Artificial esterases
Description:
The invention relates to artificially mutated, enzymatically active, recombinant esterases which are derived from a pig liver esterase sequence and possess a high enzymatic activity.
The importance of esterases as biocatalysts in organic synthesis or as additives in household products, e.g. detergents, has been increasing steadily for many years. This has led research to focus mainly on the preparation of novel esterases that are distinguished on the one hand by their high enantioselectivity in the conversion of their substrates, and on the other hand by a high enzymatic activity.
Despite their great importance and the diversity of their possible applications, only a few recombinant esterases are available. Until recently this also applied particularly to pig liver esterases, which were only obtainable as extracts. These extracts have a variety of disadvantages. Apart from fluctuations in the proportion of esterase between different batches, one particular problem is the presence of other hydrolases in the extract. Furthermore, it has so far proved impossible to separate the extracts into the individual isoenzymes contained therein, so this is another reason why the extracts have fluctuated in their enzymatic a-ctivities. In addition, extraction of the enzymes, e.g. from pig liver tissue, carries the risk of viral or toxic contamination of the extract.
To solve these problems, putative pig liver esterase genes were cloned (Matsushima, M., et al., FEES Lett. 293, 37-41
(1991); Swiss-Prot. Ace. No. Q29550), but the functional expression of an active pig liver esterase enzyme was not successful with the constructs obtained. As an alternative, attempts were also made to obtain other esterases from higher organisms, e.g. a putative glycerol ester hydrolase from the digestive tract of pigs (described in David L. et al., Eur. J. Biochem. 257, 142- 148 (1998); Smialowski-Fleter, S. et al., Eur. J. Biochem. 269, 1109-1117 (2002)). However, it has also proved impossible so far to express this recombinant putative esterase gene with enzymatic activity.
The expression of a pure recombinant functional pig liver esterase was only recently successful (WO 02/48322) . For the first time, heterologous biotechnological preparation of the recombinant esterase made it possible to obtain pure enzymes uncontaminated by other hydrolases or isoenzymes. The esterases obtained are distinguished by a very good enantioselectivity towards many substrates, which is also in marked contrast to the properties of the conventional pig liver esterase extracts.
However, for many applications, especially technical applications, it would be desirable to have enzymes with modified properties, e.g. a different enantioselectivity or an increased activity.
In an attempt to prepare the above-described putative glycerol ester hydrolase starting from the amino acid sequence of recombinant pig liver esterase, Seq. ID No. 1 (WO 02/48322), by site directed mutagenesis, a number of artificial mutant enzymes were produced and, surprisingly, the enzymatic activity and temperature stability of some of the artificial enzymes obtained is higher than that of recombinant pig liver esterase or that of the glycerol ester hydrolase produced. Furthermore, the artificial
mutants exhibit a different enantioselectivity from that of rPLE.
The object is thus achieved by recombinant esterases containing an amino acid sequence Seq. ID No. 1 or a variant thereof with at least 50%, preferably over 70%, particularly preferably over 80% and very particularly preferably over 90% sequence homology, the amino acid sequence in the case of Seq. ID No. 1 being mutated in at least one of the positional regions from 74 to 79, 90 to
95, 110 to 114, 127 to 141, 193 to 197, 234 to 239, 288 to 296 and 365 to 372. Preferably at least one of positions 76, 77, 92, 93, 112, 129, 133, 134, 138, 139, 195, 236, 237, 290, 294, 367 and 370 is mutated. There is the possibility of exchange of the amino acid given by Seq. ID No. 1, in said positions, with another natural or modified amino acid.
In the case of the amino acid sequence Seq. ID No. 1, preferred mutants contain the following amino acid exchanges:
- position 76: valine (V) with alanine (A),
- position 77: glutamic acid (E) with glycine (G) ,
- position 92: threonine (T) with isoleucine (I), - position 93: leucine (L) with proline (P) ,
- position 112: lysine (K) with arginine (R) ,
- position 129: leucine (L) with valine (V),
- position 133: proline (P) with serine (S) ,
- position 134: methionine (M) with threonine (T) , - position 138: valine (V) with leucine (L) ,
- position 139: valine (V) with alanine (A),
- position 195: proline (P) with threonine (T) ,
- position 236: valine (V) with alanine (A),
- position 237: alanine (A) with glycine (C) , - position 290: phenylalanine (F) with leucine (L) ,
- position 294: glutamine (Q) with proline (P) ,
- position 367: tyrosine (Y) with phenylalanine (F) ,
- position 370: alanine (A) with threonine (T) .
