7. NCgl 1021 Cgl 1066; transposase
8. NCglI464 Cgll521; putative transposase
9. NCgn496 Cgl 1557; transposase 10.NCgU662 Cgll733; putative transposase 11. NCgll664 Cgl 1734; transposase I2.NCgI2131 CgI22l2; tran^osase 13.NCgl2284 Cgl2367; transposase
14. NCgI2392 CgI2479; putative ttmtsposase l5,NCgl24l8 Cgl2S04; putative ttaaq>08asc 16. NCgl2420 Cgl2S06; putative transposase l7.NCgi2460 Cgl2548; predicted transposase l8.NCgl2542 Cgl2631; predicted transposase 19. NCgl2665 Cgl276l; putative transposase 20.NCgl2748 Cg^845; putative transposase 21.NCgl2850 Cgl2951; predicted transposase
[0159] Tlie IS or artificial transposon may be introduced into corynefonn bacteria using a suitable vector, for example, a plasmid repUcable in corynefomi bacteria. Specific examples of the plasmid include pHM15l9 (Agric. Biol. Chem., 48,2901-2903 (1984)), pAM330 (Agiic. Biol. OiMn., 48,2901-290^ (1984)), and plasmids obtwnod by modifying these to cany adtug-issistant gene. Furttennore, to efficiently amplify tbt IS or artificial

32
tran^son on the chromosome, a plasmid havmg a temperature-sensitive replication drigin as described in the above (1) Is preferably used (see JP-A-5-7491). The parent strain used for ttiis screening is preferably a strata that cannot cause accumulation of L-glutamic acid in the presence of excess biotin, for example, the ATCC13869 strain, which is a wild-type Strain of C. g/MroOT/c«m.
[0160] As the method of introducing the plasmid canying the IS or artificial transposon into coryneform bacteria, cssnvendon^ly used methods, such as die protoplast method (Gene, 39,281-286 (1985)), electroporation mediod (Bio/Technology, 7,1067-1070 (1989)), and the like may be used.
[0161] Introduction of the IS or artificial transposon carried on the temperature-sensitive plasmid into cotynefonn bacteria can be performed by tratisforming the coryneform bacteria with the plasmid, culturing the transformants at IS'C at which Ifae plasmids can replicate to amplify the IS or artiGcial transposon at sevoal tens to several hundreds copies per cell to enable intnxiuction of the IS or artificial transposon into the chromosome, and Hxwt culturing the cells at 34°C to remove the excess plasmids. Alternatively, a DNA fragment of only IS or artificial tran^ioson or a plasmid vector that cannot r^Iicate in coryneform bacteria (for example, plasmid vector replicable in Escherichia coli) may be u^ to introduce the IS or artificial transposon into Ihe cbmmosome of the coryneform bacteria (JP-A-7-107976, Vartes, A A., Asai, Y., Inui, ML, Kobayashi, M., Kunisu, Y. and Yukawa, R: Mol. Gen. Genet, 245,397-405 (1994)). [0162] The strain which has the IS or artificial transposon on the chromostnne is cultured in a medium containing excess biotin so to select a strain that causes accumulation of L-glutamic acid. By deterrainmg the nucleotide sequence of the yggB gene on the chromosome of this stmm, a coryneform bacterium having a mutant-type yggB gene can be obtained.
[0163] (11-3) Method ofrandomlylntzodudog 8 mut^kmhito the ygg0 gene in vitro
^' Furthermore, the rautant-typc'yggB gene can be obtained by randomiy introducing a mutation into die yggB gene in vitro, introducing the mutated gene into coryneform

33
bacterium, and screening for strains which produce L-gb^mic acid in the presence of excess biotin as a result of the presence of the mutant-type yggB gene. The parent strain useful for screening is preferably a strain that cannot cause accumulation of L-glutamic acid in the presencse of excess biotin, for example, Cotynebacterium gltOamicum ATCC13869 strain, ATCCI3032 strain, ATCC14067 stnun, and Corynebacterlian melassecola ATCCI7965 strain.
[0164] To screen for a mutant-type yggB gene, a yggB-deficient strain is preferably used. The construction of the yggB gene-disrupted strain can be performed by a method similar to the above-mentioned method in which the sacB gene is used. For example, PCR is performed using primers shown in SEQ ID NOS: 39 and 40 and the chromosomal DMA of C. glutamicum ATCC13869 as a template to ptepan an N-termbfd fragment of tiie yggB gene. Similarly, PCR is performed using synthetic DNAs of SEQ ID NOS: 41 and 42 as primers to prejpare a C-tetminal fragment. SEQ ID NOS: 40 and 41 are complementary to each other. Subsequently, PCR is performed usmg a mixture of equimolar amounts of the N-tmninai fragment and Ae C-tominal fragment as a template and syn&etic DNAs of SEQ ID NOS: 39 and 42 as primers to prepare a fragment in whidi an int^nal sequence of yggB gene is deleted.
[0165] The obtained PCR frag^ient is inserted into a plasmid for gene disruption, for example, pBS4S canying the levan sucrase gene. The obtuned plasmid is introduced into the chromosome of corynefotm bacterium, for example, C. gfutanticum ATCC13869 strsun to construct a yggB g»ie-disrupted strain.
[0166] Thai, for example, in vitro mutagenesis of die yggB gene can be pertbtmed as follows. First, yggB is cloned into a plasmid that can replicate m coryneform bacteriunu About 10 )ig of die obtained y^ gene-catiying plasmid is dissolved in a buffer containing mutagens, for example, SOO mM phosphate bufiBsr conttdning 400 mM hydroxylamine and 1 mM EDTA (pH 6.0), and he^ed at IS'C for 60 to 90 minutes to introduce a mutation into the yggB gene. After mutagoiests, the plasmid is desalted with -!SUPREC-02 (Takara Bio INC.) or the lifce, and then introduced into ATCC13869 Ayg^ ^ strain, and transibrmants are scneaed in a mediun^ coi^ainii^ an antibiotic. As a control.

34
yggp gene-carryiftg piasmid without mutagenesis is introduced into the ATCC13869 Aygg^ strain! The eoMsged transformants are inoculated into a medium containing excess biotin and cultured with shaidng, and then the concentnttion of accumulated L-glutamic acid is determined: L-glutamic acid does not substantially accumulate in a medium in whidi the wild-type yggB gene-canying plasmid-introduced strain is cultured, wliereas a significant amount of L-glutamic acid accumulates m a medium in which the mutant-type yggB gene-carrying plasmid-introduced strain is cultured. Whetho: the strain carries a mutant-type yggB gene or not can be confirmed by extracting a piasmid firom the srtram and determining the nucleotide sequence of the yggB gene. [0167] Alternatively, a mutant-type yggB gene can be obtained by ardficially introducing mutations into die yggB gene by such methods as error prone PCR, DNA shuffling, and StEP-PCR (Firth AE, Patrick WM; Bioinfonnatics. 2005 Jun 2; Statistics of protein library construction).
[0168] The methods of introducmg a mutation into the yg^ gene on the chromosome include, besides the above-mentioned method, a m^od of treating a coryneform bacterium with irradiation of X-rays or ultraviolet rays or witii a mutagm such as N-methyl-N'-nitrO'N-nitrosoguanidine, and selecting a strain which jaoduces L-glutandc acid in the presence of excess biotin. Whediw the mutant-type yggB gene has been infaroduced or not can be confirmed by determiiung the nucleotide sequence of the yggB gene on the chromosome.
[0169] Ql-A) Method of screoimg L-glutaraic acid analog-resistant strains
Mutant-type yggB genes may be obtained by culturing a ccnyneform bactnium having a wild-type yggB gene in a medium containing an L-glutamtc acid analog, and selecting L-giutamic acid anaiog-resistant strains wliich can grow in die medium. A parent strain used in this method is preferably a wild-type strain of coryneform bacterium as described above, and may be any strain having a v^dld-type yggB gene, including a strain having a pla^mitf containing a wild-type yggB g«i©. [0170] "L-glutamic acid analogs" as used herdn include y-methyl L-g^utamate,

35
a-methyl glutamic acid, p-hydro5cyglutamic acid, m^iomnesulfoxtmine, glutamic acid-7-monohydroxaraate, 2-amino-4-phosphonobutyrio acid, y-taonoethyl Uglutamate, dimethyl L-glutamate, di-t-butyl L-glutamate, monofiuoroglutamic acid, diethyl L-glutamate, D-glutamic acid, and 4-fluorogIutamic acid, and among these, 4-£luorogiutamic acid is preferably used. For example, L-glutamic acid analog^esistant strains can be obtained as follows. That is, a coiyneform bacterium is inoculated on a minimum medium contuning an L-glutamic acid analog, and colonies that have appeared afier.24-48 hours are collected. Concentration of the L-glutamic acid analog added to the medium is preferably a concentration at which a strain having a non-mutated yggB gene cannot grow and a strain having a mutated yggB gene can grow. Specifically, the concentration of 4-fliuoroglutamic acid is 1.25 raM or more, preferably 2.S niM or more, and more preferably 5 mM or more. For example, "L-glutamic acid analog-resistant strain" as used herein means that when the straui is cultured in a minimum medium containing 4-fluoroglutamic acid where the viable cell count (number of cells capable of forming colonies) of a wild-type strain is suppressed to not more than 1/100 that of when in the absence of 4-^uoroglutamic acid, the strain exhibits 1/10 or more growth of die strain cultured to the absence of 4-fluoroglutamic acid.
[0171] The obtained L-glutamlc acid analog-resistant sfxuins are inoculated hito a liquid medium containing excess biotin and cultured with diaking, followed by measurement of the concentration of L-giutamic acid v/tddi has accumulated in the medium. Whereas a strain hflving a wild-type yggB gene accumulates little L-glutamic acid, some of the L-glutamic acid analog-resistant strains accumulate a significant amount of L-glutamic acid. The yggB gene is amplified fiaai such a strain and the nucleotide sequence thereof is determmed, and ther^y, a novel mutant-type yggB gene can be obtained.
[0172] (m) Mutant-type yggB genes
iHereinafter, specific examples of ttie mutant-4ype yggB gene are described. However,'tiie mutant-type yggB gene of the present invention is not Janited to these genes. [0173] The mutant type yggB gene obtained by the above-menUpncd metiiod is not

