WO2000023609A1

WO2000023609A1 - Transaminase biotransformation process employing glutamic acid

Info

Publication number: WO2000023609A1
Application number: PCT/US1999/024322
Authority: WO
Inventors: Ian G. Fotheringham; Tao Li; Richard F. Senkpeil; David Ager
Original assignee: Nsc Technologies Llc
Priority date: 1998-10-19
Filing date: 1999-10-19
Publication date: 2000-04-27

Abstract

This invention provides a process for making an amino acid which comprises a first transaminase catalyzed reaction between a keto acid and glutamic acid under conditions appropriate to produce said amino acid and α-ketoglutarate and a second transaminase catalyzed reaction between said α-ketoglutarate and a compound chosen from the group consisting of lysine, ornithine and mixtures thereof. The reaction of the invention can be performed using cells or extracts of cells of a microorganism transformed with a vector comprising a gene which encodes for a transaminase enzyme.

Description

TITLE

TRANSAMINASE BIOTRANSFORMATION PROCESS EMPLOYING GLUTAMIC ACID

BACKGROUND OF THE INVENTION

Field of the Invention

This invention relates to an improved process for producing amino acids by means of aminotransferase biotransformation. More particularly, this invention relates to an improved process for making amino acid products by an aminotransferase biotransformation process that uses glutamic acid, in particular L-glutamic acid, as the amino donor.

Related Background Art

The process of the invention improves on prior art processes for making natural and unnatural amino acids using transaminase enzymes. Transaminases have been known in the literature for many years. See, Transaminases , Philipp Christen & David E. Metzler, ed. (1985) (John Wiley and Sons, New York) . Briefly, a transaminase reaction requires two substrates, an amino acid and a keto acid. The general scheme of a transaminase catalyzed reaction is shown below.

Keto acid Amino donor Amino acid Keto acid substrate product byproduct

A keto acid substrate is reacted with an amino acid known as the amino donor. The amino group of the amino donor is exchanged with the keto group of the keto acid substrate to form the new amino acid and a keto acid by-product. When this reaction is run commercially only the new amino acid is typically desired and the keto acid by-product is generally discarded. As can be readily seen, the specific keto acid substrate chosen will depend on the desired amino acid product. However, the amino donor can be any amino acid accepted and reacted upon by the enzyme.

The reaction is freely reversible with an equilibrium of approximately 1. The most effective way to increase the yield of product beyond 50% is to remove the keto acid byproduct as it is synthesized. In this way the reverse reaction is prevented and the synthesis of amino acid product is favored. This has never been demonstrated when the amino donor used is L-glutamate because it is difficult to remove the keto acid by-product, alpha ketoglutarate, by enzymatic or chemical conversion in an economical way.

U.S. Patent No. 4,518,692 ("Rozzell I") discloses a method for producing L-amino acids by reacting L-aspartic acid and various 2-keto acids with transaminases. The Rozzell I method uses L-aspartic acid as the amino acid to produce oxaloacetate and describes various methods of decarboxylating oxaloacetate to form pyruvate. However, the pyruvate produced in the Rozzell I method can still act as a keto donor in the reverse process to form alanine. Tokarski et al., Biotechnology Letters, Vol. 10 (1) (1988), pp. 7-10, show that alanine acts as a substrate in transaminase reactions. See also, Transaminases (1985) ; and Amino Acids; Biosynthesis and Genetic Regulation, Klaus M. Herrmann and Ronald L. Somerville ed. (1983) (Addison-Wesley Publishing, Reading MA). Tokarski, et al. studied the use of a transaminase to produce L-2-aminobutyrate from 2-ketobutyrate and alanine. The reference, however, discloses only 25-30% conversion to products, demonstrating that the reverse reaction will prevent attaining even the theoretical limit of 50%. This has long been considered an intrinsic property and a problem of transaminase reactions and is the major reason such enzyme catalyzed reactions have not been exploited more in industrial processes to produce these highly desired amino acid products. The transaminase biotransformation process described in U.S. Patent Application Serial No. 08/858,111 minimizes the production of alanine by the action of an additional enzyme acetolactate synthase. In some cases, however, even low alanine levels introduce difficulties with product isolation and recovery, as L-alanine is very water soluble. The present invention differs from these disclsoures by its use of glutamic acid as the amino donor and by providing an effective enzymatic method to remove the potential substrates of the reverse reaction from the mixture.

Others have attempted to improve the yield of transaminase reactions from 50% to near 100% using additional enzymes in the process. For example, U.S. Patent No. 4,826,766 ("Rozzell II"), discloses a transaminase catalyzed reaction that employs two transaminase enzymes and additional keto acids. In the Rozzell II process, a first transaminase enzyme catalyzes the reaction between a first amino acid and a first keto acid to produce a second amino acid and second keto acid. A second transaminase enzyme then catalyzes a further reaction of the second amino acid and a third keto acid to form the desired amino acid. The two transaminase enzymes are selected such that the first enzyme does not catalyze the second reaction and the second enzyme does not catalyze the first reaction. The Rozzell II method, however, is distinguishable from this invention in that Rozzell II requires additional keto acid and does not disclose the use of glutamic acid as the amino donor or the use of enzymes having lysine aminotrasferase or ornithine aminotransferase activity.

The disclosure of these references are hereby incorporated in their entirety into this specification by reference to describe the state of the art. Thus, a method to increase the yield of amino acids using transaminase enzymes and glutamic acid is desirable. Further, a transaminase biotransformation process that does not use L-aspartate and does not result in the production of L-alanine is also desirable.

SUMMARY OF THE INVENTION

This invention provides a process for making an amino acid which comprises a first transaminase catalyzed reaction between a keto acid and glutamic acid under conditions appropriate to produce said amino acid and α-ketoglutarate and a second transaminase catalyzed reaction between said α-ketoglutarate and a compound chosen from the group consisting of lysine, ornithine and mixtures thereof.

The invention also provides a non-naturally occurring microorganism containing a vector comprising a gene which encodes for a transaminase enzyme.

The amino acids produced by this process are useful by themselves, for example as feed additives, flavor enhancers, sweeteners, and nutritional supplements, or can be used as synthetic intermediates to be further reacted to form useful products, in particular pharmaceuticals. Amino acid products of this process are particularly useful as single enantiomer starting materials for producing chiral pharmaceuticals.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows plas id pTL17 discussed in Example 2.

