corresponding saturation conditions. The reactions were performed in triplicate and the oxidations of the substrates were followed by spectrophotometric changes (ABTS: .sub.418=36,000 M.sup.1 cm.sup.1; Propranolol: .sub.325: 1,996 M.sup.1 cm.sup.1 and benzyl alcohol: .sub.280=1,400 M.sup.1 cm.sup.1). The kinetics for propranolol were performed calculating .sub.325 experimentally at pH 7.0. In order to calculate the values of K.sub.m and k.sub.cat, values of V.sub.max were represented at substrate concentrations and the hyperbole function was adjusted (using SigmaPlot 10.0, wherein the parameter a is equal to k.sub.cat and the parameter b, to K.sub.m).

(114) HPLC Analysis

(115) The reactions were analysed by means of chromatography in reverse phase (HPLC). The equipment was composed of a tertiary pump (Varian-Agilent Technologies, USA) coupled to an autosampler (Merck Millipore, MA, USA); for the separation, a Zorbax Eclipse plus C18 column (15 cm4.6 cm) at 40 C. was used and the detection was performed using a photodiode detector (PDA) (Varian, Agilent Technologies, USA). The mobile phase selected was a gradient from 10% methanol and 90% ddH.sub.2O (in both cases with 0.1% of acetic acid) up to 90% methanol and 10% ddH.sub.2O at a flow rate of 0.8 mL/min. The reaction was quantified at 280 nm (based on HPLC standards). For the 15 minute reaction, the mixture contained 0.03 M of pure enzyme, 4 mM propranolol and 2 mM H.sub.2O.sub.2 in 50 mM pH 7.0 potassium phosphate buffer (0.5 mL of final volume). The reaction was started with the addition of H.sub.2O.sub.2 and was stopped with 20 L of 37% HCl. In order to determine the turnover rates of the variants with 5-hydroxypropranolol (product of interest), the mixture contained 0.03 M of pure enzyme, 0.4 mM 5-hydroxypropranolol and 2 mM H.sub.2O.sub.2 in 50 mM pH 7.0 potassium phosphate buffer (0.3 mL of final volume). In order to calculate the total turnover number (TTN) of the assayed variants, the assay was carried out using 0.03 M of pure enzyme, 5 mM propranolol and 2 mM H.sub.2O.sub.2 in 50 mM pH 7.0 potassium phosphate buffer and in the same manner, but also adding 4 mM ascorbic acid. In both cases, 2 mM H.sub.2O.sub.2 was added every 10 minutes, monitoring the reaction in each addition point taking different aliquots. The standard deviations were less than 5% in all cases.

(116) Analysis by Liquid Chromatography/Mass Spectrometry (LC/MS)

(117) These analyses were performed using a mass spectrometer with a Q-TOF hybrid analyser (QSTAR, ABSciex, MA, USA). Electrospray (ESI) was used as an ionisation source and, as ionising phase, methanol. In this case, the entrance system was direct injection in a HPLC 1100 (Agilent Technologies, USA). The resolution of the assay corresponds to 9,000 FWHM (Full Width at Half Maximum), accuracy at 5-10 ppm and it was performed in positive mode.

(118) Results

(119) The activity of the different UPO variants was evaluated by means of the 4-AAP assay to determine the most appropriate starting point for determining the capacity of said variants for HDM synthesis (FIG. 9). As can be observed in the figure, the variant with the greatest activity against propranolol and best ratio among its activity with and without ascorbic was JaWa (SEQ ID NO: 24, encoded by SEQ ID NO: 23), due to which it was the mutant selected for the docking assays (FIG. 10). Based on these results, wherein it was observed that the substrate interacted with a series of residues of the catalytic pocket and of the heme access channel, a region of the JaWa mutant that was in direct contact with the substrate was selected (residues D187-V248 of SEQ ID NO: 24). The objective is to obtain a mutant enzyme or variant that shows less peroxidase activity on 5-hydroxypropranolol (which is the product of the reaction with propranolol) while improving peroxygenase activity on propranolol, also taking into account that said variant must be expressed robustly in heterologous organisms and secreted in an active, soluble and very stable form. To this end, each variant obtained in the mutant libraries was subjected to double screening designed ad hoc for the purpose of obtaining the variants with the aforementioned capabilities, greater peroxygenase activity on propranolol (measured in the presence of ascorbic acid) and less peroxidase activity against 5-hydroxypropranolol (in the absence of ascorbic acid). Two libraries with different mutagenic rates (concentration of MnCl.sub.2) were analysed, identifying a single mutant in both libraries and repeatedly to that called SoLo and which comprises the nucleotide sequence SEQ ID NO: 41 that encodes the variant of SEQ ID NO: 42. Said SoLo mutant (SEQ ID NO: 42) has the F191S mutation (FIG. 11) with respect to the JaWa mutant of SEQ ID NO: 24, with a peroxygenase activity on microplate 30% higher than its parental (JaWa) and decrease in peroxidase activity of more than two fold.

(120) Two further cycles of evolution (2G and 3G) were performed using the SoLo variant (SEQ ID NO: 41) as parental, wherein no enhanced variant was detected.

(121) Both variants, JaWa (SEQ ID NO: 24) and SoLo (SEQ ID NO: 42), were produced, purified at homogeneity (Reinheitszahl [Rz] [A.sub.418/A.sub.280] value 2) and biochemically characterised.

