WO2000037631A1

WO2000037631A1 - Process for the production of enzymes with altered catalytic activity relative to an amino acid substrate and enzymes so produced

Info

Publication number: WO2000037631A1
Application number: PCT/IE1998/000113
Authority: WO
Inventors: Paul Cowper Engel; Xing-Guo Wang
Original assignee: University College Dublin, National University Of Ireland, Dublin
Priority date: 1998-12-22
Filing date: 1998-12-22
Publication date: 2000-06-29
Also published as: AU1780099A

Abstract

A process for the production of an altered microbial enzyme, especially an aminotransferase, which displays catalytic activity towards the L-enantiomer of an amino acid which is not the normal substrate for the precursor enzyme comprises treating a microorganism with a mutagenic agent, growing the microorganism so treated on a medium containing said amino acid as the sole nitrogen source and isolating surviving microorganisms containing enzymes capable of using said amino acid as a substrate. The process involves the use of a random chemical mutagen followed by a biological selection procedure and does not require prior knowledge of the protein structure of the enzyme nor site-directed mutagenesis.

Description

Process for the production of enzymes with altered catalytic- activity relative to an amino acid substrate and enzymes so produced

Technical Field

This invention relates to a process for the production of an altered microbial enzyme which displays catalytic activity towards the L- enantiomer of an amino acid which is not the normal substrate for the precursor enzyme.

Background Art

Enzymes can be altered by chemically modifying their amino acid sequence or, alternatively, by genetically manipulating the gene sequence coding therefor. Of these two approaches genetically manipulating the gene encoding the enzymes so as to cause a specific mutation is the more specific approach. Reference can be made in this regard to the work of Wilks, H.M., et al. (Science (1988); 244, 1542- 1544) who transformed the substrate specificity of a lactate dehydrogenase from Bacillus stearothermophilus for which the preferred substrate is lactate, to that of a malate dehydrogenase, for which malate is the preferred substrate (see also Wilks, H.M., et al. Biochem. Biophys. Res. Comm. (1987); 148, No.1 , 15-23).

There is a requirement in the pharmaceutical industry for a supply of enantiomerically pure amino acids for use, for example, in producing pharmacologically active substances. In particular, there is a requirement for producing enantiomerically pure non-naturally occurring amino acids, mainly neutral amino acids.

These non-naturally occurring amino acids are used, for example in the synthesis of angiotensin converting enzyme (ACE) inhibitors and in the preparation of synthetic peptides.

WO 95/1 1296 describes a multi-step process for directed modification of enzymes, which process can be used as a basis of knowledge-based synthesis of amino acids. This process requires an elucidation of the structure of at least two enzymes from a group of enzymes, comparing the binding pockets of said enzymes and determining the amino acids required to bind a substrate which is not preferred for the unmodified enzyme and using this knowledge to produce a mutation in the gene coding for the enzyme so as to produce a modified enzyme having the desired change in its binding pocket. There is a growing awareness of the limitations of a purely site-directed approach, especially the difficulty of achieving the desired mutation on the first attempt.

The work of Wilks, H.M., et al. {supra) and the invention of WO 95/1 1296 are based on the premise that if one knows and understands a protein structure then one can modify it by "knowledge-based" site- directed mutagenesis.

Another requirement of the pharmaceutical industry is for various oxo or keto acids which are generated by the reversible transamination reaction catalysed by various amino acid transaminases. For example, there is a requirement for 2-oxoglutarate in the treatment of kidney failure where amino acid metabolism is impaired.

Forced or directed evolution of enzymes is known from the work of Clarke, P.H., (( 1972) Soc. Gen. Microbial Symposium No. 24 "Evolution in the Microbial World", pp 183 - 217 The evolution of enzymes for the utilisation of novel substrates.). Clarke worked on the aliphatic amidase of Pseudomonas aeruginosa, an inducible enzyme splitting acetamide to acetate and ammonia. Clarke carried out a series of experiments in which she used forced evolution to produce changes both in the substrate specificity and in the pattern of induction and repression of the enzyme. The selective pressure applied resided in the fact that the various amides supplied represented the major source of both carbon and nitrogen for growth of the organism. By stepwise selective changes Clarke ultimately produced amidases capable of handling amides with 5-6 carbon chains and additionally amides with bulky substituents on the nitrogen atom of the amide group, such as acetanilide.

