CA1341352C - Cloning and utilization of aminotransferase genes - Google Patents

Abstract

This invention relates to a method of increasing the rate of transamination of a susceptible substrate by the gene product of the aspC, tyrB or ilvE gene comprising: a) inserting a plasmid containing aspC, tyrB or ilvE gene into a microorganism; b) expressing the aspC, tyrB or ilvE gene in the microorganism resulting in the synthesis of an active aminotransferase; c) catalyzing the transamination of the susceptible substrate by the aminotransferase. The method is useful, for example, where the substrate is phenylpyruvic acid and the amino acid is L-phenylalanine.

Description

Bac)~ground of the Invent10I1 Field of the Tm~e~Wion This invention involves the construction and use of genetically engineered plasmids. 'These plasmids are constructed using restriction endonucleases to ~~ontain the panes coding for the amipotransferase enzymes used during the synthesis of amino acids. These plasmids promote the synthesis of aminotransferases in bacterial cells. This synthesis allows increased yields of aminotransferase enzymes and the elimination of the aminotransferase reaction as a rate-limiting step in amino acid biosynthesis.
Description of the Prior Art The transmination of amino acid precursor molecules by aminotransferase enzymes is reviewed by Umbarger (Ann.
Rev. Biochemistry, 47:533-606,1978. Umbarger at p. 534 indicates he is reviewing principally the gram-negative bacteria Escherichia coli and Salmonella typhimurium, but that their pattern of amino acid synthesis and regulation is not always the universal pattern.
Aminotransferase and transaminase are used as ec;uivalent terms in this document. (resent aminotransfera:>e nomenclature defining the transaminases A, B and C and the evolution of this nomenclature is discussed by Umbarger at p. 550-552 and p. 581-582. To avoid the potential confusion of the transaminase A, B, C designation we will use the genetic terminology of as~C, tyrB and ilvE gene products as defined by Umbarger at p. 582. The aspC gene product v.il1 catalyse the t:ansamination of amino acid precursors to produce aspartate, glutamate, phenylalanine and tyrosine. The tyzB gene product will catalyze the transamination of amino acid precusors to produce pl:enylalanine, tyrosine, glutamate, aspartate and leucine. The ilvE gene product will catalyze the transamination of the amino acid precusors to produce isoleucine, valine, leucine, phenylalanine, glutamate and alanine.
The location of the as C, tyrB and ilvE genes on the genetic map of E. col.i K12 strain is disclosed and discussed by Bachmann et al. (Microbological Reviews, 44:
1-56, b3arch 1980). The aspC gene located at E. coli K12 map position 20 minutes codes for an enzyme here called aspartate aminotransferase and given the Enzyme Commission Number E.C. 2.6.1.1 (Bachmann et al, p. a).
The tyrB gene located at E. coli 1'12 map poaition 91 minutes codes for an enzyme called tyrosine or aromatic aminotransferase and given the Enzyme Commission Number E.C. 2.6.1.5 (Bachmann, et a1. p. 21). The ilvE gene located at E coli f~12 map position 84 minutes codes for an enzyme here called branched-chain-amino-acid aminotransferase and given Enzyme Commission Number E.C.

2.6.1.42 (Bachmann et al. p. 10).

1 341 ~5 2 Amino acid transamination by a~C and t~rB gene products are disclosed by Gelfand et al., (J. Bacteri-ology, 130:429-440, April 1.977) to be responsible for essentially all the aminotransferase activity required for the de novo biosynthesis of tyrosine, phenylala-nine, and aspartate in E. coli K12. However, the pres-ence of the ilvE gene product alone can reverse a phenylalanine requirement indicating it has the ability to transaminate a precursor or phenylalanine.
Transducing phage lambda has been used to carry DNA
fragments for E. coli in the as~C region (Christensen and Pedersen Mol. Gen Genet., 181: 548-551, 1981). The as~C region in bacteriophage lambda was used as a source of the gene for the ribosomal protein S1. Their ;31 gene was closed onto plasmids for research purposes. 'Phe as~C
gene, however, was not cloned into a plasmid for i:he production of transaminases. The as~.nC region was used as a marker to facilitate isolation of a transduc.ing phage in a tyrosine auxotroph lacking the as~C and ~B genes.
This transducing bacteriophage was not suitable for producing high levels of aminotransferases.
Summary of the Invention Applicants invention describes the isolation of the genes coding for the aminotransferases used during the bacterial synthesis of amino acids and the construction of plasmids containing these aminotransferase gene_>s.

~ 341 35 2 Applicant's plasmids contain one or more copies of the as~C, t~rB or ilvE genes. Applicant's invention des-cribes a use for these aminotransferase genes in the synthesis of increased enzyme concentrations of the aminotransferase in bacterial cells. Applicant's in-vention describes a method for the production of in-creased amounts of aminotransferases. Applicant's invention describes a method of improved amino acid synthesis where the aminotransferase catalyzed reaction is rate limiting. Applicants invention describes a method for the increased synthesis of L-phenylalanine wherein the aminotransferase catalyzed transamination of phenylpyruvic acid to phenylalanine is rate-limiting.
Applicant's invention describes the nucleotide sequence of the t~rB gene encoding the enzyme aromatic amino-transferase. Applicant's invention describes the amino acid sequence of the aromatic aminotransferase encoded by the gene t r~rB.
The aminotransferase genes, as~C, t~rB, ilvE, code for the synthesis of the aminotransferases which <:atalyze the transamination of the carbonyl precursors of amino acids. ABC gene codes for the transaminase A (aspar-tate aminotransferase) (EC2.6.1.1) and is catal.yt_Lcally active during the synthesis of aspartate, glutamate, phenylalanine and tyrosine. The t~rB gene codes for tyrosine or aromatic aminotransferase (EC 2.6.1.5) and is catalytically active during the synthesis of phe-nylalanine, tyrosine, glutamate, aspartate and leu-cine. The ilvE gene codes for transaminase B and is '1 341 35 2 catalytically active during the synthesis of isoleucine, valine, leuc.ine, phenylalanine and glutamate. Then the transamination reaction in bacterial amino acid synthesis becomes rate-limiting, the presence of these plasmid-borne aminotransferase genes allows for the synthesis of additional aminotransferases. Aminotransferase synthesis sufficient to overcome the rate limitation result, in an increased rate of amino acid biosynthesis.
These plasmids are used to increase the total quan-tity of aminotransferase present in a cell and it may be used as follows. The transamination reaction may be used within the cell, in a cell extract, or the extract can be utilized for further purification of the specific am-inotransferase and then used in a purified state.
Objects of the Invention It is an object of this invention to construct an extra-chromosomal element, such as a plasmid, that contains one or more genes capable of coding for t:he synthesis of an aminotransferase.
Another object of the invention is to construct a plasmid containing one or more of the following amino-transferase genes. as~C, tyrB or ilvE.
Yet another object of the invention is a method for the synthesis of aminotransferases.
Still another object of the invention is a method for the synthesis of an aminotransferase that is the product of the as~C, t~rB or ilvE gene.

Yet another object of the invention is; a method for producing amino acids using microorganisms wherein the amino acid transamination reaction becomes rate-limiting, the improvement comprising the introduction into the microorganism's chromosome additional genes coding for an amino transferase capable of transaminating the keto acid precursors of the amino acids.
", 5 a ~~+~ ~5 2 Another object of the invention is a method of in-creasing the rate of transamination of receptive amino acid precursor molecules accomplished by the inse>rtion of a plasmid containing the aspC, tyrB or :i.lvE gene into a microorganism, followed by the synthesis of a physio-logically effective concentration of aminotransfe~rase resulting in an increased rate of amino acid synthesis.
Still another object of the invention is a method of increasing the transamination of phenylpyruvic acid by the aminotransferases coded for by plasmids containing the as~C, t~rB or ilvE genes and expressed in a micro-organism.
Yet another object of the invention is a method of increasing the synthesis of an amino acid in a cell where the aminotransferase is a rate-limiting enzyme which is accomplished by the introduction and expression of a plasmid containing one or more aminotransferase genes.
Still another object of the invention is the in-creased synthesis of phenylalanine from phenylpyruvic acid by the introduction and expression of a plasmid containing an aminotransferase gene in a microorganism wherein the wild type transamination of phenylpyruvic acid is a rate-limiting step in the synthesis of phe-nylalanine.
Another object of the invention is the increased synthesis of tyrosine from p-hydroxyphenylpyruvate by the introduction and expression of a plasmid containing an aminotransferase gene in a microorganism wherein the wild type transaminase of p-hydroxyphenylpyruvic acid is a rate-limiting step in the synthesis of tyrosine.
Still yet another object of tile invention is the introduction and expression of a plasmid containing an aminotransferase gene in an enteric microorganism, one such microorganism being Escherichia coli.
Description of the Drawings Figure 1 describes the biosynthetic pathway for the synthesis of the aromatic amino acids tyrosine, phenylalanine and tryptophan.
Figure 2 is a native protein gel electropherogram showing the presence or absence of the aminotra;nsferases encoded by tyrB, as~C or ilvE genes in various bacterial strains.
Figure 3 describes the DNA nucleotide sequence of the EcoRI-BamHI fragment carrying the t~rrB gene.
Figure 4 describes the amino acid sequence of the aminotransferase encoded for by the tyrB gene.
Figure 5 illustrates an E. coli Flowchart showing intermediate strains and processes resulting in deposited strain HW159.
Figure 6 illustrates the modification of they tyrB+
plasmid.
Figure 7 illustrates the restriction map of the plasmid carrying the aspC gene.

