IMPROVED CYSTEINE-PRODUCING STRAINS

Abstract

Genetically modified microorganism strains for the fermentative production of cysteine provide higher yields of L-cysteine or L-cystine during fermentation. Cysteine production is improved in the genetically modified microorganism strains by attenuating or inactivating phosphoenolpyruvate synthase enzyme activity, alone or in combination with the overexpression of efflux proteins and proteins that reduce feedback inhibition by cysteine and by serine.

Claims

1.-12. (canceled)

13. A microorganism strain suitable for fermentative production of L-cysteine, comprising a genetically modified microorganism strain having inactivated or reduced enzyme activity relative to the activity of the corresponding wild-type enzyme of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database; and increased L-cysteine production relative to a microorganism strain having wild-type enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database, wherein the gene encoding said enzyme activity is ppsA.

14. The microorganism strain of claim 13, wherein the strain is from the Enterobacteriaceae or Corynebacteriaceae family.

15. The microorganism strain of claim 13, wherein the microorganism strain is selected from the group consisting of Escherichia coli, Pantoea ananatis and Corynebacterium glutamicum.

16. The microorganism strain of claim 13, wherein the microorganism strain is selected from the group consisting of Escherichia coli and Pantoea ananatis.

17. The microorganism strain of claim 13, wherein the microorganism strain is a strain of the species Escherichia coli.

18. The microorganism strain of claim 14, wherein the genome of the microorganism strain contains at least one mutation in the ppsA gene.

19. The microorganism strain of claim 18, wherein the mutated ppsA gene is selected from the group consisting of the ppsA gene from Escherichia coli, the ppsA gene from Pantoea ananatis, and a gene homologous to these genes, wherein a gene homologous to these genes is a DNA sequence which is at least 80% identical to these genes.

20. The microorganism strain of claim 19, wherein the coding DNA sequence of the ppsA gene is SEQ ID NO: 5.

21. The microorganism strain of claim 19, wherein the strain overexpresses a serine 0-acetyltransferase protein having a reduced feedback inhibition by cysteine; an efflux gene; and a serine feedback-resistant variant of 3-phosphoglycerate dehydrogenase.

22. The microorganism strain of claim 14, wherein the relative enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database is reduced by at least 25% in this strain in relation to the activity of the corresponding wild-type enzyme.

23. The microorganism strain of claim 14, wherein the relative enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database is reduced by at least 70% in this strain in relation to the activity of the corresponding wild-type enzyme.

24. The microorganism strain of claim 14, wherein the strain has no enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database.

25. The microorganism strain of claim 21, wherein the relative enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database is reduced by at least 25% in this strain in relation to the activity of the corresponding wild-type enzyme.

26. The microorganism strain of claim 21, wherein the relative enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database is reduced by at least 70% in this strain in relation to the activity of the corresponding wild-type enzyme.

27. The microorganism strain of claim 21, wherein the strain has no enzyme activity of the enzyme class identified by the number EC 2.7.9.2 in the KEGG database.

28. A fermentative process for producing L-cysteine, comprising: providing a microorganism strain, selected from the group consisting of Escherichia coli, Pantoea ananatis and Corynebacterium glutamicum and suitable for fermentive production of L-cysteine, wherein the strain comprises inactivated or reduced PpsA enzyme activity relative to the activity of the corresponding wild-type PpsA enzyme, and increased L-cysteine production relative to a microorganism strain having wild-type enzyme activity of the PpsA enzyme; culturing the microorganism strain under fermentation conditions to produce L-cysteine; and collecting the cysteine from the culture.

29. The process of claim 31, wherein the microorganism strain further comprises at least one mutation in a ppsA gene, selected from the group consisting of the ppsA gene from Escherichia coli, the ppsA gene from Pantoea ananatis, and a gene homologous to these genes, wherein a gene homologous to these genes is a DNA sequence which is at least 80% identical to these genes.

30. The process of claim 32, wherein the microorganism strain overexpresses a serine O-acetyltransferase protein having a reduced feedback inhibition by cysteine; the efflux gene; and a serine feedback-resistant variant of 3-phosphoglycerate dehydrogenase.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0148] The figures show the plasmids used in the examples.

[0149] FIG. 1 shows the 3.4 kb vector pKD13 used in Example 1 and Example 2.

[0150] FIG. 2 shows the 6.3 kb vector pKD46 used in Example 1 and Example 3.

[0151] FIG. 3 shows the 5 kb vector pKan-SacB used in Example 3.

[0152] FIG. 4 shows the 4.2 kb vector pACYC184 used in Example 4.

[0153] The invention will be further illustrated by the following examples without being restricted by them:

Example 1: Production of a ppsA Deletion Mutant in Escherichia coli

[0154] The parent strain used for gene isolation and for strain development was Escherichia coli K12 W3110 (commercially available under the strain number DSM 5911 from the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH).