Preferred mutants possess the indicated amino acid exchanges in positions 92, 93, 112, 138, 139, 236, 237,
367 and 370, which are distinguished by a greater activity increase towards the model substrates tested (cf. Fig. 3) . It is striking here that the preferred mutations are not located in the direct vicinity of the active centre of the esterase formed by the amino acids in positions 80, 186 and 431, as would have been expected. However, enzymes that are mutated in positions 76, 77, 195, 290 and 294 also exhibit an increased enzymatic activity towards the model substrates. Particularly preferred enzymes contain mutations in positions 76, 77, 112, 236 and 237.
Artificial mutant enzymes with a mutation in positions 112, 129, 133 and 134 are distinguished in particular by a high specific activity at elevated temperatures, preferably at above 65°C (Pig. 2) . However, mutations in positions 195, 236, 237, 290 and 294 also exhibit an increased specific activity at elevated temperatures.
The insertion of mutations in appropriate positions not only increases the activity, but also varies the enantioselectivity of the enzyme. It has also been possible in the case of specific substrates to observe an appreciable increase in enantioselectivity. In particular, the insertion of a mutation in position 77 of rPLE (Seq. ID No. 1) has a strong influence on the enantioselectivity (Fig. 6) . It is furthermore found, e.g. in the hydrolysis of carboxylic acid esters, that the use of rPLE leads preferentially to the formation of (S) enantiomers, while the artificial enzymes are preferable for the synthesis of (R) enantiomers.
The present invention also provides artificial esterases
which, in the case of Seq. ID No. 1, have partial sequences shortened by up to 75, preferably up to 50 and particularly preferably up to 25 amino acids at the N- terminal end and/or by up to 100, preferably up to 50, particularly preferably up to 25 and very particularly preferably up to 5 amino acids at the C-terminal end.
Apart from the pure proteins, the artificially mutated esterases claimed also include post-translationally modified enzymes.
Artificially mutated proteins based on the amino acid sequence Seq. ID No. 1 can be used directly as enzymes or they can multimerize as subunits with identical or different, naturally occurring or artificially mutated subunits. Thus, for example, the proteins claimed can form enzymatically active multimers with the known, naturally occurring subunits of pig liver esterases of the α, β or γ type.
The mutagenized enzymes have other advantages apart from increased enzymatic activity. In particular, the enzymes can be expressed as recombinant proteins and are thus obtainable in pure form, i.e. without interfering and not easily removable contamination due to other hydrolases or isoenzymes. Said expression takes place with maintenance of the enzyme function even with common host cells like P. pastoris . It is also easy to produce fusion proteins containing e.g. other functional domains at their C- terminal end, such as a myc-his domain for easier purification. The N terminus can likewise contain fused, functional peptide domains, of which secretory signal domains, e.g. the α-factor signal sequence or the ompA signal sequence, are of particular interest. However, the N terminus of the natural amino acid sequence (Swiss. Prot. Ace. No. Q29550) , or a variant thereof with a
homology of over 50%, preferably of over 70% and particularly preferably of over 80%, can also be fused.
The artificial mutant enzymes are prepared via expression of the corresponding mutated nucleic acid sequences. In the case of Seq. ID No. 29, the mutations are inserted by site directed mutagenesis, e.g. using the QuikChange Kit from Stratagene, with appropriate primers.
Thus the present invention also provides nucleic acids that code for the recombinant, artificially mutated pig liver esterases according to the invention, or nucleic acids that are complementary to these coding pig liver esterase nucleic acids, and that hybridize under stringent conditions, where, starting from Seq. ID No. 29, which codes for the peptide of Seq. ID No. 1, of these are mutated in at least one codon of positions 76, 77, 92, 93, 112, 129, 133, 134, 138, 139, 195, 236, 237, 290, 294, 367 and 370. Thus, for example, mutations can be inserted in positions 227, 230, 275, 278, 335, 385, 397 and 399, 401 and 402, 412, 416 and 417, 583, 707, 710, 870, 881 and 882, 1100 and 1108, preferred base exchanges being apparent from the mutated sequences Seq. ID No. 30 to 38.