Detailed Description of the Specific Embodiments The aminotransferases or tranaminases are a family of enzymes that covalently transfer an amino group from a donor molecule, such as aspartate or glutamate, to an acceptor c-keto-acid preci.irsor of an amino acid, such as oxaloacetic acid. The reaction is freely rover:>ible.
Therefore, the aminotransferases can be ut.ilizecl both in the synthesis and the degradation of amino acidw~.
Amino acid biosyntl-aesis often requires tran~;amination as the final step in biosynthesis. An example e~f such a process is the transamination of phenylpyruvic acid to produce phenylalanine with the simultaneous conversion of glutamate to a-ketoglutarate. This is shown as step 10 in ~igure 1. The biosynthesis of the aromatic amino acids phenylalanine and tyrosine both require the action of an aminotransferase as the final step. Phenylpyruvate is converted to phenylalanine by an aminotransferase that transfers an amino group from glutamate (figure 1, step 10). Similarly, an amino group from glutamate is transferred to 4-hydroxyphenylpyruvate to produce tyrosine (figure 1, step 12). Each aminotransferase has.preferred substrates for the transamination reaction. However, residual activity is present on other substrates permitting each aminotransferase to participate in more than one amino acid's biosynthesis. The E. coli transaminase encoded by the aspC gene is active on _g_ aspartate, glutamate, phenylalanine and tyrosine. The E. coli transaminase encoded by the ~rB gene is active on phenylalanine, tyrosine. glutamate, aspartate and leucine. The E. coli transaminase encoded by the ilvE
gene is active on isoleucine, valine, leucine, ph~~nyl-alanine and glutamic acid (Umbarger at p. 582).
It is recognized that other pathways exist fo:r synthesizing tyrosine and phenylalanine in other microorganisms. One such pathway is found :in cyanobacteria and P. aeruginosa and .involves the conversion of prephenate directly to pretyrosine.
Pretyrosine is then a substrate for conversion directly to tyrosine or to phenylalanine. It is recognized that analogous cloned transaminase genes for the appropriate transaminase are similarly useful in increasing the reaction synthesizing pretyrosine when there are excess levels of prephenate. This transaminase would result in increasing the rate of synthesis of tyrosine and phenyl-alanine.
The enzymatic steps :in the synthesis of the aromatic amino acids are shown in figure l, steps 1-20. Normally each enzymatic step is under regulation by controlling the synthesis of the enzyme and/or by allost.erical.ly regulating the rate of enzymatic catalysis. Wild type microorganisms normally do not over-produce any one amino acid due to these controls. Mutant strains in which the controls on the synthesis of a particular amino acid have been inactivated or by-passed may be isolated and in _g_ general these strains tend to over-produce that particu-lar amino acid. In the case of L-phe, the regulation is of necessity complex, since part of its synthetic pathway is shared with the other aromatic amino acids L-tyrosine (L-tyr) and L-tryptophane (L-trp) (see Figure 1 of the accompanying drawings).
In E. coli, as in many microorganisms, the first step in the pathway is catalyzed by the three DAHP synthetase isozymes each of which is sensitive to one of the three final products. In addition, the first steps in the pathway after the branch point at chorismate are also sensitive to feed-back inhibition. Anthranilate syn-thetase is inhibited by L-trp and the two chorismate mutase isozymes are inhibited by L-phe and L-tyr. The synthesis of a number of the enzymes concerned is also subject to regulation. The tyrosine repressor, the product of the t~rrR gene, controls expression of DAHP
synthetase (L-tyr) and chorismate mutase (L-tyr). The genes for these last two enzymes together constitute the tyrosine operon. The tryptophan repressor, the product of the tr~R gene, controls expression of all the enzymes of the try operon which convert chorismate to L-trp.
This regulator molecule also controls the expression of DAHP synthetase (L-trp). In addition, expression of the pheA gene which codes for the chorismate mutase (L-phe) and the expression of the try operon are rendered sens-hive to the level of L-phe. The trp operon .is rendered sensitive to the level of L-trp by an attenuator/lcsader peptide control system.
-lo-Mutant strains that over-produce amino acids are often isolated by selecting for resistance to appropriatf=
amino acid analogues. In the case of L-phe production, one may isolate strains derepressed for DAHP (L-tyr) and DAHP (L-trp) expression by selecting 3-fluoro-tyrosine and 5-methyl-tryptophan resistant mutants which generally have inactive tyrR and tr~.R repressors. Mutants in which the chorismate mutase (L-phe) is resistant to L-pile may be isolated by selecting for 2-thienyl-alanine re;sist-ance. A DAHP synthetase (L-phe) that is resistant to feed-back inhibition may be isolated by first construct-ing a mutant strain lacking the other two DAHP synthetase isozymes. In such a strain, the synthesis of L-tyr and L-trp is sensitive to the amount of L-phe in the medium.
Mutants able to grow in medium containing L-phe, but lacking L-tyr or L-trp, usually contain feed-back resis-tant forms of DAHP synthetase (L-phe).
Now, a combination of these approaches will result in strains that over-produce L-phe. Since the synthetic pathway is largely shared with that of L-tyr and L-trp, attention must be given to the prevention of L-tyr and L-trp build-up if one is interested in the optimisation of L-phe synthesis. One simple expedient is to incorporate a t~rA mutation into an L-phe over-producing strain so as to prevent L-tyr accumulation and to avoid the unwanted diversion of precursors to L-phe. Similarly. a deletion of the entire try operon would prevent accumulation of L-trp. Such mutants would, of course, have to be cultured 'I 341 35 2 in media that included these two amino acids so ass to permit growth.
The combination of approaches outlined above results in a strain de-regulated for the synthesis of L-phe which accumulates that amino acid in significant amounts. Now, the rate limiting steps in a biosynthetic pathway are usually those at which control is exerted. In a de-regulated strain, it is likely that a new step will be-come rate-limiting. Thus, to increase the production of L-phe still further it is necessary to identify the new rate-limiting step and to increase the activity of the relevant enzyme. This may be done in one of three ways:
(a) by isolating mutant strains in which the relevant enzyme is present in greater amounts;
(b) by isolating mutant strains in which the relevant enzyme has a higher specific activity;
(c) achieving the same end as in (a), but by cloning the gene for the enzyme of interest onto a suitable multi-copy plasmid and introducing this into -the L-phe over-producing strain.
If such an enzyme is not normally subject to regu:Lation, isolating mutants of the first two classes may be dif-ficult. The best way to achieve the increase in activ-ity is to clone the gene for the relevant enzyme.
This is achieved by ligating DNA from a suitable do-nor strain carrying the gene for the enzyme one washes to clone to a suitably prepared DNA from an appropriate vector. One can then transform a mutant strain 1<~cking '1341352 the activity of the newly cloned gene. The problem experimentally is to isolate the gene. If the lack of this activity bestows a recognizable phenotype on this recipient strain, such as auxotrophy for a particular amino acid, then one can simply select for clones carrying the gene of interest by virtue of the ability of such clones to complement the lesion in the recipient strain.
As an alternative, the gene could be cloned fir~~t into a low copy number plasmid or into a suitable bacteriophage vector and subsequently sub-cloned into the desired multi-copy plasmid. Alternatively a strain carrying a suitable prophage near the gene of interest can be constructed and a specialized transducing phage carrying the desired gene isolated by inducing the lysogenic donor strain and using the resultant lysate to transduce the recipient to prototrophy. The gene of interest can then be sub-cloned from the DNA of this specialized transducing phage such as lambda.
Should the mutation of the gene of interest not confer l0 any readily identifiable phenotype, such as auxotrophy, then the cloning of such a gene can be effected by cloning nearby genes for which a selection exists and then screening these clones for the presence of the gene of interest by assaying for the product of the desired gene after introducing the clones into a recipient deficient in that enzyme. Such an approach is facilitated by using methods that allow the cloning of large DNA fragments, such as those involving bacterio-phages or cosmids. If no -.13-readily selectable marker is known to be in the vicinity of the desired gene, then it may be possible to isolate a strain carrying the insertion of a readily selectable transposon, such as Tn 10, which encodes tetracycline resistance, near the gene of interest. Large DNA
fragments contalnlng this marker can then be cloned and the resultant phaae or cesmids screened for the presence 'of the gene of interest. The gene could then be sub-cloned into a suitable high copy number plasmid.
lp The following are the strain of E. coli used in the examples. Beside each strain designation is a summary of the bacterial genotype and source. The relationship of intermediate strains of E. coli leading to the production of the auxotroph HW159 which is both asnC and tr~rB
is shown in Figure 5. The process which converted one strain to another is also summarized on the Flowchart in Figure 5.
Strain HW159, the aspC' and tyrB mutant was deposited in The American Type Culture Collection, 12301 20 Parklawn Drive, Rockville, MD 20852, U.S.A. and has the number designation ATCC 39260.