36
particularly limited so long^ it has a function to eniiance the L-g!utamic acid-producing ability of a coryneform bactedum in the presence of excess hiotia when it is introduced into the coryneform bacterium. [0174] (III-l) Mutation in the C-terminal regicxi of the yggB gene
This mutation is introduced into the region encoding amino acids 419-S33 of SEQ ED NO: 6, 68,84 or 85, or amino acids 419-529 of SEQ ED NO: 62. For example, m SEQ ID NO: 5, this region corresponds to ttie region consisting of nucleotides 2692 to3035. This mutation may be of any type, so long as it is introduced into the region, and includes point mutations and insertion of an artificial sequence. Among these, mutations which introduce an insertion sequence (IS) or an artificial transposon are preferable. Tbe mutation may cause amino acid substitution (mis-sense mutation), frame-shift, or stop codon (non-sense mutation) as a result of the point mutation, insertion of IS, or transposon.
[0175] OU-l-l) The mutation by insertion of transposable element (2A-1 type mutation)
Examples of a mutaticm in the C-tmninal region include a mutation which inserts an transposable element such as inaction sequence (IS) next to the "G" at position 2691 of SEQ ID NO: 5. The nucleotide sequence of the mutant-type yggB gene laving dtis mutation is shown in SEQ ID NO: 7, and the amino acid sequmce of the mutant type YggB protein encoded by the gene is shown in SEQ Q) NO: 8. The IS inserted into the nucleotide sequence of SEQ ID NO: 7 has high homology to 181207 (GenBank accession No. X96962) and IS719 (GenBank accession No. E12759). In the amino acid sequence of SEQ DO NO: 8, the C-terminal region containing the Val at position 419 and thereaftor of tiieYgg protein (SEQ ID NO: 6) is replaced by a shorter IS-dmved sequence. This type of mutation, mcluding the mutations chang^g or deleting the C-terminal r^on in the amino acid sequence of SEQ ID NOS: 6,62,68, 84 and 85, is named a 2A-I-type mutation. [0176] Furthermore^ the 2A-1 -type mutation also inclines other mutations «Mch introduce anotiier IS or transposon into the same site as the 2A-1 type mutton. The position at v^ich the IS or transposon is inserted may be any position so long as it is

37
located within the above-described region. It is {ffeferable that an IS is inserte^jnto a position where the transposase can readily recognize it, or hot spots whwe an IS is easy to insert
[0177] (ni-1-2) The mutation which results In replacement of the proline with anofter amino acid (66-type and 22-type mutations)
Furdiermore, an example of a mutation in the C-tetmioal region also includes a mutation which results in replacement of the proline in C-tcmiinal region with another amino acid. The prolines at the following positions may be substituted with anoth^ amino acid in the amino acid sequence of SEQ ID NO: 6.
Pro at position 424 (424 in SEQ ID NO: 62,68,84.85)
Pro at position 437 (437 in SEQ ID NO: 62,68,84,85)
Pro at position 453 (453 in SEQ ID NO: 62,68,84,85)
Pro at position 457 (457 in SEQ ID NO: 62,68,84,85)
Pro at position 462 (462 in SEQ ID NO: 62,68,84,85)
Pro at position 469 (469i n SEQ ED NO: 62,68,84,85)
Pro at position 484 (484 in SEQ ID NO: 62,68,84,85)
Pro at position 489 (489 in SEQ ED NO: 62,68,84,85)
Pro at position 497 (497 in SEQ ID NO: 62.68,84,85)
Pro at position 515 (515 in SEQ ED NO: 62,68,84,85)
Pro at position 529 (529 in SEQ ID NO: 68,84,85,525 m SEQ ID NO:62 )
Pro at position 533 (533 in SEQ ID NO: 68,84,85,529 in SEQ ID NO:62 ) [0178] It is considered that the proline residues in the C-tenninal region of tiie YggB protein play an important role in maintainance of three-dimoisionai structure of the YggB protein (Protein Eng. 2002 Jan;15(l):29-33, J Biol Chem. 1991 Dec 25;266(36):24287-94.).
[0179] Especially, replacing the proline at position 424 and/or the proline at position 437 m SEQ ID NOT 6? 62, 68, 84 or 85 wifli anotiier amino add are piefenAI(|., [0180] Examples of tite other amino acid include Ala, Arg, Asp, Asn, Cys^ GIu, Gin,

38
Gly, His, lie. Met, l^u, Lys, Phe, Ser,-*Frp, Tyr, Val, and Thr. As the other amino acid replacing the proline at position 424, hydrophobic amino acids sudi as Ala, Gly, Val, Leu, and He are preferable, and amino acids haying brandied chain such as Leu, Val, and lie are more preferable. An example of a mutation which replaces Pro at position 424 with Leu includes a mutation which replaces "C" at position 1673 with'T" in SEQ ID NO: 67. Tlie nucleotide sequence of the mutant type yggB gene having this mutation b shown in SEQ ID NO: 69 and the amino acid sequence of the mutant type YggB protein encoded by die gene is shown in SEQ ID NO: 70.
[0181] As ^e other amino acid replacing the proline at position 437, amino acids containing a hydroxyl-group such as Thr, Ser, Tyr are prefeti^le, and amino acids liaving Ser are more preferable. An example of a mutation wliich iqiiaces Pro at position 437 with Ser includes a mutation which replaces "C" at position 2745 witii 'T' in SEQ ID N0:5. Moreover, this mutation may be accompanied by the mutation which replaces the "C at position 3060 with "T" in SEQ ID NO: 5. The nucleotide sequence of the mutant type yggB gene having this mutation is shown in SEQ ID NO: 73 and the amino acid sequence of the mutant type YggB protein encoded by the gene is shown in SEQ ID NO: 74.
[0182] (ni-2) Mutton in a transmembrane region of the yggB goie
The YggB protein encoded by the ygg^ gene is presumed to have five transmembrane regions. In the amino acid sequence of the wild-type YggB protein shown in SEQ ID NOs: 6,62,68,84 and 85, the transmembrane regions correspond to amino acids 1 to 23 (first transmembrane region), amino acids 25 to 47 (second transmembrane region), amino acids 62 to 84 (third transmembrane region), amino acids 86.to 108 (fourth transm^nbrane region), and ammo acids 110 to 132 (fiSk transmembrane region), hi SEQ ID NO: S, nucleotides encoding these regions correspond to nucleotides 1437 to 1505, nucleotides 1509 to 1577, nucleotides 1620 to 1688, nucleotides 1692 to 1760, and nucleotides 1764 to 1832, respectively. This type of mutation is prefierably introduced into these regions. Hiis kind of mutation pieforably introduces a substitution, deletion, addition, insertion, or inversion of one (M' morejmino acids into these regions witiiout causing a

39
frame shift mutation and a translation termination. Among these, mis-sense mutations
causing amino acid substitutions in the above-mmtioned regions is ptefnable. The '
number of "several" amino acids to be substituted, deleted, added, ins«ted, or inverted
means 2 to 20, preferably 2 to TO, mote preferably 2 to 5, and still more preferably 2 or 3.
The mutations causing insertion and deletion of one or several amino acids widiout a frame
shift are also preferable, and, more preferably, insertion CM: deletion of 3,6,9, 12,15, 18 or
21 nucleotides, still more preferably, deletion or insertion of 3,6, or 9 nucleotides, and
much more preferably, deletion or insertion of 3 nucleotides.
[0183] Specific examples of a mutation in the transmembrane regions include the
following:
[0184] (III-2-1) Mutation in the first transmembrane region (Al-type mutation)
This type of mutation includes one which introduces one or more amino acids between the leucine at position 14 and die tryptt^han at position 15 in the amino acid sequence shown by SEQ ID NOs: 6,62,68,84 and 85, and m<»e specifically mcludes a mutation whidi introduces three amino acids, for example, Cys-Ser-Leu, between the Leu at position 14 and die tryptophan at position IS. This mutation includes inscation of TTCATTGTG next to the G at portion 1480 in flie wild-type j^ gene of SEQ ID NO: 5. The nucleotide sequence of the mutant-type yggB gene having Uns mutation is shown in SEQ ID NO: 19 and the amino acid sequence of the mutant type YggB protein encoded by the gene is shown in SEQ ID NO: 20.
[0185] (in-2-2) Mutation in the 4th transmembrane region (19-type mutation)
This type of mutation includes r^lacit^ die Ala at position 100 widi another amino acid in the amino add sequence shown in SEQ ID NOs: 6,62,68,84 and 85. Examples of the other amino acid include Arg, Asp, Asn, Cys, Glu, Gin, Gly, Ms, De, MJet, Leu, Lys, Phe, Pro, Sec, Ttp, Tyr, Val, and Thr. Of these, amino acids containing a hydroxji-pwup such as Thr, Ser, and Tyr are prefisrable and, threonine is more preferable. A mutatixyl-gix>up such as Thr, Ser, and Tyr are preferable, and Val or Thr is preferable. This type of mutation includes one which replaces the "C at position 1768 with a *T' in the nucleotide sequence of SEQ ID NO: 5 (L30-type mutation), and a mutation which replaces the "G" at position 837 wifli an "A" in die nucleotide sequence of SEQ ED NO: 61 (8-type mutation). Tlic nucleotide sequence of the mutant-type yggB gene having an L30-type mutation is shown in SEQ ID NO: 23 and the amino acid sequence of the mutant type YggB protein encoded by this gene is shown in SEQ ID NO: 24. The nucleotide sequence of the mutaot-^ype yggB gene having 8-type mutation is shown in SEQ ID NO: 63 and tiie amino acid sequence of the mutant tj^pe Yggp fwotein encoded by this gene is shown in SEQ ID.NO: 64.
[0188] (IV) Equivalents of the mutant-type yggB gojes
The "mutant type yggB gene" used in the present invention may be a fimctionally equivalent gene that is substantially homologous to tiie above-mentioned "mutant type yggB genes", for example, a mutant type gene com|H:ising a nucleotide sequence that is able to hybridize to a nucleotide sequence conq>Iem«itaiy to at least one of the nucleotide sequences sdected from the group consisting.of tfie nucleotides 1437 to 2705 of SEQ ID NO: 7, the nucleotides 1437 to 3044 of SEQ ID MO: 19, the nucleotides 1437 to 3035 of SEQ ID NO: 21, the nucleotides 507 to 2093 of §EQ ID NO: 63, &e nudec^des 403 to