[0155] The target of gene inactivation was the coding sequence of the ppsA gene from E. coli. The DNA sequence of the ppsA gene from E. coli K12 (Genbank GeneID: 946209) is disclosed in SEQ ID NO: 1. Nucleotides 333-2711 (identified by E. coli ppsA) encode a phosphoenolpyruvate synthase protein having the amino acid sequence disclosed in SEQ ID NO: 2 (E. coli PpsA).

[0156] The E. coli ppsA gene was inactivated using Red/ET technology from Gene Bridges GmbH as detailed below (described in the user manual of the “Quick and Easy E. coli Gene Deletion Kit”, see “Technical Protocol, Quick & Easy E. coli Gene Deletion Kit, by Red®/ET® Recombination, Cat. No. K006, Version 2.3, June 2012” and the references cited therein, e.g., Datsenko and Wanner, Proc. Natl. Acad. Sci. USA 97 (2000): 6640-6645). To this end, the following plasmids were used: pKD13, pKD46 and pCP20: [0157] The 3.4 kb plasmid pKD13 (FIG. 1) is disclosed in the “GenBank” gene database under the accession number AY048744.1. [0158] The 6.3 kb plasmid pKD46 (FIG. 2) is disclosed in the “GenBank” gene database under the accession number AY048746.1. [0159] The 9.4 kb plasmid pCP20 is disclosed in Cherepanov and Wackernagel, Gene 158 (1995): 9-14.

[0160] To inactivate the ppsA gene in E. coli W3110 by homologous recombination using the Lambda Red system, the following steps were carried out: [0161] 1. E. coli W3110 was transformed with the plasmid pKD46 (so-called “Red Recombinase” plasmid, FIG. 2) and an ampicillin-resistant clone was isolated (referred to as W3110×pKD46). [0162] 2. A ppsA-specific DNA fragment suitable for inactivation thereof was produced in a PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) with DNA of the plasmid pKD13 (FIG. 1) and the primers pps-5f (SEQ ID NO: 7) and pps-6r (SEQ ID NO: 8). [0163] Primer pps-5f contained 30 nucleotides (nt) from the 5′ region of the ppsA gene (nt 333-362 in SEQ ID NO: 1) and, connected thereto, 20 nt specific for the plasmid pKD13 (referred to as “pr-1” in FIG. 1). [0164] Primer pps-6r contained 30 nt from the 3′ region of the ppsA gene (nt 2682-2711 in SEQ ID NO: 1, in reverse-complementary form) and, connected thereto, 20 nt specific for the plasmid pKD13 (referred to as “pr-2” in FIG. 1). DNA of the plasmid pKD13 was used to produce, using the primers pps-5f and pps-6r, a 1.4 kb PCR product which contained at both the 5′ end and the 3′ end a 30 nt section of DNA that was specific for the ppsA gene from E. coli W3110. Furthermore, the PCR product contained the expression cassette of the kanamycin resistance gene contained in pKD13 and, flanking the 5′ and 3′ ends of the kanamycin expression cassette, so-called “FRT direct repeats” (referred to as “FRT1” and “FRT2” in FIG. 1), short sections of DNA that were used as a recognition sequence for “FLP recombinase” (contained on the plasmid pCP20) in a later working step for removal of the antibiotic marker kanamycin. [0165] 3. The 1.4 kb PCR product was isolated and treated with the restriction endonuclease Dpn I, which is familiar to a person skilled in the art and which only cuts methylated DNA, in order to remove residual pKD13 plasmid DNA. Nonmethylated DNA from the PCR reaction is not degraded. [0166] 4. The 1.4 kb PCR product, which is specific for the ppsA gene and contains an expression cassette for the kanamycin resistance gene, was transformed into E. coli W3110×pKD46 and kanamycin-resistant clones were isolated on LBkan plates at 30° C. LBkan plates contained LB medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl), 1.5% agar and 15 mg/L kanamycin. [0167] 5. Ten of the kanamycin-resistant clones obtained were purified on LBkan plates (i.e., isolation of a clone by singularization) and checked in a PCR reaction to determine whether the kanamycin-resistance cassette had been correctly integrated in the ppsA gene. The genomic DNA used for the PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) was isolated using a DNA isolation kit (Qiagen) from cells from the cultivation of kanamycin-resistant clones of E. coli W3110 in Lbkan medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, 15 mg/L kanamycin). Genomic DNA of the E. coli W3110 wild-type strain was used as control. The primers used for the PCR reaction were pps-7f (SEQ ID NO: 9) and pps-8r (SEQ ID NO: 10). Primer pps-7f contained nt 167-188 from SEQ ID NO: 1, and primer pps-8r contained nt 2779-2800 from SEQ ID NO: 1 in reverse-complementary form. [0168] E. coli W3110 wild-type DNA yielded a DNA fragment of 2630 bp in the PCR reaction, as expected for the intact gene. By contrast, a kanamycin-resistant clone under study yielded a DNA fragment of approx. 1660 bp in the PCR reaction, as expected if the 1.4 kb PCR product had been integrated in the ppsA gene at the sites defined by the primers pps-5f and pps-6r. This result showed that the kanamycin resistance gene had been successfully integrated at the locus of the ppsA gene and that the ppsA gene had thus been inactivated. The clone containing an inactivated ppsA gene was selected and designated W3110-ΔppsA::kan. [0169] 6. To eliminate the kanamycin selection marker, W3110-ΔppsA::kan was transformed with the plasmid pCP20 and transformants were selected at 30° C. The 9.4 kb vector pCP20 is disclosed in Cherepanov and Wackernagel (1995), Gene 158: 9-14. Present on the vector pCP20 is the gene of FLP recombinase. FLP recombinase recognizes the FRT sequences flanking the expression cassette of the kanamycin resistance gene and brings about the removal of the kanamycin expression cassette. To this end, the clones obtained at 30° C. were incubated at 37° C. Under these conditions, the expression of FLP recombinase was induced and the replication of the pCP20 vector was suppressed. [0170] This step produced clones in which the ppsA gene had been inactivated and which had regained sensitivity to kanamycin (so-called “curing” of the antibiotic selection marker). The removal of the kanamycin cassette from the genome of the ΔppsA mutants allows the introduction of further mutations in order to produce double or multiple mutants. W3110-ΔppsA::kan regained kanamycin sensitivity after the treatment with the pCP20 plasmid, which was checked as follows: [0171] by plating on LB and LBkan plates: [0172] Growth on LB plates was positive, whereas growth was no longer observed on LBkan plates, which indicated the successful removal of the kanamycin cassette from the genome. [0173] by PCR reaction: [0174] To this end, genomic DNA was isolated from the kanamycin-sensitive clones (Qiagen DNA isolation kit) and used in a PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) using the primers pps-7f (SEQ ID NO: 9) and pps-8r (SEQ ID NO: 10). E. coli W3110 wild-type DNA yielded a DNA fragment of approx. 2630 bp in the PCR reaction, as expected for the intact gene. By contrast, the kanamycin-sensitive clone yielded a DNA fragment of approx. 300 bp in the PCR reaction, which corresponded to the expected size of the 5′ and 3′ fragments of the inactivated ppsA gene remaining after homologous recombination. [0175] The strain isolated from this step was designated E. coli W3110-ΔppsA. This strain is distinguished by the fact that it contained an inactivated ppsA gene and that said strain regained sensitivity to the antibiotic kanamycin.