Table 1 STR.~1IN LIST
Strain Genotype Source HW22 metE M. Edtrards BH82bsag recA,'\.in~m 334 CIQ57 b2 red 3 EamlSJ. Burke Sam7 BT3B2690 recA;'?,imm =34 CI857 b2 red 3 DamlS J. 3urke Sam?

E107 thrl leuB6 thr.~,6 thil dnaB107 deoClB. Bachmann lacYl tonA21 rpsL,67 SupE4a DG44 hsdS thil lacYl galfi2 xyl5 mtll proA2B. Bachmann argE3 hisG4 hppT29 aspCl3 rpsL31 tsx33 supE44 recB21 recC22 sbc815 MC1061 araD139(ara-leu)de17G97 M. Casadaban (lacIPO~Y)de174 galU gall: hsdR rpsL

DG30 thil proA2 argE3 hisG4 lacYl JalF~2 B. Bachmann aral4 xyl5 mtll rpsL31 tsx323 supE44 recB27 recC22 sbcBlS h sdS hppT29 ilvEl2 tyrB509 aspCl3 ES430 thi ma1B29 relAl spoil B. Bachmann HW157 araD139(ara-1eu)de17b97 This work (lacIPOZY)de174 galU galK

hsdR rpsL aspCl3 HW159 araD139(ara-leu)de17697 This work (lacIPOZY)de174 galU galK (deposited) hsdR rpsL aspCl3 tyrB507 HW225 araD139(ara-leu)de17697 This work (lacIPOZY)de174 galU galK

hsdR rpsL aspCl3 tyrB507 recA srl::TnlO

HW519 Prototroph W3110 Commercially j0 Available EXAMPLES
Construction and Use of Transaminase Plasmids For the purpose of exemplification, a novel strain of E coli that over-produces L-phe has been constructed.
This was achieved by a mufti-step process that involved systematically de-regulating the L-phe synthetic pathway.
In such de-regulated strains, it leas been found that the final step, namely the transamination of PPA to L-phe, becomes rate-limiting during fermentation and the PPA
accumulates with detrimental effect. As a further example, therefore, the gene for the tyrosine (aromatic) amino-transferase from E coli has been cloned onto a mufti-copy plasmid. Introduction of this plasmid into the L-phe over-producing strain reduces this build-up of PPA
and increases the efficiency of the fermentation. As an additional example, the gene for the aspartate aminotransferase of E coli has been cloned onto a high copy number plasmid. This enzyme also has L-phe aminotransferase activity and incorporation of this plasmid in the L-phe over-producing strain also enhances the efficiency of the process.
Strains carrying plasmids with tyrB or aspC have added utility in that they over-produce the relevant enzyme to a considerable extent. They are therefore a novel and useful source of these enzymes.
The tyrE or as C genes are also useful on a plasmid, in a suitable genetic background, by providing a selective pressure for maintenance of that plasmid. This selective pressure has advantages over conventional antibiotic selection methods.
Example 1 Preparation of Plasmid DNA
Plasmid DNA was prepared as follows (adapted from the method of J. Burke and D. Ish-Horowitz) (nucleic Acids Research 9: 2989-2998 (1981)) The desired bacterial strain was grown to saturation in 5m1 of L-Broth plus suitable antibiotics. The culture was transferred to a 15m1 corex tube and the cells deposited by centrifugation at 10,000 RPM for 1 minute. The supernatant wa.s decanted and the pellet resuspended in 500 ul of fresh solution I, (50 mP9 D-glucose, 25 MM Tris pH8.0, 1G mM EDTA and 2 mg Lysozyme ml-1) and then incubated at room temperature for 5 minutes. The cells were lysed by the addition of 1 ml fresh solution II (containing 0.2 D9 NaOH and lj; SDS).
This was mixed in gently and then incubated on ice for 5 minutes. 750 ul of cold solution III (to 29 ml. of glacial acetic acid add water to about 60 ml adjust the pH
to 4.8 with 10 m k:OH and make up to 100 ml) was then added and mixed in thoroughly. After a further 5 minutes on ice, the precipitated chromosonal DNA was pells:ted by centrifugation at 10,000 rmp for 10 minutes. '.Che supernatant was decanted to a fresh 15 ml core;K tube and 3.75 ml of Ethanol was added. The tube was keI?t on i<:e for 15 minutes. The DNA was pelleted by centrifugation at 10,000 rpm for 10 minutes. The supernatant was then poured off and the pellet thoroughly resuspended in lOmM
Tris (ph8.0), lmMEDTA, 20 ul of 3 M sodium acetate pH7.0 was added and tl:e DNA transferred to an Eppendorf micro-test tube. The DNA was extracted twice with 200 ul of ultra-pare phenol that contained 0.1°0 8-hydroxy quinol ine and lad been pr~:=-enuilibrated twice ac_fainst 1 m Tris pH8.0 and once against 100 mf~7 Tris 10 mM EDTA pH8Ø
The residual phenol was removed by four extractions with 1 ml of diethyl ether. The DNA was then precipitated by the addition of 500 ul ethanol and incubation in a _ dry-ice; ethanol bath for 10 minutes. The DNA wa,s pelleted by centrifugation at 10,000 rpm for 10 minutes. The supernatant was discarded and the pellet resuspe~nded in 500 ul of 70;o ethanol in water (v/v). The DNA was re-pelleLed by centrifugation at 10,000 rpm for 10 minutes, the supernatant taas discarded and the ~>ellet containing the DNA was dried in a vacuum dessicator for 30 minutes. The DNA was redissolved finally in 50 ul of 10 mM Tris 1 mM EDTA and stored at -20°C_ 2n This preparation cJenerally gives about 5 ug of plasmid DNA when the strains used are derived from MC1061. The preparations also contain considerable amounts of RNA and so <31i tine restriction digest=s described in this patent also include 100 ug ml-~1 RNASE
A that has been pre-treated at 90° for 15 minute's to destroy DNAses. The preparation can also be scaled up to cope with plasmid isolation from 50 ml cultures.