41
2001 of SE(T'ID NO: 69. and the nucleotides 1437 to 3035 of SEQ ID NO: 23. the nupleotides 548 to 2146 of SEQ ED NO: 73 with a probe prepared ftom these nucleotide sequences imder stringent conditions, so long as the gene has a function to enhance L-glutamic acid-producing ability of a ooryneform bacterium in the presence of excess
biotin.
[0189] "Stringent conditions" as used herein are conditions under which a so-called specific hybrid is foimed. and a non-specific hybrid is not formed. Examples of stringent conditions include, those under which DNAs having high homology hybridize to each other, for example. DNAs having a homology of not less than 70%, preferably not less than 80%, more preferably not less tiian 90%, especially preferably not less than 95%. hybridize to c^h other, and DNAs having homology lower than 70% do not hybridise to each otfxer. and those under which DNAs hybridize to each other at a salt concentration with washing typical of Southern hybridization, i.e.. washmgonce or preferably 2-3 times under 1 x SSC, 0.1% SDS at 6ff>C, preferably 0.1 x SSC, 0.1% SDS at 60»C, more preferably O.l x SSC,
0.l%SDSat68''C.
[01901 TTie mutant-type yggB gene used in the present mvention includes a gene
encoding a protein having the amino acid sequence of SEQ ID NOs: 8,20.22.24,64. 70. or 74 whereby one or more amino acids are replaced, deleted, inserted or added at one or
more positions other than the specific amino acid as described above, while maintaining a fiuiction to enhance L-glutamic acid-producing abiUty oFa coryneform bacterium in the presence of excm biotin. Although the number orsevcral" amino acid residues referred to herein may differ depending on positions in the three-dimensional structure or types of amino acid residues of the protein, it may be preferably 2 to 20. more preferably 2 to 10.
particularly preferably 2 to 5.
[01911 The yg^ gene preferably encodes a protein having the above-described specific ammo acid substitution or deletion and having homology of not less than 70%, more preferably not less than 80%. fi«ther m^re preferably not less than 90%. particularly preferably &t less than 95% to die amino acid sequence shown in SEQ m NOs: 8.20.22 24,64; 70. or 74 while maintaining a fonction to enhance L-glutamic acid-producing
or.

42
abUity of a coryneform bacterium in the ptesence of excess Wotin. The above-mentioned substitution is preferably a conservative substitution (neirtral mutation). In the case of aromatic amino acids, conservative substitutions include subatitations of phe, trp, and tyr for each other. In the case ofhydrophobic amino acids, conservative substitutions include substitutions of leu, ile, and val for each other. In (he case of polar amino acids, conservative substitutions include substitutions of gin and asn for each other. In the case of basic amino adds, conservative substitutions include substitutions of arg. lys, and his for each other. In the case of acidic amino acids, conservative substitutions are substitutions of asp and glu for each other. In tiie case of hydroxyl group-containig amino acids, conservative substitutions include substitutions of ser and thr for each other. The conservative substitutions also include: substitution of ser or thr for Ala, substitution of Gin, His or Lys for Arg, substitution of Glu, Gin, Lys, His or Asp for Asn, substitution of Asn, Glu or Gin for Asp, substitution of Ser or Ala for Cys, substitution of Asn. Glu, Lys, His.'Asp or Arg for Ghi. substitution of Gly, Asn. Gta, Lys or Asp for Glu, substitution of Pro'for Gly, substitution of Asn, Lys. Gki. Arg or Tyr for His, substitation of Leu, Met, Val
or Phe for Ile, substitution of He, Met, Val or Phe for Leu, substitution of Asn, Glu. Gin, His or Arg forLys. substitution.of Ue, Leu, Val or Phe for Met, substitution of Trp. Tyr. Met, lie or Leu for Phe, substitution ofThr or Ala for Ser,substitutionofScr or Ala for Thr.
substitution of Phe or Tyr for Trp. substitution of His. Phe or Tip for Tyr and substitution of Met, De or Leu for Val. As mentioned above, tht amino acids shown as Xaa may be substituted m the ammo acid sequence of SEQ ID NO: 85.
[0192] Especially, the following amino acids may be substituted or deleted in ti»e amino acid sequences of SEQ ID NO: 8,20.22,24,64,70, or 74.
Glu at position 48 (preferably replaced by Arg)
Asp at position 275 (preferably replaced by Ser)
Glu at position 298 (preferably replaced by Ala)
Ala at position 343 (preferably replaced by Val). Phe at position 396 (preferably replaced by Ile)
Ser at position 438 (preferably replaced by Gly)

43
Val at positioS'44S (preferably replaced by Ala) -Ala at pasfijon 454 (preferably replaced by Val) Pro at position 457 (preferably rq)laced by Ser), Ser at position 474 (preferably replaced by Asp) Val at position S17 (preferably deleted) Glu at position 518 (preferably deleted) Ala at position 519 (preferably deleted) Pro at position 520 (preferably deleted)
[0193] (V) Methods of introducing the above-described mutant-type yggB genes into corjoieform bacterium
The mutant-type yggB gene having the above-mentioned specific mutation can be obtained by conventional methods including a site-directed mutagenesis technique. The site-directed mutagenesis technique includes an overlap extension PCR method that am|>Ufies a mutant gene using a PCR primer haying a mutation (Urban, A., Neuldrchen, S. and Jaeger, K. E., A rapid and efficient method for site-directed mutagenesis using one-st^ overlap extension PCR. Nucleic Acids Res, 25,2227-8. (1997)). [0194] The coryneform bacterium of the present invention having the abov^-menttoned mutant-type yggB gene can be obtained by introducing the above-mentioned mutant-type yggB gene into a coryneform bacterium. A wild-type yggB gene on a diromosome may be replaced with the mutant-type yggB gene. The mutant-type yggB gene may be introduced into a coryneform bact^um in which a wild-type yg^ gene is disrupted. In addition, as in the case of single cross-over recombinants, the mutant type yggB g«ie may co-exist with a wild-type yggB gene in a coryneform bacterium. For example, the substitution of Hie yggB gene on the chromosome can be performed by usmg, for example, a temperature-sensitive plasmid containing a sacB gene encoding the above-mentioned levan sucrase. Furthermore, to introduce the rautagt type yggB gene Into coryneform bacterium, a vector such as a plasmid replicable in coryaeform bacteinnn or transposon comprising the mutant type yggB gene may be us««l.

44
[0195] To introduce the mutant type yggB gene into flie chromosomall^NA of the coryneform bacteria, it is also possible to perform homologous recombination by targeting a sequence that is present on the chromosomal DNA in multiple copies. Exaniples of such a sequence include a repetitive DNA and an inverted lepeat that is present on the end of a transposable element The mutant type yggB gene may exist in coryneform bacterium in a single copy or multiple copies. The introduction of (he mutant type yggB gate into coryneform bacterium can be confirmed by PCR, Southern hybridization, or the like. [0196] Fuitfaennore, the mutant type yggB gene may be under the control of a potent promoter which is derived from other genes, as described in WO00/1893S. For example, lac promoter, trp promoter, trc promote-, PS2 promoter, and so on are known as potent promoters. It is also possible to introduce substitution of nucleotides into the promoter region of the mutant-type yggB gene so that expression of mutant-type yggB gene is enhanced. The substitution of the expression regulating sequence can be poformed by using, for example, a temperature-sensitive plasmid.
[0197] (VT) L-glutamic acid analog resistance
Furthermore, the coryneform bacterium of the present invoidon may have tnoeased resistance to L-glutamic acid analogs as a result of the introduction of the mutant-type yggB gene of the present invention. "L-glutamic acid analogs" as used herein include ynme^yl L-glutamate, a-methyl glutamic acid, ^ydroxyglutamic acid, methioninesuifoximine, glutainic acid-^-monobydroxamate, 2-anaino-4-phosphonobutyric acid, Y-mono^yl L-glutamate, dimethyl L-glutamale, di-t-butyi L-glutamaie, monofluoroglutamic acid, diethyl L-glutamate, D-glutamic acid, and 4-flHoroglutamic acid. For example, an increase in resistance to L-glutamic acid analogs is confirmed by the fact that when the strain of the presmt invmtion is cultured in a minimum medium containing 4-fIuoroglutamic acid at a concoitration whidi tbs viable cell count (number of ceils capableof fiJrming colonies) of the parrait strain can be suppressed to not mare than l/lOO, the strain exhibits I/IO or more growth as compared to what cultured in die absence of 4-fluoroglutamic acid. Specifically, it is preferable that the strain has a resistance to

45
1.25 mM or more, preferably 2.5 ttiM or more, and more preferably 5 mM or more of 4-fIuorogtutamic acid.
[0198] (Vn) Furtlier modification to inactivate the gene which suppresses the function of mutant-tjpe yggB gene
The coryhefomi bacterium of ifae present invention may be fatttter modified so to inactivate a gene which suppresses the function of nuriant-type yggB gene. The "geae which suppresses the fimction of mutant-type yggB gene" means that L-giutamic acid production by a mutant-type yggB gene-introduced stxun is su]:^res8ed by amplifying die gene in the strain. An example of sudi a gene includes the symA gene (suppressor of yggB mutation). The symA gene is shown as nucleotide numbers 2051306-2051845 of the genome sequence (Genbank Accession No. NC_003450) of corynebact^um glutamicum ATCC13032 strain, and it^stered as NCgl 1867 (NP_601149. hypotiwtical prot..[gi:l9553147]). The symA gene of corynebact«-ium glutamicum ATCC13869 strain is shown in nucleotides 58S-1121 of SEQ ID NO: 86. The symA gene may be a DNA that is able to hybridize with a nucleotide sequ^ice complementary to nucleotides S8S to 1121 of SEQ ID No: 86, or a probe pr^ared from said nucleotides under stringent conditions, as long as tiie DNA suppresses a function of said mutant-type yggB gene in the coryneform bacterium.
[0199] Gene inactivBtion can be performed by (tisrupting the gene, deleting die gene, or modifying it to decrease expression of die gene. Inactivaticxi of the symA gene can be performed using a similar method as the above-described methods for decreasing enzymatic activity.
[0200] (Vin) Puith«: modification to decrease a-tcetoglutarate dehydrogenase activity In die present invention, a coryneform bacterium is preferably modified $o that die activity of a-ketoglutarate ddiydrogenase (hweinafter, refisrred;to as VKGDH") is decreased in addition to the modification using a yggB ^ne. The "«-KGDH activity is decreased" means diat the a-KGDH activity is decreased as compared widi diat of the