Figure 2 shows a schematic representation of one example of the process of this invention using L-lysine.

Figure 3 shows a schematic representation of another example of the process of this invention using L- ornithine.

DETAILED DESCRIPTION

The process of the invention is readily understood by reference to Figures 2 and 3 which provide schematic representations of examples of the process of the invention using lysine and ornithine, respectively. The process comprises a first transaminase catalyzed reaction between a keto acid and glutamic acid employing any transaminase enzyme that accepts glutamic acid as a substrate. In a preferred embodiment, the process uses L- glutamic acid as the amino donor. However, it is anticipated that racemic, i.e., D, L-glutamic acid, as well as D-glutamic acid and non-racemic mixtures of D- and L- glutamic acid would also be useful in the process. Transaminases , (1985) discloses various transaminase enzymes that accept glutamic acid as a substrate and those of ordinary skill in the art are capable of readily determining which enzymes are appropriate for use in this - process. In a preferred embodiment the transaminase is the aromatic transaminase, in particular the transaminase of E. coli encoded by the tyrB gene, and the branched chain transaminase, in particular the transaminase of E . coli encoded by the ilvE gene . Preferred embodiments of

T^-amino acid transaminase include the D-amino acid transaminase of Bacillus sphericus , Bacillus licheniformis or Bacillus sp . YM1.

The improvement provided by the invention is the introduction of a second transaminase catalyzed reaction in the process which increases the yield of the desired amino acid product by recycling the keto acid by-product, α-ketoglutarate, back to glutamic acid, thereby driving the production of more of the desired amino acid product. The second transaminase catalyzed reaction recycles the α- ketoglutarata by-product back to glutamic acid while converting the ornithine or lysine to the corresponding semi-aldehyde, which readily undergoes cyclization to form dehydroproline, when ornithine is used, and dehydro- homoproline, when lysine is used. The semialdehyde or cyclized product can be easily separated from the desired amino acid product by methods known to those of ordinary skill in the art.

In a preferred embodiment, the process of the invention further comprises reacting the semialdehyde with a "trapping agent" to facilitate its separation from the desired amino acid product. In this embodiment the semialdehyde does not form the cyclic product but is derivatized with a trapping agent which converts it to an even more stable molecule which cannot be transaminated. Several such trapping agents will be apparent to those of ordinary skill in the art. In a preferred embodiment the trapping agent is sodium bisulfite. In an equally preferred embodiment the trapping agent is an amine.

Amines that can trap the aldehyde product as an i ine are ^" preferred since the imine is typically stable under the conditions under which the process of the invention are run. Examples of amines that are useful as trapping agents for the process of this invention include, but are not limited to, primary amines such as benzylamine and diphenyl propylamine and n-butylamine.

The second transaminase catalyzed reaction employs ornithine or lysine and an enzyme which has either lysine aminotransferase activity or ornithine aminotransferase activity. Some transaminase enzymes are known to have both lysine aminotransferase activity and ornithine aminotransferase activity. In a preferred embodiment, the enzyme used in the second transaminase catalyzed reaction has lysine e-aminotransferase activity. In an equally preferred embodiment, the enzyme used in the second transaminase catalyzed reaction has ornithine δ-aminotransferase activity. Any enzyme possessing both activities would also be useful in the process. Examples of enzymes useful in the second transaminase catalyzed reaction are disclosed in P. K. Mehta, T. I. Hale, and P. Christen (1993) Eur. J. Bioche . 214(2): 549-61, which disclosure is incorporated herein by reference. Examples include, but are not limited to, N-acetyl-L-ornithine amino transferase (EC 2.6.1.11), L-ornithine amino transferase (EC 2.6.1.13), ω-alanine amino transferase (EC 2.6.1.18), GABA amino trans erase (EC 2.6.1.19), and L-Lysine amino transferase (EC 2.6.1.36). In a preferred embodiment the enzyme used in the second transaminase catalyzed reaction is the ornithine aminotransferase of Bacillus subtilis strain 168 (ATCC 33234) which is demonstrated herein to have both ornithine and lysine aminotransferase activity. In each case the preferred co- substrate is 2-ketoglutarate.

The process of the invention can be run in separate steps, but it is preferred that it be run as a "coupled" reaction, where both enzymes are added to a solution of the keto acid, glutamic acid and lysine and/or ornithine. In the coupled reaction the enzymes act in tandem producing the desired amino acid product and recycling the unwanted α-ketoglutarate by-product, theoretically enabling a 100% conversion of keto acid starting material to the desired amino acid product.

The process of this invention can be applied to produce a variety of natural and non-naturally occurring amino acids simply by selecting the appropriate keto acid. U.S. Patent No. 4,518,692 and Transaminases , (1985) disclose a broad range of keto acids which are useful in this invention. As noted above, the choice of keto acid will depend on the desired amino acid product. In a preferred embodiment, the keto acid is 2-ketobutyrate. In a separately preferred embodiment, the keto acid is tri- methyl pyruvate .

In addition to these sources, keto acids can also be prepared from readily available starting materials, including other amino acids. For example, the enzyme threonine deaminase reacts with L-threonine to produce 2- ketobutyrate . The keto acid, thus produced, is then reacted with the glutamic acid substrate according to the process described above to produce L-2-aminobutyrate. L- threonine is an inexpensive starting material available from Archer Daniels Midland (Decatur, IL) and the reaction produces 2-ketobutyrate in essentially 100% yield. See, Amino Acids: Biosynthesis and Genetic Regulation (1983). Additional reactions using various amino acid starting materials to produce various keto acids that are useful in this process are known in the art. See, Gene, (1989) Vol. 76, pp. 255-269 and Gene, (1988) Vol. 63, pp. 245-252. See also, Massad G. , et al., J.Bacteriol. , (1992) Vol. 177, pp. 5878-5883, for a general description of the activity of amino acid deaminase enzymes from Proteus mirabilis . These references are hereby incorporated in their entireties into this specification.