Example 2 (a) Cloning of the tyr B gene Isolation of cosmid clones carrying t~rB anc: DNA from the E coli strain HW 2'? (metE) was prepared according to the method of Marmur (PNAS 46:453, 1960). A:Liquots of this DNA were partially digested with the restriction endonuclease Sau3A as follows: The DNA (80 ug) was made up to 400 ul in Sau3A buffer containing 50 mM
NaCl, 6 mM Tris-HC1 (tris (hydroxymethyl) aminomethane l~ hydrochloride) pH 7.5, 5 mM MgCl2, 100 ug/ml gelatin.
The DNA solution was then split into 8 x 50 ul aliquots. Sau3A (New England Biolabs) was then serially diluted (2-fold steps) in Sau3A dilution buffer containing 50 mM I:C1, 10 mM Tris-HC1 pH 7.4, 0.1 mM EDTA
(ethylene diamine tetra acetic acid, disodium salt), 1 mM
dithiothreitol (DTT), 200 ug/ml bovine serum albumin (BSA) and 50% v/v glycerol. 5 pl of each of seven serial dilutions along with S yl of undiluted enzyme ' were then added to the DNA solution. The amount of enzyme 20 added ranged from 2.5 to 0.02 u. The digestion was continued for 1 hour after which time the reactions were stopped by a five minute incubation at 65°C. The tubes were then cooled on ic:e.
Two yl aliquots of each of the digestions were taken and were subjected to electrophoresis on a 0.7;o w/v agarose gel in TBE (5U mM tris-HC1 pH 8.3, 50 mM boric acid and 1 mM EDTA) along with suitable markers. After staining with ethidium bromide (1 ug/ml) for 15 minutes, the gel was observed under 254 nm illumination to visualise the DNA. In this way, the sample giv:Lng the greatest amount of material in the 45 kb sire range could be estimated.
The DNA from this sample was then ligated to DNA from the cosmid vector pHC79 t:~hich had been prepared as described below:
Two 25 ug aliquots of pHC 79 DNA were completely digested one o;ith the res~ri,~.ti~n e=ndonuclease Hind III
(EC 3.2.12.21) and the other with the restriction °ndonuclease SalI (EC 3.1.23.37). The Hind III
digestion was carried out in 100 ul of Hind III
digestion buffer containing 60 mM NaCl, 7 rnM MgCl2, 7 mM
Tris-HC1 pH 7.4, 100 ug;ml gelatin and 15 units of Hind III (New England Biolabs). Incubation was at 37°C for one hour. The enzyme was then inactivated by heating to 65°C for five minutes. The SalI digestion was carried out in 100 ul of SalI digestion buffer containing 150 mM NaCl, 6 mM Tris-HC1 pF3 7.9, 6 mM MgCl2, 6 mM
2-mercaptoethanol, 100 yg/m1 gelatin and 25 units of SalI (New England Biolabs). Incubation was at 37°C for one hour. The enzyme was then inactivated by heating to 65°C for five minutes.
The two aliquots were pooled and digested with 10 units calf intestinal phosphatase (PL Labs) at 37°C for 2 hours to remove S' phosphates. The pooled pHC 79 was then adjusted to 0.3 M sodium acetate pH 7.0 and phenol extracted twice, followed by four ether extractions. The DNA was precipitated with 2.5 volumes of ethanol, washed with 70% v/v ethanol, dried and finally re-.,uspende~' in 10 mM Tris-HC1, 1 mM EDTA to a final concentration of 500 vg/ml. The ligation reactions contained 1 ug of prepared pHC 79 and 1 ug of Sau3A digested E coli DDIA
in S ul of ligation buffer (10 mM Tris-HC1 pH 7.9, lOmM
L~lgCi2, 20 nnLQ dithiothreitol, 25 ug/rnl gelatin and 1 rnM
ATP). The reaction mixture was incubated at 37°C for 5 minutes, cooled and the ligat:ion initiated :~y the addition of T4 DNA lipase (EC 6.5..1.1).
to These reactions were continued for 4 hours at 15°C.
Four u1 aliquots of the ligation mixtures were then subjected to in vitro packaging into bacterio-phage a particles. Strains BHB 2688 and B1-3B 2690 were inoculated from NZY agar plates into 50 ml of NZY broth (containing i0 g NZamire (from F?umko Sheffield), 5 g yeast extract and 2.5 g NaC1/litre) and incubated with aeration at 30°C
until the E600 reached 0.3. The cultures were then raised to 45°C by immersion in a 60°C water bath and incubated at 45°C for 20 minutes without aeration. The 20 two cultures were then vigorously aerated at 37°C for 3 hours. The two cultures were pooled, chilled to 4°C and harvested by centrifugation at 7000 rpm for 2 minutes.
The pellet was washed ence in 100 ml of cold M9 minimal medium (per liter: 6g Na2 HPO 4, 3g KH2P04, 0.5g NaCl, 1.0 NH4C1; adjusted to pH7 with 8 N NaOH, after autoclaving, sterile glucose added to 0.2% ta/V, CaCl2 added to 0.1 mM and Mg S04 to 1 mM). The pellet: was washed in 5 ml cold complementation buffer (40mM Tris-HC1 '1 341 35 2 pH 8.0, 10 mM spermidine-HC1, 10 mM putrescine-HC1, 0.1%
V;'V 2-mercaptoethanol <~nd 7°o V/V dimethyl sulfoxide) and pell~ted at 5000 rpm for 30 seconds. 'fhe cells were finally resuspended in 0.5 ml cold complementa'~ion buffer and dispe:lsed into 20 ill :i'_iq~_iots in micro test tubes.
These were tro~ei: immediately in liquid nitrogen and stored at -SO°C.
For the packaging reaction, one 20 ul packaging mix was removed from the liqmid nitrogen and to it was immediately added 1 ul of 30 mM ATP. The mix was then placed on ice for 2 mivnutes. The 4 ul_ aliquots of ligated DNA were then added and mixed thoroughly. This was then incubated at 37°C for 30 minutes. After 30 minutes, 1 ul of 1 mg,/ml DNAase (Worthington) w<3s added and mia:ed in until the sample had lost its viscosity. To this packaged DNA preparation was then added 20a ul of the strain E 107 which ilad been grown to ail E600 °' about 1.0 in tryptone maltose broth. MgS04 was also added to a final concentration of 10 mM. The packaged cosmids were absorbed for 30 minutes at 30°C after which time 500 ul of L-broth (Luria broth: 1% W/V
bacto-tryptone, 0.5% W/V bacto-yeast extract, 0.5% W/V
NaCl, 1.2% W/V glucose, pH7) was added and the incubation continued for a further 30 minutes at 30°C. Th.e cells were then plated on L-agar containing 100 yg/ml carbenicillin (a-carboxy benzyl penicillin) and incubated at 30°C for 24 hours.

Carbenicillin resistant colonies were then picked to L-agar carbenicillin plates and incubated overnight at ~42°C. Strain E107 (3) carries the mutation dnaB which renders it unable to grow at 42°C. Thus, only those colonies v:~hicl: had acquired a cosmid clone carrying dnaB from the wild-type donor should be able to grow at this te:rperature. In this way, 8 co.~mid clones carrying dnaB were isolated. Since dnaB maps close to tyrB, the gene for the aromatic transaminase, and since cosmid cloning results in plasmids containing large inserts of donor DNA, it was reasonable to assume that some of the cosmid clones would also carry tyrB. That some of the clones did indeed carry t~rB was shown in 'the following way.
Cosmid DNA from each of the dnaB clones was purified and packaged into bacteriophage a particles as described above. The packaged cosmids were then introduced into the transaminase deficient strain DG 44 which carries the mutations tyrB and asDC. These two mutations together 10 bestocv a Tyr Asp phenotype on DG 44. Introduction of an intact tyrB gene would be expected to restore the ability of this strain to grow in the absence of tyrosine and aspartate. Direct screening of clones for tyrB is not easy in this strain, since it will not maintain plasmids derived from colEl, such as pHC 79. After introducing the cosmid clones into DG 44, therefore, 2 hours were allowed for recombination between the ~rB
gene on the cosmid and the mutant allelle on the bacterial -l3-chromosome. The cells were then washed and played on minimal medium lacking tyrosine and aspartate to screen for the occurrence of t~,~_B+ ~wecombinants. Four of the original dnaB clones gave tvrB recombinants in this screen and mast therefore carry at least some o:E the ~~rB
gene.
(b) Sub-cloning of restriction fracaments from cosmid dnaB
clone The ~rB cosmid clones were of limited utility owing to the large size thereof which resulted in a low copy number. Therefore, restriction fragments carrying the ty~rB gene were sub-cloned into the multi-copy p:lasmid pATl53, (see fer example, Twigg, A.J., and Shenatt, D., (1980), Nature, 283, 216-218) as follows: Two ug aliquots of dna8 cosmid number 5 which carries tyrB was digested in 20 ul reaction with the following restriction endonucleases BamHI, HindIII, SalI, SphI, ClaI, EcoRI and B~lII.
In each case, the incubation was at 3'7°C for one hour ,.20. and .the.. enzyme w.as inactivated by heating to 65°C for five minutes. For the BamHI digestion, the reaction medium contained 150 mM NaCl, 6 mM Tris-HC1 pH 7.9, 6 ;mM MgCl2, 100 ug/ml gelatin and 4 units of BamHI (New England Biolabs). Three (3) units of HindIII (New England Biolabs) were used in 60 mM NaCl, 7 mM MgCl2, 7 mM
Tris-HC1 pH 7.4 and 100 ug/ml gelatin. In the case of SalI, 5 units of enzyme obtained from New England Biolabs were used in 150 mB9 NaCl, 6 mM Tris-HC1 pH 7.9, 6 mM