46
wild-type strain or non-modified strains, such as the parent strain. The a-KODH activity can be measured according to the method of Shiio et al. (Isamu Shiio and Kyoko. Ujigawa-Takcda, Agric. Biol. Chem., 44(8), 1897-1904,1980). Althou^ it is sufficient that the a-KGDH activity is decreased as compared to a non-modified strains such as a vwld-type strain or a parent strain, it is preferable that the o-KGDH activity is decreased to about 1/2 time or less, prefra^ly about 1/4 time or less, and mote preferably about 1/10 time or less with respect to a wild-type or non-modified strain. Tlie coryneform bacterium of the present invention may not have a detectable activity of a-KGDH. [0201] The coryneform bacterium in which the a-KGDH activity is decreased can be constructed in a similar way as described above.
[0202] For example, a-KGDH activity may be decreased by introdycing a gene encoding the EIo subunit of tiie a-KGDH complex having a mutation in the thiamine pyropbo^hate-binding region (the region encoded by nucleotides 2498 to 2S84 of SBQ ID NO: 43 (686Gly-714Asp of SEQ ID NO: 44)).
[0203] Examples of the strain having a decreased activity of o-KGDH include Brevibacterium lactofermentum AS strain (W095/34672) and Brevibacterium lactofermenium AJ12821 (PERM BP-4172) strata (JP-A-06-237779). When using a coryneform bax^rium carrying the mutant-type ygg^ gene and having decreased a-KGDH activity, either decreasing a-KGDH activity or introducing the mutant type yggB gene may be performed first.
[0204] <2> Method of producing L-glutamic acid
L-giutamic acid can be produced by culturing the coryneform bacterium of the
present inveotioo in a medium to cause accumulation of L-glutamic acid in the medium
and/or in the bacterial cells, and collecting the L-glutamic acid fitom (he medhim and/or the
bacterial cells. In the production method of the present invention, L-glutamic acid is
produced preferably by^^cultoring die corynefonn bacterium of the present 'mvmtion, for
example, at 25 to 40''C for 8 to 120 hours. ^,
[0205] The culture medium may be an ordinary medium that contains a caibon sqwce, a

47
nitrogen source, an inorganic salt, and optiofudly organic raicronutrients such as amino acids and vitamins. Either a synthetic medium or a natural medium may be used. Any kind of carbon and nitrogen source may be used s6 long as they can be utilized by the stain
being cultured.
[0206] Saccharides such as glucose, glycerol, fructose, sucrose, maltose, mannose, galactose, starch hydrolysate, and molasses may be used as the carbon source. In addition, organic acids such as acetic acid and citric acid, and alcohols such as ethanol may also be used alone or in combination as a carbon source. Ammonia, ammonium salts such as ammonium sulfate, ammonium carbonate, ammonium chtoride. ammonium phosphate, and ammonium acetote. nitrates, and the like may be used as the nitrogen source. Amino acids, vitamms, fatty acids, nucleic acids, substances containing peptone, casamino acid, yeast extract, and soybean protein decomposition products may be used m a sli^t amount as the organic nutrients. When an auxotrophic mutant strain that requires an amino acid etc. for growth is used, such a required nutrient is preferably added. Phosphates, magnesium salts, calcium salts, iron salts, manganese salts, and the Uke can be used as inorganic salts. [0207] Surfectants such as Tween40. penicillin, or biotin may be added in an apprcpriate amount dependmg on a strain to be cultured. For exaniple, a sttaiB-having a
mutant-type yggB gene may be cultured in the presence of excess biotin. although such a stram may also be cultured under L-glutamic conditions containing surfectants or penicillin, or when biotin is limited.
[0208] Preferably, aerobic culturing is performed by controllmg the feimentation temperature and adjusting the pH of the culture medium to 3 to 9. When the pH decreases during the culhire. the medium is neutralized by adding alkali such as calcium carbonate or aramoniagas. Culture for about 10 to about 120 hours results in accumulation of a considerable amount of L-glutamic acid ill the medium.
[0209] Furthermore, the culture may be performed by using a liquid medium adjusted to conditions under which the pixKluced-L-glutamicacid crystallizes and precipitates. The
conditions under which ^glutamic acid cryrtallizes include pH 5.0 to 4.0. pRsfcrably pH 4.5 to 4.0. more preferably pH 4.3 to 4.0. pa^icularly preferably pH 4.0 (EP1233069.

48
EP12^070).
[0210] CoUection of L-glutamic acid from the medium after conq}ledon of the culture may be performed by convoitional methods. L-glutamic acid may be collected, for example, by removing bacterial cells from the medium and concentrating L-glutamic acid or by using ion exchange chromatography. When the culture is performed under conditions under which L-glutamic acid crystallizes and precqiitates, the crystallized L-glutamic acid can be collected, for example, by centrifugalion or filtration. In this case, L-glutamic acid dissolved in the medium may also be collected after crystallization of the dissolved L-glutamic acid.
[0211] EXAMPLES
Hereinafter, the present invention is specifically explained by refoxing to the following non-limiting examples.
[0212] Example I
< Construction of a vector for gene disruption carrymg the sacB gen^
[0213] (M) Construction of pBS3
Construction of ^ grate disruption vector carrying die sacB gene was pofcmned by using the method in WO2005/113745 and WO2005/113744. A sacB geoe (SEQ ID NO: 11) was obtained by PCR using a dbromosomai DNA of Bacillus subtilis as a template and oligonucleotides of SEQ ID NOS: 13 and 14 as prim«s. The PCR was performed using LA taq (manu&ctured by TaKaRa) as follows: one cycle of heat r^ention at 94°C for S minutes; and 25 cycles of denaturing at 94*'C for 30 seconds, annealing at 49°C for 30 seconds, and elongation at 72''C for 2 minutes. The (Stained PCR product was purified by a conventional metiiod, and then digested widi BgUI and BamHI and blunt-ended. The fragment was inserted into pHS0299 whidi had been digested with Avail and blunt-ended. The bbtedned DNA was used to transforihn competent cells of Escherichia coli JM109 (man^Bctuted by TAKARA BIO INC.)? Then, the transformed bacterial cells were spread oQ IB agar medium containing 25jtg/'ml Kanamycin Qierrinafter, abbreviated as "Km"),

49
and incubated for one night. Tliereafter, single coloniesiafere isolated as transformants. Plasmids were extracted fix>m the obtained tfansfoimants and the plasmid which had an insert of the object PCR product was named pBS3. Fig. I shows the procedure for construction of pBS3.
[0214] (1-2) Construction of pBS4S
The Smal recognition site in the kanamycin-resistant gene on pBS3 was modified by nucleotide substitution using cross-over PCR without causing amino acid substitution so that pBS3 is not cut by Smal endnuciease. First, PCR was performed using pBS3 as a template and synthetic DNAs of SEQ ID NOS: IS and 16 as primers, to thereby obtain an N-tatnind fragment of the kanamycin-resistant gene. On the othe^hand, to obtain a C-terminal fragment of die kanamycin-resistant gene, PCR was performed using pBS3 as a template and synthetic DNAs of SEQ ID NOS: 17 and 18 as primers. PCR was performed using Pyrobest DNA Polymerase (manufactured by TAKARA BIO INC.) as follows: one cycle of heat retention at 98°C for S minutes; and 2S cycles of denaturing at 98°C for 10 seconds, annealing at ST^'C for 30 seconds, and elongation at 72°C for 1 minute, to obtain the objective PCR product SEQ ID NOS: 16 and 17 are partially complem«ttary to each other and do not contain the Smal recognition site. Then, to obtain a fiiil-length fragm^t of the mutant kanamycin-resistant gene without the Smal recognition site, the above-mentioned N-tenninal and C-teiminal gene products were mixed toffither in substantially equimolar amounts. PCR was performed using the geat products as a template and syndietic DNAs of SEQ ID NOS: IS and 18 as primers to obtam a Smal site-modified kanamycin-resistant gene fragment The PCR was performed using Pyrobest DNA Polymerase (manufactured by TAKARA BIO INC.) as follows: one cycle of heat ratention at PS'C for 5 minutes; and 25 cycles of d«>atttring at PS'C foe 10 seconds, »i^pAaitp£ at S7°C for 30 seconds, and elongation at 72''C for 1.5 minutes, to thereby obtain the object PCR product
p)215] The PCR product was purified by a conventional method, and then digested wth^ Banll and then inserted into die above-described BanlLrecognition site of pBS3. The

50
resulting plasmid%as used to transform competent cells of Escherichia coli JMl69 (available from Takara Bio). That is, the transformed bacterial cells were spread on LB agar medium containing 25 pg/ml of kanamycin, and incubated for one nigfat Tbeteafier, colonies that appeared were sdected as transformants. Plasmids were isolated from the obtained transformants and the plasmid having an insert of the object PCR product was named pBS4S. Fig. 2 shows the procedure for constructing pBS4S.
[0216] Example 2

The nucleotide sequence of the odhA gene encoding the a-ketoglutarate ddiydrogenase of coryneform bacterium has already been idet^fied (Microbiology 142, 3347-33S4, (1996), GenBank accession No. D84102). Based on the nucleotide sequence of the odhA gene, the primers described in SEQ ID KOS: 1 and 1 were designed, and PCR wasperformed using the (ximers and the chromosomal DNA of the ATCC13869 stttun as a template to amplify the internal sequence of the odhA gme. The amplified PCR fragment was completely digested with BamHI and insertedto the BamHI siteof pBS4S constructed in Example I, thereby the plasmid pBS4SAsucAint was obtained (Fig. 3). [0217] pBS4SAsucAint vras introduced mto C. glutamkum ATCC13869 strain by the electric pulse method (JP-A-02-207791) and the tranformed bact«ial cells were spread over CM-Dex agar medium (5g/l glucose, lOg/1 polypqjtone, lOg/1 yeast extract, I g/| J<:H2P04, 0.4 g/1 MgS04-7H20.0.01 g/l FeS04-7H20,0.01 g/\ MnS04-4-5H20,3 g/1 urea, 1.2 g/l soybean protein hydrolysate, and 20 g/i agar, adjusted to pH 7.5 with NaOH: autoclaved at 120°C for 20 minutes) containing 25 ^ml kanamycin. After culturing at 3 l.S°C, PCR was performed usmg each of the chromosome extracted from strains v^ch appeared to confirm that these strains were single cross-over recombinants in which pBS4SAsucAint was incorporated by homologous recombination into the chromosome. Primes each having a sequence (SEQ ID NO:. 3) ^lecific to pBS4S plasmid and a sequence (SEQ "35 NO: 4) complementary to a chromosomal sequence was used for PCR. Since the sequence of pBS4S is absent in a non-recombbant strain, no fragment is

