WO2006082252A2

WO2006082252A2 - Method for the enzymatic production of alpha-ketobutyrate

Info

Publication number: WO2006082252A2
Application number: PCT/EP2006/050722
Authority: WO
Inventors: Rainer Figge; Fabien Lux; Céline RAYNAUD; Philippe Soucaille
Original assignee: Metabolic Explorer
Priority date: 2005-02-07
Filing date: 2006-02-07
Publication date: 2006-08-10
Also published as: WO2006082252A3; EP1846564A2

Abstract

This invention relates to a process for the enzymatic production of α-ketobutyrate and its derivatives, in particular isoleucine. In the described process α-ketobutyrate is obtained from activated homoserine by Ϝ-elimination. Homoserine is activated to either acylhomoserine or phosphohomoserine using homoserinetransacylase and homoserine kinase, respectively. Subsequently phosphohomoserine and/or acylhomoserine is/are transformed into phosphate or/and acylate, ammonia and α-ketobutyrate through the Ϝ-eliminase activity of cystathionine- Ϝ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfhydrylase. This reaction is favored by (i) reducing intracellular concentrations of H2S or cysteine, (ii) reducing the access of these two sulfur containing substrates to the enzyme cystathionine-Ϝ- synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfliydrylase or (iii) by using cystathionine-Ϝ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfhydrylase with high Ϝ-elimination and low cystathionine-Ϝ-synthase and/or low acylhomoserine sulfliydrylase and/or low phosphohomoserine sulfhydrylase activity. At least one of these measures is applied in microorganisms, in particular coryneform bacteria and enterobacteria, in which as a consequence α-ketobutyrate and/or its derivatives can be produced in high quantities.

Description

Method for the enzymatic production of alpha-ketobutyrate

The present invention relates to a method for the enzymatic production of α- ketobutyrate, wherein α-ketobutyrate is obtained by γ-elimination of an activated homoserine with an appropriate enzyme in an appropriate medium, in particular where the appropriate enzymes are modified enzymes, preferably enzymes that when expressed from the plasmid pMElOl-thrA* in strains MG1655 ΔmetBJ rnetAU AiIvA ΔfcfcABCDEFG the ratio of isoleucine to methionine produced at least equal to two.

Background of the invention

L- amino acids and their derivatives are used in animal nutrition, human medicine and the pharmaceutical industry. Isoleucine, which is an essential amino acid, is principally used as a starting material for drug synthesis and in nutritional supplements.

Isoleucine that includes two asymmetric carbon atoms is difficult to synthesize chemically as the pure L-stereoisomer and therefore fermentative processes for the production of isoleucine have been developed (EP0745679B1, US5474918A1, EP0685555A1). These processes rely on the natural biosynthetic pathway in which the aspartate-derived homoserine is transformed into threonine that in turn can be converted by a set of five reactions to isoleucine (see Fig. 1). Conversion of aspartate to homoserine requires the enzyme aspartate semialdehyde dehydrogenase and aspartokinase/ homoserine dehydrogenase that as a iusion protein harbors both activities. Two Aspartokinases/homoserine dehydrogenases are present in E. coli, one is encoded by the gene thrA that is part of the threonine operon (encoding also homoserine kinase (thrB) and threonine synthase (thrC), which catalyze the biosynthesis of threonine) and the other is encoded by the gene metL that is part of the methionine regulon. The threonine operon is regulated by attenuation via threonine and isoleucine. In addition, the enzyme ThrA is feedback controlled by threonine, the end product of the pathway. Synthesis of isoleucine from threonine requires its deamination to α-ketobutyrate, a reaction catalyzed by two threonine deaminases HvA and TdcB. HvA, the major enzyme under aerobic conditions is feedback controlled by isoleucine. TdcB is preferentially expressed under anaerobic conditions. Subsequently α-ketobutyrate is transformed to isoleucine by a set of 4 additional reactions catalyzed by the enzymes acetohydroxyacid synthase (encoded by ilvB, ilvN, ilvG, ilvM, ilvL, ilvH) acetohydroxy acid isomeroreductase (ilvC), dihydroxyacid dehydratase (ilvD) and branched chained amino acid aminotransferase (ilvE), all of which are shared with valine biosynthesis. In E. coli three acetohydroxyacid synthases exist that are each preferentially feedback regulated by one of the three end-products isoleucine, valine and leucine (reviewed in Neidhardt, F. C. (Ed. in Chief), R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik, W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (eds). 1996. Escherichia coli and Salmonella: Cellular and Molecular Biology. American Society for Microbiology).

In addition to the synthesis via threonine deamination, α-ketobutyrate can also be derived from activated homoserine by γ-elimination, a side reaction of cystathionine-γ- synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfhydrylase. Cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfhydrylase from different organisms will preferentially accept activated homoserines in the form of O-acetylhomoserine, O-succinylhomoserine and phosphohomoserine. In plants methionine and threonine are synthesized from phosphohomoserine, which is thus the branchpoint between the two biosyntheses. In bacteria homoserine iulfils this role. To enter the methionine specific pathway, homoserine is either activated to acetyl- or succinyl-homoserine. In both plants and bacteria activated homoserine is the substrate of cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfliydrylase that can catalyze one of the three reactions: (i) the synthesis of γ-cystathionine with cysteine, (ii) the sulfhydrylation to homocysteine with hydrogen sulfide or (iii) the γ-elimination to ammonia, phosphate and/or acylate and α-ketobutyrate in the absence of sulfur containing substrates. Cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfhydrylase vary with respect to their ratios between the three reactions. Relatively low γ- eliminase activity has been observed in plant enzymes, whereas higher activity is found in the E. coli enzyme, γ-elimination activity also depends on the presence and the accessibility of the enzyme for the sulfur containing compounds.

Object of the invention is a process for the enzymatic production of isoleucine in which α- ketobutyrate is produced from activated homoserine. This can be accomplished by either reducing the amount of intracellular H₂S and/or cysteine, limiting access of these substrates to the active site of cystathionine-γ-synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfhydrylase or by using cystathionine-γ-synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfhydrylase with intrinsically high γ-elimination and low cystathionine-γ-synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfhydrylase activity.

