Bacillus MGA3 aspartokinase II gene

Abstract

The present invention provides the isolated DNA sequence encoding the .alpha.B dimer subunit of the lysine-sensitive aspartokinase II isozyme from the thermophilic methylotrophic Bacillus sp. MGA3.

Description

BACKGROUND OF THE INVENTION

Aspartokinase (ATP:4-L-Aspartate-4-phosphotransferase [EC 2.7.2.4]) catalyses the conversion of aspartate and ATP to 4-phosphoaspartate and ADP. As shown in FIG. 1 for E. coli, aspartokinase is the first enzyme utilized in the biosynthetic pathway leading to lysine, threonine, and methionine. The biosynthesis of these nutritionally important amino acids is highly regulated. One mechanism for the regulation of this pathway is via the production of several isozymes of aspartokinase having different repressors and allosteric inhibitors. In both Escherichia coli and recently in Bacillus subtilis, three isozymes of aspartokinase differing in their sensitivity to repression and inhibition by lysine, threonine, methionine, and diaminopimelate have been identified. The three B. subtilis isozymes are feedback-inhibited by diaminopimelate, lysine, or threonine plus lysine, respectively (L. M. Graves, J. Bacteriol., 172, 218 (1990)). The lysine-sensitive aspartokinase II from B. subtilis has been purified to homogeneity by D. Moir et al., J. Biol. Chem., 252, 4648 (1977). The gene encoding this enzyme has also been cloned and sequenced, as reported by R. P. Bondaryk et al., J. Biol. Chem., 260, 592 (1985) and N. Y. Chen et al., J. Biol. Chem., 262, 8787 (1987).

Recently, F. J. Schendel et al. in J. Appl. Environ. Microbiol., 56, 963 (1990), identified homoserine auxotrophs and S-(2-aminoethyl)-cysteine (AEC) resistant mutants of a thermophilic methylotrophic Bacillus sp. which overproduce significant quantities of L-lysine at 50.degree. C. Such thermophilic methylotrophs may have advantages over other organisms for industrial use, as discussed by Al-Awadhi et al., Biotechnol. Bioeng., 36, 816, 821 (1990). In particular, the methylotrophic Bacillus MGA3 identified by F. J. Schendel et al., cited supra, may have significant advantages over other bacilli for the overproduction of lysine since it does not sporulate at high temperatures even under conditions of nutrient limitation, in contrast to lysine-producing mutants of B. licheniformis that sporulated when grown at temperatures greater than 40.degree. C. (H. Hagino et al., Biotechnol. Lett., 3, 425 (1981)).

Since both spore components, diaminopimelate and dipicolinic acid, are derived from the lysine biosynthetic pathway, as shown in FIG. 1, differences in the regulation of this pathway may occur between this thermophilic Bacillus sp. and other mesophilic bacilli. Therefore, a need exists to isolate and characterize the informational macromolecules (DNA and RNA) which function in the biosynthetic pathway to lysine, methionine and threonine in the thermotolerant Bacillus sp. MGA3. A further need exists to isolate and characterize the products, such as the enzymes, that function in these biosynthetic pathways. A further need exists to produce mutant varieties of said informational macromolecules, in order to improve the properties of the enzymes and other polypeptides encoded thereby, or to produce improved strains of thermotolerant, methylotrophic bacteria.

SUMMARY OF THE INVENTION

The present invention provides a DNA sequence in substantially pure form, which corresponds to the structural gene coding for the .alpha.B dimer subunit of lysinesensitive aspartokinase II (AKII) of the methylotrophic thermotolerant Bacillus sp. MGA3. The native form of this enzyme is an .alpha..sub.2 B.sub.2 tetramer. The DNA sequence was identified by cloning the structural gene from a genomic library via complementation of an Escherichia coli auxotrophic mutant lacking all three aspartokinase isozymes. The nucleotide sequence of the entire 2.2 Kb PstI fragment has been determined to be as depicted in FIG. 2 and a single open reading frame coding for the aspartokinase II enzyme was identified at positions 664-1885 of this fragment.

The present invention also provides a substantially pure enzyme corresponding to this form of aspartokinase II (AKII) and a substantially pure polypeptide corresponding to the .alpha.B dimer subunit of AKII. AKII is an .alpha..sub.2 .beta..sub.2 tetramer (M.sub.r 122,000) with the .beta. subunit (M.sub.r 18,000) being encoded within the .alpha. subunit (M.sub.r 45,000) in the same reading frame. The N-terminal sequence of both the .alpha. and .beta. subunit were found to be identical with those predicted from the gene sequence. The predicted AKII sequence of 411 amino acids is only 76% identical with the sequence of the B. subtilis aspartokinase II. The transcription initiation site of the AKII gene is located approximately 350 base pairs upstream of the translation start site, and putative promoter regions at -10 (TATGCT) and -35 (ATGACA) were also identified. Therefore, this gene represents a significant point of divergence of the MGA3 lysine biosynthetic pathway from the pathway operative in other mesophilic bacilli.