Stringent hybridization conditions are understood as meaning a hybridization temperature of 60°C and 0.1 x SSC in 0.1% SDS. As also described e.g. in WO 02/48322, the coding DNA sequences can be cloned into conventional vectors and expressed after the transfection of host cells with such vectors in cell culture. An example of a suitable vector is pUC19 or pCYTEX for the transformation of E. coli, or pPICZα for the transformation of the yeast Pichia pastoris. The species Aspergillus sp. , Schwanniomyces sp. , Kluyveromyces sp. , Yarrowia sp. , Arxula sp . , Saccharomyces sp. , Hansenula sp. or Pichia sp. include other interesting unicellular organisms which have
proved suitable as hosts for the biotechnological expression of recombinant enzymes. Preferred host organisms which may be mentioned in addition to P. pastoris are Saccharomyces cerevisiae, Aspergillus orycae, Schwanniomyces occidentalism Kluyveromyces lactis , Yarrowia lipolytica , Arxula adeninivrans , Pichia methanolica, Pichia guilliermondii or Hansenula polymorpha .
In the vectors the coding DNA fragments must be located in the open reading frame belonging to a promoter. Preferred promoters are especially strong promoters, e.g. the lac, lambda, T7 or T4 promoter, the rhamnose-inducible promoter or the alcohol oxidase (AOXI) promoter. The vectors can contain other functional regions. In addition to selection markers and origins of replication, gene regulating elements, e.g. operators, repressors or transcription factors, are of special interest. In particular, it is possible to use a construction of vectors that allows reversible, inducible or repressible expression of the recombinant mutated esterases.
Preferred host cells for transfection with vectors containing the mutated sequences for expression of the esterases are unicellular prokaryotic or eukaryotic organisms such as Aspergillus sp. , S. cerevisiae, Hansenula sp. , E. coli or P. pastoris.
Thus the present invention also provides the use of DNA fragments that code for mutant pig liver esterases according to the invention, and optionally for other N- and/or C-terminal domains fused thereto, for cloning into vectors. The invention also provides the use of these vectors for the transformation of cells and the use of such transformed cells or cell cultures for expression of
the recombinant artificial esterases mutated by site directed mutagenesis. The esterases expressed can be isolated e.g. in monomeric form, but it is also possible to prepare enzymatically active, multimeric pig liver esterases and for different units of the multimer to carry different mutations.
Due to the high purity of the recombinant enzymes and the increased specific activity, and in some cases also due to the modified enantioselectivity of the artificially mutated enzymes claimed, they are particularly suitable for organic synthesis. Appropriate substrates for the catalytic conversion are particularly aromatic-aliphatic and aliphatic-aliphatic esters, especially carboxylic acid esters of chiral or prochiral alcohols; the carboxylic acid component preferably contains 2 to 5 carbon atoms and can also be branched. The enzymatic-catalytic resolution of racemic carboxylic acid esters, especially acetates, is characterized by an enhanced R enantioselectivity compared with conversion with rPLE.
The optimal enzymatic activity of recombinant pig liver esterases containing monomeric subunits according to Seq. ID No. 1 is at a pH of between 5 and 10 and preferably of between 7.5 and 9.5, and at a temperature of between 20°C and 90°C, preferably of between 40°C and 80°C and particularly preferably of between 50°C and 65°C and, for enzymes which in positions 129, 133, 134, 195, 236, 237, 290 or 294, additionally of between 50°C and 80°C.
The recombinant enzymes according to the invention can also be used for the resolution of racemic carboxylic acids or for the conversion of prostereogenic compounds,
especially diols or dicarboxylic acids.