Brief summary of the invention

The invention provides a new process for the production of α-ketobutyrate and its derivatives, in particular isoleucine by utilizing cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulihydrylase, enzymes that normally are utilized in methionine biosynthesis. Enzymes that have high γ-elimination activity are preferred, since they will convert phospho and/or acylhomoserine into α- ketobutyrate. In addition, limiting conditions for the sulfur containing substrates H₂S and cysteine as well as reduced access to the active site of the acyl and/or phosphohomoserine cystathionine-γ-synthase and/or acylhomoserine sulihydrylase and/or phosphohomoserine sulihydrylase favor γ-elimination. Thus the invention provides a new pathway for the synthesis of isoleucine that deviates from the classical known pathway that relies on the synthesis of threonine and its subsequent deamination to α-ketobutyrate. As a consequence the invention makes the requirement for feedback resistant threonine deaminases obsolete.

Detailed description of the invention

The invention describes a process for the enzymatic production of α-ketobutyrate and its derivatives in particular isoleucine, in which α-ketobutyrate is derived from activated homoserine by γ-elimination. For this purpose the implicated enzymes have to be produced and active in an appropriate medium. An appropriate medium can be any solvent or mixture of solvents containing ions that allows the enzymes to be active in the production of α- ketobutyrate from activated homoserine. According to the invention activated homoserine can be phosphohomoserine or acylhomoserine, preferentially acetyl or succinylhomoserine.

The γ-elimination reaction can be performed using phosphohomoserine accepting cystathionine-γ-synthases and/or sulfhydrylases that are known in plants. Several amino acid positions are known to be relevant for the binding of the phosphogroup (Steegborn et ah, 1999, J. MoI. Biol. 290, 983-996). According to these conserved residues the inventors have deduced that cystathionine-γ-synthases and/or sulfhydrylases from the bacterial family Chloroflexaceae should also preferentially accept phosphohomoserine as a substrate. Recently it was shown that some archaebacterial species also use phosphohomoserine accepting cystathionine-γ-synthases and/or sulfhydrylases. Thus cystathionine-γ-synthases and/or sulfhydrylases may be derived from any of the organismal groups described above. Phosphohomoserine is obtained by phosphorylation of homoserine using homoserine kinase, which is preferentially homoserine kinase from E. coli or Corynebacterium glutamicum.

In another embodiment of the invention activated homoserine is acylhomoserine, preferentially acetylhomoserine. In this case the γ-elimination reaction is catalyzed by acetylhomoserine sulfhydrylases and/or cystathionine-γ-synthases that are preferentially derived from Corynebacterium glutamicum, Bacillus subtilis or Saccharomyces cerevisiae. Acetylhomoserine is obtained by acetylation of homoserine, a reaction catalyzed by homoserine acetyltransferase. Preferentially homoserine transacetylases are employed that are feedback resistant to methionine and S-adenosylmethionine. In a preferred embodiment of the invention homoserine transacetylases are obtained from Corynebacterium glutamicum, Bacillus subtilis or Saccharomyces cerevisiae or a spirochaete.

In another application of the invention activated homoserine is succinylhomoserine. Succinylhomoserine is converted into succinate, ammonia and α-ketobutyrate by the γ- elimination reaction catalyzed by succinylhomoserine sulfhydrylase and/or cystathionine-γ- synthase. Preferentially succinylhomoserine sulfhydrylase and/or cystathionine-γ-synthase is obtained from enterobacteriaceae, specifically from E. coli. Homoserine is activated to succinylhomoserine by homoserine succinyltransferase. Homoserine succinyltransferases that are feedback resistant to methionine and S-adenosylmethionine are preferred and have been described in patent application PCT IB 2004/001901. Homoserine succinyltransferases are derived from Enterobacteriaceae, preferentially from E. coli.

In the description of the present invention, enzymes are identified by their specific activities. This definition thus includes all polypeptides that have the defined specific activity also present in other organisms, more particularly in other microorganisms. Often enzymes with similar activities can be identified by their grouping to certain families defined as PFAM or COG.

PFAM (protein families database of alignments and hidden Markov models; http://www.sanger.ac.uk/Soflware/Pfam/) represents a large collection of protein sequence alignments. Each PFAM makes it possible to visualize multiple alignments, see protein domains, evaluate distribution among organisms, gain access to other databases, and visualize known protein structures.

COGs (clusters of orthologous groups of proteins; http://www.ncbi.nlm.nih.gov/CQG/) are obtained by comparing protein sequences from 43 fully sequenced genomes representing 30 major phylogenic lines. Each COG is defined from at least three lines, which permits the identification of former conserved domains.

The means of identifying homologous sequences and their percentage homologies are well known to those skilled in the art, and include in particular the BLAST programs, which can be used from the website http://www.ncbi.nlm.nih.gov/BLAST/ with the default parameters indicated on that website. The sequences obtained can then be exploited (e.g., aligned) using, for example, the programs CLUSTALW (http://www.ebi.ac.uk/clustalw/) or MULTALIN (http://prodes.toulouse.mra.fr/multalm/cgi-bm/multalin..pl), with the default parameters indicated on those websites. Using the references given on GenBank for known genes, those skilled in the art are able to determine the equivalent genes in other organisms, bacterial strains, yeasts, fungi, mammals, plants, etc. This routine work is advantageously done using consensus sequences that can be determined by carrying out sequence alignments with genes derived from other microorganisms, and designing degenerate probes to clone the corresponding gene in another organism. These routine methods of molecular biology are well known to those skilled in the art, and are described, for example, in Sambrook et al. (1989 Molecular Cloning: a Laboratory Manual. 2^nd ed. Cold Spring Harbor Lab., Cold Spring Harbor, New York.).

In addition, the invention also related to enzymes that have been modified, especially optimized for the desired activity. In the case of the invention this optimization concerns specifically enzymes that activate homoserine, in particular the reduction of feedback by end- products can increase their activity. Equally important is the optimization or an increase in γ- eliminase activity while at the same time the cystathionine-γ-synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfhydrylase activities are reduced. This optimization can be related to the amount of isoleucine versus methionine produced in a strain in which the enzymes normally producing isoleucine are lacking (*7vA, tdcB); an example is the strain MGl 655 AmetBJ met All AiIvA Atdc ABCDEFG in which the meB alleles can be introduced on a low copy plasmid (Details for these experiments are given in example 1). In a preferred embodiment of the invention these strains produce a ratio of isoleucine to methionine that is at least of a factor of two or higher. The wildtype MetB allele could be shown to produce only a ratio of 1.4. Modified enzymes may be obtained by directed evolution, preferentially coupled with enzyme modeling, site directed mutagenesis or in vivo or in vitro evolution of the enzyme as described in (Directed Enzyme Evolution Screening and Selection Methods and Directed Enzyme Evolution Library Creation Methods and Protocols, 2003, eds Arnold, F. and Georgiou G., Humana Press).