Availability of the MGA3 AKII gene, coupled with knowledge of its sequence, permits the production of mutant forms of the present AKII, via mutagenesis of the gene. Mutant forms of the MGA3 AKII gene may be useful to produce microorganisms such as new strains of bacteria, which overproduce lysine at higher levels, or under even more stringent environmental conditions. Methodologies for the mutagenesis of the MGA3 AKII gene are discussed in detail hereinbelow.

As used herein, with respect to an enzyme or a subunit thereof, the term "corresponding to aspartokinase II (AKII)" is intended to mean that the enzyme or the subunit referred to exhibits substantial sequence homology to AKII derived from MGA3 (e.g., .gtoreq.85-90%) and that the enzyme also exhibits a substantially equivalent profile of bioactivity, e.g., exhibits .gtoreq.85-90% of the lysine sensitivity exhibited by AKII from MGA3.

As used herein, with respect to a DNA sequence which encodes AKII or a subunit thereof, the term "substantially pure" means that the DNA sequence is free of other DNA sequences that occur naturally in MGA3, e.g., that it has been isolated from MGA3, via the methodologies of recombinant DNA technology, as described herein, or has bee prepared by known techniques of organic synthesis. Likewise, as used with respect to an AKII enzyme or a subunit thereof, the term "substantially pure" means that the enzyme is free of the other components of naturally occurring Bacillus, in that it has been isolated from a biological medium or has been prepared by known techniques organic synthesis or of recombinant DNA technology.

All the patents, patent documents and publications cited herein are incorporated by reference herein.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic depiction of the lysine biosynthetic pathway in E. coli wherein the following letters indicate the following enzymes: a--aspartokinase; b--aspartylsemialdehyde dehydrogenase; c--dihydrodipicolinic acid synthase; d--dihydrodipicolinic acid reductase; e--succinyloxoaminopimelate synthase; f--succinyldiaminopimelate amino transferase; g--succinyldiaminopimelate desuccinylase; h--diaminopimelate racemase; and i--mesodiaminopimelate decarboxylase.

FIGS. 2A-2D depict the nucleotide sequence of the 2.2 Kb PstI fragment of the genomic clone pAA8671 (Sequence I.D. No. 1) and the derived amino acid sequence for Bacillus MGA3 aspartokinase II .alpha.B dimer subunit (Sequence I.D. No. 2). Regions of dyad symmetry are overlined with arrows, potential ribosome binding sites are underlined, the -10 and -35 regions of the putative promoter are boxed, and the transcription initiation site is marked with an asterisk.

FIGS. 3A and 3B are a comparison of the predicted and determined N-terminal amino acid sequences for (a) the .alpha. subunit, and (b) the .beta. subunit of AK-II from B. MGA3.

DETAILED DESCRIPTION OF THE INVENTION

The invention will be described by reference to the following detailed examples, wherein the bacterial strains, vectors and recombinant plasmids used are summarized in Table 1, below.

TABLE 1__________________________________________________________________________Bacterial Strains and PlasmidsStrain Relevant Markers Reference or Source__________________________________________________________________________Escherichia coliDH5.alpha.F' F'.phi.80dlacZ.DELTA.M15 .DELTA.(lacZYA-argF)U169 recA1 end Bethesda Research Lab hsdR17(r.sub.K -, m.sub.K +) supE44.lambda..sup.- thi-1 gyrA relA1Gif106M1 F.sup.- thrA1101 supE44 .lambda..sup.- rpsL9 malT1(.lambda..sup.R) xyl-7 Barbra Bachman mtl-2 ilvA296 metL1000 arg-1000 thi-1 lysC1001BacillusMGA3 -- ATCC 53907MGA3 S-12 Hse.sup.- R. S. HansonPlasmidspUC19cm Cm.sup.r J. FuchspBR322 Tc.sup.r, Ap.sup.r F. Bolivar et al..sup.bpAA8363 Tc.sup.r, AK.sup.+a This studypAA8671 Cm.sup.r, AK.sup.+ This studypAA8802 Cm.sup.r, AK.sup.- This study__________________________________________________________________________ .sup.a AK, Aspartokinase activity. .sup.b F. Bolivar et al., Gene, 2, 95 (1977).

A. Media and Growth Conditions

Strains of E. coli were grown at 37.degree. C. in baffled Erlenmeyer flasks (Bellco) rotated at 280-320 rpm (Labline) on SOC medium (D. Hanahan, "Techniques for transformation of E. coli," in DNA Cloning: A Practical Approach, D. M. Glover, ed., IRL Press, Washington, D.C. (1985) at pages 109-135), or M9 medium (T. Maniatis et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. (1982)). Auxotrophic stains were implemented with 50 .mu.g/ml of the appropriate amino acid. Bacillus MGA3 (ATCC 53907, American Type Culture Collection, Rockville, Md., USA) was grown at 53.degree. C. in baffled flasks rotated at 350 rpm on MY medium (F. J. Schendel et al., cited supra) containing 1% methanol. Solid media contained 15 g of agar (Sigma, St. Louis, Mo.) per liter of medium. Selective media contained antibiotics at the following concentrations: 15 .mu.g tetracycline per ml, 35 .mu.g chloramphenicol per ml, 100 .mu.g ampicillin per ml, and 50 .mu.g streptomycin sulfate per ml.