The general technique of forced or directed evolution has not been applied to enzymes capable of catalysing a stereospecific transformation involving the L-enantiomer of an amino acid.

Apart from the method of WO 95/1 1296, which can be used as a means of producing L-amino acids, other existing methods of producing such L-amino acids in purified form involve fermentation methods and the resolution of racemic mixtures of amino acids. In the case of fermentation methods one can only produce naturally occurring amino acids. Methods for resolving mixtures of amino acids are costly and labour intensive involving high input of effort and potential waste in that only half of the resultant product can be utilised, unless the other half can be recycled.

Disclosure of Invention

The invention provides a process for the production of an altered microbial enzyme which displays catalytic activity towards the L- enantiomer of an amino acid which is not the normal substrate for the precursor enzyme, which process comprises treating a microorganism with a mutagenic agent, growing the microorganism so treated on a medium containing said amino acid as the sole nitrogen source and isolating surviving microorganisms containing enzymes capable of using said amino acid as a substrate.

The altered microbial enzymes produced by the process according to the invention display catalytic activity towards the L-enantiomers only of the relevant amino acids. Thus, the process according to the invention provides an efficient means of producing pure L-amino acids. The altered microbial enzymes produced in accordance with the invention preferably have a variable specificity.

Thus, as indicated above, the process according to the invention allows one to produce enzymes retaining specificity towards the L- enantiomer only of an amino acid and not its D-enantiomer with the attendant advantages.

The process according to the invention does not require prior knowledge of the protein structure nor site-directed mutagenesis. Rather, the invention may involve the use of a random chemical mutagen followed by a biological selection procedure.

Preferably, the enzyme is a transaminase/aminotransferase, hereinafter referred to generally as an aminotransferase. An aminotransferase is an enzyme that catalyses the transfer of an amino group from an amino acid to an oxo acid, thereby giving rise to the opposite pair of oxo acid and amino acid. The reaction is readily reversible, a property which is utilised in accordance with the invention.

An especially preferred precursor aminotransferase is a branched chain or aromatic amino acid aminotransferase.

The amino acid is preferably a neutral amino acid, more particularly a non-naturally occurring amino acid. Non-naturally occurring amino acids which can be produced in accordance with the invention include L-teτt.leucine and homophenylalanine and amino acids of similar size and structure.

However, the amino acid can, of course, be any naturally occurring neutral amino acid such as leucine, valine and the like. However, as these are relatively cheap, their production would not be of particular interest in the context of the invention. It will be appreciated that the present invention provides a practical route for producing any neutral amino acid for which the corresponding keto acid is available more cheaply than the amino acid itself. Suitable mutagens include, for example, sodium nitrite, hydroxylamine and nitrosoguanidine. A particularly preferred mutagen is sodium nitrite.

A preferred micro-organism for use in accordance with the invention is a strain of Escherichia coli .

The reversible nature of the aminotransferase reaction means that one may obtain two desirable products, namely the oxo-acid and the corresponding amino acid. Alternatively, one of the two reactants may be a desirable product. For example, as indicated above there is a commercial need for 2-oxoglutarate. This product is a higher value product than glutamate as such, the amino acid reactant. In general, for the common biological L-amino acids, the corresponding oxo acid is usually a more expensive product to produce (about 4-fold for glutamate, 15-20 fold for leucine, 30-fold for aspartic acid and so on). Conversely in the case of the non-biological amino acids, since they are not available in nature and require a chiral centre that is missing in the keto acid, the situation is reversed: it is the amino acid that is more expensive. Thus, L- tert.leucine is about 1000-fold more expensive than L-glutamate.

Thus, in a typical embodiment of the invention, the 2-oxoacid corresponding to the desired amino acid is used as a substrate in conjunction with L-glutamic acid.