The inventors have identified mutated versions of cystathionine-γ-synthase and/or acylhomoserine sulfhydrylase and/or phosphohomoserine sulfliydrylase with an increased ratio between in vivo γ-elimination activity and cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfhydrylase. These mutated MetB enzymes are also object of the invention. Mutated MetB enzymes have preferentially at least one mutation in the following three conserved regions or combinations thereof.

All references to amino acid positions are made based on the cystathionine-γ- synthase/succinylhomoserine sulfhydrylase encoded by the metB gene of E. coli (SEQ ID NO

I)-

>E. coli |EG10582|MetB: 386 aa - Cystathionine gamma-synthase MTRKQATIAV RSGLNDDEQY GCWPPIHLS STYNFTGFNE PRAHDYSRRG NPTRDWQRA LAELEGGAGA VLTNTGMSAI HLVTTVFLKP GDLLVAPHDC YGGSYRLFDS LAKRGCYRVL FVDQGDEQAL RAALAEKPKL VLVESPSNPL LRVVDIAKIC HLAREVGAVS WDNTFLSPA LQNPLALGAD LVLHSCTKYL NGHSDWAGV VIAKDPDWT ELAWWANNIG VTGGAFDSYL LLRGLRTLVP RMELAQRNAQ AIVKYLQTQP LVKKLYHPSL PENQGHEIAA RQQKGFGAML SFELDGDEQT LRRFLGGLSL FTLAESLGGV ESLISHAATM THAGMAPEAR AAAGISETLL RISTGIEDGE DLIADLENGF RAANKG

The relative positions of corresponding conserved regions in other cystathionine-γ- synthase or acylhomoserine sulfhydrylase or phosphohomoserine sulfliydrylase from different organisms can be found by a person skilled in the art by simple sequence alignment as represented in figure 2 with the enzymes listed below:

>gi|2852454|dbj|BAA24699.1| cystathionine gamma-synthase, Arabidopsis thaliana >gi|8439541|gb|AAF74981.1|AF082891_l cystathionine gamma-synthase isoform 1, Solarium tuberosum

>gi|2198853|gb|AAB61348.1| cystathionine gamma-synthase, Zea mays >gi|4322948|gb|AAD16143.1| cystathionine gamma-synthase precursor, Nicotiana tabacum >gi|305042|gb|AAB03071.1| cystathionine gamma-synthase, Escherichia coli >gi|21323418|dbj|BAB98046.1| O-acetylhomoserine sulfhydrylase, Corynebacterium glutamicum ATCC 13032

>gi|4191271|emb|C AA09983.1| O-succinylhomoserine sulfhydrylase, Rhizobium etli >gi|53798754|reflZP_00020132.2| COG0626: Cystathionine beta-lyases/cystathionine gamma- synthases, Chloroflexus aurantiacus

>gi|14601268|ref|NP_147803.1| cystathionine gamma-lyase Aeropyrum pernix Kl >gi|26988726|refjNP_744151.11 O-succinylhomoserine sulfhydrylase, Pseudomonas putida KT2440

>gi|15598303|ref|NP_251797.1| o-succinylhomoserine sulfhydrylase, Pseudomonas aeruginosa PAOl

>gi|53803333 |refj YP l 14900.11 O-succinylhomoserine sulfhydrylase, Methylococcus capsulatus str. Bath

>gi|4191271|emb|C AA09983.1| O-succinylhomoserine sulfhydrylase, Rhizobium etli >gi|41324877|emb|CAF19359.1| O-Acetylhomoserine (Thiol)-Lyase, Corynebacterium glutamicum ATCC 13032 >gi|38233237|ref|NP_939004.1| Putative methionine biosynthesis-related protein,

Corynebacterium diphtheriae NCTC 13129

>gi|41409555 |ref|NP_962391.11 MetB, Mycobacterium avium subsp. paratuberculosis str. klO

>gi|54025566|ref|YP_119808.1| putative O-acetylhomoserine sulfliydrylase, Nocardia farcinica IFM 10152

>gi|15896038|reilNP_349387.1| O-acetylhomoserine sulfliydrylase, Clostridium acetobutylicum ATCC 824

>gi|45657715|ref| YP OO 1801.11 O-acetylhomoserine (thiol) lyase, Leptospira interrogans serovar Copenhageni str. Fiocruz Ll-130

>gi|48866095|ref|ZP_00319952.11 COG2873: O-acetylhomoserine sulfliydrylase, Oenococcus oeni PSU-I

>gi|23024881|ref|ZP_00064071.1| COG2873: O-acetylhomoserine sulfliydrylase, Leuconostoc mesenteroides subsp. mesenteroides ATCC 8293

>gi|33865385|ref|NP_896944.1| putative O-Acetyl homoserine sulfliydrylase Synechococcus sp. WH 8102

>gi|39937649|reilNP_949925.1| homocysteine synthase, Rhodopseudomonas palustris

CGA009

>gi|33240250|reilNP_875192.1| O-acetylhomoserine sulfliydrylase, Prochlorococcus marinus subsp. marinus str. CCMP 1375]

>gi|20091539|ref|NP_617614.1| O-acetylhomoserine (thiol)-lyase, Methanosarcina acetivorans C2A

>gi|48838423|ref|ZP_00295367.1| COG2873: O-acetylhomoserine sulfliydrylase,

Methanosarcina barkeri str. iusaro

In a preferred embodiment of the invention MetB has one or several amino acid changes in conserved region 1 comprising the following amino acids: Xl -X2-X3-Y-X4-R-X5-X6-N-P-T in which

Xl represents S, R, G or is missing X2 represents F, E, N, H, Y, P X3 represents E, I, V, D, R X4 represents G, A, S, T X5 represents Y, F, R, S, T, L, I X6 represents G, T, A, S, M

In another embodiment MetB has one or several amino acid changes in conserved region 2 comprising the following sequence: Xl -X2-X3-X4-X5-G-X6-X7-X8

In which

Xl represents I, H, L, N, R

X2 represents A, S, T, G, L, V,

X3 represents P, N, T, G, E, A, V

X4 represents S, N

X5 represents F, L, I, V

X6 represents G, D

X7 represents C, V, S, T, A,

X8 represents E, K, R

In another embodiment MetB has one or several amino acid changes in conserved region 3 comprising the following sequence: Xl -X2-V/I-X3-X4-P/A-X5-X6-X7-X8 In which

Xl represents S, T X2 represents I, L, T X3 represents D, E, T, A, S, C, V, I X4 represents Q, H, V, I X5 represents A, G, S, K X6 represents I, T, S, R, V, L X7 represents M, T X8 represents S, T