B. Recombinant Genetic Methods

DNA manipulations were carried out according to T. Maniatis et al. (cited supra) unless otherwise stated. Transformations of E. coli strains DH5.alpha.F' and GM2163 were carried out according to D. Hanahan (cited supra). Electrotransformation of E. coli strain Gif106M1 was carried out using a Gene Pulser apparatus (Bio-Rad Lab; Richmond, Calif.) at 12.5 KV per cm and 25 .mu.FD capacitance. Cells were allowed to recover for one hour in SOC medium before plating. Electrocompetent E. coli Gif106M1 cells were prepared by growth in SOC to mid-log phase. One liter of cells were harvested by centrifugation at 7,000.times.g, washed twice with an equal volume of cold sterile water, and resuspended in 40 ml cold 10% glycerol. The cells were harvested by centrifugation, resuspended in 2 ml cold 10% glycerol, and 150 .mu.l samples frozen in a dry-ice ethanol bath. The cells were then stored at -80.degree. C. until needed. Restriction endonucleases, T4 DNA ligase, AMV reverse transcriptase, and bacterial alkaline phosphatase were purchased from Bethesda Research Labs (Gaithersburg, Md.) and used according to the instructions of the supplier. Bacillus MGA3 chromosomal DNA was isolated from cells grown in MY medium using the method of R. E. Yasbin et al., J. Bacteriol., 121, 269 (1975).

C. DNA Sequencing and Analysis

Nested deletions were constructed by unidirectional exonuclease III-S1 nuclease digestion (Erase-a-base, Promega Corp., Madison, Wis.). The DNA sequence was determined by the dideoxy-chain termination method of F. Sanger et al., PNAS USA, 74, 5463 (1977) for both strands using Sequenase (United States Biochemicals, Cleveland, Ohio). Analysis of the DNA sequence data was carried out using Intellagenetics software (University of Minnesota Molecular Biology Computing Center).

D. Primer Extension

Total RNA was isolated and primer extension was performed as described by F. M. Ausubel et al., in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y. (1987). For the isolation of RNA, E. coli was grown in SOC and B. MGA3 was grown in minimal methanol media (F. S. Schendel et al., cited supra). Total RNA was isolated from E. coli as described by F. M. Ausubel et al., cited supra, and from B. MGA3 as described by H. Shimotsu et al., J. Bacteriol., 166, 461 (1986). A 24-mer oligonucleotide complementary to the coding strand base pairs 383-406 was endlabeled with .sup.32 P and used as the primer. The products were analyzed on a 6% polyacrylamide-urea gel.

E. Cloning of Aspartokinase II Gene

A chromosomal library of Bacillus MGA3 DNA was constructed by partial digestion of the Bacillus MGA3 chromosomal DNA with PstI followed by ligation with the PstI digested, alkaline phosphatase treated vector, pBR322. The ligation reaction was electrotransformed into E. coli Gif106M1 cells and tetracycline-resistant transformants selected on SOC medium. The tetracycline-resistant colonies were scraped off the SOC plates, washed twice with SSC, then plated onto M9 medium. Aspartokinase II positive clones were identified by their ability to grow on M9 medium lacking lysine, threonine, and methionine.

F. Enzymatic Assays and N-terminal Sequencing of Aspartokinase II

Aspartokinase II was assayed by measuring the amount of aspartyl-.beta.-hydroxamate formed as described by M. J. M. Hitchcock et al., Biochem. Biophys. Acta, 445, 350 (1976). Determination of the apparent K.sub.i for lysine inhibition was carried out with partially purified aspartokinase II from E. coli/pAA8671. Cells were broken in a French press pressure cell at 16,000 psi, cell debris removed by centrifugation at 40,000.times.g for 1 hour, and the supernatant fractionated between 35-50% saturation with ammonium sulfate. The sample was desalted on Sephadex G-25, and apparent K.sub.i for lysine determined by varying the amount of lysine added to the assay in the presence of saturating amounts of aspartate and ATP. Determination of the N-terminal sequence of aspartokinase was carried out by automated Edman degradation at the University of Minnesota Microchemical Facility. Approximately 1 nmol of aspartokinase was run on a 14% SDS gel (U. K. Laemmuli, Nature, 227, 680 (1970)) to separate the .alpha. and .beta. subunits, then electroblotted onto Applied Biosystems (Forest City, Calif.) ProBlot PVDF membrane, following the manufacturer's instructions. The membrane was stained for 15 seconds with Coomassie Blue R-250 (ICN, Cleveland, Ohio), destained with 50% methanol, and the bands corresponding to the .alpha. and .beta. subunits excised and submitted for sequencing.