Treatment of the microorganism with the mutagenic agent is preferably carried out in varying concentrations of the mutagen until an optimal concentration is determined. Cells of the microorganism so treated are then collected and washed in a suitable washing medium to remove the mutagenic agent. The cells are then resuspended in a selective medium containing the amino acid which will be the desired product of the transamination reaction once an altered microbial enzyme with the requisite substrate specificity is isolated.

The cells are preferably successively incubated with decreasing concentrations of said amino acid until the lowest concentration of the amino acid on which the mutagen-treated microorganism will grow is identified. Because of the aforementioned limitations of a purely site- directed approach, we adopted the approach of starting as near to the target site as possible, combined with 'traditional' random mutagenesis which allows one to carry out a very large number of experiments simultaneously. With this approach there is a requirement for a means of screening so that an improved mutant can be readily detected when it occurs, as hereinafter described.

The process according to the invention results in the production of an enzyme which displays catalytic activity towards the L-enantiomer of an amino acid. Accordingly, if this enzyme is presented with a racemic mixture of said amino acid it will utilise the L-enantiomer only, enabling one to isolate the D-enantiomer. Accordingly it will be appreciated that one can obtain a pure source of a D-amino acid in accordance with the invention with the attendant advantages, because the production of pure D-amino acids is normally an expensive and painstaking task.

Accordingly, in a further aspect of the invention there is provided a process for the production of the D-enantiomer of an amino acid from a racemic mixture thereof, which comprises incubating the racemic mixture with an altered microbial enzyme as hereinbefore described, and isolating the D-enantiomer of the amino acid.

Preferably, the altered enzyme is an aminotransferase.

The invention also provides an aminotransferase which has been altered relative to a precursor enzyme so that it can catalyse the stereospecific transamination of a range of amino acids which are not the normal or efficient substrate for said precursor enzyme.

The invention also provides a whole cell extract containing an aminotransferase as hereinbefore defined.

Further, the invention provides a suspension of whole cells containing an aminotransferase as hereinbefore defined.

Brief Description of Drawings

Fig 1. is a graph of absorbance v wavelength (nm) for the products converted by TGI, the TGI transformant and the isolate, respectively as described in Example 3;

Fig. 2 is a thin layer chromatogram showing the decrease in L-tert.leucine during the growth of the bacterial cells, the subject of Example 5;

Fig. 3 is a thin layer chromatogram showing the conversion of various oxo acids into their corresponding amino acids by the 1-4 crude extract and resting cells when L-glutamate was used as one of the substrate pair as described in Example 5; Fig. 4 is a thin layer chromatogram which shows the transamination specificity of the initial bacterium as described in Example 6; and

Fig. 5 is a thin layer chromatogram which shows the transamination specificity of the isolate as described in Example 6.

Modes for Carrying Out the Invention

The invention will be further illustrated by the following Examples.

Example 1

Forced Evolution of E. coli

The target amino acid was L-t<?rt-leucine which is a 'bulky' amino acid. The recombinant p^tac plasmid DNA containing the gene of the triple mutant (K89L/A163G/S380A) was extracted from E. coli Q100 host cells by use of Wizard (Wizard is a Trade Mark) DNA miniprep kit obtained from Promega, and then transformed into E. coli TGI using the standard method. This triple mutant was created using the general approach set forth in WO 95/11296. This mutant has a new and quantitatively quite respectable activity with methionine and norleucine. Transformants were screened by growing the cells on Luria Bertani (LB) plates supplemented with 100 μg/ml ampicillin and confirmed by identification of either the size of plasmid DNA on agarose gel or characteristics of the gene product. 50 μl of E. coli TGI transformants growing at 37°C overnight was inoculated into 2 ml LB containing ) 00 μg/ml ampicillin, and then 5 mM, 10 mM, 20 mM, 50 mM and 100 mM NaNθ2 was added respectively into each tube. The bacteria treated with NaNθ2 were incubated at 37°C for 6 hours. The cells were collected by centrifugation at 4°C and washed with TEN (0.1 M NaCl, 10 mM TrisHCl (pH7.4) and 1 mM EDTA (pH8.0)) once. After resuspension in 10 ml of the selective medium (47.7 mM Na2HPθ4, 22 mM KH2PO4, 8.56 mM NaCl, 2 mM