Modified cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfliydrylase comprises at least one of the following mutations and combinations thereof. a valine at position X3 of conserved region 1 a leucine at position X5 of conserved region 1 a leucine or asparagine at position Xl of conserved region 2 an alanine, threonine or valine at position X3 of conserved region 2 an asparagine at position X4 of conserved region 2 a proline at the conserved position P/A an aspartate at position X6 of conserved region 2 a lysine or arginine at position X8 of conserved region 2 a threonine at position X7 of conserved region 3. The meiB/Y/Z genes encoding modified cystathionine-γ-synthase and/or acylhomoserine sulfliydrylase and/or phosphohomoserine sulfliydrylase may be encoded chromosomally or extrachromosomally. Chromosomally there may be one or several copies on the genome that can be introduced by methods of recombination known to the expert in the field. Extrachromosomally the genes may be carried by different types of plasmids that differ with respect to their origin of replication and thus their copy number in the cell. They may be present as 1-5 copies, ca 20 or up to 500 copies corresponding to low copy number plasmids with tight replication (pSClOl, RK2), low copy number plasmids (pACYC, pRSFlOlO) or high copy number plasmids (pSK bluescript II)

The meiB gene may be expressed using promoters with different strength that need or need not to be induced by inducer molecules. Examples are the promoter Ptrc, Ptac, Plac, the lambda promoter cl or other promoters known to the expert in the field.

MetB expression may be boosted or reduced by elements stabilizing or destabilizing the corresponding messenger RNA (Carrier and Keasling (1998) Biotechnol. Prog. 15, 58-64) or the protein (e.g. GST tags, Amersham Biosciences)

The method as claimed by the invention also includes the conduction of the above described enzymatic reaction in cells comprising genes encoding the required enzymes to produce α-ketobutyrate from activated homoserine. In addition, these cells may express genes that are required for the subsequent conversion of α-ketobutyrate to isoleucine.

Cells used in the method according to the invention are eukaryotes or prokaryotes.

In preferred embodiments, cells are microorganisms, preferably selected among S. cerevisiae, E. coli, or C. glutamicum.

The enzymes with γ-eliminase activity expressed in said cells are preferably different from the native enzymes present in the same organism. They may be either mutated genes of the same species or native or mutated genes of other species.

According to a preferred embodiment of the present invention, the cell is transformed to introduce a gene coding for an enzyme with said γ-eliminase activity.

Such gene may be introduced by different means available to the man skilled in the art: modification of the native gene by homologous recombination to introduce mutations in the enzyme encoded by the said gene to enhance γ-eliminase activity; integrating into the genome of the microorganism a foreign gene coding for the selected enzyme known to have high γ-eliminase activity; said foreign gene being under control of regulatory elements functional in the host microorganism; introducing a plasmid comprising a foreign gene coding for the selected enzyme known to have high γ-eliminase activity under control of regulatory elements functional in the host microorganism.

When the gene is integrated into the genome of the microorganism, it may advantageously be introduced in a locus selected to replace the native gene.

Methods used to transform cells, and particularly microorganisms, including homologous recombination, are well known in the art.

Production of α-ketobutyrate or the derived isoleucine may be further increased by enhancing the expression of one or the following genes. Enhancing in this context means increasing the expression of a said gene and thus the activity of the corresponding enzyme:

Gene genbankentry name aceK gl790446 isocitrate dehydrogenase kinase/phosphatase ppc gl790393 phosphoenolpyruvate carboxylase pps gl787994 phosphoenolpyruvate synthase

UvB gl790104 acetohydroxy acid synthase I, large subunit

?7vN gl790103 acetohydroxy acid synthase I, small subunit

UvG gl790202 acetohydroxy acid synthase II, large subunit gl790203

UvM gl790204 acetohydroxy acid synthase II, small subunit ilvl gl786265 acetohydroxy acid synthase III, large subunit

HvK gl786266 acetohydroxy acid synthase III, small subunit thrA gl786183 homoserine dehydrogenase/ aspartokinase pyc U51439 pyruvate carboxylase or its homologs metL gl790376 homoserine dehydrogenase/aspartokinase lysC gl790455 apartokinase asd gl789841 aspartate semialdehyde dehydrogenase aspC gl787159 aspartate aminotransferase thrB gl786184 Homoserine kinase

Production of α-ketobutyrate or the derived isoleucine may be further increased by decreasing or completely eliminating the expression of one or the following genes: Gene genbank entry name asp A gl 790581 aspartate ammonia lyase pck gl 789807 phosphoenolpyruvate carboxykinase ackA gl 788633 acetate kinase pta gl788635 phosphotransacetylase acs gl790505 acetate synthase aceE gl786304 pyruvate deydrogenase El aceF gl786305 pyruvate deydrogenase E2 lpd gl786307 pyruvate deydrogenase E3 sucC gl786948 succinyl-CoA synthetase, beta subunit sucD gl786949 succinyl-CoA synthetase, alpha subunit pykA gl788160 pyruvate kinase II pykF gl787965 pyruvate kinase I poxB gl787096 pyruvate oxidase aroF gl788953 DAHP synthetase aroG gl786969 DAHP synthetase aroH gl787996 DAHP synthetase cysA gl788761 sulfate permease cysXJ gl788764 cysteine transport system cysW gl788762 membrans bound sulphate transport system cysZ gl788753 ORF upstream of cysK. cysN gl789108 ATP suliurylase cysO gl789109 sulfate adenylyltransferase cysC gl789107 adenylylsulfate kinase cysH gl789121 adenylylsulfate reductase cysl gl789122 sulfite reductase, alpha subunit cysJ gl789123 sulfite reductase, beta subunit cysE gl790035 serine acetyltransferase cysK. gl788754 cysteine synthase cysM g2367138 O-acetyl-sulfhydrolase serA gl789279 phosphoglycerate dehydrogenase serB gl790849 phosphoserine phosphatase serC gl787136 phosphoserine aminotransferase glyA gl788902 serine hydroxymethyltransferase metF gl790377 5, 10-Methylenetetrahydrofolate reductase metB gl790375 Cystathionine gamma-synthase metC gl789383 Cystathionine beta-lyase meiΑ gl790450 B 12-dependent homocysteine-NS-methyltetrahydrofolate transmethylase metE g2367304 Tetrahydropteroyltriglutamate methyltransferase metF gl790377 5, 10-Methylenetetrahydrofolate reductase metR gl790262 Positive regulatory gene for metE and metR and autogenous regulation meύ gl790373 repressor of the methionine operon astA gl788043 Arginine succinyltransferase dapA gl788823 Dihydrodipicolinate synthase

The invention also concerns the process for the production of α-ketobutyrate or its derivatives in particular isoleucine. α-ketobutyrate or its derivatives are usually prepared by fermentation of the designed bacterial strain.