G. Results

1. Aspartokinase Isozymes From Bacillus sp. MGA3

Recent work in B. subtilis has demonstrated the existence of three aspartokinase isozymes that differ in their feedback inhibition and repression (L. M. Graves et al., cited supra). In order to determine the number of aspartokinase isozymes present in the thermophilic methylotroph B. MGA3, assays of cell extracts were carried out in the presence of lysine, threonine, or diaminopimelate alone or in combination, in accord with the methodology of J.-J. Zhang et al., J. Bacteriol., 172, 701 (1990). The results of these assays are shown in Table 2, and are consistent with the presence of three isozymes; one inhibited by diaminopimelate, one inhibited by lysine alone, and one inhibited by lysine plus threonine.

TABLE 2______________________________________Inhibition of Aspartokinase FromBacillus MGA3 by Amino AcidsAmino Acid Inhibition(5 mM) (%)______________________________________None 100.sup.aLysine 42Lysine + Threonine 85Diaminopimelate 12Diaminopimelate + Lysine 55Diaminopimelate + Lysine + Threonine 98______________________________________ .sup.a Corresponds to a specific activity of 0.011 U/mg protein.

2. Cloning the Structural Gene Coding for Aspartokinase II from Bacillus sp. MGA3

Previous studies by M. Y. Chen et al., J. Biol. Chem., 262, 8787 (1987) showed that the gene coding for aspartokinase II from Bacillus subtilis complemented E. coli Gif106M1, which lacks all three aspartokinase isozymes, (J. Theze et al., J. Bacteriol., 117, 133 (1974)), by restoring its ability to grow on minimal medium lacking lysine, threonine and methionine. To obtain the gene coding for aspartokinase from the thermophilic methylotroph Bacillus sp. MGA3, a chromosomal library was constructed by partial PstI digestion of the MGA3 chromosome. The fragments generated were cloned into pBR322, and used to transform E. coli Gif106M1 to impart tetracycline resistance. After plating onto minimal medium, 40 clones were identified that restored the ability of E. coli Gif106M1 to grow on minimal medium lacking lysine, threonine and methionine. Analysis of 16 of these clones showed that they all shared a common 2.2 Kb PstI fragment. One of these clones, pAA8363, was used for further characterization.

In order to determine if the restored ability to grow in the absence of lysine, threonine, and methionine was due to aspartokinase, enzymatic analysis of cell extracts was carried out, with the results shown in Table 3.

TABLE 3______________________________________Expression of Aspartokinase Activity in E. coli Aspartokinase ActivityStrain Plasmid (U/mg of protein)______________________________________DH5.alpha.F' none 0.0021.sup.aGif106M1 none 0.0002.sup.bGif106M1 pBR322 0.0002.sup.bGif106M1 pUC19cm 0.0001.sup.bGif106M1 pAA8363 0.022.sup.aGif106M1 pAA8363 0.021.sup.bGif106M1 PAA8802 0.0001.sup.b______________________________________ .sup.a Cells were grown in minimal M9 medium lacking lysine, threonine, and methionine. .sup.b Cells were grown in minimal M9 medium containing lysine, threonine and methionine.

As shown in Table 3, significant levels of aspartokinase activity were only found in the wild type E. coli DH5.alpha.F' and in Gif106M1 cells carrying the plasmid pAA8363. No repression of aspartokinase activity was observed when the cells were grown in the presence of 50 .mu.g/ml of lysine, threonine, and methionine (Table 3). Assays were performed in the presence of threonine, methionine, lysine, and diaminopimelate alone, and in combination, but only lysine was shown to inhibit enzyme activity, with an apparent K.sub.i of 100 .mu.M.

Inactivation of the aspartokinase activity was carried out by subcloning the 2.2 Kb PstI fragment into the PstI site of pUC19cm, followed by removal of a 0.6 Kb AvaI fragment from pAA8671. The resulting clone, pAA8802, was examined for aspartokinase activity (Table 3) as well as ability to support growth of Gif106M1 on minimal medium lacking lysine, threonine, and methionine. No significant aspartokinase activity was detected, and pAA8802 would not support growth of E. coli Gif106M1 on minimal medium lacking lysine, threonine, and methionine.

The approximate location of the aspartokinase gene and control regions on the 2.2 Kb PstI fragment was determined by creating a series of unidirectional deletions, and testing each of these for their ability to support growth of Gif106M1 on a minimal medium lacking lysine, threonine, and methionine. Aspartokinase activity was lost when deletions were made 420 base pairs from the 3' end of the fragment, and 350 base pairs from the 5' end.

3. Nucleotide and Derived Amino Acid Sequences of Aspartokinase

The entire 2.2 Kb PstI fragment was sequenced (FIG. 2). The nucleotide sequence (SEQ. I.D. No. 1) revealed one major open reading frame starting at base pair 790, however, there is no potential ribosome binding site preceding this possible start site. A preferred translation start site is apparent at position 664, where a GTG is preceded by a potential ribosome binding site (AAGGGA) underlined in FIG. 2). This translational start site was in complete agreement with the N-terminal amino acid sequence of the .alpha. subunit as shown in FIG. 3(A). A second start site preceded by a potential ribosome binding site, AGGAGG, was found in the same reading frame beginning at base pair 1399. This smaller open reading frame may correspond to the smaller .beta. subunit of aspartokinase. As shown in FIG. 3(B), this second translational start site was in complete agreement with the N-terminal sequence of the .beta. subunit. A stop codon was found at base pair 1897 resulting in predicted molecular weights for the .alpha. and .beta. subunits of 44,313 and 17,899, respectively, and these were in good agreement with the values obtained by SDS gel electrophoresis of 45,000 and 18,000, respectively. The native molecular weight of aspartokinase was found to be 122,000 by gel filtration on Sephacryl-300, which is in good agreement with the predicted molecular weight of 124,424 for an .alpha..sub.2 .beta..sub.2 tetramer.