Brief description of the Figures:
Fig. 1 shows the pH dependence of the enzymatic activity of the artificial esterases prepared. Fig. 2 shows the temperature dependence. Fig. 3 shows the measured specific activity of the mutant esterases, compared with commercially available esterase extracts and rPLE, towards methyl butyrate, compared with proline-β-naphthylamide. Fig. 4 and 5 show the enantioselectivity of hydrolysis in respect of selected substrates. sing the mutated esterases. Fig. 6 shows the Pichia pastoris expression vector pPICZαA, which contains the following important sequence elements: 1-940 bp 5'AOXl promoter 941-1207 bp α-factor signal sequence 1208-1274 bp multiple cloning site .1275-1304 bp c-myc epitope 1320-1337 bp his-tag 1341-1682 bp 3'AOXl stop site 1683-2094 bp TEFl, S. cerevisiae transcription elongation factor 1 promoter 2095-2162 bp EM7, synthetic prokaryotic promoter 2163-2537 bp Sh ble: Streptoalloteichus hindustanus bleomycin gene (responsible for zeocin resistance) 2538-2855 bp CYC1 stop site (from the S. cerevisiae CYC1 gene) 2866-3539 bp pUC ori (origin of replication for E. coli)
Some Examples are given below to illustrate the present invention:
General :
The shuttle vector pPICZαA from Invitrogen was used for cloning in E. coli and expression in P. pastoris (Fig . 6) .
The strains used for cloning and expression are listed below:
Species Strain Genotype Supplier
Escherichia DH5α supE44Δlacϋl69 (Φ801acZΔ 15) Clontech coli hsdRl7 recAl endAl gyrA96 thi- lrelAl
Escherichia XLIO-Gold TetrΔ (mcrA) 183 Δ (mcrCB-hsdSMR- Stratagene coli mrr)173 endAl supE44 thi-1 recAl gyrA96 relAl lac Hte [ F' proAB lacIqZ ΔM15 TnlO (Tetr) Amy Cam]
Pichia X33 WT Invitrogen pastoris
Tissue preparation, mRNA isolation and cloning of rPLE- cDNA
Cloning of the native pig liver esterase gene in E. coli DH5α, containing the sequence Seq. ID No. 29 coding for rPLE, is carried out as instructed in WO 02/48322 (Examples 1 and 2) , where a sequence coding for a myc epitope and a his-tag (Seq. ID No. 41) is cloned (pPICZα- mPLE*-tag) directly following the rPLE sequence Seq. ID No. 29.
The plasmid DNA was isolated from E. coli DH5α using a plasmid preparation method based on the alkaline lysis of cells and the specific binding of DNA to ion exchanger columns (QIAgen) .
Small amounts (approx. 1 μg) of the DNA were obtained using the Plasmid Miniprep Kit (QIAgen) . After separating the cells from a 3 ml overnight culture by centrifugation (16,000 x g, 1 min, bench centrifuge) and discarding the culture supernatant, isolation of the plasmid continued as instructed by the manufacturer. Larger amounts of plasmid DNA (approx. 10 μg) were prepared using the Plasmid Midiprep Kit (QIAgen) . The starting material used in this case was a 50 ml overnight culture, which was centrifuged for 10 min at 4000 x g and 4°C. The plasmid was isolated from the cell pellet according to the manufacturer' s instructions.
The resulting vector, pPICZα-mPLE*-tag, is used for site directed mutagenesis.
Example 1: Site directed mutagenesis
Mutagenesis is carried out starting from the pPICZα-mPLE*- tag template. The individual mutations coding for the amino acid exchanges are successively inserted according to Table 1, starting in each case from the nucleic acid obtained in the previous mutation step.
Site directed mutagenesis is carried out via a PCR using the QuikChange™ Site Directed Mutagenesis Kit (Stratagene) , observing the process conditions indicated by the manufacturer. The plasmid used, pPI Zα-mPLE* -tag, is amplified, with the aid of Pfu DNA polymerase, with two complementary primers containing the desired mutation, the primers indicated in Table 2 being used for the site directed mutagenesis. The reaction mixture containing the methylated template DNA is then digested for 1 h at 37°-C by the addition of 1 μl of Dpnl in order to degrade the DNA. The QIAquik PCR Purification Kit was used as instructed by the manufacturer (QIAgen) to purify the resulting
artificially mutated DNA sequences after the PCR.
17 amino acid codons in Seq. ID No. 29 were thus modified by the exchange of 21 nucleotides.