MgS04, 0.1 mM CaCl, 0.6 μM thiamine, 20 mM glucose and 0.5-20 mM L-tert-leucine (as sole nitrogen source)) containing 100 μg/ml ampicillin, the cells were incubated continually at 37°C for 2 weeks. The bacteria in each tube were harvested by centrifugation and transferred into the fresh selective medium containing 10 mM L-teτt-leucine and allowed to grow again under the same conditions. After one week's incubation, the bacteria were transferred into a fresh selective medium with half reduced concentration of L- r/-leucine. This manipulation was repeated each week until the L-teτt-leucine concentration was down to 0.5 mM. Eventually only the cells from treatment with 10 mM NaN02 grew under this condition. The surviving cells were isolated by spreading them on the LB plate supplemented with ampicillin, and the isolates were further investigated. The triple mutant was employed on the grounds that it was already well on the way to the specificity that was required for large hydrophobic amino acid side chains. 7 rt-leucine, however, is bulkier and it was considered that the necessary changes might be further back from the shell of residues immediately involved in the enzyme's active site and might be very difficult to predict with any confidence. We had a selective screen for identifying mutants, namely the ability to use ter eucine as a nitrogen source when the medium was depleted in ammonia. It was uncertain as to whether this would or would not be a viable strategy. However, to our surprise and as described in Examples 2 and 3 we found that the evolved bacteria were not, after all, using their plasmid-encoded, mutated glutamate dehydrogenase to permit growth on tert-leucine as expected, but rather had found a different way of achieving such growth. We had predicted that the route would be that the amino acid would give, via the dehydrogenase reaction, ammonia which could then be taken up to form glutamate and supply the cell's needs by transamination. However, the E. coli effectively short-circuited this approach by finding a transaminase which it could mutate directly so that the glutamate could be made from the nitrogen of the supplied amino acid without going through the intermediary of ammonia. This offers particular advantages, in that there is no requirement for NAD/NADP co-factors.

Example 2

Investigation of activities of NAD+-dependent amino acid dehvdrogenases

The activities of NAD+-linked amino acid dehydrogenases for the adapted bacterial isolates of Example 1 were investigated on a 96 well flexible assay plate using a colour reaction. Each well contained 200 μl of assay solution (0.1 M TrisHCl (pH8.0), 1 mM NAD+ 100 mM amino acid, 0.3 mg/ml iodonitrotetrazolium (INT), 0.03 mg/ml phenazine- ethosulfate (PES)) and 50 μl crude extract. In each case, 1 ml crude extract in 0.1 M potassium phosphate buffer (pH7.0) and 0.1 mM EDTA was prepared by sonication of the cells from 10 ml overnight-grown culture in LB broth. E. coli TGI and the triple mutant transformant were used as controls. The plate was incubated in the dark at room temperature for 30 minutes, and monitored directly by eye.

Activity of NAD+-dependent amino acid dehydrogenases with L- t -leucine was also investigated with a Kontron Uvikon 922 spectrophotometer, thermostatted at 25°C, by recording A340 for 10

minutes in an assay mixture containing 1 mM NAD+ and 100 mM amino acid in 0.1 M TrisHCl (pH8.0). In each assay, 50 μl crude extract was added.

The results are shown in Table 1.

As shown in Table 1 the activities of NAD⁺-dependent amino acid dehydrogenases for 6 isolates (denoted I-, - 1-₆), which were investigated by the colour reaction, show that none of the isolates displays any colour reaction with 12 amino acids, similar toTGl. However, as expected, the TGI transformant showed activities towards Nle and Met owing to the function of the triple mutant enzyme. As to activity with L-t<?rt-leucine at pH8.0, no detectable activity was found spectrophotometrically with any of the 6 isolates, the TGI transformant and TGI . Table 1 Activities of NAD⁺-dependent amino acid dehydrogenases for 6 isolates towards 12 amino acids.