51
amplified from the non-recombinant strain, whtereas a single fragn^it is amplified trom a
single cross-ov« recombinant. • .
[0218] The stagle ctoss-ovet recombinant thus obtained was named 2A-1 strain. The wild-type 13869 strain and 2A-1 sttain were inoculated in 20 ml of a flask medium (30g/l glucose, 15 g/1 ammonium sulfete, 1 gA KH2PO4,0.4 gA MgSOA-THaO, O.Ol g/1 FeS04-7H20,0.01 g/1 MnS04-4-5H20,200 ugA VBl (vitamta Bl), 300 |ig/l Biotin, and 0.48 gA soybean hydrolysates (T-N: total nitrpgen), adjusted to pH 8.0 with KOH: autoclaved at 1I5»C for 10 minutes), followed by addition of 1 gof heiat-sterilized calcium caibonate, and each ofthestiains was culhiredwith shaking at31.5»C. After the sugar was completely consumed, the concentration of ^glutamic acid which had accumulated in the medium was detemuned. The results are shown in Table 1 (OD620 is turbidity at 620nm of culture solution diluted to 101 times, and radicates the cell amount, and Glu (g/L) indicates the amount of accumulated L-glutamic acid). It was found that the 2A-1 sttain produced L-glutamic acid in the presence of an excess amount of biotin, whereas the parent strain AT^CC 13869 did not produce L-glutamic acid at alL

[0219] straitf>

In the 2A-1 strain, the odhA gene on the chromosome was disrupted by pBS4§AsucAinL By curing the plasmid from the chromosome of this strain, the odhA gene coJld^be reverted to the wild-lype one. Although the odhA gae-disrapted stiam grows very slowly in a medium containing no sugar, (b«rodhA gene-revertant strain in

52
which the OdhA gene-reverb to the wild-type one grows weU in a medium containing no sugar such as CM2B (ICTg^ polypeptone, 10 g/1 yeast extract, 5 gANaCl, 10 Mg/I Biotin, 20 g/i agar, adjusted to pH 7.0 \irith KOH). To obtain such a revertant strain, the 2A-1 strain was spread over CM2B agar medium to select growth-improved strains. The growtfa-improved strain which appeared was named 2A-lRand isolated on the CM2B agar medium and the kanamycin-sensitivity of the 2A-1R strain was examined. As a result, it was found that all of the selected strains were kanamycin-sensitive and sucrose-resistant Since the pBS4SAsucAint contains a kanamycin-resistant gene and the sacB gene encoding levan sucrase. strains harboring pBS4SAsucAint exhibit kanamycin-iesistance and suciose-sensitivity. while strains from which pBS4SAsucAint was removed exhibit kanamycin-sensitivity and sucrose-resistance.. Therefore, it was considered that the odhA gene reverted to the wild-type one in the 2A-1R strain. Detemiination of the nucleotide sequence of the OdhA gene confirmed that the strain ha? the wild-type OdhA gene.
[0221] The ability of 2A-IR strain to produce L-glutaraic acid in the presence of an excess amount of biotin was confirmed by the same method as in Example 2. The results are shown in Table2(OD620 is turbidity at 620nm of cuIturesolutiondUuted to 101 times,
and indicates the cell amount, and Glu (gfL) indicates the amount of accumulated L-glutamic acid). Although the accumulation of L-glutamic acid by the 2A-1R stiam was sUghUy decreased as compared to the 2A-1 strain, the 2A-1R strain produced a much higher amount of L-glutamic acid in the presence of an excess amount of biotin than the wild-type ATCCB869 strain (Table 2). In addition, when the shaking culture was continued after the sugar was completely consumed, decomposition of L-glutamic acid was observed in the 2A-IR strain, which proved that the OdhA gene had reverted to the
wild-type in this strain (Fig. 4).

53
[0222]

OD620(xl01) Glu(g/L)
ATCC13869 0.696 0.5
2A-1 0.332 17.1
2A-IR 0.327 14.3
Blank 0.002 0.6
[0223] Example 4
isolation of a geas that is involved in L-glutamic acid production by On the CM2B agar medium, the 2A>1R strain could fomi colonies at substantially the same rate as that of the wild-type strain ATCC 13869. However, on ftie minimum plate medium (20 g/l glucose, 2.64 g/1 ammonium sulfate, 0.5 g/1 KHbPO^, 0.5 ^ KaHPO*, 0.25 g/1 MgS04-7H20,0.01 g/l FeS04-7H20,0.01 g/l MnS04-7HaO, 0.01 g^ CaClj, 0.02 mgA CuSO*, 40 g/i MOPS, 30 mg/1 protocatechuic acid, 200 |ig/I VBrHu, 300 ng/l Biotin, 20 g/1 agar, adjusted to pH 6.7 with NaOH), the 2A-1R strdn sjiowed a coosidaably decreased colony-forming rate as cotnpared to the wild-type ATCC13869 strain. Accordingly, a gene that can recover the growdt of the 2A-1R strain in the minimum medium was found.
[0224] The chromosomiil DNA of the ATCC13869 strain was partially digested with Sau3 AI and ligated to tJie shuttle vector pVK9 that had been digested with BamHI. The obtained plasmld was precipitated with ethanol and used to transform competent cell of £. coli DH5a CTAKARA BIO INC.) by an electric pulse method. pVK9 is a shuttle vector obtained by blunt-ending the Avail site of pHSG299 (TAKARA BIO INC.) and inserting therein a fragment comprismg a sequence automaticaUy rq)Iicable m coryneform bacteria excised with BamHI and JCpnl from pHK4 (JP-A-05-00749I). The transformed cells were spread over an LB agar medium (10 g/l polypeptooe. 5 g/1 yeaA extract, 5 g/1 NaCl, 20 g/1 agar, adjusted to pH 7.0 with NaOH) containmg 25 |ig/ml kanamydn, and cultured at 37°C for one night. On die next day, all of the colonies which appeared were co|lf?cted from the

54
plate wifii a platinum loop and plasmidS were extracted to constnict a plasmid library of the ATCCi3869 strain. The plasmid library was transformed to the 2A-1R stram obtained in Example 3 by the electric pulse method, and the transfonoed cells were applied to a minimum agar medium containing 25 )ig/ml kanamycin. The strains that showed an increased colony-forming rate were selected. By extracting a plasmid firom the selected strains showing the increased colony-forming rate, it was found that the fragment having a nucleotide sequence shown in SEQ ID NO: S was inserted into the BamHI site of pVK9. The obtained plasmid was named pLSk.
[0225] Comparison of the nucleotide sequence ins«ted in the pLSlc with the already published genome sequence of Corynebacterium glutamicum ATCC13032 (Ace. No. NC_003450) showed that pL5k contained only one ORF encoding the amino acid sequence shown in SEQ ID NO: 6. [0226] The program "SOSUT" availrf>le on the Internet
(sosuL.pn>tBome.bio.tuatac.jp/sosuiframeOE.htmr as of 2004/10/07) was used to predia whether tiie ORF encodes a membrane protein. Results of analysis of the ORF by using "SOSUI" suggested that five transmembrane regions are presoit in tiiis amino acid sequence. In the amino acid seqi^ice of SEQ ID NO: 6, the transm«nbrane regions correspond to the regions of amino acids 1 to 23, amino acids 25 to 47, amino acids 62 to 84, amino acids 86 to 108, fu)d amino acids 110 to 132. DNA sequences encoding these regions correspond to the nucleotides 1437 to 1505, nucleotides 1509 to 1577, nucleotides 1620 to 1688, mideotides 1692 to [760, and nucleotides 1764 to 1832 of SEQ DD NO: 5. Each of the amino acid sequences of these regions is shown in SEQ ID NOS: 25 to 29 and Table3.

55
|b227]

No, N-teatmin
ai
position transmembrane region C-tamin
a)
position type length SEQ ID
1 I MILGVPrQYLLYSLWNWI VDTGF 23 SECONDA RY 23 25
2 25 VAIILVLAJPLIPRJGRLAMR HK 47 PRIMARY 23 26
3 62 QLAFAGVGVYIAQIVAFF MLAVS 84 PRIMARY 23 27
4 86 MQAFGFSLAGAAffATIAS AAIG 108 SECONDA RY 23 28
5 no GAQSIVADFLAGFFILTEK QFGV 132 SECONDA RY 23 29
[0228] A further search of the litetature revealed that the ORF is named YggB (NCg 1221) (FEMS Microbiology letters 218(2003)305-309).
[0229] Examples
identification of the mutation introduced into the yggB gene of the 2A-iR stnun>
The yggP gene could recover the growth of the 2A-IR strain in the minimum medium, which suggested the possibility that the ygg3 gene of the 2A-IR strain has some mutations. Accordingly, the nucleotide sequence of the yg^ gene of the 2A-IR strain was determined. The results indicated tfiat fllle 2A-IR strain, an IS was msetted into the C-terminal region of the wild-type yggB gene (Fig. 5). The nucleotide sequence of flie mutant type yggB gene derived fiom the 2A-1R straui is shown in SEQ ID NO: 7 and the 'Corresponding amino acid sequence is shown in SEQ ID NO: 8. [0230] This sugg^ed the possibility that tito ability ofthe2A4R strain to produce L-glutamic acid in the presence of an excess amount of biotm was due to the mutton in

56
the yggP gene. If^ould be noted that this mutation was presoit not cmly in the 2A-IR strain, but also m the 2A-1 strain. This mutation is presumed to have occurred as a suppressor muta^ to stably excrete L-glutamic acid from the cell when the odhA gene was disrupted. The mutation in x^ich an IS was inserted was named the 2A-1 type mutation.
fOZSI] Example 6

[0232] (6-1) Introduction of the 2A-1 (ype mutation intx> a wild-type strain and
evaluation of L-glutamic acid-producing ability (single cross-over recombinants)
PCR was performed using the chromosomal DNA of the 2A-1 stain as a template and synthetic DNAs shown in SEQ ID NOS: 9 and 10 a$ primers to amplify the fragment of yggB gene having the 2A-1 type mutadoa The amplified product was treated with SacI and mserted into the SacI site of pBS3 obtained in Example 1 to thereby obtain a plasmid contiuning the 2A-1 type mutant yggB gene (pBS3yggB2A).
[0233] TheobtainedpBS3yggB2AwasuitroducedintoC.g/«
OD620(xI01) GIu(g/L)