Techniques for Utilizing Enzymes

In the practice of this invention "conditions appropriate" to react the described enzymes with the described substrates are known to those of ordinary skill in the art. For example, cells producing transaminase enzymes and enzymes having ornithine aminotransferase or lysine aminotransferase activity may be contacted with a solution containing the keto acid and amino acid starting materials with the resulting conversion of at least a portion of the keto acid starting material in the reaction mixture to the desired amino acid product. The cells may be permeabilized to facilitate diffusion of the substrates and products into and out of the cells. This permeabilization can be accomplished by treating cells with a low concentration of a surfactant, including but not limited to Tween 80®, Triton X-100®, Nonidet P40®, cetylpyridiniu chloride, deoxycholic acid, hexadecyltrimethylammonium bromide or benzalkonium chloride. Further, organic solvents, including but not limited to N,N-dimethylformamide (DMF) , dimethyl sulfoxide (DMSO) , ethanol or acetone at low concentrations have also been used to increase permeabilization. Enzymes, including the transaminase, ornithine aminotransferase or lysine aminotransferase, may also be added to the starting reaction mixture in the form of cell extracts containing crude, partially purified, or purified enzyme. Cell extracts are prepared by methods known to those skilled in the art which provide for cell disruption and recovery of the enzyme. Cell disruption, can be accomplished by mechanical or non-mechanical means. Most often, for bacterial suspensions mechanical devices such as a French pressure cell, ultrasonication, bead mill or Manton-Gaulin homogenizer is used with the specifics of the method known to those of ordinary skill in the art. See, Scopes, R.K. "Protein Purification", (1982) (Springer-Verlag, New York) . The reaction using the cell extract is then carried out in similar fashion to the whole cell method discussed above. It is understood that "cells" and "extracts of cells" are used interchangeably herein.

The enzyme-containing cells, or extracts thereof or purified enzyme or enzyme fractions, may also be immobilized, if desired. Immobilization methods which may be used in the practice of this invention include well-known methods such as entrapment in polymeric gels, covalent attachment, crosslinking, adsorption, and encapsulation. Some examples of these methods are described by A.M. Klibanov in Science, 219:722-727 (1983) and the references therein and in Methods in Enzymology (1976), Volume 44, (K. Mosbach, editor) which are hereby incorporated by reference. In one method of immobilization disclosed in U.S. Patent No. 5,019,509, a support material containing at least 20% by weight of silica or alumina is contacted with aminoalkyl compound such as an aminoalkyl silane, polyethyleneimine, or a polyalkylamine, followed by activation with glutaralde- hyde. The enzyme-containing solution is then contacted with the activated support to produce an immobilized enzyme composition having transaminase and/or ornithine aminotransferase and/or lysine aminotransferase activity.

Other immobilization supports useful in the practice of this invention include, but are not limited to, porous glass and porous ceramics, bentonite, diatomaceous earth, charcoal Sepharose® and Sepharose® derivatives, cellulose and cellulose derivatives, polyacrylamide and polyacrylamide derivatives, polyazetidine, alginate, carrageenan, and Chro osorb®. Sepharose® (Pharmacia Fine Chemicals, Uppsala Sweden) is a bead-formed gel prepared from agarose. The manufacturer's product literature reports that in its natural state, agarose occurs as part of the complex mixture of charged and neutral polysaccharides referred to as agar. The agarose used to make Sepharose^ is obtained by a purification process which removes the charged polysaccharides to give a gel with only a very small number of residual charged groups. Those of ordinary skill in the art will appreciate that a number of other materials suitable for the immobilization of cells or extracts derived therefrom may also be useful for the immobilization of the enzymes used in the present invention. These supports can be activated, if desired, by techniques well-known in the art.

The coupled reaction to produce a desired amino acid product utilizing cells containing transaminase, ornithine aminotransferase, lysine aminotransferase or compositions comprising extracts derived from said cells, can be carried out by contacting a solution containing a keto acid and glutamic acid with the enzymes under conditions permitting the conversion of at least a portion of the keto acid to the desired amino acid. In the practice of the processes of this invention the cells contact an aqueous solution of the enzymes at a cell load in the range of about 50 mg/ml to about 300 mg/ml. In a preferred embodiment the cell load is about 100 mg/ml. When the invention is practiced using extracts of cells, the extracts are prepared from an amount of cells that would give these cell loads.

The enzymatic reactions of this invention are carried out at temperatures in the range of from about 30°C to about 50°C, and preferably at temperatures ranging from about 37°C to about 45°C. The optimal pH for the reaction ranges from about 6 to about 9, and more preferably from about 7 to about 8, with a pH of 8 being most preferred. The invention also provides a non-naturally occurring microorganism containing a vector comprising a gene which encodes for a transaminase enzyme. For example the vector can be a plaεmid, such as pTL17 described herein which contains the rocD gene which encodes for ornithine aminotransferase. In a preferred embodiment the vector can comprise more than one gene, each of which encodes for a different transaminase enzyme. In an equally preferred embodiment, the vector can comprise genes encoding for the separate transaminase enzymes required in the process of the invention. The vector is then used to transform a microorganism such as a bacterium, for example E . coli .

In a preferred embodiment the composition comprises cells or extracts of cells of E . coli W3110.

Cloning of the transaminase gene can be accomplished by the isolation of a DNA fragment encoding the polypeptide carrying the transaminase activity from a suitable donor microorganism, and incorporating the DNA fragment into a suitable vector, known to those of ordinary skill in the art. Suitable donor microorganisms include all microorganisms carrying a gene encoding a polypeptide which catalyzes either transaminase reaction of the process of this invention. Examples include, but are not limited to, bacteria from the genus Escherichia coli ( "E . coli" ) , including E . coli B, E . coli K12 , E . coli NIHJ and E . coli Tennessee; Proteus vulgaris; Bacterium cadaveris; Azotobacter vinelandii ; Rhizobium leguminosarum; R . trifolii ; or bacterium from the genus Bacillus .

Suitable cloning vectors useful in the practice of this invention normally contain an origin of replication and a selectable marker to maintain the stability of the vector in the host cell and to facilitate the identification of transformants . A description of some methods and materials useful in the cloning of the transaminase gene or genes can be found in Molecular Cloning: A Laboratory Manual [T. Maniatis, E. Fritsch, and J. Sambrook, Cold Spring Harbor Laboratory (1982)] and references therein, which are hereby incorporated by reference.