;~1gC12, 7 mM 2-mercapteethanoi and 100 ug;'ml c~~latin.
For SohI, the reaction contained 50 mM NaCl, 6 mM
Tris-HC1 pH 7.4, 6 mM I~igCl2, 10 mff 2-mercaptoet:hano:L, 100 ug;iml gelatin and 1 unit of the Sr~hI enzyme (New England Biolabs) . The r~~~act~on rniat:ure fo:- i.laI
dic_restion corresponded to that for :?in.iIII, ciig~=stion, eYCept that the enzyme u:.ed was 3 units of CIaI obtained from Boehringer Biochenicals. In the case of EcoRI 4._'i units of EcoRI (New Ennland Biolabs) were used in 100 mM
Tris-TIC1 pH 7.5, 50 mM 1'!aC.l, 5 MM MgCl3 and 100 ugjml gelatin. Lastly, the reaction medium for B~1II digestion contained 60 mM NaCl, 10 rnM Tris-HC1 pH 7.6, 10 mM
f~7gC12, 10 mM 2-mercaptoethanol and 100 ug/ml gelatin.
These digested DNA preparations were then subjected to electrcphoresis on a horizontal l;o w/v low gelling temperature (LGT) agarose gel in TEE for 6 hours at 5 V;~cm. The DNA fragments were visualised by staining the gel for 15 minutes in a 1 ug;ml solution of eth:Ldium bromide, followed by observation under a 366 nm UV light source. Iuidividual bands were excised and stored at 4"C.
One (1) ug aliquots of pAT153 DNA were digested with the same set of restriction endonucleases in 20 ul reactions, except for )III (which does not cut pAT153).
Cosmid III fragments were sub-cloned into BamHI-cut pATl53 (BamHI and BglII give the same 5' overhangs).
After digestion, the linearized vector was digested with 1 unit of calf intestinal phosphatase (PL Labs) for 1 hour, to remove 5' phosphates and prevent recircularization, incubated at 70°C for 10 minutes to inactivate the enzyme and then also run on a 1°o w/v LGT agarose gel as described for the cosmid fragments. The band cor:-esponding to plasmid linearized by each enzyme was then excised from the gel as described above.
The cosmid fragments were th~_n ligated to the appropriately-cut vector as follows. The gel slices were melted by incubation at 65°C for 10 minutes and. then cooled to 37°C. The melted gel slices were all about 100 ul. Two u1 of vector fragment and 8 ul o.f a particular cosmid frac~memt were then added to 9~0 ul of pre-warmed 1.25 x ligation buffer (62.5 mM Trig;-HC1 pH

7.8, 12.5 mP9 MgCl2, 25 mM dithiothreitol, 1.25 mM ATP
and 62.5 mgjml gelatin) and mixed thoroughly. Ligase was added and the reaction continued at 15°C overnight. The ligazed samples were t=hen re-melted at 65°C for 5 minutes, cooled to room temperature and added to 200 ul of competent cells of the E coli strain HW 8? which had been prepared as follows. An overnight culture. of HW 87 LO in L-broth was diluted 1:50 into 50 ml of fresh pre-warmed L-broth and incubated at 37°C with good aeration until the E600 reached 0.6. Th<~ cells were then pelleted by centrifugation at 10,000 rpm for 5 seconds and resuspended in 25 ml of cold 50 mM CaCl2. The cells were .Left on ice for 10 minutes after which they were re-pe:Lleted as above and re-suspended in 2 ml of cold 50 mM C;~C12.
After a further 10 minute incubation at 0°C, the cells were competent for transformation.