TertL

Leu

He

Nle - +

Glu

Gly

Tyr

Met - +

Val

Thr

Ser

Ala

Nle: L-norleucine; TertL: L-rerf-Ieucine; TG1-3M: TG I transformant containing the gene of triple mutant K89L/A163G/S380A; I: isolates.

'+" means positive colour reaction

'-" presents negative reaction. Example 3

Investigation of activity of branched chain aminotransferase

The isolated bacteria in the selective medium from Example 1 containing 5 mM L-terMeucine, TGI and the triple mutant transformant in LB were grown at 37°C until the number of bacterial cells in 10 ml culture approached to 1x10^0 which was estimated by measuring absorbance at 600 nm (1 ODGOO = 8x10^ cells/ml). An equal amount of cells was respectively harvested by centrifugation at 4°C and resuspended in 1 ml sonication buffer (50 mM potassium phosphate (pH7.0) 1 mM mercaptoethanol and 0.1 mM EDTA). After sonication and centrifugation, crude extracts were obtained by collecting supernatants.

The assay procedure involves selective extraction of the branched chain 2-oxoacid-2,4-dinitrophenylhydrazone into a suitable organic solvent. The procedure used is described below: 3 ml reaction mixture contained 40 μM L-t -leucine, 40 μM 2-oxoglutarate, 200 μM TrisHCl (pH8.6) and 200 μl crude extract. After incubation at 37°C for 10 minutes, the reaction was terminated by the addition of 0.3 % 2,4- dinitrophenylhydrazine in 2 M HC1. The reaction mixture was then left at room temperature for 10 minutes. The solution was extracted with 5 ml toluene by vigorously shaking for 20 seconds and centrifuging at 3000 rpm in MISTRAL (MISTRAL is a Trade Mark) 1000 for 5 minutes. The branched chain 2-oxoacid-2,4-dinitrophenylhydrazone in the upper phase was back extracted into 1.5 ml 10% (w/v) sodium carbonate. After centrifugation, the lower phase was then transferred carefully into a fresh tube. To 1 ml of the sodium carbonate extract, 2 ml IM NaOH was added and mixed. A wavelength scan from A ₅₀ to A ₂₀ was performed on a Kontron Uvikon 922 spectrophotometer.

The results show that when 2-oxoglutarate and L-tert-leucine were used as initial substrates, the absorbance peak (around 420 nm) of product converted by an isolate was significantly different from those (at 375 nm) for TGI and TGI transformant of the triple mutant (see Fig. 1).

Fig. 1 depicts the wavelength scanning of the mixtures of linking 2,4-dinitrophenylhydrazine with the product of converting L-tert-leucine and 2-oxoglutarate respectively by TGI, TGI transformant and I-4. TGI : is E. coli TGI ; 3M: is the original strain containing the gene of the triple mutant (K89L/A163G/S380A); and I-4: is the isolate.

When only 2-oxoglutarate or L-ter eucine was used as substrate, both reaction products showed the same absorbance peaks as that for TGI and the TGI transformant. Without substrate, the absorbance peak also located at 375 nm. Meanwhile, two branched chain 2-oxoacids, 2- oxooisocaproic acid and trimethylpyruvate were used as control. The linkage of two 2-oxoacids with 2,4-dinitrophenylhydrazine gave a value of absorbance peak 440 nm for 2-oxoisocaproic acid and 430 nm for trimethylpyruvate respectively, which were close to that (420 nm) for the product of L-tert-leucine converted by the isolate. Example 4

Reverse-phase high performance liquid chromatography

Reverse-phase high-pressure liquid chromatography was employed to determine the amount of L-tert-leucine consumed during bacterial growth in the medium in Example 1. A Beckman high-pressure liquid chromatography system and a μBondapak-Ci8 (Bondapak is a Trade Mark) column were used. The column was equilibrated with 0.1 % orthophosphoric acid (pH2.5). All solutions and samples were filtered through 0.2 μm filters. A flow rate of 2 ml/min was used which was maintained by a pressure of 1500 psi. All tests were carried out at room temperature. In each case, 20 μl of sample solution was injected. Water containing 0.1% H3PO4 was used as an eluant. All results were recorded with UV detection at A205-