A DNA fragment or fragments encoding the polypeptide or polypeptides displaying transaminase activity is then expressed. The general strategy for expression of the transaminase gene or genes involves the ligation of the DNA fragment encoding the polypeptide displaying transaminase activity into an expression vector suitable for the desired host cell, many examples of which are well known in the art. Suitable expression vectors are normally characterized by the presence of an origin of replication, a promoter or other transcription enhancing sequences, a ribosome binding site, and a transcriptional termination sequence. The expression vector may also include a gene conferring resistance to an antibiotic as a selectable marker; however, plasmids containing any other gene encoding a protein required by the host microorganism for growth and viability may be used as a selectable marker, if desired. Examples of promoters useful in recombinant DNA constructions include, but are not limited to, tryptophan, alkaline phosphatase, beta-galactosidase, beta-lactamase, and P- promoter systems; and such hybrid promoter systems composed of components or structures from two or more known promoter systems such as tac. While these are the most common promoters used in bacterial host strains, other strains and microbial species amenable to genetic manipulation and other promoters useful in those host strains may also be used.

After the ligation, the resultant vector contains the transaminase gene or genes in operative association with a promoter such that host cells transformed with the vector are capable of directing the production of transaminase. Suitable hosts for transforming with the vector include any microorganism, for example a bacterium, which allows for the expression of the gene encoding transaminase. Examples include, but are not limite.d to, bacteria of the genera Escherichia, Bacillus, Klebsiella, Pseudomonas ,

Salmonella, Proteus, Azotobacter , and Rhizobium .

In a preferred embodiment of this invention, the host strain for expression of the transaminase gene or genes is any strain of E . coli , in particular E . coli W3110, and the promoter system is the modified pheA promoter described in U.S. Patent No. 5,120,837 (Fotheringham et al.) the contents of which are incorporated herein by reference. Expression of the gene is thus accomplished by ligation of the gene into a vector for gene expression in the chosen host strain. The plasmid pP0T3 , described in Example 1 below, contains genes encoding chloramphenicol resistance, allowing the facile identification of transformed cells. Therefore, in an especially preferred embodiment of this invention, the vector used for the expression of the transaminase gene in E . coli is pP0T3 or a plasmid which is derived from pP0T3 , such as pTL17.

This invention thus also provides the novel plasmid pTL17, which is further described in Example 2 below.

If the preferred embodiments pPOT3 or its derivative pTL17 are used, expression of the enzyme is constitutive. pPOT3 contains a heat sensitive promoter, cI857 mutation, derived from the lambda cl repressor carried on the plasmid and induction of the synthesis of transaminase can be accomplished by a temperature shift, for example, raising the fermentation temperature from 30°C to 40°C. Transaminase enzyme would be expected to be produced at levels of from about 5% to about 40% of the total protein in the cell, thus yielding whole cells containing an enzyme composition with very high transaminase activity. Other heat sensitive promoters are known to those of ordinary skill in this art and these can be used in the manner described herein without departing from the scope of the invention. In a separately preferred embodiment, the gene can be incorporated into a .thermostable microorganism to identify mutations in the gene which enhance the thermal stability of the enzyme thus allowing for the operation of this process at elevated temperatures .

It is not critical to this invention whether the expression of the gene results in enzyme produced intracellularly or extracellularly, since the desired product may be conveniently recovered and, if desired, further purified in either case by methods known to those of ordinary skill in the art.

EXPERIMENTAL DETAILS

The invention will now be further illustrated by the following examples, which are not intended, and should not be interpreted, to limit the scope of the invention which is defined in the claims which follow thereafter.

Example 1

Construction of the Expression Vector pPOT3

Plasmid pBR322 was obtained from New England Biolabs (Beverly, MA) . A modified pheA promoter was inserted between unique Hindlll and Sphl sites on pBR322 to construct pIF306. Within the Hindlll to Sphl insert there exists unique BamHI and Bglll sites. The modified pheA promoter was derived from that characterized in co-owned

U.S. Patent No. 5,120,837 to Fotheringham et al. which is incorporated by reference herein, such that the sequence was as follows: Hindlll

AAGCTTTTTTGTTGACAGCGTGAAAACAGTACGGGTATAATACT AAAGTCACAAGGAGGATCCACTATGACATCGGAAAACCCGTTACT GGCGCT Haell

pIF306 was cleaved with the enzymes BamHI and Sphl . The 3.9 kB fragment was isolated and ligated to a similarly cleaved fragment containing the E . coli K12 ilvE gene which was generated by PCR from W3110 chromosome using the following oligonucleotide primers:

5 ' CGC GGA TCC ACT ATG ACC ACG AAG AAA GCT GAT TAC ATT TGG3 '

5 ' CAG CGT GCA 73C TTA TTG ATT AAC TTG ATC TAA CCA GC 3 '

The resulting vector was named pIF307. Plasmid pIF307 was cleaved with enzymes EcoRI and Pstl and the 4.1 kB fragment isolated. This was ligated to a similarly cleaved and purified 982 base pair DNA fragment containing the kanamycin resistance gene from plasmid pLG388 (pLG388 is described in U.S. Patent No. 5,120,837 and Stoker et al Gene 18: 335-341; the contents of both are incorporated herein by reference) . This was generated using PCR with the following oligonucleotide primers:

5 ' CCG GAA TTC ACG TTG TGT CTC AAA ATC TCT GAT3 '

5 ' CCG CTG CAG GCC GTC CCG TCA AGT CAG CGT AAT G3 '

The resulting plasmid was named pIF312. Plasmid pIF312 was cleaved by EcoRI and BamHI and the resulting 4.97 kB fragment was ligated to the phage lambda cI857 gene which was similarly cleaved following isolation by PCR using the Lambda ZapII vector (Stratagene, La Jolla, CA) as template and the following oligonucleotide primers: 5 ' TTTGGATCCTCCTTAGTACATGCAACC3 '

5 ' TTTGAATTCGGATGAAGATTCTTGCTCGATTGT '