Example 2: Production of a ppsA Deletion Mutant in Pantoea ananatis

[0176] The parent strain used for gene isolation and for strain development was Pantoea ananatis (commercially available under the strain number DSM 30070 from the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH).

[0177] The target of gene inactivation was the ppsA gene from Pantoea ananatis. The DNA sequence of the ppsA gene from P. ananatis (Genbank GeneID: 31510655) is disclosed in SEQ ID NO: 3. Nucleotides 417-2801 (identified by P. ananatis ppsA) encode a phosphoenolpyruvate synthase protein having the amino acid sequence disclosed in SEQ ID NO: 4 (P. ananatis PpsA).

[0178] The P. ananatis ppsA gene was inactivated using Red®/ET® technology from Gene Bridges GmbH as detailed below (described in the user manual of the “Quick and Easy E. coli Gene Deletion Kit”, see “Technical Protocol, Quick & Easy E. coli Gene Deletion Kit, by Red®/ET® Recombination, Cat. No. K006, Version 2.3, June 2012” and the references cited therein, e.g., Datsenko and Wanner, Proc. Natl. Acad. Sci. USA 97 (2000): 6640-6645). To this end, use was made of the plasmids pKD13 and pRedET. [0179] The 3.4 kb plasmid pKD13 (FIG. 1) is disclosed in the “GenBank” gene database under the accession number AY048744.1. [0180] The commercially available 9.3 kb plasmid pRedET is disclosed in the user manual of the “Quick and Easy E. coli Gene Deletion Kit”, see “Technical Protocol, Quick & Easy E. coli Gene Deletion Kit, by Red®/ET Recombination, Cat. No. K006, Version 2.3, June 2012.”