When starting growth of bacteria, 5 mM L-t τt-leucine was added into the selective medium. 1 ml culture solution was then taken out at different stages during growth of bacteria, centrifuged and filtered to eliminate bacterial bodies and debris. Reverse-phase high performance liquid chromatography shows (i) that L-terr-leucine in the medium was eluted at 2.55±0.01 minute which was the same as that for this amino acid in water; and (ii) that when the number of bacteria increased to 6x10^ cells/ml and lxlθ9 cells/ml, amount of L-tert-leucine in the medium decreased 25% and 50% respectively as compared with initial amount of L-rert-leucine in the medium. Example 5

Chromatography on Silica Gel layers

Amino acids, either substrates or products of enzymatic reaction, were separated by use of thin-layer chromatography (TLC) on Silica Gel 60 plates, and detected with ninhydrin spray, n-propanol/water (7:3) was used as solvent. When the solvent front reached a distance of 10 cm, the plate was removed from the chromatography tank, and dried in a 80- 100°C oven for 10 minutes, and then sprayed with ninhydrin reagent. Finally the plate was baked in the same oven until purple colour development occurred.

TLC (see Fig. 2) shows the results which are similar to those obtained in the HPLC assay of Example 4. The amount of L-r<?rt-leucine gradually decreased following the growth of bacteria in the medium.

When the number of cells was up to 2x10^ per millilitre, L-tert-leucine in the medium was undetectable under the conditions used.

In Fig. 2:

A = 50 mM L-tert-leucine in H2O;

B = 5 mM L-tert-leucine in the selective medium as control;

C = the medium when the number of cells was 6x10 cells/ml;

D = the medium when the number of cells was 1 x 1 cells/ml;

E = the medium when the number of cells was up to 1.2x lθ9

cells/ml; and F = the medium when the number of cells was up to 2x10^

cells/ml.

As indicated in Fig. 2, thin-layer chromatography showed a decrease of L- rt-leucine content in the selective medium during the growth of I-4 cells.

The transamination reactions for L-t<?rt-leucine and L-leucine were examined. In 1 ml reaction solution, 5 mM 2-oxoacid, 5 mM amino acid, 200 mM TrisHCl (pH8.6) and 100 μl crude extract (or 20 μl concentrated cell suspendants) were used. After incubation at 37°C for 25 minutes, the samples were centrifuged at top speed for 5 minutes and filtered with a 0.2 μm filter. Finally, the clarified supernatants were used for the assay by TLC. The results are shown in Fig. 3.

In Fig. 3:

Ci ₌ 5 mM L-teτt-leucine in water;

C2 = 50 mM L-glutamate in water;

1 = I-4 crude extract + 2-trimethylpyruvate + L-glutamate;

2 = I-4 crude extract + 2-ketoleucine + L-glutamate;

3 = I-4 crude extract + 2-ketoneopentylglycine + L-glutamate;

4 = I-4 crude extract + L-glutamate;

5 = I-4 crude extract;

6 = I-4 crude extract + 2-ketoneopentylglycine; 7 = I-4 crude extract + 2-ketoleucine;

8 = I-4 crude extract + 2-trimethylpyruvate;

9 = I-4 resting cells + 2-ketoleucine + L-glutamate;

10 = 1-4 resting cells + 2-ketoneopentylglycine + L-glutamate;

and

1 1 = 1-4 resting cells + 2-trimethylpyruvate + L-glutamate.

Fig. 3 shows that effective conversion of 2-ketoleucine, 2- ketoneopentylglycine and 2-trimethylpyruvate into their corresponding amino acids by both I-4 crude extract and resting cells when L-glutamate was used as one of the substrate pair.