The resulting plasmid was named pPT353. This plasmid was then cleaved with Pstl and Eagl and the 3.17kB fragment was isolated. Plasmid pPOTl can then be constructed by ligating this fragment to the 1.95kB Pstl to BspEI fragment of pBR322 and the PCR fragment generated from pLG338 using the oligonucleotides :

5 ' GGC GGC CGA CGC GCT GGG CTA CG3 '

5 ' CCC TCG CAA GCT CGT CCG GAG GCA AAT CGC TGA ATA TTC C3 '

This PCR results in the amplification of a 0.59kB DNA fragment. This fragment is then cleaved with BspEI and Eagl to generate the necessary sticky ends for the tri- molecular ligation to generate pPOTl.

pPOTl was then cleaved with BamHI and Sail and the 4.68 kB fragment isolated. This fragment was ligated to an oligonucleotide linker prepared by annealing the following two oligonucleotides:

5 ' GATCCTAGGTACCGGTGCGGCCGCATGCTGACTGACTGAAGATCCCGGGCGATTCT ACGCCCGGGTTTTTTATG3 '

5 ' TCGACATAAAAAACCCGGGCGTAGAATCGCCCGGGATCTTCAGTCAGTCAGCATGC GGCCCGCACCGGTACCTAG3 '

The resulting plasmid was named pPOT2. This plasmid was cleaved with Xhol and Pstl and the 3.9 kb fragment isolated. This was ligated to a fragment similarly cleaved with Xhol and Pstl containing the cat gene of HSG415 described in U.S. Patent No. 5,345,672

(Fotheringham) , the contents of which is hereby incorporated by reference. The cat gene confers chloramphenicol resistance upon the host strain and can be isolated from plasmid HSG415 using the following oligonucleotide primers:

5 ' GAC CTC GAG GCA CTT TGC GCC GAA TAA ATA CCT GTG3 '

5 ' GAC CTG CAG CAC CAG GCG TTT AAG GGC ACC AAT AAC3 •

The resulting plasmid was named pPOT3.

Example 2

Cloning and Expression of Gene Encoding Ornithine Aminotransferase of B . subtilis 168

A. Preparation of Plasmid PTL17.

B . subtilis strain 168 was obtained from the ATCC and cultured according to instructions. Chromosomal DNA of 168 was prepared using a microbial chromosomal DNA isolation kit (QIAGEN, Chatsworth, CA) . The rocD gene encoding ornithine aminotransferase was isolated by PCR from the chromosomal DNA using a High Fidelity PCR cloning kit (Clontech, Palo Alto, CA) . The oligonucleotide primers used to amplify rocD were as follows

5' -GAC GGA TCC ACT ATG ACA GCT TTA TCT AAA TCC AAA G-3 '

5 ' -CAG GGT ACC TCA TTA TGC GTT TCG CAG CAC GTG-3 '

The following PCR conditions were used:

Initial denaturing at 97 °C for 15 seconds followed by 30 cycles of (1) denaturing at 94 °C for 30 seconds, (2) annealing at 55°C for 15 seconds, and (3) extension at 72 °C for 1 minute 20 seconds. The 30 cycles were followed by incubation at 72 °C for 5 minutes to optimize full length chain formation. This reaction generated a 1.2 kb double stranded DNA fragment. Approximately 0.5 μg of this fragment was isolated by electrophoresis on a 1% agarose gel and recovered by a Gel Extraction kit (QIAGEN, Chatsworth, CA) . The recovered 1.2 kb fragment was digested with the restriction enzymes BamHI and Kpnl using manufacturer ' s recommend conditions (BIOLABS, Beverly, MA), purified using a PCR Purification Kit (QIAGEN, Chatsworth, CA) , and ligated to the 4.8kb fragment of similarly cleaved plasmid pPOT3 which was isolated and purified from a 1% agarose gel. Plasmid pPOT3 allows for temperature controlled expression of genes inserted downstream of the BamHI site. Genes inserted at this location are transcribed from an upstream copy of the lambda P_? promoter region. The promoter is regulated by a temperature sensitive derivative of the lambda cl repressor carried on the plasmid. The repressor becomes unstable when cells carrying pPOT3 are grown at 30°C expression of the inserted gene is repressed. When the temperature is raised to 42 °C the gene is expressed strongly. The ligation of the rocD fragment to the pPOT3 vector was carried out using a Takara ligation kit (Takara Biochemicals, PanVera, Madison, WI) . The ligation was carried out overnight at 4°C. The ligation product was purified using a PCR purification kit and used to transform E . coli strain JA221 (ATCC 33875) cells by electroporation using standard conditions. Immediately following electroporation the cells were incubated for 60 minutes at 30^;C with 0.5 ml of SOC medium. They were then plated on to LB agarose plates containing 10 μg/ml chlormaphenicol and incubated overnight at 30°C.

Six colonies vere isolated from the ligation plates and cultured in LB containing 10 μg/ml chloramphenicol at 30°C for 24 hours. Plasmid DNA was isolated using standard methods and characterized by restriction analysis. An isolate of pP0T3 containing the desired rocD insert was identified and designated pTL17.

B. Expression of the rocD gene in 3110.

Plasmid pTL17 was used to transform E . coli . strain W3110

(ATCC 27325) by electroporation. One isolate designated W3110/pTL17 was purified for gene expression studies. W3110/pTL17 cells were incubated with shaking in 100ml LB containing 10 ug/ml chloramphenicol at 30° C in a IL shake flask. The culture was grown to OD₆₀₀ = 0.6-0.8, at which point the incubation temperature was increased to 42 °C and continued for a further 2 hours. The cells were then recovered by centrifugation and stored at 4° C until used in the biotransformation.

Example 3

Biotransformation using W3110/PTL17

A. Reaction with L-ornithine and 2-ketoglutarate as substrates.

2-ketoglutarate (disodium salt) (9.51g), to a final concentration of 50mM, and L-ornithine hydrochloride (8.43g) to a final concentration of 50mM, were added to 80 ml water in a 250 ml reaction vessel. The pH was adjusted to 8.4. 5 g (wet weight) of W3110/pTL17 cells in 20 ml water was added to the reaction mixture. The reaction temperature was maintained at 37 °C throughout the reaction. The pH of 8.4 was maintained by feeding 1M ammonium hydroxide with a pH stat.