The PCR was performed under the following conditions :
Mixture for mutagenesis by the QuikChange method
PCR thermocycler programme for mutagenesis using the QuikChange™ method
The mutated plasmids prepared in this way are transformed in E. coli for further multiplication. The transformation of E. coli (XLIO-Gold or DH5α) with the DNA plasmids obtained is effected by the chemical treatment (Mg2+, DMSO) and heat shock of competent cells. For a maximum of 10 transformations, 50 ml of LB medium 1:100 (0.5% (w/v) of yeast extract, 1% (w/v) of tryptone, 1% (w/v) of NaCl, 2% (w/v) of agar for the agar plates) were inoculated with an E. coli overnight culture and incubated for 2 - 3 hours at 37°C and 200 rpm until the OD60o was 0.4 - 0.7. The
transformation culture was transferred under sterile conditions to a 50 ml Falcon tube and the cells were centrifuged off for 10 min at 1750 x g and 4°C. The cell pellet was resuspended in 2 ml of ice-cold TSS solution (10% (w/v) of PEG 6000, 5% (v/v) of DMSO, 50 mM MgS04, ad 100 ml of LB medium, autoclaved) , distributed into ten 1.5 ml Eppendorf tubes (200 μl each) and then cooled on ice for 5 min. 1-3 μl of the plasmid DNA are then pipetted into each 200 μl aliquot of cell suspension and the transformation mixture is placed on ice. After 20 min the cells are incubated first for 30 - 45 sec at 42°C in a water bath (heat shock) and then, after the addition of 800 μl of LB medium, for 1 h at 37°C. For positive selection of the successfully transformed cells, the cell suspension is streaked on LBLszeocin-agar plates (1% (w/v) of yeast extract, 1% (w/v) of tryptone, 0.5% (w/v) of NaCl, 2% (w/v) of agar for the agar plates) and incubated overnight at 37°C. The mutated plasmid DNA is isolated and worked up as described above for the rPLE vector. The mutated nucleic acid sequence is then sequenced for control purposes.
Part of the mutated plasmid DNA isolated is used to produce the next mutant nucleic acid sequence .
Nine mutated nucleic acid sequences were produced in this way. The coding DNA sequences of the artificial esterases are shown in Seq. ID No. 2 to 10. The mutations are listed in Table 1 on the basis of the amino acid sequences.
Table 1: Mutations in the amino acid sequences of the rPLE variants
Table 2: Sequences of the primers used for the site directed mutagenesis
The nucleotides used to insert the particular mutation are printed in bold.
Example 2: Preparation of the modified enzymes
To prepare larger amounts of linearized plasmid for the transformation of P. pastoris, the plasmids obtained in Example 1 are subjected to a preparative restriction digestion with Pmel . This is done by adding the following buffer solution to 6 μg of plasmid DNA:
dH20 ad 65 μl
Buffer 4 (Promega) 6.5 μl BSA solution 0.65 μl
Pznel 5 μl
The plasmid DNA is digested overnight at 25°C.
The digested plasmid DNA is purified using the QIAquik PCR Purification Kit as instructed by the manufacturer (QIAgen) .
For transformation in P. pastoris, 50 ml of YPD medium (yeast extract-peptone-dextrose medium: 1% (w/v) of yeast extract, 2% (w/v) of peptone, with 10% (v/v) of 10 x D solution added after autoclaving, 2% (w/v) of agar for the agar plates) are cultivated with P. pastoris X33, with shaking at 30°C, until the OD6oo s 1.0. The resulting cell suspension is centrifuged for 10 min at 1500 x g and room temperature, the supernatant is discarded and the cells are washed once with 25 ml of sterile water. The cells are centrifuged off again and taken up in 1 ml of 100 mM LiCl and the cell suspension is then transferred to a microreaction vessel. The cells are sedimented again (max. speed in the bench centrifuge, 15 s) , the supernatant is pipetted off and discarded and the cell pellet is resuspended in 400 μl of 100 mM LiCl. 50 μl portions of the resuspended cells are distributed into 1.5 ml microreaction vessels for further direct processing. Shortly before transformation, the cells are centrifuged off again and the LiCl solution above the cells is removed. For transformation, solutions are pipetted onto the cells in the following order: 240 μl of 50% (w/v) PEG, 36 μl of 1 M LiCl, 25 μl of 2 mg/ml fragmented herring sperm DNA and 5-10 μg of plasmid DNA in 50 μl of sterile water. The cells are completely dissolved by vortexing for 1 min and are subsequently incubated with the plasmid DNA for 30 min at 30°C without
shaking and then for 20-25 min at 42°C (heat shock) . The cells are centrifuged off (3500-6000 x g) , the supernatant is discarded and the pellet is resuspended in 1 ml of YPD and shaken at 30°C. After 1 h and 4 h, 25, 50 and 100 μl portions are plated on YPD agar plates containing 100 μg/ml of zeocin. The agar plates are incubated at 30°C until visible colonies appear (2-3 days) .