Example 6

Transamination reactions of cells

Transamination activities with 20 amino acids were tested using bacterial cells of the isolate. In each case, 5 mM 2-oxoglutarate, 5 mM amino acid, 200 mM TrisHCl (pH8.6) and 20 μl cells (3x10' ' cells/ml) were used, and the total volume was 1 ml. The reactions were performed at 37°C for 25 minutes. 2 μl of the supernant clarified by centrifugation was loaded on Silica Gel 60 plates The results are shown in Figs. 4 and 5 and are summarised in Table 2.

Fig. 4 represents the original strain (TGI transformant) and Fig. 5 the 1-4 cells: In each of Figs. 4 and 5: a = 2-oxoglutarate; b = 2-oxoglutarate + Val; c = 2-oxoglutarate + Norvaline; d = 2-oxoglutarate + Met; e = 2-oxoglutarate + Leu; f = 2-oxoglutarate + Norleucine; g = 2-oxoglutarate + He; h = 2-oxoglutarate + Thr; i = 2-oxoglutarate + Ser; j = 2-oxoglutarate + Ala; k = 2-oxoglutarate + Gly; 1: 2-oxoglutarate + His; m = 2-oxoglutarate + Arg; n = 2-oxoglutarate + Lys; o = 2-oxoglutarate + Cys; p = 2-oxoglutarate + Asp; q = 2-oxoglutarate + Asn; r = 2-oxoglutarate + Tyr; s = 2-oxoglutarate + Trp; t = 2-oxoglutarate + Homophenylalanine; u = 2-oxoglutarate + Phe; and v = L-glutamate. Table 2

Comparison of transamination specificity between the initial bacterium and the isolate

TG I transformant Isolate

Val +

Nva +

Met +

Leu +

Nle +

He +

Thr

Ser

Ala

Gly

His

Arg

Lys

Cys

Asp + +

Asn + +

Tyr +

Homo-Phe - +

Phe +

Claims

Claims :-

1. A process for the production of an altered microbial enzyme which displays catalytic activity towards the L-enantiomer of an amino acid which is not the normal substrate for the precursor enzyme, which process comprises treating a microorganism with a mutagenic agent, growing the microorganism so treated on a medium containing said amino acid as the sole nitrogen source and isolating surviving microorganisms containing enzymes capable of using said amino acid as a substrate.

2. A process according to Claim 1, wherein the enzyme is a transaminase.

3. A process according to Claim 1 or 2, wherein the amino acid is a neutral amino acid.

4. A process according to Claim 1 or 2, wherein the amino acid is a non-naturally occurring amino acid.

5. A process according to Claim 4, wherein 2-oxoacid corresponding to the desired amino acid is used as a substrate in conjunction with L-glutamic acid.

6. A process according to Claim 5, wherein the amino acid is homophenylalanine.

7. A process according to any preceding claim, wherein the microorganism is a strain of Escheήchia coli .

8. A process according to any one of Claims 2-7, wherein the amino acid used as the substrate is produced by reaction of its corresponding 2-oxoacid with L-glutamic acid.

9. A process for the production of the D-enantiomer of an amino acid from a racemic mixture thereof, which comprises incubating the racemic mixture with an altered microbial enzyme obtainable by a process according to any one of Claims 1-8, and isolating the D- enantiomer of the amino acid.

10. A process according to Claim 9, wherein the altered enzyme is an aminotransferase.

1 1. An aminotransferase which has been altered relative to a precursor enzyme so that it can catalyse the stereospecific transamination of a range of amino acids which are not the normal or efficient substrate for said precursor enzyme.

12. A whole cell extract containing an aminotransferase as claimed in Claim 1 1.

13. A suspension of whole cells containing an aminotransferase as claimed in Claim 1 1.

14. A process according to Claim 1 , substantially as hereinbefore described and exemplified.

15. A process according to Claim 9, substantially as hereinbefore described.

16. An aminotransferase according to Claim 1 1 , substantially as hereinbefore described and exemplified.

17. A whole cell extract according to Claim 12, substantially as hereinbefore described and exemplified.

18. A suspension of whole cells according to Claim 13, substantially as hereinbefore described.