Table 1 shows the analysis of samples taken at various time intervals to measure the formation of L-glutamate in the reaction. L-glutamate in the reaction was measured by enzymatic assay as described in Beutler, H.-O. (1985) in Methods of Enzymatic Analysis (Bergmeyer, H.U. , ed.) 3rd ed. vol. VIII, pp. 369-376.

Table 1

Time L-glut. ^conversion α-ketoglut. (mg/ml) to L-glut

13.6 50.2 68.25 25 55 74.78

B. Reaction with L-ornithine and 2-ketoglutarate as the substrates and bisulfite as the trapping reagent.

2-ketoglutarate (disodium salt) to a final concentration of 50mM, L-ornithine hydrochloride to a final concentration of 50mM, and 16.7 mmol sodium bisulfite were added to 80 ml water in a 250 ml reaction vessel. The pH was adjusted to 8.4. W3110/pTL17 cells (5 g; wet weight) in 20 ml water as added to the reaction mixture. The reaction temperature maintained at 37°C throughout the reaction. The pH of 8.4 was maintained by feeding 1 M ammonium hydroxide with a pH stat. Two portions of sodium bisulfite (1.73g each) were added after 5 hours, and 10 hours of reaction, each in a volume of 10 ml water. The reaction was continued a total of 24 hours.

Table 2 shows the analysis of samples taken at various time intervals to measure the formation of L-glutamate in the reaction. L-glutamate in the reaction was measured by biological assay as described in Beutler H.-O. (1985). Table 2

Time L-glut. %conversion ^-ketoglut. (mg/ml) to L-glut

10 19.6 33.1 23 55 93.47

Example 4

Catalytic Properties of L-Ornithine δ-Aminotransferase With Different Isomers of Lys and Ort

A. Transformation of isomers of ornithine and lysine.

In all reactions, W3110/pTL17 (0.2 g; wet cells) was used as the catalyst, 100 mM α-ketoglutarate was used as the amino acceptor and the formation of glutamate was identified by comparing with standards on TLC (Silica gel

G₂₅₄ developed in Acetonitrile\Methanol\water\formic acid = 7\2\1\0.5). Amino acids are detected with ninhydrin spray followed by heating. The R_f for glutamate is 0.32.

The transformation of isomers of ornithine were carried out in 50 mM Tris-HCl at pH 8.5. In each reaction, 100 mM D- or L-ornithine was included as the amino donor. The transformation of isomers of lysine were carried out in 50 mM Borate buffer at pH 9.5. In each reaction, 100 mM D- or L-lysine was included as the amino donor. After incubation at 37°C for 16 hours, glutamate was found in all reaction mixtures by TLC.

B. The determination of apparent kinetic parameters for different isomers of Ort and Lys.

The apparent kinetic parameters for L-lysine, D-lysine, L-ornithine, and D-ornithine were determined by using them as amino donors in the respective reactions. In addition to the respective amino donors, the standard solution to determine initial rates of L-ornithine-δ-amino transferase contained 10 mM α-ketoglutarate, 0.1 mM pyridoxal-5-phosphate, 3.3 mM o-amino benzaldehyde, and 40 μg of cell free extract of W3110/pTL17 cells in pH 8.5 Tris-HCl in a final volume of 1 ml. The reaction was started by adding the enzyme and then incubated at 37°C for 15 minutes at which time they were stopped by adding 50 μl of 6 N HC1.

The extent of reactions was determined by measuring respective dihydroquinazolinium derivatives at 440 nm from the products when D- or L-ornithines were the amino donor and at 460 n from the products when D- or L-lysines were the amino donor. The kinetic parameters were calculated by Lineweaver-Burk plot.

Table 3

The Apparent Kinetic Parameters for Isomers of Lys and Ort

Amino acids Km (mM) Vmax (μmol/min/mg protein

L-Lys 48.0 0.421

D-Lys 21.9 0.177

L-Ort 5.40 4.92

D-Ort 11.4 0.836

Example 5

Coupled Reaction

The purpose of this example is to show that the enzyme with ornithine and lysine aminotransferase activity which generates glutamate from alpha ketoglutarate can be coupled to a second aminotransferase reaction which uses glutamate as a substrate and produces α-ketoglutarate as a by-product. The resultant "re-cycling" of α-ketoglutarate drives the second reaction to a much greater yield than would be otherwise achievable.

The second reaction in the example is carried out by the branched chain aminotransferase of E.coli encoded by the ilvE gene . The reaction produces the amino acid L- tertiary-leucine from the keto acid substrate tri-methyl pyruvate ("TMPA") using glutamate ("Glu") as the amino donor and producing α-ketoglutarate as the by-product. L- tertiary leucine is an unnatural amino acid of great commercial significance in the synthesis of pharmaceuticals such as peptidomimetics.

A. Preparation of Plasmids.

1. W3110/pTL17. Prepared as described in Example 2. This plasmid carries the rocD gene encoding ornithine/ lysine aminotransferase ("OAT") and carries out the ornithine/α- ketoglutarate reaction.

2. W3110/pIF328. This plasmid carries ilvE gene encoding E . coli branched chain aminotransferase ("BCAT") and carries out the L-tertiary leucine/glutamate reaction. This plasmid was prepared as follows.

Plasmid pIF312 (described in Example 1) was cleaved with EcoRI and BamHI. The vector fragment of 4.97kB was isolated and ligated to a DNA fragment containing the pckA promoter of E . coli K12 which was generated by PCR using E . coli K12 chromosomal DNA as template. The pckA promoter is generated by PCR using the following oligonucleotides:

5 ^■ GACGAATTCACTTTACCGGTTGAATTTGC 3 '

and

5 ^• GACGGATCCTCCTTAGCCAATATGTATTGCC 3 ' The resulting 0.27kB fragment is cleaved with EcoRI and BamHI and ligated to the 4.97kB vector fragment to yield plasmid pIF313.