ATCC13869 0.625 0.3
2A-1R 0.334 15.5
13869-2A 0.582 3.6
Blank 0.002 0.6
[0236] (6-2) introduction of the 2A-1 type mutant yggB gene into the wild-type strain and evaluation of L-gtutamic acid-producing ability (double cross-over recombinants)
To construct a strain having only the mutant-type yggB gene, the 13869-2A strain was cultured in the CM-Dex liquid medium for one night and die obtained culture was spread over the S10 agar medium (100 g/l sucrose, 10 g/l polypeptone, 10 g/t yeast extract, 1 g/l KH2PO4,0.4 g/l MgS04-7H20,0.01 g/l FeSO4-7H2O,^.0l gA MnS04-4-5H20.3 g/l urea, 1.2 g/l soybean protem hydrolysates, 20 g/l agar, adjusted to pH 7.5 with NaOH: autoclaved at 120°C for 20 minutes) and cultured at 31 J°C. Among tiie colonies which appeared, tiie strain exhibiting sensitt^ty to kataamycin was isot^sd on CM2B agar medium. Chrompsomal DNAs were prepared from the strains. Thtai, PCR was poformed using synthetic DNAs shown in SEQ ID NOS: 9 and 10 as primers to confirm tiiat the strain has only a mutant-type yggB gene. The strain containing the mutant-type yggB gene in ■wUdt an IS-like sequence was inserted was named 13869-2A-7. [0237] The ability of the obtained 13869-2A-7 strain to produce L-glutamic acid in the presence of an excess amount of bioiin vna evaluated by the method described in Example 2. The results are shown in Table 5 (OD6Z0 is turbidity at 620nm of culture soludon diluted to 101 times, and mdicates the cdl amount, and Glu (g/L) indicates die amount of accumulated L-glutamic acid). The 13869-2A-7 strain producedirghitamic acid

58
equivalent to or higher thabthe 2A-1R strain, which confirmed that L-glutamic acid production of the corydtefonn bacterium in the presence of an excess amount of biotin was caused by the mutation in the yggB gene.
[0238]

OD620(xl01) Giu(g/L)
ATCC13869 0.648 0.4
2A-IR 0.420 13.8
13869-2A-7 0.414 16.1
Blank 0.002 0.7
[0239] Example 7

As a result of the screening using the above-mentioned L-glutamic acid-producing odhA gene-dismpted strain (AsucA strain), five kinds of mutatitais were identified on HOB yggB gene besides the above-mmtioned 2ArI mutation. Hoeinafter, these mutations wore named A1-type mutation, 19-type mutadon, L30-type mutation, 8-type mutation, and 66-type mutation. The mutant-type ygg^ genes having each of the Al-type mutation, 19-type mutation, and L30-type mutaticm were introduced into Hat duomosome of the ATCCl 3869 strain, and the effect of each mutation was evaluated. The mutant-type yggB gene having the 8-type mutation was introduced into the chromosome of the ATCCi4067 strain, and the effect of the mutation was evaluated. The mutant-type yggB gene having the 66^type mutation was introduced into the chromosome of the C. melassecola ATCCl 7965 strain, and the effect of die mutation was evaluated. [0240] "Hie Al type mutation is a mutation yMck mstsds "TTCATTGTG" xvsxt to the "G" at position 1480 of the wild-type y^B gene (the wild-type g»e of C ghitamicum is shown in nucleotides 1437^303 5 of SEQ ID NO: 5), and Ie 6 Amount of L-glutamlc acid produced by the control strain, and the mutant-type (A 1 -type) yggB gene-introduced strain>
strains OD620(xI01) Glu(g/L)
ATCC 13869 0!650 o!5 ~"
ATCC13869-AI 0.548 8.6

61
conventional method. It was found that the l9-type mutant yggB gene-introduced strain^,
produced L-glutamic acid in the presence of an excess amount of biotin in a greater amount
as compared to the contra! strain. . ' ■ "
[0247]
The L30 type mutation is a mutation which replaces the "C" at position 1768 of the wild type yggB gene (the wild type gene of C. glutamicum is shown in SEQ ID NQ: 5) with "T", and causes replacement of the Aiaat position 111 with Val in the amino acid sequence shown in SEQ ID NO: 6. The nucleotide sequoice of the mutant type yggB gene having this type of mutation is shown in SEQ ID NO: 23 and the amino acid sequence of the mutant type YggB protein encoded by this gene is shown in SEQ ID NO: 24. [0249] In- the same manner as in Example 7, the L30 type mutant yggB gene-introduced strain was constructed. Specifically, PCR was performed by using the syntfietic DNAs shown in SEQ ID NOS: 30 and 37 as primers and the duwmosomal DNA of ATCC13869 strain as a template to prepare an N-terminal fragm^it Similarly, PCR. was performed by . using the synth^c DNAs shown in SEQ ID NOS: 34 and 38 as primers to pnpaxe a C-termiital fragment. Subsequently, PCR was performed by using an equimoiar mixture of die N-terminal fragment and the C-terminal fragment as a tetnplate and the synthetic DNAs shown in SEQ ID NOS: 9 said 34 as primers to amplify a partial fragment of the L30 type mutant yggB gene. The obtained yggB fiagm^t was treated with SacI and inserted into the SacI site of pBS4S to obtain a piasmid for introducing this type of mutation. Tlie thus

62
obtsdned pBS4yggB-L was introduced into the cteomosome of ATCC13869 strain in the same manner as described in Example 6, andlfaen the vector portion was cured firom the diromosome. The nucleotide sequence of the yg^ gme of the obtdned kanamycin-sensttive strain was determined and the strab havbg L30 type mutant ygg|B gene was selected. Tlie L30-type mutant yg^ gene-introduced strain was named ATCC13869-L30 strain.
[0250] TheATeC13869-L30strainandthecontrolATCCl3869strainwereculturedin the same manner as in Example 2. After completion of the culture, the amount of L-glutamic acid which had accumulated in ti)e cultute broth was measured by a conventional method. The results are shown in Table 8 (OD620 is turbidity at 620nm of culture solution diluted to 101 times, and indicates the cell amount, and Glu (g/L) indicates the amount of accumulated L-glutamic acid). ATCCl 3869-L30 strain having the 1.30 type mutant yggB gene caused accumulation of L-glutamic actd in a greater amount as compared to the parent ATCCl3869 strain.
[0251]
OD620(xl0l) Glu(g/L)
. ATCC13869 0.650 0.5
ATCC13869-U0 0.389 15.9
[0252] Example 10
Evaluation of the mutant-type ygg^ gene-introduced strains under.L-ghftamic
acid-producing conditions
L-glutamic acid ptodaoiion of Coiyneform badierium is induced by addition of surfactants such as Twe«i40 or by limiting the biottn c(»iceDtiatioB. Therefore, the ATCC13869 straui and the ATCC13869-19 s&ain were cultured under a condition containing Tween40 and a condition containing a limited amount of biotin, respectively. [0253] Each of ftie strains was mocuiated ii^ 20 ml of seed caltine medium (80 g/1

63
glucose, ad g/I ammonium sulfate, 1 g/1 KH2PO4,0.4 g/1 MgS04.7H20,0.01 g/1 FeS0,.7H20,0.01 g/1 MnS04.4-5H20.200 fig/1 VBl, 60 jig/l biotin, 0.48 g/I soybean hydrolysate (T-N), adjusted to pH 8.0 with KOH: autoclaved at 115''C for 10 minutes), and then 1 g of sterilized calcium carbonate was added thoeto, followed by a shaking culture at 31 .S'C. The culture solution obtmed after complex consumption of sugars was used as a seed culture solution in the following main culture.
[02S4] For culturing with Tween40,2 ml of the seed culture solution was inoculated into 20 ml of main culture medium (80 g/1 glucose, 30 g/I ammonium sulfate, I g/1 KHjPO«, 0.4 g/1 MgS04.7H20,0.01 g^ FeS04.7H2O, O.Ol g/1 MnS04.4-5H20, 200 ^ig/l VBl, 60 Hg/1 biotin, 0.48 g/1 soybean hydrolysate (T-N), adjusted to pH 8.0 with KOH: autoclaved at 1 IS^C for 10 minirtes), and then I g of sterilized calcium carbonate was added thereto, followed by a shaking culture at 313<*C. When OD620 of culture broth dihited 101-fold reached 0.2, Tween40 was added to a final conc(»itratioh of Sg/L and the culture was continued.
(02S5] For culturing with limited biotin, 1 mi of the seed culture solution was inoculated into 20 ml of main culture medium (80 g/l glucose, 30 gA ammonium sul&te, I g/l KH2P04.0.4 g/1 MgS04.7H20,0.01 g/l FeS04.7H30.0.01 g/1 MnS04.4-5H20,200 ugA VBl, 0.48 g/1 soybean hydrolysate (T-N), adjusted to pH 8.0 with K.OH: autoclaved at 115<^ for 10 minutes), and diot 1 g of sterilized calcium carbonate was added theteto, followed by a shaking culture at 31 .S^C. Under these culture conditions, a final concentration of biotin is calculated to be about 2.9^g/L.
[0256] After a 40 hour-culture, the amount of L-glutamic acid which had accumulated in the medium was measured for the TweeD40-added culture and the biotin-limited culture. The result is shown in Table 9. It was found that the ATCC13869-19 strain produced L-glutamic acid in an amount greato: than the OKitrol stram under L-glutamtc acid-ptodudng conditiotis.

64
[0257]

Stnuns OD620(X101) Glu(g/L)
Tween40-added
ATCCl 3869 0.538 25.6
ATCCB869-I9 0.395 28.6
biotin-limited
ATCCI3869 0.462 36.0
ATCC13869-19 0,431 40.0
[0258] The wild-type ATCC13869 strain, yggB mutfflit|rains ATCC13869-19, ATCC13869-A1, ATCC13869-L30, and a strain having a plasmid containing a wild-type yggB gene (ATCCl3869/pL5k-l), and a strain having acQntrol plasmid (ATCCl 3869/pVK9) were cultured with Tween40. Each of diese strains were cultured on a CM-Dex plate medium overnight, and cells collected from 1/6 area of the plate were inoculated in 20 ml of a flask medhim (80g/i glucose, 30 g/l ammonium sulfite, 1 g/I KH2PO4,0.4 g/l MgS04-7HiO, 0.01 g/l FeS04-7H30,0.Ol g/l. MttSO4-4-5H20,200 Mfi^ VBl, 60 ng/1 biotin, and 0.48 g/l soybean hydrolysates (T-N: total nitrogen),adjusted to pH i.O with KOH: autoctaved at 115°C for 10 minutes), followed by addition of 1 g of heat-sterilized calcium caibonate, and each of die strains was cultured with sbddng at 31 .S^C. After a 5-hour culture, Tween40 was added to a final concentration of 1 g/L and the culture was continued. Table 10 shows Hit amount of cells (OD620) and the amount of L-glutamic acid which had accumulated in the medium after 24 hours. It was found diat the ATCCl3869-19 strain, ATCC13869-AI strain, ATCCI3869-L30 strain, and the straiii having a plasmid containing a wild-type yggB graie have an »ihaiKed ability to produce L-glutamic acid under L-glutamic acid-producing conditions.