[0181] To inactivate the ppsA gene in P. ananatis by homologous recombination using the Lambda Red system, the following steps were carried out: [0182] 1. P. ananatis was transformed with the plasmid pRedET (so-called “Red Recombinase” plasmid) and a tetracycline-resistant clone was isolated (referred to as P. ananatis×pRedET). [0183] 2. A ppsA-specific DNA fragment suitable for inactivation thereof was produced in a PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) with DNA of the plasmid pKD13 (FIG. 1) and the primers ppsapa-3f (SEQ ID NO: 11) and ppsapa-4r (SEQ ID NO: 12). [0184] Primer ppsapa-3f contained 49 nt from the 5′ region of the ppsA gene (nt 417-465 in SEQ ID NO: 3) and, connected thereto, 20 nt specific for the plasmid pKD13 (referred to as “pr-1” in FIG. 1). [0185] Primer ppsapa-4r contained 49 nt from the 3′ region of the ppsA gene (nt 2753-2801 in SEQ ID NO: 3, in reverse-complementary form) and, connected thereto, 20 nt specific for the plasmid pKD13 (referred to as “pr-2” in FIG. 1). [0186] DNA of the plasmid pKD13 was used to produce, using the primers ppsapa-3f andppsapa4r, a 1.4 kb PCR product which contained at both the 5′ end and the 3′ end a 49 nt section of DNA that was specific for the ppsA gene from P. ananatis. Furthermore, the PCR product contained the expression cassette of the kanamycin resistance gene contained in pKD13 and, flanking the 5′ and 3′ ends of the kanamycin expression cassette, so-called “FRT direct repeats” (referred to as “FRT1” and “FRT2” in FIG. 1), short sections of DNA that allow removal of the antibiotic marker kanamycin in ppsA deletion mutants as required. [0187] 3. The 1.4 kb PCR product was isolated and treated with the restriction endonuclease Dpn I, which is familiar to a person skilled in the art and which only cuts methylated DNA, in order to remove residual pKD13 plasmid DNA. Nonmethylated DNA from the PCR reaction is not degraded. [0188] 4. The 1.4 kb PCR product, which is specific for the ppsA gene and contains an expression cassette for the kanamycin resistance gene, was transformed into P. ananatis×pRedET and kanamycin-resistant clones were isolated on LBkan plates at 30° C. LBkan plates contained LB medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl), 1.5% agar and 15 mg/L kanamycin. [0189] 5. A kanamycin-resistant clone was purified on LBkan plates (i.e., isolation of a clone by singularization) and checked in a PCR reaction to determine whether the kanamycin-resistance cassette had been correctly integrated in the ppsA gene. [0190] The genomic DNA used for the PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) was isolated using a DNA isolation kit (Qiagen) from cells from the cultivation of the kanamycin-resistant clone of P. ananatis in Lbkan medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, 15 mg/L kanamycin). Genomic DNA of the P. ananatis wild-type strain was used as control. The primers used for the PCR reaction were ppsapa-1f (SEQ ID NO: 13) and ppsapa-2r (SEQ ID NO: 14). Primer ppsapa-1f contained nt 281-302 in SEQ ID NO: 3, and primer ppsapa-2r contained nt 2901-2922 in SEQ ID NO: 3, in reverse-complementary form. [0191] P. ananatis wild-type DNA yielded a DNA fragment of 2640 bp in the PCR reaction, as expected for the intact gene. By contrast, a kanamycin-resistant clone under study yielded a DNA fragment of approx. 1670 bp in the PCR reaction, as expected if the 1.4 kb PCR product had been integrated in the ppsA gene at the sites defined by the primers ppsapa-3f (SEQ ID NO: 11) and ppsapa-4r (SEQ ID NO: 12). This result showed that the kanamycin resistance gene had been successfully integrated at the locus of the ppsA gene and that the ppsA gene had thus been inactivated. The clone containing an inactivated ppsA Gen was selected and designated P. ananatis-ΔppsA::kan.

Example 3: Production of Escherichia coli W3110-ppsA-MHI

[0192] E. coli W3110-ppsA-MHI, characterized by mutations of the ppsA structural gene in a manner causing attenuation of enzyme activity, was produced by using the combination, known to a person skilled in the art, of Lambda Red recombination and counter-selection screening for genetic modification (see, for example, Sun et al., Appl. Env. Microbiol. (2008) 74: 4241-4245). The DNA sequence of the gene ppsA-MHI is disclosed in SEQ ID NO: 5 (ppsA-MHI), encoding a protein having the sequence as specified in SEQ ID NO: 6 (PpsA-MHI).