Plasmid pIF313 was then cleaved by Sphl and BspEI and the 4. IkB vector fragment isolated. This was then ligated to a- DNA fragment containing the Par locus of plasmid pLG338 which was generated by PCR using purified pLG338 DNA as template. Preparation of pLG338 is described in U.S. Patent No. 5,120,837, the contents of which are incorporated herein by reference. The Par locus is generated by PCR using the following oligonucleotides:

GACGCATGCACCATTCCTTGCGGCGGCG 3 ^»

and

5 ' GACTCCGGAGGCAAATCGCTGAATATTCC 3 *

The 0.97 kB fragment is cleaved by Sphl and BspEI and ligated to the 4. IkB vector fragment to yield plasmid pIF328. PIF328 was then used to transform cells of E . coli

K12 W3110 to generate the strain W3110/pIF328.

Preparation of cultures:

Cultures of W3110/pTL17 were prepared as described above in Example 2(B). Cultures of W3110/pIF328 were grown in 100ml LB containing 40mg/ml kanamycin, cells were grown to OD₆₀₀ of 1.0 without thermal induction. Cells were then recovered by centrifugation, washed and finally resuspended in a 5 ml 50mM tris/HCl buffer, pH 7.0, containing 10% glycerol, ImM DTT and ImM EDTA. Cell free extracts were prepared using a French pressure cell at 1,000 PSIG and centrifugation of lysates at 16,000G for thirty minutes. C. Reactions.

1. 40 mM TMPA + 40 mM L-Glu

(W3110/pTL17)

126 μl TMPA, and 85 mg of Glu (sodium salt) were added to 10 ml distilled water, adjusted to pH to 8.5 with 5 N NaOH after all the chemicals were dissolved. Cell free protein extract of W3110/pIF328 ("BCAT") (100 μl, 290 μg protein) was added to the reaction mixture. The final volume was adjusted to 12.5 ml by addition of distilled water. The reaction was carried out with shaking at 37 °C. At 2, 4 and 6 hours during incubation the pH was monitored and adjusted to pH 8.5. Samples were taken and analyzed for L-tertiary leucine at 3 , 5 and 16 hours according to the HPLC conditions provided below.

2. 40 mM TMPA + 40 mM L-Glu +25 mM L-Ort ( 3110/pTL17 and 3110/pIF328)

126 μl TMPA, 85 mg of Glu (sodium salt) , and 43 mg of L- ornithine (chloride) were added to 10 ml water, adjusted to pH 8.5 with 5 N NaOH after all the chemicals were dissolved. Cell free protein extracts of W3110/pTL17

("OAT") (100 μl, 290 μg protein), and W3110/pIF328 (100 μl, 300 μg of protein) were added to the reaction mixture. The final volume was adjusted to 12.5 ml by adding more water. The reaction was carried out with shaking at 37 °C. At 2 , 4 and 6 hours during incubation the pH was monitored and adjusted to pH 8.5. Samples were taken and analyzed for L-tertiary leucine at 3 , 5 and 16 hours according to the HPLC conditions provided below. 3. 40 mM TMPA + 40 mM L-Glu +50 mM L-Ort (W3110/pTL17 and 3110/pIF328)

126 μl TMPA, S5 mg of Glu (sodium salt) , and 86 mg of L- ornithine (chloride) were added to 10 ml water, adjusted to pH 8.5 with 5 N NaOH after all the chemicals were dissolved. Cell free protein extracts of W3110/pTL17 (100 μl, 290 μg protein) , and W3110/pIF328 (100 μl, 300 μg of protein) were added to the reaction mixture. The final volume was adjusted to 12.5 ml by adding more water. The reaction was carried out with shaking at 37 °C. At 2, 4 and 6 hours during incubation the pH was monitored and adjusted to pH 8.5. Samples were taken and analyzed for L-tertiary leucine at , 5 and 16 hours according to the HPLC conditions provided below.

HPLC Monitoring Conditions Type: OPA derivatized, achiral HPLC

Mobile Phase:

Buffer stock solution: 84 mM Na^PO, and 216 mM Na₂HP0₄ in water, made IL at a time.

Mobile phase buffer: 50 mL buffer stock solution diluted to IL with water, adjusted to pH 6.2 with 85% phosphoric acid.

Pump B: 2% Acetonitrile, 98% mobile phase buffer. Pump A: 100% Acetonitrile

Column: Pheno enex Columbus, 5μ, C18, 100A; 250 X 4.6 mm with matching 30 mm guard column.

Flow rate : 1.5 ml/min Detection: 338nm Gradient:

Time (min) Pump A% Pump B%

0 0 100

2 0 100

9 25 75

15 40 60

20 40 60

21 0 100

30 0 100

Run time 30 minutes. Column temperature 40C. Retention Times: Asp 6.4 min

Glu 7.5 min Ala 10.5 min TBG 13.9 min

Injection volume according to the following program:

- Draw 0 μl from water vial

- Draw 3 μl from Borate buffer vial - Draw 1 μl from OPA vial

- Draw 0 μl from water vial

- Draw 2 μl from sample vial

- Draw 0 μl from water vial

- Draw 1 μl from OPA vial - Draw 1 μl from air vial

- Mix 10 μl total in loop for 6 cycles

- Wait 4 minutes

- Inject

Approximate syringe speed:

Draw = 6 μl/min; Mix and inject 42 μl/min

Borate buffer (0.4N): 2.06g Boric acid in 250 L water, adjust to pH 10.4 using KOH.

OPA (Derivatization agent) : o-phthalaldehyde and 3- Mercaptoethanol in borate buffer purchased from Hewlett Packard (California, USA) . The results of the reactions are provided in Table 4

Table 4

Time (hrs) 16

Reaction 1

W3110/pIF328 (mM) 5.7 12.5 12.9 (BCAT)

Reaction 2

W3110/pIF328 (mM) 9.2 20.2 27.8

(BCAT)

+ W3110/pTL17

(OAT)

Reaction 3

W3110/pIF328 (mM) 7.7 21.7 29.2 (BCAT)+ W3110/pTL17 (OAT)

Reaction 1, where only one transaminase enzyme, BCAT, is present, produces 12.9mM L-tertiary leucine in 16 hours. This represents 32% conversion from the 40mM TMP substrate. Reaction 2, where a second transaminase enzyme OAT is also present and ornithine is used at 25mM, produces 27.8mM L-tertiary leucine in 16 hours. This represents 69% conversion from the 40mM TMP substrate. Reaction 3 , where the second transaminase enzyme OAT is also present and ornithine is used at 50mM produces 29.2mM L-tertiary leucine in 16 hours. This represents 73% conversion from the 40mM TMP substrate.