After addition of the ligated DNA, the cells were incubated at 0°C for a further 10 minutes, heat-shocked at 37°C for 2 minutes and finally mil:ed with 750 u:L of pre-warmed L-broth. The cells were incubated for 30 minutes at 37°C to allow phenotypic e::~_~ession of the plasmid encoded ~-lactamase gene before plating suitable aliquots onto L-agar plates containing 200 ug,%m:L
carbenicillin. The plates were incubated at 37°C
overnight. Colonies containing recombinant plasmids were identified by the sensitivity thereof to tetracvycline in the case of fragments cloned at the HindIII, SalI, SphI, BamHI and ClaI sites of pAT 153.
Recombinants isolated using the restriction endonuclease EcoRI were identified by examining the si<:e of the plasmids in individual colonies. 'This was performed as follows. Potential recombinaait colonies were patched onto L-agar plates containing 200 yg/ml carbenicillin. These were incubated overnight at 37°C.
The bacteria from about 1 cm2 of each patch were re-suspended in lytic mix COIltaining 10 mM Tris-HC1 pH 8, 1 mM EDTA, 1% w/v sodium dodecyl sulfate (SDS), 2% w/v Ficoll and 0.5% W/V bromphenol blue (Sigma chemicals) :100 ug/ml RNAse. This was then incubated at 65°C for 30 minutes. Each sample was then vortex-mixed vigorously for seconds. 50 ul aliquots of each preparation were then loaded onto a 1% w/v agarose gel and subjected to electrophoresis for 4 hours at 10 V/cm. The plasmid bands were stained with ethidium bromide as described. above and visualized under _27_ 254 nm UV illuminatlOIl. Recombinant plasmids were identified by the reduced mobility thereof compared to non-recorzbinant controls. Using these methods, a number of recombinant plasmids carrying various fragments from the oricJinal cosmid dnaB clone ~,~ez.-e isolated. '.Co facilitate the screening of these plasmi<is for those that carried tl:e intact t;,rrB gene, a new strain carrying tvrB and as_pC mutations was constructed.
Example 3 Construction of transaminase deficient strain.
It was decided to move the already characterised mutations in DG 44 into a background known to allow maintainance of the plasmids (that of strain MC 1061).
Bacteriophage P1 was grown on a mixed culture of strain HW
22 carrying random insertion of the tetracycline resistant transposon Tn 10 prepared as follows: A saturated culture of HV22 was grown overnight in lambda x'M broth (10 g bacto tryptone, 2.5 g NaCl and 0.2w/v maltose per litre).
This was diluted x 1j100 into 100 ml of fresh pre-warmed 2p lambda YM broth and incubated at 37°C until the OD600 reached 0.6. The cells were deposited by centrifugation at 10,000 rmp for five seconds and resuspended in 5 ml of lambda YM broth. NK 55, a bacteriophage lambda derivative carrying the tetracycline transposon TnlO, was then added to a multiplicity of infection of 0.2. The bacteriophage was absorbed for 45 minutes at 37°C. Then 200 ml aliquots were spread onto L-agar plates containing tetracycline (20 ug/ml) and sodium pyrophosphate (2.5 mM). The plates _2$_ were incubated at 37°C for 24 hours. Approximately 5,000 TetR ~:,lonies wire pooled by washing the colonies off the plates using a total of 5 ml of L-broth. Tluis was then diluted into 50 ml. of L-broth containing 20 ug;ml tetracycline and grown overnight at 37°C. 'The T'n10 pool was stored at -20°C after adding sterile glycerol to bring the concentration to 20~o w/v.
In order to grow bact~riophage Pl on t:he Tnl.O pool, a 200 ml aliauot of this glycerol stock ',aas diluted with 5 lU ml of L-broth including 10 ugjml tetracycline an,d incubated overnight with shaking at 37°C. To 0.2 ml of this overnight culture was added 2 x 106 plaque forming units (pfu) of phage P1 clear and 10 ul of 50 mlvl CaCl2. The cells were incubated at 37°C for five minu:.es and then added to 5 ml of L-broth + 2.5 mM CaCl2 pre-warmed to 37°C. The cells were then incubat=ed at 37°C
with vigorous aeration for four hours. The cell debris was removed by centrifugation at 10,000 rpm for five minutes and the phage-~~ontaining supernate stored over 2p . , . ~hlor~oform~at a~~C:' . . , This supernate was used to transduce DG 30 'to permit growth on minimal medium lacking tyrosine and aspartate.
(The minimal medium used was that of Vogel and :Bonner; 0.2 g/1 MgS04.7H20, 2 g/1 citric acid, 10 g/1 1:2H P02, 3.5 g/1 NaH2P04 .4H20 and 10 mM feCl2. After autoclaving, glucose was added to a final concentration of 0.2% w/v along with necessary supplements of amino acids at 50 ug/ml. The Tyr+ Asp+ transductants were then replica-plated to the same medium including 10 ugjml tetracycline to detect those transductants that had simultaneously become T'etP. A number of these TetR
+ +
Tyr Asp recombinants were purified and used to prepare nev; P1 ly_::;t:es. F:ach of tl~~~se eight °1 preparations ;:as then usec_1 to transduce DG 30 to TetR.
From each experiment, 50 TetR tramm:uctants wire picked to minimal medium lacking tyrosine and aspartate to test for linkage between the site of the Tn 10 insertion involved and as'oC or tyrB. From each of these experiments one TetR colony which remained t~B asyDC
was purified and again used to prepare a P1 lysate. These lysates were used to transduce I~7C 1061 to TetR. Cell extracts from these transductants were then run on native polyacrylamide gels and stained for L-phe transa.minase activity as described by Gelfand.
The cell extracts were prepared by growing t:he desired transductants overnight in 10 ml of L-broth at 3~7°C. The cells were harvested by cetrifugation at ,10,000 rpm for LO fifteen seconds and re-suspended in 0.5-1.0 ml of sonication buffer depending upon the size of the. pellet..
The sonication buffer used was 25 mid Tris-HC1, <:5 mM
KH2P04 pH 6.9, 0.2 mM pyr:idoxal phosphate, 0.5 nrM
dithiothreitol, 0.2 mM ED'rA and 10% v/v glyceroT~. The cell suspensions were transferred to Eppendorf rnicro test tubes and kept on ice. Each suspension was sonicated with a Davies sonicator using four five-second burst:a at setting 2. The sonicated suspensions were then cetrifuged at 15,000 rpm for five minutes at 4°C to remove cell debris and the supernatants transferred to a fresh micro test tube. The cell extracts were then kept on ice until they could be loaded on a gel or stored at =20°C.
The native gel electrophoresis was performed as follows. The gel consisted of 8.5 °o w,/v acrylamide, 0.227;0 vtiv bis-acrylarnide in 375 mL4 Tris-HC1 pH 8.3. This acrylamide stock was de-gassed and polymerized by the addition of 0.05°a w;/v ammonium persulfate and 0.016% w/v TEP7ED. After polymeri:;ation, the gels were pre-run in 37.5 mT~1 Tris-HC1 pH 8.3 for at least one hour at: 4°C.
Prior to loading the samples, the buffer was changed to running buffer consisting of 76.7 mM glycine, 1 mM
Tris-HC1 pH 8.3. About 50 ug protein from each cell extract was then loaded and subjected to electrophoresis at 10 v/cm for from four to six hours at 4°C. The protein concentration in the cell extracts was determined using the Bio-Rad protein reagent method. The gels were stained for L-phe transaminase activity as follows. The gel was washed briefly in 100 mM phosphate buffer pH 7..'> and then . ~. ..3s~_ immersed in 500,m1~of fresh staining solution containing 12.5 mM a-ketoglutarate, 0.2 mM pyridoxal phosphate, 0.6 mM nitro blue tetrazol.ium, 0.2 mM phenazine metlzosulphate, mM L-phe and 100 mM I~2HP04 pH 7.5 that had been pre-warmed to 37°C. The gel was shaken gently in the staining solution for one hour at 37°C in the dark. The gel was then washed in distilled water and left in distilled water until it could be photographed. (See figure 2.) This illustrates the preseace or ab~:ence of detectable aminotransferases in various E. coli strains.
MC 1061 extracts prepared in this way gave three staining bands characteristic of the ilvE, as~C and tvrB activities. By comparing the transaminase pattern of extracts prepared from the transductants with that of MC1061 it was possible to detect MC 1061 recombinants .lacking the characteristic aspC activity.
One mutant v;hich was devoid of a~C activity was purified and named HW1G9. This strain was then cured of the TnlO transposon by the method of Bochner et al (J.Bact 143:926). HW109 was grown overnight in L-broth.
Approximately 106 cells per plate were spread onto Bochner selective medium which was prepared as follows:
Solution A (Bacto-trypt:one lOg, Bacto yeast extract 5 g, chlortetracycline HC1 50 mg, agar 15 g, water 500 ml) and solution B (NaCl 10 g, NaH2P04.H20 10 g, glucose: 2 g, water 500 ml) were autoclaved separately for 20 min at psi. The solutions were mixed and cooled to pouring te~npe,rature.. .:5~,n1,l,,Qf "ZnC.l2._(zO,mM). and,b ml of 2 mg ml 1 fusaric acid were then added prior to pouring. The plates were incubated for 24 hr at 37° and the f~usaric acid resistant colonies isolated and purified by re-streaking on the same medium. These isolates were then tested for sensitivity to tetracycline. One tetracycline sensitive derivative of_ HW109 was chosen and named HW157.
This approach did not, however, yield tr~rB
derivatives of HW 22. To isolate a strain with Tn 10 . 1 341 35 2 lint:ed to tyrB, the procedure was as follows. S~tL-ain ES430 carries a mutation i.n malB that renders it. unable to use maltose as a carbon source. The malB is reasonably closely linked with tyrB. Therefore, ES430 was transduced with Pl. gro~,:~n on the random 'rn 1f pool in Hid 22 described abo~.:~e, selecting for Tet R on maltose P~laconkey agar plates ~uplemented v:ith 10 yc~,'m1 tetracvcline-F1 transductions were performed as follows. The recir~i~~nt ~ train vas cy-o~:w in L-broth to an OD600 of about 1Ø CaCl2 was added to a final concentration of 2.5 mM. Phage P1 clear grown on the desired donor was then added at a multiplicity of insertion of between 0.2 and 1.0 (usually lOg pfu P1 clear per ml of recipient).
The cells were incubated at 3;°C For fifteen minutes, centrifuged at 10,000 fpm for five seconds, washed in the original volume of 0.1 1'i citrate buffer pH 7.0 a.nd finally re-suspended in citrates buffer. Suitable aliquots were then plated on the selective medium.
Transductants which had simultaneously become TetR
and Mal+ were picked and purified. P1 lysates were then prepared from these strains and used to transduc:e DO a4 to TetR. The TetR colonie>s derived from each of the eight P1 lysates were then patched onto minimal agar plates lacking aspartat:e and tyrosine. In one of these experiments, good linkage between Tn 10 and tyrB was obtained. Therefore, one TetR t_yrB recombinani~ from this experiment was isolated, a P1 lysate prepared from this recombinant and used to transduce HW157 described above to TetR. Fifty of these transducants were then patched onto minimal agar supplemented with leuc:ine, but lacking L-aspartate and L-phenylalanine. In this way, a tyrB aspC derivative of MC 1061 was detected.
This strain was designated HS9.158. Finally, HW159, a tetracycline sensitive derivative of HW158 was isolated as described above for the i~>olation of Hw157. HW 159 will not grow on minimal medium supplemented with leucine since it requires aspartate and tyrosine for growth unlike the 1p parental strain t~7C1061 which only requires leucine.
Example 4 Screening of sub-clones for abili~ to complement the t~
B lesion in HW 159 All the recombinant: plasmids obtained by subcloning DNA fragments from the dnaB cosmid clone (described above) were introduced into strain H49 159 by transformation selecting on L-agar supplemented with 200 ugjml carbenicillin. These transformants were then streaked onto minimal medium supplemented only with 2U. , leucine. to test ;for .the,.ability.,:of tOe,"p.reseyce .of .qe.nes carried by the plasmid to complement the t~rB lesion in.
HW 159. One sub-clone carrying a ClaI fragment from cosmid dnaB c'_one number 5 was found to restore the ability of HW 159 to grow on minimal medium. supplemented with leucine and thus to carry the ~rB gene.
Sequencing of the t~B gene was performed by the method of Maxam and Gilbert. The sequence of the tyrB

gene is shown in figure 3. Based upon this DNA sequence the amino acid sequence for the aminotransferase encoded for by tyzB could be determined using the genetic triplet codons. This amino acid sequence is shown in figure 4.
E~:an~, le S
.p __._ Cloning of the ast~~C-cJene_ As a starting point for the cloning of the asoC c3ene, it was decided to use the specialized transducin.g phage lambda aspC2 obtained from M. Ono. The phacJe was prepared lU as follows. A culture of HW76, which is a double lysogen carrying lambda as~C and lambda CI 857 Sam 7, wa.s grown to an OD600 °f 0.6 at 30°C. The culture was then incubated at 45°C for fifteen minutes to induce the prophage and then shaken vigorously at 37°C for three hours. Cell lysis was completed by the addition of 0.5 ml of CHC1.. and the cell debris removed by centrifugation J
at 10,000 rpm for ten minlates. The phage was precipitated by the addition of NaCl to 2.4;o w/v and polyethylene glycol (average molecular weight 6,000) to 10°,o w/v. The 20 , phac~e~,was precipitated overnight, at 4°C, and, then pelleted by centrifugation at S,OOU rpm for ten minutes. The phage pellet was gently re-suspended in 10 ml of phage buffer consisting of 10 mM Tri.s-HC1 pH 7.5, 10 mM MgSO~~. The phage cvas pelleted by centrifugation for one hour at 40,000 rpm and finally re-suspended in 1 ml of phage buffer.
The phage was further purified by running on a CsCl.
block gradient as follows. The lml of phage waa applied to a two-step block gradient in celimiose nitrate tubes.
The gradient contained 1 ml of 5L~9 CsCl, 10 mL~i MgS04, 10 mM Tris-HC1 pH 8.0 and 0.1 mM EDTA overlayed with 3 ml of 3M CsCl, 10 mM r9g SO~, 10 mM Tris-HC1 pH 8.0 and 0.1 mM
EDTA. The gradient: ',:as cent: ifud='d W : a )ackmann S1~ E5 rotor for one hour at 30,000 ~-ym at 2C°C. Tloe phage band was remo-.-ed in 0.5 ml :_rc~m the side ~f the tube usitxg a 1 ml syringe with a 5,,8 inclu (1.6 cm) 25 c)uacre needle. The 0.5 ml of phaae ~.~as th=_-n mixed with 0.5 ml. of saturated CsCl solution (25°C) in 10 mM t~lg SO;I, 10 n~I~1 Trigs-HCl pH