[0193] The procedure was as follows: [0194] 1. A 2.6 kb DNA fragment comprising parts of the ppsA WT gene (nt 167 to nt 2800 in SEQ ID NO: 1), i.e., the cds and also 5′ and 3′ flanking sequences, was isolated from genomic DNA of E. coli W3110 by PCR using the primers pps-7f (SEQ ID NO: 9) and pps-8r (SEQ ID NO: 10). [0195] 2. ppsA-MHI was obtained from the ppsA WT gene by successively introducing the mutations into the ppsA WT gene by “site-directed” mutagenesis. This was done using the commercially available cloning kit “QuickChange II Site-Directed Mutagenesis Kit” from Agilent in accordance with the instructions in the user manual. [0196] 3. In order to exchange the ppsA WT gene of E. coli W3110 for ppsA-MHI, the 3.2 kb Kan-sacB cassette was first isolated from the plasmid pKan-SacB (FIG. 3) by PCR using the primers pps-9f (SEQ ID NO: 15) and pps-10r (SEQ ID NO: 16). [0197] The plasmid pKan-sacB contains expression cassettes for both the kanamycin (Kan) resistance gene and the sacB gene encoding the enzyme levansucrase. [0198] The primer pps-9f contained 30 nt starting from the start ATG of the ppsA gene (nt 333-362 in SEQ ID NO: 1) and, connected thereto, 20 nt specific for the plasmid pKan-SacB (referred to as “pr-f” in FIG. 3). [0199] The primer pps-10r contained 30 nt from the stop codon of the ppsA gene (nt 2682-2711 in SEQ ID NO: 1, in reverse-complementary form) and, connected thereto, 21 nt specific for the plasmid pKan-SacB (referred to as “pr-r” in FIG. 3). [0200] 4. E. coli W3110×pKD46 (for production thereof, see Example 1) was transformed with the ppsA-specific 3.2 kb PCR product and kanamycin-resistant clones were isolated. [0201] 5. The clones were seeded onto LBSC plates (10 g/L tryptone, 5 g/L yeast extract, 7% sucrose, 1.5% agar and 15 mg/L kanamycin). [0202] Clones containing an integrated sacB gene produced toxic levan from sucrose, and this led to growth inhibition. Such clones were selected and checked in a PCR reaction to determine whether the Kan-sacB cassette had been correctly integrated in the ppsA gene. The genomic DNA used for the PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) had been obtained previously using a DNA isolation kit (Qiagen) from cells from the cultivation of kanamycin-resistant clones of E. coli W3110 in Lbkan medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, 15 mg/L kanamycin). Genomic DNA of the E. coli W3110 wild-type strain was used as control. The primers used for the PCR reaction were pps-7f (SEQ ID NO: 9) and pps-8r (SEQ ID NO: 10). [0203] E. coli W3110 wild-type DNA yielded a DNA fragment of 2630 nt in the PCR reaction, as expected for the intact gene. By contrast, kanamycin-resistant clones yielded a DNA fragment of approx. 3400 nt in the PCR reaction, as expected if the 3.2 kb PCR product had been integrated in the ppsA gene at the sites defined by the primers pps-9f (SEQ ID NO: 15) and pps-10r (SEQ ID NO: 16). This result showed that the Kan-sacB cassette had been successfully integrated at the locus of the ppsA gene and that the ppsA gene had thus been inactivated. A clone containing an integrated Kan-sacB cassette was selected and designated W3110-ΔppsA::kan-sacB×pKD46. [0204] 6. In the next step, the Kan-sacB cassette was exchanged for the ppsA-MHI gene. To this end, a 2.5 kb DNA fragment was amplified from the ppsA-MHI DNA fragment from step 2 in a PCR reaction (“Phusion™ High-Fidelity” DNA Polymerase, Thermo Scientific™) using the primers pps-11f (SEQ ID NO: 17) and pps-12r (SEQ ID NO: 18). Primer pps-11f contained nt 300-319 in SEQ ID NO: 1, and primer pps-12r contained nt 2743-2763 in SEQ ID NO: 1, in reverse-complementary form. [0205] 7. The 2.5 kb ppsA-MI-II gene was transformed into E. coli W3110-ΔppsA::kan-sacB×pKD46 and clones were selected on LBS plates (10 g/L tryptone, 5 g/L yeast extract, 7% sucrose, 1.5% agar) without kanamycin. Only clones which no longer contained an active sacB gene could grow on LBS plates. [0206] These clones were seeded onto LBkan plates in order to select those clones which also no longer contained an active Kan gene and the growth of which was inhibited in the presence of kanamycin. [0207] Clones exhibiting positive growth in the presence of sucrose and negative growth in the presence of kanamycin were selected and checked in a PCR reaction to determine whether the Kan-sacB cassette had been correctly replaced by the ppsA-MHI gene. [0208] Genomic DNA was obtained using a DNA isolation kit (Qiagen) from cells from the cultivation in LB medium (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl). Genomic DNA of the E. coli W3110 wild-type strain was used as control. The primers used for the PCR reaction were pps-7f (SEQ ID NO: 9) and pps-8r (SEQ ID NO: 10). PCR products of the expected size of 2630 nt were analyzed by DNA sequencing (Eurofins Genomics). Clones containing a correctly integrated ppsA-MI-II gene yielded the DNA sequence as disclosed in SEQ ID NO: 5, encoding a protein corresponding to the sequence from SEQ ID NO: 6. A clone containing a correct ppsA-MI-II gene containing the mutations V126M, R427H and V434I was selected and designated E. coli W3110-ppsA-MHI.

Example 4: Generation of Cysteine Production Strains

[0209] The cysteine-specific production plasmid used was the plasmid pACYC184-cysEX-GAPDH-ORF306-serA317 derived from the parent vector pACYC184 (FIG. 4). pACYC184-cysEX-GAPDH-ORF306-serA317 is a derivative of the plasmid pACYC184-cysEX-GAPDH-ORF306 disclosed in EP 0 885 962 B1. The plasmid pACYC184-cysEX-GAPDH-ORF306 contains not only the origin of replication and a tetracycline resistance gene (parent vector pACYC184), but also the cysEX allele, which encodes a serine 0-acetyltransferase having a reduced feedback inhibition by cysteine, and the efflux gene ydeD (ORF306), the expression of which is controlled by the constitutive GAPDH promoter.