The yield of L-tertiary leucine from TMP according to the method of this invention is therefore increased more than twofold over the single enzyme reaction. No alanine was generated in any of the reactions. Both of these factors contribute to this reaction being improvements over the prior art where transaminations involving L-glutamate do not exceed 50% at best and reactions using aspartate to improve yield generate L-alanine as an undesired byproduct.

Claims

What is claimed is:

1. A process for making an amino acid which comprises a first transaminase catalyzed reaction between a keto acid and glutamic acid under conditions appropriate to produce said amino acid and α-ketoglutarate and a second transaminase catalyzed reaction between said α- ketoglutarate and a compound chosen from the group consisting of lysine, ornithine and mixtures thereof.

2. The process of claim 1 wherein the glutamic acid is L-glutamic acid.

3. The process of claim 1 wherein the transaminase in the first transaminase catalyzed reaction is an aromatic transaminase .

4. The process of claim 3 wherein the transaminase is the transaminase of E . coli encoded by the tyrB gene.

5. The process of claim 1 wherein the transaminase in the first transaminase catalyzed reaction is a branched chain transaminase.

6. The process of claim 5 wherein the transaminase is the transaminase of E . coli encoded by the ilvE gene.

7. The process of claim 1 wherein the transaminase in the first transaminase catalyzed reaction is a D-amino acid transaminase.

8. The process of claim 7 wherein the D-amino acid transaminase is chosen from the group consisting of the D-amino acid transaminase of Bacillus sphericus , Bacillus licheniformis and Bacillus sp . YM1.

9. The process of claim 1 wherein lysine is used in the second transaminase catalyzed reaction.

10. The process of claim 1 wherein ornithine is used in the second transaminase catalyzed reaction.

11. The process of claim 1 further comprising the addition of a trapping agent.

12. The process of claim 11 wherein the trapping agent is sodium bisulfite.

13. The process of claim 11 wherein the trapping agent is an amine.

14. The process of claim 13 wherein the amine is chosen from the group consisting of benzylamine, diphenyl propylamine, n-butylamine, and combinations thereof.

15. The process of claim 1 wherein the second transaminase catalyzed reaction is catalyzed by an enzyme having lysine aminotransferase activity.

16. The process of claim 15 wherein the enzyme has lysine e-aminotransferase activity.

17. The process of claim 1 wherein the second transaminase catalyzed reaction is catalyzed by an enzyme having ornithine aminotransferase activity.

18. The process of claim 17 wherein the enzyme has ornithine δ-aminotransferase activity.

19. The process of claim 1 the second transaminase catalyzed reaction is catalyzed by an enzyme having both lysine aminotransferase activity and ornithine aminotransferase activity.

20. The process of claim 1 wherein the second transaminase catalyzed reaction is catalyzed by an enzyme chosen from the group consisting of N-acetyl-L-ornithine amino transferase (EC 2.6.1.11), L-ornithine amino transferase (EC 2.6.1.13), ω-alanine amino transferase (EC 2.6.1.18), GABA amino transferase (EC 2.6.1.19), and L-Lysine amino transferase (EC 2.6.1.36).

21. The process of claim 1 wherein the second transaminase catalyzed reaction is catalyzed by ornithine aminotransferase of Bacillus subtilis strain 168 (ATCC

33234) .

22. The process of claim 1 wherein the transaminase from the first transaminase catalyzed reaction is supplied by cells producing transaminase enzymes.

23. The process of claim 1 wherein the transaminase from the second transaminase catalyzed reaction is supplied by cells producing ornithine aminotransferase or lysine aminotransferase enzymes.

24. The process of claim 1 wherein the transaminase from the first transaminase catalyzed reaction is supplied by extracts of cells producing transaminase enzymes.

25. The process of claim 1 wherein the transaminase from the second transaminase catalyzed reaction is supplied by extracts of cells producing enzymes having ornithine aminotransferase or lysine aminotransferase activity.

26. A non-naturally occurring microorganism containing a vector comprising a gene which encodes for a transaminase enzyme .

27. The microorganism of claim 26 wherein the vector is a plasmid.

28. The microorganism of claim 27 wherein the plasmid contains a gene encoding a polypeptide carrying transaminase activity.

29. The microorganism of claim 26 wherein the gene encoding the polypeptide carrying transaminase activity is - derived from bacteria chosen from the group consisting of from the genus Escherichia coli ; Pseudomonas ; Salmonella ;

Proteus vulgaris ; Bacterium cadaveris ; Azotobacter vinelandii ; Rhizobium leguminosarum ; R . trifolii ; or bacterium from the genus Bacillus .

30. The microorganism of claim 29 wherein the gene encoding the polypeptide carrying transaminase activity is derived from bacteria chosen from the group consisting of E . coli B, E . coli K12 , E . coli NIHJ and E . coli

Tennessee.

31. The microorganism of claim 26 wherein the gene encodes for ornithine aminotransferase.

32. The microorganism of claim 31 wherein the gene is rocD.

33. The microorganism of claim 32 wherein the rocD gene is obtained from B . subtilis strain 168 (ATCC 33234).

34. The microorganism of claim 26 wherein the gene encodes for lysine aminotransferase.

35. The microorganism of claim 26 wherein the gene is the tyrB gene of E . coli .

36. The microorganism of claim 26 wherein the gene is the ilvE gene of E . coli .

37. The microorganism of claim 26 wherein the gene encodes for the D-amino acid transaminase chosen from the group consisting of the D-amino acid transaminase of Bacillus sphericus , Bacillus licheniformis and Bacillus sp . YM1.

38. The microorganism of claim 27 wherein the plasmid is pTL17.

39. Plasmid pTL17.

40. A process for making an amino acid which comprises a first transaminase catalyzed reaction between a keto acid and glutamic acid under conditions appropriate to produce said amino acid and α-ketoglutarate and a second transaminase catalyzed reaction between said α- ketoglutarate and a compound chosen from the group consisting of lysine, ornithine and mixtures thereof, wherein the second transaminase catalyzed reaction is catalyzed by enzymes produced by the microorganism of claim 38.