For the identification of enzymatically active P. pastoris clones, individual transformed cells are transferred to YPD-zeocin-agar plates. After 24 hours' incubation at 30°C, 100 μl of methanol are added to the cover of the Petri dish every 24 h in order to induce esterase production. After 2-3 days the agar plates are covered with a layer of 10 ml of soft agar (0.5% of agar in water) mixed with 100 μl of α-naphthyl acetate (40 mg/ml in DMF) and 100 μl of Fast Red TR (100 mg/ml in DMSO) . α-Naphthol is produced during hydrolysis and forms a coloured complex with Fast Red. Positive clones are stained brown.
After the identification of enzymatically active clones, 10 ml of BMGY medium (2% (w/v) of peptone, 1% (w/v) of yeast extract, 1% (v/v) of glycerol, with 10% (v/v) of phosphate buffer (1 M, pH 6.0) and 0.2% (v/v) of 500 x B added after autoclaving) are inoculated with the clones in a 50 ml tube. Expression takes place on a small scale at 30°C and 200 rpm. When the OD60o is in the range 2-6 (after approx. 16 h) , the cells are centrifuged off (2000 x g, 5 min, RT) and resuspended in 2 ml of BMMY medium (2% (w/v) of peptone, 1% (w/v) of yeast extract, with 10% (v/v) of phosphate buffer (1 M, pH 6.0), 0.2% (v/v) of 500 x B and 0.5% (v/v) of methanol added after autoclaving). 20 ml of the BMMY medium were inoculated with this suspension in 100 ml flasks (until the OD was approx. 1) . The cultures were incubated at 30°C and 200 rpm and methanol (0.5% of the culture volume) was added every 24 h. Samples are
taken after 0.5, 24 and 48 h and the supernatant containing the expressed enzyme is concentrated by ultrafiltration. This is done using a Centrikons-Plus 20 30K with an Ultracel-PL membrane, which affords a 25- to 100-fold concentration of the esterase solution. The activity of the concentrated supernatant is determined by means of a pNPA assay. After concentration by ultrafiltration, the success of the enzyme expression is checked by native PAGE.
For the expression of larger amounts of enzyme, 100 ml of BMGY medium are inoculated in 500 ml flasks in an overnight culture (3 ml of YPD medium) and incubated at 30°C and 200 rpm. When the OD60o is in the range 2-6 (after approx. 16 h) , the cells are centrifuged off (2000 x g, 5 min, RT) and resuspended in 10 ml of BMMY medium. 200 to 400 ml of the BMMY medium are inoculated with this suspension in 1 1 or 2 1 flasks (until the OD is approx. 1) . The cultures are incubated at 30°C and 200 rpm and methanol (0.5% of the culture volume) is added every 24 h. Cultivation is carried out for 72 hours, after which the cells are centrifuged off and the supernatant is processed further. The activity of the expressed enzymes is determined by means of a pNPA assay (cf. Ex. 3.4 for instructions) .
The results of the pNPA assay are shown in Table 3.
Table 3: Volumetric and specific activity in the supernatant after cultivation and concentration
Example 3: Further characterization of the enzymes
3.1. Determination of the molecular weight
The molecular weight of the enzymes was determined with a
Ferguson Plot (native gel) and a gradient gel (PhastSyste ) :
With the Ferguson Plot, enzyme trimers with a molecular weight of approx. 180 kDa were found preferentially. With the gradient gel (PhastSystem) , in which the semilogarithmic plot of molecular weight against migration distance of the protein shows a linear dependence, a molecular weight of approx. 180 kDa is again found.