8.0 and 0.1 mM EDTA and mixed well in a cellulose nitrate centrifuge tube. This was then overlayed with ~ ml of 5M
CsCl in 10 mi~7 Mg SOa, 10 mM Tris-HC1 pH 8.0 and 0.1 mM
EDTA and then 1 ml of 3M CsCl in 10 mhl L4g SO~, 1.0 mM
Tris-HC1 pH 8.0 and 0.1 mL~9 EDTA. This was ac3ain centrifuged at 30,000 rpm,for OI'1e hour at 20°C. The phage was again removed from the side of the tube in 0.5 ml as before.
The DNA was extracted from the phage particles as ZO follows. To the 0.5 m1 of phage in~a l5 ml coreex centrifuge tube was added 50 ul of 2M Tris-HC1 pH 8.5, 0.2 M EDTA and 0.5 ml of formamide Laas added. After 30 minutes, 0.5 ml of water was added and the DNA
precipitated by the addition of 3 ml of ethanol. The DNA
was pelleted by centrifugation for five minutes at 10,000 rpm. The supernate was discarded and the pellets rinsed with 70% v/v ethanol. The DNA was dried in vacuo and finally dissolved in TE (10 mM Tris-HC1 and 1 mLn EDTA pa -.s 6-S.0) to give a DNA COIICentration of SOO ug/ml. The lambda aspC DNA was partially digested with the restriction endonuclease Sau3A as described above fo~_- E.
coli DNA. The DNA from all of the digests were pooled and run on a 1°o w/v low c~c:lling temperature agorose gel. in TBE. Also prepared and run on the same gel was 1 ug ef p.~~.T153 DNA ccmpletel y digested with the re:;t~-ici~ion lendonuclease BamHI and calf intestinal phosphatase as described above. The DN.4 was visualized under 366 nm U.V.
licJht as described above.
The bat:ds corresponding to linearised pAT153 and the partially restricted lambda aspC DNA between 2._'> and 6kb were excised. The DNA was extracted from the gel as follows. The gel slices were melted at 65°C fox five minutes in 15 m1 convex tubes, cooled to 37°C and diluted with two volumes of water a~ 37°C. The agarose was then removed by the addition of an equal volume oz phenol at 37°C and vortexing for three minutes. The phases were separated by cetrifugation at 10,000 rpm for five minutes 2U and the upper aqueous phases removed to clean tubes.
Three phenol extractions'were performed.~~ Tyre final supernate was then extracted four times by vorte:xing with an equal amount of diethyl ether, allowing the phases to separate and then removing the upper ether layer. The volume of the remaining aqueous phase was determined and 3M sodium acetate added to bring the salt concentration to 0.3 M. Yeast transfer RNA was added to a final concentration of 5 ug/ml as a carrier. The DNA was precipitated by the addition of 2.5 volumes of ethanol, overnight incubation at. -20°C and centrifugation at 10,000 rpm for 30 minutes. Tlve pellet was washed in 70% v;v ethanol, re-centrifuged at 10,000 rpm for 30 minutes and dried in vacuo. 'flm Dr~A was then dissolved in T'E to give a concentration of about 100 uc3;ml. The partially rests icted lambJa a PLC DNA was then lirated to E~amE3I cut pAT153 by adding 2 ul of i>repared lamb<ja asr~C DNA to 2 ul of prepared pATl53 DNA in 4 u1 of 2x ligation bt;ffar (described above). This mixture was incubated at 37°C for five minutes, cooled to room temperature and the ligation started by the addition of 1 ul of T4 DNA
ligase (New England Biolabs). Tine ligation was continued at 15°C for four hours. The entire ligation reaction mixture was then cooled on ice and added to 200 ul of competent cells prepared from strain HW 159 as described above. The transformation was performed as above and the cells plated on L-agar containing 200 ugjml carbenicillin. The plates were incubated overnight at 2U 37°C. About 5,000 colonies were obtained. These were then 'pooled, wasl-ied 'ti,;~ice in' 10 ml of 0.$5% wjv NaCl and plated onto minimal agar containing leucine and carbenicillin, but lacking tyrosine and aspartate. Only HW 159 cells which carried a recombinant plasmid containing aspC could grow on this medium.
After incubation at 37°C for twenty-four hours, about 100 colonies were obtained. Some of these were purified by successive streakings on the same medium. That these contained a recombinant plasmid was confirmed by running single colony agarose gels as described above. That the recombinant plasmids in these strains .indeed conferred the + +
Asp Tyr phenotype was confirmed by isolating the plasmids, re-transforming into H~~7 159 and showing that + +
these transformants had become Asp Tyr . That the transformants had been altered by the presence of cloned asoC as opposed to t~)w°as shown by running native acrylamide gels on cell-free extracts and staining for transaminase activity as described above. These experiments confirmed that the plasmids indeed carried an intact aspC gene.
The actual level of aminotransferase activity is measured in arbitrary units indicating relative activity.
Relative aminotransferase activity on various strains containing the t~rB and a ~C genes on plasrnid pAT153 is shown in table 2. The assay was performed using a sigma R
aspartate transaminase assay kit. (Sigma Technical Bulletin No. 56-UV, revised August 1980) .> . v: ~. ~..::. r ., . ; ; , ~ ~ : r:, . . ., ,_ : , . . ~.,.... . . ... . .

Table 2 Transaminase Levels b7easured-Usilyg-Ast~artate_As Subs-trate Sigma Aspaytate _T.ransaminase-Assa_y_ ISit mg Frotein Umits~t~IlI1- Rei ative i4ctivity t~IC 1061 40 1 fisJ157 25 0.6 fitd 15 8 I~ID -HW158 pHC79 15 0.37 HSa158 p tyrB X94 12 . 3 Ht°7158 paspCl-1 1738 -I3.5 E~arnnie 6 Further Analysis of the tvrB Clone The ClaI insert carrying tyrB is approt:imately 4.5 kb. A preliminary restriction map was constructed by the commonly used technique of double restriction endonucle<3se digestion followed by gel electrophoresis ~tci'analyse thE=
restriction fragments obtained. In one orientation (see.
figure 6) the ClaI insert had EcoRI, BamHI and NruI sites in positions that facilitated the reduction in plasmid size by in vitro deletion. For example, the small EcoRI
fragment was removed as follows. One ug of ptyrB DNA
was digested to completion with EcoRI in a final volume of 20 ul. The enzyme was inactivated by incubation at 65°