The transcription initiation site was found by primer extension to correspond to the `A` residue at position 297 in both the B. MGA3 and from the cloned gene in E. coli DH5.alpha.F'/pAA8671. The sequences TATGCT and ATGACA near the -10 and -35 regions correspond to a putative aspartokinase promoter (boxed in FIG. 2). Two regions of dyad symmetry with .DELTA.G's of -18.6 and -11.1 kcal are found in the intervening sequence between transcription initiation and the translation start site (FIG. 2), and the second region contains a series of T residues following the hairpin loop typical of a rho-independent terminator. Another region of dyad symmetry with a .DELTA.G-23.2 kcal occurs distal to the coding region, but lacks a run of T residues following the hairpin loop common to rho-independent terminators.

4. Amino Acid Sequence Comparisons of Aspartokinase

Sequence data are now available for six microbial aspartokinase isozymes, three E. coli (M. Cassan et al., J. Biol. Chem., 261 1052 (1986) (K12); M. Katinka et al., PNAS USA, 73, 5730 (1980); M. M. Zakin et al., J. Biol. Chem., 258, 3028 (1983)), the Bacillus subtilis aspartokinase II (N. Y. Chen et al., cited supra), and Saccharomyces cerevisiae (J. A. Rafalsk et al., J. Biol. Chem., 263, 2146 (1988). The deduced amino acid sequence for B. MGA3 aspartokinase II (SEQ. I.D. No. 2) was compared with the proposed alignment for the B. subtilis aspartokinase II, and the three E. coli aspartokinase isozymes, the S. cerevisiae isozyme and the E. coli isozymes, the S. cerevisiae isozyme, and the E. coli isozyme. Some similarity exists between the deduced amino acid sequence of B. MGA3 aspartokinase and the B. subtilis aspartokinase II, with 76% of amino acid residues being identical. When the amino acid sequence of B. MGA3 aspartokinase is compared with the three E. coli aspartokinases and the S. cerevisiae enzyme, less similarity is found. Only 29, 23, 20, and 17% of its amino acid residues are identical to those of E. coli aspartokinase III, I, II and the S. cerevisiae aspartokinase, respectively. These findings support the assignment of MGA3 to the genus Bacillus, as discussed by F. J. Schendel et al., cited supra.

H. Discussion

Complementation of the E. coli strain Gif106M1, a mutant in all three aspartokinase isozymes, resulted in the selection of only the gene coding for aspartokinase II from B. MGA3, and neither of the genes coding for aspartokinase I or III. This is probably due to the inability of E. coli to recognize either the Bacillus promoters or Shine-Dalgarno sequences for these two isozymes (L. Band et al., DNA, 3, 17 (1984); G. Lee et al., Mol. Gen. Genet., 180, 57 (1980)). The proposed -10, TATGCT, and -35 regions, ATGACA, are similar to the compiled -10, TATAAT, and -35, TTGACA, regions from several B. subtilis genes (as reported by C. P. Moran et al., Mol. Gen. Genet., 186, 339 (1982)), and to the -10, TAAAAT, and -35, TTGTCC, regions of the B. subtilis aspartokinase II gene (N. Y. Chen et al., J. Biol. Chem., 262, 8787 (1987)). The expression of the gene coding for aspartokinase II in E. coli results from transcription initiation at the same site as in B. MGA3, and is probably due to the similarity of the -10 and -35 regions to the consensus sequences of E. coli -10, TATAAT, and -35, TTGACA, regions. In addition, the proposed Shine-Dalgarno sequences for the aspartokinase II .alpha. and .beta. subunits, AAGGGA and AGGAGG, respectively, are both very similar to the consensus sequence, AAGGAG, of B. subtilis (C. P. Moran et al., Mol. Gen. Genet., 186, 339 (1982)). These proposed ribosome binding sites are also very similar to the E. coli consensus sequence, AGGAGG (J.-C. Patte et al., Biochem. Biophys. Acta., 136, 245 (1967)).