3.2. Determination of the isoelectric point
The isoelectric focusing of the enzymes was carried out with a PhastSystem. This was done using two different gels, one with a pH range of 3 - 9 and the other with a pH range of 4 - 6.5. In addition, for the purposes of comparison, all the isoelectric points were also calculated theoretically with a computer program {DNAsis) . The results are listed in Table 4.
Table 4: Results of the pi determination
3.3. Determination of the protein content
The protein content was determined by SDS-PAGE. Different concentrations of BSA were used for the standard curve. The intensity of the protein bands was measured with the computer program NIH-Imager after the gel had been scanned, and this was used to calculate the protein content (Table 5) . Conventional methods (Bradford, BCA test) have proved unsuitable for determining the protein content in the supernatant of a Pichia culture because interfering peptide-containing constituents are present in the medium.
Table 5: Protein contents of the recombinant enzymes
The kinetic data were determined by means of a pNPA assay (concentration range 0.1-5 mM) . Km and Vmax were calculated from a Lineweaver-Burk plot .
Table 6: Kinetic data of the recombinant enzymes
Determination of the esterase activities by means of a pNPA assay
The hydrolysis of p-nitrophenyl acetate (pNPA) is used for the photometric determination of esterase activity. The product formed is p-nitrophenol, which is quantified at a wavelength of 410 nm. The molar extinction coefficient (ε) is pH-dependent and the extinction coefficients determined for different pH values are indicated below.
Extinction coefficients for p-nitrophenol as a function of pH
The reaction rate is measured over one minute at one- second intervals in a 1 ml cuvette at room temperature. The assay solution used consists of 800 μl of sodium phosphate buffer (50 mM, pH 7.5), 100 μl of enzyme solution and 100 μl of substrate solution (pNPA in DMSO, 10 mM for standard measurements or from 0.1 to 100 mM for determinations of enzyme kinetics) . The autohydrolysis of pNPA in buffer is measured in each case as a blank. 1 ϋ of esterase activity corresponds to the amount of enzyme that liberates 1 μl of p-nitrophenol in one minute.
Example 4: pH profiles of the enzymatic activity
The pH profiles were determined by means of a pNPA assay (performed as in Ex. 3) . The activity was measured in phosphate buffer at the appropriate pH values. The profiles are very similar, with the exception of the mutant GEH-PLEe. The highest activity was observed for all the enzymes in the pH range from 8 to 9 (Table 7, Fig. 1).
Table 7: Volumetric and specific activity of the recombinant enzymes at optimum pH
The pH profiles of the recombinant mutant enzymes are represented graphically in Fig. 1. The highest activity was taken as 100% for each enzyme and compared with the other activities.
Example 5: Temperature profiles
The temperature profiles were determined by the pH stat method (carried out as in Ex. 6) at different temperatures. Tributyrin was used as the substrate. 0.5 ϋ (pNPA) of enzyme was used in each case for 30 ml of substrate-containing emulsion. The butyric acid liberated was titrated with 0.01 M NaOH, the pH being kept constant at 7.5 (cf. Table 8, Fig. 2 for results). As the number of mutations increases, so too does the temperature stability of the enzymes: at 70°C rPLE is no longer active and rGEH is at its optimum.
Table 8: Specific activity of the recombinant enzymes at optimum temperature
The temperature profiles of the recombinant mutant enzymes are shown in Fig. 2. The highest activity was taken as 100% for each enzyme and compared with the other activities.
Example 6: Hydrolysis of methyl butyrate and proline-β- naphthylamide
The hydrolysis of methyl butyrate was carried out by the pH stat method, a known amount of enzyme being added to a substrate solution (30 ml) consisting of 2% w/v of gum arabic and 5% v/v of substrate in water, emulsified with a homogenizer. The acid liberated in the reaction is back- titrated with sodium hydroxide solution in order to keep the pH constant. This is done by titrating the liberated butyric acid with 0.01 M NaOH, the pH being kept constant at 7.5 (at 37°C) .