According to the invention, the terms 'culture' and 'fermentation' are used indifferently to denote the growth of a microorganism on an appropriate culture medium containing a simple carbon source.

According to the invention a simple carbon source is a source of carbon that can be used by those skilled in the art to obtain normal growth of a microorganism, in particular of a bacterium. In particular it can be an assimilable sugar such as glucose, galactose, sucrose, lactose or molasses, or by-products of these sugars. An especially preferred simple carbon source is glucose. Another preferred simple carbon source is sucrose.

Those skilled in the art are able to define the culture conditions for the microorganisms according to the invention. In particular the bacteria are fermented at a temperature between 20⁰C and 55°C, preferentially between 25°C and 40⁰C, and more specifically about 30⁰C for C. glutamicum and about 37°C for E. coli.

The fermentation is generally conducted in fermenters with an inorganic culture medium of known defined composition adapted to the bacteria used, containing at least one simple carbon source, and if necessary a co-substrate necessary for the production of the metabolite.

In particular, the inorganic culture medium for E. coli can be of identical or similar composition to an M9 medium (Anderson, 1946, Proc. Natl. Acad. ScL USA 32:120-128), an M63 medium (Miller, 1992; A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York) or a medium such as defined by Schaefer et al. (1999, Anal. Biochem. 270: 88-96).

Analogously, the inorganic culture medium for C. glutamicum can be of identical or similar composition to BMCG medium (Liebl et al., 1989, Appl. Microbiol. Biotechnol. 32: 205-210) or to a medium such as that described by Riedel et al. (2001, J. MoI. Microbiol. Biotechnol. 3: 573-583). The media can be supplemented to compensate for auxotrophies introduced by mutations.

After fermentation α-ketobutyrate or its derivatives are recovered and purified if necessary. The methods for the recovery and purification of α-ketobutyrate or its derivatives such as isoleucine in the culture media are well known to those skilled in the art.

Description of figures:

Fig. 1 Metabolic pathway used for isoleucine production. Four different pathways are indicated: the original pathway that starts with the production of threonine, three new pathways that utilize activated homoserines, i.e. phosphohomoserine, acetylhomoserine and succinylhomoserine.

Fig. 2 Alignment of cystathionine-γ-synthases and acylhomoserine sulfliydrylases and phosphohomoserine sulfhydrylases, as indicated on page 6-8. Highly conserved residues are indicated in red, conserved residues in blue

Example 1:

Construction of an isoleucine producing mutant

To produce isoleucine via activated homoserine, homoserine can be activated by homoserine-transsuccinylase to succinylhomoserine. Subsequently succinylhomoserine is cleaved by cystathionine-γ-synthase/sulfhydrylase to succinate, NH₃, and α-ketobutyrate. In Escherichia coli this reaction sequence requires the expression of the genes metA and meiB, encoding homoserine-transsuccinylase and cystathionine-γ-synthase/sullhydrylase. Both genes are repressed by the repressor protein MetJ and its corepressor S-adenosyl methionine (SAM). In addition, the homoserine transsuccinylase activity is feedback controlled by methionine and SAM. To bypass these repression mechanisms a AmetJ mutant was constructed and feedback resistant mutants of MetA (encoded by met A^) were introduced into the Amefl mutant. These constructions have been described in patent application PCT IB2004/001901.

Under aerobic conditions threonine is deaminated to α-ketobutyrate by the enzyme threonine deaminase encoded by the UvA gene. To demonstrate that isoleucine can truly be produced via activated homoserine, the strain ΔmetBJ metA^®* was constructed starting with the metA^&τ already described (metA* 11 in this case). To inactivate the metB and met] gene the homologous recombination strategy described by Datsenko & Wanner (2000) was used. This strategy allowed the insertion of a chloramphenicol resistance cassette, while deleting most of the genes concerned. For this purpose the following oligonucleotides were used: DmetJR (SEQ ID NO 2) tgacgtaggcctgataagcgtagcgcatcaggcgattccactccgcgccgctcttttttgctttagtattcccacg tctcTGTAGGCTGGAGCTGCTTCG with

- a region (lower case) homologous to the sequence (4125596-4125675) of the gene meti (reference sequence on the website http://genolist.pasteur.fr/Colibri/),

- a region (upper case) for the amplification of the chloramphenicol resistance cassette (reference sequence in Datsenko, K.A. & Wanner, B.L., 2000, PNAS, 97: 6640-6645), DmetJBF (SEQ ID NO 3) tatgcagctgacgacctttcgcccctgcctgcgcaatcacactcatttttaccccttgtttgcagcccggaagcca ttttCAGGCACCAGAGTAAACATT with

- a region (lower case) homologous to the sequence (4127460-4127381) of the gene meiB and a region (4126116-4126197) homologous to the promoter ofmetL

- a region (upper case) for the amplification of the chloramphenicol resistance cassette.

The oligonucleotides DmetJBR and DmetJBF were used to amplify the chloramphenicol resistance cassette from the plasmid pKD3. The PCR product obtained was then introduced by electroporation into the strain MGl 655 metA^&τ (pKD46), described in patent application PCT IB2004/001901, in which the expressed Red recombinase enzyme permited the homologous recombination. The chloramphenicol resistant transformants were then selected and the insertion of the resistance cassette was verified by a PCR analysis with the oligonucleotides MetJR and MetJF defined below. The strain retained was designated MG1655 ΔmetJB::Cm metA^&τ

MetJR (SEQ ID NO 4): ggtacagaaaccagcaggctgaggatcagc (homologous to the sequence from 4125431 to 4125460).

MetLR (SEQ ID NO 5): aaataacacttcacatcagccagactactgccaccaaattt (homologous to the sequence from 4127500 to 4157460).

The chloramphenicol resistance cassette was then eliminated. The plasmid pCP20 carrying recombinase FLP acting at the FRT sites of the chloramphenicol resistance cassette was introduced into the recombinant strains by electroporation. After a series of cultures at 42°C, the loss of the chloramphenicol resistance cassette was verified by a PCR analysis with the same oligonucleotides as those used previously.