The large, >300 nucleotides, intervening sequence that exists between the transcription initiation and translation start sites (FIG. 2), may function in the control of aspartokinase II expression in the presence of lysine. Unlike the control sequence for the B. subtilis aspartokinase II, that contains characteristics similar to attenuators from several E. coli amino acid biosynthetic operons, as shown by R. Kolter et al., Ann. Rev. Genet., 16, 113 (1982), no open reading frame preceded by a ribosome binding site that contained a lysine rich peptide was found. This also explains why attenuation of aspartokinase II was not observed when E. coli Gif106M1/pAA8363 was grown in the presence of lysine (Table 3). In contrast, growth inhibition due to 22 .mu.M lysine wa observed with E. coli Gif106M1 carrying a single copy plasmid containing the gene encoding the B. subtilis aspartokinase II (N. Y. Chen et al., J. Biol. Chem., 263, 9526 (1988)). While part of this inhibition may have been due to feedback inhibition, since the aspartokinase II from B. subtilis had a K.sub.i 100 .mu.M (30), it is likely that some of the growth inhibition resulted from attenuation of the aspartokinase gene.

EXAMPLE I

Mutagenesis of Aspartokinase II Gene

Site-directed mutants were constructed by in vitro second strand synthesis (Altered Sites, Promega Corp., Madison, Wis.) or by the method of T. A. Kunkel et al., PNAS USA, 82, 488 (1985) (Muta-Gene, Bio-Rad, Richmond, Calif.) using a mismatched oligonucleotide primer of 18-24 base pairs. A 19 base pair primer corresponding to the sequence 5'-TTTTGTTCTAATGTTACTT was used to change the `T` and `G` at positions 1400 and 1401 to `A` and `T` respectively. This results in an amino acid change from methionine to leucine at position 246 in the protein sequence. In addition, this amino acid substitution eliminates the initiation codon for the synthesis of the .beta. subunit resulting the synthesis of only the .alpha. subunit. Analysis of cell extracts containing this altered (.alpha..sub.2) enzyme revealed that the aspartokinase activity and inhibition by lysine was essentially identical to the wild type (.alpha..sub.2 .beta..sub.2) protein. This result was similar to the result obtained by Chen and Paulus, cited above.

In vitro plasmid mutagenesis was carried out using hydroxylamine as described by C. Wolf et al., J. Bacteriol., 170, 4509 (1988). One .mu.g of pAA8671 DNA was treated with 100 uL of 0.4 M hydroxylamine in 0.5 M potassium phosphate (pH 6.0) for 36 hours at 37.degree. C. The sample was then dialyzed for 12 hours against 4 L of 1 mM EDTA (pH 7.0). Electrocompetent E. coli Gif106M1 cells were then transformed by electroporation with 1 ul of the dialyzed sample. The cells were allowed to recover for 1 hour in SOC, then plated onto minimal media containing 10 g/l lysine, and grown for 24 hours at 37.degree. C. Ten colonies were selected, grown at 37.degree. C. for 16 hours in TB containing 35 ug/ml chloramphenicol. The cells were collected by centrifugation, suspended in 50 mM potassium phosphate, and sonicated for 30 seconds to disrupt the cells. Aspartokinase II enzyme assays were preformed in the presence and absence of 100 mM lysine. Two clones, 9234 and 9236, showed decreased sensitivity to lysine with apparent K.sub.I 's of 10 and 100 mM, respectively, compared to a K.sub.I of 100 uM for wild type enzyme.

The DNA coding for these altered enzymes was sequenced and only a single base pair change was found in each case. In pAA9234, a `T` replaces the `C` at position 1790 resulting in a single amino acid change from a serine to leucine at position 376 of the protein sequence. In pAA9236, a `T` replaces the `C` at position 1730 resulting in a single amino acid change from an alanine to a valine at position 356 of the protein sequence.

From the results of these mutagenesis experiments, specific mutations in the .alpha. subunit alone should result in altered lysine feedback inhibition in a wide variety of transformants.

The invention has been described with reference to various specific and preferred embodiments and techniques. However, it should be understood that many variations and modifications may be made while remaining within the spirit and scope of the invention.