The amount of sodium hydroxide solution consumed is used to calculate the amount of liberated acid that correlates with the activity of the enzyme, 1 ϋ corresponding to the amount of enzyme that liberates 1 μmol of acid per minute. 1 U (from pNPA assay) of enzyme is used in each case for 30 ml of substrate-containing emulsion. The results of the hydrolysis are shown in Table 9.
The hydrolysis of proline-β-naphthylamide is followed photometrically by means of a Fast Garnet assay using 0.5 ϋ (pNPA) of enzyme in each case. β-Naphthylamine liberated in the reaction forms a pink complex with Fast Garnet, which can be detected at 520 nm (ε = 24.03-103 M"1 cm-1) .
The reaction mixture (500 μl) , consisting of Tris/HCl (0.1 M, pH 8.0), 50 μl of substrate solution (0.2 mM in DMSO) and esterase solution (0.4 - 0.5 U (from pNPA assay) ) , is reacted in a thermomixer at 37°C. After 30 min, Fast Garnet solution (1.5 ml, 15% w/v, first dissolved with ethanol and then topped up with 4% Brij solution) is added and the absorption is measured immediately.
1 U of amidase activity corresponds to the amount of enzyme that liberates 1 μmol of β-naphthylamine in one minute .
The results of the hydrolysis are likewise shown in Table 9.
Table 9: Results of the hydrolysis of methyl butyrate and proline-β-naphthylamide (as reference substance) with all the recombinant enzymes
The results of the reactions are also given in Fig. 3, which shows the specific activity of commercially available esterases (PLE Fluka and PLE Chirazyme E2) and recombinant esterases towards methyl butyrate/proline-β- naphthylamide. It is seen that the activity of mutants a to f and h is greater than that of rPLE.
Example 7 : Study of the stereoselectivities of the enzymatic hydrolysis
The hydrolysis is carried out in 1.5 ml reaction vessels in a thermomixer (Eppendorf) at 37°C. 0.5 U of esterase (based on the pNPA test) is used in each case for 1 ml of substrate solution (10 mM in sodium phosphate buffer pH 7.5, 50 mM) . Samples are taken from the reaction mixture at the times indicated in the Tables. The reaction is stopped by extracting the mixture with methylene chloride and drying the organic phase over anhydrous sodium sulfate.
The enantiomeric purity and the conversion are determined by gas chromatography. The results are shown in Tables 10 to 15 and in Fig. 4.
Table 10: Enantioselectivity of different recombinant esterases in the hydrolysis of 1-phenyl-l- ethyl acetate (all the enzymes exhibit an (R) preference)
Table 11: Enantioselectivity of different recombinant esterases in the hydrolysis of 1-phenyl-l- propyl acetate (all the enzymes exhibit an (R) preference)
Table 12: Enantioselectivity of different recombinant esterases in the hydrolysis of l-phenyl-3- butyl acetate
Table 13: Enantioselectivity of different recombinant esterases in the hydrolysis of l-phenyl-2- propyl acetate (all the enzymes exhibit an (S) preference)
Table 14: Enantioselectivity of different recombinant esterases in the hydrolysis of l-phenyl-2- pentyl acetate
Table 15: Enantioselectivity of different recombinant esterases in the hydrolysis of l-phenyl-2- butyl acetate (all the enzymes exhibit an (S) preference)
Fig. 4 is a graphical representation of the results obtained. Above the X-axis: (S) preference, below: (R) preference. Values above the X-axis indicate an S preference of the particular enzyme and values below the X-axis indicate an R preference. The R preference of the esterase was increased, or the S-preference reduced, by the inserted mutations. The mutated enzymes tested are thus particularly suitable for the preparation of R enantiomers.
Example 8 : Further study of the enantioselectivity
To delimit more precisely the mutations which have a particularly strong influence on the enantioselectivity of the hydrolysis, two other artificial mutant enzymes, Seq. ID No. 39 (PLE-GEHal) and 40 (PLE-GEHa2) , were prepared as described in Examples 1 and 2. 1-Phenylethyl acetate (grey) and 1-phenyl-l-ethyl acetate (black) were used as the model substrates and converted enzymatically as described in Example 7. The E value is calculated for a conversion of approx. 50%. The results are illustrated in Fig. 5, which shows that the amino acid exchange in position 77, in particular, is responsible for the increase in enantioselectivity in respect of the R enantiomers.