65

ttf259]

The S-lype mutation is a mutation which replaces the "G" at position 837 of SEQ ID NO: 61 with an "A", and causes replacement of the Ala at position 111 with Thr in Ifae amino acid sequence of SEQ ID NO: 62. Ttie nucleotide sequence of the mutant-type yggB gene havug liiis type of mutation is shown in nucleotides S07--2093 of SEQ ID NO: 63 and the amino acid sequence of the mutant-type Yggp3 protein encoded by this gene is shown in SEQ ID NO: 64.
[0261] In the same manner as in Example 7, the 8-type mutant yggB goae-mtroduced strain is constructed. Specifically, PCR is performed by using the synthetics DNAs shown in SEQ ID NOS: 30 and 6S as primers and the cbFomosomal DNA otBrevibacterium flavum ATCC14067 strain as a temple to prcf>are an N-tominal fragment Similarly, PCR is performed by using the synthetic DNAs shown in SBQ ID NOS: 34 and 66 as primers to prepare a C-tcrminal fi»gmenL Subsequently, PCR is perforined by using an equimolar mbcture of the N-terminal fragment and the C4erminal fivgmoit as a template and the synthetic DNAs diown in SEQ ID NOS: 9 and 34 as primers to amplify a partial fragment of the 8-type mutant yggB gene. The obtabed yg^ gene fragment is treated with SacI and inserted into the SacI site oCpBS4Sto obtain a plasmid for mtroducing this
fof mutation. The thus obtained pBS4yggB8 is btroduced mta the chromosome of ATCC14067 strain in the same manner as described in Example 6 and then the vector

66
portion is cured fixim the chromosome. The nucleotide sequence of the yggB gene of the obtained kanamycui-sensitive strain is determined and die strain having the 8-type mutant yggB gene was selected. The S-type mutant yggB gtaie-intcoduced strain is named A.TCC14067-yggB8 strain.
[0262] Example 12
[0263]
The 66-type mutation is a mutation w^ich replaces die "C" at position 1673 of SEQ ID NO; 67 with a'T', and causes r^lacemmt of the Pro at position 424 with Leu in the amino acid sequence of SEQ ID NO: 68. The nucleotide sequence of the mutant type-yggB gene having this type of mutation b ^own in SEQ ID NO: 69 and the amino acid sequence of the mutant type Yg^ protein encoded by this gene is shown in SEQ ID NO: 70.
[0264] In the same manner as in Example 7, the 66-type mutant yg^ gene-introduped stnia is construct. Specifically, PCR is performed by using the synthetic DNAs shown in SEQ ID NOS: 30 and 71 as primers and the chromdsomal DNA of C. melassecola ATCC1796S stnun as a VextxiMt to prepare an N-terminal fragnwnt Simiiarty, PCR is performed by using the synthetic DNAs shown in SEQ ID NOS: 34 and 72 as primos to prepare a C-terminal fragment Subsequently, PCR is performed by using an equimolar mixture of the N-tetminal fragment and the C-terminal fragment as a template and the synthetic DNAs shown in SEQ ID NOS: 9 and 34 as primns to anq>Ufy a partial fragment of the 66-type mutant yg^ gene. The obtained yggB fragment is treated with SacI and inserted into the SacI site of pBS4S to obtain a plasmid for introducmg diis type of mutation. - The thus ol^ined pBS4yggB66 is introduced into die chromosome of ATCC17965 strain in the same msaxas as described in-Example 6 and then die vector portion is cured from the chromosome. The nucleotide sequence of the yggB gene of the obtainied kanamycin-sensitive strain is determined and the strain having dw 66-type mutant yggB gene was selected. The 66-type mutant yggB g«)e-introduced strtun is named

67
ATCCl7965-yggB66 Strain. ....
[0265] Example 13

Mutant-type yggB genes may be obtained by introducing a random mutation into the wild-type yggB gene in vitro, transforming a ooryneform bacterium with the mutation-introduced yggB gene, and selecting mutant strains capable of producing L-glutamic acid widiout addition of sur&ctants or penicillin in die presence of an excess amount of biotin. [0266] (13-1) Construction of a yggB gene-disnipted strain
To perform screening for mutant-type yggB genes, first a yggB gene-disrupted stram was constructed. PCR was performed by using the syntiietic DNAs shown in SEQ ID NOS: 39 and 40 as primers and the chromosomed DNA of ATCC13g69 stnun as a template to prepare an N-traminal fragment. Sinularly, PCR was performed by using the synthetic DNAs shown in SEQ ID NOS: 41 and 42 as primers to prepare a C-terminal fragment. SEQ ID NO: 40 and SEQ ED NO: 41 are complementary to each othor. Subsequendy, PCR was performed by using an equimolar mixture of the N-tenntnal fragment and die C-terminal fragment as a template and the synthetic DNAs shown in SEQ ID NOS: 39 and 42 as primers to obtun a fragmoit containing a yggB gene in which the ORF is deleted.
[0267] The obtsuned PCR fragment was treated with SacI and mserted into the SacI site of pBS4S to obtain a plasmld useful for disrupting the yg^ gene. The pBS4AyggB dius obtained was introduced into the chromosome of ATCC13869 strain in the same manner as described in Example 6 and die vector portion is cured from die duomosome. PCR \Vas performed by using the chromosomal DNA of the obtained kanamycin-sensitive strain as a template and the syntiietic DNAs of SEQ D NOS: 39 and 42 as primers to confirm tiiat the yggB gene was disrupted. The obtained ygg^-disnipted strain was named ATCC13869AyggB stiain.

68
[0268] (13-2) in vitro screening ofmutant-typeyggB genes
Mutagenesis of the yggB gene was p^ormed as follows. First, the above-described pLSk plasmid was treated widi Xhol and Sail and self-Iigated to remove die region other dian the yggB gene, and thereby die plasmid pLSkXS was obtained. A Sail recognition site does not exist on the oucleodde sequence of SEQ ID NO: S but is present on the multi-cloning site of pBSS. About 10 pg of the obtained pLSkXS was dissolved in SOO mM phosphate buffer containing 400 mM hydroxylamine and 1 mM EDTA (pH 6.0), and heated at 75°C for 30 to 90 minutes to introduce a mutation. The plasmid after mutagenesis treatment was desalted using SUPREC-02 (manufactured by TAKARA BIO INC.) and flien introduced into ATCC13869AyggB strain by die raediod described in Example 6. transformed cells were screened on die C1M2B medium containing 2S |ig/ml of icanamycin. As a control, pLSkXS.witbout mutagenesis treatment was introduced into die ATCC13869AyggB strain. The ajq>6ared transformants are inoculated into 2 m(of a liquid CM2BGU2 medium (CKf2B medium described in Example 3 fiudier containing 10 g/1 glucose and IS g/1 urea) and cultured at 3 l.S°C for 5 hours widi shaking, followed by determination of the concentration of L-glutamic acid which had accumulated in the culture brodi.
[0269] Table 11 shows the result of cuhuring of the strain obtained by transforming die ATCC13869AyggB strain widi the mutated pLSkXS on «» CM2BGm medium. Three strains which cause accumulation of more than Ig/L of L-glutamic acid were obtained among the traosformants transformed with 60, or 90-minute mutated plasmids. The amount of L-giutamic acid contained in die starting mecfium is 0,16g/L, and die amount of L-glutamic acid which had acx;umulated by die control ATCC13869AyggB/pL5kXS (without mutagenesis treatment) strain was 0.3 Ig/L.
[0270] Table 12 shows the results of otltuiing transformants obtained by transforming die ATCC13869AyggB strain widi mutated pLSkXS on die CM2BGU medium, wfaidi has die same composition'«s the above-mentioned CM2BOU2 medium except diat die concentration of urea is l.Sg^. One clone M^ich causes acoimulation of more dun Ig/L of L-glutamic acid was obtained among the transformants transformed with 90-minute

(B
mutated plasmids.
[02713

Bki aisiHiaHitattan
(g/D Number of diones

' Tipis of autactoii«ci«

SOnin . eOniln. 90ittin
. ." s-iuSo.4. :■•• ■ i^-' ■:■■■■■ .36 39
d..4^.eiu$a,6 . ■ 9:- • 11, ■;-.6..
d.6<&iu£oi8 • ■ a. 0 ■' 1
0,8

8 hi apcu mulct ion'' Number of ciohas

IwHt'pf IMltaSMMWis

eOiiiin' 90Bdn
GluSO.7 45 41
0.7
. Strajlaa / \-.' . : Oi?620 (klOl). Glu(QyL)
ATCClSeeMyggB/pLSlQCSt?"' irMliw»t) .0^253 ,0.58
ATCC1^869-^Y
(14-1) Evaluation of resistance to 4-fiki
ThB yggB gene and odfaA gene-double mutant steain were prqiared by mtroducing a.nnita^ype odhA g«ie into the above-mentioned ATCC13869-L30 stnun. [0282] First, each of die mutations shown in Table 14 was introduced into die odhA geqe ^coding Elo subunit of a-KGDH on the duomosonie of die ATCC13869-L30 strain, bi T{j>le 14, nucleotide sequences of the region corcespoading to nucleotides 2528 to 2562 of SEQ ID NO: 43 in each of the mutant-type odhA gene are shown. In T^le 15, amino aicid sequences of die region corresponding to amino acids 696 to 707 of SEQ ED NO: 44 in each of the amino acid sequences encoded by the mutant-type odhA genes are shown. [0283] The L30sucA8 strain in which the odhA gwe having die mideotide sequence of SEQ ID NO: 45 is introduced can be obtained as follows. The mutant odhA gene fragment is pr^>aied by PCR using primers of SEQ ID NOS: 53 and 54. The.obtained fragment is digested widi^BamHI and ligated to dw BamHI site of plasniid pKF19m which is attadied to Mutah-Sup«FExptess Km (Takara Bio). Th«>, PCR is performed using a primer of SEQ ID NO: 55 having a phosphorylated 5'-end cpd the selection primer of Mutan-Super