[0210] Furthermore, pACYC184-cysEX-GAPDH-ORF306-serA317 additionally contains the serA317 gene fragment, which is cloned after the ydeD (ORF306) efflux gene and which encodes the N-terminal 317 amino acids of the SerA protein (total length: 410 amino acids). The E. coli serA gene is disclosed in the “GenBank” gene database with the gene ID 945258. serA317 is disclosed in Bell et al., Eur. J. Biochem. (2002) 269: 4176-4184, referred to therein as “NSD:317”, and encodes a serine feedback-resistant variant of 3-phosphoglycerate dehydrogenase. The expression of serA317 is controlled by the serA promoter.

[0211] The strains E. coli W3110, E. coli W3110-ΔppsA, E. coli W3110-ppsA-MHI, P. ananatis and P. ananatis-ΔppsA::kan were each transformed with the plasmid pACYC184-cysEX-GAPDH-ORF306-serA317 (referred to as pCYS in the following examples). Transformation was carried out according to the prior art by means of electroporation, as described in EP 0 885 962 B1.

[0212] Plasmid-bearing transformants were selected on LBtet agar plates (10 g/L tryptone, 5 g/L yeast extract, 5 g/L NaCl, 1.5% agar, 15 mg/L tetracycline). Selected transformants were checked for the transformed pCYS plasmid by plasmid isolation by means of the QIAprep Spin Plasmid Kit (Qiagen) and restriction analysis. Transformants containing a correctly incorporated plasmid pCYS were cultivated to check ppsA enzyme activity (Example 5) and to determine cysteine production (Example 6 and Example 7).

Example 5: Determination of ppsA Enzyme Activity

[0213] What was determined was the ppsA enzyme activity of the E. coli strains W3110, W3110-ΔppsA, W3110-ppsA-MHI, each transformed with the production plasmid pCYS (Example 4). Cells from the shake-flask cultivation of the three strains in 50 ml of SM1 medium (for the composition thereof, see Example 6) were pelleted by centrifugation for 10 min and washed once with 10 ml of 0.9% (w/v) NaCl. The cell pellets were taken up in 10 ml of assay buffer (100 mM Tris-HCl, pH 8.0; 10 mM MgCl.sub.2) and a cell extract was prepared.

[0214] The cell homogenizer FastPrep-24™ 5G from MP Biomedicals was used. To this end, 2×1 ml of cell suspension were disrupted in 1.5 ml tubes prefabricated by the manufacturer and containing glass beads (“Lysing Matrix B”) (3×20 sec at a shaking frequency of 6000 rpm with a sec pause each time between the intervals). The resulting homogenate was centrifuged and the supernatant was used as cell extract for determining activity.

[0215] The protein content of the extract was determined by means of a Qubit 3.0 Fluorometer from Thermo Fisher Scientific using the “Qubit® Protein Assay Kit” according to the manufacturer's instructions.

[0216] To determine ppsA enzyme activity, the phosphate detection kit “Malachite Green Phosphate Assay Kit” from Sigma Aldrich (catalog number MAK307) was used in accordance with the manufacturer's instructions. The basis thereof is the conversion of pyruvate with ATP to form phosphoenolpyruvate in equilibrium reaction (4) by ppsA enzyme activity. This produces stoichiometric amounts of phosphate, which is used for determining activity. [0217] The assays contained 100 μg of cell extract, 4 mM Na pyruvate and 4 mM ATP in 1 ml of assay buffer (100 mM Tris-HCl, pH 8.0; 10 mM MgCl.sub.2). [0218] The various assays were incubated at 30° C. [0219] 0 min, 10 min, 20 min, 30 min and 60 min after the start of incubation, 50 μl of the respective assay were removed, added to 750 μl of H.sub.2O, and lastly admixed with 200 μl of reagent from the “Malachite Green Phosphate Assay Kit”. [0220] After 30 min of incubation, the amount of phosphate formed was determined photometrically by determination of the absorbance at 620 nm, with the aid of a phosphate standard curve and according to the manufacturer's instructions. Lastly, ppsA enzyme activity in U/ml extract (1 U=μmol substrate turnover/min) was determined from the measured amount of phosphate, based on the time of sampling from the respective assay. Specific ppsA enzyme activity was calculated by basing the ppsA enzyme activity on 1 mg of total protein of the cell extract (U/mg protein).

TABLE-US-00001 TABLE 1 Determination of ppsA enzyme activity Specific ppsA Relative enzyme activity (in Strain activity relation to W3110 × pCys) W3110-ΔppsA × pCYS 0.00 U/mg  0% W3110-ppsA-MHI × pCYS 0.42 U/mg 26.8%  W3110 × pCYS 1.58 U/mg 100%

Example 6: Cysteine Production in a Shake Flask

[0221] As a preculture for cultivation in a shake flask, 3 ml of LB medium (10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl) which additionally contained 15 mg/L tetracycline were inoculated with the respective strain and incubated in a shaker at 30° C. and 135 rpm for 16 h. The strains studied were E. coli W3110, W3110-ΔppsA, W3110-ppsA-MHI, and, in a second experiment, P. ananatis and P. ananatis-ΔppsA::kan, each transformed with the production plasmid pCYS (Example 4).