for 5 minutes. A 2 ul ;sliquot of this digest waa them diluted into 50 ul of lipase buffer, one ul of T4 DNA
lipase added and the DNA ligated for 16 hours at 15°.
Ligaticn at this low DNA concentration leads pre~dominani~ly to the re-circularisation of tlw:e plasn~id .,~it.hout the small EcoRI fragment. Th~s Dr7A was then transformed into HW87 selecting for carbenicillin resista:~ce as described above. Indi~~idual transfozmants were picked, purified and analysed on single colony lysate gels as described above.
Most of the transformants contained r~lasmids that exhibited higher mobility than the parental plas;mid ptyrB. Plasmid DNA from several of these clones was isolated and the loss of the small EcoRI fragment confirmed by EcoRI digestion and analysis of these digests by electrophoresis on a 1% agarose gel. All the deleted plasmids gave just one band correspornding to the large EcoRI fragment of ptyrB. The parental plasmid gave the expected two bands.
The EcoRI deleted ptyrB was then subjected to a similar series of manipulations to remove the BamHI
'fragment and this EcoRI BamHI deleted plasmid~was~further subjected to deletion using the enzyme NruI in the following buffer. (100 mM NaCl 6 mM Tris pH7.4 6 mM
MgCl2 5 mM S-mercaptoethanol 100 ug/ml gelatin) All three plasmids were purified and used to transform HW225 (tyrB aspC recA) to carbenicillin resistance. A.11 the transformants simultaneously acquired the ability to grow on minimal medium in the absence of L-tyrosine and L-aspartate. This demonstrates that the tyrB gene is still intact on all three deleted plasmids.
The BamHI-EcoRI fragment from the double deleted ptyrB
was subjected to sequence analysis following the protocols described by i~ia:w:m and GilbeLt (Mar: am, .~,. M. , and W. Gilbert, 1977, Ft~IAS 51: 3S2-389; Maxam, A. M., and W. Gilbert, 1977, P~II~ 65: CH 57 a°°-'59, Pant I). One substantial open reading frame was discovered that could code for a protein of 380 or 396 amino acids in length. The precise start point for translation is not known. There is a typical ribosome bindlIlg site about 10 nucleotides upstream from an in phase GTG (valine codon). Alternatively there is an in phase ATG
(methionine) but this is not preceded by a typical ribosome binding site. The DNA sequence of the Ecol;I-Bam.HI fragment and the amino-acid sequence deduced from it are presented in figures 3 and 4.
Mao of paspC
The restriction map of the smallest aspC sub-clone (paspC
3-~) was elucidated by a combination of single a.nd double l0 restriction digests as described above see figure 7. The 'aspC sub-c'Iones do not pos~sess-restric~ion siteee that allow the insert to be excised cleanly since they were made by ligating Sau3A fragments into the BamHI site of pAT 153, a procedure which does not regnerate the BamHI site.
The paspC3-4 (pME98) was used to transform the E. coli prototroph HW519 and during fermentation produce>d level.s of transaminase equivaleni~ to or better than that of paspCl-1.
The E. toll HW519 carrying pME98 has the deposii~ number ATCC 39501.

1 :341 35 2 Exam le 6 Increased Field of amino Acids In the reactions ~ecruiring aminotransferase enzymes in figure 1, reactions #10 and #13, presence of additional aminotransfe:.-ase en~vm~~:: f,~cilitate~- amino ac:d synthesis. This increa<;ed levol of the aminotrar~sferase enz.~mes synthesised by the ash C, tyr 6 or i lyE gene enables a cell to overcome a rate-limited reaction at the transamination step.
A plasmid of the preseat invention, carrying the asp C, ~r B or ilv E gene is inserted into a bacterial cell by techniques well known in biochemistry. This inserted gone produces a messenger RNA which catalyzes they synthesis of increased levels of aminotrar_sferase~s when the bacterial cell is aro~.an in a suitable media. The presence of these increased levels of aminotransi:arases catalyzes increased levels of amino acids. The amino acids are then harvested from the cells and media by purification procedures well known in fermentation science.
2U The addition of the tyrB containing plasmid into E.Coli 13281 (U:S. pateilt'2;°73,304resulted i~ri ari 11 increase in L-phenylalanine production as shown in Table 3.

Table 3 Phenylalanine Production in E. Coli (Strain 13281) 210_Flas:nid Pl.us_tyr8__Plasmid L-Phe Produced 1C0,°~ 111°0 Culture grcw,i at 32°C for 48 hou_s.
... . , .. , , ~ . , . . . _: ~~:_:, , . . .

Claims (11)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A method of increasing the rate of transamination of a susceptible substrate by the gene product of the asPC, tyrB or ilvE gene comprising:

a) inserting a plasmid containing asPC, tyrB or ilvE
gene into a microorganism, other than Escherichia coli:

b) expressing the asPC, tvrB or ilvE gene in the microorganism resulting in the synthesis of an active aminotransferase;

c) catalyzing the transamination of the susceptible substrate by the aminotransferase.

2. The method of claim 1 wherein the said receptive substrate is phenylpyruvic acid.

3. In a method using a microorganism, other than Escherichia coli, for producing an amino acid wherein an amino acid transamination reaction becomes rate-limiting, the improvement comprising the introduction into a microorganism of a plasmid containing an asPC gene or a tyrB gene or an ilvE
gene, or any combination thereof.

4. Method of claim 3 wherein the amino acid is L-phenylalanine and its keto-acid precursor is phenylpyruvic acid.

5. The method of claim 3 wherein the amino acid is tyrosine and its keto acid precursor is p-hydroxyphenylpy-ruvate.

6. The method of claim 3 wherein the aminotransferase is the product of the asPC gene.

7. The method of claim 3 wherein the aminotransferase is the product of the tyrB gene.

8. The method of claim 3 wherein the aminotransferase is the product of the ilyE gene.

9. In a method using a microorganism , other than Escherichia coli, for producing an amino acid wherein an amino acid transamination reaction becomes rate-limiting, the improvement comprising the introduction into a microorganism's chromosome of an asPC gene or a tyrB gene or an ilvE gene, or any combination thereof.

10. The method of claim 3 wherein the said microorga-nism is an enteric microorganism.

11. A tyrB aminotransferase comprising the following amino acid sequence:
N-VAL-PHE-GLN-LYS-VAL-ASP-ALA-TYR-ALA-GLY-ASP-PRO-ILE-LEU-THR-LEU-MET-GLU-ARG-PHE-LYS-GLU-ASP-PRO-ARG-SER-ASP-LYS-VAL-ASN-LEU-SER-ILE-GLY-LEU-TYR-TYR-ASN-GLU-ASP-GLY-ILE-ILE-PRO-GLN-LEU-GLN-ALA-VAL-ALA-GLU-ALA-GLU-ALA-ARG-LEU-ASN-ALA-GLN-PRO-HIS-GLY-ALA-SER-LEU-TYR-LEU-PRO-MET-GLU-GLY-LEU-ASN-CYS-TYR-ARG-HIS-ALA-ILE-ALA-PRO-LEU-LEU-PHE-GLY-ALA-ASP-HIS-PRO-VAL-LEU-LYS-GLN-GLN-ARG-VAL-ALA-THR-ILE-GLN-THR-LEU-GLY-GLY-SER-GLY-ALA-LEU-LYS-VAL-GLY-ALA-ASP-PHE-LEU-LYS-ARG-TYR-PHE-PRO-GLU-SER-GLY-VAL-TRP-VAL-SER-ASP-PRO-THR-TRP-GLU-ASN-HIS-VAL-ALA-ILE-PHE-ALA-GLY-ALA-GLY-PHE-GLU-VAL-SER-THR-TYR-PRO-TRP-TYR-ASP-GLU-ALA-THR-ASN-GLY-VAL-ARG-PHE-ASN-ASP-LEU-LEU-ALA-THR-LEU-LYS-TRP-LEU-PRO-ALA-ARG-SER-ILE-VAL-LEU-LEU-HIS-PRO-CYS-CYS-HIS-ASN-PRO-THR-GLY-ALA-ASP-LEU-THR-ASN-ASP-GLN-TRP-ASP-ALA-VAL-ILE-GLU-ILE-LEU-LYS-ALA-ARG-GLU-LEU-ILE-PRO-PHE-LEU-ASP-ILE-ALA-TYR-GLN-GLY-PHE-GLY-ALA-GLY-MET-GLU-GLU-ASP-ALA-TYR-ALA-ILE-ARG-ALA-ILE-ALA-SER-ALA-GLY-LEU-PRO-ALA-LEU-VAL-SER-ASN-SER-PHE-SER-LYS-ILE-PHE-SER-LEU-TYR-GLY-GLU-ARG-VAL-GLY-GLU-LEU-SER-VAL-MET-CYS-GLU-ASP-ALA-GLU-ALA-ALA-GLY-ARG-VAL-LEU-GLY-GLN-LEU-LYS-ALA-THR-VAL-ARG-ARG-ASN-TYR-SER-SER-PRO-PRO-ASN-PHE-GLY-ALA-CLN-VAL-VAL-ALA-ALA-VAL-LEU-ASN-ASP-GLU-ALA-LEU-LYS-ALA-SER-TRP-LEU-ALA-GLU-VAL-GLU-GLU-MET-ARC-THR-ARG-ILE-LEU-ALA-MET-ARG-GLN-GLU-LEU-VAL-LYS-VAL-LEU-SER-THR-GLU-MET-PRO-GLU-ARG-ASN-PHE-ASP-TYR-LEU-LEU-ASN-GLN-ARG-GLY-MET-PHE-SER-TYR-THR-GLY-LEU-SER-ALA-ALA-GLN-VAL-ASP-ARC-LEU-ARG-GLU-GLU-PHE-GLY-VAL-TYR-LEU-ILE-ALA-SER-GLY-ARG-MET-CYS-VAL-ALA-GLY-LEU-ASN-THR-ALA-ASN-VAL-GLN-ARG-VAL-ALA-LYS-ALA-PHE-ALA-ALA-VAL-MET-TER-C.