Subsequently the UvA gene was deleted using the same strategy and the ibllowing oligonucleotides: DiIvAF (SEQ ID NO 6) GgctgactcgcaacccctgtccggtgctccggaaggtgccgaatatttaagagcagtgctgcgcgcgccggtttacgaggTGTAG GCTGGAGCTGCTTCG with

- a region (lower case) homologous to the sequence (3952954-3953033) of the gene UvK (reference sequence on the website http://gcnolist.pastcur.fr/Colibri/),

- a region (upper case) for the amplification of the chloramphenicol resistance cassette (reference sequence in Datsenko, K.A. & Wanner, B.L., 2000, PNAS, 97: 6640-6645), DiIvAR (SEQ ID NO 7) cctgaacgccgggttattggtttcgtcgtggcaatcgtagcccagctcattcagccgggtttcgaaatccggttcatggCATATGA ATATCCTCCTTAG with

- a region (lower case) homologous to the sequence 3954478-3954400) of the gene HvK

The oligonucleotides DiIvAR and DiIvAF were used to amplify the chloramphenicol resistance cassette from the plasmid pKD3. The PCR product obtained was then introduced by electroporation into the strains MGl 655 ΔmetBJ metK^ (pKD46) and ΔmetJ metK^ (pKD46) in which the expressed Red recombinase enzyme permitted the homologous recombination. The chloramphenicol resistant transformants were then selected and the insertion of the resistance cassette was verified by a PCR analysis with the oligonucleotides ilvAR and ilvAF defined below. The resulting strains are ΔmetBJ metK^ (metK* 11 in this case) ΔilvA and ΔmetJ metK³* (metK*l 1 in this case) ΔilvA. ilvAR (3954693-3954670) (SEQ ID NO 8): gccccgaaccggtgcgtaaccgcg ilvAF (3952775-3952795) (SEQ ID NO 9): ggtaagcgatgccgaactggc

In addition to HvA the threonine dehydratase (TdcB) is known to catalyze the deamination of threonine to α-ketobutyrate under anaerobic or microaerobic conditions. To eliminate the possible contribution of this enzyme to α-ketobutyrate production the gene was deleted from the genome of the strains ΔmetJ metK^ AiIvK and ΔmetBJ metK^ AiIvK. TdcB is part of the operon tdcABCDEFG that was deleted in a similar way as described for previous mutants using the four oligonucleotides described below. DtdcGR and DtdcAF were used to amplify the cassette and tdcGR and tdcGF for the verification. DtdcGR (3255915-3255993) (SEQ ID NO 10) gctgacagcaatgtcagccgcagaccactttaatggccagtcctccgcgtgatgtttcgcggtatttatcgttcatatcCAT ATGAA TATCCTCCTTAG DtdcAF (3264726-3264648) (SEQ ID NO 11) GgtaattaacgtaggtcgttatgagcactattcttcttccgaaaacgcagcacctggtagtctttcaggaagtcattagTGTAGGCTGGAG

CTGCTTCG tdcGR (3255616-3255640) (SEQ ID NO 12) gcgtctgcaatgacgcctttattcg tdcAF (3264922-3264899) (SEQ ID NO 13)

Cgccataaaatatggttatccccg

Into the resulting strains Ameii metA^ AiIvA ΔfcfcABCDEFG and AmeiBJ metA^ AiIvA ΔfcfcABCDEFG the plasmid pSBl (for construction see below) was introduced to boost the production of homoserine, which can be transformed into α-ketobutyrate via its activated form.

To further boost the production of homoserine a thrA* allele with reduced feed-back resistance to threonine was expressed from the plasmid pCL1920 (Lerner & Inouye, 1990, NAR 18, 15 p 4631) using the promoter Vtrc. For the construction of plasmid pSBl (pMElOl- thrA* 1) thrA was PCR amplified from genomic DNA using the following oligonucleotides: AspHlthrA (SEQ ID NO 14): ttaTCATGAgagtgttgaagttcggcggtacatcagtggc SmalύxA (SEQ ID NO 15): ttaCCCGGGccgccgccccgagcacatcaaacccgacgc The PCR amplified fragment was cut with the restriction enzymes BspϊU and Smaϊ and cloned into the Ncoϊ I Smaϊ sites of the vector pTRC99A (Stratagene). For the expression from a low copy vector the plasmid pMElOl was constructed as follows. The plasmid pCL1920 was PCR amplified using the oligonucleotides PMElOlF and PMElOlR and the BstZllϊ- Xmnϊ fragment from the vector pTRC99A harboring the lacϊ gene and the Ptrc promoter was inserted into the amplified vector. The resulting vector and the vector harboring the thrA gene were restricted by Apaϊ and Smaϊ and the thrA containing fragment was cloned into the vector pMElOl. To relieve ThrA from feed-back inhibition the mutation F318S was introduced by site-directed mutagenesis (Stratagene) using the oligonucleotides ThrAF F318S for and ThrAR F318 S, resulting in the vector pME 101 -thrA* 1 , called pSB 1. PMElOlF (SEQ ID NO 16): Ccgacagtaagacgggtaagcctg PMElOlR (SEQ ID NO 17): Agcttagtaaagccctcgctag

ThrAF F318S (Smal) (SEQ ID NO 18): Ccaatctgaataacatggcaatgtccagcgtttctggcccggg ThrAR F318S (Smal) (SEQ ID NO 19): Cccgggccagaaacgctggacattgccatgttattcagattgg

Recombinant plasmids were verified by DNA sequencing. Example 2:

Fermentation of the isoleucine producing mutants

Strains were initially analyzed in small Erlenmeyer flask cultures using modified M9 medium (Anderson, 1946, Proc. Natl. Acad. Sd. USA 32:120-128) that was supplemented with 5 g/1 MOPS and 5 g/1 glucose. Spectomycin was added if necessary at a concentration of 50 mg/1. An overnight culture was used to inoculate a 30 ml culture to an OD600 of 0.2. After the culture had reached an OD600 of 4.5 to 5, 1,25 ml of a 50% glucose solution and 0.75 ml of a 2 M MOPS (pH 6.9) were added and the culture was agitated for 1 hour.

Extracellular metabolites were analyzed during the batch phase. Amino acids were quantified by HPLC after OPA/Fmoc derivatization and other relevant metabolites were analyzed using GC-MS after silylation.

The strain MG 1655 metA^Ατ AmetB] AiIvA pSBl was grown in the presence of either 2 mM methionine or 2 mM methionine and 2 mM isoleucine. In the absence of isoleucine the strain failed to grow demonstrating that meiB is required for the production of isoleucine if the threonine deaminase encoding gene UvA is deleted.