[0222] Main culture: Thereafter, a portion of the respective preculture was transferred to a 300 ml Erlenmeyer flask (baffled) containing 30 ml of SM1 medium containing 15 g/L glucose, 5 mg/L vitamin B1 and 15 mg/L tetracycline.

[0223] Composition of the SM1 medium: 12 g/L K2HPO.sub.4, 3 g/L KH.sub.2PO.sub.4, 5 g/L (NH.sub.4).sub.2SO.sub.4, 0.3 g/L MgSO.sub.4×7 H.sub.2O, 0.015 g/L CaCl.sub.2×2 H.sub.2O, 0.002 g/L FeSO.sub.4×7 H.sub.2O, 1 g/L Na.sub.3 citrate×2 H.sub.2O, 0.1 g/L NaCl; 1 ml/L trace element solution.

[0224] Composition of the trace element solution: 0.15 g/L Na.sub.2MoO.sub.4×2 H.sub.2O, 2.5 g/L H.sub.3BO.sub.3, 0.7 g/L CoCl.sub.2×6 H.sub.2O, 0.25 g/L CuSO.sub.4×5 H.sub.2O, 1.6 g/L MnCl.sub.2×4 H.sub.2O, 0.3 g/L ZnSO.sub.4×7 H.sub.2O.

[0225] The main culture was inoculated with enough preculture to establish an initial cell density OD.sub.600/ml (optical density of the main culture, measured at 600 nm) of 0.025/ml. Starting from this, the entire 30 ml batch was incubated at 30° C. and 135 rpm for 24 h.

[0226] After 24 h, samples were taken and the cell density OD.sub.600/ml and the total cysteine content in the culture supernatant were determined, the colorimetric assay by Gaitonde (Gaitonde, M. K. (1967), Biochem. J. 104, 627-633) being used for quantitative determination of cysteine. It should be borne in mind that, under the highly acidic reaction conditions, this assay does not distinguish between cysteine and the condensation product of cysteine and pyruvate, 2-methylthiazolidine-2,4-dicarboxylic acid (thiazolidine), that is described in EP 0 885 962 B1. L-cystine, which is formed by oxidation of two cysteine molecules according to equation (2), is likewise detected as cysteine in the assay by reduction with dithiothreitol in dilute solution at pH 8.0. The results are reported in Table 2 for the E. coli strains mentioned and in Table 3 for the P. ananatis strains.

TABLE-US-00002 TABLE 2 Cell density and total cysteine content after a culture time of 24 h in a shake flask Cell density Cysteine Strain OD.sub.600/ml (g/L) W3110 7.0 0.00 W3110 × pCYS 3.4 0.46 W3110-ppsA-MHI × pCYS 4.8 0.73 W3110-ΔppsA × pCYS 5.2 0.72

TABLE-US-00003 TABLE 3 Cell density and total cysteine content after a culture time of 24 h in a shake flask Cell density Cysteine Strain OD.sub.600/ml (g/L) P. ananatis × pCYS 2.5 0.09 P. ananatis-ΔppsA::kan × pCYS 2.4 0.31

Example 7: Cysteine Production in a Fermenter

[0227] A comparison was made between E. coli W3110×pCYS, W3110-ppsA-MHI×pCYS and W3110-ΔppsA×pCYS in production-scale fed-batch fermentation.

[0228] Preculture 1:

[0229] 20 ml of LB medium containing 15 mg/L tetracycline were inoculated with the respective strain in a 100 ml Erlenmeyer flask and incubated on a shaker (150 rpm, 30° C.) for 7 h.

[0230] Preculture 2:

[0231] Thereafter, the entire preculture 1 was transferred to 100 ml of SM1 medium supplemented with 5 g/L glucose, 5 mg/L vitamin B1 and 15 mg/L tetracycline (for the composition of SM1 medium, see Example 6).

[0232] The cultures were shaken in Erlenmeyer flasks (1 L volume) at 30° C. for 17 h at 150 rpm (Infors incubator shaker). Following this incubation, the cell density OD.sub.600/ml was between 3 and 5.

[0233] Main Culture:

[0234] Fermentation was carried out in a “DASGIP® Parallel Bioreactor System for Microbiology” fermenter from Eppendorf. Culture vessels with a total volume of 1.8 L were used. The fermentation medium (900 ml) contained 15 g/L glucose, 10 g/L tryptone (Difco), 5 g/L yeast extract (Difco), 5 g/L (NH.sub.4).sub.2SO.sub.4, 1.5 g/L KH.sub.2PO.sub.4, 0.5 g/L NaCl, 0.3 g/L MgSO.sub.4×7 H.sub.2O, 0.015 g/L CaCl.sub.2)×2 H.sub.2O, 0.075 g/L FeSO.sub.4×7 H.sub.2O, 1 g/L Na.sub.3 citrate×2 H.sub.2O and 1 ml of trace element solution (see Example 6), 0.005 g/L vitamin Bl and 15 mg/L tetracycline.