__________________________________________________________________________SEQUENCE LISTING(1) GENERAL INFORMATION:(iii) NUMBER OF SEQUENCES: 2(2) INFORMATION FOR SEQ ID NO: 1:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 2223 base pairs(B) TYPE: Nucleic Acid(C) STRANDEDNESS: Single(D) TOPOLOGY: Linear(ii) MOLECULE TYPE: Genomic DNA(vi) ORIGINAL SOURCE: 2.2 Kb Pst frag. of PAA8671(ix) FEATURE:(A) NAME/KEY: Aspartokinase II Gene(B) LOCATION: 1 to 2223(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:TTACGCCAAGCTTGCATGCCTGCAGCGAATCCAAGATGAAGTGCACAGATTTGCGATTAC60TTTCCACCGTCAATTGCGGGGGAAAAATGCTTTTCAATCGTTATTGGA CGATATACCAGG120AATTGGTGAAAAACGGAAAAAACTGCTTCTTAAACAATTTGGTTCCGTAAAAAAAATGAA180GGAAGCAACAATGGCGGAAATTACATCTGTCGGCATTCCGGCAAATGTTGCAAAAGAATT240GATGAAAAAGTTGCATGAATGACAT TGTCATAAATCAGGTCGTATGCTATACTGAAAAAA300ATTTTATAGTGTAATCACTTTAGAGCATTAAAGTGAAGATAGAGGTGCGAACTTCATCAG360TAAAAGCTTGGAGAAGAATGAGCTTCAATGAAAAGCTTTGAAAGGGAACGTTCGCCGAAG420TG AAGAAAAACTCATTTTTTTCTTTGCTGGTCCTGCATTTAAGAGATGCCGGATTGTCAA480GGCGGTGCCGCCTTGGAGAGCTATCTCACTGTGTCTGCGTATTTTACTACGTTATCCACA540GCAATGAGGTAGCTTTCTCATTGCTGTTTTTTATTAAATTAAAAACAG CTTCATTGAGAA600AGCTAGTTATACATAAAATGGCGGCACTTCTTTGATTAATTTCATAGAAAGAAGGGAAAA660AAAGTGGGATTAATTGTCCAAAAGTTTGGCGGAACATCTGTTGGCTCC708ValGlyLeuIleValGln LysPheGlyGlyThrSerValGlySer151015GTTGAGCGCATCTTAAACGTTGCCAATCGGGTAATTGAAGAAAAAAAG756ValGluArgIleLeuAsn ValAlaAsnArgValIleGluGluLysLys202530ACCGGAAATGACGTTGTTGTGGTTGTTTCTGCAATGGGGAAGACAACA804AsnGlyAsnAspValVa lValValValSerAlaMetGlyLysThrThr354045GATGAGCTTGTCGATTTAGCAAAACAAATTTCAGCACATCCACCAAAG852AspGluLeuValAspLeuA laLysGlnIleSerAlaHisProProLys505560CGCGAAATGGATATGCTTCTTACAACCGGAGAGCAAGTGACGATTTCG900ArgGluMetAspMetLeuLeuThr ThrGlyGluGlnValThrIleSer657075CTTTTGGCTATGGCATTGAATGAAAAAGGCTATGAGGCCATTTCCTAT948LeuLeuAlaMetAlaLeuAsnGluLysGlyTyr GluAlaIleSerTyr80859095ACTGGATGGCAGGCAGGAATTACAACTGAACCTGTTTTTGGGAACGCG996ThrGlyTrpGlnAlaGlyIleThrThrGl uProValPheGlyAsnAla100105110AGAATATTAAATATCGAAACCGAAAAAATTCAAAAACAGCTAAACGAA1044ArgIleLeuAsnIleGluThrGluLysI leGlnLysGlnLeuAsnGlu115120125GGAAAAATTGTCGTAGTTGCCGGCTTCCAAGGTATTGATGAGCACGGA1092GlyLysIleValValValAlaGlyPheGln GlyIleAspGluHisGly130135140GAAATTACGACTCTTGGGAGAGGCGGATCCGATACTACGGCTGTAGCA1140GluIleThrThrLeuGlyArgGlyGlySerAspThr ThrAlaValAla145150155CTTGCTGCGGCTTTGAAAGCCGAAAAATGTGATATTTACACCGATGTT1188LeuAlaAlaAlaLeuLysAlaGluLysCysAspIleTyrThrAs pVal160165170175ACTGGAGTTTTTACTACAGATCCGCGCTATGTAAAGTCGGCTAGGAAG1236ThrGlyValPheThrThrAspProArgTyrValLysSerA laArgLys180185190CCTGCTTCTATTTCATATGATGAAATGCTTGAACTTGCGAATCTTGGT1284LeuAlaSerIleSerTyrAspGluMetLeuGluLeuAla AsnLeuGly195200205GCGGGCGTCCTTCATCCAAGAGCAGTAGAATTTGCGAAAAATTACGGA1332AlaGlyValLeuHisProArgAlaValGluPheAlaLysAsn TyrGly210215220ATTACTTTGGAGGTGCGCTCCAGTATGGAACGAGAAGAAGGGACGATC1380IleThrLeuGluValArgSerSerMetGluArgGluGluGlyThrIl e225230235ATTGAGGAGGAAGTAACAATGGAACAAAATCTTGTTGTCCGGGGAGTA1428IleGluGluGluValThrMetGluGlnAsnLeuValValArgGlyVal240 245250255GCTTTTGAAGATGAAATCACTCGAGTAACAGTTTTTGGATTGCCAAAC1476AlaPheGluAspGluIleThrArgValThrValPheGlyLeuProAsn 260265270TCATTAACGAGTTTATCTACTATTTTTACGACACTTGCTCAAAATCGC1524SerLeuThrSerLeuSerThrIlePheThrThrLeuAlaGlnAsnArg 275280285ATTAATGTTGATATCATCATCCAAAGTGCAACTGATGCTGAAACAACA1572IleAsnValAspIleIleIleGlnSerAlaThrAspAlaGluThrThr 