(122) As can be observed in FIG. 12, the SoLo variant of SEQ ID NO: 42 showed very similar kinetic thermostability to that of the JaWa mutant (SEQ ID NO: 24).

(123) The propranolol transformation reaction performed by the wild AaeUPO enzyme (SEQ ID NO: 2), and the PaDa-I (SEQ ID NO: 18), JaWa (SEQ ID NO: 24) and SoLo (SEQ ID NO: 42) variants in the absence of ascorbic acid and was analysed using HPLC-PDA is included in FIG. 13. It can be observed that both JaWa and SoLo are those that produce the largest amount of 5-hydroxypropranolol, in addition to having 99% of regioselectivity, since traces of neither 4-hydroxypropranolol nor N-desisopropyl propranolol (DYP) were detected.

(124) The kinetic value of AaeUPO, JaWa and SoLo for propranolol, and for ABTS and H.sub.2O.sub.2 (Table 3) were determined.

(125) TABLE-US-00011 TABLE 3 Kinetic parameters for the variants of the invention and for wild AaeUPO. Kinetic Substrate constants AaeUPO1 PaDa-I JaWa SoLo ABTS Km (M) 25-0 2.5 48.8 4.5 181 22 568 91 K.sub.cat (s.sup.1) 221 6 395 13 125 5 365 23 K.sub.cat/K.sub.m (s.sup.1M.sup.1) 8.8 10.sup.6 6.9 10.sup.5 8.2 10.sup.6 6.0 10.sup.5 6.9 10.sup.5 6.3 10.sup.4 6.4 10.sup.5 6.7 10.sup.4 Propranolol Km (M) 2,239 333 2,268 220 244 92 391 97 K.sub.cat (s.sup.1) 150 12 212 11 765 76 497 35 K.sub.cat/K.sub.m (s.sup.1M.sup.1) 6.7 10.sup.4 4.8 10.sup.3 9.3 10.sup.4 4.3 10.sup.3 3.1 10.sup.6 0.9 10.sup.5 1.3 10.sup.6 0.2 10.sup.5 Naphthalene Km (M) 156 20 578 106 127 27 789 96 K.sub.cat (s.sup.1) 92 3 229 17 78 3 337 20 K.sub.cat/K.sub.m (s.sup.1M.sup.1) 5.9 10.sup.5 5.9 10.sup.4 4.0 10.sup.5 4.0 10.sup.4 6.2 10.sup.5 1.1 10.sup.4 4.3 10.sup.5 2.8 10.sup.4 H.sub.2O.sub.2 Km (M) 1,370 162 486 55 1,250 153 1,430 153 K.sub.cat (s.sup.1) 290 15 238 8 446 23 446.23. K.sub.cat/K.sub.m (s.sup.1M.sup.1) 2.1 10.sup.5 1.5 10.sup.4 5.0 10.sup.5 4.2 10.sup.4 3.1 10.sup.5 1.8 10.sup.4 3.1 10.sup.5 1.8 10.sup.4

(126) As can be observed in Table 3, both the JaWa (SEQ ID NO: 24) and SoLo (SEQ ID NO: 42) variants increased the k.sub.cat/K.sub.m (catalytic efficiency) values for propranolol by two orders of magnitude. It can also be observed that the JaWa (SEQ ID NO: 24) and SoLo (SEQ ID NO: 42) variants show a reduction in peroxidase activity, measured with ABTS, of one order of magnitude in catalytic efficiency, being the affinity to the substrate, in the case of the SoLo variant, three fold worse with respect to its parental. The values for H.sub.2O.sub.2 with benzyl alcohol were not affected. As in the case of the propranolol between JaWa and SoLo, JaWa has kinetic constants similar to AaeUPO with the naphthalene as substrate, differentiating itself in the total turnover values, which are higher for JaWa.

(127) Since the kinetics with propranolol of the JaWa and SoLo variants are very similar, the turnover rates were calculated with 5-hydroxypropranolol as a substrate in the absence of ascorbic acid, in order to evaluate the peroxidase activity of each variant against its propranolol reaction product. In FIG. 14 it can be observed that JaWa and AaeUPO oxidise practically the entire product, but SoLo is capable of maintaining approximately 50% thereof without oxidising. It follows that the SoLo variant (SEQ ID NO: 42), has significantly reduced its peroxidase activity on its own product, allowing higher performances in the production of this propranolol metabolite.

(128) When the reaction was monitored for long reaction times with the addition of 2 mM H.sub.2O.sub.2, the total turnover numbers (TTNs) were determined, obtaining a value of 45,000 for SoLo, 15,000 for JaWa and 3,000 for AaeUPO in the absence of ascorbic acid; and in the presence of ascorbic acid, 62,000 for SoLo, 48,000 for JaWa and 14,000 for AaeUPO (Table 4). This implies that, even by adding ascorbic acid to the reaction, the independent use of this antioxidant in the reaction medium is possible, simplifying the process. (FIG. 15).

(129) TABLE-US-00012 TABLE 4 Determination of the total turnover numbers (TTNs) for the variants of the invention and for wild AaeUPO. TTNs With ascorbic acid Without ascorbic acid AaeUPO 14,000 3,000 JaWa 48,000 15,000 SoLo 62,000 45,000