73
Express Km, and the obtained PCR product is used to transform supChiB. coli strain, sudi as MVI184 strain, to obtain a plasmid containing the mutant odhA fiagment This fragment is inserted into die pBS4S plasmid and the obtained plasnid is used to transform die ATCCl 3869-L30 strain to thereby obtain a strain in whidi the plasmid is integrated into \ts chromosome. Then, a strain which is resistant to sucrose and sensitive to kanamycin is selected from these strains. The nucleotide sequence of the odhA g«te of di^ selected stnains is detennined and the straui in which fimction of o-KGDH is deCcient by a frameshift mutation in the odhA gene is selected as ATCC13869-L30sucA8 (odhAS) strain. •
[0284] The other odhA mutant strains can be obtained by the similar procedures usmg the ATCCl 3869-L30 strain.
[0285] The sucASOl strain in which a mutant odhA gme havmg a nucleotide sequence of SEQ ID HO; 47 is introduced can be obtained by a stmilar mediod as described above in which a primer of SEQ ID NO: S6 having a phosphorylated S'-«id is used instead of a primcrof SEQ ID NO: 55.
[0286] The sucASOS strain in vMcii a mutant odhA genxt having a nucleotide sequ«aice of SEQ ID NO: 49 is introduced can be obtained by a similar method as described above in which a primer of SEQ ID "NO: 57 having a phosphoi^^bted 5'-end is used instead of a primer of SEQ ED NO; 55.
[0287] The sucA77 strain in which a mutant odhA gate having a nucleotide sequence of SEQ ID NO: 51 is introduced can be obtained by a sbnilar rn^od as desoibed above in which a primer of SEQ ID NO: 58 having a phosphorylated 5'-end is used instead of a primer of SEQ ID NO: 55.
[0288] The. LBOsucAS strain does not have mtraceliular ot-KGDH because die sucAS mutation is a frame-shift mutation v^ich causes an immahire truncation of die a-KQDH protein. On die other band, sucASOl strain, sucASOS strain, and 8UcA77 strain have decreased but some o-KGDH activity because these mutations are not frame-shift mutations iuid do not cause immature truncadon of the ot-KGpH pptein.

74
[0289] Table 14: partialtnucleotide acid sequrace of odhA mutant genes

Strains

Nucleotide sequence of odhA gene

ATCC13869-L30

CTG GCT AAG CTG CGI GGC TAG GAG GTC GGA GGC ACC

L30sucA8
L30sucA80l
L30sucA805

CTG GOT AAG CTG CGT
CTG GCT AAG CTG CGT
CTG GCT AAA AGC TGC

C GAG GTC GGA GGC ACC
CTC GAG GTC GGA GGC ACC
GTC GAG GTC GGA GGC ACC

L30sucA77

CTG GCT ATA AGC TGC

GTC GAG GTC GGA GGC ACC

[0290] Table 15: amino acid sequence of odhA mutants

Strains

Amino acid sequence of Elo subunit

Wild
L30sucA8(flsucA)
Ii30sucA80I
L30sucA805
L30SUCA77

Leu Ala Lys Leu Arg Gly Tyr Asp Val Gly Gly Thr
Leu Ala Lys Leu Arg
Arg Arg Arg Arg His
Leu Ala Lys Leu Arg — Leu Asp Val Gly Gly Thr "Leu Ala Lys Ser Cys — Val Asp Val Gly Gly Thr
Leu Ala lie Ser Cys Val Asp Val Gly Gly Thr

[0291 ]
L-glutamic acid productivity of the obtained ygg gene and odhA gene-double mutant strain was evaluated by culturing diese strains in a Saki^dii flask. Each of fiie strdns listed in Table 14 was cultured at 31 .S°C ovonight on CM-Dex agar medium, and then 1/6 of the culture was transferred to 20 ml of a medium contiuning 60 g/1 glucose, 22.5 g/1 (NH4) 2SO4,1 gA KH2PO4,0.4 g/1 MgS04-7H2t), 0.01 g/1 FeSO^THjO, 0.01 g/1 MnS04-4-5H20,200 ng/l vitamin Bl, 0.48 g/1 soybean protein hydiolysate, and 300 ng/1 biot'm (adjusted to pH 8.6 with KOH), added with CaCOs and cultured with stirring at llSrpmat31.S°C. The amount ofaccumulated L-glutamic acid ater 19 hours of culture is ^own hi Table 16. The sucASOl, sucASOS, and sucA77 strains exhibited higher L-glutamic add productivity than the ATCC13869-L30 Wain canying a wild-type odhA goa&and the sucAS strain carrying odhA gene with a ftame-^ft mutation. These results showed that L-glutaniic dctd is efficiently produced by tegulat'mg o^KQDH activity by

75
introducing Kiutations-lnto the proximate of thiamine pyrophosphate binding region of the odhAgene.
[0292] Table 16: L-glutamJc acid production by odhA mutant strains

Strain L-glutatnic acid (gfL)
ATCX:i3869-L30 4.9
L30sucA8 19.8
L30sucA80I 22.1
L30sucA805 23.8
L30sucA77 21.6
[0293] Example 16

The odhA gene was disrupted in the ATCC14067strBJn and the ATCC14067yg^8 strain, respectively and the obtained sbains were cultured. First, a plaamid Sat disrupting the odhA gene was constructed.. PCR was praformed using tfie synthetic oligonucleotides shown in SEQ ID NOS: 77 and 78 as priniras and chromosomal DNA of ATCC14067 strain as a template to amplify a fragment covering the N-terminal region of the odhA goie. Anotho: PCR was performed using this synthetic oligonucleotides shown in SEQ ID KOS: 79 and 80 as primers and chromosomai DNA of ATCC14067 strain as a tni^>late to amplify a fragment covering C-terminal region of the 0(fiiA guie. Subsequmtiy, PCR was performed using a mixture of equimolar amounts of the N4enninal fragment and the C-terminal fragment as a template and synthetic DNAs of SEQ ID NOS: 81 and 82 as priihers to prepare a fragment in which an internal sequoice of the ocfiiA gene is deleted. The obtained PCR product was digested with BamHI and inserted into the pBS4S coDSthicted in Example 1 to obtain a plasmid pBSAsucA47.
[0294] ATCG14067 strain and ATCC14067ygg3B8 of Example 11 were transfcwmed with the pBSAsucA47 in Uie same way as described in Example 6 to intioducethe deletion-4ype odhA gene into die chromosome and remove only the veotor portion. Strains

76
in which the odhA gene was (Usrupted.,wete selected from kanamycin-sensiUve strains by PCR using the primers of SEQ ID NOS: 77 and 80. The tiius obtained strains were named ATCC14067AodhA strain and ATCC14067AodhA yggB8 strain, respectively. [0295] Table 17 shows tiie results of cultivation of the ATCC l4067AodhA strain and the ATCCl4067AodhA yggB8 strain according to flje method described in Example 3. It was found that introductJonof the 8-type mutant yggB g«ie enhanced L-gJutamic acid producing ability of the odhA gene-disrupted sttain.
[0296]
OD620(xl01) Glu(g/I.)
ATCC14067^oahA ■ 0.270 STT
ATCCl4067^odhA yggB8 0.242 22.0
[0297] Example 17
^Disruption of the symA gene in the yggB-mutant strain>
The symA gene was disrupted in the 2A-1R strain constracted in Example 3 and having an IS inserted mto the yggB gene, and the obtained strain was cultured in comparison with the 2A-1R strain. The nucleotide se^uoice of the symA gene from the ATCC 13869 strain is shown in SEQ ID NO: 86, and the amino acid sequence is shown in SEQ ID NO: 87. First, a plasmid for &e purpose of disn4>ting the symA gene was constructed. PCR was performed using the synfliedc oU^mucleotides shown in SEQ ID NOS: 88 and 89 as inimers and dmmiosomal DNA of ATCC13869 strain as a template to amplify a fragmoit covering N-terminal region of the symA gene. Another PCR was performed using the synthetic oligonucleotides shown in SEQ ID NOS: 90 and 91 as primers and chromosomal DNA of ATCC13869 strain as a tonplate to amplify a fragment covering the C-teimipal regjon of the symA gene. Sub^ueotly, PCR was performed using a mixture of equimolar amounts of the N-temsinal fragment and the C-terminal fragment as a template and synthetic DNAs of SEQ ID NOS: 88 and 91 as primers to prepare a fragmwit in which an internal sequ«ice of the SymA'gene is deleted. The

77
. obtained PCR product was digested with BamHI and inserted into tiie pBS4S constructed inExample 1 toobtab aplasinidpBSAI867.
[0298] T1ie2A-IR strain was transformed with die pBSA 1867 in the same way as described in Example 6 to introduce the del^'on-type SymA gene into the chromosome .. and remove only the vector portion. Strains in which the SymA gene was disrupted were selected from kanamycin-sensitive strains by PCR using the pnmets of SEQ ED NOS: 88 and 91. The thus obtained strains were named 2A-IRASymA strain. [0299] Table 18 shows the results of cultivation of the 2A-lRASymA strain and the 2A-IR strain according to the method described in Example 3. It was found that deletion of the SymA gene iiirtber enhanced L-glutamic acid producing ability of the mutant-type yggB gene-introduced strain.
[0300]

1 1—l^
pL5lc containing a wild-type yggB gene off a corynefoirm baderium was used to introduce the wild-type yggB goie into coryneform bacterium. A plasmid having a similar structure as pL5k can also be constructed by performing PCR using primers of SEQ ID Nos: 59 and 60 and chromosomal DNA of ATCC13869 stndn as a template, digesting the amplified product with BamHI, and inserting die resulting fiagmeat into the BamHI site of pVK9. pVK9 is a shuttle vedtor which was obtained by blunt-eiiding the Avail ate of pHSG299 (Takara Bio) and &sertmg tiierein a fragment Involved in autonomous Implication in coryneform bact«ia'excised bom pHK4 (JP-A-05-007491) with BamHI and-"Kpni.

78
{0302] A Corynebacterium glutamicum ATCC13869 strain was transformed with pL5k by .the electric pulse method (JP-A-2-207791), The transformed cells were spread over a CM2B plate medium (10 g/'l polypeptone, 10 g/1 yeast extract, 5 g^ NaCl, 20 gi^l agar, adjusted to pH 7.0 with KOH) and cultured at 31 J^C for one night On the nejct day, colonies which appeared were purified on a CM2B plate medium containing 25 jig/ml kanamycin to obtain a wild-type yggB-gene amplified strain. Plasmids were extracted from the transformants" by a conventional method to confirm that the target plasmld was introduced. The wild-type yggB-gene amplified strain thus obtained was named ATCC13869/pL5k. As a control strain, ATCC13869/pVK9 introduced with pVK9 was constructed in ^e same manner as described above.
[0303J Example 19
Iified strain (ATCC13869/pLSk) caused accumulation of more L-glutamic acid than the vector-introduced straip. (Table 21)
[031OJ Table 21 Amount of L-glutamic acid accumulated und^ penicillin G-«dded conditions