290295300AATTTATCTTTTTCCATAAAGAGCGACGATTTAGAAGAAACAATGGCC1620AsnLeuSerPheSerIleLysSerAspAspLeuGluGluThrMetAla305 310315GTCCTCGAAAACAATAAAAATTTGCTTAACTACCAAGGGATTGAATCG1668ValLeuGluAsnAsnLysAsnLeuLeuAsnTyrGlnGlyIleGluSer320325 330335GAAACGGGATTAGCAAAAGTATCGATTGTCGGTTCAGGAATGATCTCT1716GluThrGlyLeuAlaLysValSerIleValGlySerGlyMetIleSer34 0345350AACCCTGGAGTCGCAGCTAAAATGTTTGAAGTGCTTGCTTTAAATGGA1764AsnProGlyValAlaAlaLysMetPheGluValLeuAlaLeuAsnGly355 360365ATCCAAGTGAAAATGGTCAGCACTTCAGAAATAAAAGTATCGACGGTT1812IleGlnValLysMetValSerThrSerGluIleLysValSerThrVal370 375380GTTGAAGAAAGCCAGATGATCAAGGCAGTAGAAGCGCTTCATCAAGCA1860ValGluGluSerGlnMetIleLysAlaValGluAlaLeuHisGlnAla385390 395TTTGAACTGTCGGGATCCGCTGTTAAATCGGAACGCTAACGCCTAT1906PheGluLeuSerGlySerAlaValLysSerGluArg400405410ATTATAAA GAAAAACTTGAGGCTGACCCATAAGGTCCTGGCTCGCGTTTGCAGTTACTAA1966ATATTGTAGAAACAGTAATCATGTTTTTTAATATTTAGTAACTGAGAGTGCCTGGCTCTT2026AGTCTTGGGTCAGCCTTTATCCATAAATCATGGCTTTACGACGTCTTTTTTGT CCCACTT2086AACCGTTATTAGCACCTTTGATCCCTTTTTACGAGGGTGTTCAAACGCTTCAGCAATTAC2146TTTTTTTTGCTGTTCAATTTGCTGGGCAATAAATCCCGCTTCCAACTGAAAAGAGATATC2206TTTTTTTGACTGCAGGT 2223(2) INFORMATION FOR SEQ ID NO: 2:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 411 amino acids(B) TYPE: Amino Acid(D) TOPOLOGY: Linear(ii) MOLECULE TYPE: Polypeptide(ix) FEATURE:(A) NAME/KEY: Aspartokinase II dimer subunit( B) LOCATION: 1 to 411(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:ValGlyLeuIleValGlnLysPheGlyGlyThrSerValGlySerVal151015GluArgIleLeuAsnValAlaAsnArgVa lIleGluGluLysLysAsn202530GlyAsnAspValValValValValSerAlaMetGlyLysThrThrAsp3540 45GluLeuValAspLeuAlaLysGlnIleSerAlaHisProProLysArg505560GluMetAspMetLeuLeuThrThrGlyGluGlnValThrIleSerLeu6 5707580LeuAlaMetAlaLeuAsnGluLysGlyTyrGluAlaIleSerTyrThr859095Gly TrpGlnAlaGlyIleThrThrGluProValPheGlyAsnAlaArg100105110IleLeuAsnIleGluThrGluLysIleGlnLysGlnLeuAsnGluGly1 15120125LysIleValValValAlaGlyPheGlnGlyIleAspGluHisGlyGlu130135140IleThrThrLeuGlyArgGlyGlySe rAspThrThrAlaValAlaLeu145150155160AlaAlaAlaLeuLysAlaGluLysCysAspIleTyrThrAspValThr165 170175GlyValPheThrThrAspProArgTyrValLysSerAlaArgLysLeu180185190AlaSerIleSerTyrAspGluMetLeuGluL euAlaAsnLeuGlyAla195200205GlyValLeuHisProArgAlaValGluPheAlaLysAsnTyrGlyIle210215220 ThrLeuGluValArgSerSerMetGluArgGluGluGlyThrIleIle225230235240GluGluGluValThrMetGluGlnAsnLeuValValArgGlyValAla 245250255PheGluAspGluIleThrArgValThrValPheGlyLeuProAsnSer260265270LeuTh rSerLeuSerThrIlePheThrThrLeuAlaGlnAsnArgIle275280285AsnValAspIleIleIleGlnSerAlaThrAspAlaGluThrThrAsn290 295300LeuSerPheSerIleLysSerAspAspLeuGluGluThrMetAlaVal305310315320LeuGluAsnAsnLysAsnLeuL euAsnTyrGlnGlyIleGluSerGlu325330335ThrGlyLeuAlaLysValSerIleValGlySerGlyMetIleSerAsn340 345350ProGlyValAlaAlaLysMetPheGluValLeuAlaLeuAsnGlyIle355360365GlnValLysMetValSerThrSerGluIleLysVal SerThrValVal370375380GluGluSerGlnMetIleLysAlaValGluAlaLeuHisGlnAlaPhe38539039540 0GluLeuSerGlySerAlaValLysSerGluArg405410

Claims

1. An isolated DNA fragment from methylotrophic, thermotolerant Bacillus sp. MGA3 which encodes a polypeptide corresponding to the .alpha..beta. dimer subunit of lysine-sensitive aspartokinase II of the methylotrophic, thermotolerant Bacillus sp. MGA3.

2. A DNA sequence corresponding to SEQ. I.D. No. 1.