The following results were obtained for the strains, MGl 655 Ameύ metA^&τ pSBl and MG1655 Ameύ metA^ AiIvA ΔfcfcABCDEFG pSBl grown without the addition of amino acids. Clearly the deletion of ΔilvA ΔtdcABCDEFG does not reduce isoleucine production indicating that isoleucine is produced by γ-elimination catalyzed by the MetB enzyme.

Table 1. Concentrations of extracellular metabolites (mmol/gDW) after batch fermentation of strains MG1655 Ameύ metA^ pSBl and and MG1655 Ameύ metA^ AiIvA ΔfcfcABCDEFG pSBl. Abbreviations: ocKB α-Ketobutyrate , KMV 2-keto 3-methyl valerate, ISO isoleucine, THR threonine; ND not detected.

Strains that produced substantial amounts of metabolites of interest were subsequently tested under production conditions in 300 ml fermentors (DASGIP) using a fed batch protocol. For this purpose the fermentor was filled with 145 ml of modified minimal medium containing 10 μM IPTG and inoculated with 5 ml of preculture to an optical density (OD600nm) between 0.5 and 1.2.

The temperature of the culture was maintained constant at 37 ⁰C and the pH was permanently adjusted to values between 6.5 and 8 using an NH₄OH solution. The agitation rate was maintained between 200 and 300 rpm during the batch phase and was increased to up to 1000 rpm at the end of the fed-batch phase. The concentration of dissolved oxygen was maintained at values between 30 and 40% saturation by using a gas controller. When the optical density reached a value between three and five the fed-batch (medium containing 10 μM IPTG) was started with an initial flow rate between 0.3 and 0.5 ml/h and a progressive increase up to flow rate values between 2.5 and 3.5 ml/h. At this point the flow rate was maintained constant for 24 to 48 hours. The media of the fed was based on minimal media containing glucose at concentrations between 300 and 500 g/1.

When the concentration of biomass reached values between 30 and 60 g/1 the fermentation was stopped and the extracellular isoleucine concentration was determined using HPLC.

Table 2. Isoleucine titer of strain MGl 655 AmetJ metA^ AiIvA ΔfcfcABCDEFG pSBl after fed batch fermentation.

Example 3:

Construction of cystathionine-γ-synthase/ sulfhydrylase mutants with increased γ- elimination activity

To increase the activity of cystathionine-γ-synthase/ sulfliydrylase (MetB) for γ- elimination, several mutations were introduced into regions that are involved in the binding of the substrate cysteine.

To this end Escherichia coli metB was PCR-amplified from genomic DNA using the oligonucleotides MetBF and MetBR (numbers in parentheses correspond to positions on the E. coli genome). The PCR fragment was restricted by Pst\ and Hindlϊϊ and cloned into pUC18 into the same restriction sites. MetBF (4125957^125982) (SEQ ID NO 20) Ttagacagaactgcagcgccgctccattcagccatgagatac MetBR (4127500^127469) (SEQ ID NO 21) Cgtaacgcccaagcttaaataacacttcacatcagccagactactgcc

Subsequently mutations were introduced into the Escherichia coli cystathionine-γ- synthase/ sulfliydrylase that resulted in the following amino acid changes: E325L, E325V, E325T, E325W and E325P using the following oligonucleotides. Mutant E325A had been previously obtained by a selection procedure that is described elsewhere (patent FR2851255). A pair of oligonucleotides was used for the introduction of each mutation using site directed mutagenesis according to Stratagene's Quick change™ site directed mutagenesis KIT. The mutations are indicated in the name of the oligonucleotide. E325PF (SEQ ID NO 22) Cgttgtttacgctggcgccgtcattagggggagtggaaag E325PR (SEQ ID NO 23) ctttccactccccctaatgacggcgccagcgtaaacaacg E32LF (SEQ ID NO 24) Cgttgtttacgctggcgctgtcattagggggagtggaaag E325LR (SEQ ID NO 25) Ctttccactccccctaatgacagcgccagcgtaaacaacg E325VF (SEQ ID NO 26) cgttgtttacgctggcggtgtcattagggggagtggaaag E325VF (SEQ ID NO 27) ctttccactccccctaatgacaccgccagcgtaaacaacg E325TF (SEQ ID NO 28) cgttgtttacgctggcgacctcattagggggagtggaaag E325TF (SEQ ID NO 29) ctttccactccccctaatgaggtcgccagcgtaaacaacg E325WF (SEQ ID NO 30) cgttgtttacgctggcgtggtcattagggggagtggaaag E325WR (SEQ ID NO 31) ctttccactccccctaatgaccacgccagcgtaaacaacg

The resulting modified meiB genes were verified by sequencing. Clones were restricted with Pst\ and Hindlϊϊ and the meiB containing fragments cloned into the same sites of vector pSBl. Then plasmids were transformed into the strain MG 1655 AmeiBJ met All AiIvA ΔfcfcABCDEFG. Example 4:

Fermentation of strains with mutant MetB enzymes that increase the ratio of isoleucine to methionine produced

The strains MGl 655 AmetBJ metAM AiIvA ΔfcfcABCDEFG expressing the different metB alleles were analyzed for isoleucine production as described in example 2. The strains with the mutant metB alleles showed an increased ratio of isoleucine to methionine produced, indicating that γ-elimination activity of the corresponding enzymes has been increased while cystathionine-γ-synthase and sulfliydrylase activities remain equal or decrease.

Table 3 Methionine and isoleucine production (mmol/g Dw) of strain DmetBJ metA*l l pME101-tΛrA*l-metBxx, where metBxx is either wildtype metB (metBwt) or metB with the amino acid changes indicated.

Example 5

Fermentation under sulfur limiting conditions α-ketobutyrate production can be farther increased by fermenting strains with high γ-eliminase activity under suliur limiting conditions. These can either be achieved by reducing the assimilation of inorganic sulfur through mutations in the cys operon or by fermenting the strains under sulfur limiting conditions. Under these conditions γ-elimination will be favored and higher isoleucine titers can be obtained.

Claims

1) A method for the enzymatic production of α-ketobutyrate, wherein α- ketobutyrate is obtained by γ-elimination of an activated homoserine with an appropriate enzyme in an appropriate medium.

2) The method as claimed in claim 1, wherein activated homoserine is phosphohomoserine.

3) The method as claimed in claim 2, wherein the appropriate enzyme is cystathionine-γ-synthase and/or phosphohomoserine sulfhydrylase.