[0235] The pH in the fermenter was initially adjusted to 6.5 by pumping in a 25% NH.sub.4OH solution. During the fermentation, the pH was maintained at a value of 6.5 by automatic correction with 25% NH.sub.4OH. For inoculation, 100 ml of preculture 2 were pumped into the fermenter vessel. The initial volume was therefore about 1 L. The cultures were initially stirred at 400 rpm and aerated with compressed air sterilized via a sterile filter at an aeration rate of 2 vvm (volume of air per volume of culture medium per minute). Under these starting conditions, the oxygen probe was calibrated to 100% saturation prior to inoculation.

[0236] The target value for the O.sub.2 saturation during the fermentation was set to 30%. After the O.sub.2 saturation had fallen below the target value, a regulation cascade was started in order to bring the O.sub.2 saturation back up to the target value. This involved first increasing the gas supply continuously (to a maximum of 5 vvm) and then increasing the stirring speed continuously (to a maximum of 1500 rpm).

[0237] The fermentation was carried out at a temperature of 30° C. After a fermentation time of 2 h, a sulfur source in the form of a sterile 60% (w/v) stock solution of sodium thiosulfate×5 H.sub.2O was fed in at a rate of 1.5 ml per hour.

[0238] Once the glucose content in the fermenter had fallen from an initial 15 g/L to approx. 2 g/L, a 56% (w/w) glucose solution was continuously metered in. The feeding rate was adjusted such that the glucose concentration in the fermenter no longer exceeded 2 g/L from then on. Glucose was determined using a glucose analyzer from YSI (Yellow Springs, Ohio, USA).

[0239] The fermentation time was 48 h. Thereafter, samples were taken from the fermentation batch and separate determination of the content of L-cysteine and the derivatives derived therefrom in the culture supernatant (primarily L-cysteine and thiazolidine) and in the precipitate (L-cystine) was carried out. For this purpose, use was made of the colorimetric assay by Gaitonde in each case (Gaitonde, M. K. (1967), Biochem. J. 104, 627-633). The L-cystine present in the precipitate first had to be dissolved in 8% (v/v) hydrochloric acid before it could be quantified in the same way. Lastly, the total amount of cysteine was determined as the sum total of cysteine in the pellet and in the supernatant.

[0240] As summarized in Table 4, the cell density OD.sub.600/ml of the strains studied was comparable, although somewhat higher for the control strain W3110×pCYS. By contrast, volume production of cysteine (in g/L) was significantly higher both in W3110-ppsA-MHI×pCYS and in W3110-ΔppsA×pCYS (by a factor of approx. 3) than in the control strain W3110×pCYS containing the wild-type ppsA gene.

[0241] Under the controlled fermentation conditions, the result therefore achieved for the production scale is that attenuation of activity or inactivation in respect of ppsA enzyme activity leads to significantly improved cysteine production and is therefore a suitable measure for improving strains, which result has not been described previously and is also unexpected for a person skilled in the art on account of the prior art.

TABLE-US-00004 TABLE 4 Cell density and total cysteine content after a culture time of 24 h in a fermenter Cell density Cysteine Strain OD.sub.600/ml (g/L) W3110 × pCYS 95.6 8.7 W3110-ppsA-MHI × pCYS 85.0 26.4 W3110-ΔppsA × pCYS 85.4 25.0

ABBREVIATIONS USED IN THE FIGURES

[0242] bla: Gene conferring resistance to ampicillin ((3-lactamase) [0243] rrnB term: rrnB terminator for transcription [0244] kanR: Gene conferring resistance to kanamycin [0245] ORI: Origin of replication [0246] pr-1: Binding site 1 for primer [0247] pr-2: Binding site 2 for primer [0248] FRT1: Recognition sequence 1 for FLP recombinase [0249] FRT2: Recognition sequence 2 for FLP recombinase [0250] araC: araC gene (repressor gene) [0251] P araC: Promoter of the araC gene [0252] P araB: Promoter of the araB gene [0253] Gam: Lambda phage Gam recombination gene [0254] Bet: Lambda phage Bet recombination gene [0255] Exo: Lambda phage Exo recombination gene [0256] ORI101: Temperature-sensitive origin of replication [0257] RepA: Gene for plasmid replication protein A [0258] sacB: Levansucrase gene [0259] pr-f: Binding site f for primer (forward) [0260] pr-r: Binding site r for primer (reverse) [0261] OriC: Origin of replication C [0262] IHF: Binding site for DNA binding protein IHF (“Integration Host Factor”) [0263] CamR: Gene conferring resistance to chloramphenicol [0264] TetR: Gene conferring resistance to tetracycline [0265] P15A ORI: Origin of replication