4) The method as claimed in claim 3, wherein the cystathionine-γ-synthase and/or phosphohomoserine sulfhydrylase is obtained from plants, Chloroflexaceae or archaea.

5) The method as claimed in one of claims 2 to 4, wherein phosphohomoserine is obtained by enzymatic transformation of homoserine with a homoserine kinase.

6) The method as claimed in claims 4, wherein the homoserine kinase is obtained from E. coli or Corynebacterium glutamicum.

7) The method as claimed in claim 1, wherein activated homoserine is acylhomoserine.

8) The method as claimed in claim 7, wherein the appropriate enzyme is acetylhomoserine sulfliydrylase and/or cystathionine-γ-synthase.

9) The method as claimed in claim 8, wherein the cystathionine-γ-synthase and/or acetylhomoserine sulfhydrylase is obtained from Corynebacterium glutamicum, Bacillus subtilis or Saccharomyces cerevisiae.

10) The method as claimed in one of claims 7 to 9, wherein acylhomoserine is obtained by enzymatic transformation of homoserine with homoserine transacetylase.

11) The method as claimed in claim 10, wherein homoserine transacetylase is feedback resistant to methionine and S-adenosyl-methionine.

12) The method as claimed in one of claims 10 or 11, wherein the homoserine transacetylase is obtained from Corynebacterium glutamicum, Saccharomyces cerevisiae or from a spirochete.

13) The method as claimed in claim 1, wherein activated homoserine is succinylhomoserine.

14) The method as claimed in claim 13, wherein the appropriate enzyme is succinylhomoserine sulfhydrylase and/or cystathionine-γ-synthase. 15) The method as claimed in claim 14, wherein the cystathionine-γ-synthase and/or succinylhomoserine sulihydrylase is derived from Enter obacteriaceae, preferentially from E. coli.

16) The method as claimed in one of claims 13 to 15, wherein succinylhomoserine is obtained by enzymatic transformation of homoserine with homoserine transsuccinylase.

17) The method as claimed in claim 16, wherein homoserine transsuccinylase is feedback resistant to methionine and S-adenosyl-methionine.

18) The method as claimed in one of claims 16 or 17, wherein the homoserine transsuccinylase is obtained from Enterobacteriaceae, preferentially from E. coli.

19) The method as claimed in one of claims 1 to 18, wherein the appropriate enzymes are modified enzymes.

20) The method as claimed in claim 19, wherein modified enzymes expressed from the plasmid pMElOl-thrA* in strain MG1655 AmetBJ metAl 1 AiIvA ΔfcfcABCDEFG have a ratio of isoleucine to methionine produced equal to two or higher.

21) The method as claimed in one of claims 19 or 20, wherein modified cystathionine-γ-synthase and/or acylhomoserine sulihydrylase and/or phosphohomoserine sulihydrylase comprise at least one amino acid mutation when compared to the wild-type sequence in one of the following conserved regions and combinations thereof: conserved region 1 comprising the following amino acids: Xl -X2-X3-Y-X4-R-X5-X6-N-P-T in which

Conserved region 2 comprising the following sequence: Xl -X2-X3-X4-X5-G-X6-X7-X8 In which

Xl represents I, H, L, N, R X2 represents A, S, T, G, L, V, X3 represents P, N, T, G, E, A, V X4 represents S, N X5 represents F, L, I, V X6 represents G, D

X7 represents C, V, S, T, A,

X8 represents E, K, R

Conserved region 3 comprising the following sequence: Xl -X2-V/I-X3-X4-P/A-X5-X6-X7-X8 In which

Xl represents S, T X2 represents I, L, T X3 represents D, E, T, A, S, C, V, I X4 represents Q, H, V, I X5 represents A, G, S, K X6 represents I, T, S, R, V, L X7 represents M, T X8 represents S, T.

22) The method as claimed in claim 21, wherein the modified cystathionine-γ- synthase and/or acylhomoserine sulihydrylase and/or phosphohomoserine sulihydrylase comprise at least one of the following mutations and combinations thereof: a valine at position X3 of conserved region 1 a leucine at position X5 of conserved region 1 a leucine or asparagine at position Xl of conserved region 2 an alanine, threonine, valine, tryptophane, praline or leucine at position X3 of conserved region 2 an asparagine at position X4 of conserved region 2 an aspartate at position X6 of conserved region 2 a lysine or arginine at position X8 of conserved region 2 a proline at the conserved position P/A of conserved region 3 a threonine at position X7 of conserved region 3.

23) The method as claimed in one of claims 1 to 22, wherein α-ketobutyrate is further converted into one of its derivatives, in particular isoleucine.

24) The method as claimed in any one of claims 1 to 23, wherein the enzymatic reaction is conducted in cells comprising genes coding for the appropriate enzymes as described in claims 3, 4, 8, 9, 14, 15, 19, 20, 21 and 22 , eventually genes coding for enzymes as described in claims 5, 6, 10, 11, 12, 16, 17 and 18, and eventually genes encoding enzymes appropriate for the conversion of α-ketobutyrate into one of its derivatives, in particular isoleucine. 25) The method as claimed in claim 24, wherein at least one of the said genes is enhanced, in particular overexpressed.

26) The method as claimed in one of claims 24 or 25, wherein at least one of the following genes is enhanced, in particular overexpressed: aceK,ppc, pps, UvB, UvN, UvG, HvM, Hvϊ, HvH, thrA,pyc, meiL, lysC, asd, aspC, thrB,

27) The method as claimed in one of claims 24 to 26, wherein one or several of the following genes are attenuated, in particular deleted: asp A, pck, ackA, pta, acs, aceE, aceF, Ipd, sucC, sucD, pykA, pykF, poxB, aroF, aroG, αroH, cysA, cysXJ, cysW, cysZ, cysN, cysO, cysC, cysH, cysl, cysJ, cysΕ, cysK, cysM, serA, serB, serC, glyA, metF, meiB, metC, meiR, metE, meiF, metK, meύ, astA, dap A.

28) The method as claimed in one of claims 24 to 27, wherein the cell is a microorganism, preferably selected among E. coli, C. glutamicum or S. cerevisiae.

29) The method as described in claim 28 wherein the microorganism is used for the fermentative production of α-ketobutyrate or its derivatives.

30) The method as claimed in one of claims 1 to 29, wherein α-ketobutyrate or its derivatives, in particular isoleucine, are concentrated in the medium or in the cells of the microorganisms.

31) The method as claimed in one of claims 1 to 30, wherein α-ketobutyrate or its derivatives, in particular isoleucine, are isolated.

32) A cell as disclosed in one of claims 24 to 28.